Work careers in adults separated temporarily from their parents in childhood during World War II

This is a self-archived version of an original article. This version may differ from the original in pagination and typographic details. Author(s): Title: Year: Version: Copyright: Rights: Rights url: Please cite the original version: CC BY-NC-ND 4.0 https://creativecommons.org/licenses/by-nc-nd/4.0/ Work careers in adults separated temporarily from their parents in childhood during World War II © 2019 Elsevier Inc. Accepted version (Final draft) Salonen, Minna K.; von Bonsdorff, Mikaela; Kautiainen, Hannu; von Bonsdorff, Monika; Kajantie, Eero; Wasenius, Niko S.; Pesonen, Anukatriina; Räikkönen, Katri; Eriksson, Johan G. Salonen, M. K., von Bonsdorff, M., Kautiainen, H., von Bonsdorff, M., Kajantie, E., Wasenius, N. S., Pesonen, A., Räikkönen, K., & Eriksson, J. G. (2019). Work careers in adults separated temporarily from their parents in childhood during World War II. Journal of Psychosomatic Research, 118, 63-68. https://doi.org/10.1016/j.jpsychores.2019.01.014 2019 WORK CAREERS IN ADULTS SEPARATED TEMPORARILY FROM THEIR PARENTS IN CHILDHOOD DURING WORLD WAR II Running title: Temporary separation from parents in childhood and work career PhD Minna K Salonen1,2, PhD Mikaela B von Bonsdorff 1,3, MSc Hannu Kautiainen 1,7, PhD. Monika E von Bonsdorff 3, MD, PhD. Eero Kajantie2,4,5, PhD Niko S Wasenius1, Professor, PhD Anukatriina Pesonen6, Professor, PhD Katri Räikkönen6 , Professor, MD, PhD Johan G Eriksson1,2,7,8,9 1 Folkhälsan Research Center, Helsinki, Finland 2 National Institute for Health and Welfare, Department of Public Health Solutions, Chronic Disease Prevention Unit, Helsinki, Finland 3 Gerontology Research Center, Faculty of Sport and Health Sciences, University of Jyväskylä, Jyväskylä, Finland 4 Hospital of Children and Adolescents, Helsinki University Central Hospital and University of Helsinki, Helsinki, Finland 5 PEDEGO Research Unit, MRC Oulu, Oulu University Hospital and University of Oulu, Oulu, Finland 6 University of Helsinki, Department of Psychology and Logopedics, Helsinki, Finland 7 University of Helsinki and Helsinki University Hospital, Department of General Practice and Primary Health Care, Helsinki, Finland 8 Singapore Institute for Clinical Sciences (SICS), Agency for Science and Technology (A*STAR), Singapore 9 Obstretics & Gynecology, Yong Loo Lin School of Medicine, National University of Singapore and National University Health System, Singapore Corresponding author: PhD Minna K Salonen, National Institute for Health and Welfare, Chronic Disease Prevention Unit, P.O.X 30, FI-00271 Helsinki, Finland, email: [email protected] ABSTRACT Introduction: Traumatic experiences, such as separation from parents in childhood causing early life stress (ELS) may increase the risk of adverse long-term health outcomes and biological agerelated changes. This may have an impact on work career. Our aim was to examine long term consequences of ELS due to temporary separation from parents during World War II (WWII) in relation to work career. Material and Methods: The Helsinki Birth Cohort Study comprises 13345 individuals born in Helsinki, Finland, between the years 1934-1944. From the original cohort, 1781 individuals were identified as being separated temporarily from their parents due to World War II. Information on date and type of pension was provided by the Finnish Centre for Pensions and the Social Insurance Institution of Finland. The cohort members either transitioned into old age pension at the statutory retirement age or retired earlier and transitioned into disability, unemployment, part-time pension or died before retirement. Results: Those who were separated were more likely to have transitioned into disability pension (RRR: 1.26: 95% CI: 1.06-1.48), especially due to diseases of the musculoskeletal system (RRR: 1.57; 95% CI: 1.20-2.07), or into unemployment pension (RRR: 1.25; 95% CI: 1.02-1.53) compared with those not separated from their parents. Longer duration of separation was associated with early exit from the workforce compared with non-separation. Conclusions: Exposure to ELS may have an impact upon lifetime work career. Early interventions preventing exposure to ELS or mitigating its negative effects may prolong future work careers along with healthier aging across the life-span. Key words: aging, chronic disease, labor force, life cycle, life stress, separation Introduction Severe, traumatic conditions and experiences early in life such as war, temporary separation from parents due to war, childhood abuse, neglect or severe illness in the family are important sources of early life stress (ELS) and may lead to an increased risk of adverse health outcomes across the life span (1). For example, separation from parents due to World War II (WWII) is a known risk factor for mental health disorders in later life (2, 3, 4, 5) that could persist across generations (6) and may induce physiological and biological changes and pathologies (such as poor physical and psychosocial functioning, frailty, type 2 diabetes and cardiovascular disease) (7, 8). The results of a recent extensive study by Hughes et al. that examined the mentioned associations between ELS and unfavorable health outcomes show that those having multiple childhood adversities were at higher risk for a variety of poor health outcomes and health related behaviors (9). Furthermore, other recently published studies showed that childhood adversities were associated with permanent sickness, part-time employment and earlier retirement age at 55 years (10) and that childhood psychosocial adversity and low adult socioeconomic status were additive risk factors for work disability (11). Many governments have introduced pension reforms in order to deal with continuously growing public expenditures due to the greying of the population in OECD countries. For example, in 1950’s there were more than seven working-aged persons for each retired person but by the year 2050, the ratio will be two for one (12). To control for increasing costs, early exit from labor market has been reduced through e.g. changes to retirement legislation and by rising the statutory retirement age (13). Understanding the way in which work career in later life may be influenced by childhood ELS is of importance in terms of the designing and implementing proactive, early intervention aiming to prolong future work careers along with healthier aging across the life course. To the best of our knowledge, no studies have investigated the association between ELS due to parental separation during wartime evacuations and subsequent work career exits. Thus, by using data across the life course from the Helsinki Birth Cohort Study (HBCS), we aimed to study whether age of statutory retirement and early exit from the workforce due to disability pension, unemployment pension, part time pension or death before retirement differed between those who were exposed to ELS and those who were not. In the HBCS, data on war time separation were used to define ELS. During WWII a proportion of the HBCS cohort had been temporarily separated from their families due to evacuations of children to temporary foster families in Sweden or Denmark. We also examined whether those who transitioned into disability pension due to mental disorders, the diseases of musculoskeletal system, cardiovascular diseases, and diseases of nervous system or “all other reasons” differed according to wartime separation status. Materials and methods HBCS comprises 13 345 men and women who were born in Helsinki, Finland at Helsinki University Central Hospital or City Maternity Hospital between the years 1934 and 1944 (14, 15). In the year 1971 all residents of Finland were assigned an individual identification number which was used to link data from several national registers upto the end of year 2013. Using the Register of the Finnish National Archives, managed by the Ministry of Social Affairs and Health between the years 1939 and 1946, 1781 (13.3 %) individuals from the original cohort were identified as having been separated temporarily from their biological parents and sent abroad as children due to WWII (2, 16). Although the register is of high quality and includes most of the evacuations, some children were also evacuated privately through personal contacts (2, 16). Of the original cohort, we excluded those with missing data on retirement (n=112), those who had migrated before retirement (n=507) and those who had reported being evacuated, but whose information could not be found in the register of the Finnish National Archives (n=189) (2, 16). Hence, the final study population included 12 537 men and women of whom 1645 had been evacuated abroad in childhood. The study complies with the guidelines of the Declaration of Helsinki and was approved by the Ethics Committee of Epidemiology and Public Health of the Hospital District of Helsinki and Uusimaa and the National Public Health Institute. Early exit from the workforce While the average retirement age in OECD countries in 2014 was 64.6 years for men and 63.1 years for women (13), in 2011 in Finland, the average age of statutory retirement was 60.5 years and of those who transitioned into disability retirement, a third had retired by the age of 52 years (17). In Finland, the National Pension Act was carried out in 1937 in order to secure pension in case of old age or disability. All Finnish residents who had lived in Finland for at least 3 years after they turned 16 years are covered by the pension system. Until 2005, the statutory retirement age was 63-65, and from 2005 to 2016 63-69 years but certain professions, such as firemen, pilots and army officers, had younger retirement ages. Approximately 5% of the cohort continued working after the age of 65 years (e.g. entrepreneurs in the private sector), and there are no retirement data available for these cohort members (18). Among those who transitioned into old-age pensions, 90% retired at the age of 58-65 years (19). Information on date and type of pension was provided by the Finnish Centre for Pensions and the Finnish Social Insurance Institution. Information on death was obtained from the Finnish Population Register Centre. In the present study early exit from the workforce meant that the person transitioned into retirement before statutory pension age due to disability pension, unemployment or part-time pension or they died before retiring and thus did not reach a full work career. Disability pension is granted when a person is unable to continue working due to a medically confirmed illness whilst a part-time pensioner is partly working and partly retired based on their own whish (a person could retire on a part-time pension at the age of 61, providing their working life before retirement was long enough),or impaired working capacity due to medical condition. Criteria to qualify for pensions are described in more detail elsewhere (17, 18, 19, 20). Medically confirmed main causes of disability pensions were coded as according to the International Classification of Diseases (ICD) 9th and 10th versions into the following groups: mental disorders (ICD 9 codes 290-319 and ICD 10 codes F00-F99), diseases of musculoskeletal system (710-739 and M00-M99), cardiovascular diseases (400-459 and I00-I99), diseases of the nervous system (320-359 and G00-G99) and all other diagnoses. Socioeconomic background Childhood socioeconomic status was based on father’s highest occupation at birth, child welfare and school healthcare records and was coded as high official, low official and manual workers based on the original social classification system by Statistics Finland (21). Information on adult socioeconomic status was available from a register kept by the Statistics Finland, recorded at 5-year intervals between the years 1970-2000. We used the highest achieved occupation and it was coded as high official, low official, self-employed and manual workers. In a similar way, educational attainment was available at 5-year intervals between the years 1970-2000 and the highest achieved educational level is used in the analysis. Statistical methods Characteristics of the study population between the separated and non-separated were compared using t-test for continuous variables and χ2-square test for categorical variables. Cox proportional hazard model was used to examine the associations between separation status during WWII and early exit from the workforce vs. statutory pension. The analyses were stratified for sex and year of birth, and adjustments were made for birth order, maternal age and childhood socioeconomic status. Results from the Cox model analysis were non-significant (HR: 1.04 with 95% CI: 0.97-1.12; p=0.229). In order to reduce the effects of confounding and mitigate against the potential selection bias (22) multinomial logistic regression analyses using inverse probability of treatment weighting method (IPTW) were used to calculate the relative risk ratios (RRR) and 95% confidence intervals (CI) for the types of early exit from the workforce according to separation status, old age pension being the reference category. Models were adjusted for the following covariates: sex (when both sexes were combined), birth order, maternal age (both based on the birth records of maternal hospitals), childhood socioeconomic status and year of birth. Birth order was categorised as follows: firstborn, the second, the third or later. Further adjustments for adulthood socioeconomic status, or alternatively educational attainment, did not alter the results. In a similar way, multinomial logistic regression using IPTW-method (based on the covariates mentioned above) was used to calculate RRRs for the main causes of disability pension. Binary logistic regression was used to calculate odds ratios (OR) for disability pension (all types combined) as well as for early exit from the workforce according to age at the first separation (if one had multiple periods of separations) and duration of separation; non-evacuated formed the reference category. The analyses according age at separation was firstly adjusted for sex, and secondly, in addition to sex, also for birth order, maternal age, childhood socioeconomic status as well as year of birth, and thirdly, also for duration of separation. The analyses for early exit from the workforce according to the duration of separation were adjusted in a similar way as the calculations according to age at separation, except the duration of separation was replaced with the age of separation. All the analysis included tests of interaction to assess whether the effects of separation varied by sex or childhood socioeconomic background. When interactions were not found, the results are presented as both sexes combined. Analyses were carried out with IBM SPSS (Statistical Package for Social Sciences) version 25 and for the treatment weighting models using STATA 14.1 (StataCorp LP, College Station, TX). Results Characteristics of the study population are presented in Table 1. Boys who were separated tended to be firstborns to younger mothers and were more likely born before WWII. They belonged to a lower socioeconomic class in childhood, were less educated and had lower socioeconomic status in adulthood than their non-separated peers. In a similar way the separated girls had younger mothers and tended to be more frequently firstborns, belonged to a lower childhood socio economic class and were less educated than the non-separated girls. However, there were no statistically significant differences according to separation status and adult socioeconomic status among the girls. Figure 1 shows the probabilities and RRRs for both sexes combined and then separately for men and women for the four types of early exits from the workforce according to separation status with old age pension being the reference group. Relative to non-separated, those who were separated had higher risk ratios of transitioning into disability or unemployment pension. The analyses stratified according to gender revealed that in contrary to the women, these associations were statistically non-significant among men with an exception that having been separated decreased the RRRs of transitioning into part-time pension, the interaction between sex and separation status on transitioning into part-time pension being statistically significant (p=0.004). Findings of the associations between wartime separation status and the main causes of disability pension are shown in Table 2. Compared to those who were not separated as children, the separated ones had an increased risk of transitioning into disability pension due to diseases of musculoskeletal system (RRR 1.57 95% CI 1.20-2.04). Associations between age at and duration of wartime separation and transitioning into disability pension are shown in Table 3. Results from the analyses using in the first model show that a higher age at separation, from early childhood (> 4 years) and all the way through school-age (> 7 years), was associated with higher likelihood of transitioning into disability pension. However, after further adjustments, statistically significant associations were attenuated to non-significant, especially after 5. Santavirta T, Santavirta N, Betancourt TS, Gilman SE. Long term mental health outcomes of Finnish children evacuated to Swedish families during the second world war and their non-evacuated siblings: cohort study. BMJ 2015;350:g7753. 6. Santavirta T, Santavirta N, Gilman SE. Association of the world war II Finnish evacuation of children with psychiatric hospitalization in the next generation. JAMA Psychiatry 2018;75:21-7. 7. Alastalo H, Räikkönen K, Pesonen A-K, Osmond C, Barker DJ, Kajantie E et al. Cardiovascular health of Finnish war evacuees 60 years later. Ann Med. 2009;41:66-72. 8. Alastalo H, Von Bonsdorff MB, Räikkönen K, Pesonen A-K, Osmond C, Barker DJ et al. Early life stress and physical and psychosocial functioning in late adulthood. PloS One. 2013;8:e69011. 9. Hughes K, Bellis MA, Hardcastle KA, Sethi D, Butchart A, Christopher Mikton et al. The effect of multiple adverse childhood experiences on health: a systematic review and metaanalysis. Lancet Public Health 2017; 2: e356–66. 10. Fahy AE, Stansfeld SA, Smuk M, Lain D, van der Horst M, Vickerstaff S et al. Longitudinal associations of experiences of adversity and socioeconomic disadvantage during childhood with labor force participation and exit in later adulthood. Soc Sci Med. 2017;83:80-87. 11. Halonen JI, Kivimäki M, Vahtera J, Pentti J, Virtanen M, Ervasti J et al. Childhood adversity, adult socioeconomic status and risk of work disability: a prospective cohort study. Occup Environ Med. 2017;74:659-66. 12. OECD (2011) Pensions at a Glance 2011: Retirement income systems in OECD and G20 countries. Paris: OECD Publishing. 13. OECD (2015) Pensions at a Glance 2015: OECD and G20 indicators. Paris: OECD Publishing. 14. Eriksson JG, Forsén T, Tuomilehto J, Winter PD, Osmond C, Barker DJ. Catch-up growth in childhood and death from coronary heart disease: longitudinal study. BMJ. 1999;318:42731. 15. Barker DJ, Osmond C, Forsén TJ, Kajantie E, Eriksson JG. Trajectories of growth among children who have coronary events as adults. N Engl J Med. 2005;353:1802-9. 16. Pesonen A-K, Räikkönen K, Feldt K, Heinonen K, Osmond C, Phillips DI et al. Childhood separation experience predicts HPA axis hormonal responses in late adulthood: a natural experiment of World War II. Psychoneuroendocrinology. 2010;35:758-67. 17. Official Statistics of Finland (2013) Pensioners and insured in Finland 2011. Helsinki: Finnish Center for Pensioners. 18. von Bondorff MB, Törmäkangas T, Salonen MK, von Bonsdorff ME, Osmond C, Kajantie E et al. Johan G. Early life origins of all-cause and cause specific disability pension: findings from the Helsinki Birth Cohort Study. PlosOne. 2015;10:e0122134 19. von Bonsdorff MB, von Bonsdorff ME, Haanpää M, Salonen M, Mikkola TM, Kautiainen H et al. Work-loss years among people diagnosed with diabetes: a reappraisal from a life course perspective. Acta Diabetol. 2018;55:485-91. 20. https://www.etk.fi/en/the-pension-system/pension-security/earnings-related-pensionbenefits/part-time-pension/ 21. Central Statistical Office of Finland (1989). Classification of socioeconomic groups: handbooks17 Helsinki, Finland: Central Statistical Office of Finland. 22. Cattaneo MD. Efficient semiparametric estimation of multi-valued treatment effects of under ignorability. Journal of econometrics. 2010; 155:138-54. 23. Anda RF, Felitti FJ, Bremner JD, Walker JD, Whitfield C, Perry BD et al. The enduring effects of abuse and related adverse experiences in childhood. Eur Arch psychiatry Clin Neurosci. 2006;256:174-86. 24. Clark C, Caldwell T, Power C, Stansfeld SA. Does the influence of childhood adversity on psychopathology persist across the lifecourse? A 45-year prospective epidemiologic study. Ann Epidemiol. 2010;20:385-94. 25. Dube SR, Felitti VJ, Dong M, Chapman DP, Giles WH, Anda RF. Childhood abuse, neglect, and household dysfunction and the risk of illicit drug use: the adverse childhood experiences study. Pediatrics. 2003;111:564-72. 26. Teicher MH, Andersen SI, Polcari A, Anderson CM, Navalta CP, Kim DM. The neurobiological consequences of early stress and childhood maltreatment. Neurosci Biobehav Rev. 2003;27:33-44. 27. Pechtel P, Pizzagalli DA. Effects of early life stress on cognitive and affective function: an integrated review on human literature. Psychopharmacology. 2011;214:55-70. 28. Ford E, Clark C, Stansfeld SA. The influence of childhood adversity on social relations and mental health at mid-life. J Affect Disord. 2011;133:320-7. 29. Harkonmäki K, Korkeila K, Vahtera J, Kivimäki M, Suominen S, Sillanmäki L et al. Childhood adversities as a predictor of disability retirement. J Epidemiol Community Health. 2007;61:479-84. 30. Joensuu M, Mattila-Holappa P, Ahola K, Ervasti J, Kivimäki M, Kivekäs et al. Clustering of adversity in young adults on disability pension due to mental disorders: a latent class analysis. Soc Psychiatry Psychiatr Epidemiol. 2016;51:281-7. 31. Carpenter LL, Tyrka AR, Ross NS, Khoury L, Anderson GM, Price LH. Effect of childhood emotional abuse and age on cortisol responsitivity in adulthood. Biol Psychiatry. 2009;66:69-75. 32. Gunnar, MR, Frenn K, Wewerka SS, Van Ryzin MJ. Moderate versus severe early life stress: associations with stress reactivity and regulation in 10-12-year-old children. Psychoneuroendocrinology. 2009;34:62-75. 33. Yehuda R, Golier JA, Harvey PD, Stavitsky K, Kaufman S, Grossman RA et al. Relationship between cortisol and age-related memory impairments in Holocaust survivors with PTSD. Psychoneuroendocrinology. 2005;30:678–87. 34. Salonen J, Takala M. Working career and income of part-time pensioners in Finland. Finnish Centre for Pensions, Working papers 2011:2. http://www.etk.fi/en/julkaisu/workingcareer-and-income-of-part-time-pensioners-in-finland-2/ 35. Pesonen AK, Räikkönen K, Kajantie E, Heinonen K, Osmond C, Forsén T et al. Intergenerational social mobility following early life stress. Ann Med. 2011;43:320-8. 36. Bowlby J. The making and breaking of affectional bonds. I. Aetiology and psychopathology in the light of attachment theory. An expanded version of the Fiftieth Maudsley Lecture, delivered before the Royal College of Psychiatrists, 19 November 1976. Br J Psychiatry. 1977;130:201–10. doi:10.1192/bjp.130.3.201. 37. Carstensen J, Andersson D, Andre´ M, Engström S, Magnusson H, Borgquist LA. How does comorbidity influence healthcare costs? A population-based cross-sectional study of depression, back pain and osteoarthritis. BMJ Open 2012;2:e000809. doi:10.1136/bmjopen2011-000809. 38. Breivik H, Collett B, Ventafridda V, Cohen R, Gallacher D. Survey of chronic pain in Europe: prevalence, impact on daily life, and treatment. Eur J Pain. 2006;10:287-333. 39. Kwak J, Kim HK, Kim T, Jang SH, Lee KH, Kim MJ et al. The Prevalence and Characteristics of Depression in Work-related Musculoskeletal Disease. Ann Rehabil Med. 2012;36:836-40. 40. Official Statistics of Finland. Pensioners and insured in Finland 2011. 2013; Finnish Centre for Pensioners, Helsinki. 41. The UN Refugee Agency (UNCHR), Global Trends, Forced displacement in 2017. http://www.unhcr.org/globaltrends2017/ Table I. Characteristics of the study population (mean and standard deviation unless stated otherwise) according to separation status in men and women Men Women N Separation status Separation satus 12 537 Yes, n=903 No, n=5756 p-value Yes, n=742 No, n=5136 p-value Maternal age (years) 12 530 27.9 (5.6) 28.4 (5.4) <0.01 27.9 (5.4) 28.6 (5.5) 0.001 Birth order (%) 12 532 < 0.001 < 0.001 1st 42.7 50.1 41.6 48.8 2nd 26.7 28.1 28.6 29.3 > 3rd 30.6 21.1 29.8 21.9 Whether born at the time of WWII < 0.001 < 0.001 No, born before the 30th of November 1939 (%) 3755 28.9 72.9 24.1 75.9 Yes, born after the 30th of November 1939 (%) 8782 7.9 92.1 7.6 92.4 Age at separation (years) 1491 4.6 (2.4) - - 4.8 (2.4) -0.19 Duration of separation (years) 1457 1.8 (1.1) - - 1.7 (1.0) -0.06 Childhood socioeconomic status (%) 12 211 < 0.001 < 0.001 High official 11.1 19.0 10.9 17.3 Com paris ons were carrie d out betwe en those with and witho ut separ ation status Low official 21.8 24.2 22.7 24.9 Manual worker 67.2 56.8 66.4 57.8 Educational attainment (%) 12093 < 0.001 < 0.001 Basic or less 48.9 37.8 52.9 45.0 Upper secondary 25.1 26.0 23.2 23.6 Lower tertiary 18.5 22.9 18.1 22.8 Upper tertiary 7.6 13.3 5.9 8.5 Adult socioeconomic status (%) 11 866 < 0.001 0.07 High official 11.1 16.0 8.0 9.5 Low official 21.1 24.1 53.2 51.7 Self-employed 10.6 10.5 7.2 9.7 Manual worker 57.1 49.4 31.6 29.1 Table II. Relative Risk Ratios (RRR) and 95% confidence intervals for the main causes of disability pension among the men and women according to separation status old-age pension serving as a reference category. *N/n indicates for the proportions of people belonging to old-age pension and type of disability pension group (N=non-separated and n=separated; one may have more than one diagnosis for the disability pension) Inverse probability of treatment weighted multinomial logistic regression analysis based on sex, maternal age, birth order, childhood socioeconomic status and year of birth was applied N/n* RRR 95% CI p-value Old-age pension 4097/524 ref. ref. ref. Mental 1145/191 0.97 0.71 – 1.32 0.844 Musculoskeletal 945/216 1.57 1.20 – 2.04 <0.001 Cardiovascular 543/116 1.04 0.73 – 1.5 0.817 Nervous 293/51 1.29 0.85 – 1.96 0.230 Other 1100/215 1.16 0.93 – 1.43 0.182 Table III. Odds ratios and 95% confidence intervals (CI) for the disability pension according to the age at separation and duration of separation in men and women Model 1a Model 2a Model 3a OR 95% CI OR 95% CI OR 95% CI Disability pension Age at separation (years) Non separated ref. ref. ref. ref. Infancy (˂ 2) 1.25 0.92-1.70 1.18 0.86-1.61 1.00 0.66-1.51 Toddlerhood (2-4) 1.13 0.92-1.38 1.04 0.81-1.28 0.89 0.64-1.24 Early childhood (4-7) 1.11 0.91-1.09 1.04 0.85-1.28 0.89 0.64-1.23 School age (≥ 7) 1.40 1.09-1.780 1.32 1.02-1.71 1.16 0.83-1.62 Model 1b Model 2b Model 3b Disability pension Duration of separated (years) Non separated ref. ref. ref. ref. ≤ 1 0.96 0.75-1.24 0.92 0.71-1.19 0.82 0.56-1.22 1 - 2 1.26 1.07 - 1.49 1.21 1.02 - 1.43 1.08 0.76 - 1.53 2 - 3 1.08 0.82-1.42 1.01 0.76-1.35 0.91 0.61-1.36 ˃ 3 1.47 1.09 - 1.99 1.23 0.9 0 - 1.68 1.12 0.75 - 1.67 Model 1a is adjusted for sex and year of birth Model 2a is adjusted for sex, maternal age, birth order, childhood socioeconomic status and year of birth Model 3a is adjusted for sex, maternal age, birth order, childhood socioeconomic status, year of birth and duration of separation Model 1b is adjusted for sex and year of birth Model 2b is adjusted for sex, maternal age, birth order, childhood socioeconomic status and year of birth Model 3b is adjusted for sex, maternal age, birth order, childhood socioeconomic status, year of birth and age at separation Table IV. Odds ratios and 95% confidence intervals (CI) for early exit from the workforce according to the age at separation and duration of separation in men and women Model 1a Model 2a OR 95% CI OR 95% CI Early exit from the workforce Age at separation (years) Non separated ref. ref. Infancy ( ˂ 2 ) 1.29 0.96 - 1.75 0.98 0. 65 - 1.48 Toddlerhood (2-4) 1.24 1.02-1.51 0.95 0.69-1.30 Early childhood (4 - 7 ) 1.23 1.00 - 1.51 0.96 0.69 - 1.33 School age (≥ 7) 1.15 0.89-1.49 0.85 0.60-1.21 Model 1b Model 2b Early exit from the workforce Duration of separated (years) Non separated ref. ref. ≤ 1 1.08 0.85-1.37 1.13 0.77-1.66 1 - 2 1.22 1.04-1.44 1.26 0.89-1.78 2 - 3 1.17 0.90-1.54 1.17 0.79-1.74 ˃ 3 1.79 1.27-2.53 1.67 1.08-2.57 Model 1a is adjusted for sex and year of birth Model 2a is adjusted for sex, maternal age, birth order, childhood socioeconomic status, year of birth and duration of separation Model 1b is adjusted for sex and year of birth Model 2b is adjusted for sex, maternal age, birth order, childhood socioeconomic status, year of birth and age at separation