- Email: [email protected]
Associations of childhood circumstances with physical and mental functioning in adulthood
ARTICLE IN PRESS
Social Science & Medicine 62 (2006) 1831–1839 www.elsevier.com/locate/socscimed
Associations of childhood circumstances with physical and mental functioning in adulthood Tomi Ma¨kinen, Mikko Laaksonen, Eero Lahelma, Ossi Rahkonen Department of Public Health, University of Helsinki, Helsinki, Finland Available online 27 September 2005
Abstract The aim of this study was to examine the associations of parental education and specific childhood circumstances with adult physical and mental functioning. Self-reported data were collected in 2000, 2001 and 2002 among middle-aged women (n ¼ 7171) and men (n ¼ 1799) employed by the City of Helsinki. Functioning was measured by the physical and mental component summaries of the Short-Form 36 Health Survey (SF-36). The lowest quartile of the scores on each component summary was considered to indicate limited functioning. Adult socio-economic circumstances were measured by their own education. Among women parental education was inversely associated with physical functioning. The association remained after adjusting for specific childhood circumstances but disappeared after adjustment for own education. In contrast, parental education was positively associated with mental functioning among women, and the association remained after adjusting for specific childhood circumstances and the own education. Among women childhood adversities such as own chronic disease, parental mental problems, economic difficulties and having been bullied in childhood were associated with both physical and mental functioning. Parental drinking problems were associated with adult mental functioning among women. Among men, chronic disease, economic difficulties and having been bullied in childhood were associated with physical functioning. Parental mental problems, economic difficulties and having being bullied in childhood were also associated with mental functioning among men. These results suggest that the effect of parental education on physical functioning is mediated through one’s own education. The association between parental education and mental functioning and the effects of several specific childhood circumstances may suggest a latency effect. Some evidence of cumulative effects of childhood and adulthood circumstances were found among women in physical functioning. Specific childhood circumstances are therefore important determinants of adult functioning. These circumstances provide detailed information on the association of childhood circumstances with adult functioning over and above parental education. r 2005 Elsevier Ltd. All rights reserved. Keywords: Childhood; Adulthood; Parental education; Physical functioning; Mental functioning; SF-36; Finland
Introduction Corresponding author. Tel.: +358 919127608;
fax: +358 919127570. E-mail addresses: tomi.makinen@helsinki.fi (T. Ma¨kinen), mikko.t.laaksonen@helsinki.fi (M. Laaksonen), eero.lahelma@helsinki.fi (E. Lahelma), ossi.rahkonen@helsinki.fi (O. Rahkonen). 0277-9536/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.socscimed.2005.08.040
Recently there has been much interest in the importance of early living conditions on adult health (Galobardes, Lynch & Davey Smith, 2004; Harper et al., 2002; Hayward & Gorman, 2004; Huurre, Aro & Rahkonen, 2003; Pensola &
ARTICLE IN PRESS 1832
T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
Valkonen, 2002). This interest is inspired, among other things, by David Barker’s hypothesis of biological programming of disease during the fetal period (Barker, 1995). This fetal origins hypothesis was later extended to also include infancy. Furthermore social circumstances in childhood have been introduced as another possible explanation for the association between early living conditions and adult health (Hayward & Gorman, 2004; Hill, Thomson Ross, Mudd, & Blow, 1997; Rahkonen, Lahelma, & Huuhka, 1997; Va˚gero¨ & Illsley, 1995). Social circumstances in childhood may have direct or indirect effects on adult health, and for example parental socioeconomic position influences the environment where childhood is spent. Childhood social circumstances may also contribute to the educational pathway of a person, and to the lifestyles and health habits he or she will adopt. Parental socioeconomic position can be used as a general indicator of social and economic circumstances in childhood. Several studies have shown that low socio-economic position of the parents is associated with poor self-rated health, illness, mortality and other health problems in later life (Galobardes et al., 2004; Harper et al., 2002; Huurre et al., 2003; Lundberg, 1993; Notkola, Punsar, Karvonen, & Haapakoski, 1985; Pensola & Valkonen, 2002; Rahkonen et al., 1997). More specific indicators of childhood circumstances are likely to provide detailed information on how childhood circumstances are associated with adult health. Some studies have shown that specific childhood circumstances, such as economic difficulties (Hill et al., 1997; Lundberg, 1993; Van de Mheen, Stronks, Van de Bos, & Mackenbach, 1997; Rahkonen et al., 1997), parental drinking problems (Hill et al., 1997) and other family problems (Lundberg, 1993; Rodgers, 1994) are associated with adult health. Examining the effects of childhood circumstances on adult health, later circumstances need to be taken into account as well. From the life course perspective three general models have been proposed to exemplify the possible influences of childhood and adult circumstances on adult health (BenShlomo & Kuh, 2002; Graham, 2002; Hertzman, 1999; Kuh & Ben-Shlomo, 1997; Kuh, Ben-Shlomo, Lynch, Hallqvist, & Power, 2003). First, the latency model suggests that childhood circumstances as such are associated with adult health regardless of adult circumstances. Second, the pathway model suggests that childhood circumstances indirectly affect adult health through adult circumstances.
Third, the cumulative model suggests that both childhood and adulthood circumstances are important to health. Several adverse circumstances increase the risk of ill health as negative influences may cumulate over the life course. The aim of this study was to examine the associations of parental education and specific childhood circumstances with adult physical and mental functioning. Health-related functioning has so far been largely neglected as an outcome in studies on early effects on later health. Nonetheless, functioning covers important domains of health, since it takes into account the consequences of health status for well-being and the quality of life (Martikainen, Stansfield, Hemingway, & Marmot 1999). Functioning provides important information on how people cope with the challenges of their living circumstances. The particular focus of our study was on a broad variety of childhood circumstances. Parental social class is often examined but only generally reflects childhood circumstances, whereas several specific childhood circumstances are likely to give a more detailed description of childhood living circumstances. The three life-course models were used to highlight the possible routes through which childhood circumstances may be associated with adult functioning. Methods Data The data were derived from the Helsinki Health Study (HHS) cohort consisting of middle-aged women and men employed by the City of Helsinki (Lahelma, Martikainen, Rahkonen, Roos, & Saastamoinen, 2005). Three separate cross-sectional baseline surveys were conducted in 2000, 2001 and 2002. A self-administered questionnaire was sent to each employee who during the year of the survey reached the age of 40, 45, 50, 55 or 60 years. The overall response rate was 67%. The data include 7171 women and 1799 men, reflecting the distribution of women and men employed by the City of Helsinki. The data generally represent the target population but younger people and manual workers are slightly underrepresented (Lallukka et al., 2002). Measures Parental education was used as the indicator of socio-economic position in childhood. In previous
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
study parental education has been used as an indicator of childhood socio-economic circumstances (Harper et al., 2002). Educational level of both parents was asked, and the higher one was chosen to indicate parental education. Parental education was divided into three groups: primary school or less (low education), secondary school or vocational training (intermediate education) and matriculation or a university degree (high education). Specific childhood circumstances were measured by asking seven questions concerning experiences before the age of sixteen. These circumstances were chronic disease, parental divorce, father’s or mother’s death, father’s or mother’s mental problem, parental drinking problem, economic difficulties in the family and having been bullied in childhood. Response alternatives were yes and no. A missing answer was classified among those not having experienced the adverse childhood circumstance when the respondent had replied (yes or no) to any other specific question. The respondent’s own education was used as an indicator of adult socio-economic circumstances. The level of education was divided into three groups: primary school (low education), secondary school or vocational training (intermediate education) and matriculation or university degree (high education). Preliminary analyses also included occupational class and marital status but they were found unrelated to adult functioning after own education was taken into account and are therefore not reported here. Adult health functioning was measured by the physical and mental component summaries of the Short-Form 36 Health Survey (SF-36) (Ware, 2003; Ware & Kosinski 2001). The SF-36 is a 36-item questionnaire that covers issues relating to physical, psychological and social functioning. The SF-36 includes eight subscales: physical functioning, role limitations due to physical problems, social functioning, bodily pain, general mental health, role limitations due to emotional problems, vitality and general health perceptions. These eight subscales can be summarized into physical and mental functioning component summaries, used as the outcome measures in this study. The SF-36 basically indicates the self-assessed impact of ill-health and diseases on functioning and well-being. This study used the lowest quartile of SF-36 scores to indicate limited functioning. This cut-off point is somewhat arbitrary, but agrees with several previous studies
1833
(Hemingway, Nicholson, Stafford, Roberts, & Marmot 1997; Kuper, Singh-Manoux, Siegrist, & Marmot 2002; Stansfeld, Bosma, Hemingway, & Marmot 1998; Stansfeld, Head, Fuhrer, Wardle, & Cattell 2003). The scores were calculated separately for women and men. Women with limited physical functioning had a physical component summary score below 43.6 and men below 46.4. For women the mental functioning component summary score in the lowest quartile was below 45.9 and for men below 46.3. Statistical methods All analyses were made separately for women and men. Logistic regression analysis was used to examine the associations of parental education and specific childhood circumstances with adult functioning. Model 1 presents individual effects of parental education and each of the adverse childhood circumstances on limited adult functioning after adjustment for age. In Model 2 parental education and adverse childhood circumstances were adjusted for each other, and in Model 3 the associations were further adjusted for own education. These analyses suggest a latency effect if the association of parental education or adverse childhood circumstances with adult functioning remains after adjusting for own education. A pathway effect is suggested if the association disappears after adjusting for own education. To test cumulative effects interactions between childhood circumstances and own education were examined. The results from the cumulative models are presented as prevalence percentages with 95% confidence intervals. Results Table 1 presents the distributions of the explanatory variables among women and men. All age groups were equally large, with the exception of the oldest one which was slightly smaller. The two most common specific childhood circumstances for women and men were parental drinking problems and economic difficulties, followed by parental divorce and the death of a parent. Half of the parents had low education while half of the respondents had high education. Associations of parental education and specific childhood circumstances with limited adult physical functioning among women and men are presented
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
1834
Table 1 Distribution of age, parental education, specific childhood circumstances, combined childhood circumstances and own education among women (n ¼ 7171) and men (n ¼ 1799) Women (%) Age (years) 40 45 50 55 60
Men (%)
21 22 22 24 11
18 19 21 27 15
100
100
54 26 20
49 25 25
100
100
Specific childhood circumstances Own chronic disease Parental divorce Parental death Parental mental problem Parental drinking problem Economic difficulties Having been bullied in childhood
6 10 13 4 18 17 7
9 12 12 5 18 17 10
Combined childhood circumstances Low parental education and one or more specific childhood circumstance Low parental education and no specific childhood circumstances High parental education and one or more specific childhood circumstance High parental education and no specific childhood circumstances
27 26 20 26
27 23 23 28
100
100
13 30 57
14 27 59
100
100
Total Parental education Low Intermediate High Total a
Total Own education Low Intermediate High Total a
Percentages indicate the proportion of those who reported a specific childhood circumstance. For example, 10% of female respondents’ parents had been divorced and 90% had not been divorced.
in Table 2. Women whose parents had low education reported limited adult physical functioning more often than women whose parents had high education. The association between parental education and adult functioning remained unchanged after adjusting for specific childhood circumstances but disappeared after own education was adjusted for. Of the childhood circumstances own chronic disease, parental mental problems, economic difficulties and having been bullied were all associated with limited physical functioning in adulthood. However, there were no associations of parental
divorce, death of a parent or parental drinking problems with limited adult physical functioning. All associations of specific childhood circumstances with limited adult physical functioning remained after parental as well as own education were adjusted for. Among men, parental education was not associated with adult physical functioning (Table 2). However, as for women, own chronic disease, economic difficulties and having been bullied in childhood were associated with limited adult physical functioning, but parental mental problems
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
1835
Table 2 Associations of parental education and specific childhood circumstances with limited adult physical functioning among women and men. Logistic regression analysis (OR ¼ odds ratio) and 95% confidence intervals (CI) Women
Men
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
OR CI
OR CI
OR CI
OR CI
OR CI
OR CI
Parental education High Intermediate Low
1.00 1.00 1.00 1.00 1.00 1.00 1.35 (1.14–1.61) 1.32 (1.11–1.57) 1.11 (0.93–1.33) 0.92 (0.67–1.27) 0.92 (0.66–1.27) 0.77 (0.55–1.08) 1.46 (1.25–1.61) 1.39 (1.19–1.63) 1.03 (0.87–1.22) 1.25 (0.95–1.64) 1.21 (0.92–1.59) 0.90 (0.67–1.22)
Specific childhood circumstancesa Own chronic disease Parental divorce Parental death Parental mental problem Parental drinking problem Economic difficulties Having been bullied in childhood
1.48 0.93 0.84 1.31 1.03 1.51 1.64
a
(1.20–1.82) (0.76–1.13) (0.71–1.00) (1.03–1.66) (0.88–1.20) (1.31–1.75) (1.34–2.00)
1.49 0.94 0.82 1.34 1.02 1.46 1.63
(1.21–1.84) (0.77–1.15) (0.70–0.98) (1.05–1.70) (0.87–1.19) (1.25–1.70) (1.33–1.99)
1.54 0.87 0.78 1.38 1.02 1.44 1.54
(1.24–1.90) (0.72–1.07) (0.66–0.92) (1.09–1.76) (0.87–1.19) (1.24–1.67) (1.26–1.89)
1.62 0.94 1.04 0.89 1.06 1.55 1.42
(1.13–2.33) (0.65–1.34) (0.74–1.44) (0.49–1.63) (0.78–1.44) (1.16–2.08) (0.99–2.02)
1.66 0.96 1.03 0.89 1.05 1.50 1.44
(1.13–2.35) (0.67–1.38) (0.74–1.43) (0.49–1.62) (0.78–1.43) (1.12–2.02) (1.00–2.06)
1.54 0.86 0.96 0.93 1.06 1.46 1.47
(1.07–2.23) (0.60–1.24) (0.69–1.34) (0.50–1.70) (0.78–1.44) (1.08–1.97) (1.02–2.11)
The reference category (OR ¼ 1.00) was those reporting no specific childhood circumstances.
were not. Associations of these three specific childhood circumstances with limited adult physical functioning remained even after adjustments. Women whose parents had high education reported limited adult mental functioning more often than women whose parents had low education (Table 3). This association was unaffected by the adjustments for specific childhood circumstances and for own education. Among women, parental mental problem, chronic disease, parental drinking problem, economic difficulties and having been bullied in childhood were associated with limited adult mental functioning. These associations remained after parental as well as own education were adjusted for. Among men, the association of parental education with adult mental functioning was similar to that among women, but it failed to reach statistical significance (Table 3). Among men parental mental problems, economic difficulties and having been bullied in childhood were associated with limited adult mental functioning. In contrast to women, there were no associations of childhood chronic disease and parental drinking problems with limited adult mental functioning among men. Also among men all associations between specific childhood circumstances and limited adult mental functioning remained after adjusting for parental and own education.
In order to examine cumulative effects of childhood and adulthood circumstances on adult functioning, the categorizations of the variables were simplified. For parental education the categories of intermediate and high education were collapsed. Specific childhood circumstances were combined and dichotomized into those having experienced none of the specific childhood circumstances and those having experienced one or more specific childhood circumstances. Own low and intermediate education were also collapsed. Specific childhood circumstances and parental education were then cross-classified to produce four combinations. Results from the cumulative analyses confirmed that specific childhood circumstances increased the prevalence for limited adult physical functioning but the role of parental education was modest (Table 4). Among those with low own education the prevalence for limited physical functioning was higher among those with poor combined childhood circumstances. Among those with high parental education the prevalence for limited functioning was higher for those with one or more specific childhood circumstances, regardless of their parental education. The interaction between combined childhood circumstances and own education was statistically significant among women (p ¼ 0:02) but not among men. Among women with combined childhood circumstances and low own education
ARTICLE IN PRESS 1836
T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
Table 3 Associations of parental education and specific childhood circumstances with limited adult mental functioning among women and men. Logistic regression analysis (OR ¼ odds ratio) and 95% confidence intervals (CI) Women
Men
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
OR CI
OR CI
OR CI
OR CI
OR CI
OR CI
Parental education High Intermediate Low
1.00 1.00 1.00 1.00 1.00 1.00 0.81 (0.70–0.95) 0.77 (0.66–0.90) 0.79 (0.67–0.93) 0.78 (0.58–1.06) 0.73 (0.54–0.99) 0.75 (0.54–1.03) 0.78 (0.68–0.90) 0.72 (0.63–0.83) 0.76 (0.65–0.89) 0.79 (0.61–1.03) 0.71 (0.54–0.94) 0.74 (0.56–1.00)
Specific childhood circumstancesa Own chronic disease Parental divorce Parental death Parental mental problem Parental drinking problem Economic difficulties Having been bullied in childhood
1.41 0.90 0.87 2.01 1.32 1.46 1.62
a
(1.14–1.74) (0.75–1.09) (0.74–1.04) (1.61–2.50) (1.14–1.52) (1.26–1.69) (1.33–1.97)
1.40 0.89 0.89 1.97 1.33 1.53 1.64
(1.13–1.72) (0.74–1.08) (0.75–1.05) (1.58–2.47) (1.15–1.54) (1.32–1.77) (1.35–1.99)
1.40 0.90 0.90 1.96 1.33 1.53 1.66
(1.13–1.72) (0.74–1.09) (0.76–1.07) (1.57–2.45) (1.15–1.54) (1.32–1.77) (1.36–2.01)
1.00 1.18 1.03 1.77 1.23 1.71 1.78
(0.68–1.49) (0.85–1.65) (0.73–1.44) (1.05–2.99) (0.92–1.65) (1.28–2.30) (1.27–2.51)
0.97 1.16 1.03 1.76 1.27 1.80 1.77
(0.66–1.45) (0.83–1.63) (0.73–1.44) (1.04–2.98) (0.94–1.70) (1.33–2.42) (1.26–2.50)
0.98 1.18 1.04 1.75 1.27 1.81 1.77
(0.66–1.46) (0.84–1.66) (0.74–1.46) (1.04–2.95) (0.94–1.70) (1.34–2.43) (1.26–2.49)
The reference category (OR ¼ 1.00) was those reporting no specific childhood circumstances.
Table 4 Limited adult physical and mental functioning by combined childhood circumstances and own education among women and men. Prevalence percentages and 95% confidence intervals (CI) Women
Men
Own education
Own education
All
Low
High
All
Low
High
Combined childhood circumstances
CI
CI
CI
CI
CI
CI
Low parental education and one or more specific childhood circumstance Low parental education and no specific childhood circumstances High parental education and one or more specific childhood circumstance High parental education and no specific childhood circumstances
30 (29–31)
36 (34–38)
21 (20–22)
32 (30–34)
39 (35–43)
23 (20–26)
25 (24–26)
31 (29–33)
17 (16–18)
23 (21–25)
29 (26–32)
17 (15–19)
25 (24–26)
30 (28–32)
23 (22–24)
25 (23–27)
26 (23–29)
25 (22–28)
19 (18–20)
29 (27–31)
17 (16–18)
19 (17–21)
27 (24–30)
16 (14–18)
Total %
25 (24–26)
33 (31–35)
19 (18–20)
25 (23–27)
33 (30–36)
19 (17–21)
29 (28–30)
28 (26–30)
30 (29–31)
26 (24–28)
25 (22–28)
28 (25–31)
19 (18–20)
27 (25–29)
21 (20–22)
19 (17–21)
19 (16–22)
19 (17–21)
30 (29–31)
29 (27–31)
31 (30–32)
33 (31–35)
30 (27–33)
34 (31–37)
22 (21–23)
22 (21–23)
22 (21–23)
20 (18–22)
17 (14–20)
20 (18–22)
25 (24–26)
24 (22–26)
26 (25–27)
25 (23–27)
23 (20–26)
25 (22–28)
Limited physical functioning
Limited mental functioning Combined childhood circumstances Low parental education and one or more specific childhood circumstance Low parental education and no specific childhood circumstances High parental education and one or more specific childhood circumstance High parental education and no specific childhood circumstances Total %
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
existed more limited physical functioning than among women with combined childhood circumstances and high own education. Mental functioning was more often limited among those who had experienced one or more specific childhood circumstances regardless of their parental education. The association was similar in both categories of own education as well as in women and men. Discussion This study focused on the associations of childhood circumstances with adult physical and mental functioning among middle-aged women and men employed by the City of Helsinki. Parental education was used as a generic indicator of socioeconomic circumstances in childhood and specific childhood circumstances were measured with seven indicators. Three life course models, latency, pathway and cumulative model, were utilized to highlight the nature of the associations between childhood circumstances and adult functioning. Among women low parental education was associated with limited adult physical functioning and high parental education with limited adult mental functioning. This inverse association between parental education and adult physical functioning disappeared after adjusting for own education whereas the corresponding association for mental functioning remained. This suggests that own education may mediate the effects of childhood circumstances on adult physical functioning. Previous studies (Hallqvist, Lynch, Bartley, Lang, & Blane, 2004; Hertzman, Power, Matthews, & Manor, 2001; Laaksonen, Rahkonen, Martikainen, & Lahelma, 2005; Lundberg, 1993; Pensola & Martikainen, 2003) equally support a similar mediated pathway. However, it remains an open question why high parental education was associated with limited adult mental functioning even after adjustment for specific childhood circumstances and own education. Several specific childhood circumstances remained associated with adult physical and mental functioning after adjusting for parental and own education. This may be taken as a suggestion for a latency effect. Own childhood chronic disease, parental mental problems, economic difficulties and having been bullied in childhood potentially have long lasting consequences for physical and mental functioning. A previous study has examined how specific childhood circumstances are associated
1837
with adult functioning (Lynch, Kaplan, & Shema, 1997). However, corresponding studies using other health outcomes have supported specific childhood circumstances being markedly associated with adult health (Hill et al., 1997; Lundberg, 1993; Van de Mheen et al., 1997; Rahkonen et al., 1997; Rodgers, 1994). Close to same childhood circumstances were associated with both adult physical and mental functioning. Own chronic disease, parental mental problems, economic difficulties and having been bullied in childhood were all associated with both physical and mental functioning. An exception was parental drinking problems which were only associated with mental functioning. There were minor differences in physical and mental functioning between women and men. Parental mental problems were associated with women’s but not men’s physical functioning. Among men own childhood chronic disease was not associated with mental functioning unlike among women. It has been found that children with poorly educated parents are likely to have fewer resources and receive less social support from their parents than children with better educated parents (Huurre, Eerola, Rahkonen, & Aro, submitted). Parents also influence their children’s health habits and there is evidence that childhood social position is associated with smoking (Brunner, Shipley, Blane, Davey Smith, & Marmot, 1999; Power et al., 2005; Tyas & Pederson, 1998) and obesity (Brunner et al., 1999) in adulthood. These findings confirm that adverse psycho-social circumstances in childhood may be critical for early socialization and later health behaviours. We did find some evidence on cumulative effects of childhood and adulthood circumstances among women on physical functioning. However, adverse childhood circumstances may be more important to physical functioning than parental education or own education. Several previous studies have examined the cumulative model (Hallqvist et al., 2004; Hart, Davey Smith, & Blane, 1998; Holland et al., 2000; Lynch et al., 1997; Power, Manor, & Matthews, 1999; Power, Matthews, & Manor, 1998; Wamala, Lynch, & Kaplan, 2001) and shown that low childhood social class, economic difficulties, shortage of early life resources and own chronic disease together with adulthood circumstances cumulatively affect adult health and health-related risks (Hart et al., 1998; Holland et al., 2000; Power et al., 1999; Power et al., 1998). However,
ARTICLE IN PRESS 1838
T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
accumulation of adverse circumstances over the life course is likely to vary by gender (Power et al., 1998) and by childhood social class (Holland et al., 2000). Regarding the three life course models it has recently been questioned whether the latency, pathway and accumulation effects can in fact be distinguished from each other (Hallqvist et al., 2004). Nevertheless, our study renders some evidence on the latency, pathway and cumulative model. There are some limitations in our study to be kept in mind when interpreting the results. In a crosssectional design the childhood circumstances were asked retrospectively. All information on childhood and adulthood circumstances was self-reported and collected from middle-aged women and men. Distressed people may recall their life history in adverse terms and their present perceptions may have influenced their response through ‘negative affectivity’ (Watson, 1988). However, a recent study suggested that retrospective data on childhood circumstances are likely to be generally reliable (Dube, Williamson, Thompson, Felitti, & Anda, 2004; Krieger, Okamoto, & Selby, 1998). Since we only studied employees it is possible that those with the poorest functioning have dropped out as they may have already left the labour force. However, the large size of our data allowed an examination of even relatively rare specific childhood circumstances. In conclusion, parental education and specific childhood circumstances showed clear and strong associations with adult physical and mental functioning. However, especially among women parental education had different associations with adult physical and mental functioning. Several specific childhood circumstances were associated with physical and mental functioning in both women and men. Parental education is likely to follow a pathway effect on adult physical functioning. Our results suggest that specific childhood circumstances are likely to have latency effects on adult physical and mental functioning. Therefore these circumstances provide more detailed evidence than generic parental social position on how childhood is associated with adult functioning. Ideally the specific childhood circumstances are studied side by side with parental social position. Acknowledgements Helsinki Health Study is supported by grants from the Academy of Finland (#53245, #205588 and #105952) and the Finnish Work Environment
Fund (#103310). The Academy of Finland also supports Mikko Laaksonen (#204894) and Ossi Rahkonen (#45664 and #210435). We would like to thank the City of Helsinki and the members of the Helsinki Health Study research group. References Barker, D. J. P. (1995). Fetal origins of coronary heart disease. British Medical Journal, 311, 171–174. Ben-Shlomo, Y., & Kuh, D. (2002). A life course approach to chronic disease epidemiology: Conceptual models, empirical challenges and interdisciplinary perspectives. International Journal of Epidemiology, 31, 285–293. Brunner, E., Shipley, M. J., Blane, D., Smith, G. D., & Marmot, M. G. (1999). When does cardiovascular risk start? Past and present socioeconomic circumstances and risk factors in adulthood. Journal of Epidemiology and Community Health, 53, 757–764. Dube, S., Williamson, D. F., Thompson, T., Felitti, V., & Anda, R. (2004). Assessing the reliability of retrospective reports of adverse childhood experiences among adult HMO members attending a primary care clinic. Child Abuse & Neglect, 28, 729–737. Galobardes, B., Lynch, J. W., & Davey Smith, G. (2004). Childhood socioeconomic circumstances and cause-specific mortality in adulthood: Systematic review and interpretation. Epidemiological Reviews, 26, 7–21. Graham, H. (2002). Building an inter-disciplinary science of health inequalities: The example of lifecourse research. Social Science & Medicine, 55, 2005–2016. Hallqvist, J., Lynch, J., Bartley, M., Lang, T., & Blane, D. (2004). Can we disentangle life course processes of accumulation critical period and social mobility? An analysis of disadvantaged socio-economic positions and myocardial infarction in the Stockholm heart epidemiology program. Social Science & Medicine, 58, 1555–1562. Harper, S., Lynch, J., Hsu, W.-L., Everson, S., Hillemeier, M., Raghunathan, T., Salonen, J., & Kaplan, G. (2002). Life course socioeconomic conditions and adult psychosocial functioning. International Journal of Epidemiology, 31, 395–403. Hart, C., Davey Smith, G., & Blane, D. (1998). Inequalities in mortality by social class measured at 3 stages of the life course. American Journal of Public Health, 88, 471–474. Hayward, M. D., & Gorman, B. K. (2004). The long arm of childhood: The influence of early life social conditions on men’s mortality. Demography, 41, 87–107. Hemingway, H., Nicholson, A., Stafford, M., Roberts, R., & Marmot, M. (1997). The impact of socioeconomic status on health functioning as assessed by the SF-36 questionnaire: The whitehall II study. American Journal of Public Health., 87, 1484–1490. Hertzman, C. (1999). The biological embedding of the early experience and its effects on health in adulthood. Annals of the New York Academy of Sciences, 869, 85–95. Hertzman, C., Power, C., Matthews, S., & Manor, O. (2001). Using an interactive framework of society and lifecourse to explain self-rated health in early adulthood. Social Science & Medicine, 53, 1575–1585.
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839 Hill, E., Thomson Ross, L., Mudd, S., & Blow, F. (1997). Adulthood functioning: The joint effects of parental alcoholism, gender and childhood socio-economic stress. Addiction, 92, 583–596. Holland, P., Berney, L., Blane, D., Davey Smith, G., Gunnel, D. J., & Montgomery, S. M. (2000). Life course accumulation of disadvantage: Childhood health and hazard exposure during adulthood. Social Science & Medicine, 50, 1285–1295. Huurre, T., Aro, H., & Rahkonen, O. (2003). Well-being and health behaviour by parental socioeconomic status. A follow up study of adolescents aged 16 until age 32 years. Social Psychiatry and Psychiatric Epidemiology, 38, 249–255. Huurre, T., Eerola, M., Rahkonen, O., & Aro H. Does social support affect the relationship between socio-economic status and depression? A longitudinal study from adolescence to adulthood, submitted for publication. Krieger, N., Okamoto, A., & Selby, J. V. (1998). Adult female twins’ recall of childhood social class and father’s education: A validation study for public health research. American Journal of Epidemiology, 147, 704–708. Kuh, D., & Ben-Shlomo, Y. (1997). A life course approach to chronic disease Epidemiology. Oxford Medical Publications. Kuh, D., Ben-Shlomo, Y., Lynch, J., Hallqvist, J., & Power, C. (2003). Life course epidemiology. Glossary. Journal of Epidemiology and Community Health, 57, 778–783. Kuper, H., Singh-Manoux, A., Siegrist, J., & Marmot, M. (2002). When reciprocity fails: Effort–reward imbalance in relation to coronary heart disease and health functioning within the whitehall II study. Occupational and Environmental Medicine, 59, 777–784. Laaksonen, M., Rahkonen, O., Martikainen, P., & Lahelma, E. (2005). Multiple dimensions of socioeconomic position and self-rated health. The contribution of childhood socioeconomic circumstances, adult socioeconomic status and material resources. American Journal of Public Health, 95, 1403–1409. Lahelma, E., Martikainen, P., Rahkonen, O., Roos, E., & Saastamoinen, P. (2005). Occupational class inequalities across key domains of health: Results from the Helsinki Health Study. European Journal of Public Health, Advance access published, July 13, doi:10.1093/eurpub/cki022. Lallukka, T., Aittoma¨ki, A., Piha, K., Roos, E., Kivela¨, K., & Silventoinen, K. (2002). Postikyselytutkimukseen vastanneiden edustavuus sosioekonomisten tekijo¨iden ja sairauspoissaolotietojen mukaan: Helsingin kaupungin henkilo¨sto¨n terveystutkimus. Sosiaalila¨a¨ketieteellinen Aikakauslehti, 39, 164–171. Lundberg, O. (1993). The impact of childhood living conditions on illness and mortality in adulthood. Social Science & Medicine, 36, 1047–1052. Lynch, J. W., Kaplan, G. A., & Shema, S. J. (1997). Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. New England Journal of Medicine, 337, 547–552. Martikainen, P., Stansfield, S., Hemingway, H., & Marmot, M. (1999). Determinants of socioeconomic differences in change in physical and mental functioning. Social Science & Medicine, 49, 499–507. Van de Mheen, H., Stronks, K., Van de Bos, J., & Mackenbach, J. P. (1997). The contribution of childhood environment to the explanation of socio-economic inequalities in health in
1839
adult life: A retrospective study. Social Science & Medicine, 44, 13–24. Notkola, V., Punsar, S., Karvonen, MJ., & Haapakoski, J. (1985). Socioeconomic conditions in childhood and mortality and morbidity caused by coronary heart disease in adulthood in rural Finland. Social Science & Medicine, 21, 517–523. Pensola, T., & Martikainen, P. (2003). Cumulative social class and mortality from various causes of adult men. Journal of Epidemiology and Community Health, 57, 745–751. Pensola, T., & Valkonen, T. (2002). Effect of parental social class, own education and social class on mortality among young men. European Journal of Public Health, 12, 29–36. Power, C., Manor, O., & Matthews, S. (1999). The duration and timing of exposure: Effects of socio-economic environment on adult health. American Journal of Public Health, 89, 1059–1065. Power, C., Matthews, S., & Manor, O. (1998). Inequalities in selfrated health: Explanations from different stages of life. Lancet, 351, 1009–1013. Power, C., Graham, H., Due, P., Hallqvist, J., Joung, I., Kuh, D., & Lynch, J. (2005). The contribution of childhood and adult socioeconomic position to adult obesity and smoking behaviour: An international comparison. International Journal of Epidemiology, advance access published on January 19. Rahkonen, O., Lahelma, E., & Huuhka, M. (1997). Past or present? Childhood living conditions and current socioeconomic status as determinants of adult health. Social Science & Medicine, 44, 327–336. Rodgers, B. (1994). Pathways between parental divorce and adult depression. Journal of Child Psychology and Psychiatry, 35, 1289–1308. Stansfeld, S. A., Bosma, H., Hemingway, H., & Marmot, M. (1998). Psychosocial work characteristics and social support as predictors of SF-36 health functioning: The whitehall II study. Psychosomatic Medicine, 60, 247–255. Stansfeld, S. A., Head, J., Fuhrer, R., Wardle, J., & Cattell, V. (2003). Social inequalities in depressive symptoms and physical functioning in the whitehall II study: Exploring a common cause explanation. Journal of Epidemiology and Community Health, 57, 361–367. Tyas, S. L., & Pederson, L. L. (1998). Psychosocial factors related to adolescent smoking: A critical review of the literature. Tobacco Control, 7, 409–420. Va˚gero¨, D., & Illsley, R. (1995). Explaining health inequalities: Beyond black and barker. European Sociological Review, 11, 219–241. Wamala, S. P., Lynch, J. W., & Kaplan, G. A. (2001). Women’s exposure to early and later life socioeconomic disadvantage and coronary heart disease risk: The Stockholm female coronary risk study. International Journal of Epidemiology, 30, 275–284. Ware, J. SF-36s Health survey update. www.sf-36.org/tools/ SF36bookschapter.shtml (13.5.2003), 1–17. Ware, J., & Kosinski, M. (2001). SF-36 Physical and mental health summary scales: A manual for users of version 1 (2nd ed.). Lincoln Rhode Island: QualityMetric Incorporated. Watson, D. (1988). Intraindividual and interindividual analyses of positive and negative affect: Their relation to health complaints, perceived stress, and daily activities. Journal of Personality & Social Psychology, 54, 1020–1030.
Social Science & Medicine 62 (2006) 1831–1839 www.elsevier.com/locate/socscimed
Associations of childhood circumstances with physical and mental functioning in adulthood Tomi Ma¨kinen, Mikko Laaksonen, Eero Lahelma, Ossi Rahkonen Department of Public Health, University of Helsinki, Helsinki, Finland Available online 27 September 2005
Abstract The aim of this study was to examine the associations of parental education and specific childhood circumstances with adult physical and mental functioning. Self-reported data were collected in 2000, 2001 and 2002 among middle-aged women (n ¼ 7171) and men (n ¼ 1799) employed by the City of Helsinki. Functioning was measured by the physical and mental component summaries of the Short-Form 36 Health Survey (SF-36). The lowest quartile of the scores on each component summary was considered to indicate limited functioning. Adult socio-economic circumstances were measured by their own education. Among women parental education was inversely associated with physical functioning. The association remained after adjusting for specific childhood circumstances but disappeared after adjustment for own education. In contrast, parental education was positively associated with mental functioning among women, and the association remained after adjusting for specific childhood circumstances and the own education. Among women childhood adversities such as own chronic disease, parental mental problems, economic difficulties and having been bullied in childhood were associated with both physical and mental functioning. Parental drinking problems were associated with adult mental functioning among women. Among men, chronic disease, economic difficulties and having been bullied in childhood were associated with physical functioning. Parental mental problems, economic difficulties and having being bullied in childhood were also associated with mental functioning among men. These results suggest that the effect of parental education on physical functioning is mediated through one’s own education. The association between parental education and mental functioning and the effects of several specific childhood circumstances may suggest a latency effect. Some evidence of cumulative effects of childhood and adulthood circumstances were found among women in physical functioning. Specific childhood circumstances are therefore important determinants of adult functioning. These circumstances provide detailed information on the association of childhood circumstances with adult functioning over and above parental education. r 2005 Elsevier Ltd. All rights reserved. Keywords: Childhood; Adulthood; Parental education; Physical functioning; Mental functioning; SF-36; Finland
Introduction Corresponding author. Tel.: +358 919127608;
fax: +358 919127570. E-mail addresses: tomi.makinen@helsinki.fi (T. Ma¨kinen), mikko.t.laaksonen@helsinki.fi (M. Laaksonen), eero.lahelma@helsinki.fi (E. Lahelma), ossi.rahkonen@helsinki.fi (O. Rahkonen). 0277-9536/$ - see front matter r 2005 Elsevier Ltd. All rights reserved. doi:10.1016/j.socscimed.2005.08.040
Recently there has been much interest in the importance of early living conditions on adult health (Galobardes, Lynch & Davey Smith, 2004; Harper et al., 2002; Hayward & Gorman, 2004; Huurre, Aro & Rahkonen, 2003; Pensola &
ARTICLE IN PRESS 1832
T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
Valkonen, 2002). This interest is inspired, among other things, by David Barker’s hypothesis of biological programming of disease during the fetal period (Barker, 1995). This fetal origins hypothesis was later extended to also include infancy. Furthermore social circumstances in childhood have been introduced as another possible explanation for the association between early living conditions and adult health (Hayward & Gorman, 2004; Hill, Thomson Ross, Mudd, & Blow, 1997; Rahkonen, Lahelma, & Huuhka, 1997; Va˚gero¨ & Illsley, 1995). Social circumstances in childhood may have direct or indirect effects on adult health, and for example parental socioeconomic position influences the environment where childhood is spent. Childhood social circumstances may also contribute to the educational pathway of a person, and to the lifestyles and health habits he or she will adopt. Parental socioeconomic position can be used as a general indicator of social and economic circumstances in childhood. Several studies have shown that low socio-economic position of the parents is associated with poor self-rated health, illness, mortality and other health problems in later life (Galobardes et al., 2004; Harper et al., 2002; Huurre et al., 2003; Lundberg, 1993; Notkola, Punsar, Karvonen, & Haapakoski, 1985; Pensola & Valkonen, 2002; Rahkonen et al., 1997). More specific indicators of childhood circumstances are likely to provide detailed information on how childhood circumstances are associated with adult health. Some studies have shown that specific childhood circumstances, such as economic difficulties (Hill et al., 1997; Lundberg, 1993; Van de Mheen, Stronks, Van de Bos, & Mackenbach, 1997; Rahkonen et al., 1997), parental drinking problems (Hill et al., 1997) and other family problems (Lundberg, 1993; Rodgers, 1994) are associated with adult health. Examining the effects of childhood circumstances on adult health, later circumstances need to be taken into account as well. From the life course perspective three general models have been proposed to exemplify the possible influences of childhood and adult circumstances on adult health (BenShlomo & Kuh, 2002; Graham, 2002; Hertzman, 1999; Kuh & Ben-Shlomo, 1997; Kuh, Ben-Shlomo, Lynch, Hallqvist, & Power, 2003). First, the latency model suggests that childhood circumstances as such are associated with adult health regardless of adult circumstances. Second, the pathway model suggests that childhood circumstances indirectly affect adult health through adult circumstances.
Third, the cumulative model suggests that both childhood and adulthood circumstances are important to health. Several adverse circumstances increase the risk of ill health as negative influences may cumulate over the life course. The aim of this study was to examine the associations of parental education and specific childhood circumstances with adult physical and mental functioning. Health-related functioning has so far been largely neglected as an outcome in studies on early effects on later health. Nonetheless, functioning covers important domains of health, since it takes into account the consequences of health status for well-being and the quality of life (Martikainen, Stansfield, Hemingway, & Marmot 1999). Functioning provides important information on how people cope with the challenges of their living circumstances. The particular focus of our study was on a broad variety of childhood circumstances. Parental social class is often examined but only generally reflects childhood circumstances, whereas several specific childhood circumstances are likely to give a more detailed description of childhood living circumstances. The three life-course models were used to highlight the possible routes through which childhood circumstances may be associated with adult functioning. Methods Data The data were derived from the Helsinki Health Study (HHS) cohort consisting of middle-aged women and men employed by the City of Helsinki (Lahelma, Martikainen, Rahkonen, Roos, & Saastamoinen, 2005). Three separate cross-sectional baseline surveys were conducted in 2000, 2001 and 2002. A self-administered questionnaire was sent to each employee who during the year of the survey reached the age of 40, 45, 50, 55 or 60 years. The overall response rate was 67%. The data include 7171 women and 1799 men, reflecting the distribution of women and men employed by the City of Helsinki. The data generally represent the target population but younger people and manual workers are slightly underrepresented (Lallukka et al., 2002). Measures Parental education was used as the indicator of socio-economic position in childhood. In previous
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
study parental education has been used as an indicator of childhood socio-economic circumstances (Harper et al., 2002). Educational level of both parents was asked, and the higher one was chosen to indicate parental education. Parental education was divided into three groups: primary school or less (low education), secondary school or vocational training (intermediate education) and matriculation or a university degree (high education). Specific childhood circumstances were measured by asking seven questions concerning experiences before the age of sixteen. These circumstances were chronic disease, parental divorce, father’s or mother’s death, father’s or mother’s mental problem, parental drinking problem, economic difficulties in the family and having been bullied in childhood. Response alternatives were yes and no. A missing answer was classified among those not having experienced the adverse childhood circumstance when the respondent had replied (yes or no) to any other specific question. The respondent’s own education was used as an indicator of adult socio-economic circumstances. The level of education was divided into three groups: primary school (low education), secondary school or vocational training (intermediate education) and matriculation or university degree (high education). Preliminary analyses also included occupational class and marital status but they were found unrelated to adult functioning after own education was taken into account and are therefore not reported here. Adult health functioning was measured by the physical and mental component summaries of the Short-Form 36 Health Survey (SF-36) (Ware, 2003; Ware & Kosinski 2001). The SF-36 is a 36-item questionnaire that covers issues relating to physical, psychological and social functioning. The SF-36 includes eight subscales: physical functioning, role limitations due to physical problems, social functioning, bodily pain, general mental health, role limitations due to emotional problems, vitality and general health perceptions. These eight subscales can be summarized into physical and mental functioning component summaries, used as the outcome measures in this study. The SF-36 basically indicates the self-assessed impact of ill-health and diseases on functioning and well-being. This study used the lowest quartile of SF-36 scores to indicate limited functioning. This cut-off point is somewhat arbitrary, but agrees with several previous studies
1833
(Hemingway, Nicholson, Stafford, Roberts, & Marmot 1997; Kuper, Singh-Manoux, Siegrist, & Marmot 2002; Stansfeld, Bosma, Hemingway, & Marmot 1998; Stansfeld, Head, Fuhrer, Wardle, & Cattell 2003). The scores were calculated separately for women and men. Women with limited physical functioning had a physical component summary score below 43.6 and men below 46.4. For women the mental functioning component summary score in the lowest quartile was below 45.9 and for men below 46.3. Statistical methods All analyses were made separately for women and men. Logistic regression analysis was used to examine the associations of parental education and specific childhood circumstances with adult functioning. Model 1 presents individual effects of parental education and each of the adverse childhood circumstances on limited adult functioning after adjustment for age. In Model 2 parental education and adverse childhood circumstances were adjusted for each other, and in Model 3 the associations were further adjusted for own education. These analyses suggest a latency effect if the association of parental education or adverse childhood circumstances with adult functioning remains after adjusting for own education. A pathway effect is suggested if the association disappears after adjusting for own education. To test cumulative effects interactions between childhood circumstances and own education were examined. The results from the cumulative models are presented as prevalence percentages with 95% confidence intervals. Results Table 1 presents the distributions of the explanatory variables among women and men. All age groups were equally large, with the exception of the oldest one which was slightly smaller. The two most common specific childhood circumstances for women and men were parental drinking problems and economic difficulties, followed by parental divorce and the death of a parent. Half of the parents had low education while half of the respondents had high education. Associations of parental education and specific childhood circumstances with limited adult physical functioning among women and men are presented
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
1834
Table 1 Distribution of age, parental education, specific childhood circumstances, combined childhood circumstances and own education among women (n ¼ 7171) and men (n ¼ 1799) Women (%) Age (years) 40 45 50 55 60
Men (%)
21 22 22 24 11
18 19 21 27 15
100
100
54 26 20
49 25 25
100
100
Specific childhood circumstances Own chronic disease Parental divorce Parental death Parental mental problem Parental drinking problem Economic difficulties Having been bullied in childhood
6 10 13 4 18 17 7
9 12 12 5 18 17 10
Combined childhood circumstances Low parental education and one or more specific childhood circumstance Low parental education and no specific childhood circumstances High parental education and one or more specific childhood circumstance High parental education and no specific childhood circumstances
27 26 20 26
27 23 23 28
100
100
13 30 57
14 27 59
100
100
Total Parental education Low Intermediate High Total a
Total Own education Low Intermediate High Total a
Percentages indicate the proportion of those who reported a specific childhood circumstance. For example, 10% of female respondents’ parents had been divorced and 90% had not been divorced.
in Table 2. Women whose parents had low education reported limited adult physical functioning more often than women whose parents had high education. The association between parental education and adult functioning remained unchanged after adjusting for specific childhood circumstances but disappeared after own education was adjusted for. Of the childhood circumstances own chronic disease, parental mental problems, economic difficulties and having been bullied were all associated with limited physical functioning in adulthood. However, there were no associations of parental
divorce, death of a parent or parental drinking problems with limited adult physical functioning. All associations of specific childhood circumstances with limited adult physical functioning remained after parental as well as own education were adjusted for. Among men, parental education was not associated with adult physical functioning (Table 2). However, as for women, own chronic disease, economic difficulties and having been bullied in childhood were associated with limited adult physical functioning, but parental mental problems
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
1835
Table 2 Associations of parental education and specific childhood circumstances with limited adult physical functioning among women and men. Logistic regression analysis (OR ¼ odds ratio) and 95% confidence intervals (CI) Women
Men
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
OR CI
OR CI
OR CI
OR CI
OR CI
OR CI
Parental education High Intermediate Low
1.00 1.00 1.00 1.00 1.00 1.00 1.35 (1.14–1.61) 1.32 (1.11–1.57) 1.11 (0.93–1.33) 0.92 (0.67–1.27) 0.92 (0.66–1.27) 0.77 (0.55–1.08) 1.46 (1.25–1.61) 1.39 (1.19–1.63) 1.03 (0.87–1.22) 1.25 (0.95–1.64) 1.21 (0.92–1.59) 0.90 (0.67–1.22)
Specific childhood circumstancesa Own chronic disease Parental divorce Parental death Parental mental problem Parental drinking problem Economic difficulties Having been bullied in childhood
1.48 0.93 0.84 1.31 1.03 1.51 1.64
a
(1.20–1.82) (0.76–1.13) (0.71–1.00) (1.03–1.66) (0.88–1.20) (1.31–1.75) (1.34–2.00)
1.49 0.94 0.82 1.34 1.02 1.46 1.63
(1.21–1.84) (0.77–1.15) (0.70–0.98) (1.05–1.70) (0.87–1.19) (1.25–1.70) (1.33–1.99)
1.54 0.87 0.78 1.38 1.02 1.44 1.54
(1.24–1.90) (0.72–1.07) (0.66–0.92) (1.09–1.76) (0.87–1.19) (1.24–1.67) (1.26–1.89)
1.62 0.94 1.04 0.89 1.06 1.55 1.42
(1.13–2.33) (0.65–1.34) (0.74–1.44) (0.49–1.63) (0.78–1.44) (1.16–2.08) (0.99–2.02)
1.66 0.96 1.03 0.89 1.05 1.50 1.44
(1.13–2.35) (0.67–1.38) (0.74–1.43) (0.49–1.62) (0.78–1.43) (1.12–2.02) (1.00–2.06)
1.54 0.86 0.96 0.93 1.06 1.46 1.47
(1.07–2.23) (0.60–1.24) (0.69–1.34) (0.50–1.70) (0.78–1.44) (1.08–1.97) (1.02–2.11)
The reference category (OR ¼ 1.00) was those reporting no specific childhood circumstances.
were not. Associations of these three specific childhood circumstances with limited adult physical functioning remained even after adjustments. Women whose parents had high education reported limited adult mental functioning more often than women whose parents had low education (Table 3). This association was unaffected by the adjustments for specific childhood circumstances and for own education. Among women, parental mental problem, chronic disease, parental drinking problem, economic difficulties and having been bullied in childhood were associated with limited adult mental functioning. These associations remained after parental as well as own education were adjusted for. Among men, the association of parental education with adult mental functioning was similar to that among women, but it failed to reach statistical significance (Table 3). Among men parental mental problems, economic difficulties and having been bullied in childhood were associated with limited adult mental functioning. In contrast to women, there were no associations of childhood chronic disease and parental drinking problems with limited adult mental functioning among men. Also among men all associations between specific childhood circumstances and limited adult mental functioning remained after adjusting for parental and own education.
In order to examine cumulative effects of childhood and adulthood circumstances on adult functioning, the categorizations of the variables were simplified. For parental education the categories of intermediate and high education were collapsed. Specific childhood circumstances were combined and dichotomized into those having experienced none of the specific childhood circumstances and those having experienced one or more specific childhood circumstances. Own low and intermediate education were also collapsed. Specific childhood circumstances and parental education were then cross-classified to produce four combinations. Results from the cumulative analyses confirmed that specific childhood circumstances increased the prevalence for limited adult physical functioning but the role of parental education was modest (Table 4). Among those with low own education the prevalence for limited physical functioning was higher among those with poor combined childhood circumstances. Among those with high parental education the prevalence for limited functioning was higher for those with one or more specific childhood circumstances, regardless of their parental education. The interaction between combined childhood circumstances and own education was statistically significant among women (p ¼ 0:02) but not among men. Among women with combined childhood circumstances and low own education
ARTICLE IN PRESS 1836
T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
Table 3 Associations of parental education and specific childhood circumstances with limited adult mental functioning among women and men. Logistic regression analysis (OR ¼ odds ratio) and 95% confidence intervals (CI) Women
Men
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
Model 1 Age adjusted
Model 2 Mutually adjusted
Model 3 Model 2 + own education
OR CI
OR CI
OR CI
OR CI
OR CI
OR CI
Parental education High Intermediate Low
1.00 1.00 1.00 1.00 1.00 1.00 0.81 (0.70–0.95) 0.77 (0.66–0.90) 0.79 (0.67–0.93) 0.78 (0.58–1.06) 0.73 (0.54–0.99) 0.75 (0.54–1.03) 0.78 (0.68–0.90) 0.72 (0.63–0.83) 0.76 (0.65–0.89) 0.79 (0.61–1.03) 0.71 (0.54–0.94) 0.74 (0.56–1.00)
Specific childhood circumstancesa Own chronic disease Parental divorce Parental death Parental mental problem Parental drinking problem Economic difficulties Having been bullied in childhood
1.41 0.90 0.87 2.01 1.32 1.46 1.62
a
(1.14–1.74) (0.75–1.09) (0.74–1.04) (1.61–2.50) (1.14–1.52) (1.26–1.69) (1.33–1.97)
1.40 0.89 0.89 1.97 1.33 1.53 1.64
(1.13–1.72) (0.74–1.08) (0.75–1.05) (1.58–2.47) (1.15–1.54) (1.32–1.77) (1.35–1.99)
1.40 0.90 0.90 1.96 1.33 1.53 1.66
(1.13–1.72) (0.74–1.09) (0.76–1.07) (1.57–2.45) (1.15–1.54) (1.32–1.77) (1.36–2.01)
1.00 1.18 1.03 1.77 1.23 1.71 1.78
(0.68–1.49) (0.85–1.65) (0.73–1.44) (1.05–2.99) (0.92–1.65) (1.28–2.30) (1.27–2.51)
0.97 1.16 1.03 1.76 1.27 1.80 1.77
(0.66–1.45) (0.83–1.63) (0.73–1.44) (1.04–2.98) (0.94–1.70) (1.33–2.42) (1.26–2.50)
0.98 1.18 1.04 1.75 1.27 1.81 1.77
(0.66–1.46) (0.84–1.66) (0.74–1.46) (1.04–2.95) (0.94–1.70) (1.34–2.43) (1.26–2.49)
The reference category (OR ¼ 1.00) was those reporting no specific childhood circumstances.
Table 4 Limited adult physical and mental functioning by combined childhood circumstances and own education among women and men. Prevalence percentages and 95% confidence intervals (CI) Women
Men
Own education
Own education
All
Low
High
All
Low
High
Combined childhood circumstances
CI
CI
CI
CI
CI
CI
Low parental education and one or more specific childhood circumstance Low parental education and no specific childhood circumstances High parental education and one or more specific childhood circumstance High parental education and no specific childhood circumstances
30 (29–31)
36 (34–38)
21 (20–22)
32 (30–34)
39 (35–43)
23 (20–26)
25 (24–26)
31 (29–33)
17 (16–18)
23 (21–25)
29 (26–32)
17 (15–19)
25 (24–26)
30 (28–32)
23 (22–24)
25 (23–27)
26 (23–29)
25 (22–28)
19 (18–20)
29 (27–31)
17 (16–18)
19 (17–21)
27 (24–30)
16 (14–18)
Total %
25 (24–26)
33 (31–35)
19 (18–20)
25 (23–27)
33 (30–36)
19 (17–21)
29 (28–30)
28 (26–30)
30 (29–31)
26 (24–28)
25 (22–28)
28 (25–31)
19 (18–20)
27 (25–29)
21 (20–22)
19 (17–21)
19 (16–22)
19 (17–21)
30 (29–31)
29 (27–31)
31 (30–32)
33 (31–35)
30 (27–33)
34 (31–37)
22 (21–23)
22 (21–23)
22 (21–23)
20 (18–22)
17 (14–20)
20 (18–22)
25 (24–26)
24 (22–26)
26 (25–27)
25 (23–27)
23 (20–26)
25 (22–28)
Limited physical functioning
Limited mental functioning Combined childhood circumstances Low parental education and one or more specific childhood circumstance Low parental education and no specific childhood circumstances High parental education and one or more specific childhood circumstance High parental education and no specific childhood circumstances Total %
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
existed more limited physical functioning than among women with combined childhood circumstances and high own education. Mental functioning was more often limited among those who had experienced one or more specific childhood circumstances regardless of their parental education. The association was similar in both categories of own education as well as in women and men. Discussion This study focused on the associations of childhood circumstances with adult physical and mental functioning among middle-aged women and men employed by the City of Helsinki. Parental education was used as a generic indicator of socioeconomic circumstances in childhood and specific childhood circumstances were measured with seven indicators. Three life course models, latency, pathway and cumulative model, were utilized to highlight the nature of the associations between childhood circumstances and adult functioning. Among women low parental education was associated with limited adult physical functioning and high parental education with limited adult mental functioning. This inverse association between parental education and adult physical functioning disappeared after adjusting for own education whereas the corresponding association for mental functioning remained. This suggests that own education may mediate the effects of childhood circumstances on adult physical functioning. Previous studies (Hallqvist, Lynch, Bartley, Lang, & Blane, 2004; Hertzman, Power, Matthews, & Manor, 2001; Laaksonen, Rahkonen, Martikainen, & Lahelma, 2005; Lundberg, 1993; Pensola & Martikainen, 2003) equally support a similar mediated pathway. However, it remains an open question why high parental education was associated with limited adult mental functioning even after adjustment for specific childhood circumstances and own education. Several specific childhood circumstances remained associated with adult physical and mental functioning after adjusting for parental and own education. This may be taken as a suggestion for a latency effect. Own childhood chronic disease, parental mental problems, economic difficulties and having been bullied in childhood potentially have long lasting consequences for physical and mental functioning. A previous study has examined how specific childhood circumstances are associated
1837
with adult functioning (Lynch, Kaplan, & Shema, 1997). However, corresponding studies using other health outcomes have supported specific childhood circumstances being markedly associated with adult health (Hill et al., 1997; Lundberg, 1993; Van de Mheen et al., 1997; Rahkonen et al., 1997; Rodgers, 1994). Close to same childhood circumstances were associated with both adult physical and mental functioning. Own chronic disease, parental mental problems, economic difficulties and having been bullied in childhood were all associated with both physical and mental functioning. An exception was parental drinking problems which were only associated with mental functioning. There were minor differences in physical and mental functioning between women and men. Parental mental problems were associated with women’s but not men’s physical functioning. Among men own childhood chronic disease was not associated with mental functioning unlike among women. It has been found that children with poorly educated parents are likely to have fewer resources and receive less social support from their parents than children with better educated parents (Huurre, Eerola, Rahkonen, & Aro, submitted). Parents also influence their children’s health habits and there is evidence that childhood social position is associated with smoking (Brunner, Shipley, Blane, Davey Smith, & Marmot, 1999; Power et al., 2005; Tyas & Pederson, 1998) and obesity (Brunner et al., 1999) in adulthood. These findings confirm that adverse psycho-social circumstances in childhood may be critical for early socialization and later health behaviours. We did find some evidence on cumulative effects of childhood and adulthood circumstances among women on physical functioning. However, adverse childhood circumstances may be more important to physical functioning than parental education or own education. Several previous studies have examined the cumulative model (Hallqvist et al., 2004; Hart, Davey Smith, & Blane, 1998; Holland et al., 2000; Lynch et al., 1997; Power, Manor, & Matthews, 1999; Power, Matthews, & Manor, 1998; Wamala, Lynch, & Kaplan, 2001) and shown that low childhood social class, economic difficulties, shortage of early life resources and own chronic disease together with adulthood circumstances cumulatively affect adult health and health-related risks (Hart et al., 1998; Holland et al., 2000; Power et al., 1999; Power et al., 1998). However,
ARTICLE IN PRESS 1838
T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839
accumulation of adverse circumstances over the life course is likely to vary by gender (Power et al., 1998) and by childhood social class (Holland et al., 2000). Regarding the three life course models it has recently been questioned whether the latency, pathway and accumulation effects can in fact be distinguished from each other (Hallqvist et al., 2004). Nevertheless, our study renders some evidence on the latency, pathway and cumulative model. There are some limitations in our study to be kept in mind when interpreting the results. In a crosssectional design the childhood circumstances were asked retrospectively. All information on childhood and adulthood circumstances was self-reported and collected from middle-aged women and men. Distressed people may recall their life history in adverse terms and their present perceptions may have influenced their response through ‘negative affectivity’ (Watson, 1988). However, a recent study suggested that retrospective data on childhood circumstances are likely to be generally reliable (Dube, Williamson, Thompson, Felitti, & Anda, 2004; Krieger, Okamoto, & Selby, 1998). Since we only studied employees it is possible that those with the poorest functioning have dropped out as they may have already left the labour force. However, the large size of our data allowed an examination of even relatively rare specific childhood circumstances. In conclusion, parental education and specific childhood circumstances showed clear and strong associations with adult physical and mental functioning. However, especially among women parental education had different associations with adult physical and mental functioning. Several specific childhood circumstances were associated with physical and mental functioning in both women and men. Parental education is likely to follow a pathway effect on adult physical functioning. Our results suggest that specific childhood circumstances are likely to have latency effects on adult physical and mental functioning. Therefore these circumstances provide more detailed evidence than generic parental social position on how childhood is associated with adult functioning. Ideally the specific childhood circumstances are studied side by side with parental social position. Acknowledgements Helsinki Health Study is supported by grants from the Academy of Finland (#53245, #205588 and #105952) and the Finnish Work Environment
Fund (#103310). The Academy of Finland also supports Mikko Laaksonen (#204894) and Ossi Rahkonen (#45664 and #210435). We would like to thank the City of Helsinki and the members of the Helsinki Health Study research group. References Barker, D. J. P. (1995). Fetal origins of coronary heart disease. British Medical Journal, 311, 171–174. Ben-Shlomo, Y., & Kuh, D. (2002). A life course approach to chronic disease epidemiology: Conceptual models, empirical challenges and interdisciplinary perspectives. International Journal of Epidemiology, 31, 285–293. Brunner, E., Shipley, M. J., Blane, D., Smith, G. D., & Marmot, M. G. (1999). When does cardiovascular risk start? Past and present socioeconomic circumstances and risk factors in adulthood. Journal of Epidemiology and Community Health, 53, 757–764. Dube, S., Williamson, D. F., Thompson, T., Felitti, V., & Anda, R. (2004). Assessing the reliability of retrospective reports of adverse childhood experiences among adult HMO members attending a primary care clinic. Child Abuse & Neglect, 28, 729–737. Galobardes, B., Lynch, J. W., & Davey Smith, G. (2004). Childhood socioeconomic circumstances and cause-specific mortality in adulthood: Systematic review and interpretation. Epidemiological Reviews, 26, 7–21. Graham, H. (2002). Building an inter-disciplinary science of health inequalities: The example of lifecourse research. Social Science & Medicine, 55, 2005–2016. Hallqvist, J., Lynch, J., Bartley, M., Lang, T., & Blane, D. (2004). Can we disentangle life course processes of accumulation critical period and social mobility? An analysis of disadvantaged socio-economic positions and myocardial infarction in the Stockholm heart epidemiology program. Social Science & Medicine, 58, 1555–1562. Harper, S., Lynch, J., Hsu, W.-L., Everson, S., Hillemeier, M., Raghunathan, T., Salonen, J., & Kaplan, G. (2002). Life course socioeconomic conditions and adult psychosocial functioning. International Journal of Epidemiology, 31, 395–403. Hart, C., Davey Smith, G., & Blane, D. (1998). Inequalities in mortality by social class measured at 3 stages of the life course. American Journal of Public Health, 88, 471–474. Hayward, M. D., & Gorman, B. K. (2004). The long arm of childhood: The influence of early life social conditions on men’s mortality. Demography, 41, 87–107. Hemingway, H., Nicholson, A., Stafford, M., Roberts, R., & Marmot, M. (1997). The impact of socioeconomic status on health functioning as assessed by the SF-36 questionnaire: The whitehall II study. American Journal of Public Health., 87, 1484–1490. Hertzman, C. (1999). The biological embedding of the early experience and its effects on health in adulthood. Annals of the New York Academy of Sciences, 869, 85–95. Hertzman, C., Power, C., Matthews, S., & Manor, O. (2001). Using an interactive framework of society and lifecourse to explain self-rated health in early adulthood. Social Science & Medicine, 53, 1575–1585.
ARTICLE IN PRESS T. Ma¨kinen et al. / Social Science & Medicine 62 (2006) 1831–1839 Hill, E., Thomson Ross, L., Mudd, S., & Blow, F. (1997). Adulthood functioning: The joint effects of parental alcoholism, gender and childhood socio-economic stress. Addiction, 92, 583–596. Holland, P., Berney, L., Blane, D., Davey Smith, G., Gunnel, D. J., & Montgomery, S. M. (2000). Life course accumulation of disadvantage: Childhood health and hazard exposure during adulthood. Social Science & Medicine, 50, 1285–1295. Huurre, T., Aro, H., & Rahkonen, O. (2003). Well-being and health behaviour by parental socioeconomic status. A follow up study of adolescents aged 16 until age 32 years. Social Psychiatry and Psychiatric Epidemiology, 38, 249–255. Huurre, T., Eerola, M., Rahkonen, O., & Aro H. Does social support affect the relationship between socio-economic status and depression? A longitudinal study from adolescence to adulthood, submitted for publication. Krieger, N., Okamoto, A., & Selby, J. V. (1998). Adult female twins’ recall of childhood social class and father’s education: A validation study for public health research. American Journal of Epidemiology, 147, 704–708. Kuh, D., & Ben-Shlomo, Y. (1997). A life course approach to chronic disease Epidemiology. Oxford Medical Publications. Kuh, D., Ben-Shlomo, Y., Lynch, J., Hallqvist, J., & Power, C. (2003). Life course epidemiology. Glossary. Journal of Epidemiology and Community Health, 57, 778–783. Kuper, H., Singh-Manoux, A., Siegrist, J., & Marmot, M. (2002). When reciprocity fails: Effort–reward imbalance in relation to coronary heart disease and health functioning within the whitehall II study. Occupational and Environmental Medicine, 59, 777–784. Laaksonen, M., Rahkonen, O., Martikainen, P., & Lahelma, E. (2005). Multiple dimensions of socioeconomic position and self-rated health. The contribution of childhood socioeconomic circumstances, adult socioeconomic status and material resources. American Journal of Public Health, 95, 1403–1409. Lahelma, E., Martikainen, P., Rahkonen, O., Roos, E., & Saastamoinen, P. (2005). Occupational class inequalities across key domains of health: Results from the Helsinki Health Study. European Journal of Public Health, Advance access published, July 13, doi:10.1093/eurpub/cki022. Lallukka, T., Aittoma¨ki, A., Piha, K., Roos, E., Kivela¨, K., & Silventoinen, K. (2002). Postikyselytutkimukseen vastanneiden edustavuus sosioekonomisten tekijo¨iden ja sairauspoissaolotietojen mukaan: Helsingin kaupungin henkilo¨sto¨n terveystutkimus. Sosiaalila¨a¨ketieteellinen Aikakauslehti, 39, 164–171. Lundberg, O. (1993). The impact of childhood living conditions on illness and mortality in adulthood. Social Science & Medicine, 36, 1047–1052. Lynch, J. W., Kaplan, G. A., & Shema, S. J. (1997). Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. New England Journal of Medicine, 337, 547–552. Martikainen, P., Stansfield, S., Hemingway, H., & Marmot, M. (1999). Determinants of socioeconomic differences in change in physical and mental functioning. Social Science & Medicine, 49, 499–507. Van de Mheen, H., Stronks, K., Van de Bos, J., & Mackenbach, J. P. (1997). The contribution of childhood environment to the explanation of socio-economic inequalities in health in
1839
adult life: A retrospective study. Social Science & Medicine, 44, 13–24. Notkola, V., Punsar, S., Karvonen, MJ., & Haapakoski, J. (1985). Socioeconomic conditions in childhood and mortality and morbidity caused by coronary heart disease in adulthood in rural Finland. Social Science & Medicine, 21, 517–523. Pensola, T., & Martikainen, P. (2003). Cumulative social class and mortality from various causes of adult men. Journal of Epidemiology and Community Health, 57, 745–751. Pensola, T., & Valkonen, T. (2002). Effect of parental social class, own education and social class on mortality among young men. European Journal of Public Health, 12, 29–36. Power, C., Manor, O., & Matthews, S. (1999). The duration and timing of exposure: Effects of socio-economic environment on adult health. American Journal of Public Health, 89, 1059–1065. Power, C., Matthews, S., & Manor, O. (1998). Inequalities in selfrated health: Explanations from different stages of life. Lancet, 351, 1009–1013. Power, C., Graham, H., Due, P., Hallqvist, J., Joung, I., Kuh, D., & Lynch, J. (2005). The contribution of childhood and adult socioeconomic position to adult obesity and smoking behaviour: An international comparison. International Journal of Epidemiology, advance access published on January 19. Rahkonen, O., Lahelma, E., & Huuhka, M. (1997). Past or present? Childhood living conditions and current socioeconomic status as determinants of adult health. Social Science & Medicine, 44, 327–336. Rodgers, B. (1994). Pathways between parental divorce and adult depression. Journal of Child Psychology and Psychiatry, 35, 1289–1308. Stansfeld, S. A., Bosma, H., Hemingway, H., & Marmot, M. (1998). Psychosocial work characteristics and social support as predictors of SF-36 health functioning: The whitehall II study. Psychosomatic Medicine, 60, 247–255. Stansfeld, S. A., Head, J., Fuhrer, R., Wardle, J., & Cattell, V. (2003). Social inequalities in depressive symptoms and physical functioning in the whitehall II study: Exploring a common cause explanation. Journal of Epidemiology and Community Health, 57, 361–367. Tyas, S. L., & Pederson, L. L. (1998). Psychosocial factors related to adolescent smoking: A critical review of the literature. Tobacco Control, 7, 409–420. Va˚gero¨, D., & Illsley, R. (1995). Explaining health inequalities: Beyond black and barker. European Sociological Review, 11, 219–241. Wamala, S. P., Lynch, J. W., & Kaplan, G. A. (2001). Women’s exposure to early and later life socioeconomic disadvantage and coronary heart disease risk: The Stockholm female coronary risk study. International Journal of Epidemiology, 30, 275–284. Ware, J. SF-36s Health survey update. www.sf-36.org/tools/ SF36bookschapter.shtml (13.5.2003), 1–17. Ware, J., & Kosinski, M. (2001). SF-36 Physical and mental health summary scales: A manual for users of version 1 (2nd ed.). Lincoln Rhode Island: QualityMetric Incorporated. Watson, D. (1988). Intraindividual and interindividual analyses of positive and negative affect: Their relation to health complaints, perceived stress, and daily activities. Journal of Personality & Social Psychology, 54, 1020–1030.