European Journal of Public Health, 1–5
? The Author 2013. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.
The effect of the late 2000s financial crisis on suicides
in Spain: an interrupted time-series analysis
James A. Lopez Bernal1, Antonio Gasparrini2, Carlos M. Artundo3, Martin McKee1
1 Department of Public Health and Policy, London School of Hygiene and Tropical Medicine, London WC1H 9SH, UK
2 Department of Medical Statistics, London School of Hygiene and Tropical Medicine, London WC1H 9SH, UK
3 Andalusian School of Public Health (ASPH), Granada, Spain
Correspondence: James A. Lopez Bernal, Department of Public Health and Policy, London School of Hygiene and
Tropical Medicine, 15-17 Tavistock Place, London WC1H 9SH, UK, Tel: +44 20 7927 2106, Fax: +44 20 7927 2701,
Background: The current financial crisis is having a major impact on European economies, especially that of Spain.
Past evidence suggests that adverse macro-economic conditions exacerbate mental illness, but evidence from the
current crisis is limited. This study analyses the association between the financial crisis and suicide rates in Spain.
Methods: An interrupted time-series analysis of national suicides data between 2005 and 2010 was used to
establish whether there has been any deviation in the underlying trend in suicide rates associated with the
financial crisis. Segmented regression with a seasonally adjusted quasi-Poisson model was used for the analysis.
Stratified analyses were performed to establish whether the effect of the crisis on suicides varied by region, sex
and age group. Results: The mean monthly suicide rate in Spain during the study period was 0.61 per 100000 with
an underlying trend of a 0.3% decrease per month. We found an 8.0% increase in the suicide rate above this
underlying trend since the financial crisis (95% CI: 1.009–1.156; P=0.03); this was robust to sensitivity analysis.
A control analysis showed no change in deaths from accidental falls associated with the crisis. Stratified analyses
suggested that the association between the crisis and suicide rates is greatest in the Mediterranean and Northern
areas, in males and amongst those of working age. Conclusions: The financial crisis in Spain has been associated
with a relative increase in suicides. Males and those of working age may be at particular risk of suicide associated
with the crisis and may benefit from targeted interventions.
macroeconomic changes may increase suicides.1
associations have been found between economic instability and
both mental ill health and suicide,2–8with past economic crises,
such as the collapse of the Soviet Union and the Southeast Asian
economic crisis in the late 1990s, associated with increases in
suicides.9,10The current global financial crisis is widely regarded
as the worst since the Great Depression of the 1930s and has had
a severe deleterious effect on the Spanish economy, culminating in
a request for Eurozone support for its banking sector.11Following a
decade of expansion, economic growth began to slow in 2007
and gross domestic product (GDP) began to contract from the
second quarter of 2008, ushering in a recession lasting seven
2007 and 2012, unemployment trebled from 8% to 24%
(Supplementary Appendix 2), reaching the highest rate in the
European Union, with the greatest increases seen in young males
(age 20–34): from 11% to 50%.13Since the onset of the financial
crisis, there have been major cuts to social spending. Savings
have included redundancies and salary reductions for health
care personnel, changes to drug-prescribing policies, closure of
facilities, reductions in opening times and delays in payments to
Evidence on the influence of the current financial crisis on mental
health, either in Spain or elsewhere, is limited, largely because delays
in data availability mean that the full impact cannot yet be assessed.
The lack of any sign that the crisis is ending makes it important that
its effects be understood so that mitigating interventions can be
implemented. One previous study has formally examined the effect
of the financial crisis on suicides, finding that observed suicide
counts in England were above those that would have been
n his seminal 1897 work, Le Suicide, Durkheim proposed that
expected based on underlying trends, particularly amongst men.17
There have also been reported increases in suicides in Greece, Italy
and the European Union in general and possible increases in mental
illness in Spain and Greece, although underlying trends were not
accounted for.18–24A further study used joinpoint regression as
part of their analysis of several mental health indicators in South
Australia but found no change in trends of any psychological distress
measures associated with the crisis.25
In this study, we investigate the relationship between the current
financial crisis and suicides in Spain using an interrupted time-series
analysis. We also evaluate how this relationship varies by geograph-
ical area, sex and age.
An interrupted time-series analysis was used to compare suicide
rates before the financial crisis with those subsequently. A
financial crisis may impact on health through a variety of
mechanisms, including job loss (or anticipation thereof), reduced
working hours and debt. Consequently, it is appropriate to look at
the financial crisis as a discrete, if multifaceted, event rather than
using one or more intermediary variables. Suicide data were
obtained from the ‘Instituto National de Estadı ´stica’ (INE),
Spain’s national statistics institute.13,17Suicides are deaths coded
as X60-X84 (ICD-10).26
In Spain, suicides are determined
following judicial review of any deaths that may have a possible
accidental or violent cause. Monthly suicide data were used to
maximize the data points available since the financial crisis.27Full
monthly suicide data, disaggregated by region and age-group, were
only available from 2005; in addition, the data are published with a
2-year delay, so only data for 72 months were available (January
2005 to December 2010). Population denominators were obtained
from the INE official population figures.
The European Journal of Public Health Advance Access published June 25, 2013
by guest on June 26, 2013
The ‘intervention’ of interest in this study was the financial crisis.
For unplanned events, the timing of the intervention must be estab-
lished from data that are independent of the time-series data being
analysed.28We based the timing on observed changes in GDP, the
measure most commonly used in defining a recession.10,29In Spain,
GDP began to contract from the second quarter of 2008
(Supplementary Appendix 1),12so the period up to and including
March 2008 was considered pre-financial crisis and April 2008
onwards, post financial crisis.
Segmented regression was used to estimate the effect of the financial
crisis on suicides.27A Poisson distribution of monthly suicide counts
was assumed offset by population data to model rates. Adjustments
were made for the length of the month and for any seasonal effect,
the latter by using a harmonic term based on the month of the year
that included two sine/cosine pairs.30Initial analyses suggested a
moderate degree of overdispersion (dispersion parameter=1.46),
so a more flexible quasi-Poisson model was used for all analyses.31
Durbin–Watson test. Further models were tested in a sensitivity
analysis, including a model allowing for both a step change and a
change in the trend; a two-step model whereby a recession period
was modelled as the duration that GDP contracted—second quarter
of 2008 until the last quarter of 2009, then returning to growth
until the end of the dataset (December 2010) (Supplementary
Appendix 1); and a model with the crisis period starting from July
2007, the point where unemployment began to rise (Supplementary
Appendix 2), rather than April 2008.
A control analysis was also undertaken using mortality from
accidental falls as the outcome to distinguish any association with
the financial crisis from other concurrent events.27This outcome
was chosen as being unlikely to be affected by the financial crisis
but has similarities to suicides in that both require judicial review
and have short lag times compared with other causes of death such
as chronic diseases.
Further stratified analysis was conducted to investigate whether
changes in suicides varied by region, sex and age group. The
individual autonomous regions of Spain could not be examined
separately, as the number of suicides per month was too low (<30
for most regions), which would have led to too much random vari-
ability. The regions were therefore grouped into three areas based on
geographical location and economic similarities: Northern Spain
(comprising Galicia, Asturias, Cantabria, the Basque Country,
Navarra, La Rioja and Aragon), central Spain (comprising Castilla
and Leon, Castilla La Mancha, Extremadura and Madrid), the
Mediterranean area and Canary Islands (comprising Catalonia,
Valencia, Murcia, Andalucia, the Balearic Islands and the Canary
Islands). Broadly speaking, regions in the Northern area have large
manufacturing sectors; the Central area have a relatively large agri-
cultural sector, although Madrid, which relies more on its financial
and service sectors, is an anomaly here; and the Mediterranean area
has a large service sector, in particular tourism, and a relatively large
real estate sector. All regions also have large construction and public
service sectors.13Age was grouped into younger economically active
ages (age 15–39), older economically active ages (age 40–64) and
post state retirement age (age 65 plus).
All analyses were conducted using the statistical packages R 2.15.0
and STATA version 11.
also tested for usingthe
Table 1 shows summary data of the average monthly suicide counts
and rates over the 72 months of the time series for the different
groups studied. The highest monthly suicide rates for the
categories of area sex and age group were in Northern Spain,
males and those aged ?65 years, respectively.
The principal model for all of Spain (figure 1) suggested that, over
the period studied, the underlying trend was of a 0.3% decrease in
the suicide rate per month (95% CI: 0.995–0.998; P<0.001). There
was an 8.0% step increase in the suicide rate associated with
Durbin–Watson statistic showed no evidence of autocorrelation
(DW 2.10, P=0.421). The control analysis showed no evidence of
a change in mortality from accidental falls associated with the
financial crisis (step change: RR 1.031; 95% CI: 0.939–1.132;
P=0.525) (Supplementary Appendix 3).
Results of the stratified analyses are presented in table 2 with the
plots in Web Supplementary Appendices 4–6. All results are
concordant in suggesting an increase in suicide rates. Although the
stratified analyses suggest a greater increase in the Mediterranean
and Northern areas, in males and in younger age groups, the low
statistical power when testing for interaction and the associated
P-values (test for interaction: area P=0.868, sex P=0.263, age
P=0.923) prevent any firm conclusions being made on a differential
effect by sub-groups.
All models tested in the sensitivity analysis also showed a relative
increase in the suicide rate during the financial crisis period
outlying observations (December 2006, December 2009 and
November 2010) had little impact on the results.
Figure 1 Trend in monthly suicide rates for all of Spain before and
since the financial crisis
Table 1 Mean monthly suicide data for the duration of the time
rate (per 100000)
277 (31)45.7 (1.1)0.61 (0.07)
Mean (SD). Note: Total for all Spain is not equal to the sum of the
area totals, as Ceuta and Melilla (two autonomous Spanish cities in
North Africa) are not included in the latter.
2 of 5
European Journal of Public Health
by guest on June 26, 2013
These results suggest that the onset of the financial crisis has been
associated with a relative increase in the suicide rate in Spain above
the expected underlying trend. If this association was causal, the
financial crisis may account for around 21 suicides per month in
Spain or around 680 suicides since the crisis so far (up to the end of
2010). Although we were underpowered to test for interaction, the
relative increase in the suicide rate appears to have been greatest in
the Mediterranean and Northern areas of Spain, amongst males and
amongst those of working age.
Comparison with previous studies
The underlying downward trend in suicides seen during the study
period is consistent with the long-term trend in most European
countries. Numerous explanations have been offered for this
downward trend including improved mental health provision,
greater use of antidepressants and policies such as changes to para-
cetamol pack sizes and withdrawal from certain drugs from over the
counter sales.32Only one previous study has examined the effect of
the current financial crisis on suicides while taking into account such
underlying trends; this found around 1000 more suicides in England
during the recession than would be expected based on historical
trends. Similar to our results this increase was found to be greater
amongst males.17The reported increases in suicides in Greece, Italy
and the European Union as a whole also provide some support to our
results, although these reports did not account for underlying trends
in suicides.19,21,22Again, consistent with our results for suicides, the
Spain found increases in primary care attendance for various mental
illnesses in 2010–11 compared with 2006–07, including mood,
anxiety, somatoform and alcohol-related disorders.23Only one
study so far, in South Australia, has shown no evidence of an
increase in mental health indicators related to suicide since the
financial crisis, in this case depression, suicidal ideation and other
psychological distress measures.25This may perhaps be explained by
a low proportion of unemployed people in their sample compared
with the population as a whole but also the limited effect that the
globalfinancial crisishassofarhadonAustralia comparedwithSpain
and other countries in Europe.25The increase in suicide rates is also
supported by evidence of increases during past economic crises in
Asia and Russia.9,10In addition, other studies find that associations
between adverse macro-economic conditions and suicide are
strongest amongst males and younger age groups, consistent with
our findings.3–6No other studies were found that have investigated
the differential effect of the current financial crisis on suicide or
other mental health indicators between age groups in Spain or
abroad; nor were there any studies investigating regional differences
in the effect within Spain; therefore, our findings cannot directly
be compared with other results.
Strengths and limitations
Our study has a number of strengths. Interrupted time-series is
regarded as a powerful quasi-experimental design for assessing the
longitudinal impact of an intervention, as it enables both random
month on month fluctuations and the underlying trend to be
accounted for in the analysis.27In addition, it does not suffer from
some of the biases and confounders of other observational studies.
Other variables associated with suicides that may have changed over
the same time period, including seasonal fluctuations, were adjusted
for during the analysis. Any residual confounders would have to be
events that occurred at the same time as the financial crisis but were
not a manifestation of the crisis. While the lack of any association
between the financial crisis and deaths from accidental falls cannot
exclude all such events, it does help to exclude many, such as unrec-
ognized changes to systems of death registration or classification, or
any other events that would have impacted on both causes of death.
The results of this study were also robust to sensitivity analysis for the
timing of the financial crisis period, with all models that were tested
showing a relative increase in suicides above the underlying trend
during the financial crisis period. In addition, there was no
evidence of autocorrelation in our model. Finally, whilst the use of
suicides as an outcome measure may provide a very incomplete
picture of the impact of the financial crisis on mental health, it is
an objective measure that is less subject to responder bias, observer
bias and validity issues compared with other outcomes such as self-
reported mental health indicators or suicidal ideation.
There are also potential limitations to this study. Firstly, the
number of suicides per month in the stratified analysis was
relatively low, leading to relatively wide confidence intervals and
requiring the amalgamation of autonomous regions into three
larger areas. Secondly, there is no established definition for the
timing of the financial crisis in Spain. As recommended for imper-
fectly identifiable events, an independent indicator was used to
determine the onset of the crisis, in this case GDP.28Nonetheless,
it is clear that some manifestations of the financial crisis in Spain
began before this point including the increase in unemployment
rates. However, various other definitions of the financial crisis
period were modelled in the sensitivity analysis, and all of these
also showed evidence of an increase in the suicide rate above the
underlying trend during the crisis, so it is unlikely that any incorrect
timing of the ‘intervention’ impacted on the results. Thirdly, no lag
period was included in the model, consistent with evidence from
elsewhere that there is no significant lag effect.17,22However, if a lag
did exist, the modelled effect would have been attenuated so the
effect of the financial crisis on the suicide would, if anything, have
been underestimated. Finally, studies have suggested that deaths
from events with undetermined intent (ICD-10: Y10-Y3426)
should also be included when analysing suicides to avoid underesti-
mation.33However, these were not available in a form disaggregated
by the variables required (including region and age group). There
were only between 1 and 19 deaths per month nationally from events
with undetermined intent during the period of study, so even if all
had been a suicide, it would have had little impact on the results.
Interpretation and implications
Our study alone cannot establish whether the association found
between the financial crisis and suicides is causal; however, this ex-
planation is supported by the relatively large magnitude of effect,
Table 2 Suicide rates during the financial crisis period compared
with before the financial crisis
PopulationStep change (RR) 95% CIP-value
RR: rate ratio; 95% CI: 95% confidence interval.
P-values are two-sided. All results based on a step change model,
adjusted for seasonality, with the pre-crisis period from January
2005 to March 2008 and the financial crisis period from April
2008 to December 2010.
The effect of the late 2000s financial crisis on suicides
3 of 5
by guest on June 26, 2013
consistency with previous studies, coherence with existing theory
dating back to Durkheim1
and the existence of numerous
plausible mechanisms for the relationship between financial crises
and suicides (figure 2).34–36Perhaps the most researched mechanism
is through the effects of the financial crisis on unemployment and its
association with suicide. Spain has experienced a dramatic rise in
unemployment during the financial
Appendix 2); numerous studies have found strong associations
between increasing unemployment and suicides in the past.3,5,37
Initial evidence from the current financial crisis in other European
countries appears to suggest the same pattern is occurring.17
required, the groups in the stratified analysis that showed evidence
of increases in the suicide rate above the underlying trend appear to
coincide relatively well with those that have experienced the biggest
absolute increases in unemployment rate between the start and the
end of the study period, including the Mediterranean region
of Spain, males and those of younger working age groups
(Supplementary Appendix 8).
This study is the first to show the association between the financial
crisis and suicides in Spain; it is also perhaps the strongest evidence
to date of the detrimental impact that the current financial crisis
may be having on mental health. The association found with suicides
is likely to represent only the tip of the iceberg of a possible effect of
the financial crisis on wider mental health. In addition, by 2010 the
biggest effects of the financial crisis in Spain had yet to manifest
themselves, as many of the major social spending cuts had only just
been introduced.14It may therefore be that, if the association found
in this study is causal, a further increase in suicide rates will already
have occurred and will continue unless mitigating interventions are
introduced promptly. Potential interventions include active labour
market programmes, family support programmes and debt relief
programmes.36The results of the stratified analysis undertaken in
this study may help to identify which groups should be targeted with
Although this study has provided important insight into one of the
harmful impacts the financial crisis may be having on health, there
remains a need for further research. Firstly, a number of potentially
at risk groups have been identified during the stratified analysis and
possible reasons for the increased risk in these groups have been
hypothesized, but we were underpowered to test whether these
were true differences or due to chance. Further investigation is
needed to try to establish whether these subgroups are truly at
greatest risk and if so why this is the case. Secondly, given the
global nature of the financial crisis, further international studies
are required to establish whether similar effects are being seen
with suicide rates elsewhere to help identify whether certain
policies implemented by some countries may have a protective
effect. Thirdly, further research is needed to identify the effectiveness
of mitigating interventions and how they are best implemented.
Supplementary data are available at EURPUB online.
Ethical approval was obtained from the London School of Hygiene
and Tropical Medicine research ethics committee.
We would like to thank Isabel Ruiz, MaJose ´ Sa ´nchez and Esther
Molina from the Andalusian School of Public Health for their
comments and suggestions.
There was no specific funding for this study; J.L.B. was supported by
a National Institute of Healthcare Research Academic Clinical
Fellowship; A.G. was supported by a Medical Research Council
Research Methodology Fellowship.
Conflicts of interest: None declared.
? In Spain, the financial crisis has been associated with a
substantial increase in suicide rates over and above the
? Public health interventions focussing on mitigating the
impact of the financial crisis on mental health and suicide
should be established.
? The effect of the financial crisis on suicides appears to be
greatest amongst men and amongst those of working age;
these groups may benefit most from targeted interventions.
Figure 2 Proposed mechanisms for the relationship between economic instability and mental illness and suicides
4 of 5
European Journal of Public Health
by guest on June 26, 2013
1 Durkheim E. Le Suicide. Felix Alcan: Paris, 1897.
2 Brenner MH. Mental Illness and the Economy. Oxford: Harvard U. Press, 1973.
3 Tapia Granados J. Recessions and Mortality in Spain, 1980–1997. Eur J Popul
4 Berk M, Dodd S, Henry M. The effect of macroeconomic variables on suicide.
Psychol Med 2006;36:181–9.
5Platt S, Micciolo R, Tansella M. Suicide and unemployment in Italy: description,
analysis and interpretation of recent trends. Soc Sci Med 1992;34:1191–201.
6 Gerdtham UG, Johannesson M. Business cycles and mortality: results from Swedish
microdata. Soc Sci Med 2005;60:205–18.
7 Stuckler D, Basu S, Suhrcke M, et al. The public health effect of economic crises and
alternative policy responses in Europe: an empirical analysis. Lancet
8 Ruhm CJ. Are recessions good for your health? Q J Econ 2000;115:617–50.
9Pridemore WA, Chamlin MB, Cochran JK. An interrupted time-series analysis of
Durkheim’s social deregulation thesis: the case of the Russian federation. Justice Q
10 Chang SS, Gunnell D, Sterne JA, et al. Was the economic crisis 1997-1998
responsible for rising suicide rates in East/Southeast Asia? A time-trend analysis for
Japan, Hong Kong, South Korea, Taiwan, Singapore and Thailand. Soc Sci Med
11 Temin P. The great recession and the great depression. National Bureau of Economic
Research, Inc. Cambridge, Massachusetts, USA: NBER Working Papers: 15645, 2010.
12 Organisation for Economic Cooperation and Development. OECD.StatExtracts,
2012 [cited 2012 04/07/2012]. Available at: http://stats.oecd.org/Index.
13 INE. Instituto Nacional de Estadı ´stica [National Statistics Institute]. [cited 2012 04/
07/2012]. Available at: http://www.ine.es/.
14 Gene-Badia J, Gallo P, Hernandez-Quevedo C, Garcia-Armesto S. Spanish health
care cuts: penny wise and pound foolish? Health Policy 2012;106:23–8.
15 Houston M, Day M, de Lago M, Zarocostas J. Health services across Europe face
cuts as debt crisis begins to bite. BMJ 2011;343:d5266.
16 Vogler S, Zimmermann N, Leopold C, de Joncheere K. Pharmaceutical policies in
European countries in response to the global financial crisis. South Med Rev
17 Barr B, Taylor-Robinson D, Scott-Samuel A, et al. Suicides associated with the 2008-
10 economic recession in England: time trend analysis. BMJ 2012;345:e5142.
18 Economou M, Madianos M, Theleritis C, et al. Increased suicidality amid economic
crisis in Greece. Lancet 2011;378:1459–60.
19 Mattei G, Ferrari S, Rigatelli M. Economic recession in Italy: a review of short-term
effects on healthJournal of Psychosomatic Research, Conference: 14th Annual
Meeting of the European Association for Consultation Liaison Psychiatry
and Psychosomatics, EACLPP 2011 Budapest Hungary. Conference
Start: 20110630 Conference End: 20110702. Conference Publication: (var.pagings).
J Psychosom Res 2011;70:606.
20 Sargent-Cox K, Butterworth P, Anstey KJ. The global financial crisis and psycho-
logical health in a sample of Australian older adults: a longitudinal study. Soc Sci
21 Kentikelenis A, Karanikolos M, Papanicolas I, et al. Health effects of financial crisis:
omens of a Greek tragedy. Lancet 2011;378:1457–8.
22 Stuckler D, Basu S, Suhrcke M, et al. Effects of the 2008 recession on health: a first
look at European data. Lancet 2011;378:124–5.
23 Gili M, Roca M, Basu S, et al. The mental health risks of economic crisis in
Spain: evidence from primary care centres, 2006 and 2010. Eur J Public Health
24 Economou M, Madianos M, Peppou LE, et al. Major depression in the Era of
economic crisis: a replication of a cross-sectional study across Greece. J Affect Disord
25 Shi Z, Taylor AW, Goldney R, et al. The use of a surveillance system to measure
changes in mental health in Australian adults during the global financial crisis.
Int J Public Health 2011;56:367–72.
26 WHO. International Statistical Classification of Diseases and Health Related Problems.
Geneva: World Health Organisation, 2004.
27 Wagner AK, Soumerai SB, Zhang F, Ross-Degnan D. Segmented regression analysis
of interrupted time series studies in medication use research. J Clin Pharm Ther
28 Gilmour S, Degenhardt L, Hall W, Day C. Using intervention time series analyses to
assess the effects of imperfectly identifiable natural events: a general method and
example. BMC Med Res Methodol 2006;6:16.
29 Gurfinkel EP, Bozovich GE, Dabbous O, et al. Socio economic crisis and mortality.
Epidemiological testimony of the financial collapse of Argentina. Thromb J
30 Hunsberger S, Albert PS, Follmann DA, Suh E. Parametric and semiparametric
approaches to testing for seasonal trend in serial count data. Biostatistics
31 Hoef JMV, Boveng PL. Quasi-Poisson vs. negative binomial regression: how should
we model overdispersed count data? Ecology 2007;88:2766–72.
32 Mackenbach J, McKee M. Successes and Failures of Health Policy in Europe Over Four
Decades: Diverging Trends, Converging Challenges. Buckingham: Open University
33 Ohberg A, Lonnqvist J. Suicides hidden among undetermined deaths. Acta
Psychiatrica Scandinavica 1998;98:214–8.
34 Zivin K, Paczkowski M, Galea S. Economic downturns and population mental
health: research findings, gaps, challenges and priorities. Psychol Med
35 Lester BY. Learnings from Durkheim and beyond: the economy and suicide. Suicide
Life Threat Behav 2001;31:15–31.
36 World Health Organization Regional Office for Europe. Impact of Economic Crises
on Mental Health. Copenhagen, Denmark: WHO Regional Office For Europe, 2011.
37 Yip PSF, Caine ED. Employment status and suicide: the complex relationships
between changing unemployment rates and death rates. J Epidemiol Commun Health
The effect of the late 2000s financial crisis on suicides
5 of 5
by guest on June 26, 2013