Technical ReportPDF Available

Eurofound. Health and Well-being at Work: A Report Based on the Fifth European Working Conditions Survey

January 2012

Report number: EF1302EN
Affiliation: Laboratorio Revelli, University of Torino and ASL TO3 Piemonte

Authors:

Chiara Ardito

European Commission

Roberto Leombruni

University of Turin

Lia Pacelli

University of Turin

Angelo d'Errico

Local Health Unit TO3, Piedmont Region Italy

Average gender gap in well-being, by country

…

Job satisfaction by country (%)

…

Workers who are satisfied or very satisfied with their working conditions, by sector and gender, 2010 (%)

…

Average level and variability of well-being by job satisfaction level and gender

…

Distribution of workers reporting 'bad' and 'very bad' health, by number of mentioned health problems and gender (%)

…

Figures - uploaded by Chiara Ardito

Content may be subject to copyright.

Content uploaded by Chiara Ardito

Content may be subject to copyright.

Health and well-being at work

Wyattville Road, Loughlinstown, Dublin 18, Ireland. - Tel: (+353 1) 204 31 00 - Fax: 282 42 09 / 282 64 56

email: information@eurofound.europa.eu - website: www.eurofound.europa.eu

Click for contents

A report based on the

fifth European Working Conditions Survey

Authors: Chiara Ardito, LABORatorio Revelli; Angelo d’Errico, ASL TO3 Piemonte; Roberto Leombruni,

LABORatorio Revelli and University of Turin; Lia Pacelli, LABORatorio Revelli and University of Turin

Research institute: LABORatorio Riccardo Revelli, Turin

Research managers: Agnès Parent-Thirion, Greet Vermeylen, Gijs van Houten and John Hurley

Project team for fifth EWCS: Agnès Parent-Thirion, Greet Vermeylen, Maija Lyly-Yrjänäinen, Gijs van Houten,

Isabella Biletta and Sophia MacGoris

Project: European Working Conditions Survey

When citing this report, please use the following wording:

Eurofound (2012), Health and well-being at work: A report based on the fifth European Working Conditions

Survey, Dublin.

Contents
Executive summary
Introduction
1. Health and well-being across Europe
2. Psychosocial work environment
3. Determinants of well-being
4. Work environment and health
5. Absenteeism and presenteeism
6. Conclusions
Bibliography
Annex: The European Working Conditions Survey series
1
3
7
23
35
43
55
65
67
83

Abbreviations used in the report

EU-OSHA European Agency for Safety and Health and Work

EWCS European Working Conditions Survey

ISCO International Standard Classification of Occupations

LFS Labour Force Survey (Eurostat)

NACE Nomenclature générale des activités économiques dans les Communautés européennes (General

industrial classification of economic activities within the European Communities)

ILO International Labour Organization

Country codes

EU27

The order of countries follows the EU protocol based on the alphabetical order of the geographical names of countries

in the original language.

BE Belgium FR France AT Austria

BG Bulgaria IT Italy PL Poland

CZ Czech Republic CY Cyprus PT Portugal

DK Denmark LV Latvia RO Romania

DE Germany LT Lithuania SI Slovenia

EE Estonia LU LuxembourgSK Slovakia

IE Ireland HU Hungary FI Finland

EL Greece MT Malta SE Sweden

ES Spain NL Netherlands UK United Kingdom

Other countries

HR Croatia MK former Yugoslav Republic of Macedonia1

ME Montenegro TR Turkey

Potential candidates

AL Albania XK Kosovo2

Other

NO Norway

Country groups

EC12 12 EU Member States prior to enlargement in 1995

EU15 15 EU Member States prior to enlargement in 2004

EU27 Current 27 EU Member States

1MK corresponds to ISO code 3166. This is a provisional code that does not prejudge in any way the definitive nomenclature for

this country, which will be agreed following the conclusion of negotiations currently taking place under the auspices of the United

Nations (http://www.iso.org/iso.country_codes/iso_3166_code_lists.htm).

2This code is used for practical purposes and is not an official ISO code.

Introduction

Health and well-being are key dimensions of the policy debate on how to improve the lives of individuals in society.

Health and well-being have an intrinsic value and are fundamental to the concept of progress of individuals and the

functioning of society because of their direct link with issues such as labour force participation, productivity and

sustainability. This report examines the relationship between work, health and well-being, based on findings from

Eurofound’s fifth European Working Conditions Survey (EWCS).

Work is central to a person’s well-being, as it both provides an income and is a means of broader social advancement.

Work and well-being are closely related, in that the good or bad quality of working conditions have a direct impact on

an individual’s quality of life. Work is also central to health, due to specific risk factors in the workplace which may lead

to injuries and professional diseases, work-related illnesses or long-term health consequences.

Policy context

Health and well-being at work are key elements of the overall Europe 2020 strategy for growth, competitiveness and

sustainable development. A healthy economy depends on a healthy population. Without this, employers lose out on

worker productivity and citizens are deprived of potential longevity and quality of life. This is especially important in

view of the current debate on demographic ageing of the European population.

Safeguarding the entitlement of all to work and ensuring that people of different health capacities can engage in paid

work was an objective set by EU Member States in both the Lisbon and Europe 2020 strategies. The European Treaties

legislation and policy measures recognise the importance of preserving the health and safety of workers, and maintaining

their well-being. Directive 89/391/EEC on measures to improve the safety and health of workers states that work should

be adapted to individuals and not the other way around.

Both depression and work-related stress are the focus of increasing attention, as they can lead to lower well-being and

eventually result in the incapacity to work. In 2004, the EU-level social partners concluded a European Framework

Agreement on Work-Related Stress to identify, prevent and manage problems related to work-related stress. In 2008,

the European Commission, along with relevant social partners and stakeholders, signed the European Pact for Mental

Health and Well-being, highlighting the importance of mental health and well-being for a strong and competitive

Europe.

Key findings

nPoor general health is mentioned by 2.5% of European workers while 47% report more than two health problems,

with a strong connection between the physical and mental dimensions.

nIn total, 60% of the workers who declare very good or good health are confident in their ability to do the same job

at the age of 60, while the proportion is significantly lower among those with poor health.

nJob quality is strongly and positively associated with well-being. Among its many dimensions, intrinsic job quality

and job prospects (job security, career progression, contract quality) have the most impact on well-being. As job

quality deteriorates, variability in well-being increases substantially: when faced with poor job quality conditions

large differences in the capacity to cope emerge.

Executive summary

nUnskilled workers and those in transportation, hotels and manufacturing report very demanding work situations and

insufficient control over their work. Individuals facing these ‘high-strain’ working conditions report the lowest well-

being. Social support from co-workers is the main factor helping them to cope.

nAmong the indicators mostly associated with poor health and well-being are atypical/variable working hours,

disruptive interruptions, exposure to restructuring, environmental hazards and job insecurity. On the positive side,

support, ‘rewards’ (feelings of pay fairness and of career advancement chances) and skills are important protective

factors.

nWorkers in transportation and construction are subject to the worst dimensions of the psychosocial work

environment. Employment status and gender also have a significant impact.

nUsing the mental well-being index (WHO-5) designed by the World Health Organization as a measure of emotional

and psychological well-being, 23% of workers in Europe report low levels of well-being and should be assessed for

depression, and 6% are likely to suffer from depression, with women reporting lower levels than men.

nSome 40% of workers in Europe report having been absent from work due to sickness. Absence is significantly

higher with higher job security/job protection, hinting at possible opportunistic behaviour among workers. It also

increases with psychosocial factors linked to lower well-being at work (bullying, discrimination, emotional

demands).

nA total of 41% of men and 45% of women reported having worked while ill (‘presenteeism’) at least one day in the

previous 12 months. This phenomenon is more frequent among high-grade, over-committed white-collar workers

with high autonomy and engagement with their job. The positive association observed with exposure to work

intensity, verbal abuse or discrimination, handling chemicals, awkward postures and shift work seems to indicate that

presenteeism is also increased by several unfavourable working conditions.

Policy recommendations

nPolicy interventions targeting health, well-being and safety of workers can have a significant impact if the focus is

on employment quality, the psychosocial work environment and organisational factors.

nEmployment quality is identified as a key element for workers, with a high influence on their well-being. Poor job

quality leads to worryingly low levels of well-being for individuals less capable of coping with it. The policy focus

should go beyond the average relationship of work and well-being and target a range of individual situations.

nPhysical and mental health and work safety have a weak association with traditional dimensions that tend to steer the

debate, such as industry, firm size or even job contract. The main split is between manual and non-manual

occupations; the main associations are to be found with the psychosocial work environment and organisational

determinants. Once these are taken into consideration, even cross-country differences tend to disappear.

nIn relation to the psychosocial work environment and organisational factors, the multiplicity of individual situations

should also be a policy target: low-skilled manual workers are those likely to benefit most from improvements in job

design and a more supportive work environment.

nLow well-being and poor health have a high societal cost in terms of absenteeism and presenteeism. Working

conditions have a role over and above their link with health and well-being: good working conditions are indicative

not only of better health, but also of less opportunistic behaviours in the case of absenteeism, and a lower incidence

of presenteeism.

Health and well-being at work

The health and well-being of individuals are two dimensions around which researchers and policymakers are

re-arranging the debate on how to foster the progress of societies. Health and well-being have an intrinsic value, which

should be part of the very definition of progress, and also a societal one because of their direct connection with issues

such as labour force participation, productivity and sustainability.

The aim of this report is to contribute to this debate, building on Eurofound’s European Working Conditions Surveys

(EWCS), which have proven to be a valuable source of information on the topic since the early 1990s. Offering a very

detailed view of working conditions, the surveys provide the unique opportunity to study the relationship of work with

many health dimensions and, in the fifth EWCS, with a measure of emotional and psychological well-being of

individuals.

The available literature already shows that work and well-being are closely related, and have significance beyond their

role as a means for economic and social advancement.

Work is also central to health. Literature on public health has pointed out the pivotal role work has in relation to health

issues. Specific risk factors present in a job may lead to injuries and professional diseases, to work-related illnesses, or

to other long-term health consequences. Work also has a role in determining the socioeconomic status of the individual,

which in turn has been identified as one of the main determinants of health and health inequalities (Marmot, 2005). Work

can also contribute positively to health and well-being.

Although the cross-sectional nature of the survey hinders a true causal investigation, the study maintains the view that

the relationship between work on the one side and well-being and health on the other are bi-directional. It is also true

that bad health and low well-being at work have economic consequences, for both direct (reduction in the labour supply,

costs of work illnesses) and indirect reasons (loss of motivation and capacity, increased expenses from the health and

social protection systems). The EWCS makes it possible to examine a highly relevant policy issue on this topic: the

association between health, working conditions and absenteeism/presenteeism.

Concepts and measures of well-being

Indeed, while many aspects of health rely on very precise definitions and accepted indicators, the attempt to measure

well-being has been taken up by a more recent surge of initiatives, such as the World Health Organization Quality of

Life group (WHO-QOL, 1995); the French Commission on the Measurement of Economic Performance and Social

Progress (Stiglitz et al, 2009); the European Commission ‘Beyond GDP’ initiative (European Commission, 2009). An

opinion expressed by the European Economic and Social Committee points out that progress is still at an early stage

(European Economic and Social Committee, 2011), but the debate is lively.

This helps to explain why terms such as well-being, happiness, life satisfaction and positive emotions are sometimes

used synonymously, though such constructs are theoretically separable and show different patterns. The fifth EWCS

offers two views on the theme: a single-item indicator of an individual’s overall judgement about the specific domain of

work, as measured by job satisfaction; and a multidimensional indicator, the well-being index (WHO-5) proposed by the

World Health Organization, originally developed as a measure of emotional and psychological well-being and a screener

for depression (WHO, 1990).

Introduction

Policy context

Health and well-being at work are key foundation stones of the overall Europe 2020 strategy for growth, competitiveness

and sustainable development. A healthy economy depends on a healthy population. Without this, employers lose worker

productivity and citizens are deprived of potential longevity and quality of life. This is doubly important as the European

population ages in the coming decades.

Preserving the entitlement of all European citizens to work and ensuring that people of varying health capacities can

participate in paid work was an objective set by EU Member States in the Lisbon strategy and reiterated in the Europe

2020 strategy. The European Treaties legislation and policy measures recognise the importance of preserving the health

and safety of workers, and maintaining their well-being. According to Directive 89/391/EEC on the introduction of

measures to encourage improvements in the safety and health of workers, work should be adapted to suit individuals and

not the other way around.

In this regard, better health is one way of addressing the economic challenges of Europe. It may help to support the

financial sustainability of the European social model and strengthen social cohesion. Therefore, health promotion is not

just the responsibility of the health sector because, as a Health in All Policies (HIAP) approach3emphasises, societal

objectives are best achieved when all actors include health and well-being as key components of their objectives.

With these objectives, the European Commission promoted a whole range of actions for safety and health at work. In

2007, the Commission defined the Community strategy for the period 2007–2012. This strategy was intended to provide

an integrated framework within which Member States can deliver their national policies and stakeholders can promote

common initiatives. The primary objective of the Community strategy 2007–2012, among others, was a sustainable and

uniform reduction in accidents at work by 25% by 2012. A new health and safety strategy is being drafted at the moment.

The concept of ‘well-being for all’ is fundamental to the definition of social cohesion promoted by the Council of Europe

(2008). Social cohesion is defined as ‘the capacity of a society to ensure well-being for all its members, minimising

disparities, and [it] accentuates the importance of social actors’ joint responsibility for its attainment’. The concept of

‘well-being for all’ was first introduced by the Council of Europe in its revised Strategy for Social Cohesion as the

ultimate goal of a modern society, emphasising the fact that well-being must be shared by all members of society and

cannot be attained at an individual level.4

Depression and work-related stress are, at the present time, an increasingly important cause of incapacity for work.

Acknowledging the importance of this, the European-level social partners concluded a European Framework Agreement

on Work-Related Stress in 2004 to raise awareness of the phenomenon among employers, employees and their

representatives.

Health and well-being at work

3See the White Paper presented by the Commission, ‘Together for Health: A Strategic Approach for the EU 2008–2013’, p. 6,

available online at http://ec.europa.eu/health-eu/doc/whitepaper_en.pdf.

4A new strategy for Social Cohesion, European Committee for Social Cohesion (CDCS), Revised strategy for Social Cohesion,

approved by the Committee of Ministers of the Council of Europe on 31 March 2004, at

http://www.coe.int/t/dg3/socialpolicies/socialcohesiondev/source/RevisedStrategy_en.pdf.

Health and well-being at work

In June 2008, the European Commission, alongside relevant social partners and stakeholders, signed the ‘European Pact

for Mental Health and Well-being’, which highlights the importance of mental health and well-being for a strong and

competitive European Union. The pact recognises that good mental health and well-being are key resources for the

European Union to meet the objectives of the Lisbon strategy, promote growth and jobs, achieve social cohesion and

make considerable gains towards sustainable development. One of the five main priority areas of the pact is ‘Mental

health in workplace settings’, which acknowledges and underpins the important role that companies have in promoting

and enabling well-being at work and in creating a more competitive and productive Europe.

Report outline

This report addresses all the aspects mentioned of the relationship between work, health and well-being. It should be

read as a complement to the fifth EWCS overview report (Eurofound, 2012a).

Chapter 1 sets the stage with a statistical portrait about health and various measures of well-being across Europe.

Chapter 2 is ancillary to the study of the connections between risk factors, well-being and health that will be delivered

in the rest of the report. The aim is to take the risk factors that are already available in the EWCS and add a set of

indicators of the psychosocial work environment, which has been identified as one of the most important risk factors in

contemporary and future society, and has a relation with health and well-being that is mainly – but not exclusively –

revealed in a work-related stress condition.

Chapter 3 is devoted to the relation between work and well-being, mainly based on several quality of work and

employment indicators that can be computed from the EWCS questionnaire.

Chapter 4 explores in a multivariate context the relationships between several physical and mental health conditions on

the one side, and risk factors on the other, with particular attention to the psychosocial work environment.

Chapter 5 is devoted to the other direction of causality. Poor health status has a direct impact on work in terms of sickness

absence. Many factors may however mediate this relationship, lengthening the absence (perhaps due to opportunistic

behaviours and/or to a poor fit of the job with individual needs and skills) or shortening it (perhaps due to work

environments that better accommodate workers’ needs). At the limit, there is the flip side of work absence, namely

presenteeism when employees go to work despite being sick, a behaviour which may entail short- to long-term costs for

both employers and employees.

Well-being at work

Cross-national comparisons of well-being have long been hampered by the absence of adequate data (Diener and

Biswas-Diener, 2002; Bracke et al, 2008). In the last years the situation for Europe improved, thanks to Eurofound’s

European Quality of Life Survey (EQLS), the European Social Survey (ESS) and, since its last edition, Eurofound’s

European Working Conditions Survey (EWCS). All these included questions on several aspects of well-being as well as

on many of its determinants in the domains of employment, economic resources, family life, community life, health,

housing and the local environment.

Published findings from the EQLS generally show striking disparities across European nations (Eurofound, 2012c): the

average level of subjective well-being is higher in the EU15 (and highest in the Nordic countries), intermediate in the

new Member States and lower in accession countries. This geography is similar when different indicators are used, such

as emotional well-being, life satisfaction and happiness. Since the main focus in this report will be on work, it is useful

to derive from the EQLS, as a background, an indication on how workers perform compared to individuals with a

different economic status (Figure 1). This shows that employed people have the second highest level of well-being after

people in education. Both among men and women, those unable to work due to long-term illness or disability have the

lowest level of well-being. Women’s well-being is worse than that of men in all groups except the long-term unemployed

and full-time homemakers.

Figure 1: Subjective well-being (WHO-5 mental well-being index), by employment status and gender, EU27, 2012 (%)

Source: EQLS 2011

The fifth EWCS, with its focus on a broad range of work-related features not dealt with by the other surveys, gives the

opportunity to investigate the interplay between working conditions and individual well-being as measured by a

multidimensional indicator, the World Health Organization’s WHO-5 index on psychological well-being, and a single

question about an individual’s overall judgement of their work domain measured by job satisfaction.

Health and well-being across Europe

66 62

62 60 60

Employed or

self-employed Unemployed

< 12 months Unemployed

>= 12 months Unable to work

due to illness or

disability

Full-time

homemaker In education Retired

Men Women

Psychological well-being: WHO-5 index

The mental well-being index (WHO-5) was originally proposed by the World Health Organization as a measure of

emotional and psychological well-being and a screener for depression (WHO, 1990). It is a short questionnaire covering

five positively worded items, related to positive mood (good spirits, relaxation), vitality (being active and waking up

fresh and rested) and general interests (being interested in things), all experienced over the previous two weeks. The

index range is 0–100 and a higher score means better well-being.5The main facts about the subpopulation of workers

basically confirm those reported for the whole population in previous European surveys (see Figure 2). The outcome by

geographic location is still the same: Nordic and continental countries have higher well-being levels, the eastern nations

rank in the intermediate positions and non-European nations present the lowest levels. Kosovo, Malta and the former

Yugoslav Republic of Macedonia exceptionally stand out from the averages of their reference groups. Above all, Kosovo

ranks first, having a level higher than the traditional winners of this unusual competition. If the point made in Veenhoven

(2000) is accepted, that political factors and personal freedom are important drivers of happiness, the more likely

explanation of this may be connected to the very recent past: the declaration of independence in 2008 after decades of

conflict is likely to be the strongest determinant of their very high score.

Figure 2 also shows the prevalence of very low levels of WHO-5-measured mental well-being. The WHO-5 index

suggests that a score of 52 or below is indicative of poor well-being and is an indication for testing for depression under

other specific scales, such as the ICD-10. A score of 28 or below indicates likely depression and warrants further

assessment to confirm it. Many countries paired to a below-average WHO-5 score have a worryingly high share of

workers below the suggested thresholds. Kosovo and the former Yugoslav Republic of Macedonia, in spite of a very high

average score, have high shares of people below the thresholds, too.

Figure 2: Subjective well-being and its critical values (WHO-5 index), by country (%)

Source: EWCS 2010

Health and well-being at work

5Each of the five items is rated on a 6-point Likert scale from 0 (= at no time) to 5 (= all of the time). Adding the five scores and

multiplying by 4 creates an index ranging from 0 (worst thinkable well-being) to 100 (best thinkable well-being)

(www.who-5.org).

66.33

LT AL TR CZ LV HU HR IT SK BG RO SI PT CY PL EE ME EL AT FR UK DE FI MK BE LU MT NL SE NO ES DK IE XK EU27

WHO-5 % WHO-5<=28 % WHO-5<=52

Health and well-being at work

Available literature says that women tend to report higher happiness but worse scores on mental health assessment scales

(Alesina et al, 2004; Eurofound data), although a few studies report no gender differences (for instance, Louis and Zhao,

2002). Among workers, the EWCS confirms a gender gap in WHO-5 (2.4 points in the EU27, see Figure 3) and also

striking country disparities.

Figure 3: Average gender gap in well-being, by country

Source: EWCS 2010

The gap is visible across the different occupations and sectors, with some exceptions (service and sales workers and in

the transportation sector) (see Figures 4 and 5).

2.43

-4

-2

ME SK AT TR AL FI PL XK ES IT MT NL LU IE DE CZ SE DK HR FR BE EL SI NO EE HU LV CY LT UK BG RO MK PT EU27

Figure 4: Average well-being by occupation (ISCO-08-1) and gender (%)

Source: EWCS 2010

Figure 5: Average well-being by sector (NACE Rev.2) and gender (%)

Source: EWCS 2010

Health and well-being at work

Managers Professionals Technicians

and associate

professionals

Clerical

support

workers

Service and

sales workers Skilled

agricultural,

forestry and

fisheries

Craft and

trades

workers

Plant and

machine

operators,

and

assemblers

Elementary

occupations

Women Men

Agriculture Industry Construction Wholesale,

retail, food and

accommodation

Transport Financial

services Public

administration

and defence

Education Health Other services

Women Men

Health and well-being at work

Job satisfaction

Job satisfaction is explored here as a single-item worker’s well-being indicator. It is a relevant item, since it is a predictor

of overall well-being and individual behaviours such as resigning from a job, productivity and absenteeism (Clark,

2009).

The country breakdown presented in Figure 6 shows a general pattern very similar to that of WHO-5 and confirms the

findings of previous waves of the EWCS. The Nordic countries on average show high levels of job satisfaction, with

Denmark and the UK recording the highest levels. The difference between the old and new Member States (those that

joined the EU after 2004) is remarkable, even if less striking than in 2005. The share of satisfied or very satisfied workers

in most EU15 Member States is above the EU27 average, while the share in most of the new Member States is much

lower. However, some exceptions can be highlighted such as the EU15 southern countries, which have a lower relative

score compared to the average (Italy, Spain, Greece and France). By contrast, three of the new Member States (Cyprus,

Malta and Poland) are above the EU average.

Figure 6: Job satisfaction by country (%)

Source: EWCS 2010

Looking at demographics, the percentage of workers reporting themselves as satisfied or very satisfied slightly increases

with age: older people seem more satisfied than younger people (Figure 7). Interestingly, the gender gap is reversed when

looking at WHO-5, where women are consistently more satisfied than men in all age groups.

As with the WHO-5, a positive association exists between level of qualification and job satisfaction (Figure 8):

professions such as agricultural worker and plant and machine operator suffer from a lower level of satisfaction. Among

managers, clerical, service and elementary workers, women are more satisfied than men.

100

AL TR MK EL XK ME LT HR HU LV SI BG RO EE FR IT CZ SK ES EU

27 PL PT MT CY SE LU DE FI BE IE AT NO NL UK DK

Very satisfied Satisfied Not very satisfied Not at all satisfied

Figure 7: Workers who are satisfied or very satisfied with their working conditions, by age and gender, 2010 (%)

Source: EWCS 2010

Figure 8: Workers who are satisfied or very satisfied with their working conditions, by occupation and gender, 2010 (%)

Source: EWCS 2010

Health and well-being at work

Under 35 35–49 50+

Women Men

100

Managers Professionals Technicians and

associate

professionals

Clerical support

workers Service and sales

workers Skilled

agricultural,

forestry and

fisheries

Craft and related

trades workers Plant and

machine

operators, and

assemblers

Elementary

occupations

Women Men

Health and well-being at work

Workers employed in education, financial services and the public sectors are those more satisfied with their working

conditions (see Figure 9). A large gap in favour of women is found in the transportation and construction sectors.

Nonetheless, such a big difference is easily explained, since in these two male-dominated sectors women are mainly

employed as clerical support staff and thus are less exposed to both physical and psychosocial risks.

Figure 9: Workers who are satisfied or very satisfied with their working conditions, by sector and gender, 2010 (%)

Source: EWCS 2010

The relationship between job satisfaction and overall well-being has already been investigated in a number of studies,

which proposed different and competing theoretical approaches. The EWCS 2010 results seem coherent with the so-

called spill-over theory (Spector, 1997), which suggests that feelings in specific dimensions of life affect feelings in other

domains as well. Therefore, having a high level of job satisfaction is a strong predictor of overall well-being as measured

by the WHO-5 (Figure 10). The differences in WHO-5 scores between unsatisfied and satisfied workers is striking: the

gap is about 30 points for both genders. In the most synthetic way, this shows how strongly correlated the two measures

are and how important work is for the well-being of individuals. In Figure 10, the variability of well-being at different

levels of job satisfaction is also reported.

100

Agriculture Industry Construction Wholesale,

retail, food and

accommodation

Transport Financial

services Public

administration

and defence

Education Health Other services

Women Men

Figure 10: Average level and variability of well-being by job satisfaction level and gender

Note: CV is the WHO-5 variability as measured by the coefficient of variation.

Source: EWCS 2010

General health conditions and main job features

To arrive at a first impression of European workers’ general health, the study relies on two questions. One prompts a

self-assessment of general health conditions (Q68), and in particular the analysis looks at the distribution of those

reporting ‘bad’ and ‘very bad’ health.

A subsequent question (Q69) asks respondents to mention which specific health problems the person has experienced in

the previous year, listing 13 possibilities (including ‘injury’) plus a residual ‘other’. Here the focus is on the distribution

of individuals mentioning more than two health problems. The threshold is justified because among the possibilities

there are health complaints known to be very common: for instance, 46% of workers report ‘backache’, and 43% report

‘muscular pains in upper limbs’. Hence a threshold of just one mentioned problem would have made it difficult to

distinguish between serious conditions and milder ones.

The distribution across occupations of the two health measures is displayed in Figure 11. Bad general health is mentioned

by 2.5% of European workers, while 47% mention more than two health problems. In this second case women are more

likely to do so than men (50% compared to 45%). Clearly the two dimensions signal different perceived health

conditions, a milder and more common one in the case of more than two health problems, a more severe and uncommon

one in the case of those reporting ‘bad’ and ‘very bad’ health.

The focus on job characteristics such as occupation and type of employment can shed light on the possible correlation

between health and working conditions. As expected, manual workers are more affected by health problems (Figure 11);

skilled agricultural workers stand out as the worst affected in this regard, closely followed by those in elementary

Health and well-being at work

45.5

76.1

44.6

73.6

100

Not at all

satisfied Not very

satisfied Satisfied Very satisfied

CV (0-100)

WHO-5 (0-100)

Mean WHO-5 Men Mean WHO-5 Women CV Men CV Women

Health and well-being at work

occupations. Women are more likely to report more than two health problems compared to men in all occupations, with

the exception of clerical jobs. However, bad health is reported more evenly across genders. There might be a

psychological dimension here, where the perception of several illnesses is not coupled with the perception of bad health.

However, Figure 12 shows a polarisation in the perception of women: those reporting bad health are either reporting very

few health problems or a huge number of them, compared to men.

Figure 11: Prevalence of individuals with bad health status among all workers, by occupation and gender (%)

Source: EWCS 2010

1.4 2.3 1.6 1.6 1.4 2.3 1.8 1.5 2.3

8.1 12.0

2.3 4.2 2.4 4.8 4.0 4.1

Men

Women

Men

Women

Men

Women

Men

Women

Men

Women

Men

Women

Men

Women

Men

Women

Men

Women

Managers Professionals Technicians Clerical Service Skilled

agricultural Craft Plant and

machine

operators

Elementary

More than two health problems Bad and very bad health

Figure 12: Distribution of workers reporting ‘bad’ and ‘very bad’ health, by number of mentioned health problems and

gender (%)

Source: EWCS 2010

The type of contract a worker has is a relevant dimension in the shaping of general health conditions. As Figure 13

shows, there is a clear divide between those working within a firm, and those working without a contract or as self-

employed without employees. ‘Several health problems’ and ‘bad health’ are more common in the second group of

workers. The gender gap in the mention of several health problems is present and quite constant in all types of

employment. By contrast, bad health is mentioned more frequently by men working without a contract or as self-

employed without employees.

Health and well-being at work

012345678910111213

Men Women

Health and well-being at work

Figure 13: Type of employment of workers with bad self-reported health, by employment type and gender (%)

Source: EWCS 2010

Self-employed workers without employees (about 13% of European workers) present several characteristics. The

category is male dominated (65% of the total), mostly made up of agricultural and craft workers (16% and 13%

respectively). Women are more spread across occupations, but the higher prevalence is in services and agriculture (8%

and 7%).

In almost all occupations, values of bad health indicators are higher among self-employed workers (Figure 14) than

among the whole population of workers (Figure 11), pointing to working conditions where safety regulations are better

monitored and enforced, for those working inside firms with other employees.

1.8 2.3 1.8 1.6 3.3 2.8

5.0 4.6

1.8 2.6

Men Women Men Women Men Women Men Women Men Women

Permanent Temporary No contract Self-employed with no

employees Self-employed with employees

More than two health problems Bad and very bad health

Figure 14: Prevalence of individuals with bad health status among the self-employed without employees, by occupation

and gender (%)

Source: EWCS 2010

Health itself is a strong determinant of overall well-being, maybe one of the most important. As shown in Figure 15, the

psychological well-being of individuals is strongly correlated with how they evaluate their own health status. The 40-

percentage point difference between those who are in very good and very bad health is sizeable. Those who consider

their health either bad or very bad show worryingly low levels of psychological well-being, far below the threshold of

52 suggested by the WHO-5 as indicative of very poor well-being and depression.

Health and well-being at work

3.0 0.8 1.0 0.3 1.6 1.4 0.0 0.0 2.8 3.1

10.7

15.5

3.0 3.0 3.0 0.0

11.3

1.6

Men Women Men Women Men Women Men Women Men Women Men Women Men Women Men Women Men Women

Managers Professionals Technicians Clerical Service Skilled

agricultural Craft Plant and

machine

operators

Elementary

More than two health problems Bad and very bad health

Health and well-being at work

Figure 15: Average well-being by health status (%)

Source: EWCS 2010

Work sustainability

The support for active ageing, including appropriate working conditions for elders and non-elders, adequate incentives

to work and discouragement of early retirement, is at the centre of most European employment policy. The question

about respondents’ ability to do the same job at the age of 60 (Q75) is an immediate source of information, which gives

a valuable insight on the theme.

Differences in responses across Europe are remarkable (Figure 16). Those who report themselves not able or not willing

to do the same job at the age of 60 go from a very low figure of 20% in the Netherlands to 70% and 80% in Slovenia

and Turkey respectively.

Very good Good Fair Bad Very bad

Women Men

Figure 16: Prevalence of workers unable or unwilling to do the same job at age 60, by country (%)

Source: EWCS 2010

Work sustainability is quite similar among genders, but higher among older workers than younger ones. This is quite a

standard result, due a healthy worker effect and to the fact that how workers imagine they are likely to be at the age of

60 becomes clearer as they approach that age. Self-employed workers report sustainable work less frequently than

employed workers.

Figure 17 shows the strong correlation between work sustainability and health and well-being. Among workers who

declare very good or good health, 60% are confident in their ability to do the same job at the age of 60, while the

proportion is significantly lower among those in bad or very bad health.

Health and well-being at work

NL DE IE NO DK UK SE FI CY IT EE EU

27 LV DE AT RO LT CZ MT SK LU BG AL PL ES HU HR FR EL ME XK PT MK SI TR

Health and well-being at work

Figure 17: Distribution of perceived ability to do same job at age 60, by health status and job satisfaction level (%)

Source: EWCS 2010

62 57

18%

21 28

47%

17 15 20 21

35%

100

Very good Good Fair Bad Very bad Very satisfied Satisfied Not very

satisfied Not at all

satisfied

Health Job satisfaction

Yes, I think so No, I do not think so I wouldn't want to

This chapter is devoted to the so-called psychosocial work environment, a risk factor for the health and well-being of

workers that the European Agency for Safety and Health at Work (EU-OSHA) has identified as one of the most important

in contemporary and future society, suggesting that it deserves ‘top priority’ among other work issues (EU-OSHA,

2000a). After introducing the topic, it is possible to build a set of indicators for assessing the topic using the many

questions present in the EWCS, with the double aim of providing original empirical evidence on this important risk

factor, and to be able to refer back to it throughout the rest of the report in the investigation of the determinants of health

and well-being.

Work-related stress and the psychosocial work environment

Work-related stress is a pattern of reactions occurring when workers experience prolonged exposure to work demands

that are not matched to their knowledge, skills or abilities, and which challenge their ability to cope (EU-OSHA, 2000b;

Eurofound, 2005). The public health relevance of the issue is well established: research has proven that stress at work is

associated with a number of physical and psychological negative effects at individual level such as cardiovascular

diseases, musculoskeletal diseases, immunological problems and mental health problems (anxiety and depression

disorders) (EU-OSHA, 2009).

Several models have been created to illustrate these links. One is presented below in Figure 18, which summarises the

stress process identifying the causes of stress, short-term stress reactions, long-term consequences of stress and

individual characteristics, as well as their interaction. The model highlights also the importance of individual

characteristics that determine how workers perceive their working environment and what is expected of them. Generally,

it is accepted that most individuals are well adapted to cope with short-term exposure to pressure, which can be

considered as positive, but have greater difficulty in coping with prolonged exposure to intensive pressure (ETUC et al,

2004).

Figure 18: Model of causes and consequences of work-related stress

Source: Eurofound, 2005; EU-OSHA 2009 (Adapted from Kompier and Marcelissen, 1990)

Psychosocial work environment

RISKS FOR WORK-RELATED

STRESS:

- High workload

- Low control

- Low support

- Job insecurity

- Long working hours

- etc.

STRESS REACTIONS:

- Physiological

- Behavioural (productivity,

reporting sick, smoking, etc.)

- Emotional

- Cognitive

LONG-TERM CONSEQUENCES:

For the worker:

- High blood pressure

- Affective disorders

- Musculoskeletal disorders, etc.

For companies:

- Increased absenteeism

- Tardiness

- Increased turnover

- Reduced performance and

productivity

- etc.

INDIVIDUAL CHARACTERISTICS:

- Gender

- Age

- Education

- Competitiveness

- Overcommitment

- Self-confidence

- etc.

Regarding risks factors, several classifications have been offered, the majority of which include organisational factors

such as job intensity and workload, job control and social support, as suggested by the Karasek ‘Demand-Control’ model

(Karasek, 1979). This model states that the greatest risks to physical and mental health are faced by workers who have

to cope with high psychological workload, demands or pressures, combined with low control or decision latitude in

meeting those demands (Karasek, 1979) and lack of social support (Karasek and Theorell, 1990). By looking

simultaneously at demands and control, it is possible to classify jobs into four categories:

nactive jobs (high demands and high control);

nhigh-strain jobs (high demands and low control);

nlow-strain jobs (low demands and high control);

npassive jobs (low demands and low control).

The psychosocial work environment is a more general construct which, in addition to job demand and control, entails a

wide set of items related to work organisation and job content, type of production and tasks, interpersonal relations and

so on, covering a large range of potential stressors. A large proportion of employees in Europe report being exposed to

many of these psychosocial stressors at work, and the consequences are believed to be very significant for workers,

workplaces and society (Kristensen et al, 2005).

One of the most comprehensive instruments for the assessment of the psychosocial work environment is the Copenhagen

Psychosocial Questionnaire (COPSOQ), which is used by a rapidly increasing number of researchers and work

environment professionals (Rugulies et al, 2010). The questionnaire is theory-based, but not attached to one specific

theory. In its longest version there are more than 144 items exploring 20 workplace psychosocial dimensions based on

the most influential theories on work-related stress. These include theories centred on the individual perception of the

work environment, the organisational climate and the interpersonal relationships at work, such as the Michigan

organisational stress model (Caplan et al, 1975); or the ‘effort–reward imbalance’ model (Siegriest, 1996); theories more

focused on the assessment of objective job characteristics, rather than perceived stressors, such as the demand-control

model (Karasek, 1979); or the action-theory model (Frese and Zapf, 1994) – following the idea that ‘there should not be

any significant “white spots” in the picture painted’ (Kristensen et al, 2005). In general, COPSOQ scales on the

psychosocial work environment have demonstrated good internal consistency (Kristensen et al, 2005), high validity

(Bjorner and Pejtersen, 2010) and high test-retest reliability (Thorsen and Bjorner, 2010).

The indices

The richness of the EWCS allowed the researchers to reproduce as closely as possible the scales included in the

COPSOQ questionnaire, with a few exceptions. For example, for the sake of comparability with previous results, the

demand and control dimensions of the Karasek model have been modelled using as the reference the Job Content

Questionnaire (JCQ).6

Health and well-being at work

6See http://www.jcqcenter.org. The best-known scales (decision latitude, psychological demands and social support) are used to

measure the high-demand/low-control/low-support model of job strain development. There are several versions of it, but the

recommended one is the standard 49 questions of the JCQ.

Health and well-being at work

Based on the scales developed in the COPSOQ II long version (Pejtersen et al, 2010), 28 questions present in the EWCS

questionnaire were used to reproduce 12 indices organised in four domains: demands; work organisation and work

content; interpersonal relations and leadership; work-individual interface.7Selected items were first normalised so that

they have a 0–1 range, then grouped in a summative index with equal weights for the items, and normalised to 0–100.

A brief description of them follows here (further details are in Table 1).

Demands at work

Four different scales of demand were constructed based on six questions, with three questions assessing exposure to

psychological demand, as defined by the demand-control model, and one item each for cognitive demands, emotional

demands and demands for hiding emotions.

Psychological demand corresponds to Karasek’s definition of ‘high workload and conflicting demands’ and has been

constructed using three items assessing work intensity and one item on role ambiguity, given the absence in the EWCS

of specific information on conflicting demands.

Cognitive and emotional demands were assessed in the COPSOQ by four and three questions, respectively (Table 2).

For cognitive demands, unfortunately, none of the questions in the EWCS resembled those used in the COPSOQ, so the

analysis had to rely for its construction on a single item asking whether their own job involved performing complex

tasks. This question captured only in part the wider dimension of cognitive demand, involving ‘attention, perception,

memory, decision-making and/or volitional motor action’ (Maki and McIlory, 2007).

Both emotional demands and demand for hiding emotions were measured by means of a single question, which was in

both cases very similar to one of the questions included in the corresponding COPSOQ scales. For emotional demands,

the high item correlations with the total scale (0.65–0.80) would indicate that the item employed is probably a good

proxy of the total scale, whereas this may not be true for demand for hiding emotions, for which the item correlation

with the total scale was much lower (0.31–0.45) (Kristensen et al, 2005). However, the item used to measure demand

for hiding emotions (‘your job requires that you hide your feelings’) appears central in the assessment of this dimension,

compared to the other two in the COPSOQ scale (see Table 1).

The constructs of ‘emotional demands’ and ‘demand for hiding emotions’ are based on work by Hochschild (1983), who

defined emotional labour as the requirement for workers employed in jobs implying relationships with clients,

customers, pupils or patients to display certain emotional expressions as part of the tasks. Therefore, the concept of

emotional labour refers to the quality of interactions between employees and clients. Later research has better

characterised this psychosocial dimension, recognising four different dimensions: frequency of appropriate emotional

display, attentiveness to required display rules, variety of emotions to be displayed and emotional dissonance (Morris

and Feldman, 1996). In particular, emotional dissonance, defined as the conflict between genuinely felt emotions and

emotions required to be displayed in organisations (Middleton, 1989), is believed to be the main factor responsible for

the adverse health events reported in many studies.

High levels of cognitive demand have been found prospectively associated with an increased risk of sickness absence

(Rugulies et al, 2010) and of non-fatal occupational injury (Nakata et al, 2006); furthermore, in a Danish survey on

hospital workers, an inverse association was observed between exposure to high cognitive demands and mental health

score (Aust et al, 2007).

7Indeed, in spite of efforts in this study, a limited number of questions could be used compared to the COPSOQ, which means the

indicators cannot be interpreted as a full implementation of COPSOQ scales.

High emotional demand has also been found to be associated with low job satisfaction (Rutter and Fielding, 1988;

Martinez-Inigo et al, 2007) and with different health outcomes, including sickness absence (Clausen et al, 2012;

Rugulies et al, 2010), exhaustion (Bakker et al, 2004; Morris and Feldman, 1997), burnout (Zapf et al, 1999), depression

(Muntaner et al, 2006), fatigue and psychological distress (Bültmann et al, 2002).

Work organisation and job contents

This dimension corresponds to the Karasek’s ‘job control’ concept and the two scales created try to approximate the

control subscales of the JCQ. The first scale measures skills discretion and development. It is assessed by four items

measuring whether the employee’s main job entailed ‘solving unforeseen problems on your own’, ‘learning new things’

or ‘monotonous and repetitive tasks’. The second scale (decision authority) entails the possibility to ‘influence the order

of tasks, method and speed of work’, ‘take a break when you wish’, ‘influence decisions, work target definition’, etc.

Interpersonal relations and leadership

Here indicators are included for support from colleagues and supervisors, social climate and job rewards.

Social support from co-workers is measured through one question in the EWCS (‘your co-workers help and support

you’). This item is expected to represent the core aspect of this dimension and is similar to the most salient of the three

used in the COPSOQ.

Social support from supervisors is investigated here through four items: one item investigating this dimension in a direct

way, as well as three statements on the quality of leadership. In the COPSOQ the latter is assessed as a separate

psychosocial factor.

Research on social support at work was fostered on the one hand from early observations of an increase in total

mortality, especially from cardiovascular diseases, among less socially integrated people (Berkman and Syme, 1979;

Kaplan et al, 1988; House et al, 1988), and on the other hand from studies that found higher risks of different health

outcomes associated with inadequate workplace social support and to social isolation (Haynes and Feinleib, 1980; Rose

et al, 1979; Medaile et al, 1973). Following these reports and other results suggesting that workplace social support has

a moderating effect on the impact of stressful working conditions (LaRocco et al, 1980; Karasek et al, 1982), Johnson

and Hall (1988) demonstrated, in a sample of the Swedish general population, that the risk of cardiovascular diseases

associated with exposure to high strain, as defined in the Karasek model, was strongly enhanced by low social support.

Although the type of social support at work considered by Johnson and Hall (1988) was limited to co-workers’ support,

in most subsequent studies the support of supervisors and of co-workers has been investigated separately (LaRocco et

al, 1980).

Reviews of studies published in the last two decades on the health effects of social support have mostly found that

workers who lack social support are at significantly increased risk of coronary heart disease (Eller et al, 2009), common

mental disorders (Stansfeld and Candy, 2006), depression (Bonde, 2008; Netterstrøm et al, 2008) and neck pain (Da

Costa and Vieira, 2010). The buffering role of social support on the effect of job strain has also been confirmed by these

reviews, at least in part.

Another concept akin to supervisor support, that is quality of leadership, has been found to have a protective effect on

sickness absence and retirement for disability (Kuoppala et al, 2008).

The quality of the social community at work is a construct partially overlapping with that of social support from co-

workers as it covers the opportunity for pleasant and meaningful contacts, for feeling part of a greater social system, and

for getting and giving strategic information about one’s own performance and informal power position in the workplace

Health and well-being at work

(Schabracq, 2003). This dimension was assessed in the EWCS by two items addressing the social climate at work, plus

an item on engagement of the employee in the organisation.

The psychosocial dimension of reward belongs to the effort–reward model, which stems from the social exchange theory

(Cosmides and Tooby, 1992) and proposes as a stress determinant the imbalance between high job demands, in terms of

workload and commitment, and low rewards, the latter distributed through three remuneration systems: money, esteem

and career development (including job security). According to the model, the prolonged experience of a lack of

reciprocity, in terms of high expenses and low gains, would produce negative health effects (Siegriest, 1996). This model

also postulates that ‘overcommitment’, a term referring to a personal characteristic implying excessive work engagement

and the need for approval from others, would enhance susceptibility of workers with these characteristics to the effect

of the imbalance.

The reward dimension in the COPSOQ is assessed through three questions focusing on ‘social justice’ or ‘esteem’,

whereas, as mentioned above, in the effort–reward model (Siegrist et al, 2004), rewards are distributed by three

transmitter systems: money, esteem and career opportunities. This psychosocial dimension was assessed using the

Siegrist model as a reference, but since questions on esteem are not available in the EWCS, the reward dimension is

based only on money and career opportunities.

An effort–reward score was built dividing the overall score of psychological job demand, defined according to the

Karasek model, by the score of the reward dimension.

Work–individual interface dimension

The last two scales permit the investigation of how well organisation of work fits with individual needs and

commitments. This dimension is explored through an indicator of work–life balance and job security.

The dimension of work–family conflict refers to a condition in which work and family domains interfere so much with

each other that one exerts a negative effect on the other (Greenhaus and Beutell, 1985). According to the US National

Institute for Occupational Safety and Heath, work–family conflict is one of the 10 most important stress factors at work

(Kelloway et al, 1999). The two prevalent theories on which the work–family conflict dimension is based are the

spillover or generalisation hypothesis, which postulates the carry-over of alienation from work into alienation from non-

work (Kabanoff and O’Brien, 1980), and the role strain hypothesis, which assumes that managing multiple roles is

difficult and inevitably creates ‘strain’. According to researchers belonging to the latter school, work–family conflict is

‘a form of inter role conflict in which the role pressures from the work and family domains are mutually incompatible

in some respect’ (Greenhaus and Beutell, 1985).

Several studies have found work–family conflict associated with different health outcomes, including poor self-rated

general health (Frone et al, 1996; Hammer et al, 2004; Hämmig et al, 2009), burnout (Netemeyer et al, 1996; Kinnunen

and Mauno, 1998), psychological distress or low psychological well-being (O’Driscoll et al, 1992; Parasuraman et al,

1996).

In the COPSOQ, work–family conflict is measured through a set of four questions assessing either directly the presence

of such a conflict (‘do you often feel a conflict between your work and your private life …?’) or its negative effects in

terms of energy and time subtracted from non-work activities. In the EWCS questionnaire, work–family conflict is

assessed through one direct question about the possibility of easily reconciling work and family life, and three more

indirect questions that explore this dimension mainly through the presence of working time constraints.

Job security was assessed through a single question asking workers whether it was possible that they might lose their job

in the next six months. The narrow time limit indicated in this statement is expected to have increased the specificity of

the assessment of job security, as supported by the fact that less than 20% of workers reported agreement or strong

agreement that it was possible they might lose their job. Nonetheless, the information collected through responses to this

question differs from that available in the COPSOQ, which also investigates workers’ perception of the threat of job loss

and their future employability.

The diffusion of flexible employment in most developed countries in recent decades has determined an overall decrease

of job security in working populations, in particular in Europe (Sverke et al, 2000). Job insecurity is based on individual

perceptions, which may be different even among people in the same employment situation, and is related to the threat

of involuntary job loss (Greenhalgh and Rosenblatt, 1984). Therefore, subsequent employability of the workers may be

an effect modifier of job insecurity, because the threat is expected to be greater for those characterised by lower

possibilities of re-employment. Workers with a low socioeconomic status are also expected to be more susceptible to the

effect of job insecurity, given that they generally have a lower amount of savings and assets to compensate job loss, as

well as lower skills and education, which in turn would reduce their future employability (Gallie et al, 1998; Artazcoz

et al, 2005). Research on the health effects of job insecurity has grown fast after it was demonstrated, within the

Whitehall II study, that workers threatened with privatisation of their department were at higher risk of developing minor

psychiatric morbidity (Ferrie et al, 1995). Since then, job insecurity has been confirmed as being consistently associated

with poor mental health (Rugulies et al, 2006; Swaen et al, 2004; D’Souza et al, 2003), as well as with poor self-rated

physical health (Ferrie et al, 1998; Cheng et al, 2005), ischaemic heart disease (Lee et al, 2004) and lower sickness

absence (Kivimäki et al, 2007).

Table 1: Scheme for the implementation of the psychosocial work environment indices

Health and well-being at work

Scale Reference questions used Questions used (EWCS)

Domain: Demands at work

Psychological job

demands

JCQ - abbreviated list of the recommended Format (49 q.):

l‘work fast’

l‘work hard’

l‘no excessive work’

l‘enough time’

l‘conflicting demands’

lQ45A and Q45B - high speed and tight deadline

lQ51G - have enough time

lQ51K - know what is expected of you at work

Cognitive

demands

COPSOQ II long version:

lDo you have to keep your eye on lots of things while you work?

lDoes your work require that you remember a lot of things?

lDoes your work demand that you are good at coming up with

new ideas?

lDoes your work require you to make difficult decisions?

lQ49E - Complex tasks

Emotional

demands

COPSOQ II long version:

lDoes your work put you in emotionally disturbing situations?

lDo you have to relate to other people’s personal problems as part

of your work?

lIs your work emotionally demanding?

lDo you get emotionally involved in your work?

lQ51M – You get emotionally involved in your work

Demands for

hiding emotions

COPSOQ II long version:

lAre you required to treat everyone equally, even if you do not

feel like it?

lDoes your work require that you hide your feelings?

lAre you required to be kind and open towards everyone –

regardless of how they behave towards you?

lQ51P - Your job requires that you hide your feelings

Health and well-being at work

Scale Reference questions used Questions used (EWCS)

Domain: Work organisation and job contents

Skill discretion

and development

JCQ - abbreviated list of the recommended Format (49 q.):

l‘learn new things’

l‘repetitive work’

l‘requires creative’

l‘high skill level’

l‘variety’

l‘develop own abilities’

lQ49C - solving unforeseen problems

lQ49D - monotonous tasks

lQ49F - learning new things

lQ44(AorB) - short repetitive tasks of less than

1/10 minutes

Decision authority COPSOQ II long version:

lDo you have a large degree of influence concerning your work?

lDo you have a say in choosing who you work with?

lCan you influence the amount of work assigned to you?

lDo you have any influence on what you do at work?

lQ50(A–C) - able to choose order/methods/speed of

work

lQ51C - you are consulted before targets for your

work are set

lQ51E - you have to say in the choice of your

working partners

lQ51I - apply own ideas in your work

lQ51O - you can influence decisions

Domain: Interpersonal relations and leadership

Social support

from colleagues

COPSOQ II long version:

lHow often do you get help and support from your colleagues?

lHow often are your colleagues willing to listen to your problems

at work?

lHow often do your colleagues talk with you about how well you

carry out your work?

lQ51A - your colleagues help and support you

Social support

from supervisors

JCQ - abbreviated list of the recommended Format (49 q.):

l‘supervisor is concerned’

l‘supervisor pays attention’

l‘hostile supervisor’

l‘helpful supervisor’

l‘supervisor good organiser’

lQ58(A–E) - leadership quality variables

lQ51B - your manager support and helps you

Social community COPSOQ II long version:

lIs there a good atmosphere between you and your colleagues?

lIs there good co-operation between your colleagues at work?

lDo you feel part of a community at your place of work?

lQ77D - I feel at home

lQ77E - I have very good friends

lQ77G - The organisation motivates me

Job rewards COPSOQ II long version:

lIs your work recognised and appreciated by the management?

lDoes the management at your workplace respect you?

lAre you treated fairly at your workplace?

lQ77B - I am well paid

lQ77C - Possibility for career advancement

Domain: Work–individual interface

Work–life balance COPSOQ II long version:

lDo you often feel a conflict between your work and your private

life…?

lDo you feel that your work drains too much energy...?

lDo you feel that your work takes too much time...?

lDo your friends and family tell you that you work too much?

lQ39 - chose working time arrangements set

lQ41 - working hours fit in with your family

commitments

lQ42 - worked in your free time

lQ43 - difficulty in taking a couple of hour off

Job insecurity COPSOQ II long version:

lAre you worried about becoming unemployed?

lAre you worried about new technology making you redundant?

lAre you worried about it being difficult for you to find another

job if you became unemployed?

lAre you worried about being transferred to another job against

your will?

lQ77A - I might lose my job

Application of the demand-control model to EWCS data

First, the Karasek model is replicated using the composite indicators of ‘autonomy’ and ‘intensity of work’,

corresponding to Karasek’s concepts of ‘job control’ and ‘job demands’, to form an initial picture of the matter.

In general terms, the classification of countries according to this model is confirmative of previous findings from

Eurofound’s European Working Conditions Survey (Eurofound, 2007c). Malta and Nordic countries, particularly

Finland, Norway and Sweden, are in the active jobs group, which is identified as the best organisation leading to high

performance without negative consequences for working conditions, since greater demands on the worker are

counterbalanced by greater autonomy and control over job content, reducing the potentially detrimental impact of work

intensity. Conversely, Cyprus, Greece, Turkey and the former Yugoslav Republic of Macedonia approach most closely

the high strain group, the form of work organisation that has the most negative impact on working conditions.

Figure 19 gives a concise representation of what happened from 2005 to 2010, grouping countries according to how they

moved in the demand–control space. A very general movement towards higher levels of control emerges. There are just

very few countries where workers’ level of control has lowered in a statistically significant way. The second major

movement is towards lower levels of intensity, again with few exceptions. The two movements are often combined: in

fact, about half of the countries moved towards the low-strain group. This is somehow in contrast with the long trend

observed since 1991 (Eurofound, 2007a), which was towards higher intensity, but is consistent with the current weak

macroeconomic situation. A high proportion of workers in 2010 may fairly have worked fewer extra hours than usual;

many are under some income support measures, such as partial unemployment or temporary layoff schemes, which keep

them in employment but either with shorter working time or not working.

Figure 19: Changes in average scores of job demand and control, by country, 2005–2010

Note: Axes are fixed at 2010 mean levels. Employees only.

Source: EWCS 2005 and 2010

Health and well-being at work

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Control (autonomy)

Demand (intensity)

Low strain

Passive

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

0.00.10.20.30.40.50.60.70.8

Control (autonomy)

Demand (intensity)

Lowstrain

Passive

Acve

High strain

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

0.00.10.20.30.40.50.60.70.8

Control (autonomy)

Demand (intensity)

Low strain

Passive

Acve

High strain

1.8

2.0

2.2

2.4

2.6

2.8

3.0

3.2

3.4

3.6

3.8

0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8

Control (autonomy)

Demand (intensity)

Low strain

Passive

Acve

High strain

Health and well-being at work

Figure 20 illustrates how occupation and sector fit into the Karasek model. Unskilled occupations and sectors are

dominant in the high-strain work organisation: workers in the transportation, hotel and manufacturing sectors

particularly have to face very high demanding situations without the possibility of compensating for such stressful

conditions with a sufficient extent of control over the content of their work.8Financial and real estate workers, and to a

greater extent managers, are the only ones to fit into the active category. Professionals, those working in public

administration and services are closest to the low-strain work organisation category. Finally, the agricultural, services

and retail sectors are closest to the ‘passive work organisation model’.

Figure 20: Job demand and control average scores, by occupation and sector

Note: Employees only.

Source: EWCS 2010

Managers

Professional

Technicians

Clerical

Service

Agricultural

workers

Craft

Plant operator

Elementary

occupations

Agriculture,

hunting

Mining,

manufacturing

Electricity

Construction

Wholesale

Hotel

Transport

Financial

Real

estate

Other

services

Public administration

1.5

2.0

2.5

3.0

3.5

4.0

4.5

0.30 0.35 0.40 0.45 0.50 0.55 0.60 0.65

Job control (Autonomy)

Job demands (Intensity)

Low strain

Passive High strain

EWCS Mean

control

EWCS Mean

demands

Acve

8Looking at the changes from 2005 and 2010, a further reduction in the level of autonomy for those working in elementary

occupations and transportation was observed.

Table 2 shows which is the average psychological well-being in the four groups, considering also the social support

dimension. As predicted by the model, it is high-strain situations which lead to the lowest well-being; the same group is

the one where social support has the biggest importance.

Table 2: Average well-being (WHO-5%), by Karasek groupings (%)

Note: The Job Content Questionnaire (JCQ) centre suggests identifying the groups by the exclusion of that segment of the population

that is closest to the population mean (mid-population), by dividing both the psychological demands and the decision latitude scales

into quartiles. This method is used here, but by dividing into tertiles so as not to lose too many observations.

Source: EWCS 2010

Descriptive analysis of psychosocial work environment indices

Table 3 offers a descriptive picture of psychosocial indicators by gender and age categories. Looking at the ‘demand’

scales it is clear that the situation is split between genders: women show higher levels of emotional demands, as well as

demand for hiding emotions, while men are more exposed to cognitive and psychological demands. This is true in all

age groups, even if differences decrease with age. The most demanding situation at work concerns middle-aged male

workers, who experience the highest level in almost all the demanding features.

Table 3: Average psychosocial work environment scores, by age and gender

Note: All scales are normalised to 100 scores.

Source: EWCS 2010

The possibility of using and developing their own skills and competences at work is higher for men than women at all

ages. Both men and women experience a positive trend across age for level of autonomy, which tends to increase from

56 to 64 and from 54 to 61, respectively.

Health and well-being at work

With social support Without social support Gap (%points)

Mid-population 66.69 59.99 6.70

Low strain 70.10 66.54 3.56

Passive 69.37 62.13 7.24

Active 68.82 62.17 6.65

High strain 66.17 55.40 10.77

Total 68.34 60.37 7.97

Variable

Under 35 35–49 Over 50

Men Women Men Women Men Women

Psychological demand 34.9 30.2 33.1 29.9 28.9 26.9

Cognitive demands 59.1 48.0 62.8 53.1 59.5 53.2

Emotional demands 44.8 51.2 51.2 57.1 52.1 58.5

Demands for hiding emotions 35.8 41.3 36.1 42.5 35.5 38.7

Skill discretion 63.6 62.4 66.2 64.1 65.4 62.5

Decision authority 56.5 54.2 62.5 59.4 64.1 61.6

Support from colleagues 74.2 74.4 72.5 72.9 71.5 72.3

Support from supervisors 66.3 70.7 60.6 65.7 53.9 61.6

Social community 65.9 66.9 67.0 68.7 68.6 69.9

Job rewards 50.5 48.7 48.8 45.5 46.0 41.8

Work–life balance 64.8 65.9 64.7 66.6 68.4 69.2

Job security 67.3 67.1 70.4 71.8 73.3 73.5

Health and well-being at work

Regarding interpersonal relations at work, women report a slightly better working climate and higher support from

managers, although lower rewards, compared to men. However, the higher support from managers among women may

also be attributable to the lower proportion of women in managerial positions. There is an interesting pattern which

emerges in both genders: the satisfaction of social relationships at work increases and the rewards dimension decreases

when workers become older.

Work–family conflicts are slightly more spread among male workers than women. This may be seen as a quite

paradoxical result, but it is most likely due to the cross-sectional design of the study. It is likely that a selection process

affects the gender composition, resulting in a lower proportion of women employed, compared to men who are or are

not associated with processes of adaptive preferences that would influence the level reported. It may therefore be that

dissatisfied men are working, whereas potentially dissatisfied women occupy some other labour force status.

Psychological demand at work is highest among craft workers, plant and machine operators (Table 4). Levels of

emotional demand are higher among professionals, especially women, managers and agricultural workers. The need to

hide emotion concerns, in particular, managers, professionals and service workers.

Table 4: Average psychosocial work environment scores, by occupation and gender

Note: All scales are normalised to 100 scores. M = Male, F = Female.

Source: EWCS 2010

As expected, more highly qualified occupations correspond to higher cognitive demands and higher decision latitude and

skill discretion. It is also the case that both decision latitude and skill discretion are lower among women than men in most

occupational groups and reach worryingly low levels among female plant and machine operators (35 and 44, respectively).

In the social relationship domain, lower-skilled occupations report lower social support from colleagues and a less

satisfying working climate. As expected, support from managers is lower among managerial occupations and skilled

Variable

Manager Professional Technician Clerical Service

Skilled

agriculture Craft Plant Elementary

M F M F M F M F M F M F M F M F M F

Psychological

demand 33 32 30 28 32 32 33 29 29 26 28 28 35 36 36 42 34 29

Cognitive

demands 75 72 80 71 76 68 56 56 43 34 47 35 69 49 43 43 32 23

Emotional

demands 58 60 61 70 50 56 44 47 49 54 61 64 46 52 43 43 36 41

Demands for

hiding emotions 46 45 44 48 37 44 36 38 45 45 25 21 29 33 33 33 29 29

Skill discretion 74 74 81 79 75 72 62 63 62 58 62 56 62 52 54 44 48 40

Decision

authority 84 81 73 66 65 60 53 54 56 53 76 75 57 54 44 35 45 51

Support from

colleagues 78 79 74 76 71 74 71 72 73 74 73 69 76 71 68 70 70 63

Support from

supervisors 45 55 65 72 71 72 72 76 64 66 16 11 63 54 65 65 62 62

Social

community 75 76 71 71 69 70 64 68 66 68 67 72 67 65 62 60 60 61

Job rewards 58 57 58 52 56 50 50 49 46 42 33 28 48 40 41 35 39 36

Work–life

balance 68 68 66 64 68 68 67 70 62 65 73 74 65 70 60 63 66 70

Job security 75 74 78 77 74 74 70 69 68 67 79 80 66 66 64 60 63 63

agricultural workers. Skilled agricultural workers also have the lowest levels of rewarding aspects of their work, such as

the possibility of a career and satisfaction with salary (33 and 28, for men and women respectively).

Table 5 shows that psychological demands are generally higher among those working without a contract or with an

atypical contract. Self-employed people with employees face high psychological demands. Emotional demand affects

self-employed workers to a greater extent than employees, and mainly women.

Table 5: Average psychosocial work environment score, by type of employment and gender

Source: EWCS 2010

As might be expected, decision authority is largely higher among the self-employed, although levels of skill discretion

(such as learning new things, solving unforeseen problems) are quite similar among the self-employed and employees

with a permanent contract.

The characteristics of the social environment are very positive for the self-employed with employees; this may reflect

the different perception between employers and employees of the quality of the working climate, which is viewed as

more favourable by the former, probably because of the position of power they occupy in their companies.

In terms of the rewarding dimension of work, the worst situation is faced by employees without a contract or with an

atypical contract, and by the self-employed without employees. Employees with fixed or temporary attachment to work

also have the lowest levels of job security, as expected. The possibility of reconciling work–life commitments is higher

among self-employed, particularly those without employees, probably because of their higher control of their work

schedule, which allows a better arrangement of work and family duties. On the contrary, employees without permanent

contracts are those with the greatest difficulties in balancing work–life domains.

By sector of activity, the most exposed workers are those in transportation. They face the worst conditions for several of

the psychosocial work environment indicators: they have to cope with quite high psychological demand, although with

low support from colleagues or supervisors (most of them probably work alone driving trucks or coaches) and low

rewarding aspects, such as pay, career advancement or job security. Another sector with worryingly poor psychosocial

work environment conditions is construction, with highly demanding jobs without correspondingly high discretion,

autonomy, satisfaction or reward.

Health and well-being at work

Variable

Self-employed,

without employees

Self-employed, with

employees

Employee,

permanent contract

Employee, fixed or

temporary contract

Employee, no

contract/other type

of contract

Male Women Male Women Male Women Male Women Male Women

Psychological demand 29 25 34 27 33 30 33 30 40 32

Cognitive demands 59 48 66 65 64 56 52 43 49 37

Emotional demands 58 62 64 67 47 56 44 50 46 46

Demands for hiding emotions 36 36 44 41 35 43 34 40 42 41

Skill discretion 66 64 69 74 67 65 60 59 56 55

Decision authority 83 81 86 83 57 56 49 48 48 50

Support from colleagues 60 60 81 82 74 74 73 74 72 71

Support from supervisors 0 0 0 0 77 78 76 75 77 74

Social community 68 74 78 78 68 69 62 64 60 62

Job rewards 44 43 56 55 51 47 46 45 43 40

Work–life balance 73 76 69 69 64 66 62 65 61 65

Job security 76 76 80 77 73 74 51 51 60 62

Several surveys found that the following factors are those that people mostly mention as determinants of well-being:

material conditions and consumption, a fulfilling family life such as being married and having children (Conceiçao and

Bandura, 2008; Cantril, 1965; Frey and Stutzer, 2002). Personal and family health is also a significant determinant, as

well as work-related issues (Eurofound, 2007b; Poggi et al, 2011). Although international and domestic issues are rarely

mentioned, studies have found that these have a significant influence on people’s happiness, too (Frey and Stutzer, 2002).

Van Hoorn (2007) suggests classifying this wide variety of determinants in six broad categories: personality; contextual

and situational; demographic; institutional; environmental; economic factors.

Work is just one of these factors, but a pivotal one, since it not only provides an individual with an adequate income to

fulfil material needs but also gives people a sense of identity, meaning and accomplishment. Indeed, people in

employment report a significantly higher average level of life satisfaction than those who are unemployed (Clark, 2009).

But having or not having a job is just a part of the story. The quality of work varies significantly among sectors and

workforce groups. There are certain jobs (for instance, those which do not allow for personal development, or which are

dangerous and unhealthy) whose detrimental effect on well-being is even higher than that of unemployment (Grün et al,

2008).

For this reason, the relationship between well-being and quality of work and employment is taken as a starting point,

considering several indicators of quality. Then the analysis goes deeper into two aspects, which are of particular

importance – income and job insecurity/employability.

Quality of work and well-being

The importance of quality of work on individual well-being is clear, but there is not a single definition of it. The

perspective adopted here takes an objective approach stemming from the seminal theoretical framework provided by

Maslow’s need–satisfaction model (1954). The main tenet is that people have basic needs they seek to fulfil through

work, which include the need for survival (pay, security), social needs (need for interpersonal interaction, membership,

friendship), individual needs (need for self-esteem and autonomy), and self-actualisation needs (Beham et al, 2006). In

this view, job quality is constituted by generic elements that meet universal needs. The extent of those needs will differ

according to a person’s circumstances, including the social and physical environment in which a person lives, but their

universality is why many authors, among which are Green (2006 and 2011), Muñoz de Bustillo and Fernandez (2005)

and others, suggest that an objective concept of work and employment quality should be investigated.

Eurofound (2002) provides a useful framework for this purpose, which conceives quality of work as a multidimensional

construct that includes four key areas: ensuring career and job security; maintaining and promoting the health and well-

being of workers; competence development; combining work and non-working life. Green (2006) develops the idea that

a ‘good job’ is one that allows workers to achieve well-being and to achieve a range of personal goals, offering a high

capability to do and be things that they value. The following section considers the four dimensions listed above.

Eurofound’s report Trends in job quality in Europe (2012b) describes the methodology for building these indicators and

provides results. The report looks at indexes at the level of the job, distinguishes extrinsic job features, such as ‘earnings’

and ‘prospects’, as well as a larger set of intrinsic features of the work itself, ‘intrinsic job quality’ – which is further

divided into ‘a safe physical environment’, ‘a secure and trusting social environment’, ‘skills and autonomy’ as well as

‘work intensity which constitutes a negative feature towards job quality’ – and ‘working time quality’:

n‘Earnings’ refer to the level of monetary rewards associated with the job. The target indicator is monthly earnings.

n‘Job prospects’ refer to those aspects that contribute to a person’s material and psychological needs for employment

continuity and self-esteem. The key figures composing the index are the perception of job security, whether there are

prospects of career advancement and the type of contract.

Determinants of well-being

n‘Intrinsic job quality’ refers to aspects of the work itself and of its environment. Here there are four core sets of

features included: the quality of the work itself (skill, autonomy, organisational involvement), the social environment

in which workers are situated (support from manager and colleagues, absence of abuses and positive climate), the

security and quality of the physical environment (low risk exposure), and the intensity of work (tight deadlines,

number of pressure sources, demand for emotional and value conflicts).

n‘Working time quality’ measures the extent to which a job allows a better conciliation of work and family duties,

taking into account number of hours worked, the flexibility of work arrangements and working schedule (night,

weekend and shift work).

In Figure 21, the average level and variability of well-being are reported for different levels of the four quality indices.

Three main facts can be highlighted. First, for all indicators there is a clear positive relationship between well-being and

quality. This is expected as each index is composed of factors selected because they have been proved to be associated

with health outcomes in prospective longitudinal studies.

Second, the aspects that are more effective in shaping workers’ well-being are the intrinsic job quality as well as

employment quality, with a well-being gap between the highest and lowest quality level of 19 and 16 points respectively.

It is worth noting that these aspects of quality are not monetary.

Finally, there is a negative relationship between quality and variability of well-being. As is the case for job satisfaction,

variability tends to decrease when quality improves. This means that once very good working conditions are achieved

individuals have consistent levels of well-being. It is in facing bad job quality conditions that differences in the

individual and/or collective capacity to cope emerge. There are clearly many individuals who are capable of

compensating their situation and people with worryingly low levels of well-being.

Figure 21: Average level and variability of well-being, by quality of work indicators

Note: WHO-5 variability (red lines) is measured by the coefficient of variation.

Source: EWCS 2010; Indices are based on Eurofound 2012b

Health and well-being at work

0.39

0.26

0.2

0.4

0.6

0.8

12345678910

Deciles of monthly earnings

0.45

0.24

0.2

0.4

0.6

0.8

12345678910

Deciles of intrinsic job quality

0.43

0.24

0.2

0.4

0.6

0.8

123456789

Values of employment quality

0.37 0.30

0.2

0.4

0.6

0.8

12345678910

Deciles of work–life balance

Health and well-being at work

The overall facts depicted here hold for different genders and age groups. However, some interesting specificities do

emerge (see Table 6). For women, intrinsic job features are more relevant than employment quality in determining well-

being, while for men the two dimensions present the same correlation with well-being. Interestingly, working time

quality affects both genders’ well-being in the same way. The variability of well-being at very low level of quality of

work is quite similar between the two.

More pronounced differences emerge according to age groups. Job quality is more effective in shaping the well-being

of older workers, since for them the correlation between well-being and quality of work is stronger for all the indicators.

Furthermore, older workers show a higher variability in the well-being scores when quality of work is very bad. This is

a policy-relevant result for the issue of ‘managing an ageing workforce’ to promote the participation of older workers

and discouragement of early retirement.

Table 6: Correlation between well-being and quality of work, and variability of well-being at very low levels of quality

of work indicators, by gender and age

Source: EWCS 2010

Income and happiness

Although the job quality index on earnings shows a relatively weak association with well-being, the relationship

deserves a closer look. Easterlin, in a seminal work, was the first to find some puzzling evidence, showing that over time

happiness does not increase when a country’s income increases (Easterlin, 1974). This is what has long been claimed as

the happiness–income paradox, a result supported in a number of other studies (Easterlin, 1974, 2005; Easterlin et al,

2010; Myers, 1995). Actually, some recent works provide convincing evidence that self-reported measures of well-being

such as happiness or life satisfaction do rise with income also in a longitudinal perspective (Stevenson and Wolfers,

2008; Diener and Biswas-Diener, 2002), but the relationship does not follow a linear shape, the effect dampening down

with income.

The point is that the relation between monetary income, satisfaction of needs and well-being is not always clear-cut.

Several theories have been proposed to disentangle this relation, such as the diminishing marginal utility of money

(Veenhoven, 2006; Frey and Stutzer, 2002) and the ‘values shift’ theory (Inglehart, 1990). Another set of processes that

work as confounding factors between individuals’ real experience and its evaluation are the so-called ‘relative

comparison’, ‘expectation’ and ‘adaption’ processes, as summarised by Michalos in his ‘multiple discrepancy theory’ of

satisfaction (1985).

Correlation between well-being

and job quality

Variability of well-being at very

low levels of job quality

Men Women Men Women

Monthly earnings 0.09 0.11 0.38 0.4

Intrinsic job quality 0.27 0.25 0.44 0.45

Employment quality 0.2 0.25 0.41 0.43

Work–life balance 0.1 0.1 0.37 0.37

Under 50 Over 50 Under 50 Over 50

Monthly earnings 0.1 0.14 0.38 0.41

Intrinsic job quality 0.24 0.29 0.44 0.48

Employment quality 0.22 0.24 0.42 0.44

Work–life balance 0.09 0.12 0.36 0.42

The richness of the fifth EWCS data makes it possible to shed some light on the issue, since it provides both a subjective

and objective view of the respondents’ income. Figure 22 correlates well-being and two additional income measures, the

first of which is the difficulty of the workers’ family to make ends meet. This is an objective measure of income which,

compared to monthly earnings, proxies in a more direct way the role of work for satisfying basic needs. The gradient is

still the expected one: the greater the difficulties, the lower the well-being. However, it is interesting to note that the gap

in well-being between the lowest and the highest groups is far more pronounced compared to the gap associated with

monetary earnings. Second, there is almost no heterogeneity among the different country groupings considered in

Figure 22: the difficulties in making ends meet smooth well-being differences among European workers.

The other measure considered in Figure 22 (‘I am well paid for the work I do’) is subjective. Again, the gradient is the

expected one: workers with higher income satisfaction present higher levels of emotional well-being. The interesting

point is that country differences are more apparent: the role of (satisfaction about) income is stronger in the less affluent

countries of the sample, such as eastern European countries (where the correlation is r=0.29), while Scandinavian

countries show the lowest association (r=0.15). Scandinavian countries are at the same time among the richest in the

sample and the ones with lower inequality in income distribution. The lower correlation in these countries looks coherent

with the social comparison and the ‘values shift’ theories.

Figure 22: Average well-being, by material living conditions and country groups (%)

Note: Values shown only for eastern European and Scandinavian countries.

Source: EWCS 2010

Effect of the risk of unemployment on individual well-being

A strong negative relationship between individual well-being and unemployment has been a pillar result of the empirical

research: not having a job when you want one reduces well-being more than any other single factor, including important

negative ones such as divorce and separation (Clark and Oswald, 1994).

Although the effect of unemployment on well-being is not directly observable in the EWCS, the effect of the risk of

unemployment may be studied. There is evidence that this effect is of considerable size, too (Di Tella et al, 2001). This

point has to a certain extent already been touched on in previous analyses on job quality – it is a part of the relationship

between the ‘prospects’ indicator and well-being – but it deserves a closer look.

Health and well-being at work

65 68 69 71

Strongly

disagree

Disagree Neither

agree nor

disagree

Agree Strongly

agree

WHO-5 (0–100)

I am well paid for the work I do

Anglo-Saxon

Connental

Eastern

Scandinavian

Southern

70 70

72 74

With

great

diﬃculty

With

diﬃculty

With

some

diﬃculty

Fairly

easily

Easily Very

easily

WHO-5 (0–100)

Thinking of your household’s total monthly i ncome, is your

household able to make ends meet…?

Anglo-Saxon

Connental

Eastern

Scandinavian

Southern

Health and well-being at work

When looking separately at the average well-being associated with having a permanent contract (versus a temporary one)

and of perceived job security (versus job insecurity), there is a greater role for job insecurity compared to the type of

contract (Figure 23)9, particularly for women. Nonetheless, for both genders the fear of losing a job is associated with a

remarkable drop in average well-being.

Figure 23: Average well-being, by contract and job security, stratified by gender (%)

Source: EWCS 2010

It is important however to contextualise the relationship, both when compared to the macroeconomic conditions and with

the employability of the worker: both may change dramatically the emotional impact of the risk of losing the job.

Figure 24 deals with the first point. As both the EQLS and EWCS include the WHO-5 index and the same unemployment

risk questions, it is possible to compare the drop in well-being associated with perceived job insecurity in very different

macroeconomic conditions: in 2007 (EQLS), months before the world economy was hit by the financial crisis, and in

2010 (EWCS), beyond its lowest point but still distant from a full recovery.

9Job security is based on Q77A: ‘How much do you agree or disagree with the following statement: I might lose my job in the next

6 months’, with five possible answers: strongly agree, agree, neither agree nor disagree, disagree and strongly disagree. Those

defined as ‘job insecure’ were those answering strongly agree or agree, and those defined as ‘job secure’ were those answering

neither agree nor disagree, disagree or strongly disagree.

64 64

Women Men

Permanent contract Temporary contract Secure job Insecure job

Figure 24: Prevalence of perceived job insecurity (bottom) and its effect on workers’ well-being (top), 2007 and 2010

Note: EQLS 2007 data refer exclusively to workers. The well-being loss is calculated with reference to those who answered ‘very

unlikely’.

Sources: EWCS 2010, EQLS 2007

As expected, the prevalence of job insecurity has increased since 2007: ‘very unlikely’ answers have fallen sharply and

‘likely’ and ‘very likely’ have increased (bottom-left axis). At the same time, the negative effect of unemployment risk

on well-being has increased severely (upper-right axis). Taking those answering ‘very unlikely’ as a term of reference,

the well-being loss due to job insecurity in 2010 has increased by more than three times compared to 2007 for workers

who say that they are likely or very likely to lose their job. It is worth noting that there is a significant gradient in the

2010 figures. The well-being loss was quite similar among different workers’ groups in 2007, whereas in 2010 the more

insecure a worker was, the higher was the loss of well-being.

The second key aspect to consider is that of employability, as in the ability of individuals to find and sustain employment.

Employability clearly modifies the impact that job insecurity may have on well-being: less employable people will be

more prone to distress and more hurt by the fear of being unemployed. It has been estimated that an increase in

employability from zero to 100% reduces the detrimental effect of job insecurity on well-being by more than half (Green,

2011).

Health and well-being at work

-10

-9

-8

-7

-6

-5

-4

-3

-2

-1

Very unlikely Unlikely Neither unlikely nor

likely Likely Very likely

WHO-5 loss (% points)

% Prevalence

2007 - Prevalence % 2010 - Prevalence % 2007 - WHO-5 loss 2010 - WHO-5 loss

Reference

category

Health and well-being at work

Here employability is considered using the following EWCS question: ‘If I were to lose or quit my current job, it would

be easy for me to find a job at a similar salary’.10 As shown in Figure 25, insecure workers have considerable losses in

well-being compared to secure workers, with the high-skilled clerical workers suffering the biggest fall in well-being

(9.3 points), and the high-skilled manual workers the least (6.2 points). It is interesting to note that the role of

employability in reducing the detrimental effect of job insecurity is quite different across occupations. The results show

that high-skilled clerical workers suffer a loss that is not much modified by their employability level. In contrast, high-

skilled manual workers are almost untouched by the fear of losing their current job when employability is high.

Figure 25: Effect of perceived job insecurity on well-being, by employability and occupational groups

Note: Employability difference in well-being loss is not statistically significant for high-skilled clerical workers.

Source: EWCS 2010

10 The question presents five possible answers going from ‘strongly disagree’ to ‘strongly agree’. ‘High employability’ refers to those

who answer ‘agree or strongly agree’ and ‘low employability’ refers to those who answer ‘disagree or strongly disagree’.

-9.3

-6.7

-6.2

-7.9

-7.1

-2.6

-0.7

-3.1

-10

-9

-8

-7

-6

-5

-4

-3

-2

-1

High-skilled clerical Low-skilled clerical High-skilled manual Low-skilled manual

WHO-5 loss (% points)

Low employability High employability

This chapter explores the association between the health of workers and the psychosocial work environment. Research

has proven that stress at work is associated with a number of physical and psychological negative effects at the individual

level, such as cardiovascular diseases, musculoskeletal disorders, immunological problems and mental health problems

(anxiety and depression disorders).

As for the health outcomes, the focus is first on the self-assessment of general health conditions. In particular, the study

assesses those individuals reporting ‘bad’ or ‘very bad’ health, and those individuals mentioning more than two health

problems. The analysis then focuses deeper on specific aspects of physical and mental health arising as musculoskeletal

symptoms or as anxiety and depression disorders. Finally, there is an analysis of the prevalence and determinants of work

accidents.

The aim here is to identify whether, once controlling for the job and individual characteristics, there are specific

predictors for each health and safety outcome among the many domains of the psychosocial work environment, as

implemented along the domains proposed in Chapter 2. Also, a specific focus on country and industry effects will be

conducted.

The first section discusses the main determinants of health used in the analysis. The second describes the health and

safety outcomes: general health, musculoskeletal symptoms, anxiety and depression disorders, and work accidents. Then

the predictors of general health are presented that can be singled out among psychosocial work environment domains as

well as among individual and job characteristics, while the subsequent paragraphs do the same, focusing in turn on

musculoskeletal symptoms, anxiety and depression disorders, as well as work accidents.

The main results of the analyses are presented in various tables and figures; additional results, which are presented and

commented on only in the text, are available upon request.

Determinants of health and safety at the workplace

Key determinants of health and safety at the workplace include:

nindividual characteristics (age and gender), as they are naturally linked to health;

nhuman capital endowment (years of education, experience and tenure, training spells) measuring the level of general

as well as job-specific knowledge the worker has;

ngeneral job characteristics (industry, firm size, job contract, occupation) as catch-all features of working conditions.11

Crucially, in addition to these determinants, the EWCS makes it possible to observe and analyse the impact of factors

relating to work organisation, as well as of the psychosocial work environment domains. The psychosocial work

environment domains are defined according to how they were discussed in Chapter 2, while factors relating to work

organisation are defined as follows:

nHours of work: number of hours; unsociable working hours (such as during the night, evenings, on Sundays); having

to face hours variability.12

Work environment and health

11 Earnings are not included because of the numerous non-responses, to avoid losing statistical power as non-responses would be

dropped from the analysis and to avoid a selection bias in the results. Furthermore, it was possible to include many and detailed

job features so that they can capture the effect of earnings as well.

12 Q37 A to E, the last reversed, grouped in one factor.

nPace of work: pace of work determines whether pace of work is dependent on other factors13; presence of frequent

and disruptive interruptions; piece-rate pay.

nPhysical factors: working outside; environmental hazards (such as exposure to chemicals, cold14); posture-related

hazards (such as working in awkward postures, lifting, standing15).

nSpecific features: need to travel for work (work at clients’, patients’, customers’ premises, or in a car); work with

clients; having to cope with the introduction of new processes or restructurings.

nOther challenges: second job, whether occasional work or not.

The aim of this exercise is to look inside the ‘black box’ of the general definition of a job (occupation, contract) to

determine the most significant associations between specific features of the content of the job and health/safety

outcomes. The same set of determinants were the focus for all the outcomes in the analysis, to be able to compare their

different impact on general as well as physical and mental health, and on safety.

Health and safety outcomes

‘Bad health’ and ‘2+ health problems’ were described in Chapter 1, where it emerged that bad health is mentioned by

2.5% of European workers, while on average 50% of female and 45% of male European workers mention more than two

health problems. Here, the analysis also takes into account individuals mentioning ‘backache’, ‘muscular pains in

shoulders, neck and/or upper limbs’ and ‘muscular pains in lower limbs (hips, legs, knees, feet, etc.)’. Because of their

high correlation, the three outcomes are assessed jointly, generating a single indicator signalling the existence of at least

one of the three problems and with a prevalence in the population of around 60%.16

Health and well-being at work

Statistical approach

A multivariate analysis is conducted, in order to measure the change in the prevalence of the health outcome when

changing the value of one of its determinants and holding the value of the other determinants as constant (the so-called

‘marginal effects’).

This implies that ‘residual’ country (or industry) effects can be read as catching different levels of prevalence of the

health outcomes as if all individual and job characteristics were the same in all countries (or industries); hence they are

labelled ‘residual’; in other words, due to historical or institutional differences or to any other relevant element not

included in the multivariate models used here.

All analyses have been performed weighting the data according to the sample fraction in each country.

13 The number of external inputs – all Q46 modalities – inflated on a 0–100 scale.

14 All modalities recorded by Q23, grouped in one factor. All factors vary on a 0–100 scale.

15 Q24 A to D modalities grouped in one factor.

16 The impact of the abovementioned determinants is remarkably similar on the three outcomes when analysed separately, so that it

is safe to analyse them jointly.

Health and well-being at work

Individuals are then considered who mention ‘depression or anxiety’, with a prevalence among European workers of

about 9.5%. The two outcomes, jointly asked in the EWCS, have been chosen as best indicators of mental health, as they

have a clear pathological dimension and often prompt medical intervention, while fatigue or insomnia can be mentioned

more freely and can be biased by self-perception of own general well-being.

Finally, the analysis looks at days of absence from work due to work-related accidents. In the EWCS, the absence is self-

reported, and includes short absence, contrary to most national registers. The average prevalence of work accidents by

length of absence is shown in Table 7.

Table 7: Prevalence of work accidents by length of absence

Source: EWCS 2010

As expected, workers mentioning more than two health problems or reporting bad health are much more likely to

experience both musculoskeletal symptoms and depression or anxiety disorders, or to have experienced recently (during

the year of the interview) an accident at work (Table 8). Results point to an indissoluble connection between physical

and mental health.

Table 8: Average prevalence of musculoskeletal symptoms, depression or anxiety disorders and work accidents (%)

Source: EWCS 2010

General health

This section analyses the impact of each determinant as listed in the previous paragraph on both ‘3+ health problems’

and ‘bad health’.

The findings reveal that the prevalence of ‘3+ health problems’ is higher among women. In Chapter 1 the same feature

was observed, on average, but more can be learned here. In fact, the analysis now looks at ‘marginal effects’ (see box

on ‘Statistical approach’): the prevalence of ‘3+ health problems’ is higher among women once they are considered as

if they were holding the same kind of job men hold; in other words, their prevalence is higher not because they hold jobs

of a specific kind (the so-called ‘composition effect’), but because of some specific feature that is unobservable.

Days off White-collar workers (%) Blue-collar workers (%) All (%)

None 97.15 93.78 95.96

1+ 2.85 6.22 4.04

1–3 1.07 1.93 1.37

4+ 1.78 4.29 2.67

Days off Musculoskeletal symptoms Depression or anxiety Work accidents

3+ health problems

No 30.5 1.4 2.4

Yes 93.0 17.7 5.9

Bad health

No 59.5 8.5 3.9

Yes 89.1 34.8 8.5

All workers 60.2 9.2 4.0

Ageing, as expected, is associated with an increase in the prevalence of general health problems, again holding job

characteristics as if they were constant across age groups. The following paragraphs show that gender and age have the

same impact also on musculoskeletal symptoms, on anxiety and depression disorders, and on work accidents.

Human capital endowment has no significant link with the outcomes. This means that education, tenure and training sort

workers among jobs (putting high human capital workers in healthier jobs), but then human capital on its own has no

impact on general health. The same is true for the type of contract, with the exception of self-employed and informal

workers, who experience a higher prevalence of ‘bad health’ and ‘3+ health problems’ respectively. Workers employed

in elementary occupations face worse general health conditions compared to all other occupational groups.

Observable job features usually do not go beyond those discussed up to this point. The EWCS makes it possible to

deepen the analysis much further. It emerges that it is not the number of worked hours as such that hampers general

health, but it is working during unsociable hours or facing variable working hours, as well as facing disruptive

interruptions, that does so. Other features of work organisation hamper one or both aspects of general health of workers:

having to travel, work with clients, face restructurings or environmental hazards. The next paragraphs, focusing on

specific health outcomes, will explore whether these specific job features are more disruptive for physical or mental

health; or whether they are more disruptive for manual or non-manual workers.

Crucially, it appears that the psychosocial work environment has a significant impact on the general health of workers.

The prevalence of ‘3+ health problems’ increases with high demands at work, and job insecurity, but also with high

decision authority. It decreases with skill discretion, good interpersonal relations and a good work–life balance. The

same associations appear in association with ‘bad health’, even though skill discretion and decision authority bear no

significant impact on it. Several of these associations will appear again alongside physical and mental health.

It is worth noting that holding constant individual and job features, residual country effects17 seem to group countries

in ‘low prevalence’, ‘average prevalence’ and ‘high prevalence’ of general health problems (Figure 26), although bad

health seems to be more evenly distributed across countries.18 On the contrary, again holding constant individual and

job features, residual industry effects19 are not significantly different from each other.20 This last result is new and

important.

Health and well-being at work

17 In total, 33 country dummies are included in the model. Poland is the reference country (excluded category).

18 Not reported.

19 A total of 20 industry dummies are included in the model. Manufacturing is the reference industry (excluded category).

20 Confidence intervals are overlapping.

Health and well-being at work

Figure 26: Residual country effects from multivariate analysis on reporting three or more health problems

Note: Vertical bars refer to confidence intervals at 95%. The Y axis shows the percentage-point differences in prevalence (expressed

in decimals) with respect to the reference country Poland.

Source: EWCS 2010

Physical health outcomes: backache and muscular pains

To analyse the impact of each determinant on this physical health dimension (backache and muscular pains) all workers

were analysed together, then separated into white-collar and blue-collar categories to acknowledge the different nature

of manual and non-manual tasks; they were further separated into two groups: those whose WHO-5 mental health index

is above the median and those whose WHO-5 index is below the median.21 The last split is to control for the existence

of reverse causality between low well-being and the propensity to ‘complain about everything’. This is relevant, as

backache or muscular pains are quite generic problems, not subject to a specific diagnosis, so they can be mentioned

freely according to individual propensity to voice physical health issues. Table 9 shows that the average prevalence of

reported increases in backache and muscular pains when individual well-being is lower, confirming the importance of

controlling for this possible source of reverse causality. Hence the results to be discussed as more reliable will be those

arising from the models that separate manual and non-manual workers and exclude low WHO-5 individuals.

Individual characteristics are relevant: women mention backache or muscular pains more often; age increases the

likelihood of facing backache or muscular problems, as expected. Human capital endowment, as measured by education,

is associated with a lower prevalence of backache or muscular pains among white-collar workers only.22

-0.4

-0.3

-0.2

-0.1

0.1

0.2

0.3

IE UK EL ES HR AT BG RO DE HU MT NO XK NL SK LU AL CZ BE SE TR SI DK CY FR IT MK ME LT LV EE PT FI

21 WHO-5 value at the European level computed by occupation: 68 for blue-collar workers and 72 for white-collar workers.

22 Tenure increases backache and muscular pains, but only among low well-being blue-collar workers, a result at risk of reverse

causality, that is that it might be ascribed to a general fatigue of individuals having had to do – maybe unwillingly – a manual job

for a long time.

The formal definition of the job (firm size, contract, public/private legal setup) and even occupation – once separating

blue-collar and white-collar workers – has no significant impact on the prevalence of backache or muscular pains. Also,

no significant differences emerge among industries, with very few exceptions.23 This is a novel and important finding,

and it shows that when controlling for the actual content of the job, its formal definition becomes non-informative. This

is rarely done, since in most datasets only industry, firm size, contract and occupation are observable. This calls for the

use of rich datasets like the EWCS, and a shift in policy focus to the actual content of the job.

Looking into the requirements of the ‘job’ specifically, the following emerges. Among physical factors, environmental

hazards increase backache or muscular pains significantly for all workers. Table 9 shows that their effect on backache

or muscular pains is stronger at higher values of well-being, and more so for white-collar workers than for blue-collar

workers. The size of the impact is not small: for instance, given an average marginal effect in the population of 0.4

percentage points, an increase in the environmental hazards from 0 to a score of 50 (half of its range) leads to an increase

of 20% in the prevalence of backache or muscualr pains, which is one third of the average prevalence (59.9%).

Work organisation proves important for the physical health of individuals. Piece-rate pay increases the prevalence of

backache or muscular pains among blue-collar workers (the only group facing the possibility of this pay setting) and

suggests that this kind of production incentive might put too much strain on manual workers, and that workers might

be persuaded to trade health for a higher pay. No other aspects of work organisation change the prevalence of backache

or muscular pains among blue-collar workers. White-collar workers, nevertheless, report an increase in the prevalence

of backache or muscular pains with the introduction of new processes or restructurings in the workplace, maybe

because the organisation and implementation of these innovations is a source of stress for them. Working with clients,

also potentially stressful, increases their prevalence of backache or muscular pains, while having to travel for work

decreases it.

Psychosocial work environment domains are decisive in explaining the prevalence of backache or muscular pains,

confirming the double nature of these physical health outcomes, both physical and psychological. Furthermore,

controlling for well-being is crucial in disentangling reverse and direct causality of each determinant, but chiefly of

psychosocial determinants. It is estimated that ‘psychological demand’ increases backache or muscular pains (coherently

with Linton, 2001; Ariens et al, 2001) for white-collar workers. Coherently with this first result, ‘skill discretion’ is

protective, while – at odds with it – ‘decision authority’ appears to increase backache or muscular pains. Decision

authority increased the prevalence of ‘3+ health problems’ as well, and it appears that it also increases the prevalence of

depression and anxiety disorders. This unexpected effect might arise because decision authority can be associated with

stressful heavier responsibility. In fact, the assumption of linearity of the effect of demand and control on health in

Karasek’s model has been criticised, based on the considerations that too much decision authority may be as stressful as

having too little of it, and that decision authority may have a U-shaped relationship with health (Warr, 1994). In the

estimates applied here, a linear relationship is imposed and hence only one side of the U-shape can emerge.

A strong and consistent result is related to ‘rewards’ that decrease backache or muscular pains for all workers. The effect

increases with skill and well-being and it is not negligible, its size being about two-thirds of the effect of environmental

hazards (Table 9). ‘Cognitive demand’ (and ‘work with clients’), which is a potential stressor, increase backache or

muscular pains among white-collar workers (as in Da Costa and Vieira, 2010). Finally, a good work–life balance

Health and well-being at work

23 Exceptions are finance at the low extreme and ICT at the high one.

Health and well-being at work

decreases backache or muscular pains among blue-collar workers, while job insecurity increases it among white-collar

workers.24 It is worth noting that the actual work contract (temporary or no contract at all) had no impact on backache

or muscular pains, while the subjective perception of job insecurity did have. This is very important, as it denies that

physical health is not associated with job security.

Table 9: Prevalence and marginal effects of selected determinants of musculoskeletal problems

Source: EWCS 2010

Lastly, once controlling for all the above determinants, the residual country effects are remarkably similar. As Figure 27

shows there is just a small group of countries showing ‘high backache or muscular pains’ (Nordic countries, Italy and

Portugal) and a small group showing ‘low backache or muscular pains ’ (Anglo-Saxon and some Balkan countries). This

result reinforces the conclusion that it is the actual job content that determines physical health, more than its formal setup

or even the institutional environment, as no institutional/cultural/political regularity seems to emerge in the ordering of

countries in Figure 27.

Figure 27: Residual country effects from multivariate analysis on reporting musculoskeletal problems

Note: Vertical bars refer to confidence intervals at 95%. The Y axis shows the percentage-point differences in prevalence (expressed

in decimals) with respect to the reference country Poland.

Source: EWCS 2010

All

Blue-collar

Low WHO-5

Blue-collar

High WHO-5

White-collar

Low WHO-5

White-collar

High WHO-5

Average prevalence (%)

59.9 76.5 58.2 63.6 45.3

Marginal effects (percentage points)

Environmental hazards 0.40 0.29 0.40 0.36 0.47

Decision authority 0.09 n.s. 0.13 0.09 0.08

Rewards -0.27 -0.19 -0.27 -0.23 -0.25

-0.4

-0.3

-0.2

-0.1

0.0

0.1

0.2

0.3

IE BG EL AL UK XK RO HU ES SK LU MT LT TR MK CZ HR AT CY ME SI FR DE BE NL LV SE IT EE NO DK PT FI

24 The other psychological determinants have no impact or a significant impact only among low well-being workers, so the reverse

causality effect cannot be excluded and this weakens the results.

Mental health outcomes: depression or anxiety

It is well known in the literature that depression and anxiety have a gender dimension (Kuehner, 2003; Piccinelli and

Wilkinson, 2000; Pigott, 1999), as confirmed in Table 10, where average prevalence by gender and occupation is

reported. Hence in the analysis – after considering all workers together25 – they are then separated by gender. Workers

are further separated by occupation, to fully recognise the different nature of manual and non-manual tasks; as Table 10

shows, gender differences in the prevalence of depression and anxiety are huge among manual workers (12.9%

compared to 6.7%), while they are much narrower among non-manual workers (10.7% compared to 9.3%).26

Once the sample is split by gender and occupation, several regularities emerge. First, also in the case of mental health,

the formal definitions of the job (firm size, occupation, contract and so forth) have no significant impact on the

prevalence of depression and anxiety; even among industries no significant differences emerge. This confirms the novel

and important finding discussed in case of physical health: once the analysis can control for the actual content of the job,

its formal definition becomes non-informative.

The only exception is the work contract of white-collar workers, which is associated with a higher prevalence of

depression and anxiety when it is not a dependent-work contract, when it is again confirmed that those not working

inside a firm (but as self-employed or without a contract altogether) face worse health conditions. In general, self-

employed and workers with no contract face worse health conditions both in general terms and with reference to

depression and anxiety. Nevertheless, they enjoy a lower prevalence of musculoskeletal symptoms and of work

accidents, maybe due to their lower exposure to risk.

Different patterns emerge between manual and non-manual workers facing determinants related to the physical content

of the job: working in the open air decreases the prevalence of depression and anxiety among blue-collar workers, while

being exposed to environmental hazards increases it among white-collar workers. Among job organisation determinants

able to influence both genders and both occupational classes, it emerges that the effect of introducing new processes or

restructuring, as well as of facing disruptive interruptions, all increase depression and anxiety significantly. The ability

of determining own pace of work reduces the prevalence of depression and anxiety among women.

Psychosocial work environment domains are crucial determinants of mental health, as expected. The effect of exposure

to high demands on the risk of depression and anxiety appears particularly relevant, coherently with the findings of

several studies (Rugulies et al, 2006; Paterniti et al, 2002; Wang, 2004; Shields, 2006; Kawakami et al, 1992; Virtanen

et al, 2007; Bonde, 2008; Netterstrøm et al, 2008). A specific gender pattern emerges: women are more sensitive to low

decision authority (manual workers) and high psychological demand (non-manual workers), both increasing depression

and anxiety. Male workers are more sensitive to skill discretion (decreasing depression and anxiety), cognitive demand

and work with clients.27 All white-collar workers are vulnerable to the need to hide emotions and to the feeling of job

insecurity (as in Stansfeld and Candy, 2006). A good work–life balance decreases depression and anxiety. Crucially, a

Health and well-being at work

25 To single out the effect of gender, all workers have to be considered together. It is estimated that the prevalence of depression and

anxiety among women is significantly higher than among men also holding individual and job characteristics as if they were the

same across genders; this confirms what is known in the literature.

26 The split by WHO-5 values (as applied when analysing backache and muscular pains) would be tautological, as depression and

WHO-5 are strictly correlated by definition.

27 This might explain the negative impact of unhealthy positions on depression and anxiety (that is, it decreases the prevalence of

depression and anxiety) for white-collar women: it includes ‘standing’ and ‘lifting people’, both related to work with

clients/patients; something that seems not depressing for women.

Health and well-being at work

good social community at work and having rewards decreases depression and anxiety for all workers, and a higher

emotional demand increases it for all.

Table 10 reports the marginal effects of these three crucial determinants; their impact is not small, as they are measured

on a 1–100 scale. The reduction in the prevalence of depression and anxiety due to a good social community at work is

quite homogeneous across genders and occupations: on average moving from a 0 score to a 50 score reduces the

prevalence of depression and anxiety by more than six percentage points. Nonetheless, women are more sensitive to

rewards and emotional demands compared to men: the marginal effects are almost double in size for women.

Finally, once conditioning for all the above determinants, the residual country effects are quite similar (Figure 28),

although countries’ residual heterogeneity in mental health is higher compared to physical health.

Table 10: Prevalence and marginal effects of selected determinants of mental health problems

Source: EWCS 2010

Figure 28: Residual country effects from multivariate analysis on reporting mental health problems

Note: Vertical bars refer to confidence intervals at 95%. The Y axis shows the percentage-point differences in prevalence (expressed

in decimals) with respect to the reference country Poland.

Source: EWCS 2010

All

Blue-collar

women

Blue-collar

men

White-collar

women

White-collar

men

Average prevalence (%)

9.4 12.9 6.7 10.8 9.4

Marginal effects (percentage points)

Social community at work -0.13 -0.12 -0.10 -0.13 -0.16

Rewards -0.06 -0.07 -0.04 -0.08 -0.06

Emotional demand 0.06 0.08 0.03 0.07 0.05

-0.1

-0.05

0.05

0.1

0.15

0.2

0.25

0.3

0.35

0.4

DE AT EL IE ME NL RO HU SK TR ES LT BE SI IT MK XK DK MT UK HR CZ AL LU FR BG FI PT LV SE NO EE CY

Safety: work-related accidents

The analysis concludes by focusing on a specific risk for the health of individuals who work: the risk of experiencing a

work accident. It can jeopardise physical health temporarily or permanently, it can be followed by depression or anxiety,

and it can jeopardise employability.

As already mentioned, in the literature a cut-off at three days of absence due to a work accident is applied, so that minor

accidents are excluded from any analysis. This is because most countries’ legislation does not cover short absences and

they are consequently not recorded. In the EWCS however, the absence is self-reported, and includes all absence lengths.

This provides a unique chance to shed some light on minor accidents.

Three groups are defined: (i) no accident, no absence, (ii) mild accident, short absence of 1–3 days, (iii) serious accident,

standard absence of 4 days or more. A general pattern emerges28: the prevalence of mild accidents is linked to

determinants (individual and job characteristics) in the same way that serious accidents are, although with smaller effect

size. Hence, pooling workers taking short leave to the group of workers taking no days off, that is those not experiencing

even mild accidents, is misleading as they do not share the same reaction to determinants. Based on this evidence, it is

possible to estimate the impact of the set of determinants on work accidents, considering all lengths of absences (one

day or more), pooling mild and serious accidents, and analysing them against the group of workers experiencing no

accidents at all.

In this analysis, after considering all workers together, they are separated by occupation due to the – obviously different

– set of risks faced by manual and non-manual workers. Afterwards, workers are excluded whose tenure is shorter than

one year, to control for risk exposure. Mainly temporary work agency and some temporary contract workers are excluded

by this selection. Average prevalence of work accidents is displayed in Table 11. As expected, blue-collar workers face

the highest risk of work accidents, while this experience is infrequent among white-collar workers. The exclusion of

short-tenure workers increases the one-year prevalence of work accidents but only to a quite limited extent.

From the empirical analysis it appears that no individual characteristic has a significant and robust impact on the

prevalence of work accidents, while the set of formal describers of the job retains a very limited link with work accidents:

only firm size increases the prevalence of work accident, and only for white-collar workers; no occupation or industry

effects emerge. Hence, even in the case of work accidents, controlling for the actual content of the job is sufficient to

cancel the impact of the formal definition of the job itself.29

Among determinants linked to job organisation the following emerges. Central is the role of environmental hazards for

all workers, as must be the case for work accidents (Table 11). The effect of exposure to environmental hazards doubles

among manual workers compared to non-manual workers. Among manual workers, a change of the factor from 0 to 50

increases the prevalence of work accidents by almost five percentage points.

Health and well-being at work

28 As part of the analysis, an econometric model (ordered probit) is applied that is more robust to misreporting and measurement

errors in the outcome of interest with respect to a duration model.

29 Not having a permanent contract decreases the prevalence of work accidents, but the result is weakened when excluding short-

tenure workers; in fact the decrease remains significant only for workers with no contract and self-employees, for whom the very

concept of tenure is blurred.

Health and well-being at work

Blue-collar and white-collar workers face different patterns of determinants of work accidents among factors related to

work organisation. Not working at night, evening and on Sundays decreases work accidents for white-collar workers. A

second occasional job and, most interestingly, travelling for work both increase the prevalence of work accidents among

blue-collar workers. This last finding is important, as accidents while travelling are seldom (or only recently) recorded

in administrative data and hence this effect is difficult to observe. Table 11 shows that those travelling for work face a

prevalence of work accidents 2.5 percentage points higher than the others.

Psychosocial work environment domains are crucial also in this context. The positive role of rewards emerges also in

this case, decreasing the prevalence of work accidents among all groups of workers30 by more than one percentage point

if increased from a 0 to a 50 score (Table 11). A higher cognitive demand increases work accidents for non-manual

workers while a higher decision authority is protective. A good work–life balance decreases work accidents among

manual workers.

Table 11: Prevalence and marginal effects of selected determinants of work accident absence (one day or more)

Note: (*) dummy 0/1

Source: EWCS 2010, data weighted

Finally, after controlling for all the above determinants, residual country effects highlight a few significant differences:

Romania, Bulgaria and Hungary enjoy a lower prevalence of work accidents, while France, Germany, Slovenia, Finland

and Belgium face higher prevalence of work accidents (Figure 29).

30 All but long tenure blue-collar workers, though.

All

blue-collar workers

Blue-collar workers,

long tenure

All

white-collar workers

White-collar workers,

long tenure

Average prevalence (%)

4.3 7.3 7.6 2.9 3.0

Marginal effects (percentage points)

Environmental hazards 0.064 0.091 0.097 0.052 0.052

Travel (*) 1.007 2.376 2.438 n.s. n.s.

Rewards -0.023 -0.034 n.s. -0.017 -0.020

Figure 29: Residual country effects from multivariate analysis on reporting an absence of one day or more because of a

work accident

Note: Vertical bars refer to confidence intervals at 95%. The Y axis shows the percentage-point differences in prevalence (expressed

in decimals) with respect to the reference country Poland.

Source: EWCS 2010

Health and well-being at work

-0.06

-0.04

-0.02

0.02

0.04

0.06

0.08

0.1

0.12

RO BG HU LV SK EE EL AL LT HR ME CZ TR IE NL MK DK NO UK SE ES PT CY LU IT MT XK AT FR DE SI FI BE

Sickness absence has high costs for society in terms of lost productivity and workers’ compensation, and places a burden

on most social security systems. This is why a growing body of literature is investigating which are the main health-

related, economic and institutional drivers of it (OECD, 2010; Livanos and Zangelidis, 2010). The role played by

working conditions is twofold. There is an indirect relationship, which passes through the impact they have on the

physical and psychological health of individuals: the better the working conditions, the better the health of workers, the

lower – all things being equal – the frequency of sick leave. Available literature however has shown that sickness absence

is only partially determined by health (Marmot et al, 1995; Andrea et al, 2003; Farrel and Stam, 1988), whereas social,

cultural and individual factors appear to play an important role.

The relationship that working conditions have with the health of workers has already been the subject of the previous

chapter. Here, it is interesting to investigate whether they play an additional, direct role in the individual decision – given

their health status – to take sick leave.

At the same time, illness or health conditions do not always lead to work absence: sickness presenteeism is the concept

used ‘to designate the phenomenon of people, despite complaints and ill health that should prompt rest and absence from

work, still turning up at their jobs’ (Aronsson et al, 2000). The importance of presenteeism relies primarily on the

associated reduced productivity at work, whose costs have been estimated to exceed those attributable to both medical

expenses and sickness absence (Hemp, 2004). Furthermore, the results of a few longitudinal studies suggest that

presenteeism may increase the risk of developing health disorders (Kivimäki et al, 2005; Bergstrom et al, 2009).

The fifth EWCS gives the rather unique opportunity to contribute to the knowledge of the theme since it explicitly asks

individuals whether they did work when they were sick during the previous 12 months.

This chapter first gives an overview on sickness absence/presence across Europe. Secondly, it attempts to disentangle

the role of working conditions among the non-health factors that have a relation with both phenomena.

Overview on absenteeism and presenteeism

To describe the diffusion of absenteeism and presenteeism, two measures are used: the average days that have been spent

in the two statuses are counted31; the number of people who have been involved are counted (the so-called prevalence).

To have a first overview as complete as possible, the prevalences of sickness absence and sickness presenteeism are

measured as the proportion of subjects with at least one day of sickness absence and at least one day of sickness

presenteeism during the previous 12 months. Later on, higher thresholds on the number of days will be considered.

Most aggregate figures say that the prevalence of sickness absence in Europe is 40%, with an average number of five

days of absence per year; both appear consistent with another Eurofound study on absence from work (Eurofound,

2010a).

Absenteeism and presenteeism

31 Days of sickness absence in the previous 12 months were computed by subtracting from the days of absence for health problems

(Q72) the days of absence due to work injuries (Q73), after recoding records with missing information to zero (no days absent

because of injury) in the latter variable, given the high proportion of subjects with missing information (63.7%). Days of sickness

presenteeism in the previous 12 months were obtained from answers to Q74b, after excluding subjects who reported not having

been sick in the previous year (13.2%).

Both prevalence and average days indicate a higher risk among women: they report higher levels of absences in all

countries with few exceptions (Figure 30). This regularity is present in most of the literature, which explains the

difference with a number of biological and social reasons, such as pregnancy and the double burden posed by combining

work and family duties. The same figure notes the higher levels of absence in the northern, compared to the southern,

region of Europe, which again is consistent with previous findings (Bonato and Lusinyan, 2004).

Figure 30: Mean days of sickness absence, by country and gender

Source: EWCS 2010

Looking at occupational class, white-collar workers reported higher prevalence of sickness absence (above 40%)

compared to blue-collar workers (around 35%). Mean days of absence were highest among low-skilled blue-collar

workers and lowest among high-skilled white-collar workers (5.8 and 3.8 days respectively).

A factor that is often quoted as a driver of absenteeism is job insecurity: the higher the market pressure that individuals

feel, the less they will tend to be absent from the workplace. This general tendency actually emerges from the data:

prevalence of absence was highest among permanent employees (46%) and lowest among the self-employed (23%),

while employees with a temporary or no contract showed intermediate values (38% and 31%, respectively).

Regarding presenteeism, in the sample 41% male and 45% female workers reported to have worked while ill at least one

day in the previous 12 months. Considering both genders together, prevalence of presenteeism ranked highest in

Montenegro, followed by Slovenia, Malta, Denmark and Sweden (all well above 50%), and lowest in Italy, Portugal,

Poland and Bulgaria (23%–25%). Average days of presenteeism were 3.1 in the whole sample, again with a slightly

higher figure for women (3.4 days) than men (2.9 days) (see Figure 31).

Health and well-being at work

XK EL IE TR RO ME AL LV ES LU MT IT MK CY UK BG PT EE HU LT HR FR SK DE DK NL SE AT BE CZ PL SI FI NO

Women Men

Health and well-being at work

Figure 31: Mean days of presenteeism, by country and gender

Source: EWCS 2010

Prevalence of presenteeism was higher among high-skilled white-collar workers (around 50%), compared to the other

occupational classes (35%–38%), a pattern that was observed also for mean days of presenteeism.

Determinants of sickness absence and presenteeism

The approach now turns to a multivariate analysis in order to assess the role of working conditions among all the non-

health determinants that have been pointed out in the literature. The goal is to explain the prevalence of the two measures

that have been employed so far. However, a higher number of days has been chosen as a threshold in order to have a

more clear-cut representation of the phenomena. For sickness absence, the probability of having spent five or more days

of sickness absence in the previous year has been used, as suggested by Eurostat in the methodology for the European

Statistics on Accidents at Work (ESAW). For sickness presenteeism, the outcome was based on at least two days of

presence while ill, as in most previous research on sickness presenteeism (Aronsson et al, 2000; Aronsson and

Gustafsson, 2005; Elstad and Vabø, 2008; Bergstrom et al, 2009; Heponiemi et al, 2010).

Respondents who reported having been employed for less than one year in the actual company or organisation (8.9% of

the sample)32 were excluded from the analyses. Since they have a different exposure to the risk of being absent from

work, their inclusion would hamper a clear interpretation of the results. The resulting population for the analysis on

sickness absence was composed of 37,353 individuals, 52% of which were men, whereas the analysis on sickness

presenteeism included 32,554 people (51% men) (subjects reporting to have not been sick in the previous year were

excluded).

IT BG EL CY IE MT AL LV PT XK AT UK PL SE ES LU LT DE TR CZ NL DK HU FR EE MK SK BE NO HR RO FI ME SI

Women Men

32 As well as those with missing information in this item (2.5%). Individuals were also excluded who were employed in the armed

forces (less than 1%), plus those who reported their house as the main place of work (about 3%).

In the analyses, several sociodemographic and work-related characteristics were investigated as possible determinants

of sickness absence or sickness presenteeism, including organisational features and exposure to psychosocial,

environmental and ergonomic hazards. They are defined as in Chapter 4, with the addition of some specific determinants

detailed in the following box.

Sickness absence

Sickness absence for at least five days during the previous year was reported by 22.5% of men and 28.1% of women.

Among significant risk factors, several were consistent between genders, including poor general or mental health,

younger age, longer seniority, being a permanent employee or a low-skilled clerical worker, exposure to repetitive

movements of arm/hand and working full time (Table 12, first two columns).

Among men, the risk of sickness absence was also significantly increased among workers employed in firms with more

than 50 workers, and by exposure to discrimination, introduction of new technology, high emotional demand and second-

hand smoke, whereas very long work hours (more than 48 hours per week), high psychological demand, high supervisor

support and being responsible for the work of more than 10 people were associated with a decreased risk. Among

women, prevalence of sickness absence was also significantly higher among workers belonging to the public sector,

high-skilled manual workers (compared to high-skilled clerical workers) and for workers reporting job rotation implying

different skills or bullying. On the contrary, being a fixed-term worker (compared to permanent workers), standing and

high skill discretion were found to be associated with a reduced risk of sickness absence.

The effect of age on sickness absence still appears to be controversial in the literature (Voss et al, 2001; Mastekaasa,

2000; Taimela et al, 2007; Alavinia et al, 2009), and the reverse relationship observed here supports a small, but

significant, negative effect of increasing age on the risk of sickness absence in both genders, so that older age is linked

to a lower prevalence of sickness absence.

Health and well-being at work

Possible determinants of sickness absence and presenteeism

In the analyses, the following variables were investigated as possible determinants of sickness absence and

presenteeism, in addition to those discussed in Chapter 4:

nworking time autonomy (three categories: no control, control on either speed or breaks, control on both speed and

breaks);

npreferred working hours compared to the current situation (less, same, more than currently);

nresponsibility for the work of other people (0, 1–5, 6–10, >10 workers);

ndiscrimination (positive answer to any of seven yes/no questions);

nbullying, verbal abuse, threat, physical violence or sexual harassment at the workplace (all yes/no questions);

nhousehold composition (single without children, single parent, couple without children, couple with child(ren));

ntotal unpaid working hours per week (0, 1–3, 4–7, 8–14, 15–28, 29+);

nself-perceived general health (very good, good, fair, poor, very poor);

nmental health (WHO-5 indicator treated as a continuous variable);

nincome (income tertiles based on the distribution of income across Europe).

Health and well-being at work

The association with general and mental health (better health decreases the prevalence of sickness absence) was

expected, although it explained only a minor part of the variability in sickness absence of the sample. It has been actually

reported that sickness absence is only in part determined by illness or health conditions (Marmot et al, 1995; Andrea et

al, 2003; Farrel and Stam, 1988). In particular, although these factors were only partially consistent between genders, the

increased risk related to larger firm size, company seniority, permanent work and employment in the public sector

appears coherent with the results of previous studies, where higher rates of sickness absence were found among workers

employed in large firms (Barmby and Stephan, 2000; Voss et al 2001) or in the public administration (Ercolani, 2006;

Scoppa, 2010), and among permanent, compared to temporary, staff (Bourbonnais et al, 1992; Virtanen et al, 2003;

Bradley et al, 2007) and self-employed workers (Benavides et al, 2000; Scoppa, 2010). These work-related

characteristics have been interpreted by Ichino and Riphahn (2004) as associated with sickness absence because of their

relationship with employment protection, based on their analysis of different case studies in Germany and Italy.

In both genders a modest association with occupational social class was found. Furthermore, such differences were

scarcely influenced by adjustment for working conditions or health status. Also, no significant differences were observed

in any gender for the other indicators of socioeconomic status employed in the analyses, namely income and educational

attainment. This does not seem consistent with other reports showing an inverse gradient in sickness absence by

socioeconomic status, independently of the indicator used (Alexanderson et al, 1994; Feeney et al, 1998; Godin and

Kittel, 2004; Ala-Mursula et al, 2002; Fuhrer et al, 2002). The lower risk of sickness absence among subjects reporting

fewer than 20 work hours per week, compared to those working full time, may indicate that a shorter duration of work

is beneficial in reducing absenteeism, for example through increased time for private and family life and a consequent

improvement in work–life balance. The low risk of sickness absence observed for working more than 48 hours per week

seems instead related to an attitude of strong commitment toward one’s job, which would combine long work hours and

low sickness absence.

The absence of an association with shift work appears consistent with the results in the literature, given that most studies

did not find a significant association (Kleiven et al, 1998; Voss et al, 2001; Tüchsen et al, 2008; Eriksen et al, 2003),

although a few reported an increased risk (Bourbonnais et al, 1992; Morikawa et al, 2001).

Regarding physical factors in the workplace, exposure to various ergonomic hazards has been reported to increase

sickness absence rates, in particular heavy lifting and repetitive movements (Voss et al, 2001), manual material handling

(Alavinia et al, 2009) and awkward postures (Labriola et al, 2006; Alavinia et al, 2009), which appears partially

consistent with Eurofound findings.

Concerning psychosocial factors at work, the two dimensions more consistently associated with the risk of sickness

absence in the literature are low job control (North et al, 1996; Gimeno et al, 2004b; Godin and Kittel, 2004; Alavinia

et al, 2009; Arola et al, 2003) and low social support (Moreau et al, 2004; Nielsen et al, 2006; Piirainen et al, 2003;

Stansfeld et al, 1997). Of these, only the skill discretion component of job control was associated with sickness

absence in this study among women, and only support from supervisors among men. High strain (Gimeno et al,

2004b; Moreau et al, 2004), effort–reward imbalance (Head et al, 2007; Godin and Kittel, 2004) and high emotional

demand (Melchior et al, 2005) are other workplace psychosocial factors previously reported as risk factors of sickness

absence, among which only emotional demand was found to increase the risk of sickness absence in the present study,

and only among men.

It appears difficult to assess whether the moderate inverse association between sickness absence and psychological

demand observed among men is consistent with previous studies, because this relationship presents very conflicting

results in the literature: excluding studies focusing specifically on very long spells of absence (more than eight weeks),

some showed a protective effect of demand (North et al, 1996; Boedeker, 2001), most of them did not find an association

(Niedhammer et al, 1998; Andrea et al, 2003; Godin and Kittel, 2004; Moreau et al, 2004; Rugulies et al, 2007; Munch-

Hansen et al, 2008; Josephson et al, 2008; Roelen et al, 2009), whereas a few observed a moderate increase in risk

(Kivimäki et al, 2003; Gimeno et al, 2004b; Ala-Mursula et al, 2005; Rahuala et al, 2007).

The increased risk of sickness absence associated with discrimination (among men) and bullying (among women)

supports the association between sickness absence and bullying/threat of violence observed by other authors (Kivimäki

et al, 2000; Voss et al, 2001).

Although part of the observed associations between workplace hazards and sickness absence may be attributable to the

causal effect of these factors on health, this is not considered the main mechanism underlying the increased risk of

sickness absence for being exposed to occupational factors. It has been rather suggested that sickness absence is one of

the possible ways of coping with working activities that expose workers to high stress or high physical demands, through

the reduction of time of exposure to workplace hazards (Kristensen, 1991; Johansson and Palme, 1996). An alternative

interpretation of the positive association between absenteeism and unfavourable working conditions is provided by the

social exchange theory, according to which workers do not feel fully compensated by their wages for being exposed to

high levels of physical and psychosocial factors and shirk because they are not satisfied with their salary (Akerlof and

Yellen, 1986).

No association was found with work–life balance, nor with family characteristics, in spite of the fact that there is

consistent evidence in the literature about an increased risk of sickness absence for work–family conflicts (Goff et al,

1990; Hammer et al, 2003; Gignac et al 1996; Jansen et al, 2006), although it is less conclusive for the presence of

children in the household (Brooke and Price, 1989; Van den Heuvel, 1997; Mastekaasa, 2000).

Using national unemployment rates specific to age group and gender, which were available for 26 countries (the

elaboration on Eurostat’s Labour Force Survey, 2009), a significantly lower sickness absence prevalence was observed

for the highest quartile of unemployment compared to the lowest. This finding appears to be in agreement with studies

where sickness absence has been found to increase in periods with low unemployment (Virtanen et al, 2005) and to

decrease with economic recession and growing unemployment (Alexanderson, 1998). In both genders, prevalence of

sickness absence and sickness presenteeism were still significantly different among several countries after adjusting for

other risk factors significantly associated. This suggests that sickness absence and sickness presenteeism are strongly

influenced by factors acting at the national level, probably including cultural attitudes towards absenteeism, but also

national regulations on sickness absence, in particular those concerning wage replacement.33

Sickness presenteeism

Sickness presenteeism for at least two days during the previous 12 months was reported by 36.3% of men and 40.4% of

women. The prevalence of sickness presenteeism observed in the sample was lower than that reported by other

researchers using the same definition of presenteeism (two or more days of presence in the previous year), where

sickness presenteeism was found to be around 50% or above (Aronsson and Gustafsson, 2005; Hansen and Andersen,

2008; Elstad and Vabø, 2008; Bergstrom et al, 2009). Virtually all these studies were conducted in the Scandinavian

countries, where sickness presenteeism is expected to be more common, but, even limiting the comparison to the same

countries, in the sample the prevalence was still lower (44% for both genders).

Health and well-being at work

33 Results not reported in Table 12.

Health and well-being at work

As for sickness absence, older workers were at lower risk of sickness presenteeism (Table 12, second two columns), after

taking into account differences in health, consistently with previous reports (Aronsson et al, 2000; Aronsson and

Gustafsson, 2005; Hansen and Andersen, 2008). General perceived health was associated with presenteeism, but relative

risks of poor health were about half those observed for sickness absence, whereas the strength of association with mental

health was comparable to that for sickness absence. A negative association between health and sickness presenteeism has

been confirmed by several studies and it has been suggested that sickness presenteeism may be a proxy for debilitating

chronic diseases, which would affect work capacity of the individuals (Hansen and Andersen, 2008).

High-skilled clerical workers displayed a 20% higher risk of sickness presenteeism compared to subjects in other

occupational classes; this finding appears to be in contrast with the results of other studies, where no difference or a

higher risk of sickness presenteeism was found among workers in lower occupational classes or with lower educational

level (Hansen and Andersen, 2008; Heponiemi et al, 2010; Aronsson et al, 2000).

Among men, sickness presenteeism was mainly associated with a large set of psychosocial exposures, although each

showing a rather small increase or decrease in risk, including decision authority, skill discretion, psychological and

cognitive demand, need for hiding emotions, working with clients, supervisor support, rewards, verbal abuse and

work–life balance. Among women, fewer associations with psychosocial exposures were found, and some of them were

replaced in the model by others sharing a similar concept (work intensity instead of psychological demand,

discrimination instead of verbal abuse). Also, in both genders long working hours and shift work increased the risk of

sickness presenteeism.

A high risk of sickness presenteeism associated with high quantitative demand (defined as time pressure or overload)

has been observed also by other studies (Aronsson and Gustafsson, 2005; Hansen and Andersen, 2008), as well as a high

risk for long working hours and shift work (Hansen and Andersen, 2008). Regarding the control dimension, Aronsson

and Gustafsson (2005) and Johansson and Lundberg (2004) found a positive association with high control, whereas no

difference in risk was observed by Hansen and Andersen (2008) after adjusting for other factors. For other psychosocial

exposures, no results seem to be available in the literature; however, the associations found with demand for hiding

emotions among men and with emotional demand among women seem consistent with the high risk of sickness

presenteeism observed in sectors involving care, such as education and healthcare, where exposure to these factors is

very common. Occupations involving care of or provision of help to others may imply a tie with the client/patient/pupil,

which is believed to predispose workers to go to work despite illness (Aronsson et al, 2000). Among other findings, a

possible explanation of the negative association observed with rewards is that being free to stay at home when sick may

be perceived as a component of the reward dimension.

In general, the characteristics found associated with sickness presenteeism, especially high occupational class, long

working hours, high psychological and cognitive demand, high decision authority and skill discretion, seem to indicate

the profile of a high-grade, over-committed white-collar worker with high autonomy and who is very engaged with their

own job, which is characterised by complex and intense activities often involving a relationship with other people.

Presenteeism has been associated by Aronsson and Gustafsson (2005) with a personality feature which they defined as

‘individual boundarylessness’, meaning the difficulty people with this characteristic have in saying no to other people’s

demands, which appears also similar to the definition of ‘over-commitment’ proposed by Siegrist (1996). Nonetheless,

the positive association observed with exposure to work intensity, verbal abuse or discrimination, handling chemicals,

awkward postures and shift work seems to indicate that sickness presenteeism is also increased by several unfavourable

working conditions to which blue-collar workers are typically more exposed. It is plausible that workers employed in

more strenuous and hazardous jobs reported higher presenteeism because they have greater difficulties in performing

their work duties, when sick, compared to people employed in less exposed jobs.

Among personal risk factors, involvement in unpaid work activities, in great part represented by housework and care for

children, elders and disabled people, was found to increase the risk of presenteeism, in agreement with previous reports

of associations with number of children in the household or living with a sick spouse (Kristensen, 1991; Hansen and

Andersen, 2008).

The results presented here do not support previous observations suggesting a higher risk of presenteeism associated with

smaller firm size (Virtanen et al, 2002), higher levels of cooperation with colleagues (Grinyer and Singleton, 2000),

working non-standard hours and job insecurity (Virtanen et al, 2002).

Regarding job insecurity, it seems worth noting that no difference in sickness presenteeism was found between

permanent and fixed-term workers in this study, as in most other studies (Aronsson et al, 2000; Hansen and Andersen,

2008; Heponiemi et al, 2010). It has been suggested that the increased risk of sickness absence reported by several

authors among permanent workers, compared to temporary workers, may be actually attributable to higher presenteeism

among the latter. These results do not confirm this hypothesis, indicating that the observed differences in sickness

absence between permanent and temporary workers are unlikely to be explained by their differences in presenteeism.

Table 12: Relative risks of sickness absence (5+ days in previous year) and of sickness presenteeism (2+ days in previous

year), by gender

Health and well-being at work

Sociodemographic and work-related characteristics Sickness absence Sickness presenteeism

Men

Women

Men

Women

Age (10 years increase) 0.95 0.92 0.93 0.94

General health (ref. very good) 1 1 1 1

good 1.35 1.38 1.18 1.25

fair 1.71 1.78 1.41 1.49

bad 2.69 3.05 1.59 1.49

very bad 3.94 3.26 1.39 1.5

WHO-5 mental score (10 score increase, 100=high) 0.96 0.96 0.96 0.94

Occupational social class (ref. high-skilled clerical workers) 1 1 1 1

low-skilled clerical workers 1.31 1.15 0.80 0.92

high-skilled manual workers 1.05 1.26 0.80 0.85

low-skilled manual workers 1.03 1.08 0.85 0.88

Company seniority (ref. 0–4 years) 1 1

5–9 years 1.02 1.16

10+ years 1.15 1.12

Type of employment (ref. permanent employees) 1 1

self-employed 0.65 0.64

fixed-term or other contract 0.91 0.81

Sector (ref. private sector) 1 1

public sector 1.04 1.16

other 1.47 1.02

Firm size (ref. 1 worker) 1

2–9 workers 1.18

10–49 workers 1.22

50–249 workers 1.33

250+ workers 1.33

Health and well-being at work

Sociodemographic and work-related characteristics Sickness absence Sickness presenteeism

Men

Women

Men

Women

Shift work (10 score increase, 100=high) 1.04 1.03

Hours worked per week (ref. 35–40 hours) 1 1

<=20 hours 0.55 0.68

21–34 hours 1.17 0.98

41–47 hours 0.91 0.97

48+ hours 0.80 0.95

Preferred hours (ref. same as currently) 1

less than currently 1.14

more than currently 1.16

Total unpaid work hours per week (ref. = 0) 1 1

1–3 h/week 1.10 1.16

4–7 h/week 1.15 1.10

8–14 h/week 1.11 1.20

15–28 h/week 1.11 1.27

29+ h/week 1.16 1.18

Work–life balance (ref. low tertile) 1

middle tertile 0.95

high tertile 0.89

Second-hand smoke (ref. no) 1.32

Handling biological fluids or wastes (full-time exposure vs.no exposure) 0.84

Handling chemicals (full-time exposure vs. no exp.) 1.23

Awkward postures (full-time exposure vs. no exp.) 1.17

Standing (full-time exposure vs. no exp.) 0.88

Repetitive movements of arm/hand (ref. no) 1.20 1.17

Introduction of new technologies (ref. no) 1.10 1.13

Job rotation with different skills (ref. no) 1.14

Cognitive demand (ref. low tertile) 1 1

middle tertile 1.11 1.12

high tertile 1.19 1.08

Emotional demand (ref. low tertile) 1 1

middle tertile 1.05 1.16

high tertile 1.15 1.15

Psychological demand (ref. low tertile) 1 1

middle tertile 0.85 1.15

high tertile 0.88 1.16

Skill discretion (ref. low tertile) 1 1

middle tertile 0.91 1.03

high tertile 0.92 1.13

Responsibility for workers (ref. 0 workers) 1

1–5 workers 0.94

6–10 workers 0.90

>10 workers 0.77

Note: Four separate Poisson robust regression models, each including only significant variables with p<0.05 (in bold significant

modalities; in italics protective effects). RR = relative risk.

Health and well-being at work

Sociodemographic and work-related characteristics Sickness absence Sickness presenteeism

Men

Women

Men

Women

Decision authority (ref. low tertile) 1 1

middle tertile 1.21 1.18

high tertile 1.38 1.25

Supervisor support (ref. low tertile) 1 1 1

middle tertile 0.93 0.86 0.87

high tertile 0.88 0.77 0.86

Rewards (ref. low tertile) 1

middle tertile 0.84

high tertile 0.84

Need for hiding emotions (ref. low tertile) 1

middle tertile 1.12

high tertile 1.10

Discrimination (ref. no discrimination) 1.11

Bullying (ref. no bullying) 1.17

Working with clients (ref. low tertile) 1

middle tertile 1.14

high tertile 1.19

This report investigated the many relationships between the well-being and health of European workers and their

working conditions. First, it produced some descriptive evidence on the issue, which confirms the main findings of

already available literature, and highlights huge differences in the level of health and well-being across countries and

across individuals’ and job characteristics. The rest of the report is devoted to a close analysis of the main drivers – and

of some consequences – of those differences.

Regarding well-being, the findings show a clear relationship with several indicators of quality of work and employment.

Non-monetary measures of quality are of prominent importance. The employability of workers is also a key determinant,

particularly in times of recession, when subjective perception of job security is at its lowest. Furthermore, a negative

relationship between quality levels and variability of well-being was consistently found: once very good quality

conditions are achieved, individuals have high levels of well-being with few exceptions, while when facing bad quality

of work and employment, large differences emerge in the capacity to cope.

To perform a similar investigation on health and its determinants at work, it was necessary to draft a comprehensive set

of indicators of the psychosocial work environment, which has been identified as one of the most important risk factors

in contemporary and future society.

Among the several measures of health, specific health conditions were a particular focus, namely musculoskeletal

diseases and mental health conditions. In addition, occupational injuries were also a key focus, investigating as possible

key determinants individual characteristics (age and gender), human capital endowment (education, tenure, training),

formal definition of the job (industry, firm size, job contract, occupation, hours worked), physical hazards, factors related

to work organisation and the psychosocial work environment as implemented in the indicators.

Among the most interesting results is the fact that the formal definition of the job has no significant impact on many of

the health outcomes considered. This is a novel and important finding, which shows that once the analysis controls for

the actual content of the job (physical, psychosocial and organisational determinants), its formal definition becomes an

empty box. It is worth noting that, at the same time, for musculoskeletal diseases among white-collar workers, job

insecurity is a risk factor: while the actual contract (temporary or no contract at all) had no impact on them, the subjective

perception of precariousness has.

A similar result holds also for country differences: they consistently tend to vanish after controlling for all the above

determinants. This result reinforces the above statement, that it is the actual job content that determines physical health,

more than its formal setup or even the institutional environment (as no institutional, cultural or political regularity seem

to emerge in the ordering of countries).

Psychosocial dimensions reveal themselves to be a decisive factor, and not only, as might be expected, in cases of anxiety

or depression. High ‘psychological demand’, for instance, increases musculoskeletal diseases among white-collar

workers; high ‘skill discretion’ decreases them among all workers, while ‘decision authority’ increases them for both

blue-collar and white-collar workers. The positive role of rewards emerges as a protective factor for all health outcomes

considered and also decreases work accidents among all groups of workers. Indeed, work accidents – together with

musculoskeletal disease – also show clear associations with many physical hazards that can be measured in the EWCS,

such as environmental hazards, awkward postures and travelling for work.

A final part of the report is devoted to the other direction of causality. Poor health status has a direct impact on work in

terms of sickness absence. Many factors may however mediate the relationship, lengthening the absence or shortening

it. At the limit, there is the flip side of work absence, namely presenteeism, a behaviour that may entail short- to long-

term costs for both employers and employees.

Conclusions

Among significant risk factors for sickness absence, several are consistent between genders, including poor general or

mental health (although they explain only a minor part of the variability in sickness absences), seniority in the company

longer than 10 years and being a permanent employee. Among men, sickness absence is significantly increased by

exposure to discrimination and high emotional demand, highlighting again a crucial role for psychosocial factors.

Among women, sickness absence is higher among workers belonging to the public sector, and lower among fixed-term

contract workers. It is also higher among women reporting job rotation implying different skills, repetitive movements

of the arm/hand or bullying; lower for those enjoying high skill discretion; hence the psychosocial dimensions are

crucial, as for men.

Regarding presenteeism, for both genders it is highest in the Anglo-Saxon and the Scandinavian countries and lowest in

eastern and southern European countries. Some characteristics are more clearly associated with presenteeism, especially

high occupational class, long working hours, high psychological and cognitive demand, high decision authority and skill

discretion. These characteristics seem to indicate the profile of a high-grade, over-committed white-collar worker with

high autonomy and who is very engaged with their own job, which is characterised by complex and intense activities

often involving relationships with other people. Nonetheless, the positive association observed with exposure to work

intensity, verbal abuse or discrimination, handling chemicals, awkward postures and shift work seems to indicate that

presenteeism is also increased by several unfavourable working conditions. Typically more exposed to these conditions

are blue-collar workers and workers in health sector occupations involving the provision of care or help to others, and

that may imply a connection with the patient, which is believed to predispose workers to go to work despite illness.

This reinforces the central role that (quality of) work and employment plays directly and indirectly in relation to well-

being and growth. This calls for the development of a wider agenda to improve and monitor job quality, since it will

prove very important for the future of Europe.

Health and well-being at work

All Eurofound publications are available at www.eurofound.europa.eu.

Akerlof, G.A. and Yellen, J.L. (1986), Efficiency wage models of the labor market, Cambridge University Press,

Cambridge.

Ala-Mursula, L., Vahtera, J., Kivimäki, M., Kevin, M.V. and Pentti, J. (2002), ‘Employee control over working times:

associations with subjective health and sickness absences’, Journal of Epidemiology and Community Health, Vol. 56,

No. 4, pp. 272–278.

Ala-Mursula, L., Vahtera, J., Linna, A., Pentti, J. and Kivimäki, M. (2005), ‘Employee worktime control moderates the

effects of job strain and effort–reward imbalance on sickness absence: the 10-town study’, Journal of Epidemiology and

Community Health, Vol. 59, No. 10, pp. 851–857.

Alavinia, S.M., van den Berg, T.I., van Duivenbooden, C., Elders, L.A. and Burdorf, A. (2009), ‘Impact of work-related

factors, lifestyle, and work ability on sickness absence among Dutch construction workers’, Scandinavian Journal of

Work, Environment and Health, Vol. 35, No. 5, pp. 325–333.

Albisinni, M. and Ricci, G. (2007), Forze di Lavoro. Media 2007 [Labour Force, Media 2007], Istat, Rome, 2007.

Alesina, A., Di Tella, R. and MacCulloch, R. (2004), ‘Inequality and happiness: Are Europeans and Americans

different?’, Journal of Public Economy, Vol. 88, No. 9, pp. 2009–2042.

Alexanderson, K., Leijon, M., Akerlind, I., Rydh, H. and Bjurulf, P. (1994), ‘Epidemiology of sickness absence in a

Swedish county in 1985, 1986 and 1987. A three-year longitudinal study with focus on gender, age and occupation’,

Scandinavian Journal of Social Medicine, Vol. 22, No. 1, pp. 27–34.

Alexanderson, K. (1998), ‘Sickness absence: a review of performed studies with focused on levels of exposures and

theories utilized’, Scandinavian Journal of Social Medicine, Vol. 26, No. 4, pp. 241–249.

Andrea, H., Beursken, A.J.H.M., Metsemaker, J.M.F., van Amelsvoort, L.G.P.M., van den Brant, P.A. and van Schayck,

C.P. (2003), ‘Health problems and psychosocial work environment as predictors of long term sickness absence in

employees who visited the occupational physician and/or general practitioner in relation to work: a prospective study’,

Occupational and Environmental Medicine, Vol. 60, No. 4, pp. 295–300.

Ariens, G.A., van Mechelen, W., Bongers, P.M., Bouter, L.M. and van der Wal, G. (2001), ‘Psychosocial risk factors for

neck pain: a systematic review’, American Journal of Industrial Medicine, Vol. 39, No. 180–193.

Arola, H., Pitkänen, M., Nygård, C.H., Huhtala, H. and Manka, M.L. (2003), ‘The connection between age, job control

and sickness absences among Finnish food workers’, Occupational Medicine, Vol. 53, No. 3, pp. 229–230.

Aronsson, G., Gustafsson, K. and Dallner, M. (2000), ‘Sick but yet at work. An empirical study of sickness

presenteeism’, Journal of Epidemiology and Community Health, Vol. 54, No. 7, pp. 502–509.

Aronsson, G. and Gustafsson, K. (2005), ‘Sickness presenteeism: prevalence, attendance-pressure factors, and an outline

of a model for research’, Journal of Occupational and Environmental Medicine, Vol. 47, 958–966.

Bibliography

Artazcoz, L., Benach, J., Borrell, C. and Cortes, I. (2005), ‘Social inequalities in the impact of flexible employment on

different domains of psychosocial health’, Journal of Epidemiology and Community Health, Vol. 59, pp. 761–767.

Askitas, N. and Zimmermann, K.F. (2011), Health and well–being in the crisis, IZA Discussion Paper No. 5601, Institute

for the Study of Labour (IZA), Bonn.

Aust, B., Rugulies, R., Skakon, J., Scherzer, T. and Jensen, C. (2007), ‘Psychosocial work environment of hospital

workers: validation of a comprehensive assessment scale’, International Journal of Nursing Studies, Vol. 44, No. 5,

pp. 814–825.

Bakker, A., Demerouti, E. and Verbeke, W. (2004), Using the job demands–resources model to predict burnout and

performance, Human Resources Management, Vol. 43, No. 1, pp. 83–104.

Barmby, T. and Stephen, G. (2000), ‘Worker absenteeism: Why firm size may matter’, The Manchester School, Vol. 68,

No. 5, pp. 568–577.

Barros, A.J. and Hirakata, V.N. (2003), ‘Alternatives for logistic regression in cross-sectional studies: an empirical

comparison of models that directly estimate the prevalence ratio’, BMC Medical Research Methodology, Vol. 3, p. 21.

Beham, B., Drobnič, S. and Verwiebe, R. (2006), Literature review: Theoretical concepts and methodological

approaches of quality of life and work, Utrecht University, Utrecht.

Benavides, F.G., Benach, J., Diez-Roux, A.V. and Roman, C. (2000), ‘How do types of employment relate to health

indicators? Findings from the second European survey on working conditions’, Journal of Epidemiology and Community

Health, Vol. 54, No. 7, pp. 494–501.

Bergström, G., Bodin, L., Hagberg, J., Lindh, T., Aronsson, G. and Josephson, M. (2009), ‘Does sickness presenteeism

have an impact on future general health?’, International Archives of Occupational and Environmental Health, Vol. 82,

No. 10, pp. 1179–1190.

Berkman, L.F. and Syme, S.L. (1979), ‘Social networks, host resistance, and mortality: A nine-year follow-up study of

Alameda county residents’, American Journal of Epidemiology, Vol. 109, No. 2, pp. 186–204.

Bjorner, J.B. and Pejtersen, J.H. (2010), ‘Evaluating construct validity of the second version of the Copenhagen

Psychosocial Questionnaire through analysis of differential item functioning and differential item effect’, Scandinavian

Journal of Public Health, Vol. 38, No. 3 supplementary, pp. 90–105.

Boedeker, W. (2001), ‘Associations between workload and diseases rarely occurring in sickness absence data’, Journal

of Occupational and Environmental Medicine, Vol. 43, No. 12, pp. 1081–1088.

Bonato, L. and Lusinyan, L. (2004), Work absence in Europe, IMF Working Paper, European Department, 04/193, 2004.

Bonde, J.P. (2008), ‘Psychosocial factors at work and risk of depression: a systematic review of the epidemiological

evidence’, Occupational and Environmental Medicine, Vol. 65, No. 7, pp. 438–445.

Health and well-being at work

Bourbonnais, R., Vinet, A., Vézina, M. and Gingras, S. (1992), ‘Certified sick leave as a non-specific morbidity

indicator: a case-referent study among nurses’, British Journal of Industrial Medicine, Vol. 49, No. 10, pp. 673–678.

Bracke, P., Levecque, K. and Van de Velde, S. (2008), ‘The psychometric properties of the CES-D 8 depression inventory

and the estimation of cross-national differences in the true prevalence of depression’, in Harkness, J.A. et al (eds.),

International conference on survey methods in multinational, multiregional, and multicultural contexts (3MC), Wiley,

New York.

Bradley, S., Green, C. and Leeves, G. (2007), ‘Employment contracts and effort: why do temporary workers take less

absence?’, Lancaster University Management School, Working Paper 026/2007.

Brooke, P.P. and Price, J.L. (1989), ‘The determinants of employee absenteeism: an empirical test of a causal model’,

Journal of Occupational Psychology, Vol. 62, No. 1, pp. 1–19.

Bültmann, U., Kant, I.J., Schröer, C.A. and Kasl, S.V. (2002), ‘The relationship between psychosocial work

characteristics and fatigue and psychological distress’, International Archives of Occupational and Environmental

Health, Vol. 75, No. 4, pp. 259–266.

Cantril, H. (1965), The pattern of human concerns, Rutgers, University Press, New Brunswick, NJ.

Caplan, R., Cobb, S., French, J. and Harrison, R. (1975), Job demands and worker health: Main effects and occupational

differences, NIOSH, Washington DC.

Cheng, Y., Chen, C.W., Chen, C.J. and Chiang, T.L. (2005), ‘Job insecurity and its association with health among

employees in the Taiwanese general population’, Social Science and Medicine, Vol. 61, No. 1, pp. 41–52.

Clark, A.E. and Oswald, A.J. (1994), ‘Unhappiness and unemployment’, The Economic Journal, Vol. 104, No. 424,

pp. 648–665.

Clark, A.E., Frijters, P. and Shields, M.A. (2008), ‘Relative income, happiness, and utility: an explanation for the

Easterlin Paradox and other puzzles’, Journal of Economic Literature, Vol. 46, No. 1, pp. 95–144.

Clark, A.E. (2001), ‘What really matters in a job? Hedonic measurement using quit data’, Labour Economics, Vol. 8,

No. 2, pp. 223–242.

Clark, A.E. (2009), Work, jobs and well-being across the millennium, IZA Discussion Papers 3940, Paris.

Clausen, T., Nielsen, K., Carneiro, I.G. and Borg, V. (2012), ‘Job demands, job resources and long-term sickness absence

in the Danish eldercare services: a prospective analysis of register-based outcomes’, Journal of Advanced Nursing,

Vol. 68, No. 1, pp. 127–136.

Conceiçao, P. and Bandura, R. (2008), Measuring subjective wellbeing: A summary review of the literature, United

Nations Development Program, New York, available at:

http://web.undp.org/developmentstudies/docs/subjective_wellbeing_conceicao_bandura.pdf.

Cosmides, L. and Tooby, J. (1992), ‘Cognitive adaptations for social exchange’, in Barkow, J.H., Cosmides, L. and

Tooby, J. (eds.), The adapted mind: Evolutionary psychology and the generation of culture, Oxford University Press,

New York, pp. 163–228.

Da Costa, B.R. and Vieira, E.R. (2010), ‘Risk factors for work-related musculoskeletal disorders: A systematic review

of recent longitudinal studies’, American Journal of Industrial Medicine, Vol. 53, No. 3, pp. 285–323.

D’Errico, A., Cardano, M., Landriscina, T., Marinacci, C., Pasian, S., Petrelli, A. and Costa, G. (2011), ‘Workplace stress

and prescription of antidepressant medications: a prospective study on a sample of Italian workers’, International

Archives of Occupational and Environmental Health, Vol. 84, pp. 413–424.

D’Souza, R.M., Strazdins, L., Lim, L.L., Broom, D.H. and Rodgers, B. (2003), ‘Work and health in a contemporary

society, demands, control, and insecurity’, Journal of Epidemiology and Community Health, Vol. 57, pp. 849–854.

Dickinson, C.E., Campion, K., Foster, A.F., Newman, S.J., O’Rourke, A.M. and Thomas, P.G. (1992), ‘Questionnaire

development: an examination of the Nordic musculoskeletal questionnaire’, Applied Ergonomics, Vol. 23, pp. 197–201.

Diener, E. and Biswas-Diener, R. (2002), ‘Will money increase subjective well-being? A literature review and guide to

needed research’, Social Indicators Research, Vol. 57, pp. 119–169.

Diener, E. and Biswas-Diener, R. (2003), ‘Findings on subjective well-being’, paper presented at the workshop on

Measuring empowerment: cross-disciplinary perspectives at the World Bank on 4–5 February 2003, Washington DC.

Di Tella, R., MacCulloch, J.R. and Oswald, A.J. (2001), ‘Preferences over inflation and unemployment: Evidence from

surveys of happiness’, American Economic Review, Vol. 91, pp. 335–341.

Easterlin, R.A., McVey, L.A., Switek, M., Sawangfa, O. and Zweig, J.S. (2010), ‘The happiness-income paradox

revisited’, Proceedings of the National Academy of Sciences of the USA, Vol. 107, pp. 22463–22468.

Easterlin, R.A. (2005), ‘Diminishing marginal utility of income? Caveat Emptor’, Social Indicators Research, Vol. 70,

pp. 243–255.

Easterlin, R.A. (1974), ‘Does economic growth improve the human lot? Some empirical evidence’, in David, P.A. and

Reder M.W. (eds.), Nations and households in economic growth: Essays in honour of Moses Abramowitz, Academic

Press, New York and London, pp. 89–125.

Eller, N.H., Netterstrøm, B., Gyntelberg, F., Kristensen, T.S., Nielsen, F., Steptoe, A. and Theorell, T. (2009), ‘Work-

related psychosocial factors and the development of ischemic heart disease: a systematic review’, Cardiology Review,

Vol. 17, pp. 83–97.

Elstad, J.I. and Vabø, M. (2008), ‘Job stress, sickness absence and sickness presenteeism in Nordic elderly care’,

Scandinavian Journal of Public Health, Vol. 36, pp. 467–474.

Eng, A., 't Mannetje, A., McLean, D., Ellison-Loschmann, L., Cheng, S., Pearce, N. (2011), ‘Gender differences in

occupational exposure patterns’, Occupational and Environmental Medicine, Vol. 68, No. 12, pp. 888–894.

Health and well-being at work

Ercolani, M.G. (2006), UK Employees’ Sickness Absence: 1984–2005, Discussion Papers 06-02, Department of

Economics, University of Birmingham, Birmingham, UK.

Eriksen, W., Bruusgaard, D. and Knardahl, S. (2003), ‘Work factors as predictors of sickness absence: a three month

prospective study of nurses’ aides’, Occupational and Environmental Medicine, Vol. 60, pp. 271–278.

ETUC, UNICE, UEAPME and CEEP (2004), European framework agreement on work-related stress, Brussels,

available at http://www.etuc.org/IMG/pdf_Framework_agreement_on_work-related_stress_EN.pdf.

Eurofound (2002), Quality of work and employment in Europe: Issues and challenges, Foundation Paper No. 1 February

2002, Publications Office of the European Union, Luxembourg.

Eurofound, EWCO (2005), Work-related stress, Dublin.

Eurofound (2007a), Fifteen years of working conditions in the EU: Charting the trends, Publications Office of the

European Union, Luxembourg.

Eurofound (2007b), First European Quality of Life Survey: Quality of work and life satisfaction, Publications Office of

the European Union, Luxembourg.

Eurofound (2007c), Fourth European Working Conditions Survey, Publications Office of the European Union,

Luxembourg.

Eurofound, EWCO (2010a), Absence from work, Dublin.

Eurofound (2010b), Subjective well-being in Europe: Second European Quality of Life Survey, Publications Office of

the European Union, Luxembourg.

Eurofound (2011), Shifts in the job structure in Europe during the Great Recession, Publications Office of the European

Union, Luxembourg.

Eurofound (2012a), Fifth European Working Conditions Survey, Publications Office of the European Union,

Luxembourg.

Eurofound (2012b), Trends in job quality in Europe, Publications Office of the European Union, Luxembourg.

Eurofound (2012c), Third European Quality of Life Survey – Quality of life in Europe: Impacts of the crisis, Publications

Office of the European Union, Luxembourg.

European Agency for Safety and Health at Work (EU-OSHA) (2000a), Future occupational safety and health research

needs and priorities in the Member States of European Union, Publications Office of the European Union, Luxembourg.

European Agency for Safety and Health at Work (EU-OSHA) (2000b), Research on work-related stress, Publications

Office of the European Union, Luxembourg.

European Agency for Safety and Health at Work (EU-OSHA) (2009), OSH in figures: stress at work – facts and figures,

Publications Office of the European Union, Luxembourg.

European Commission (2001), ‘Employment and social policies: a framework for investing in quality’, COM/2001/313

final, Brussels.

European Commission (2002), ‘Adapting to change in work and society: a new Community strategy on health and safety

at work 2002–2006’, COM/2002/118 final, Brussels.

European Commission (2008), ‘Measuring the quality of employment in the EU’, in Employment in Europe 2008,

Publications Office of the European Union, Luxembourg.

European Commission (2009), ‘GDP and beyond: measuring progress in a changing world’, COM/2009/0433 final,

Brussels.

European Economic and Social Committee (2011), Opinion of the European Economic and Social Committee on the

‘Communication from the Commission to the Council and the European Parliament – Beyond GDP – Measuring

progress in a changing world’ COM/2009/0433 final, 19 January 2011, Official Journal of the European Union, Brussels.

Farrel, D. and Stam, C.L. (1988), ‘Meta-analysis of the correlates of employee attendance’, Human Relations, Vol. 41,

pp. 211–227.

Feeney, A., North, F., Head, J., Canner, R. and Marmot, M. (1998), ‘Socioeconomic and sex differentials in reason for

sickness absence from the Whitehall II Study’, Occupational and Environmental Medicine, Vol. 55, pp. 91–98.

Ferrie, J.E., Shipley, M.J., Marmot, M.G., Stansfeld, S. and Smith, G.D. (1998), ‘An uncertain future: the health effects

of threats to employment security in white-collar men and women’, American Journal of Public Health, Vol. 88,

pp. 1030–1036.

Ferrie, J.E., Shipley, M.J., Marmot, M.G., Stansfeld, S. and Smith, G.D. (1995), ‘Health effects of anticipation of job

change and non-employment: longitudinal data from the Whitehall II study’, BMJ, Vol. 311, pp. 1264 –1269.

Festinger, L. (1954), ‘A theory of social comparison processes’, Human Relations, Vol. 7, pp. 117–140.

Frese, M. and Zapf, D. (1994), ‘Action as the core of work psychology: a German approach’, in Triandis, H.C., Dunnette,

M.D. and Hough, L.M. (eds.), Handbook of Industrial and Organizational Psychology, 2nd ed., Vol. 4, pp. 271–340.

Frey, B.S. and Stutzer, A. (2002), Happiness and economics: how the economy and institutions affect well-being,

Princeton University Press, Princeton, NJ.

Frone, M.R., Russell, M. and Barnes, G.M. (1996), ‘Work–family conflict, gender, and health-related outcomes: A study

of employed parents in two community samples’, Journal of Occupational Health and Psychology, Vol. 1, 57–69.

Fuhrer, R., Shipley, M.J., Chastang, J.F., Schmaus, A., Niedhammer, I., Stansfeld, S.A., Goldberg, M. and Marmot, M.G.

(2002), ‘Socioeconomic position, health, and possible explanations: a tale of two cohorts’, American Journal of Public

Health, Vol. 92, pp. 1290–1294.

Gallie, D., White, M., Cheng, Y. and Tomlinson, M. (1998), Restructuring the employment relationship, Clarendon Press,

Oxford.

Health and well-being at work

Gignac, M.A.M., Kelloway, E.K. and Gottlieb, B.H. (1996), ‘The impact of caregiving on employment: a mediational

model of work–family conflict’, Canadian Journal of Aging, Vol. 15, pp. 525–542.

Gimeno, D., Benavides, F.G., Amick III, B.C., Benach, J. and Martínez, J.M. (2004a), ‘Psychosocial factors and work-

related sickness absence among permanent and non-permanent employees’, Journal of Epidemiology and Community

Health, Vol. 58, pp. 870–876.

Gimeno, D., Benavides, F.G., Benach, J. and Amick III, B.C. (2004b), ‘Distribution of sickness absence in the European

Union countries’, Occupational and Environmental Medicine, Vol. 61, pp. 867–869.

Glenn, N.D. and Weaver, C.N. (1981), ‘The contribution of marital happiness to global happiness’, Journal of Marriage

and Family, Vol. 43, pp. 161–168.

Godin, I. and Kittel, F. (2004), ‘Differential economic stability and psychosocial stress at work: associations with

psychosomatic complaints and absenteeism’, Social Science and Medicine, Vol. 58, pp. 1543–1553.

Goff, S.J., Mount, M.K. and Jamison, R.L. (1990), ‘Employer supported child care, work/family conflict, and

absenteeism: a field study’, Personnel Psychology, Vol. 43, No. 4, pp. 793–809.

Graham, C., Chattopadhyay, S. and Picon, M. (2010), ‘Adapting to adversity: Happiness and the 2009 economic crisis

in the United States’, Sociological Research, Vol. 77, pp. 715–748.

Green, F. (2011), ‘Unpacking the misery multiplier: How employability modiﬁes the impacts of unemployment and job

insecurity on life satisfaction and mental health’, Journal of Health Economics, Vol. 30, pp. 265–276.

Green, F. (2006), Demanding work. The paradox of job quality in the affluent economy, Princeton University Press,

Woodstock.

Greenhalgh, L. and Rosenblatt, Z. (1984), ‘Job insecurity: Toward conceptual clarity’, Academy of Management Review,

Vol. 3, pp. 438–448.

Greenhaus, J.H. and Beutell, N.J. (1985), ‘Sources of conflict between work and family roles’, Academy of Management

Review, Vol. 10, pp. 76–88.

Grinyer, A. and Singleton, V. (2000), ‘Sickness absence as risk-taking behaviour: a study of organisational and cultural

factors in the public sector’, Health, Risk and Society, Vol. 2, pp. 7–21.

Grün, C., Hauser, W. and Rhein, T. (2008), ‘Finding a job: Consequences for life satisfaction and interactions with job

quality. Evidence from German and British panel data’, conference presentation, 30th General Conference of the

International Association for Research in Income and Wealth, Slovenia.

Hammer, LB., Bauer, T.N. and Grandey, A.A. (2003), ‘Work–family conflict and work-related withdrawal behaviors’,

Journal of Business and Psychology, Vol. 17, pp. 419–436.

Hammer, T.H., Saksvik, P.O., Nytro, K., Torvatn, H. and Bayazit, M. (2004), ‘Expanding the psychosocial work

environment: Workplace norms and work–family conflict as correlates of stress and health’, Journal of Occupational

Health and Psychology, Vol. 9, pp. 83–97.

Hämmig, O., Gutzwiller, F. and Bauer, G.F. (2009), ‘Work–life conflict and associations with work- and nonwork-related

factors and physical and mental health outcomes: a nationally representative cross-sectional study in Switzerland’,

BioMed Central Public Health, Vol. 9, pp. 435.

Hansen. C.D. and Andersen, J.H. (2008), ‘Going ill to work – what personal circumstances, attitudes and work-related

factors are associated with sickness presenteeism’, Social Science and Medicine, Vol. 67, pp. 956–964.

Haynes, S.G. and Feinleib, M. (1980), ‘Women, work and coronary heart disease: prospective findings from the

Framingham heart study’, American Journal of Public Health, Vol. 70, pp. 133–141.

Head, J., Kivimäki, M., Siegrist, J., Ferrie, J.E., Vahtera, J., Shipley, M.J. and Marmot, M.G. (2007), ‘Effort–reward

imbalance and relational injustice at work predict sickness absence: the Whitehall II study’, Journal of Psychosomatic

Research, Vol. 63, pp. 433–440.

Hemp, P. (2004), ‘Presenteeism: at work – but out of it’, Harvard Business Review, Vol. 82, pp. 49–58.

Heponiemi, T., Elovainio, M., Pentti, J., Virtanen, M., Westerlund, H., Virtanen, P., Oksanen, T., Kivimäki, M. and

Vahtera, J. (2010), ‘Association of contractual and subjective job insecurity with sickness presenteeism among public

sector employees’, Journal of Occupational and Environmental Medicine, Vol. 52, pp. 830–835.

Hildebrandt, V.H., Bongers, P.M., Van Dijk, F.J.H., Kemper, H.C.G. and Dul, J. (2001), ‘Dutch musculoskeletal

questionnaire: description and basic qualities’, Ergonomics, Vol. 44, pp. 1038–1055.

Hochschild, A.R. (1983), The managed heart: commercialization of human feelings, University of California Press,

Berkley.

Hooftman, W.E., van der Beek, A.J., Bongers, P.M. and van Mechelen, W. (2005), ‘Gender differences in self-reported

physical and psychosocial exposures in jobs with both female and male workers’, Journal of Occupational and

Environmental Medicine, Vol. 47, pp. 244–252.

Hosmer, D.W. and Lemeshow, S. (2000), Applied logistic regression, 2nd ed., John Wiley and Sons, New York.

House, J.S., Landis, K.R. and Umberson, D. (1988), ‘Social relationships and health’, Science, Vol. 241, pp. 540–545.

Huppert, F. (2009), ‘Measuring well-being across Europe: Description of the ESS well-being module and preliminary

findings’, Social Indicators Research, Vol. 91, pp. 301–315.

Ichino, A. and Riphahn, R.T. (2004), ‘Absenteeism and employment protection: Three case studies’, Swedish Economic

and Political Review, Vol. 11, pp. 95–114.

Jansen, N.W., Kant, I.J., van Amelsvoort, L.G., Kristensen, T.S., Swaen, G.M. and Nijhuis, F.J. (2006), ‘Work–family

conflict as a risk factor for sickness absence’, Occupational and Environmental Medicine, Vol. 63, pp. 488–494.

Health and well-being at work

Johansson, G. and Lundberg, I. (2004), ‘Adjustment latitude and attendance requirements as determinants of sickness

absence or attendance. Empirical tests of the illness flexibility model’, Social Science and Medicine, Vol. 58, pp.

1857–1868.

Johansson, P. and Palme, M. (1996), ‘Do economic incentives affect work absence? Empirical evidence using Swedish

micro data’, Journal of Public Economics, Vol. 59, pp. 195–218.

Johnson, J.V. and Hall, E.M. (1988), ‘Job strain, work place social support, and cardiovascular disease: a cross-sectional

study of a random sample of the Swedish working population’, American Journal of Public Health, Vol. 78,

pp. 1336–1342.

Josephson, M., Lindberg, P., Voss, M., Alfredsson, L. and Vingård, E. (2008), ‘The same factors influence job turnover

and long spells of sick leave – a three-year follow-up study of Swedish nurses’, European Journal of Public Health,

Vol. 18, pp. 380–385.

Kabanoff, B. and O’Brien, G. (1980), ‘Work and leisure: a task attributes analysis’, Journal of Applied Psychology,

Vol. 65, pp. 596–609.

Kaplan, G.A., Salonen, J.T., Cohen, R.D., Brand, R.J., Syme, S.L. and Puska, P. (1988), ‘Social connections and

mortality from all causes and from cardiovascular disease: Prospective evidence from eastern Finland’, American

Journal of Epidemiology, Vol. 128, pp. 370–380.

Karasek, R.A. (1979), ‘Job demands, job decision latitude and mental strain: implications for job redesign’,

Administrative Science Quarterly, Vol. 24, pp. 285–308.

Karasek, R. and Theorell, T. (1990), Healthy work: Stress, productivity and the reconstruction of working life, Basic

Books, New York.

Karasek, R., Triantis, K. and Chaudhry, S. (1982), ‘Co-worker and supervisor support as moderators of associations

between task characteristics and mental strain’, Journal of Occupational Behaviour, Vol. 3, pp. 147–160.

Kawakami, N., Haratani, T. and Araki, S. (1992), ‘Effects of perceived job stress on depressive symptoms in blue-collar

workers of an electrical factory in Japan’, Scandinavian Journal of Work and Environmental Health, Vol. 18,

pp. 195–200.

Kelloway, E.K., Gottlieb, B.H. and Barham, L. (1999), ‘The source, nature, and direction of work and family conflict:

a longitudinal investigation’, Journal of Occupational Health and Psychology, Vol. 4, pp. 337–346.

Kinnunen, U. and Mauno, S. (1998), ‘Antecedents and outcomes of work–family conflict among employed women and

men in Finland’, Human Relations, Vol. 51, pp. 157–177.

Kivimäki, M., Elovainio, M., Vahtera, J. and Ferrie, J.E. (2003), ‘Organisational justice and health of employees:

prospective cohort study’, Occupational and Environmental Medicine, Vol. 60, pp. 27–34.

Kivimäki, M., Elovainio, M. and Vahtera, J. (2000), ‘Workplace bullying and sickness absence in hospital staff’,

Occupational and Environmental Medicine, Vol. 57, pp. 656–660.

Kivimäki, M., Head, J., Ferrie, J., Hemingway, H., Shipley, M., Vahtera, J. et al (2005), ‘Working while ill as a risk factor

for serious coronary events: the Whitehall II study’, American Journal of Public Health, Vol. 95, pp. 98–102.

Kivimäki, M., Vahtera, J., Thomson, L., Griffiths, A., Cox, T. and Pentti, J. (1997), ‘Psychosocial factors predicting

employee sickness absence during economic decline’, Journal of Applied Psychology, Vol. 82, pp. 858–872.

Kleiven, M., Bøggild, H. and Jeppesen, H.J. (1998), ‘Shift work and sick leave’, Scandinavian Journal of Work and

Environmental Health, Vol. 24, No. 3, pp. 128–133.

Kompier, M.A.J. and Marcelissen, F.H.G. (1990), Handboek werkstress: systematische aanpak voor de bedrijfspraktijk

[Manual on work stress: a systematic approach to business practice], NIA, Amsterdam.

Kristensen, T. (1991), ‘Sickness absence and work strain among Danish slaughterhouse workers: an analysis of absence

from work regarded as coping behaviour’, Social Science and Medicine, Vol. 32, pp. 15–27.

Kristensen, T.S., Borg, V. and Hannerz, H. (2002), ‘Socioeconomic status and psychosocial work environment: results

from a Danish national study’, Scandinavian Journal of Public Health, Vol. 59, pp. 41–48.

Kristensen, T.S., Hannerz, H., Høgh, A. and Borg, V. (2005), ‘The Copenhagen Psychosocial Questionnaire – a tool for

the assessment and improvement of the psychosocial work environment’, Scandinavian Journal of Work and

Environmental Health, Vol. 31, 438–449.

Kroenke, K., Spitzer, R.L. and Williams, J.B. (2001), ‘The PHQ-9: Validity of a brief depression severity measure’,

Journal of General Internal Medicine, Vol. 16, pp. 606–613.

Kuehner, C. (2003), ‘Gender differences in unipolar depression: an update of epidemiological findings and possible

explanations’, Acta Psychiatrica Scandinavica, Vol. 108, pp. 163–174.

Kuoppala, J., Lamminpää, A., Liira, J. and Vainio, H. (2008), ‘Leadership, job well-being, and health effects –

a systematic review and a meta-analysis’, Journal of Occupational and Environmental Medicine, Vol. 50, pp. 904–915.

Labriola, M., Lund, T. and Burr, H. (2006), ‘Prospective study of physical and psychosocial risk factors for sickness

absence’, Occupational Medicine (London), Vol. 56, pp. 469–474.

LaRocco, J., House, J. and French, J. (1980), ‘Social support, occupational stress and health’, Journal of Health and

Social Behavior, Vol. 21, pp. 202–218.

Lee, S., Colditz, G.A., Berkman, L.F. and Kawachi, I. (2004), ‘Prospective study of job insecurity and coronary heart

disease in US women’, Annals of Epidemiology, Vol. 14, pp. 24–30.

Leontaridi, R.M. and Sloane, P.J. (2001), Measuring the quality of jobs: Promotion prospects, low pay and job

satisfaction, LoWER [European Low-Wage Employment Research Network] working paper, No 7, April 2001,

Amsterdam.

Linton, S.J. (2001), ‘Occupational psychological factors increase the risk for back pain: a systematic review’, Journal of

Occupational Rehabilitation, Vol. 11, pp. 53–66.

Health and well-being at work

Livanos, I. and Zangelidis, A. (2010), Sickness absence: A pan-European study, MPRA Paper 22627, University Library

of Munich, Germany.

Louis, V.V. and Zhao, S. (2002), ‘Effects of family structure, family SES, and adulthood experiences on life satisfaction’,

Journal of Family Issues, Vol. 23, pp. 986–1005.

Maki, B.E. and McIlroy, W.E. (2007), ‘Cognitive demands and cortical control of human balance-recovery reactions’,

Journal of Neural Transmission, Vol. 114, pp. 1279–1296.

Markussen, S. (2007), Trade-offs between health and absenteeism in welfare states: striking the balance, Discussion

Paper 19/2007, Department of Economics, University of Oslo.

Marmot, M., Feeney, A., Shipley, M., North, F. and Syme, S.L. (1995), ‘Sickness absence as a measure of health status

and functioning: from the UK Whitehall II study’, Journal of Epidemiology and Community Health, Vol. 49,

pp. 124–130.

Marmot, M. (2005), ‘Social determinants of health inequalities’, Lancet, Vol. 365, pp. 1099–1104.

Martinez-Inigo, D., Totterdell, P., Alcover, C.M. and Holman, D. (2007), ‘Emotional labour and emotional exhaustion:

Interpersonal and intrapersonal mechanisms’, Work Stress, Vol. 21, pp. 30–47.

Maslow, A.H. (1954), Motivation and personality, Harper, New York.

Mastekaasa, A. (2000), ‘Parenthood, gender and sickness absence’, Social Science and Medicine, Vol. 50, 1827–1842.

Mastrangelo, G., Perticaroli, S., Camipo, G., Priolo, G., Leva, A., de Merich, D. et al (2008), ‘Condizioni di lavoro e di

salute e attività di prevenzione in un campione di 5000 lavoratori del Veneto esaminati con questionario mediante

intervista telefonica’ [Working conditions and health and prevention activities in a sample of 5000 workers in the Veneto,

by telephone interview questionnaire], Medicina del Lavoro (Industrial Medicine), Vol. 99, Suppl. 1, pp. 9–30.

Medalie, J., Kahn, H.A., Neufeld, H., Rise, E. and Goldbourt, U. (1973), ‘Five-year myocardial infarction incidence. II.

Association of single variables to age and birthplace’, Journal of Chronic Disease, Vol. 26, pp. 329–349.

Melchior, M., Krieger, N., Kawachi, I., Berkman, L.F., Niedhammer, I. and Goldberg, M. (2005), ‘Work factors and

occupational class disparities in sickness absence: findings from the GAZEL cohort study’, American Journal of Public

Health, Vol. 95, 1206–1212.

Messing, K. (1999), Integrating gender in ergonomic analysis: strategies for transforming women’s work, Trade Union

Technical Bureau for Health and Safety, Brussels.

Messing, K. (1998), One-eyed science: occupational health and women workers, Temple University Press, Philadelphia.

Michalos, A.C. (1985), ‘Multiple discrepancies theory (MDT)’, Social Indicators Research, Vol. 16, pp. 347–413.

Middleton, D.R. (1989), ‘Emotional style: The cultural ordering of emotions’, Ethos, Vol. 17, pp. 187–201.

Moreau, M., Valente, F., Mak, R., Pelfrene, E., de Smet, P., De Backer, G. et al (2004), ‘Occupational stress and incidence

of sick leave in the Belgian workforce: the Belstress study’, Journal of Epidemiology and Community Health, Vol. 58,

pp. 507–516.

Morikawa, Y., Miura, K., Ishizaki, M., Nakagawa, H., Kido, T., Naruse, Y. et al (2001), ‘Sickness absence and shift work

among Japanese factory workers’, Journal of Human Ergology (Tokyo), Vol. 30, pp. 393–398.

Morris, J.A. and Feldman, D.C. (1997), ‘Managing emotions in the workplace’, Journal of Managerial Issues, Vol. 9,

pp. 257–274.

Morris, J.A. and Feldman, D.C. (1996), ‘The dimensions, antecedents, and consequences of emotional labor’, Academy

of Management Review, Vol. 21, pp. 986–1010.

Munch-Hansen, T., Wieclaw, J., Agerbo, E., Westergaard-Nielsen, N. and Bonde, J.P. (2008), ‘Global measure of

satisfaction with psychosocial work conditions versus measures of specific aspects of psychosocial work conditions in

explaining sickness absence’, BioMed Centre Public Health, Vol. 8, p. 270.

Muñoz de Bustillo, L.R. and Fernandez, M.E. (2005), ‘Job satisfaction as an indicator of the quality of work’, Journal

of Socio-Economics, Vol. 34, pp. 656–673.

Muntaner, C., Li, Y., Xue, X., Thompson, T., Chung, H. and O’Campo, P. (2006), ‘County and organizational predictors

of depression symptoms among low-income nursing assistants in the USA’, Social Science and Medicine, Vol. 63,

pp. 1454–1465.

Myers, D.G. and Diener, E. (1995), ‘Who is happy?’, Psychological Science, Vol. 6, pp. 10–19.

Netemeyer, R.G., Boles, J.S. and McMurrian, R. (1996), ‘Development and validation of work–family conflict and

family–work conflict scales’, Journal of Applied Psychology, Vol. 81, pp. 400–410.

Netterstrøm, B., Conrad, N., Bech, P., Fink, P., Olsen, O., Rugulies, R. et al (2008), ‘The relation between work-related

psychosocial factors and the development of depression’, Epidemiology Review, Vol. 30, pp. 118–132.

New Economic Foundation (NEF) (2009), National accounts of well-being: bringing real wealth onto the balance sheet,

available at: http://www.nationalaccountsofwellbeing.org.

Niedhammer, I., Bugel, I., Goldberg, M., Leclerc, A. and Guéguen, A. (1998), ‘Psychosocial factors at work and sickness

absence in the Gazel cohort: a prospective study’, Occupational and Environmental Medicine, Vol. 55, pp. 735–741.

Nielsen, M.L., Rugulies, R., Smith-Hansen, L., Christensen, K.B. and Kristensen, T.S. (2006), ‘Psychosocial work

environment and registered absence from work: Estimating the etiologic fraction’, American Journal of Industrial

Medicine, Vol. 49, pp. 187–196.

Nordander, C., Ohlsson, K., Balogh, I., Rylander, L., Pålsson, B. and Skerfving, S. (1999), ‘Fish processing work: the

impact of two sex dependent exposure profiles on musculoskeletal health’, Occupational and Environmental Medicine,

Vol. 56, pp. 256–264.

Health and well-being at work

North, F.M., Syme, S.L., Feeney, A., Shipley, M. and Marmot, M. (1996), ‘Psychosocial work environment and sickness

absence among British civil servants: the Whitehall II study’, American Journal of Public Health, Vol. 86, pp. 332–340.

O’Driscoll, M.P., Ilgen, D.R. and Hildreth, K. (1992), ‘Time devoted to job and off-job activities, interrole conflict, and

affective experiences’, Journal of Applied Psychology, Vol. 77, pp. 272–279.

OECD (2010), OECD Employment Outlook 2010; Moving beyond the jobs crisis, OECD, Paris.

Origo, F. and Pagani, L. (2009), ‘Flexicurity and job satisfaction in Europe: The importance of perceived and actual job

stability for well-being at work’, Laboratory Economics, Vol. 16, pp. 547–555.

Parasuraman, S., Purohit, Y.S., Godshalk, V.M. and Beutell, N.J. (1996), ‘Work and family variables, entrepreneurial

career success, and psychological well-being’, Journal of Vocational Behavior, Vol. 48, pp. 275–300.

Paterniti, S., Niedhammer, I., Lang, T. and Consoli, S.M. (2002), ‘Psychosocial factors at work, personality traits and

depressive symptoms. Longitudinal results from the GAZEL Study’, British Journal of Psychiatry, Vol. 181,

pp. 111–117.

Pejtersen, J.H., Bjorner, J.B. and Hasle, P. (2010), ‘Determining minimally important score differences in scales of the

Copenhagen Psychosocial Questionnaire’, Scandinavian Journal of Public Health, Vol. 38, Suppl. 3, pp. 33–41.

Pejtersen, J.H., Kristensen, T.S., Borg, V. and Bjorner, J.B. (2010), ‘The second version of the Copenhagen Psychosocial

Questionnaire’, Scandinavian Journal of Public Health, Vol. 38, Suppl. 3, pp. 8–24.

Piccinelli, M. and Wilkinson, G. (2000), ‘Gender differences in depression. Critical review’, British Journal of

Psychiatry, Vol. 177, pp. 486–492.

Pigott, T.A. (1999), ‘Gender differences in the epidemiology and treatment of anxiety disorders’, Journal of Clinical

Psychiatry, Vol. 60, Suppl. 18, pp. 4–15.

Piirainen, H., Räsänen, K. and Kivimäki, M. (2003), ‘Organizational climate, perceived work-related symptoms and

sickness absence: a population-based survey’, Journal of Occupational and Environmental Medicine, Vol. 45,

pp. 175–184.

Poggi, A., Bizzotto, G., Devicienti, F., Vesan, P. and Villosio, C. (2011), Quality of life in Europe: Empirical evidence,

WALQING [Work and Life Quality in New and Growing Jobs] working paper 2011.4, Moncalieri, available at

http://www.academia.edu/1362005/walqing_working_paper_2011.4_Quality_of_Life_in_Europe.

Powdthavee, N. (2007), ‘Are there geographical variations in the psychological cost of unemployment in South Africa?’,

Social Indicators Research, Vol. 80, No. 3, pp. 629–652.

Rauhala, A., Kivimäki, M., Fagerstrom, L., Elovainio, M., Virtanen, M., Vahtera, J. et al (2007), ‘What degree of work

overload is likely to cause increased sickness absenteeism among nurses? Evidence from the RAFAELA patient

classification system’, Journal of Advanced Nursing, Vol. 57, pp. 286–295.

Roelen, C.A., Koopmans, P.C., Bültmann, U., Groothoff, J.W. and van der Klink, J.J. (2009), ‘Psychosocial work

conditions and registered sickness absence: a three-year prospective cohort study among office employees’,

International Archives of Occupational and Environmental Health, Vol. 82, pp. 1107–1113.

Rose, R., Hurst, M. and Herd, A. (1979), ‘Cardiovascular and endocrine responses to work and the risk of psychiatric

symptoms among air traffic controllers’, in Barrett, J. (ed.), Stress and Mental Disorder, Raven Press, New York.

Rugulies, R., Bültmann, U., Aust, B. and Burr, H. (2006), ‘Psychosocial work environment and incidence of severe

depressive symptoms: prospective findings from a 5-year follow-up of the Danish work environment cohort study’,

American Journal of Epidemiology, Vol. 163, pp. 877–887.

Rugulies, R., Christensen, K.B., Borritz, M., Villadsen, E., Bültmann, U. and Kristensen, T.S. (2007), ‘The contribution

of the psychosocial work environment to sickness absence in human service workers: Results of a 3-year follow-up

study’, Work Stress, Vol. 21, pp. 293–311.

Rugulies, R., Aust, B. and Pejtersen, J.H. (2010), ‘Do psychosocial work environment factors measured with scales from

the Copenhagen Psychosocial Questionnaire predict register-based sickness absence of 3 weeks or more in Denmark?’,

Scandinavian Journal of Public Health, Vol. 38, Suppl. 3, pp. 42–50.

Rutter, D.R. and Fielding, P.J. (1988), ‘Sources of occupational stress: an examination of British prison officers’, Work

Stress, Vol. 2, pp. 291–299.

Schabracq, M.J. (2003), ‘Organisational culture, stress and change’, in Schabracq, M.J., Winnubst, J.A.M. and Cooper,

C.L. (eds.), The handbook of work and health psychology, John Wiley and Sons, Chichester (UK), pp. 37–62.

Scoppa, V. (2010), ‘Worker absenteeism and incentives: Evidence from Italy’, Managerial and Decision Economics,

Vol. 31, pp. 503–515.

Seashore, S.E. (1974), ‘Job satisfaction as an indicator of the quality of employment’, Social Indicators Research,

Vol. 1, pp. 135–168.

Shields, M. (2006), ‘Stress and depression in the employed population’, Health Reports (Statistics Canada), Vol. 17,

pp. 11–29.

Siegrist, J. (1996), ‘Adverse health effects of high-effort/low-reward conditions’, Journal of Occupational Health and

Psychology, Vol. 1, pp. 27–41.

Siegrist, J., Starke, D., Chandola, T., Godin, I., Marmot, M., Niedhammer, I. and Peter, R. (2004), ‘The measurement of

effort–reward imbalance at work: European comparisons’, Social Science and Medicine, Vol. 58, pp. 1483–1499.

Spector, P.E. (1997), Job satisfaction: application, assessment, cause, and consequence, Sage Publications, Thousand

Oakes, CA.

Stansfeld, S.A. and Candy, B. (2006), ‘Psychosocial work environment and mental health – a meta-analytic review’,

Scandinavian Journal of Work and Environmental Health, Vol. 32, 443–462.

Health and well-being at work

Stansfeld, S.A., Rael, E.G., Head, J., Shipley, M. and Marmot, M. (1997), ‘Social support and psychiatric sickness

absence: a prospective study of British civil servants’, Psychological Medicine, Vol. 27, pp. 35–48.

Stevenson, B. and Wolfers, J. (2008), Economic growth and subjective well-being: Reassessing the Easterlin paradox,

NBER [National Bureau of Economic Research] Working Paper No. 14282, Cambridge, MA.

Stiglitz, J.E., Sen, A.K. and Fitoussi, J.P. (2009), Report by the Commission on the measurement of economic and social

progress, CMEPSP, Paris, available at http://www.stiglitz-sen-fitoussi.fr/en/index.htm.

Suls, J., Martin, R. and Wheeler, L. (2002), ‘Why, with whom and with what effect?’, Directions in Psychological

Science, Vol. 11, pp. 159–163.

Sverke, M., Gallagher, D.G. and Hellgren, J. (2000), ‘Alternative work arrangements: Job stress, well-being and pro-

organizational attitudes among employees with different employment contracts’, in Isaksson, K., Hogstedt, C., Eriksson,

C. and Theorell, T. (eds.), Health effects of the new labour market, Plenum, New York, pp. 145–167.

Swaen, G.M., Bültmann, U., Kant, I. and van Amelsvoort, L.G. (2004), ‘Effects of job insecurity from a workplace

closure threat on fatigue and psychological distress’, Journal of Occupational and Environmental Medicine, Vol. 46,

pp. 443–449.

Taimela, S., Läärä, E., Malmivaara, A., Tiekso, J., Sintonen, H., Justén, S. et al (2007), ‘Self-reported health problems

and sickness absence in different age groups predominantly engaged in physical work’, Occupational and

Environmental Medicine, Vol. 64, pp. 739–746.

Thorsen, S.V. and Bjorner, J.B. (2010), ‘Reliability of the Copenhagen Psychosocial Questionnaire’, Scandinavian

Journal of Public Health, Vol. 38, Suppl. 3, pp. 25–32.

Tüchsen, F., Christensen, K.B. and Lund, T. (2008), ‘Shift work and sickness absence’, Occupational Medicine

(London), Vol. 58, pp. 302–304.

Van den Heuvel, A. (1997), ‘Absence because of family responsibilities: an examination of explanatory factors’, Journal

of Family Economic Issues, Vol. 18, pp. 273–297.

Van Hoorn, A. (2007), ‘A short introduction to subjective well-being: its measurement, correlates and policy uses’,

conference presentation, Is happiness measurable and what do those measures mean for policy?, 2–3 April, Rome.

van Veldhoven, M. and Meijman, T.F. (1994), Questionnaire on the experience and assessment of work: VBBA—English

version, The Foundation for Quality in Occupational Health Care, Amsterdam.

Veenhoven, R. (2000), ‘Freedom and happiness: a comparative study in 44 nations in the early 1990s’, in Diener, E. and

Sush, E.M. (eds.), Culture and subjective well-being, MIT Press, Cambridge, MA.

Veenhoven, R. (1991), ‘Is happiness relative?’, Social Indicators Research, Vol. 24, pp. 1–34.

Veenhoven, R. (2006), ‘Rising happiness in nations,1946–2004. A replay to Easterlin’, Social Indicators Research,

Vol. 79, pp. 421–436.

Virtanen, M., Honkonen, T., Kivimäki, M., Ahola, K., Vahtera, J., Aromaa, A. et al (2007), ‘Work stress, mental health

and antidepressant medication findings from the Health 2000 Study’, Journal of Affective Disorders, Vol. 98,

pp. 189–197.

Virtanen, M., Kivimäki, M., Elovainio, M., Vahtera, J. and Cooper, C. (2002), ‘Contingent employment, health and

sickness absence’, Scandinavian Journal of Work and Environmental Health, Vol. 27, pp. 365–372.

Virtanen, M., Kivimäki, M., Elovainio, M., Vahtera, J. and Ferrie, J.E. (2003), ‘From insecure to secure employment:

changes in work, health, health-related behaviours, and sickness absence’, Occupational and Environmental Medicine,

Vol. 60, pp. 948–953.

Virtanen, M., Kivimäki, M., Elovainio, M., Virtanen, P. and Vahtera, J. (2005), ‘Local economy and sickness absence:

prospective cohort study’, Journal of Epidemiology and Community Health, Vol. 59, pp. 973–978.

Voss, M., Floderus, B. and Diderichsen, F. (2001), ‘Physical, psychosocial, and organisational factors relative to sickness

absence: a study based on Sweden Post’, Occupational and Environmental Medicine, Vol. 58, pp. 178–184.

Wang, J. (2004), ‘Perceived work stress and major depressive episodes in a population of employed Canadians over 18

years old’, Journal of Nervous and Mental Disease, Vol. 192, pp. 160–163.

Ware, J.E., Kosinski, M. and Keller, S.D. (1996), ‘A 12-item short-form health survey: construction of scales and

preliminary tests of reliability and validity’, Medical Care, Vol. 34, pp. 220–233.

Warr, P. (1994), ‘A conceptual framework for the study of work and mental health’, Work Stress, Vol. 8, pp. 84–97.

World Health Organization (1990), ‘Diabetes mellitus in Europe: a problem at all ages and in all countries. A model for

prevention and self care’, (Meeting proceedings), Giornale Italiano di Diabetologia e Metabolia, Vol. 10.

World Health Organization and Quality of Life Group (WHO-QOL Group) (1995), ‘The World Health Organization

Quality of Life Assessment (WHOQOL): Position paper from the World Health Organization’, Social Science and

Medicine, Vol. 41, pp. 1403–1409.

Wunsch, G., Duchene, J., Thiltges, E. and Salhi, M. (1996), ‘Socioeconomic differences in mortality: a life course

approach’, European Journal of Population, Vol. 12, pp. 167–185.

Zapf, D., Vogt, C., Seifert, C., Mertini, H. and Isic, A. (1999), ‘Emotion work as a source of stress: the concept and

development of an instrument’, European Journal of Work and Organizational Psychology, Vol. 8, pp. 371–400.

Health and well-being at work

The European Working Conditions Survey (EWCS), established in 1990, is one of the few sources of information

providing an overview of working conditions in Europe for the purposes of:

nassessing and quantifying working conditions of both employees and the self-employed across Europe on a

harmonised basis;

nanalysing relationships between different aspects of working conditions;

nidentifying groups at risk and issues of concern, as well as progress made;

nmonitoring trends by providing homogeneous indicators on these issues;

ncontributing to European policy development on quality of work and employment issues.

The EWCS was carried out in 1991, 1995, 2000 (with an extension to the then-candidate countries in 2001 and 2002),

2005 and 2010. The growing range of countries covered by each wave reflects the expansion of the European Union.

The first wave in 1991 covered only 12 countries, the second wave in 1995 covered 15 countries, and from the third

wave in 2000–2002 onwards, all 27 current EU Member States were included. Other countries covered by the survey

include Turkey (in 2002, 2005 and 2010), Croatia and Norway (in 2005 and 2010), Switzerland (in 2005), and Albania,

Kosovo, Montenegro and the former Yugoslav Republic of Macedonia (in 2010).

The fifth EWCS

The fieldwork for the fifth EWCS was carried out between January and June of 2010.34 In total, 43,816 face-to-face

interviews were carried out, with workers in 34 European countries answering questions on a wide range of issues

regarding their employment situation and working conditions.

The target population consisted of all residents in the 34 countries aged 15 or older (aged 16 or older in Norway, Spain

and the UK) and in employment at the time of the survey. People were considered to be in employment if they had

worked for pay or profit for at least one hour in the week preceding the interview (ILO definition).

The scope of the survey questionnaire has widened substantially since the first wave, aiming to provide a comprehensive

picture of the everyday reality of men and women at work. Consequently, the number of questions and issues covered

in the survey has expanded in each subsequent wave. By retaining a core of key questions, the survey allows for

comparison over time. By using the same questionnaire in all countries, the survey allows for comparison across

countries.

The main topics covered in the questionnaire for the fifth EWCS were job context, working time, work intensity,

physical factors, cognitive factors, psychosocial factors, violence, harassment and discrimination, work organisation,

skills, training and career prospects, social relationships, work–life balance and financial security, job fulfilment, and

health and well-being.

Annex: The European Working Conditions

Survey series

34 Fieldwork continued until 17 July 2010 in Belgium, due to the extended sample size, and until 29 August 2010 in Norway, due to

organisational issues.

New questions were introduced in the fifth wave to enable more in-depth analysis of psychosocial risks, workplace social

innovation, precarious employment and job security, place of work, work–life balance, leadership styles, health, and the

respondent’s household situation. The questionnaire also included new questions addressed specifically to self-employed

workers (such as financial security). Gender mainstreaming has been an important concern when designing the

questionnaire. Attention has been paid to the development of gender-sensitive indicators as well as to ensuring that the

questions capture the work of both men and women. Revisions to the questionnaire are developed in cooperation with

the tripartite stakeholders of Eurofound.

Sample

In each country, a multistage, stratified random sampling design was used. In the first stage, primary sampling units

(PSUs) were sampled, stratifying according to geographic region (NUTS 2 level or below) and level of urbanisation.

Subsequently, households in each PSU were sampled. In countries where an updated, high-quality address or population

register was available, this was used as the sampling frame. If such a register was not available, a random route procedure

was applied. In the fifth EWCS, for the first time, the enumeration of addresses through this random route procedure was

separated from the interviewing stage. Finally, a screening procedure was applied to select the eligible respondent within

each household.

The target number of interviews was 1,000 in all countries, except Slovenia (1,400), Italy, Poland and the UK (1,500),

Germany and Turkey (2,000), France (3,000) and Belgium (4,000). The Belgian, French and Slovenian governments

made use of the possibility offered by Eurofound to fund an addition to the initial sample size.

Fieldwork outcome and response rates

The interviews were carried out face to face in the respondents’ homes. The average duration of the interviews was 44

minutes. The overall response rate for the fifth wave was 44%, but there is considerable variation in response rates

between countries, varying between 31% in Spain and 74% in Latvia.

Weighting

Weighting was applied to ensure that results based on the fifth EWCS data could be considered representative for

workers in Europe.

nSelection probability weights (or design weights): To correct for the different probabilities of being selected for the

survey associated with household size. People in households with fewer workers have a greater chance of being

selected into the sample than people in households with more workers.

nPost-stratification weights: To correct for the differences in the willingness and availability to participate in the

survey between different groups of the population. These weights ensure that the results accurately reflect the

population of workers in each country.

nSupra-national weights: To correct for the differences between countries in the size of their workforce. These

weights ensure that larger countries weigh heavier in the EU-level results.

Quality assurance

Each stage of the fifth EWCS was carefully planned, closely monitored and documented, and specific controls were put

in place. For instance, the design phase paid close attention to information gathered in a data user survey on satisfaction

with the previous wave and on future needs, and an assessment was made of how the survey could better address the

topics that are central to European policymaking.

Health and well-being at work

In order to ensure that the questions were relevant and meaningful for stakeholders as well as respondents in all European

countries, the questionnaire was developed by Eurofound in close cooperation with a questionnaire development expert

group. The expert group included members of the Foundation’s Governing Board, representatives of the European Social

Partners, other EU bodies (the European Commission, Eurostat and the European Agency for Safety and Health at

Work), international organisations (the OECD and the ILO), national statistical institutes, as well as leading European

experts in the field.

Access to survey datasets

The Eurofound datasets and accompanying materials are stored with the UK Data Archive (UKDA) in Essex, UK and

promoted online via the Economic and Social Data Service (ESDS) International.

The data is available free of charge to all those who intend to use it for non-commercial purposes. Requests for use for

commercial purposes will be forwarded to Eurofound for authorisation.

In order to download the data, you must register with the ESDS if you are not from a UK university or college. For more

information, please consult the ESDS page on how to access data.

Once you are registered, the quickest way to find Eurofound data is to open the Catalogue search page, select Data

Creator/Funder from the first drop-down list and enter in the words ‘European Foundation’ in the adjacent search

box. Once Eurofound’s surveys are listed, you can click on the name of the relevant survey for more information and

download it using your user name and password.

For more information

The overview report as well as detailed information and analysis from the EWCS are available on the Eurofound

website at www.eurofound.europa.eu. This information is updated regularly.

For further queries, please contact Sophia MacGoris in the Working Conditions and Industrial Relations unit.

European Foundation for the Improvement of Living and Working Conditions (Eurofound), Wyattville Road,

Loughlinstown, Dublin 18, Ireland. Email: smg@eurofound.europa.eu.

EF/13/02/EN

Scar on my Heart. Effects of Unemployment Experiences on Coronary Heart Disease

Article

Full-text available

Nov 2015
INT J MANPOWER

Purpose The purpose of this paper is to study the impact of unemployment on coronary heart diseases (CHD) in Italy on a sample of male manual workers in the private sector. Design/methodology/approach The authors investigate the association between CHD and different unemployment experiences (ever unemployed; short, mid and long cumulative unemployment), exploiting a large Italian administrative database on careers and health. The study design is based on the balancing of individuals' characteristics during a 12-year pre-treatment period; the measurement of unemployment occurrence during a seven-year treatment period; the observation of CHD occurrence during a five-year follow up. The workers characteristics and the probability of receiving the treatment are balanced by means of propensity score matching. Standard diagnostics on the balancing assumption are discussed and satisfied, while the robustness to violations of the unconfoundedness assumption is evaluated by a simulation-based sensitivity analysis. Findings The authors find a significant increase of CHD probability was found among workers who experience more than three years of unemployment (relative risks (RR)=1.91, p <0.1), and among those who exit unemployment starting a self-employment activity (RR=1.70, p <0.1). Using different selections of the study population, a clear pattern emerges: the healthier and more labour market attached are workers during pre-treatment, the greater is the negative impact of long-term unemployment on health (RR=2.79, p <0.01). Originality/value The very large representative sample ( n =69,937) and the deep longitudinal dimension of the data (1985-2008) allowed the authors to minimize the risks of health selection and unemployment misclassification. Moreover, the adopted definition of unemployment corrected some undercoverage and misclassification issues that affect studies based on a purely administrative definition and that treat unemployment as a unique career event disregarding the duration of the experience.

Reference intervals of work ability and productivity loss and their use in patients with inflammatory rheumatic and musculoskeletal diseases

Article

Full-text available

Jan 2025

Objectives To establish reference intervals (RIs) for work ability, at-work productivity loss and overall productivity loss in the general working population and to compare work ability and at-work productivity loss of patients with inflammatory rheumatic and musculoskeletal disease (iRMD) with this population. Methods Cross-sectional analysis among patients with iRMDs and population controls without iRMDs having paid work and participating in a Dutch cohort study. They reported on three work outcomes: work ability (0–10), at-work productivity loss and overall productivity loss (0%–100%). A generalised additive model for location, shape and scale parameters was used to establish age-specific RIs and percentile curves for controls. The proportion of patients and controls below each percentile curve was compared. Results 413 controls were included; 73% female, with mean age 53 (SD 10) years, 60% had high education, mean work ability was 8.7 (1.6), at-work productivity loss 6.3% (7.2) and overall work productivity loss 11% (25.6). Percentile curves illustrated that work ability and at-work/overall work productivity loss were worse with increasing age. For instance, for work ability, the 95% RI for 22 to 29-year individuals was 5.9–10, while for individuals between 50 and 59 years, it was 4.9–9.1. Patients compared with controls had worse work outcomes, especially for at-work and overall productivity loss. Conclusion Work ability and productivity are not perfect in the general population, based on the newly developed RIs for the three work outcomes. This calls for caution to not overestimate the iRMD impact on work outcomes. Nevertheless, iRMD patients have worse work ability and higher work productivity loss, compared with controls.

Absenteeism, unemployment and employment protection legislation: evidence from Italy

Article

Full-text available

Dec 2014

Efficiency wages theories argue that the threat of firing, coupled with a high unemployment rate, is a mechanism that discourages employee shirking in asymmetric information contexts. Our empirical analysis aims to test the role of unemployment as a worker discipline device, considering the different degree of job security offered by the Italian Employment Protection Legislation to workers employed in small and large firms. Controlling for a number of individual and firm characteristics, we investigate the relationship between worker’s absences which act as a proxy for employee shirking and local unemployment rate (at the provincial level). We find a strong negative association between unemployment and absenteeism rate, larger in magnitude in small firms due presumably to a significantly lower protection from dismissals in these firms. As an indirect test of the role of unemployment as worker discipline device, we show that public sector employees, almost impossible to fire, do not react to the local unemployment. JEL codes: J41; M51; J45

Trends and Patterns in Occupational Health and Safety in Ireland

Book

Full-text available

May 2015

Is Sickness Presenteeism a Risk Factor for Depression? A Danish 2-Year Follow-Up Study

Article

May 2014
J OCCUP ENVIRON MED

Objective: To examine the prospective association between sickness presenteeism (SP), that is, working while ill, and the onset of depression. Methods: We carried out a two-wave (2006 to 2008) questionnaire-based study among 1271 employees from 60 Danish workplaces. Sickness presenteeism was assessed by asking participants to report the number of days that they went to work despite illness in the preceding year. Results: Multivariate logistic regression revealed that, after controlling for several health-related variables and other relevant confounders, reporting 8 or more days of SP was associated with an increased risk of depression among initially nondepressed participants (odds ratio, 2.45; 95% confidence interval, 1.06 to 5.64). No significant sex-related differences were observed in this relationship. Conclusion: Adding to previous evidence on the health effects of SP, this study suggests that working while ill may also be a significant risk factor for the development of depression.

Freedom and happiness: A comparative study in 44 nations in the early 1990s

Article

Full-text available

Jan 2000

Ruut Veenhoven

Sources of conflict between work and family roles, Academy of Management Review, Vol

Article

Full-text available

Jan 1985

An examination of the literature on conflict between work and family roles suggests that work-family conflict exists when: (a) time devoted to the requirements of one role makes it difficult to fulfill requirements of another; (b) strain from participation in one role makes it difficult to fulfill requirements of another; and (c) specific behaviors required by one role make it difficult to fulfill the requirements of another. A model of work-family conflict is proposed, and a series of research propositions is presented.

'Job demands, decisions latitude and mental strain: Implications for job redesign'

Article

Jan 1979

Robert Karasek

Who is happy?

Article

Jan 1995

Economic Growth and Subjective Well-Being: Reassessing the Easterlin Paradox

Article

Jan 2008

Managing emotions in the workplace

Article

Jan 1997

Job demands, job decision latitude, and mental strain: Implications for job design

Article

Jan 1979

Robert Karasek

Nations and Households in Economic Growth: Essays in Honor of Moses Abramovitz

Article

Jan 1974

RICHARD A. EASTERLIN

Publisher Summary This chapter discusses the association of income and happiness. The basic data consist of statements by individuals on their subjective happiness, as reported in thirty surveys from 1946 through 1970, covering nineteen countries, including eleven in Asia, Africa, and Latin America. Within countries, there is a noticeable positive association between income and happiness—in every single survey, those in the highest status group were happier, on the average, than those in the lowest status group. However, whether any such positive association exists among countries at a given time is uncertain. Certainly, the happiness differences between rich and poor countries that one might expect on the basis of the within-country differences by economic status are not borne out by the international data. Similarly, in the one national time series studied, for the United States since 1946, higher income was not systematically accompanied by greater happiness. As for why national comparisons among countries and over time show an association between income and happiness that is so much weaker than, if not inconsistent with, that shown by within-country comparisons, a Duesenberry-type model, involving relative status considerations as an important determinant of happiness, is suggested.

The Handbook of Work and Health Psychology: Second Edition

Book

Dec 2004

Workplace health is now recognised as having major legal, financial and efficiency implications for organizations. Psychologists are increasingly called on as consultants or in house facilitators to help design work processes, assess and counsel individuals and advise on change management. The second edition of this handbook offers a comprehensive, authoritative and up-to-date survey of the field with a focus on the applied aspects of work and health psychology. An unrivalled source of knowledge and references in the field, for students and academics, this edition also reflects the need to relate research to effective and realistic interventions in the workplace. Editors are outstanding leaders in their fields Focuses on linking research to practice Over 50% new chapters. New topics include Coping, The Psychological Contract and Health, Assessment and Measurement of Stress and Well-Being, the Effects of Change, and chapters of Conflict and Communication.

Applied Logistic Regression

Article

Jan 1994

Eurofound. Health and Well-being at Work: A Report Based on the Fifth European Working Conditions Survey

Figures

Recommended publications

Is Knowledge Work Better for Us? Knowledge Workers, Good Work, and Well-Being

EU-OSHA — European Agency for Safety and Health at Work | Summary Report of the Second European Surv...

Survey report The importance of people management Analysis of its impact on employees

Percentile Model and Usual Gait Speed Reference Values in Older Adults: A Cross-Sectional Secondary...