Does losing a parent early influence the education you obtain? A nationwide cohort study in Denmark

Does losing a parent early influence the education you obtain? A nationwide cohort study in Denmark Abstract Background Health inequalities are rooted in education and we investigate the association between early parental death and attainment across the educational spectrum. Methods Using total population data on Danes born between 1982 and 2000 (n = 1 043 813), we assess incidence rate ratios (RRs) by gender for attainment of each educational level (basic school, high school or vocational training, bachelor degree or professional programme, and university graduate degree) according to loss of a parent before the age of 18 years. We adjust for family income, education and psychiatric illness and examine parent’s gender, cause of death and child’s age at time of death as potential moderators. Results Bereaved people had significantly lower attainment rates than non-bereaved people: basic school (RR = 0.95; 95% CI: 0.93–0.97 for men and 0.96; 0.94–0.98 for women), high school or vocational training (0.78; 0.76–0.80 for men and 0.82; 0.80–0.84 for women), bachelor degree or professional programme (0.74; 0.70–0.79 for men and 0.83; 0.79–0.86 for women) and university graduate degree (0.77; 0.68–0.86 for men and 0.77; 0.69–0.86 for women). Parent’s gender, cause of death and child’s age at the death did not modify the associations. Conclusions As education impacts population health, support for bereaved school children may be more important than realized. children, education, employment and skills, epidemiology Introduction Education has a profound impact on population health, as success in the classroom predicts income and health behaviours, which subsequently influence not only physical and psychological morbidity but also, ultimately, mortality.1 An estimated 5% of children in Western countries experience parental death before they reach adulthood2 and research has shown that early parental death could have deleterious consequences for educational outcomes. For example, a 2014 cohort study in Sweden of 772 117 people showed that those who had been bereaved as children had up to twice the odds of failing basic school as those who had not.3 This problem warrants attention because of the strong link between education and health. Disparities in vulnerability to disease also have social consequences, which create health inequalities that can be passed on to the next generation.4,5 While education comprises many levels, most studies have not taken a life-course perspective and have focused only on fragmented aspects of education, with outcomes ranging from exam grades to enrolment in school.3,6–10 Furthermore, some studies addressed only paternal deaths,11,12 or were cross-sectional or based on self-reported survey data,7–10 which may potentially be biased. The roles of factors such as the deceased parent’s gender, cause of death and the child’s age at the time of the death are also not well studied. This knowledge is important because it can help identify potentially vulnerable sub-groups of children. For example, younger children who experience parental death may be more vulnerable than older children because of their early stage of cognitive development and limited capacity to cope with grief.13 Children who have lost to suicide may also be more severely affected because of the social stigma that often follows, which may complicate the grieving process and affect the support available to the grieving child from surrounding social environment.14 Finally, many studies were conducted in countries where education requires financial resources7,8,12 and the financial effect of parental death might obscure the study of other factors that potentially link the loss to educational outcomes. In summary, there is a gap in the literature regarding the association between early parental death and long-term educational attainment, with consistent measures at each level over time, to facilitate comparison and understanding of this important area. In this study, we investigate the association between early parental death and attaining a specific level of education, from basic school to university graduate studies. The Danish educational system is publically funded up to university level, and since 1970, all students over the age of 18 years can apply for a monthly allowance, which minimizes the economic effect on educational attainment. This provides a unique opportunity to disentangle economic factors from other loss-related factors in investigating our objective. We studied males and females separately, as educational outcomes traditionally differ by gender,15 and investigated the parent’s gender, cause of death and the child’s age at time of the death as potential moderators. In the context of free education, we hypothesized moderate effects of early parental death on attainment, where the role of the identified moderators will be especially apparent. Methods Study population We conducted a nationwide registry-based cohort study. All residents of Denmark have been recorded in the Central Population Register (CPR) since 1 April 196816 with a unique personal identification number that ensures accurate linkage of information among all national registries. As the Danish Education Registry was first started in 1981,17 we defined a study population that comprised all people born in Denmark between 1 January 1982 and 31 December 2000 whose parents could be identified at the time of their birth. From the CPR, we extracted information on gender, dates of birth, migration and death and family relationships. Parent–child links were identified through the Danish Fertility Database,18 which has complete information on the parents of all children born after 1960. We identified 1 238 740 children and their parents. We followed only those who were alive and living in Denmark and for each level of analysis, the following numbers were eligible: basic school, n = 1 133 356; high school or vocational training, n = 919 595; bachelor degree or professional programme, n = 771 166; and university graduate degree, n = 612 589. Due to missing values for parental variables, we had to exclude the following numbers of children: 89 543 (basic school), 73 296 (high school or vocational training), 60 617 (bachelor degree or professional programme) and 48 104 (university graduate degree). Thus, in total, we followed 1 043 813 people for attained compulsory school, 846 299 for attained high school or vocational training, 710 549 for attained bachelor degree or professional programme and 564 485 for attained university graduate degree. The study protocol was approved by the Danish Data Protection Agency (J.nr. 2009-41-3506). Ethics approval was not required, as the study was based on anonymised registers, and we did not use human blood or tissue samples. Parental death We obtained the causes of death of parents from the Danish Register of Cause of Death, which contains information on all deaths since 1970.19 We categorized cause of death by the World Health Organization’s International Classification of Diseases (ICD-10) codes for suicide (X60-X84, Y870) and for all other causes.20 We chose to focus on suicide because suicide have been shown to be distinct from other causes of death14 and may indicate a family who is psychosocially vulnerable pre-death and has been linked to worse health and well-being outcomes.21,22 Educational attainment Education in Denmark is free, except for books and certain administrative fees at tertiary level, and the Danish student grants and loans scheme allows all students over the age of 18 to apply for a monthly allowance.23 The scheme was started in 1970, as part of a socio-democratic policy to ensure equal access to education regardless of financial background. The highest pre-tax monthly allowance is currently DKK 5941 (approximately GBP 680) for tertiary-level students who do not live at home.23 We linked information on the study population and their parents to the Danish Education Registry, which contains individual information about all educational programmes in the country since 1981.17 We obtained information on type and level of education, whether it was completed and the year of completion. Parental education, which may be a confounder, was defined as the highest educational level of either parent and was obtained for the year before the birth of the child. The Danish educational system comprises 10 years of compulsory basic school (primary and lower secondary education for children aged 6–15 years), upper secondary education (high school or vocational or technical training) and tertiary education (consisting of 3–3.5-year bachelor degree or professional programmes and graduate Masters and PhD university programmes).24 We grouped educational level into four categories corresponding to the International Standard Classification of Education (ISCED-2011) codes: basic school (codes 1–2), high school or vocational training (3–4), bachelor degree or professional programme (5–6) and university graduate degree (7–8).25 Covariates Parental education, income and psychiatric illness were identified as potential confounders as family socioeconomic and psychiatric factors have been linked to both early parental death and child educational outcomes.22,26 Information on parental education, income and psychiatric illness was obtained for the year before the birth of the child from the Danish Education Register,17 Danish Income Statistics Register,27 the National Prescription Registry28 and the Central Psychiatric Register.29 We used the highest attained education level of either parent. Disposable household income per person was categorized as low (first quartile), middle (second and third quantiles) or high (fourth quartile), calculated in Danish kroner after taxation and adjusted for the number of people in the household and for inflation with the year 2000 as reference. Psychiatric illness was defined as first psychiatric hospitalization, outpatient treatment or use of antidepressants by either parent. We obtained information on all psycho-pharmaceutical prescriptions corresponding to Anatomical Therapeutic Chemical (ATC) code N06A for antidepressants30 and on hospital admissions for diagnoses related to all psychiatric disorders, court-ordered psychiatric treatment, substance use and suicide attempts. Information on education and income was available from 1981, on psychiatric hospitalization from 1969, and on antidepressant use and outpatient psychiatric treatment only from 1995. Statistical analyses For each educational level, we calculated person-years at risk from the year before the minimum possible age of attaining that specific level: age of 13 years for compulsory basic school, 16 years for high school or vocational training, 18 years for bachelor degree or professional programme and 20 years for university graduate degree. Exposure was defined as parental death before age 18 or age at entry, whichever came first. All people were followed until the first year in which they attained the investigated level of education, death, emigration or 31 December 2013, which was the end of the study period. The absolute level of educational attainment for all four levels in the study population and the cumulative probability of attaining the associated level before a given age was estimated non-parametrically according to gender and exposure status, taking into account competing risk from child’s death using the Aalen Johansen estimator31 calculated with Cuminc SAS macro.32 Exposure was defined as parental death before age 18 or age at entry, whichever came first. Incidence rate ratios (RRs) for educational attainment according to early parental death with 95% confidence intervals (CIs) were estimated in multiplicative Poisson models. We performed separate analyses for men and women. Person-time was stratified according to exposure status into two groups (parental death and no parental death), attained age into six groups (<20, 20–24, 25–29 and ≥30 years) and calendar periods into five groups (1988–92, 1993–97, 1998–2002, 2003–7 and 2008–13) in order to adjust for potential confounding from current age and calendar period with changes in the monthly allowance and admission requirements over time. In the analyses for basic school, the groups of attained age were <15, 15–19, 20–24, 25–29 and ≥30 years. Models were fitted with and without adjustment for parental education, income and psychiatric illness at the time of the child’s birth. To determine whether associations between parental death and educational attainment were moderated by which parent was lost, cause of death (suicide or other) and age at time of the loss (0–5, 6–11, 12–17 years), we fitted models that allowed interaction between exposure and each of the three factors, providing parameter estimates in each group. P values for no moderation of the effect measure were calculated in likelihood ratio tests. Results The crude estimated proportions of bereaved and non-bereaved people who completed each level of education showed that those who had been bereaved versus those who had not been bereaved differed only slightly in completing basic school (Fig. 1). For the other educational levels, however, bereaved men and women had consistently lower rates of attainment than non-bereaved people. The rates of attainment dropped as the level of education increased in both the bereaved and non-bereaved populations, with men having overall lower rates of attainment than women. Fig. 1 View largeDownload slide Proportions of boys and girls who complete each educational level by age and bereavement status in the study population. Fig. 1 View largeDownload slide Proportions of boys and girls who complete each educational level by age and bereavement status in the study population. RRs for attainment of each educational level are presented by gender in Tables 1–4. In comparison with non-bereaved people, those who had been bereaved were less likely to finish compulsory school (RR = 0.95, 95% CI: 0.93–0.97 for men; RR = 0.96, 95% CI: 0.94–0.98 for women), high school or vocational training (RR = 0.78, 95% CI: 0.76–0.80 for men; RR = 0.82, 95% CI: 0.80–0.84 for women), bachelor degree or professional programme (RR = 0.74, 95% CI: 0.70–0.79 for men; RR = 0.83, 95% CI: 0.79–0.86 for women) and university graduate degree (RR = 0.77, 95% CI: 0.68–0.86 for men; RR = 0.77, 95% CI: 0.69–0.86 for women). At all levels, neither the gender of the deceased parent, cause of death nor the child’s age at time of the death were found to modify our results. The only exception was completion of a bachelor degree or professional programme by women, wherein those who experienced the loss of a parent at or below the age of 5 years or between the ages of 13–17 years fared worse than those who experienced the loss between the ages of 6 and 12 years (P = 0.04). Table 1 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining basic school according to bereavement status (N = 1 043 813)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 1 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining basic school according to bereavement status (N = 1 043 813)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 2 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining high school/vocational training according to bereavement status (N = 846 299)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 2 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining high school/vocational training according to bereavement status (N = 846 299)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 3 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining bachelor/professional programmes according to bereavement status (N = 710 549)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 3 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining bachelor/professional programmes according to bereavement status (N = 710 549)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 4 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining graduate university (N = 564 485)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 4 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining graduate university (N = 564 485)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Discussion Main findings of this study This is the first study is to investigate the association between early parental death and attaining a specific level of education across the educational spectrum, from basic school and into university graduate studies. We found that early parental death may have long-term implications for educational attainment, regardless of the level of education, which parent died, the cause of death or the child’s age at the time. We also found that as the level of education became higher, the differences in attainment between the bereaved and non-bereaved increased in both genders, with up to a 26% lowered rate of attaining a bachelor or professional degree among bereaved males when compared to non-bereaved males. These results remained even after adjustment for potential confounders including family socioeconomic status and psychiatric illness in the family. Our results emphasizes the importance of studying education in a life-course perspective, as previous studies only studied specific fragmented aspects of education in adult life. What is already known Parental death was found previously to affect certain educational outcomes largely through family income and socioeconomic status,7,15 and a study in Sweden found that differences in mean grades in compulsory basic school disappeared after adjustment for economic and psychosocial factors.3 Little is known about the potential moderating effects of parent gender, cause of death and child’s age at time of death on educational outcomes, although some previous studies have shown that maternal death has a greater effect on cognitive skills and socio-emotional development than paternal death, which may be expected to negatively affect educational attainment.6,33 What this study adds While the finding from the above-mentioned Swedish study may be comparable to our finding of only a 4–5% difference in completion of basic school between bereaved and non-bereaved people, our results on subsequent attainment of higher education by gender showed up to a 26% lowered rate among bereaved males. This indicates a possible accumulation of effects over time stemming from academic and social vulnerabilities arising after the loss, and points towards the need for interventions that are not just grief-based, but also school-based. Furthermore, we did not find a moderating effect of the gender of the deceased parent on educational attainment. Our findings may indicate equalization of parental roles in younger generations, with fathers and mothers playing equally important roles in nurturing their child’s cognitive development. It is unclear why the cause of death and the child’s age at time of the death did not appear to moderate educational attainment, as suggested previously. We can only conclude that experiencing the death of a parent has long-term effects on educational attainment, regardless of how and when the loss occurred. Since at least one child out of every 20 will experience parental death and the link between educational attainment and health and well-being is well-established,1,34 we cannot ignore the potential impact of early parental death on educational attainment. Interventions need to target the child-related, parent-related and/or school-related mechanisms involved.35 Child-related mechanisms include the impact of grief on the child’s cognitive capability, psychosocial well-being, and the potential exacerbation of behavioural problems.22,36 Just as important is the role of the parent in nurturing the development of their children, as posited by social cognitive theory and attachment theory.37,38 When one parent dies, the surviving parent may be unable to compensate for the loss of parental support and supervision because of extra economic and household responsibilities.39 This may affect the child’s development and the parent’s involvement in school activities, which are both important predictors of schooling outcomes.40 In fact, findings from the Family Bereavement Programme, a 12-session family-based grief intervention, have shown that improved parenting skills mediated the effect of the intervention on academic expectations and GPA.41 Finally, support at schools for bereaved children need co-ordination, especially in the long-term, when many bereaved children and adolescents finally take in the reality of their loss and begin to struggle, but when teachers may have forgotten about the death.2,35 The present study is methodologically robust, as it drew on a large, virtually complete national population and had a longitudinal design. The use of registry-based data that had been collected independently of the study hypotheses minimized the selection bias inherent in self-reported data and allowed us to adjust for important confounders such as parental income, educational level and psychiatric illness. We also discriminated among sub-groups that have been suggested previously to be especially vulnerable by distinguishing between parent gender and causes of death, and combined it with a developmental approach by taking child’s age at the death into account. Limitations Potential limitations include possible residual confounding, e.g. from parental education and psychiatric illness, which was obtained only from the year before the birth of the child. Also, due to our study design, which aimed only at establishing whether there was an association between parental death and attaining a specific level of education across time, we are unable to specify whether children do not attain a particular level of education because they did not start it. For a finer-grained perspective, future studies may build on our results by investigating educational attainment only among those who start a specific level. We are also cautious about the generalizability of our results to populations who live in societies that may differ greatly from that of industrialized countries like Denmark. Conclusion The importance of education for health and opportunities cannot be underestimated, as it equips individuals with the skills and knowledge necessary to thrive and contribute to society. Although the Danish educational system supports equal opportunity regardless of financial background, we found that early parental death has deleterious consequences for educational attainment, which may initiate processes that affect life-long well-being. Thus, support for bereaved children may be more important than currently recognized. Scholarships might be valuable in buffering the financial impact of parental death in non-Nordic countries, while interventions that include extra supervision at school, family counselling and mentoring programmes for bereaved students might be developed and evaluated in future studies. Funding This work was supported by the Danish Cancer Society (Kræftens Bekæmpelse), a non-profit patient organization. References 1 Zimmerman EB, Woolf SH, Haley A. Understanding the relationship between education and health. In: Kaplan R, Spittel M, David D (eds). Population Health: Behavioral and Social Science Insights . Rockville, MD: Healthcare Research and Quality and Office of Behavioral and Social Sciences Research, National Institutes of Health, 2015. 2 Akerman R, Statham J. Bereavement in Childhood: The Impact on Psychological and Educational Outcomes and the Effectiveness of Support Services . London: Childhood Wellbeing Research Centre, 2014. 3 Berg L, Rostila M, Saarela J et al.  . Parental death during childhood and subsequent school performance. Pediatrics  2014; 133( 4): 682– 9. 4 Woolf SH, Braveman P. Where health disparities begin: the role of social and economic determinants—and why current policies may make matters worse. Health Aff  2011; 30( 10): 1852– 9. 5 Manor O, Matthews S, Power C. Health selection: the role of inter- and intra-generational mobility on social inequalities in health. Soc Sci Med  2003; 57( 11): 2217– 27. 6 Chen SH, Chen Y-C, Liu J-T. The impact of unexpected maternal death on education: first evidence from three national administrative data links. Am Econ Rev  2009; 99( 2): 149– 53. 7 Case A, Ardington C. The impact of parental death on school outcomes: longitudinal evidence from South Africa. Demography  2006; 43( 3): 401– 20. 8 Ainsworth M, Beegle K, Koda G. The impact of adult mortality and parental deaths on primary schooling in North-Western Tanzania. J Dev Stud  2005; 41( 3): 412– 39. 9 Abdelnoor A, Hollins S. The effect of childhood bereavement on secondary school performance. Educ Psychol Pract  2004; 20( 1): 43– 54. 10 Gertler P, Levine DI, Ames M. Schooling and parental death. Rev Econ Stat  2004; 86( 1): 211– 25. 11 Steele F, Sigle-Rushton W, Kravdal Ø. Consequences of family disruption on children’s educational outcomes in Norway. Demography  2009; 46( 3): 553– 74. 12 Fronstin P, Greenberg DH, Robins PK. Parental disruption and the labour market performance of children when they reach adulthood. J Popul Econ  2001; 14( 1): 137– 72. 13 Webb NB. The child and death. In: Webb NB (ed). Helping Bereaved Children: A Handbook for Practitioners , 3rd edn. New York: Guilford Press, 2010: 3– 21. 14 Kaplow JB, Howell KH, Layne CM. Do circumstances of the death matter? Identifying socioenvironmental risks for grief-related psychopathology in bereaved youth. J Trauma Stress  2014; 27( 1): 42– 9. 15 Swallen KC, Haas SA. Early Parental Death: Parental SES and Achieved SES at Midlife . Madison, WI: Center for Demography and Ecology, University of Wisconsin-Madison, 2002. 16 Pedersen CB, Gøtzsche H, Møller JØ et al.  . The Danish civil registration system. A cohort of eight million persons. Dan Med Bull  2006; 53: 441– 9. 17 Jensen VM, Rasmussen AW. Danish education registers. Scand J Public Health  2011; 39( 7 Suppl): 91– 4. 18 Blenstrup LT, Knudsen LB. Danish registers on aspects of reproduction. Scand J Public Health  2011; 39( 7 Suppl): 79– 82. 19 Helweg-Larsen K. The Danish register of causes of death. Scand J Public Health  2011; 39( 7 Suppl): 26– 9. 20 World Health Organization. International Statistical Classification of Diseases and Related Health Problems . Geneva: World Health Organization, 2010. 21 Li J, Vestergaard M, Cnattingius S et al.  . Mortality after parental death in childhood: a nationwide cohort study from three Nordic countries. PLoS Med  2014; 11( 7): e1001679. 22 Appel CW, Johansen C, Christensen J et al.  . Risk of use of antidepressants among children and young adults exposed to the death of a parent. Epidemiology  2016; 27( 4): 578– 85. 23 Danish Agency for Higher Education. SU—The Danish Students’ Grants and Loans Scheme. 2015; http://www.su.dk/english/state-educational-grant-and-loan-scheme-su/. 24 Ministry for Children EaGE. Overview of the Danish Education System. 2015; http://eng.uvm.dk/Education/Overview-of-the-Danish-Education-System?allowCookies=on. (11 November 2015, date last accessed). 25 OECD/Eurostat/UNESCO Institute for Statistics. ISCED 2011 Operational Manual . European Union: OECD Publishing, 2015. DOI:http://dx.doi.org/10.1787/9789264228368-en 26 Eccles JS. Influences of parents’ education on their children’s educational attainments: the role of parent and child perceptions. London Rev Educ  2005; 3( 3): 191– 204. 27 Baadsgaard M, Quitzau J. Danish registers on personal income and transfer payments. Scand J Public Health  2011; 39( 7 Suppl): 103– 5. 28 Kildemoes HW, Sorensen HT, Hallas J. The Danish national prescription registry. Scand J Public Health  2011; 39( 7 Suppl): 38– 41. 29 Mors O, Perto GP, Mortensen PB. The Danish psychiatric central research register. Scand J Public Health  2011; 39( 7 Suppl): 54– 7. 30 WHO Collaborating Centre for Drug Statistics Methodology. International Language for Drug Utilization Research. 2016; http://www.whocc.no/. (10 November 2016, date last accessed). 31 Aalen OO, Johansen S. An empirical transition matrix for non-homogeneous Markov chains based on censored observations. Scand J Stat  1978; 5( 3): 141– 50. 32 Rosthoj S, Andersen PK, Abildstrom SZ. SAS macros for estimation of the cumulative incidence functions based on a Cox regression model for competing risks survival data. Comput Methods Programs Biomed  2004; 74( 1): 69– 75. 33 Adda J, Björklund A, Holmlund H. The Role of Mothers and Fathers in Providing Skills: Evidence From Parental Deaths . Florence: European University Institute, 2011. 34 Elovainio M, Rosenström T, Hakulinen C et al.  . Educational attainment and health transitions over the life course: testing the potential mechanisms. J Public Health  2016; 38( 3): e254– 62. 35 Dyregrov A. Educational consequences of loss and trauma. Educ Child Psychol  2004; 21: 77– 84. 36 Hamdan S, Melhem NM, Porta G et al.  . Alcohol and substance abuse in parentally bereaved youth. J Clin Psychiatry  2013; 74( 8): 828– 33. 37 Bowlby J. Loss: Sadness and Depression . New York: Basic Books, 1980. 38 Bandura A. Social Foundations of Thought and Action, a Social Cognitive Theory . Englewood Cliffs, NJ: Prentice-Hall, 1986. 39 Valle G, Tillman KH. Childhood family structure and romantic relationships during the transition to adulthood. J Fam Issues  2014; 35( 1): 97– 124. 40 Spera C. A review of the relationship among parenting practices, parenting styles, and adolescent school achievement. Educ Psychol Rev  2005; 17( 2): 125– 46. 41 Schoenfelder EN, Tein JY, Wolchik S et al.  . Effects of the Family Bereavement Program on academic outcomes, educational expectations and job aspirations 6 years later: the mediating role of parenting and youth mental health problems. J Abnorm Child Psychol  2015; 43( 2): 229– 41. © The Author(s) 2018. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Public Health Oxford University Press

Does losing a parent early influence the education you obtain? A nationwide cohort study in Denmark

Loading next page...
 
/lp/ou_press/does-losing-a-parent-early-influence-the-education-you-obtain-a-uk0wSpCCMv
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com
ISSN
1741-3842
eISSN
1741-3850
D.O.I.
10.1093/pubmed/fdy070
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Health inequalities are rooted in education and we investigate the association between early parental death and attainment across the educational spectrum. Methods Using total population data on Danes born between 1982 and 2000 (n = 1 043 813), we assess incidence rate ratios (RRs) by gender for attainment of each educational level (basic school, high school or vocational training, bachelor degree or professional programme, and university graduate degree) according to loss of a parent before the age of 18 years. We adjust for family income, education and psychiatric illness and examine parent’s gender, cause of death and child’s age at time of death as potential moderators. Results Bereaved people had significantly lower attainment rates than non-bereaved people: basic school (RR = 0.95; 95% CI: 0.93–0.97 for men and 0.96; 0.94–0.98 for women), high school or vocational training (0.78; 0.76–0.80 for men and 0.82; 0.80–0.84 for women), bachelor degree or professional programme (0.74; 0.70–0.79 for men and 0.83; 0.79–0.86 for women) and university graduate degree (0.77; 0.68–0.86 for men and 0.77; 0.69–0.86 for women). Parent’s gender, cause of death and child’s age at the death did not modify the associations. Conclusions As education impacts population health, support for bereaved school children may be more important than realized. children, education, employment and skills, epidemiology Introduction Education has a profound impact on population health, as success in the classroom predicts income and health behaviours, which subsequently influence not only physical and psychological morbidity but also, ultimately, mortality.1 An estimated 5% of children in Western countries experience parental death before they reach adulthood2 and research has shown that early parental death could have deleterious consequences for educational outcomes. For example, a 2014 cohort study in Sweden of 772 117 people showed that those who had been bereaved as children had up to twice the odds of failing basic school as those who had not.3 This problem warrants attention because of the strong link between education and health. Disparities in vulnerability to disease also have social consequences, which create health inequalities that can be passed on to the next generation.4,5 While education comprises many levels, most studies have not taken a life-course perspective and have focused only on fragmented aspects of education, with outcomes ranging from exam grades to enrolment in school.3,6–10 Furthermore, some studies addressed only paternal deaths,11,12 or were cross-sectional or based on self-reported survey data,7–10 which may potentially be biased. The roles of factors such as the deceased parent’s gender, cause of death and the child’s age at the time of the death are also not well studied. This knowledge is important because it can help identify potentially vulnerable sub-groups of children. For example, younger children who experience parental death may be more vulnerable than older children because of their early stage of cognitive development and limited capacity to cope with grief.13 Children who have lost to suicide may also be more severely affected because of the social stigma that often follows, which may complicate the grieving process and affect the support available to the grieving child from surrounding social environment.14 Finally, many studies were conducted in countries where education requires financial resources7,8,12 and the financial effect of parental death might obscure the study of other factors that potentially link the loss to educational outcomes. In summary, there is a gap in the literature regarding the association between early parental death and long-term educational attainment, with consistent measures at each level over time, to facilitate comparison and understanding of this important area. In this study, we investigate the association between early parental death and attaining a specific level of education, from basic school to university graduate studies. The Danish educational system is publically funded up to university level, and since 1970, all students over the age of 18 years can apply for a monthly allowance, which minimizes the economic effect on educational attainment. This provides a unique opportunity to disentangle economic factors from other loss-related factors in investigating our objective. We studied males and females separately, as educational outcomes traditionally differ by gender,15 and investigated the parent’s gender, cause of death and the child’s age at time of the death as potential moderators. In the context of free education, we hypothesized moderate effects of early parental death on attainment, where the role of the identified moderators will be especially apparent. Methods Study population We conducted a nationwide registry-based cohort study. All residents of Denmark have been recorded in the Central Population Register (CPR) since 1 April 196816 with a unique personal identification number that ensures accurate linkage of information among all national registries. As the Danish Education Registry was first started in 1981,17 we defined a study population that comprised all people born in Denmark between 1 January 1982 and 31 December 2000 whose parents could be identified at the time of their birth. From the CPR, we extracted information on gender, dates of birth, migration and death and family relationships. Parent–child links were identified through the Danish Fertility Database,18 which has complete information on the parents of all children born after 1960. We identified 1 238 740 children and their parents. We followed only those who were alive and living in Denmark and for each level of analysis, the following numbers were eligible: basic school, n = 1 133 356; high school or vocational training, n = 919 595; bachelor degree or professional programme, n = 771 166; and university graduate degree, n = 612 589. Due to missing values for parental variables, we had to exclude the following numbers of children: 89 543 (basic school), 73 296 (high school or vocational training), 60 617 (bachelor degree or professional programme) and 48 104 (university graduate degree). Thus, in total, we followed 1 043 813 people for attained compulsory school, 846 299 for attained high school or vocational training, 710 549 for attained bachelor degree or professional programme and 564 485 for attained university graduate degree. The study protocol was approved by the Danish Data Protection Agency (J.nr. 2009-41-3506). Ethics approval was not required, as the study was based on anonymised registers, and we did not use human blood or tissue samples. Parental death We obtained the causes of death of parents from the Danish Register of Cause of Death, which contains information on all deaths since 1970.19 We categorized cause of death by the World Health Organization’s International Classification of Diseases (ICD-10) codes for suicide (X60-X84, Y870) and for all other causes.20 We chose to focus on suicide because suicide have been shown to be distinct from other causes of death14 and may indicate a family who is psychosocially vulnerable pre-death and has been linked to worse health and well-being outcomes.21,22 Educational attainment Education in Denmark is free, except for books and certain administrative fees at tertiary level, and the Danish student grants and loans scheme allows all students over the age of 18 to apply for a monthly allowance.23 The scheme was started in 1970, as part of a socio-democratic policy to ensure equal access to education regardless of financial background. The highest pre-tax monthly allowance is currently DKK 5941 (approximately GBP 680) for tertiary-level students who do not live at home.23 We linked information on the study population and their parents to the Danish Education Registry, which contains individual information about all educational programmes in the country since 1981.17 We obtained information on type and level of education, whether it was completed and the year of completion. Parental education, which may be a confounder, was defined as the highest educational level of either parent and was obtained for the year before the birth of the child. The Danish educational system comprises 10 years of compulsory basic school (primary and lower secondary education for children aged 6–15 years), upper secondary education (high school or vocational or technical training) and tertiary education (consisting of 3–3.5-year bachelor degree or professional programmes and graduate Masters and PhD university programmes).24 We grouped educational level into four categories corresponding to the International Standard Classification of Education (ISCED-2011) codes: basic school (codes 1–2), high school or vocational training (3–4), bachelor degree or professional programme (5–6) and university graduate degree (7–8).25 Covariates Parental education, income and psychiatric illness were identified as potential confounders as family socioeconomic and psychiatric factors have been linked to both early parental death and child educational outcomes.22,26 Information on parental education, income and psychiatric illness was obtained for the year before the birth of the child from the Danish Education Register,17 Danish Income Statistics Register,27 the National Prescription Registry28 and the Central Psychiatric Register.29 We used the highest attained education level of either parent. Disposable household income per person was categorized as low (first quartile), middle (second and third quantiles) or high (fourth quartile), calculated in Danish kroner after taxation and adjusted for the number of people in the household and for inflation with the year 2000 as reference. Psychiatric illness was defined as first psychiatric hospitalization, outpatient treatment or use of antidepressants by either parent. We obtained information on all psycho-pharmaceutical prescriptions corresponding to Anatomical Therapeutic Chemical (ATC) code N06A for antidepressants30 and on hospital admissions for diagnoses related to all psychiatric disorders, court-ordered psychiatric treatment, substance use and suicide attempts. Information on education and income was available from 1981, on psychiatric hospitalization from 1969, and on antidepressant use and outpatient psychiatric treatment only from 1995. Statistical analyses For each educational level, we calculated person-years at risk from the year before the minimum possible age of attaining that specific level: age of 13 years for compulsory basic school, 16 years for high school or vocational training, 18 years for bachelor degree or professional programme and 20 years for university graduate degree. Exposure was defined as parental death before age 18 or age at entry, whichever came first. All people were followed until the first year in which they attained the investigated level of education, death, emigration or 31 December 2013, which was the end of the study period. The absolute level of educational attainment for all four levels in the study population and the cumulative probability of attaining the associated level before a given age was estimated non-parametrically according to gender and exposure status, taking into account competing risk from child’s death using the Aalen Johansen estimator31 calculated with Cuminc SAS macro.32 Exposure was defined as parental death before age 18 or age at entry, whichever came first. Incidence rate ratios (RRs) for educational attainment according to early parental death with 95% confidence intervals (CIs) were estimated in multiplicative Poisson models. We performed separate analyses for men and women. Person-time was stratified according to exposure status into two groups (parental death and no parental death), attained age into six groups (<20, 20–24, 25–29 and ≥30 years) and calendar periods into five groups (1988–92, 1993–97, 1998–2002, 2003–7 and 2008–13) in order to adjust for potential confounding from current age and calendar period with changes in the monthly allowance and admission requirements over time. In the analyses for basic school, the groups of attained age were <15, 15–19, 20–24, 25–29 and ≥30 years. Models were fitted with and without adjustment for parental education, income and psychiatric illness at the time of the child’s birth. To determine whether associations between parental death and educational attainment were moderated by which parent was lost, cause of death (suicide or other) and age at time of the loss (0–5, 6–11, 12–17 years), we fitted models that allowed interaction between exposure and each of the three factors, providing parameter estimates in each group. P values for no moderation of the effect measure were calculated in likelihood ratio tests. Results The crude estimated proportions of bereaved and non-bereaved people who completed each level of education showed that those who had been bereaved versus those who had not been bereaved differed only slightly in completing basic school (Fig. 1). For the other educational levels, however, bereaved men and women had consistently lower rates of attainment than non-bereaved people. The rates of attainment dropped as the level of education increased in both the bereaved and non-bereaved populations, with men having overall lower rates of attainment than women. Fig. 1 View largeDownload slide Proportions of boys and girls who complete each educational level by age and bereavement status in the study population. Fig. 1 View largeDownload slide Proportions of boys and girls who complete each educational level by age and bereavement status in the study population. RRs for attainment of each educational level are presented by gender in Tables 1–4. In comparison with non-bereaved people, those who had been bereaved were less likely to finish compulsory school (RR = 0.95, 95% CI: 0.93–0.97 for men; RR = 0.96, 95% CI: 0.94–0.98 for women), high school or vocational training (RR = 0.78, 95% CI: 0.76–0.80 for men; RR = 0.82, 95% CI: 0.80–0.84 for women), bachelor degree or professional programme (RR = 0.74, 95% CI: 0.70–0.79 for men; RR = 0.83, 95% CI: 0.79–0.86 for women) and university graduate degree (RR = 0.77, 95% CI: 0.68–0.86 for men; RR = 0.77, 95% CI: 0.69–0.86 for women). At all levels, neither the gender of the deceased parent, cause of death nor the child’s age at time of the death were found to modify our results. The only exception was completion of a bachelor degree or professional programme by women, wherein those who experienced the loss of a parent at or below the age of 5 years or between the ages of 13–17 years fared worse than those who experienced the loss between the ages of 6 and 12 years (P = 0.04). Table 1 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining basic school according to bereavement status (N = 1 043 813)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 1 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining basic school according to bereavement status (N = 1 043 813)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  10 907  0.94  0.92–0.96  0.95  0.93–0.97    10 792  0.96  0.94–0.97  0.96  0.94–0.98     Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Cause of parent’s death   Suicide  1281  0.94  0.89–1.00  0.96  0.90–1.00  0.84  1277  0.94  0.89–0.99  0.94  0.89–0.99  0.42   Other  9626  0.94  0.92–0.96  0.95  0.93–0.97  9515  0.96  0.94–0.98  0.96  0.94–0.98   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Parent who died   Father  7656  0.94  0.91–0.96  0.95  0.93–0.97  1.00  7681  0.95  0.93–0.97  0.96  0.94–0.98  0.95   Mother  3251  0.94  0.91–0.98  0.95  0.92–0.98  3111  0.96  0.93–1.00  0.96  0.93–1.00   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  3668  0.93  0.90–0.96  0.94  0.91–0.97  0.57  3559  0.96  0.93–1.00  0.97  0.94–1.00  0.44   6–12 years  7239  0.94  0.92–0.96  0.95  0.93–0.98  7233  0.95  0.93–0.97  0.96  0.93–0.08   13–17 years  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A  N/A   Not bereaved  468 025  1.00  –  1.00  –  454 419  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 2 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining high school/vocational training according to bereavement status (N = 846 299)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 2 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining high school/vocational training according to bereavement status (N = 846 299)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  5649  0.70  0.68–0.72  0.78  0.76–0.80    5970  0.75  0.73–0.77  0.82  0.80–0.84     Not bereaved  194 031  1.00  –  1.00  –    190 904  1.00  –  1.00  –    Cause of parent’s death   Suicide  611  0.71  0.65–0.77  0.79  0.73–0.86  0.59  646  0.75  0.69–0.81  0.82  0.76–0.88  0.93   Other  5038  0.70  0.68–0.72  0.78  0.75–0.80  5324  0.75  0.73–0.77  0.82  0.80–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Parent who died   Father  3914  0.69  0.67–0.71  0.78  0.76–0.80  0.82  4210  0.74  0.72–0.76  0.83  0.80–0.85  0.21   Mother  1735  0.73  0.70–0.76  0.77  0.74–0.81  1760  0.76  0.73–0.80  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  1288  0.68  0.64–0.72  0.77  0.72–0.81  0.67  1339  0.71  0.67–0.74  0.80  0.76–0.84  0.14   6–12 years  2654  0.71  0.68–0.74  0.79  0.76–0.82  2885  0.77  0.74–0.80  0.84  0.81–0.87   13–17 years  1707  0.71  0.68–0.75  0.77  0.74–0.81  1746  0.75  0.72–0.79  0.80  0.76–0.84   Not bereaved  194 031  1.00  –  1.00  –  190 904  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 3 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining bachelor/professional programmes according to bereavement status (N = 710 549)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 3 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining bachelor/professional programmes according to bereavement status (N = 710 549)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  1271  0.68  0.64–0.72  0.74  0.70–0.79    1983  0.76  0.73–0.80  0.83  0.79–0.86     Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Cause of parent’s death   Suicide  142  0.76  0.64–0.90  0.86  0.73–1.01  0.07  215  0.78  0.68–0.89  0.82  0.72–0.94  0.95   Other  1129  0.67  0.63–0.71  0.73  0.69–0.78    1768  0.76  0.73–0.80  0.83  0.79–0.87   Not bereaved  39 334  1.00  –  1.00  –    53 791  1.00  –  1.00  –  Parent who died   Father  849  0.64  0.60–0.69  0.72  0.67–0.77  0.09  1349  0.73  0.69–0.77  0.81  0.77–0.85  0.15   Mother  422  0.76  0.69–0.83  0.80  0.72–0.88  634  0.84  0.78–0.91  0.87  0.80–0.94   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  Age at parent’s death (years)   ≤ 5 years  229  0.66  0.58–0.76  0.74  0.65–0.85  0.99  351  0.72  0.64–0.79  0.79  0.71–0.88  0.04   6–12 years  471  0.67  0.61–0.73  0.74  0.67–0.81  808  0.81  0.75–0.87  0.89  0.83–0.95   13–17 years  571  0.69  0.63–0.75  0.75  0.69–0.81  824  0.74  0.69–0.80  0.79  0.74–0.85   Not bereaved  39 334  1.00  –  1.00  –  53 791  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 4 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining graduate university (N = 564 485)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Table 4 Rate ratios (RRs) and their 95% confidence intervals (CIs) for attaining graduate university (N = 564 485)   Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –    Men  P  Women  P  N attained  RRa  CI  RRb  CI  N attained  RRa  CI  RRb  CI  Attained   Bereaved  312  0.70  0.62–0.78  0.77  0.68–0.86    329  0.71  0.64–0.80  0.77  0.69–0.86     Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Cause of parent’s death   Suicide  38  0.83  0.61–1.14  0.93  0.68–1.28  0.22  45  0.88  0.66–1.18  0.89  0.66–1.19  0.31   Other  274  0.68  0.61–0.77  0.75  0.66–0.84  284  0.69  0.62–0.78  0.75  0.67–0.85   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Parent who died   Father  205  0.66  0.57–0.76  0.74  0.74–0.64  0.32  220  0.68  0.60–0.78  0.77  0.67–0.88  0.91   Mother  107  0.79  0.66–0.96  0.83  0.69–1.00  109  0.78  0.65–0.95  0.78  0.64–0.94   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  Age at parent’s death (years)   ≤5 years  56  0.69  0.53–0.89  0.76  0.59–0.99  0.79  70  0.81  0.64–1.02  0.88  0.70–1.11  0.43   6–12 years  111  0.67  0.55–0.80  0.73  0.61–0.88  125  0.70  0.59–0.84  0.76  0.64–0.91   13–17 years  145  0.73  0.62–0.86  0.80  0.68–0.94  134  0.68  0.58–0.81  0.73  0.61–0.86   Not bereaved  9577  1.00  –  1.00  –  9897  1.00  –  1.00  –  aAdjusted for age and calendar year. bFurther adjusted for family income, education and psychiatric illness (at birth). Discussion Main findings of this study This is the first study is to investigate the association between early parental death and attaining a specific level of education across the educational spectrum, from basic school and into university graduate studies. We found that early parental death may have long-term implications for educational attainment, regardless of the level of education, which parent died, the cause of death or the child’s age at the time. We also found that as the level of education became higher, the differences in attainment between the bereaved and non-bereaved increased in both genders, with up to a 26% lowered rate of attaining a bachelor or professional degree among bereaved males when compared to non-bereaved males. These results remained even after adjustment for potential confounders including family socioeconomic status and psychiatric illness in the family. Our results emphasizes the importance of studying education in a life-course perspective, as previous studies only studied specific fragmented aspects of education in adult life. What is already known Parental death was found previously to affect certain educational outcomes largely through family income and socioeconomic status,7,15 and a study in Sweden found that differences in mean grades in compulsory basic school disappeared after adjustment for economic and psychosocial factors.3 Little is known about the potential moderating effects of parent gender, cause of death and child’s age at time of death on educational outcomes, although some previous studies have shown that maternal death has a greater effect on cognitive skills and socio-emotional development than paternal death, which may be expected to negatively affect educational attainment.6,33 What this study adds While the finding from the above-mentioned Swedish study may be comparable to our finding of only a 4–5% difference in completion of basic school between bereaved and non-bereaved people, our results on subsequent attainment of higher education by gender showed up to a 26% lowered rate among bereaved males. This indicates a possible accumulation of effects over time stemming from academic and social vulnerabilities arising after the loss, and points towards the need for interventions that are not just grief-based, but also school-based. Furthermore, we did not find a moderating effect of the gender of the deceased parent on educational attainment. Our findings may indicate equalization of parental roles in younger generations, with fathers and mothers playing equally important roles in nurturing their child’s cognitive development. It is unclear why the cause of death and the child’s age at time of the death did not appear to moderate educational attainment, as suggested previously. We can only conclude that experiencing the death of a parent has long-term effects on educational attainment, regardless of how and when the loss occurred. Since at least one child out of every 20 will experience parental death and the link between educational attainment and health and well-being is well-established,1,34 we cannot ignore the potential impact of early parental death on educational attainment. Interventions need to target the child-related, parent-related and/or school-related mechanisms involved.35 Child-related mechanisms include the impact of grief on the child’s cognitive capability, psychosocial well-being, and the potential exacerbation of behavioural problems.22,36 Just as important is the role of the parent in nurturing the development of their children, as posited by social cognitive theory and attachment theory.37,38 When one parent dies, the surviving parent may be unable to compensate for the loss of parental support and supervision because of extra economic and household responsibilities.39 This may affect the child’s development and the parent’s involvement in school activities, which are both important predictors of schooling outcomes.40 In fact, findings from the Family Bereavement Programme, a 12-session family-based grief intervention, have shown that improved parenting skills mediated the effect of the intervention on academic expectations and GPA.41 Finally, support at schools for bereaved children need co-ordination, especially in the long-term, when many bereaved children and adolescents finally take in the reality of their loss and begin to struggle, but when teachers may have forgotten about the death.2,35 The present study is methodologically robust, as it drew on a large, virtually complete national population and had a longitudinal design. The use of registry-based data that had been collected independently of the study hypotheses minimized the selection bias inherent in self-reported data and allowed us to adjust for important confounders such as parental income, educational level and psychiatric illness. We also discriminated among sub-groups that have been suggested previously to be especially vulnerable by distinguishing between parent gender and causes of death, and combined it with a developmental approach by taking child’s age at the death into account. Limitations Potential limitations include possible residual confounding, e.g. from parental education and psychiatric illness, which was obtained only from the year before the birth of the child. Also, due to our study design, which aimed only at establishing whether there was an association between parental death and attaining a specific level of education across time, we are unable to specify whether children do not attain a particular level of education because they did not start it. For a finer-grained perspective, future studies may build on our results by investigating educational attainment only among those who start a specific level. We are also cautious about the generalizability of our results to populations who live in societies that may differ greatly from that of industrialized countries like Denmark. Conclusion The importance of education for health and opportunities cannot be underestimated, as it equips individuals with the skills and knowledge necessary to thrive and contribute to society. Although the Danish educational system supports equal opportunity regardless of financial background, we found that early parental death has deleterious consequences for educational attainment, which may initiate processes that affect life-long well-being. Thus, support for bereaved children may be more important than currently recognized. Scholarships might be valuable in buffering the financial impact of parental death in non-Nordic countries, while interventions that include extra supervision at school, family counselling and mentoring programmes for bereaved students might be developed and evaluated in future studies. Funding This work was supported by the Danish Cancer Society (Kræftens Bekæmpelse), a non-profit patient organization. References 1 Zimmerman EB, Woolf SH, Haley A. Understanding the relationship between education and health. In: Kaplan R, Spittel M, David D (eds). Population Health: Behavioral and Social Science Insights . Rockville, MD: Healthcare Research and Quality and Office of Behavioral and Social Sciences Research, National Institutes of Health, 2015. 2 Akerman R, Statham J. Bereavement in Childhood: The Impact on Psychological and Educational Outcomes and the Effectiveness of Support Services . London: Childhood Wellbeing Research Centre, 2014. 3 Berg L, Rostila M, Saarela J et al.  . Parental death during childhood and subsequent school performance. Pediatrics  2014; 133( 4): 682– 9. 4 Woolf SH, Braveman P. Where health disparities begin: the role of social and economic determinants—and why current policies may make matters worse. Health Aff  2011; 30( 10): 1852– 9. 5 Manor O, Matthews S, Power C. Health selection: the role of inter- and intra-generational mobility on social inequalities in health. Soc Sci Med  2003; 57( 11): 2217– 27. 6 Chen SH, Chen Y-C, Liu J-T. The impact of unexpected maternal death on education: first evidence from three national administrative data links. Am Econ Rev  2009; 99( 2): 149– 53. 7 Case A, Ardington C. The impact of parental death on school outcomes: longitudinal evidence from South Africa. Demography  2006; 43( 3): 401– 20. 8 Ainsworth M, Beegle K, Koda G. The impact of adult mortality and parental deaths on primary schooling in North-Western Tanzania. J Dev Stud  2005; 41( 3): 412– 39. 9 Abdelnoor A, Hollins S. The effect of childhood bereavement on secondary school performance. Educ Psychol Pract  2004; 20( 1): 43– 54. 10 Gertler P, Levine DI, Ames M. Schooling and parental death. Rev Econ Stat  2004; 86( 1): 211– 25. 11 Steele F, Sigle-Rushton W, Kravdal Ø. Consequences of family disruption on children’s educational outcomes in Norway. Demography  2009; 46( 3): 553– 74. 12 Fronstin P, Greenberg DH, Robins PK. Parental disruption and the labour market performance of children when they reach adulthood. J Popul Econ  2001; 14( 1): 137– 72. 13 Webb NB. The child and death. In: Webb NB (ed). Helping Bereaved Children: A Handbook for Practitioners , 3rd edn. New York: Guilford Press, 2010: 3– 21. 14 Kaplow JB, Howell KH, Layne CM. Do circumstances of the death matter? Identifying socioenvironmental risks for grief-related psychopathology in bereaved youth. J Trauma Stress  2014; 27( 1): 42– 9. 15 Swallen KC, Haas SA. Early Parental Death: Parental SES and Achieved SES at Midlife . Madison, WI: Center for Demography and Ecology, University of Wisconsin-Madison, 2002. 16 Pedersen CB, Gøtzsche H, Møller JØ et al.  . The Danish civil registration system. A cohort of eight million persons. Dan Med Bull  2006; 53: 441– 9. 17 Jensen VM, Rasmussen AW. Danish education registers. Scand J Public Health  2011; 39( 7 Suppl): 91– 4. 18 Blenstrup LT, Knudsen LB. Danish registers on aspects of reproduction. Scand J Public Health  2011; 39( 7 Suppl): 79– 82. 19 Helweg-Larsen K. The Danish register of causes of death. Scand J Public Health  2011; 39( 7 Suppl): 26– 9. 20 World Health Organization. International Statistical Classification of Diseases and Related Health Problems . Geneva: World Health Organization, 2010. 21 Li J, Vestergaard M, Cnattingius S et al.  . Mortality after parental death in childhood: a nationwide cohort study from three Nordic countries. PLoS Med  2014; 11( 7): e1001679. 22 Appel CW, Johansen C, Christensen J et al.  . Risk of use of antidepressants among children and young adults exposed to the death of a parent. Epidemiology  2016; 27( 4): 578– 85. 23 Danish Agency for Higher Education. SU—The Danish Students’ Grants and Loans Scheme. 2015; http://www.su.dk/english/state-educational-grant-and-loan-scheme-su/. 24 Ministry for Children EaGE. Overview of the Danish Education System. 2015; http://eng.uvm.dk/Education/Overview-of-the-Danish-Education-System?allowCookies=on. (11 November 2015, date last accessed). 25 OECD/Eurostat/UNESCO Institute for Statistics. ISCED 2011 Operational Manual . European Union: OECD Publishing, 2015. DOI:http://dx.doi.org/10.1787/9789264228368-en 26 Eccles JS. Influences of parents’ education on their children’s educational attainments: the role of parent and child perceptions. London Rev Educ  2005; 3( 3): 191– 204. 27 Baadsgaard M, Quitzau J. Danish registers on personal income and transfer payments. Scand J Public Health  2011; 39( 7 Suppl): 103– 5. 28 Kildemoes HW, Sorensen HT, Hallas J. The Danish national prescription registry. Scand J Public Health  2011; 39( 7 Suppl): 38– 41. 29 Mors O, Perto GP, Mortensen PB. The Danish psychiatric central research register. Scand J Public Health  2011; 39( 7 Suppl): 54– 7. 30 WHO Collaborating Centre for Drug Statistics Methodology. International Language for Drug Utilization Research. 2016; http://www.whocc.no/. (10 November 2016, date last accessed). 31 Aalen OO, Johansen S. An empirical transition matrix for non-homogeneous Markov chains based on censored observations. Scand J Stat  1978; 5( 3): 141– 50. 32 Rosthoj S, Andersen PK, Abildstrom SZ. SAS macros for estimation of the cumulative incidence functions based on a Cox regression model for competing risks survival data. Comput Methods Programs Biomed  2004; 74( 1): 69– 75. 33 Adda J, Björklund A, Holmlund H. The Role of Mothers and Fathers in Providing Skills: Evidence From Parental Deaths . Florence: European University Institute, 2011. 34 Elovainio M, Rosenström T, Hakulinen C et al.  . Educational attainment and health transitions over the life course: testing the potential mechanisms. J Public Health  2016; 38( 3): e254– 62. 35 Dyregrov A. Educational consequences of loss and trauma. Educ Child Psychol  2004; 21: 77– 84. 36 Hamdan S, Melhem NM, Porta G et al.  . Alcohol and substance abuse in parentally bereaved youth. J Clin Psychiatry  2013; 74( 8): 828– 33. 37 Bowlby J. Loss: Sadness and Depression . New York: Basic Books, 1980. 38 Bandura A. Social Foundations of Thought and Action, a Social Cognitive Theory . Englewood Cliffs, NJ: Prentice-Hall, 1986. 39 Valle G, Tillman KH. Childhood family structure and romantic relationships during the transition to adulthood. J Fam Issues  2014; 35( 1): 97– 124. 40 Spera C. A review of the relationship among parenting practices, parenting styles, and adolescent school achievement. Educ Psychol Rev  2005; 17( 2): 125– 46. 41 Schoenfelder EN, Tein JY, Wolchik S et al.  . Effects of the Family Bereavement Program on academic outcomes, educational expectations and job aspirations 6 years later: the mediating role of parenting and youth mental health problems. J Abnorm Child Psychol  2015; 43( 2): 229– 41. © The Author(s) 2018. Published by Oxford University Press on behalf of Faculty of Public Health. All rights reserved. For permissions, please e-mail: journals.permissions@oup.com This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Public HealthOxford University Press

Published: Apr 19, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off