ArticlePDF Available

Parenthood and Later Life Health: An International Life Course Analysis of Parents and Childless Adults Aged 50 and Older

Authors:

Abstract and Figures

This study investigates how women’s and men’s fertility history affect their health in later life and if this relationship varies across countries and cohorts. We use life history data and current health status of persons aged 50 and over from the Survey of Health, Ageing and Retirement in Europe (SHARE) for 13 countries. Country-fixed effects regressions show that parenthood itself and the number of children have little impact on later life health, but fertility timing is important. Moreover, significant country and cohort differences show that the health implications of timing depend upon the socio-historic context.
Content may be subject to copyright.
DOI 10.1515/sjs-2018-0015
© 2018. This work is licensed under the Creative Commons Attribution-NonCommercial-
NoDerivatives 4.0 License. (CC BY-NC-ND 4.0)
Swiss Journal of Sociology, 44 (2), 2018, 327–354
Parenthood and Later Life Health: An International Life Course
Analysis of Parents and Childless Adults Aged 50 and Older
Nadine Reibling* and Katja Möhring**
Abstract: is study investigates how womens and men’s fertility history aect their health in
later life and if this relationship varies across countries and cohorts. We use life history data
and current health status of persons aged 50 and over from the Survey of Health, Ageing and
Retirement in Europe (SHARE) for 13 countries. Country-xed eects regressions show
that parenthood itself and the number of children have little impact on later life health, but
fertility timing is important. Moreover, signicant country and cohort dierences show that
the health implications of timing depend upon the socio-historic context.
Keywords: parenthood, health, life course, SHARELIFE, welfare state
Elternschaft und Gesundheit im höheren Lebensalter: Eine internationale
Lebenslaufanalyse von Eltern und kinderlosen Erwachsenen im Alter 50+
Zusammenfassung: Der Beitrag untersucht, wie sich die Elternschaft auf die Gesundheit von
Männern und Frauen im höheren Alter auswirkt und ob sich der Zusammenhang zwischen
Elternschaft und Gesundheit für verschiedene Länder und Geburtskohorten unterscheidet.
Die Studie verwendet Lebenslauf- und Gesundheitsdaten für Personen im Alter 50+ aus dem
Survey of Health, Ageing and Retirement in Europe (SHARE). Fixed-eects-Regressionen für
13 Länder zeigen, dass die Elternschaft und die Anzahl der Kinder geringen Einuss auf die
Gesundheit im höheren Alter haben, der Zeitpunkt der ersten Elternschaft aber bedeutsam
ist. Signikante Länder- und Kohortenunterschiede zeigen, dass die Bedeutung des Zeitpunkts
vom jeweiligen sozio-historischen Kontext abhängt.
Schlüsselwörter: Elternschaft, Gesundheit, Lebenslauf, SHARELIFE, Wohlfahrtsstaat
Parentalité et santé à un âge avancé : une analyse internationale des parents et des
personnes sans enfants à âgés de 50 ans et plus
Résumé : Cet article examine comment la parentalité inuence la santé des femmes et des
hommes à un âge avancé et si cette relation dière selon les pays et les cohortes de naissance.
Cette étude utilise des données biographiques et des informations sanitaires issues de l’enquête
SHARE, portant sur des personnes de 50ans et plus. Des régressions xed-eects pour
13pays indiquent que la parentalité et le nombre d’enfants ont peu d’inuence sur la santé
à un âge avancé, mais que l’âge à la naissance est important. Des diérences signicatives
entre pays et cohortes indiquent en outre que l’importance de l’âge à la naissance dépend du
contexte social et historique.
Mot-clés : parentalité, santé, parcours de vie, SHARELIFE, État-providence
* Department of Social Sciences, Siegen University, D-57068 Siegen,
reibling@soziologie.uni-siegen.de.
** School of Social Sciences, University of Mannheim, D-68131 Mannheim,
moehring@uni-mannheim.de.
328 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
1 Introduction1
An increasing number of studies aims to link family and health trajectories in a life
course framework (Mayer 2009). is rising eort strives both from a growing inter-
est in the sociology of health to identify early-life predictors of non-communicable
diseases/mortality (Burton-Jeangros etal. 2015) and in family sociology to under-
stand the consequences of family transitions and their timing (Naucket al. 2009).
Moreover, investigating family trajectories can help to enlighten the distal causes
of social disparities in health, since partnership and parenthood are, for instance,
strongly associated with one’s socioeconomic development over the life course
(Furstenberg Jr. etal. 1987; Taylor 2009).
Becoming a parent is one of the most fundamental and transformative life
course transitions. As a biological event, pregnancy and childbirth can have direct
consequences for the health of mothers. However, it also aects various dimensions
of social life including labor force participation, socio-economic development, marital
quality, gender equality, leisure activities and social integration which are important
determinants of physical and mental health for both men and women (indirect
eects). As a result, there has been a multidisciplinary interest in the analysis of
how becoming and being a parent aects individual’s health.
e changing fertility behaviour during the second half of the 20th century
with a decreasing number of children per couple and a rising age of mothers and
fathers at rst birth has further fuelled the debate on the health consequences of
parenthood. e total fertility rate dropped in the United States from2.48 in
1970 to2.04 in 2003 and decreased even further in many European countries,
e. g., in Spain from2.88 in 1970 to1.27 in 2003. e percent of births to women
aged40 and older doubled during that period, for instance, in the US, Denmark,
and Sweden. And in all countries the share of rst and second births among the
age group40 and older has increased substantially (Billari etal. 2007). Clinicians
express strong concerns over the health consequences of delayed childbearing (Breart
1997; Ben-David etal. 2016) which fuel the public debate of this issue.
While all advanced, industrialized nations share these demographic trends,
there are still marked dierences in the fertility rate and timing of parenthood across
countries. Moreover, welfare states oer dierent contexts for parenthood: Nordic
welfare states, for instance, provide strong support for families through generous
parental leave and public childcare, while in other regions, such as SouthernEurope,
the welfare states provide less help to families’ care-taking responsibilities. Com-
paring the eects of parenthood on health across studies from dierent countries
1 Nadine Reibling received funding for this article as a member of the HiNEWS project– Health
Inequalities in European Welfare States– funded by NORFACE (New Opportunities for Research
Funding Agency Cooperation in Europe) Welfare State Futures programme (grant reference:
42-14-110). For more details on NORFACE, see http://www.norface.net/11.
Parenthood and Later Life Health 329
SJS 44 (2), 2018, 327–354
suggests that institutional contexts may not only shape fertility behaviour, but also
inuence how parenthood aects parents’ health.
Considering the potential signicance of both demographic trends and
institutional contexts, this paper investigates life course eects of parenthood for
health across countries and cohorts. We use life history data and health status
information from the Survey of Health, Ageing and Retirement in Europe (SHARE)
for13 European countries. e analysis group comprises women and men older
than50years. Based on country-xed-eects regressions, we compare parenthood
eects across three health measures: depressive feelings, self-rated health status, and
the number of chronic diseases.
2 Theory
e advent of parenthood denes a core transition in the life course that may
catalyse and interlock sets of social and biological consequences. e timing of
that transition implicitly gauges how smoothly the social and biological spheres
come together, with lifelong consequences for health. (Mirowsky2002, 340)
We start by reviewing theories that have been developed for explaining the link
between the fertility history and health. ese theories not only make predictions
regarding the overall eect of parenthood on health, but also consider the health
consequences of parity, i. e. number of children, and their timing. ree strands
of theoretical mechanisms can be identied in this body of literature: biological
explanations, social mechanisms, and selection processes.
2.1 Biological explanations
Evolutionary and developmental frameworks suggest direct biological eects of
parenthood for the physical health of mothers. e disposable soma theory argues
that there is a “trade-o between longevity and reproduction,” because reproduction
uses resources that would have otherwise been available to the somatic maintenance
of the mother’s body (Westendorp and Kirkwood 1998). us, this framework
suggests that mothers are in worse health than childless women and that this health
disadvantage increase with the number of pregnancies women have. Empirical sup-
port for this perspective can be found in the fact that both in historic (Westendorp
and Kirkwood 1998) and in recent cohorts (Hurt etal. 2006), a very high number
of children is associated with increased mortality.
Biological arguments have also been important in the literature on fertility
timing, because there is a biological fertility limit that ranges for women from
around16 to the early40s. erefore, a developmental perspective suggests that
the “optimum age at rst birth [is] shortly after the reproductive system is ready,
330 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
while the organism enjoys the energy and resilience of youth” (Mirowsky 2005,32).
Clinicians are concerned about teenage births, because they often occur prior to the
complete development of the reproductive system and are therefore associated with
a higher risk for complications and health risks for mothers and children (Ozalp
etal. 2003). However, the primary concern of bio-medical frameworks have been
the health risks for mothers and children arising from late childbearing such as
birth defects, stillbirths, and maternal morbidity and mortality (Kozuki etal. 2013;
Restrepo-Méndez etal. 2015; Lisonkova etal., 2017). Older mothers are also
more likely to receive intensive medical interventions such as caesarean deliveries
( Ben-David etal. 2016).
However, clinical research has also highlighted health advantages of mother-
hood particularly the long-term protection against breast cancer that arises from
multiple and early births (Russo etal. 2005). On the other hand, motherhood
and higher parity have also been shown to increase the risk for other cancer types
(Muñoz etal. 2002).
In sum, biological explanations suggest mostly direct negative eects for
mothers’ physical health. ese negative eects should be more pronounced the
more children a woman has and if births occur either too early or too late.
2.2 Social mechanisms
Social scientists have stressed the various social correlates of parenthood which can
exhibit positive or negative indirect eects on mothers’ and fathers’ health. Unlike
biological frameworks, social scientists also consider implications for parents’ mental
health. On the one hand, parenthood can be viewed as a benecial transition (role
accumulation/enhancement) (Sieber 1974; Marks 1977) that provides meaning to
one’s life, increases social integration and responsibility towards one’s own health.
For instance, parents are less likely to engage in risk behaviour (Arnett 1998). On
the other hand, parenthood can lead to role strain (Goode 1960) which creates
stress and promotes unhealthy behaviour such as low physical activity, less sleep,
and worse dietary habits (Nomaguchi and Bianchi 2004).
Later life health might to some extent also reect the costs and benets that
adult children have for parents (Umberson etal. 2010). e relative importance
of current eect of children and accumulated life course inuences of parenthood
also depend on the dimension of health. For instance, physical health (e. g., the
development of chronic conditions such as heart disease) will reect more strongly
long-term strains and resources associated with parenthood.
Since parenthood can entail benets as well as strains, research has highlighted
the importance of timing. Earlyparenthood can lead to a process of cumulative
disadvantage (DiPrete and Eirich 2006), because young parents are more likely
to experience a disadvantaged socio-economic development and more disruptive
partnerships (Taylor 2009). erefore, in contrast to bio-clinical research, social
Parenthood and Later Life Health 331
SJS 44 (2), 2018, 327–354
scientists see delaying childbirth as benecial for mothers’ and fathers’ health (Lacey
etal. 2017), because at a later age individuals have acquired educational, nancial,
and social resources that help them to cope with the costs and stresses of parenthood.
ere has been a debate to what extent there is a limit to the benets of delay-
ing childbirth. Some theorists suggest that from a resource perspective delaying as
long as possible is benecial (Mirowsky 2005). Others have argued that such a limit
might exist particularly for women. If women have births later than social norms
expect or later than they personally envisioned, this can create psycho-social stress
with negative implications for their health (Carlson 2011). Psycho-social stress is
particularly likely if late childbearing reects experienced diculties with conception.
In sum, social mechanisms suggest that parenthood can be good for parents’
health if the benets outweigh the costs and that this is more likely with lower
parity and delayed childbearing. Unlike the biological explanations, the indirect
eects postulated by social mechanisms also suggest an impact of fatherhood on
health. While delaying childbirth is expected to be generally benecial for men,
there may be a limit to the benets of late childbirth for women if they feel that
they had their children too late considering the present normative expectations or
their own life plans.
2.3 Selection
ere are also a variety of selection processes that may aect the relationship between
parenthood and health in later life. First, health selection could aect whether persons
become parents, but also parity and timing so that potentially the eect of all three
characteristics on later life health is underestimated. Parents could be a healthier
group than childless persons, since healthier women are more likely to marry and
have children (Brockmann and Klein 2004). Parents with more children could be
healthier than parents with less children, since biological fertility is associated with
health. Finally, older mothers/fathers could be a healthier group than younger
mothers/fathers, because they are still able to conceive at an age where other men and
women have reached the end of fecundity (Yi and Vaupel 2004). e signicance
of the health selection eect is seen as important in historic populations (Hurt etal.
2006). In contemporary samples they are seen as less relevant (Huijts etal. 2013),
because fertility is stronger determined by social factors since it can be controlled
via contraceptives (Hurt etal. 2006). Second, social selection processes can aect
fertility. Men and women with a higher socioeconomic status face higher opportunity
costs and might thus more often be childless or have less children. Moreover, a low
socioeconomic status seems to increase the likelihood of an early birth.
In sum, selection mechanisms could lead to an overestimation of the nega-
tive consequences of childlessness and early age at rst birth for later life health,
while the costs of higher parity and late age at birth for later life health might be
underestimated.
332 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
3 Empirical evidence
In the following part, we review the empirical evidence on the relationship between
characteristics of the fertility history and health outcomes.
3.1 Parenthood
Studies on the general eect of parenthood on health nd either no health dierences
between parents and childless persons (Eggebeen and Knoester 2001; Mirowsky 2005;
Teachman 2010; Kroll etal. 2016) or a health advantage for parents on a variety
of health outcomes (Grundy and Tomassini 2005; Kohler etal. 2005; Helbig etal.
2006; Hurt etal. 2006; Teachman 2010; Read etal. 2011; Gibney etal. 2015).
Interestingly, this is in contrast to research on happiness and life satisfaction which
usually nds that parents with resident children are unhappier than childless persons
(Hansen 2012). Age at rst birth (Mirowsky 2005) and marital status (Nomaguchi
and Milkie 2003) are important moderators of the eect of parenthood on health.
Moreover, the eect of parenthood/childlessness seems to vary cross-nationally
(Huijts etal. 2013; Tanaka and Johnson 2014).
3.2 Parity
ere is mixed evidence concerning the implications of the number of births for
health. While a number of studies nd excess mortality (Hurt etal. 2006) and worse
health for parents with a higher number of children (Kington etal. 1997; Kohler
etal. 2005; Read etal. 2011), others nd no associations (Henretta 2007; Spence
2008). Comparative studies show interesting– albeit contradictory – variations
across contexts. In a three country comparison, Grundy (2009) reports that higher
parity is associated with higher mortality in England/Wales and the US, but lower
mortality in Norway. She suggests that the availability of public childcare might
lead to the health benets in Norway. In contrast, Hank’s (2010) comparison of
East and West German women nds that in West Germany higher parity is related
to better health, while higher parous East German women have worse health in
later life, even though childcare was widely available in the Eastern, but not in the
Western part of Germany.
3.3 Early age at first birth
A large number of studies conrms that early childbearing is associated with poorer
physical (Kington etal. 1997; Mirowsky 2002; Mirowsky 2005; Henretta 2007;
Spence 2008; Taylor 2009; Barban 2013; O’Flaherty etal. 2015) and mental health
for mothers (Mirowsky and Ross 2002; Spence 2008; Carlson 2011; Read and
Grundy 2011). e evidence base for fathers is smaller. Most Anglo-Saxon studies
also suggest negative health implications of early births for men (Mirowsky 2002;
Parenthood and Later Life Health 333
SJS 44 (2), 2018, 327–354
Mirowsky and Ross 2002; Grundy and Tomassini 2005; O’Flaherty etal. 2015) while
a study based on German data nds no eects of timing for fathers (Hank 2010).
3.4 Late age at first birth
e health eects of late childbearing are more controversial. For mothers, late age at
rst births is associated with a higher risk for breast cancer (Merrill etal. 2005), health
limitations (Read etal. 2011), and sometimes a poorer physical ( Mirowsky2002;
Mirowsky 2005) and mental health (Mirowsky and Ross 2002; Spence 2008; Carl-
son 2011). However, late births neither have a negative eect on overall mortality
( Mirowsky 2005; Henretta 2007; Grundy 2009) nor on the likelihood to have (any)
cancer (Henretta 2007). Late childbearing can even decrease the risks for certain
cancers such as endometrial and cervical cancer (Merrill etal. 2005). For fathers, a
late age at birth is either not associated with health (Hank 2010) or seems to con-
vey health benets (Mirowsky 2002; Mirowsky 2005; Mirowsky and Ross 2002).
e health implications of parenthood have been studied quite extensively.
However, the existing evidence base is ambiguous with respect to several charac-
teristics except the negative consequences of early births. Only a small number of
studies have also considered fathers. Reviewing the evidence already suggests that the
parenthood eects on health might not be universal, but depend upon the historical
or social context. is is why the aim of this study is to systematically compare the
later life consequences of parenthood across cohorts and countries.
4 Institutional and historical context
4.1 The 1920 to 1959 birth cohorts
e cohorts to be examined in this study are born between 1920 and 1959. ey cover
an interesting period of fertility trends in Europe: About half of our sample consists
of pre-WWII cohorts and the other half was born after the war. e youngest cohort
reached adolescence right before WWII. In our sample, less than 1% of the births
occurred before or during the war. us, this study analyses primarily post-WWII
fertility patterns occurring mostly between 1950 and 1980. e earlier cohorts from
1920 to 1944 are the parents of the baby boom generation. ey display a higher
fertility and a lower age at rst birth than previous cohorts. For instance, in Austria
the total fertility rate was 2.4 in 1950 compared to 1.96 in 1930, in France 2.93 in
1950 compared to 2.27 in 1930, and in Denmark 2.58 compared to 2.30 in 1930
(Tomka 2013). e later cohorts who reached adolescence not before the 1960s
already exhibit part of the demographic change towards lower and later fertility in
the Western countries that continued into the 21stcentury.
334 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
4.2 Implications of the social-historical context
e majority of existing work on parenthood and later life health comes from
single-country studies. However, comparing the results across studies from dier-
ent countries (Hansen 2012) and recent comparative work (Grundy 2009; Hank
2010; Huijts etal. 2013; Gibney etal. 2015) suggests that the wider social context
could aect how parenthood and its timing aects later life health. us, scholars
have become increasingly interested in exploring the “historical, cultural and social
variations” in the relationship between parenthood and health (Mirowsky and Ross
2002,1293). is interest strongly relates to the comparative health inequalities
literature which has investigated how contextual factors such as the institutional
arrangement of the welfare state moderates social inuences on health (Bambra
2006; Beckeld etal. 2015). With respect to parenthood, we argue that the social
context could moderate both the biological and the social mechanisms that link
parenthood and health.
First, while the biological consequences of parenthood are based on physio-
logical processes, the social context can aect the likelihood of certain risks as
well as their impact on health. For instance, better nutrition and living standards
have contributed to a reduction in maternal mortality (Scalone 2014). Even more
important was the progress in the safety of maternal care for the decreasing biological
risks of pregnancy and childbirth (Loudon 2000). e establishment of universal
healthcare systems in European countries strongly increased the access to maternity
care (Kennedy etal. 2015). Since both living standards and access to maternity
care substantially improved in the second half of the 20thcentury, we can expect
that parenthood, higher parity, and early or late age at rst birth have been less
detrimental to later life health in the younger than in the older cohorts.
Second, the social context might also aect the social costs and benets that
parenthood entails. A number of scholars have suggested that welfare states, par-
ticularly transfers and services provided to families, might increase the benets and
reduce the costs of parenthood (Curtis and Phipps 2004; Aassve etal. 2005; Aitken
etal. 2015). is might be especially true for women, for whom family policy also
suggests independence from their partner and “the capacity to form and maintain an
autonomous household” (Orlo 1993, 319; original was in italics). is could be
especially important for younger mothers who are more likely to have extra-marital
births (Taylor 2009). us, we expect that countries with more generous family
policy and services, i. e. that provide more public care services and more generous
parental leave arrangements, such as Scandinavian and Eastern European welfare
states, should show more positive eects of parenthood and less negative eects of
early childbirth than regimes that mostly rely on family support such as conservative
and Mediterranean welfare states.
Parenthood and Later Life Health 335
SJS 44 (2), 2018, 327–354
4.3 Hypotheses
is study tests the following hypotheses: (1) childless individuals are in worse
health than parents (Parenthood), (2) a higher number of children is associated
with worse health for mothers and less so for fathers (Parity), (3) early parenthood
is detrimental to parents’ health (Timing-Early), (4) for mothers, health improves
with increasing age at childbirth until a certain age and then declines, while fathers’
health continuously improves with delaying childbirth (Timing-Late), (5) the parent-
hood characteristics are most strongly associated with chronic conditions which
incorporates experiences over a long period of time and least with depression which
can change more quickly based on current life circumstances (Health-Outcomes),
(6) negative eects of higher parity and young age at rst birth are smaller in
younger cohorts (Cohort), and (7)negative eects of higher parity and lower age at
rst birth are smaller in welfare states with more generous family policy and services
( Scandinavia < Eastern < Continental < Southern) (Country).
5 Data, operationalization, and methods
5.1 Data
e empirical analysis is based on data from the Survey of Health, Ageing and
Retirement in Europe (SHARE) waves4 (2010, Release 1.1.1) and 5 (2013,
Release1.0.0) and information on the marital, fertility, and employment history
from SHARELIFE (wave3, 2008/2009, Release1) (Schröder 2011; Börsch-Supan
etal. 2013).2 Health information and the control variables are drawn from wave4
and wave5 for those who did not participate in wave 4 and combined with the
life history information from wave 3. e analysis sample consists of 5 577men
and 6 242women who were older than 50years at the time of the interview for
SHARELIFE, hence born in 1920 to 1959. e sample comprises the 13countries
Austria, Belgium, Czech Republic, Denmark, France, East Germany, West Germany,
Italy, e Netherlands, Poland, Spain, Sweden and Switzerland.
2 is paper uses data from SHARE Waves1, 2,3 (SHARELIFE),4 and 5 (DOIs: 10.6103/
SHARE.w1.500, 10.6103/SHARE.w2.500, 10.6103/SHARE.w3.500, 10.6103/SHARE.
w4.500, 10.6103/SHARE.w5.500), see Börsch-Supan etal. (2013) for methodological details.
e SHARE data collection has been primarily funded by the European Commission through
FP5 (QLK6-CT-2001-00360), FP6 (SHARE-I3: RII-CT-2006-062193, COMPARE: CIT5-
CT-2005-028857, SHARELIFE: CIT4-CT-2006-028812) and FP7 (SHARE-PREP: N°211909,
SHARE-LEAP: N°227822, SHARE M4: N°261982). Additional funding from the German
Ministry of Education and Research, the U. S. National Institute on Aging (U01_AG09740-
13S2, P01_AG005842, P01_AG08291, P30_AG12815, R21_AG025169, Y1-AG-4553-01,
IAG_BSR06-11, OGHA_04-064) and from various national funding sources is gratefully
acknowledged (see www.share-project.org).
336 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
5.2 Operationalization
As dependent variables we use three health indicators. e number of chronic
conditions is used to operationalize physical health. A measure for self-rated health is
used ranging from1 (excellent) to5 (poor). To operationalize mental health we use
the EURO Depression Scale provided in SHARE (Prince etal. 1999). e scale is an
additive index reporting the number of depressive symptoms, ranging from1 to12
with higher values reecting greater levels of depressive feelings. We operationalize the
fertility history by means of two indicators: the number of children in three groups:
no children (used as reference category), one to two children, three or more children;
and the age at the birth of the rst child. e life-course factors span the period
from the year 1935, when the oldest sample members were15years old, to 2008,
when the youngest became59years. To take into account health selectivity with
respect to fertility behaviour, we control for respondents’ childhood health and their
parents’ mortality. e information on childhood health status refers to the period
from birth until the respondent become16years old and comprises three variables:
self-rated health; whether parents smoked; and whether the respondent was conned
to bed or home for one month or longer due to illness or disability. e variable on
parents’ survival is dierentiated in three categories for each parent: mother/father
died before age60(used as reference category); died between ages60–74; and died
at age older than75or is not deceased at the time of the interview. It provides a
proxy information on an individual’s genetic health disposition.
To analyse cohort dierences, we assigned the respondents to two cohorts:
those born between 1920 and 1944 are the older cohort; born between 1945 and
1959 are the younger cohort. To dierentiate welfare state types, we use a catego-
risation in four groups following Castles and Obinger (2008): the largest group
of Continental countries comprises Austria, Belgium, France, West Germany, e
Netherlands, and Switzerland3; Scandinavian/Social-democratic are Denmark and
Sweden; Southern European countries are Italy and Spain; the Central and Eastern
European group comprises countries with a socialist legacy, those are the Czech
Republic, East Germany, and Poland. As exogenous control variables we include
age and marital status. Furthermore, we control for several socioeconomic status
indicators including years of education, whether a respondent was active on the labour
market, homeownership, and the logarithmized household income. TableA.1 in the
Appendix gives an overview of all variables included in the statistical estimations.
3 Switzerland and the Netherlands are ambiguous cases which have been assigned to dierent
regimes in welfare state typologies. In this more historic perspective and in the absence of other
liberal countries, we consider their assignment to the Continental cluster as most adequate for
our data.
Parenthood and Later Life Health 337
SJS 44 (2), 2018, 327–354
5.3 Methods
Multilevel regressions were estimated using Country Fixed Eects models. ese
regression models are appropriate for data-sets with a small number of macro-level
units (N < 20), since they control for the residual variance on the country level
( Allison 2009; Möhring 2012; Maas and Hox 2005). e regression equation for
these kind of models is (linear model):
yy xxuu
ij ij kkij jNjN
=+ ++ +++−−0111111
ββαα
... ... ij
e+
xed part random part
with
yij :
: Individual-level outcome of observation i in country j
y0:
: Intercept over all countries
β
kkij
x:
: Estimator of individual-level variable number k of observation i in country j
αα
11 11
uu
jnjn
++
−−
... :
: xed eects for the N–1 countries
eij :
: Residual variance for observation i within country j
As the metric of the outcome variable varies, we use dierent regression model
specications: negative binominal models to analyse the number of chronic condi-
tions; ordered logit models for the self-rated health with ve values; and OLSlinear
regression models for the continuous EURO Depression scale. All regression models
are estimated separately for men and women and include the variables on the fertil-
ity history (number of children; age at rst birth), on childhood and adolescence
health until age16(self-rated health; periods of conned to bed/home for more than
onemonth); parents’ health behaviour when the respondent was young ( parents
smoked); parents’ mortality (mother’s survival status; father’s survival status), as
well as the control variables, socioeconomic status, and the country dummies as
described above.
6 Results
Table 1 includes the results of the multivariate models for our three dependent
variables separated by gender. In all models, we control for childhood health condi-
tions, parents’ mortality, and several indicators associated to the respondents’ current
socio-economic status. Our main variables of interest are the number of children
and age at the birth of the rst child.
First of all, dierences between parents and childless individuals as well as
according to the number of children are weak. Only depressive feelings for women
and chronic conditions for men are signicantly related with childlessness and parity.
Women with one or two children have on average a0.65scale points reduction in the
number of depressive symptoms compared to childless women. Additional analyses
338 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
Table 1 Regressions of fertility history on later life health outcomes for women and men aged 50 and above (coefficients
for self-rated health and depression, marginal effects for chronic conditions)
Variables Women Men
Chronic conditions Self-rated health Depression Chronic conditions Self-rated health Depression
Number of children (RC: No children)
1–2 children –0.269
(0.164)
–0.362
(0.246)
–0.645*
(0.297)
0.407*
(0.202)
–0.081
(0.249)
–0.013
(0.262)
3+ children –0.254
(0.165)
–0.355
(0.247)
–0.594*
(0.298)
0.371
(0.202)
–0.024
(0.250)
0.071
(0.263)
Age at 1st Child –0.134***
(0.031)
–0.241***
(0.042)
–0.226***
(0.050)
–0.101***
(0.027)
–0.116**
(0.039)
–0.187***
(0.040)
Age at 1st Child squared 0.002***
(0.001)
0.004***
(0.001)
0.004***
(0.001)
0.001**
(0.000)
0.002**
(0.001)
0.003***
(0.001)
Childhood health status (birth until age 16)
Self-rated health 0.124***
(0.018)
0.286***
(0.025)
0.159***
(0.029)
0.056**
(0.018)
0.212***
(0.026)
0.092***
(0.027)
Parents smoked 0.041
(0.037)
0.009
(0.049)
0.096
(0.060)
0.069
(0.038)
0.091
0.053)
0.113*
(0.056)
Confined to bed/home for 1+ month 0.121*
(0.055)
–0.144
(0.076)
0.087
(0.091)
0.167**
(0.057)
0.037
(0.083)
0.023
(0.087)
Parents’ mortality
Mother‘s survival status (RC: Died before age 60)
Died age 60–74 0.159*
(0.064)
–0.295**
(0.090)
–0.281**
(0.107)
–0.095
(0.065)
–0.176+
(0.095)
–0.215*
(0.099)
Died age 75+/Not deceased 0.031
(0.056)
–0.316***
(0.078)
–0.227*
(0.092)
–0.112*
(0.056)
–0.185*
(0.082)
–0.129
(0.086)
Father‘s survival status (RC: Died before age 60)
Died age 60–74 –0.089
(0.051)
–0.201**
(0.071)
–0.181*
(0.085)
–0.028
(0.051)
–0.010
(0.073)
–0.013
(0.077)
Died age 75+/Not deceased –0.177***
(0.047)
–0.225***
(0.065)
–0.213**
(0.078)
–0.183***
(0.048)
–0.220**
(0.068)
–0.108
(0.071)
Continuation of table 1 on the next page.
Parenthood and Later Life Health 339
SJS 44 (2), 2018, 327–354
Variables Women Men
Chronic conditions Self-rated health Depression Chronic conditions Self-rated health Depression
Controls
Married –0.058
(0.041)
–0.063
(0.056)
–0.065
(0.067)
–0.021
(0.050)
–0.060
(0.072)
–0.078
(0.075)
Log. household income 0.000
(0.015)
–0.069***
(0.020)
–0.055*
(0.025)
0.004
(0.016)
–0.090***
(0.023)
–0.138***
(0.024)
Age 0.033***
(0.002)
0.043***
(0.003)
0.028***
(0.004)
0.024***
(0.002)
0.037***
(0.004)
0.033***
(0.004)
Years of Education –0.026***
(0.005)
–0.046***
(0.006)
–0.026***
(0.007)
–0.005
(0.004)
–0.045***
(0.006)
–0.022***
(0.006)
Employed –0.592***
(0.065)
–0.364***
(0.073)
–0.098
(0.088)
–0.619***
(0.064)
–0.425***
(0.078)
0.088
(0.081)
Owner –0.169***
(0.041)
–0.195***
(0.057)
–0.273***
(0.068)
–0.215***
(0.044)
–0.268***
(0.063)
–0.119+
(0.066)
Constant 5.708***
(0.813)
4.179***
(0.730)
Cut-point 1 –5.276***
(0.682)
–3.840***
(0.710)
Cut-point 2 –3.803***
(0.680)
–2.325**
(0.708)
Cut-point 3 –1.842**
(0.679)
–0.336
(0.707)
Cut–point 4 0.107
(0.679)
1.540*
(0.708)
N6 242 6 242 6 242 5 577 5 577 5 577
Log Likelihood –1.06e + 04 –8 378.405 –9 071.983 –7 527.678
Chi square 1 135.238 1 600.559 707.597 1 122.799
R squared 0.113 0.075
Adjusted R squared 0.108 0.070
Notes: Standard errors in parentheses; *** p 0.001, ** p 0.01, * p 0.05; model specification varies by the metric of the outcome variable: chronic conditions: negative binomial (marginal
effects reported), self-rated health: ordered logit; depression scale: OLS; higher values indicate worse health.
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
Continuation of table 1.
340 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
have shown that this result only applies to older women (age65and above), whereas
for the younger of age50to64no signicant dierences in depressive symptoms
between childless women and mothers exist (see TableA.2 in the Appendix). e
reverse relationship exists for men’s physical health: men with one or two children
have on average a0.41scale points higher number of chronic conditions, while there
is no signicant dierence between men with three or more children and childless
men. To sum up, we do not nd evidence to support our rst hypothesis: parents
are not in better health than childless individuals, the only exception being depressive
feelings among women. With respect to physical health of men we even nd those
without children to be in a better condition than parents. erefore, Hypothesis2
assuming worse health for individuals with a high number of children is only sup-
ported with respect to men’s chronic conditions.
Timing of the rst birth, however, is signicantly associated with later life
health for both genders. As the linear and the quadratic term are signicant, the
relationship of age at rst birth and later life health appears to be u-shaped for
all health outcomes. Accordingly, for mothers and fathers, health improves with
increasing age at childbirth until a certain age and then declines. While the eect
strength for chronic conditions is similar for men and women, gender dierences exist
for self-rated health and depressive feelings. For both outcomes, the interrelation
of age at rst child and later life health is much stronger for women compared to
men. Also for men, the curvilinear relationship is not as pronounced as for women.
e relationship of men’s physical health in later life and age at rst birth appears
to be almost linear with negative health consequences only for very young fathers
(Figure1). To sum up, early as well as late parenthood have detrimental health
consequences for both genders. erefore, our Hypothesis3 is supported, while
the u-shaped relationship assumed in Hypothesis4 only for women in fact applies
to both genders. Only for chronic conditions we nd an almost linear relationship
for men indicating that health continuously improves with delaying childbirth.
e relationship of parity and later life health diers between birth cohorts for
women (Figure2). While for those born between 1920 and 1944 signicant health
dierences exist according to number of children, the relationship is insignicant
in the younger cohorts born between 1945 and 1959. For the older cohorts, the
later life health status is the better, the more children a women has. For men, the
number of children is neither in the older nor in the younger cohort signicantly
related to any later life health outcome. Figure1 shows the marginal eects of age
at rst child for both genders and all health outcomes separated for birth cohorts of
1920 to 1944 and 1945 to 1959. For all outcomes, signicant dierences between
the two cohorts exist indicating better health outcomes for the younger cohort.
e cohort dierences are smallest for depressive symptoms and largest for chronic
conditions for both genders. Apart from depressive symptoms for men, the nega-
tive eect of lower age at rst birth is smaller in the younger cohort. Accordingly,
Parenthood and Later Life Health 341
SJS 44 (2), 2018, 327–354
we nd support for Hypothesis6 stating that the detrimental health eects of high
parity and low age at rst birth are smaller in the younger cohort. Signicant dif-
ferences between individuals’ health according to number of children can only be
found among women born 1920 to 1944, and this eect mainly stems from the
worse later life health of childless women in this cohort.
An analysis separated for welfare state types gives more indication on the
prevalence of the u-shaped relationship between age at rst child and later life health.
Figure3 depicts the marginal eects of age at rst child for both genders and all
health outcomes separated for the four country groups Continental, Scandinavian,
Figure 1 Marginal effects of the health outcomes according to of age at first
child for different cohorts
Women
Chronic conditions Self-rated health Depression
Marginal effects
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Age at first child
45
35
25
15 20 30 40
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Age at first child
45
35
25
15 20 30 40
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Age at first child
45
35
25
15 20 30 40
Men
Chronic conditions Self-rated health Depression
Marginal effects
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Age at first child
45
35
25
15 20 30 40
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Age at first child
45
35
25
15 20 30 40
0.0
0.5
1.0
1.5
2.0
2.5
3.0
3.5
4.0
4.5
Age at first child
45
35
25
15 20 30 40
born 1920–1944 born 1945–1959
Notes: Regression models include all variables as presented in Table 1; for simplified presentation results for self-rated health
based on linear regression models.
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
342 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
Eastern and Southern European states. e dierences between these regions are
largest for self-rated health, and small for chronic conditions, especially women’s. In
Eastern and Southern European countries later life health appears to be generally on
a lower level, however, the relationship of fertility timing and health is less intense
in these countries. e pronounced u-shaped relationship of age at rst child and
health outcomes, which emerged in the combined regressions, applies only to women
in Continental and Scandinavian countries, while we nd rather linear relationships
in Eastern and Southern countries. For men, only in continental welfare states a
clear u-shaped pattern emerges for all health outcomes. Dierences in later life
health according to the number of children are insignicant for both genders in
all welfare state types with the only exception being depression among women in
Figure 2 Predicted values of the health outcomes according to number of
children for different cohorts
Women Men
born 1920–1944 born: 1945–1959 born 1920–1944 born: 1945–1959
1–2 children
–2 –1 0 1 2 –2 –1 0 1 2 –2 –1 0 1 2 –2 –1 0 1 2
born 1920–1944 born: 1945–1959 born 1920–1944 born: 1945–1959
3+ children
–2 –1 0 1 2 –2 –1 0 1 2 –2 –1 0 1 2 –2 –1 0 1 2
Chronic Subjective health Depression
Notes: Regression models include all variables as presented in Table 1.
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
Parenthood and Later Life Health 343
SJS 44 (2), 2018, 327–354
Continental countries. Here, childless women have an increased likelihood to suf-
fer from depressive symptoms (FiguresA.1 andA.2 in the Appendix). To sum up,
we do not nd clear evidence to support our last hypothesis that negative eects of
higher parity and lower age at rst birth are smaller in welfare states providing more
generous family policy and services. With respect to age at rst child, a clear age
structured pattern can only be found in Continental and Scandinavian countries.
e nal models we report in Table1 all include indicators for childhood health
and parents’ mortality to control to some degree for selection eects into fertility
behavior. For both genders and all later life health outcomes, the general health
status during childhood is a strong predictor: those who suered from bad health
Figure 3 Marginal effects of the health outcomes according to of age at first
child for different regions
Women
Chronic conditions Self-rated health Depression
Marginal effects
0
1
2
3
4
5
Age at first child
45403530252015
0
1
2
3
4
5
Age at first child
45403530252015
0
1
2
3
4
5
Age at first child
45403530252015
Men
Chronic conditions Self-rated health Depression
Marginal effects
0
1
2
3
4
5
Age at first child
45403530252015
0
1
2
3
4
5
Age at first child
45403530252015
0
1
2
3
4
5
Age at first child
45403530252015
Eastern Southern
Continental Scandinavian
Notes: Regression models include all variables as presented in Table 1; for simplified presentation results for self-rated health
based on linear regression models.
Source: Own calculations using SHARE waves 4-5 and SHARELIFE.
344 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
at young ages have an increased likelihood for worse physical and mental health in
later life than their counterparts with no health problems during childhood. We
also nd a positive signicant relationship between severe illnesses in childhood
(being conned to bed or home for one month or more) and chronic conditions
in later life. Furthermore, parents’ mortality is mostly signicantly related to their
children’s health in later life: if parents lived longer or are still alive, childrens health
status is usually better. Only for men’s mental health parents’ mortality is irrel-
evant and the relationship between women’s chronic conditions and their mothers’
mortality is reversed. e results indicate that genetic disposition (operationalized
by own parents’ longevity) and childhood conditions have a signicant impact on
an individuals’ later life health. However, none of the fertility history indicators
(parity and timing of rst birth) changes their signicance or eect direction after
including the variables for childhood and parents’ health. is gives some indica-
tion that selection eects do not play a large role or dierent types of selection
eects cancel each other out for the outcomes we are interested in. An exception
are the women of the older birth cohort: the fact that here childless women suer
from worse mental and physical health than their counterparts with children hints
to involuntary childlessness due to predispositions. However, generally it would
be necessary to have more detailed health history information covering the whole
life span to rule out all selectivity eects.
7 Conclusion
Based on life course data from the Survey of Health, Ageing and Retirement
(SHARE), we investigated the role of fertility history for later life health across two
birth cohorts and 13European countries. e central ndings are that parenthood
and the number of children have little impact on health in later life, while the tim-
ing of the rst birth is important across health outcomes for both men and women.
First, our results indicate that mothers are in better (mental) health in later life
than childless women, while there is no relationship for men. Mothers with three
or more children do not dier from those with one or two children. is conrms
earlier research which found either no eects or benets of parenthood for later
life health (e. g., Eggebeen and Knoester 2001; Teachman 2010, see section3.1).
Previous research also indicated that parenthood eects vary across countries (e. g.
Grundy 2009). Indeed, our study shows that the health benets of parenthood
are limited to the older cohort (born between 1920 and 1945) and mothers in
Continental Europe.
Second, fertility timing is important for all three health outcomes. Delaying
childbirth until 30 years at age of rst birth is benecial for both men and women,
but more so for mothers. However, late childbearing ( > 35) is also more detrimental
Parenthood and Later Life Health 345
SJS 44 (2), 2018, 327–354
to women’s than to men’s health. e lesson we learn from this– conrming previ-
ous research (e. g. Mirowsky 2002; Mirowsky 2005, see section 3.4), is that delaying
childbirth until the 30s is benecial to parents’ later life health. Considering that in
our sample,55%of the women and 25% of the men had their rst child before25,
we might expect that the demographic trends towards later childbirth since the 1960s
(2014 the mean age of women at rst birth was28.9 in the European Union), has
been a positive development with respect to population health.
Finally, our study adds to the literature insights that can be drawn from
comparing the timing eect across dierent health outcomes, cohorts, and welfare
regimes. On the one hand, this comparison suggests that rst births between25
and30 for women and (less consistently) between25 and35 for men are associ-
ated with the best later life health outcomes in almost all of our analysis groups.
is universal pattern indicates that biological mechanisms partially account for
the relationship between timing and later life health, particularly for mothers. On
the other hand, there is substantial variation in the degree to which timing matters
for health across contexts. is indicates that social mechanisms play a powerful
role in this relationship. In terms of historical context, we nd that timing of rst
birth has become less important in younger cohorts. is suggests– in line with
our theoretical expectations– that in times of greater prosperity, social security and
availability of safe maternity care an early age at rst birth will be less problematic
for later life health. Our study did not directly test the relative importance of bio-
logical, social, and selection mechanisms. However, the wide variation that timing
eects have across contexts and the decreased importance over historical time sug-
gests that the inevitable health costs of both teenage and late childbearing should
not be exaggerated (Furstenberg 2007), since the social context strongly aects their
implications for health. Identifying the underlying institutional and normative vari-
ables of these context eects will provide insights into the causes of timing eects
and potentially opportunities for social intervention. However, one also needs to
consider that paternal age has also other implications, most importantly probably
the health and developmental outcomes of children. Children from teenage parents
are more likely to have poorer behavioral and psychological outcomes (Shaw etal.
2006). A higher maternal age has mixed eects for children. On the one hand it
increases the risks for chromosomal abnormalities and adverse neonatal outcomes
for a small group of children (Kozuki etal. 2013; Restrepo-Méndez etal. 2015),
but is generally associated with better cognitive ability of children in more recent
cohorts (Goisis etal. 2017).
Our aim was to explore the role of one contextual variable, welfare state
arrangements, in our country-comparative analysis of timing eects. However,
the patterns that we found across countries were more complex and did not con-
rm our expectations that the generosity of parental leave and childcare services
are the primary mechanism behind cross-national variation in timing eects. For
346 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
women, timing matters to a much larger degree in Continental and Scandinavian
countries than in Eastern or Southern Europe. For men, timing matters mostly
in Continental Europe, while for the other country groups the results vary across
outcomes, but mostly show less impact of timing. e implication of the result
that there are stronger dierences between Continental/Scandinavian compared to
Eastern/Southern European countries might suggest two venues for further research
of context eects. First, fertility trends in the number and timing of children
developed quite dierently in these two country groups than in Continental Europe
and Scandinavia which show similar trends as the Anglo-Saxon countries which
are most researched for parenthood health eects (Tomka 2013; Freijka 2016).
Second, Eastern and Southern European countries had political dictatorships after
WWII, while Continental Europe and Scandinavia were democratic. To gain deeper
insights into the relative importance of these factors and welfare state development,
it seems particularly important for further research to dierentiate between country
groups and cohorts and if possible include younger cohorts, e. g. who experienced
democratization in Southern Europe.
e results of our study are restricted by several limitations. First, as we rely
on cross-sectional information on the health outcomes, we are unable to track health
changes throughout later life. So, our study cannot inform about the change and
duration of specic health conditions in later life. Second, the life course informa-
tion provided in SHARELIFE may suer from retrospective memory bias. e
respondents who were between the age of50 and91 were asked to recall information
from the period that they were15 to49, which may not be remembered accurately.
However, the approach using calendar interviews in SHARELIFE is likely to have
limited this bias (Schröder, 2011). Finally, our results may be biased due to selec-
tive mortality.
8 References
Aassve, Arnstein, Stefano Mazzuco, and Letizia Mencarini. 2005. Childbearing and Well-Being: A
Comparative Analysis of European Welfare Regimes. Journal of European Social Policy 15(4):
283–299.
Aitken, Zoe, Cameryn C. Garrett, Belinda Hewitt, Louise Keogh, Jane S. Hocking, and Anne M.
Kavanagh. 2015. e Maternal Health Outcomes of Paid Maternity Leave: A Systematic Review.
Social Science & Medicine 130: 32–41.
Allison, Paul D. 2009. Fixed Eects Regression Models. ousand Oaks: SAGE Publications.
Arnett, Jerey J. 1998. Risk Behavior and Family Role Transitions During the Twenties. Journal of Youth
and Adolescence 27(3): 301–320.
Bambra, Clare. 2006. Health Status and the Worlds of Welfare. Social Policy and Society 5(1): 53–62.
Barban, Nicola. 2013. Family Trajectories and Health: A Life Course Perspective. European Journal of
Population/Revue européenne de Démographie 29(4): 357–385.
Parenthood and Later Life Health 347
SJS 44 (2), 2018, 327–354
Beckeld, Jason, Clare Bambra, Terje A. Eikemo, Tim Huijts, Courtney McNamara, and Claus Wendt.
2015. An Institutional eory of Welfare State Eects on the Distribution of Population Health.
Social eory & Health 13(3–4): 227–244.
Ben-David, Alon, Saralee Glassner, Eyal Schi, Aliza S. Zahav, Valentina Boyko, and Liat Lerner-Geva.
2016. Pregnancy and Birth Outcomes Among Primaparae at Very Advanced Maternal Age: At
What Price? Maternal and Child Health Journal 20: 833–842.
Billari, Francesco C., Hans-Peter Kohler, Gunnar Andersson, and Hans Lundström. 2007. Approaching
the Limit: Long-Term Trends in Late and Very Late Fertility. Population and Development Review
33(1): 149–170.
Börsch-Supan, Axel, Martina Brandt, Christian Hunkler, orsten Kneip, Julie Korbmacher, Frederic
Malter, Barbara Schaan, Stephanie Stuck, and Sabrina Zuber. 2013. Data Resource Prole: e
Survey of Health, Ageing and Retirement in Europe (SHARE). International Journal of Epide-
miology 42(4): 992–1001.
Breart, Gérard. 1997. Delayed Childbearing. European Journal of Obstetrics & Gynecology and Reproductive
Biology 75(1): 71–73.
Brockmann, Hilke and omas Klein. 2004. Love and Death in Germany: e Marital Biography and
Its Eect on Mortality. Journal of Marriage and Family 66(3): 567–581.
Burton-Jeangros, Claudine, Stéphane Cullati, Amanda Sacker, and David Blane (eds.). 2015. A Life
Course Perspective on Health Trajectories and Transitions. Cham: Springer.
Carlson, Daniel L. 2011. Explaining the Curvilinear Relationship Between Age at First Birth and
Depression Among Women. Social Science & Medicine 72(4): 494–503.
Castles, Francis G. and Herbert Obinger. 2008. Worlds, Families, Regimes: Country Clusters in European
and OECD Area Public Policy. West European Politics 31(1/2): 321–344.
Curtis, Lori and Shelley Phipps. 2004. Social Transfers and the Health Status of Mothers in Norway and
Canada. Social Science & Medicine 58(12): 2499–2507.
DiPrete, omas A. and Gregory M. Eirich. 2006. Cumulative Advantage as a Mechanism for Inequality:
A Review of eoretical and Empirical Developments. Annual Review of Sociology 32(1): 271–297.
Eggebeen, David J. and Chris Knoester. 2001. Does Fatherhood Matter for Men? Journal of Marriage
and Family 63(2): 381–393.
Freijka, Tomas. 2016. e Demographic Transition Revisited: A Cohort Perspective. Max Planck Institute
for Demographic Research Working Paper, Rostock.
Furstenberg, Frank F. 2007. Destinies of the Disadvantaged: e Politics of Teen Childbearing. New York:
Russell Sage Foundation.
Furstenberg Jr., Frank F., Jeanne Brooks-Gunn, and S. Philip Morgan. 1987. Adolescent Mothers in Later
Life. New York, NY, US: Cambridge University Press.
Gibney, Sarah, Liam Delaney, Mary Codd, and Tony Fahey. 2015. Lifetime Childlessness, Depressive
Mood and Quality of Life Among Older Europeans. Social Indicators Research 130(1): 1–19.
Goisis, Alice, Daniel C. Schneider, and Mikko Myrskylä. 2017. e Reversing Association Between
Advanced Maternal Age and Child Cognitive Ability: Evidence From ree UK Birth Cohorts.
International Journal of Epidemiology 46(3): 850–859.
Goode, William J. 1960. A eory of Role Strain. American Sociological Review 25(4): 483–496.
Grundy, Emily. 2009. Women’s Fertility and Mortality in Late Mid Life: A Comparison of ree
Contemporary Populations. American Journal of Human Biology: e Ocial Journal of the Human
Biology Council 21: 541–547.
Grundy, Emily and Cecilia Tomassini. 2005. Fertility History and Health in Later Life: A Record Linkage
Study in England and Wales. Social Science & Medicine 61(1): 217–228.
Hank, Karsten. 2010. Childbearing History, Later-Life Health, and Mortality in Germany. Population
Studies 64(3): 275–291.
348 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
Hansen, omas. 2012. Parenthood and Happiness: A Review of Folk eories Versus Empirical Evidence.
Social Indicators Research 108(1): 29–64.
Helbig, Sylvia, omas Lampert, Michael Klose, and Frank Jacobi. 2006. Is Parenthood Associated
With Mental Health? Findings From an Epidemiological Community Survey. Social Psychiatry
and Psychiatric Epidemiology 41(11): 889–896.
Henretta, John C. 2007. Early Childbearing, Marital Status, and Women’s Health and Mortality After
Age 50. Journal of Health and Social Behavior 48(3): 254–266.
Huijts, Tim, Gerbert Kraaykamp, and Subbaya V. Subramanian. 2013. Childlessness and Psychological
Well-Being in Context: A Multilevel Study on 24 European Countries. European Sociological
Review 29(1): 32–47.
Hurt, Lisa S., Carine Ronsmans, and Sara L. omas. 2006. e Eect of Number of Births on
Women’s Mortality: Systematic Review of the Evidence for Women Who Have Completed eir
Childbearing. Population Studies 60(1): 55–71.
Kennedy, Patricia, Naonori Kodate, and Nadine Reibling. 2015. Welfare Regimes, Healthcare Regimes
and Maternity Services and Policy: A Comparative Perspective. Pp. 4–19 in Maternity Services
and Policy in an International Context: Risk, Citizenship, and Welfare Regimes, edited by Patricia
Kennedy and Naonori Kodate. Abingdon: Routledge.
Kington, Raynard, Lee Lillard, and Jeannette Rogowski. 1997. Reproductive History, Socioeconomic
Status, and Self-Reported Health Status of Women Aged 50 Years or Older. American Journal of
Public Health 87(1): 33–37.
Kohler, Hans-Peter, Jere R. Behrman, and Axel Skytthe. 2005. Partner Plus Children = Happiness? e
Eects of Partnerships and Fertility on Well-Being. Population and Development Review 31(3):
407–445.
Kozuki, Naoko, Anne C. C. Lee, Mariangela F. Silveira, Ayesha Sania, Joshua P. Vogel, Linda Adair,
Fernando Barros, Laura E. Cauleld, Parul Christian, Wafaie Fawzi, Jean Humphrey, Lieven
Huybregts, Aroonsri Mongkolchati, Robert Ntozini, David Osrin, Dominique Roberfroid, James
Tielsch, Anjana Vaidya, Robert E. Black, Joanne Katz, and Child Health Epidemiology Reference
Group (CHERG) Small-for-Gestational Age Preterm Birth Working Group. 2013. e Associa-
tions of Parity and Maternal Age With Small-for-Gestational Age, Preterm, and Neonatal and
Infant Mortality: A Meta-Analysis. BMC Public Health 13(3): 5.
Kroll, Lars Eric, Stephan Mueters, Petra Rattay, and omas Lampert. 2016. Employment, Family
and Health in Men of Working Age in Germany Results of the GEDA Studies 2009–2012.
Bundesgesundheitsblatt-Gesundheitsforschung-Gesundheitsschutz 59(8): 932–941.
Lacey, Rebecca, Amanda Sacker, Steven Bell, Meena Kumari, Diana Worts, Peggy McDonough, Diana
Kuh, and Anne McMunn. 2017. Work-Family Life Courses and BMI Trajectories in ree British
Birth Cohorts. International Journal of Obesity 41: 332–339.
Lisonkova, Sarka, Jayson Potts, Giulia M. Muraca, Neda Razaz, Yessar Sabr, Wee-Shian Chan, and
Michael S. Kramer. 2017. Maternal Age and Severe Maternal Morbidity: A Population-Based
Retrospective Cohort Study. PLoS Medicine 14(5): e1002307.
Loudon, Irvine. 2000. Maternal Mortality in the Past and Its Relevance to Developing Countries Today.
e American Journal of Clinical Nutrition 72(1): 241–246.
Maas, Cora J. M, and Joop J. Hox. 2005. Sucient Sample Sizes for Multilevel Modeling. Methodology
1(3): 86–92.
Marks, Stephen R. 1977. Multiple Roles and Role Strain: Some Notes on Human Energy, Time and
Commitment. American Sociological Review 42(6): 921–936.
Mayer, Karl Ulrich. 2009. New Directions in Life Course Research. Annual Review of Sociology 35(1):
413–433.
Merrill, Ray M., Stephanie Fugal, Lelinneth B. Novilla, and Matthew C. Raphael. 2005. Cancer Risk
Associated With Early and Late Maternal Age at First Birth. Gynecologic Oncology 96(3): 583–593.
Parenthood and Later Life Health 349
SJS 44 (2), 2018, 327–354
Mirowsky, John. 2005. Age at First Birth, Health, and Mortality. Journal of Health and Social Behavior
46(1): 32–50.
Mirowsky, John and Catherine E Ross. 2002. Depression, Parenthood, and Age at First Birth. Social
Science & Medicine 54(8): 1281–1298.
Mirowsky, John. 2002. Parenthood and Health: e Pivotal and Optimal Age at First Birth. Social
Forces 81(1): 315–349.
Möhring, Katja. 2012. e Fixed Eects Approach as Alternative to Multilevel Models for Cross-National
Analyses. Paper presented at the 10th ESPAnet Conference, Edinburgh, Scotland, September 6–8,
2012.
Muñoz, Nubia, Silvia Franceschi, Cristina Bosetti, Victor Moreno, Rolando Herrero, Jennifer S. Smith,
Keerti V. Shah, Chris J. L. M. Meijer, and F. Xavier Bosch. 2002. Role of Parity and Human
Papillomavirus in Cervical Cancer: e IARC Multicentric Case-Control Study. e Lancet
359(9312): 1093–1101.
Nauck, Bernhard, Chin-Chun Yi, Johannes Huinink, and Michael Feldhaus. 2009. Family Research
From the Life Course Perspective. International Sociology 24(3): 299–324.
Nomaguchi, Kei M. and Suzanne M. Bianchi. 2004. Exercise Time: Gender Dierences in the Eects
of Marriage, Parenthood, and Employment. Journal of Marriage and Family 66(2): 413–430.
Nomaguchi, Kei M. and Melissa A. Milkie. 2003. Costs and Rewards of Children: e Eects of Becoming
a Parent on Adults’ Lives. Journal of Marriage and Family 65(2): 356–374.
O’Flaherty, Martin, Janeen Baxter, Michele Haynes, and Gavin Turrell. 2015. Age at First Birth and
Long-Term Health in Life Course Perspective: Contextualizing the Health Eects of Fertility Timing.
e Australian Research Council Life Course Centre Working Paper Series 28.
Orlo, Ann Shola. 1993. Gender and the Social Rights of Citizenship: e Comparative Analysis of
Gender Relations and Welfare States. American Sociological Review 58(3): 303–328.
Ozalp, Sinan, Huseyin Mete Tanir, Turgay Sener, Sibel Yazan, and Ali E. Keskin. 2003. Health Risks for
Early ( ≤ 19) and Late ( ≥ 35) Childbearing. Archives of Gynecology and Obstetrics 268(3): 172–174.
Prince, Martin J., Reischies F., A. T. F. Beekman, R. Fuhrur, C. Jonker, S. L. Kivela, B. A. Lawtor, A. Lobo,
H. Magnusson, M. Fichter, H. van Oyen, M. Roelands, I. Skoog, C. Turrina, J. R. M. Copeland.
1999. Development of the EURO-D Scale– a European, Union Initiative to Compare Symp-
toms of Depression in 14 European Centres. e British Journal of Psychiatry 174(4): 330–338.
Read, Sanna and Emily Grundy. 2011. Fertility History and Quality of Life in Older Women and Men.
Ageing & Society 31(1): 125–145.
Read, Sanna, Emily Grundy, and Douglas A. Wolf. 2011. Fertility History, Health, and Health Changes
in Later Life: A Panel Study of British Women and Men Born 1923–1949. Population Studies
65(2): 201–215.
Restrepo-Méndez, Maria C., Debbie A. Lawlor, Bernardo L. Horta, Alicia Matijasevich, Iná S. Santos,
Ana M. B. Menezes, Fernando, C. Barros, and Cesar G. Victora. 2015. e Association of
Maternal Age With Birthweight and Gestational Age: A Cross-Cohort Comparison. Paediatric
and Perinatal Epidemiology 29: 31–40.
Russo, Jose, Raquel Moral, Gabriela A. Balogh, Daniel Mailo, and Irma H. Russo. 2005. e Protective
Role of Pregnancy in Breast Cancer. Breast Cancer Research: BCR 7(3): 131–142.
Scalone, Francesco. 2014. Eects of Nutritional Stress and Socio-Economic Status on Maternal Mortality
in Six German Villages, 1766–1863. Population Studies 68(2): 217–236.
Schröder, M. (ed.). 2011. Retrospective Data Collection in the Survey of Health, Ageing and Retirement in
Europe. SHARELIFE Methodology. Mannheim: Mannheim Research Institute for the Economics
of Aging (MEA).
350 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
Shaw, Mary, Debbie A. Lawlor, and Jake M. Najman. 2006. Teenage Children of Teenage Mothers:
Psychological, Behavioral, and Health Outcomes From an Australian Prospective Longitudinal
Study. Social Science & Medicine 62(10): 2526–2539.
Sieber, Sam D. 1974. Toward a eory of Role Accumulation. American Sociological Review 39(4): 567–578.
Spence, Naomi J. 2008. e Long-Term Consequences of Childbearing: Physical and Psychological
Well-Being of Mothers in Later Life. Research on Aging 30(6): 722–751.
Tanaka, Kimiko and Nan E. Johnson. 2014. Childlessness and Mental Well-Being in a Global Context.
Journal of Family Issues 37(8): 1027–1045.
Taylor, Julie Lounds. 2009. Midlife Impacts of Adolescent Parenthood. Journal of Family Issues 30(4):
484–510.
Teachman, Jay. 2010. Family Life Course Statuses and Transitions: Relationships With Health Limitations.
Sociological Perspectives 53(2): 201–219.
Tomka, Béla. 2013. A Social History of Twentieth-Century Europe. Abingdon: Routledge.
Umberson, Debra, Tetyana Pudrovska, and Corinne Reczek. 2010. Parenthood, Childlessness, and
Well-Being: A Life Course Perspective. Journal of Marriage and Family 72(3): 612–629.
Westendorp, Rudi. G. and omas B. Kirkwood. 1998. Human Longevity at the Cost of Reproductive
Success. Nature 396(6713): 743–746.
Yi, Zeng and James Vaupel. 2004. Association of Late Childbearing With Healthy Longevity Among
the Oldest-Old in China. Population Studies 58(1): 37–53.
Parenthood and Later Life Health 351
SJS 44 (2), 2018, 327–354
9 Appendix
Table A.1 Summary statistics for the variables used for the
statistical estimations
Variables Women Men
Mean Standard
Deviation
Minimum Maximum Mean Standard
Deviation
Minimum Maximum
Number of chronic diseases 1.90 1.60 013 1.68 1.47 010
Self-perceived health 3.15 1.07 1 5 3.08 1.05 1 5
EURO depression scale 2.71 2.34 012 1.87 1.99 012
Age at 1st Child 24.78 4.41 15 46 27.73 4.74 17 55
Age at 1st Child squared 633.60 238.26 225 2 116 791.55 289.55 289 3 025
Number of children
No children 0.01 0.09 0 1 0.01 0.10 0 1
1–2 children 0.62 0.48 0 1 0.61 0.49 0 1
3+ children 0.37 0.48 0 1 0.38 0.49 0 1
Childhood health status 2.09 1.01 1 5 2.03 1.00 1 5
Parents smoked 0.62 0.48 0 1 0.66 0.48 0 1
confined to bed/
home 1+ month
0.11 0.32 0 1 0.11 0.31 0 1
Mother‘s survival status
Died age 60–74 0.11 0.31 0 1 0.11 0.31 0 1
Died age 60–74 0.18 0.39 0 1 0.19 0.39 0 1
Died age 75+/
Not deceased
0.71 0.46 0 1 0.71 0.45 0 1
Father‘s survival status
Died age 60–74 0.18 0.38 0 1 0.18 0.39 0 1
Died age 60–74 0.29 0.46 0 1 0.30 0.46 0 1
Died age 75+/
Not deceased
0.53 0.50 0 1 0.51 0.50 0 1
Age 67.90 9.04 52 91 68.94 8.58 52 91
Birth cohort
Born 1920–1944 0.51 0.50 0 1 0.57 0.50 0 1
Born 1945–1959 0.49 0.50 0 1 0.43 0.50 0 1
Log. HH income 9.95 1.34 1.73 13.86 10.18 1.24 1.73 13.86
Years of Education 10.04 4.49 025 10.88 4.87 025
Employed 0.17 0.38 0 1 0.18 0.38 0 1
Owner 0.72 0.45 0 1 0.78 0.42 0 1
Married 0.66 0.47 0 1 0.85 0.36 0 1
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
352 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
Figure A.1 Predicted values of the health outcomes according to number of
children for men in different regions
Men
Continental Scandinavian Southern Eastern
1–2 children
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
Continental Scandinavian Southern Eastern
3+ children
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
Chronic Subjective health Depression
Notes: Regression models include all variables as presented in Table 1.
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
Parenthood and Later Life Health 353
SJS 44 (2), 2018, 327–354
Figure A.2 Predicted values of the health outcomes according to number of
children for women in different regions
Women
Continental Scandinavian Southern Eastern
1–2 children
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
Continental Scandinavian Southern Eastern
3+ children
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
–4 –3 –2 –1 0 1 2 3
Chronic Subjective health Depression
Notes: Regression models include all variables as presented in Table 1.
Source: Own calculations using SHARE waves 4-5 and SHARELIFE.
Variables Women Men
Mean Standard
Deviation
Minimum Maximum Mean Standard
Deviation
Minimum Maximum
Owner 0.72 0.45 0 1 0.78 0.42 0 1
Married 0.66 0.47 0 1 0.85 0.36 0 1
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
354 Nadine Reibling and Katja Möhring
SJS 44 (2), 2018, 327–354
Table A.2 Regressions of fertility history on later life health outcomes for
women and men aged 50 and above (coefficients for self-rated
health and depression, marginal effects for chronic conditions)
with interaction effects of age in groups and number of children
Variables Women Men
Chronic
conditions
Self-rated
health
Depression Chronic
conditions
Self-rated
health
Depression
Age groups
(RC: Age 50–64)
Age 65–79 0.928
(0.498)
0.968
(0.627)
1.836*
(0.750)
0.040
(0.464)
0.922
(0.595)
–0.141
(0.609)
Age 80–91 1.589***
(0.457)
1.843**
(0.562)
2.645***
(0.706)
–0.248
(0.523)
0.088
(0.615)
–0.301
(0.668)
Number of children
(RC: No children)
ref. ref. ref. ref. ref. ref.
1–2 children 0.236
(0.400)
0.114
(0.402)
0.610
(0.525)
0.132
(0.293)
–0.005
(0.349)
–0.295
(0.374)
3+ children 0.291
(0.402)
0.212
(0.405)
0.774
(0.528)
0.118
(0.296)
0.060
(0.353)
–0.156
(0.378)
Interaction effects
1–2 children * Age 65–79 –0.482
(0.500)
–0.661
(0.630)
–1.667*
(0.753)
0.235
(0.467)
–0.793
(0.599)
0.279
(0.613)
1–2 children * Age 80–91 –0.808
(0.461)
–0.785
(0.568)
–1.917**
(0.713)
0.835
(0.528)
0.864
(0.625)
1.185+
(0.677)
3+ children * Age 65–79 –0.524
(0.502)
–0.766
(0.632)
–1.797*
(0.756)
0.231
(0.469)
–0.762
(0.602)
0.224
(0.616)
3+ children * Age 80–91 –0.861
(0.464)
–0.930
(0.572)
–2.174**
(0.718)
0.767
(0.530)
0.807
(0.627)
1.078
(0.680)
Age at 1st Child –0.121***
(0.031)
–0.219***
(0.042)
–0.209***
(0.050)
–0.094***
(0.027)
–0.104**
(0.039)
–0.178***
(0.040)
Age at 1st Child squared 0.002***
(0.001)
0.004***
(0.001)
0.003***
(0.001)
0.001**
(0.000)
0.002**
(0.001)
0.003***
(0.001)
Further regression output omitted
N6 242 6 242 6 242 5 577 5 577 5 577
R squared 0.113 0.077
Adjusted R squared 0.109 0.072
Notes: Standard errors in parentheses; *** p 0.001, ** p 0.01, * p 0.05; including all control variables as in main regressions
(Table 1); model specification varies by the metric of the outcome variable: chronic conditions: negative binomial (marginal
effects reported), self-rated health: ordered logit; depression scale: OLS; higher values indicate worse health.
Source: Own calculations using SHARE waves 4–5 and SHARELIFE.
Arnaud Frauenfelder
Eva Nada
Géraldine Bugnon
Ce qu’enfermer des
jeunes veut dire
Enquête dans un centre
éducatif fermé
www.editions-seismo.ch
info@editions-seismo.ch
Cet ouvrage revisite la « question carcérale » en
décortiquant ses enjeux contemporains. En Suisse
comme en Europe, la délinquance juvénile attise les
sensibilités publiques et nourrit les discours sécu-
ritaires. Parallèlement, les mineur·e·s sont progres-
sivement reconnus comme des sujets de droit, dont
il faut protéger l’intégrité physique et morale. Ce
contexte politique et moral contraint les institutions
d’enfermement à garantir plus de « dignité » dans
la vie quotidienne des jeunes (réduction des temps
d’encellulement, prise en charge pluridisciplinaire),
tout en imposant davantage de « fermeté » (ren-
forcement des aménagements sécuritaires, sanc-
tions disciplinaires).
Les auteur·e·s explorent cette ambivalence à partir
d’une enquête de terrain réalisée dans un centre
éducatif fermé de Suisse romande. Comment les
acteurs professionnels s’approprient-ils leur espace
de travail ? Quelle raison d’être confèrent-ils à leur
mission professionnelle ? L’ouvrage analyse les
rivalités de territoires, met en évidence la diversité
des conceptions éducatives et les différents rap-
240 pages, SFr. 32.—
ports de l’institution à l’environnement extérieur. Il
s’attache ainsi à saisir les formes de recomposition
de l’économie morale de l’enfermement des jeunes.
Cette analyse sociologique de la justice pénale des
mineur·e·s « par le bas » souligne combien les pra-
tiques de l’État, dans cet univers particulier, s’expri-
ment d’abord par le travail de ses agents.
Arnaud Frauenfelder, sociologue, est professeur ordi-
naire à la Haute école de travail social, HETS
(HES-SO/Genève) et responsable du Centre de recher-
che sociale (CERES).
Eva Nada, doctorante en sociologie, est adjointe
scientifique à la Haute école de travail social, HETS
(HES-SO/Genève) et chercheuse associée à l’Institut
de sociologie (UNINE).
Géraldine Bugnon, docteure en sociologie, est adjoin-
te scientifique à la Haute école de travail social, HETS
(HES-SO/Genève), chercheuse associée au Centre
romand de recherche en criminologie (UNINE) et à
l’Institut de recherches sociologiques (UNIGE).
Terrains des sciences sociales
... This is analogous to the concept of health selection, prevalent in social science studies of the cost of reproduction, whereby reproducing individuals are a non-random subset of the population with respect to health, and individuals with many children may further represent a robust subset of very healthy individuals, potentially obscuring trade-offs. This may be more important in historic populations, and less relevant in contemporary populations with access to modern contraception (Hurt, Ronsmans, & Thomas, 2006;Reibling & Möhring, 2018). ...
... Such meta analyses allow simultaneous comparison of studies from different disciplines, theoretical, and methodological approaches. A study of the relationship between reproduction and health in later life across 13 European countries, found that number of children born had little effect on health in later life, but fertility timing was important, with differences between countries and cohorts suggesting the importance of the socio-historic context (Reibling & Möhring, 2018). Similarly, Sironi (2019) found that, across 11 European countries, age at first birth was more relevant than parity for health outcomes in later life. ...
... Studies of the cost of reproduction in humans have variously focused on biological explanations (the disposable soma theory, section 2), social mechanisms (that influence both physical and mental health of parents) and selection processes (such as health selection, whereby reproduction is not random with respect to the health of individuals, reviewed in Reibling & Möhring 2018). The interplay between these three is difficult to untangle in human studies, but in one recent attempt, Barclay and Kolk (2019) set out to distinguish between physiological and social explanations for a cost of reproduction in both sexes, by comparing biological and adoptive parents in contemporary Sweden, and found that the mortality risk for adoptive parents was always lower than for biological parents, with a U-shaped pattern between parity and mortality in biological parents. ...
Article
Full-text available
Life history theory has become a prominent framework in the evolutionary social sciences, and the concept of trade-offs, the cornerstone of life history theory in studies on non-human taxa, has likewise been widely adopted. Yet, human life history research often assumes trade-offs without demonstrating them. This is not surprising given the practical difficulties in measuring trade-offs in long-lived animals, like humans. Four main methods are used to demonstrate trade-offs: phenotypic correlations, experimental manipulations, genetic correlations and correlated responses to selection. Here, I discuss challenges with these methods along with potential solutions. For example, individual heterogeneity within a population in quality or access to resources can mask underling trade-offs, and this can be accounted for by careful experimental manipulation or proper statistical treatment of observational data. In general, trade-offs have proven more difficult than expected to measure, and evidence across species is mixed, but strong evidence exists in some cases. I use the key trade-off between reproduction and survival to exemplify methods, challenges and solutions, and review the mixed evidence for a cost of reproduction in humans. I conclude by providing directions for future research. Promising avenues are opening thanks to recent advances in quantitative genetic and genomic methods coupled with the availability of high-quality large-scale datasets on humans from different populations, allowing the study of the evolutionary implications of life history trade-offs in humans.
Conference Paper
Full-text available
Multilevel models which combine individual and contextual factors are increasingly popular in comparative social science research. However, their application in country-comparative studies is often associated with several problems. First of all, some data-sets utilized for multilevel modelling include only a small number (N<25) of macro level units, and therefore, the estimated models have a small number of degrees of freedom on the country level. If models are correctly specified paying regard to the small level-2 N, only few macro level indicators can be controlled for. Furthermore, the introduction of random slopes and cross-level interaction effects is then hardly possible. Consequently, (1.) these models are likely to suffer from omitted variable bias regarding the country level estimators and (2.) the advantages of multilevel modelling cannot be fully exploited. The fixed effects approach is a valuable alternative to the application of conventional multilevel methods in country-comparative analyses. This method is also applicable with a small number of countries and avoids the omitted variable bias through controlling for country level heterogeneity. Following common practice in panel regression analyses, the moderator effect of macro-level characteristics can be estimated also in fixed effects models by means of cross-level interaction effects. Despite the advantages of the fixed effects approach, up to now it is rarely used for the analysis of cross-national data. In this paper, I compare the fixed effects approach with conventional multilevel regression models and give practical examples using data of the International Social Survey Programme (ISSP) from 2006. As it turns out, the results of both approaches regarding the effect of cross-level interactions are similar. Thus, fixed effects models can be used either as an alternative to multilevel regression models or to assess the robustness of multilevel results.
Article
Full-text available
Background One of the United Nations’ Millennium Development Goals of 2000 was to reduce maternal mortality by 75% in 15 y; however, this challenge was not met by many industrialized countries. As average maternal age continues to rise in these countries, associated potentially life-threatening severe maternal morbidity has been understudied. Our primary objective was to examine the associations between maternal age and severe maternal morbidities. The secondary objective was to compare these associations with those for adverse fetal/infant outcomes. Methods and findings This was a population-based retrospective cohort study, including all singleton births to women residing in Washington State, US, 1 January 2003–31 December 2013 (n = 828,269). We compared age-specific rates of maternal mortality/severe morbidity (e.g., obstetric shock) and adverse fetal/infant outcomes (e.g., perinatal death). Logistic regression was used to adjust for parity, body mass index, assisted conception, and other potential confounders. We compared crude odds ratios (ORs) and adjusted ORs (AORs) and risk differences and their 95% CIs. Severe maternal morbidity was significantly higher among teenage mothers than among those 25–29 y (crude OR = 1.5, 95% CI 1.5–1.6) and increased exponentially with maternal age over 39 y, from OR = 1.2 (95% CI 1.2–1.3) among women aged 35–39 y to OR = 5.4 (95% CI 2.4–12.5) among women aged ≥50 y. The elevated risk of severe morbidity among teen mothers disappeared after adjustment for confounders, except for maternal sepsis (AOR = 1.2, 95% CI 1.1–1.4). Adjusted rates of severe morbidity remained increased among mothers ≥35 y, namely, the rates of amniotic fluid embolism (AOR = 8.0, 95% CI 2.7–23.7) and obstetric shock (AOR = 2.9, 95% CI 1.3–6.6) among mothers ≥40 y, and renal failure (AOR = 15.9, 95% CI 4.8–52.0), complications of obstetric interventions (AOR = 4.7, 95% CI 2.3–9.5), and intensive care unit (ICU) admission (AOR = 4.8, 95% CI 2.0–11.9) among those 45–49 y. The adjusted risk difference in severe maternal morbidity compared to mothers 25–29 y was 0.9% (95% CI 0.7%–1.2%) for mothers 40–44 y, 1.6% (95% CI 0.7%–2.8%) for mothers 45–49 y, and 6.4% for mothers ≥50 y (95% CI 1.7%–18.2%). Similar associations were observed for fetal and infant outcomes; neonatal mortality was elevated in teen mothers (AOR = 1.5, 95% CI 1.2–1.7), while mothers over 29 y had higher risk of stillbirth. The rate of severe maternal morbidity among women over 49 y was higher than the rate of mortality/serious morbidity of their offspring. Despite the large sample size, statistical power was insufficient to examine the association between maternal age and maternal death or very rare severe morbidities. Conclusions Maternal age-specific incidence of severe morbidity varied by outcome. Older women (≥40 y) had significantly elevated rates of some of the most severe, potentially life-threatening morbidities, including renal failure, shock, acute cardiac morbidity, serious complications of obstetric interventions, and ICU admission. These results should improve counselling to women who contemplate delaying childbirth until their forties and provide useful information to their health care providers. This information is also useful for preventive strategies to lower maternal mortality and severe maternal morbidity in developed countries.
Article
Full-text available
Background: Studies on advanced maternal age-defined here as age 35 or older-and children's cognitive ability report mixed evidence. Previous studies have not analysed how the time period considered in existing studies influences the association. Methods: We analysed trends in the association between maternal age and cognitive ability using data from the 1958 National Child Development Study ( n = 10 969), the 1970 British Cohort Study ( n = 9362) and the 2000-2002 Millennium Cohort Study ( n = 11 600). The dependent variable measures cognitive ability at age 10/11 years. Cognitive scores were standardised to a mean of zero and a standard deviation of one. Results: For the 1958-70 cohort studies, maternal ages 35 -39 were negatively associated with children's cognitive ability compared with maternal ages 25-29 (1958 cohort β = -0.06 standard deviations (SD) 95% confidence interval (CI): -0.13, -0.00; 1970 cohort β = -0.12 SD 95% CI: -0.20, -0.03). By contrast, for the 2000-2002 cohort study maternal ages 35-39 were positively associated with cognitive ability (β = 0.16 SD 95% CI: 0.09, 0.23). For maternal ages 40+, the pattern was qualitatively similar. These cross-cohort differences were explained by the fact that in the earlier cohorts advanced maternal age was associated with high parity, whereas in the 2000-2002 cohort it was associated with socioeconomically advantaged family background. Conclusions: The association between advanced maternal age and children's cognitive ability changed from negative in the 1958 and 1970 cohorts to positive in the 2000-2002 cohort because of changing parental characteristics. The time period considered can constitute an important factor in determining the association between maternal age and cognitive ability.
Article
Full-text available
BACKGROUND/OBJECTIVES Combining work and family responsibilities has previously been associated with improved health in mid-life, yet little is known about how these associations change over time (both biographical and historical) and whether this extends to body mass index (BMI) trajectories for British men and women. The purpose of this study was to investigate relationships between work-family life courses and BMI trajectories across adulthood (16–42 years) for men and women in three British birth cohorts. SUBJECTS/METHODS Multiply imputed data from three nationally representative British birth cohorts were used—the MRC National Survey of Health and Development (NSHD; 1946 birth cohort, n=3012), the National Child Development Study (NCDS; 1958 birth cohort, n=9614) and the British Cohort Study (BCS; 1970 birth cohort, n=8140). A typology of work-family life course types was developed using multi-channel sequence analysis, linking annual information on work, partnerships and parenthood from 16 to 42 years. Work-family life courses were related to BMI trajectories using multi-level growth models. Analyses adjusted for indicators of prior health, birthweight, child BMI, educational attainment and socioeconomic position across the life course, and were stratified by gender and cohort. RESULTS Work-family life courses characterised by earlier transitions to parenthood and weaker long-term links to employment were associated with greater increases in BMI across adulthood. Some of these differences, particularly for work-family groups, which are becoming increasingly non-normative, became more pronounced across cohorts (for example, increases in BMI between 16 and 42 years in long-term homemaking women: NSHD: 4.35 kg m–2, 95% confidence interval (CI): 3.44, 5.26; NCDS: 5.53 kg m–², 95% CI: 5.18, 5.88; BCS: 6.69 kg m–², 95% CI: 6.36, 7.02). CONCLUSIONS Becoming a parent earlier and weaker long-term ties to employment are associated with greater increases in BMI across adulthood in British men and women.
Article
This paper deals with the fertility transition, one of the two essential components of the demographic transition. The analysis demonstrates that by applying the cohort perspective new insights are obtained about how the fertility transition unfolded. Within the overall framework of the fertility transition there were four distinct pathways of fertility trends. Combining these findings with those of other scholars shows that the demographic transition has not yet led to an equilibrium of relatively stable low mortality and stable low fertility.The four fertility transition pathways are the following: (1) The Western fertility transition pathway characterized by major cohort total fertility rate (CTFR) fluctuations; (2) the South European fertility transition pathway characterized by an almost uninterrupted CTFR decline; (3) the Central and East European fertility transition pathway characterized by stable CTFRs in the 1920s to 1950s cohorts and a decline in the 1960s and 1970s cohorts; (4) the East and South-East Asia fertility transition pathway characterized by a late start in the mid-20th century with rapidly declining CTFRs.The exploration of societal conditions shaping fertility trends in the 19th and 20th centuries confirms Notestein’s conclusions that the causes are a complex combination of “technological, social, economic, and political developments” as well as cultural and ideational effects, and that it is “impossible to be precise about the various causal factors”. At times the primary factors were economic, as in the Great Depression of the 1930s and the 1960s post-war prosperity in Western countries. However, these economic factors also had many political, cultural, social, policy and other important facets. In Central and Eastern Europe the primary factors during the era of state socialism were the political system and social policies. The patriarchal nature of societies was the prime factor shaping fertility trends in Southern Europe and in East and South-East Asia.
Book
This landmark study traces the life histories of approximately 300 teenage mothers and their children over a seventeen-year period. From interview data and case studies, it provides a vivid account of the impact of early childbearing on young mothers and their children. Some remarkable and surprising results emerge from this unique study of the long term adaptation to early parenthood. It also offers refreshing insights into the unexplored connections between mothers' careers and the development of their children. Adolescent Mothers in Later Life will be an invaluable resource for all those interested in teenage pregnancy.
Article
Background: Today's men want to participate in their children's upbringing more than in the past, but they are heavily involved in their occupation at the same time. This article describes the significance of parenthood, partnership and occupation in relation to health and health behaviours among men of working age in Germany. Data: We summarised data from the "German Health Update" (GEDA) studies conducted in 2009, 2010 and 2012. Data on 18,465 men aged 18-64 years were available, 11,429 of which were living with children. We included mental health problems, general health awareness, sports activity and smoking as outcomes. Results: Full-time employees working more than 48 h per week and unemployed persons had mental health problems more frequently (OR 1.44 and 2.35, p < 0.05) than full-time employees working 48 h or fewer. Similar associations can be shown for health awareness, physical activity and smoking. Concerning partnership and parenthood, the associations were considerably weaker: men living together with children and a partner in the household were overall less burdened and their behaviour was also healthier than single men without children. After simultaneous consideration of employment status, parenthood and partnership, our results show that the unemployed and employees with long working hours were the most burdened. Discussion: The results provide supporting evidence regarding health problems of men in Germany due to unemployment and long working hours that are of importance for their health whether they are living with a partner and/or with children or not. The association between health and occupation was stronger than between health and fatherhood or partnership.