Elsevier

The Lancet

Volume 361, Issue 9355, 1 February 2003, Pages 363-367
The Lancet

Articles
Mortality in parents after death of a child in Denmark: a nationwide follow-up study

https://doi.org/10.1016/S0140-6736(03)12387-2Get rights and content

Summary

Background

Little is known about the effect of parental bereavement on physical health. We investigated whether the death of a child increased mortality in parents.

Methods

We undertook a follow-up study based on national registers. From 1980 to 1996, we enrolled 21 062 parents in Denmark who had a child who had died (exposed cohort), and 293 745 controls—ie, parents whose children were alive, and whose family structure matched that of the exposed cohort. Natural deaths were defined with ICD8 codes 0000–7969 and ICD10 codes A00–R99, and unnatural deaths with codes 8000–9999 and V01–Y98. We used Cox's proportional-hazards regression models to assess the mortality rate of parents up to 18 years after bereavement.

Findings

We observed an increased overall mortality rate in mothers whose child had died (hazards ratio 1·43, 95% CI 1·24–1·64; p<0·0001). An excess mortality from natural causes (1·44, 1·15–1·78; p<0·0001) was noted in mothers only during the 10th–18th year of follow-up. Mothers had increased mortality rates from unnatural causes throughout follow-up, with the highest rate recorded during the first 3 years (3·84, 2·48–5·88; p<0·0001). Bereaved fathers had only an early excess mortality from unnatural causes (1·57, 1·06–2·32; p=0·04). Mothers who lost a child due to an unnatural death or an unexpected death had a hazard ratio of 1·72 (1·38–2·15; p=0·0040) and 1·67 (1·37–2·03; p=0·0037), respectively.

Interpretation

The death of a child is associated with an overall increased mortality from both natural and unnatural causes in mothers, and an early increased mortality from unnatural causes in fathers.

Introduction

Bereavement is a severe life event1, 2, 3 that leads to increased mortality in widowed populations.3, 4, 5, 6, 7 The death of a child is especially stressful, even more so than the death of a spouse,2, 3, 8 but its effect on mortality of other family members is unclear.3, 4, 9, 10, 11 Results of most studies9, 10, 11 indicate that there is no overall association between death of a child and mortality in parents. However, an increased mortality among single mothers9 and a worse cancer survival in fathers have been reported.11

Stressful life events can affect the sympathetic nervous system, the hypothalamic-pituitary-adrenal axis, the neuroendocrine systems, and the immune systems, which could result in various diseases.12, 13 Psychological stress could also raise the risk of unnatural death.6, 14, 15 It might, furthermore, lead to more adverse lifestyle behaviours, which could in turn affect mortality.15, 16, 17, 18, 19

Our aim was to examine whether or not the death of a child increased mortality in parents. We postulated that risk of mortality would change dependent on duration of follow-up,3, 4, 5, 6, 7, 9 and that the effect would be modified by the nature of the death of the child and by sociodemographic factors of possible relevance for coping with stress.3, 20, 21

Section snippets

Participants and protocol

We did a follow-up study based on cohorts identified from national registers in Denmark. For each year, from 1980 to 1996, we identified all children who died at age younger than 18 years and their family members (exposed cohort) by searching the Register of Population Statistics.22 We recorded the number of parents and the number of children in each of these families, and grouped the children in accordance with their age on Jan 1 of the year that their sibling died (<1 year old, 1–2, 3–6, 7–9,

Results

Table 1 shows the characteristics of the 12 512 children who died between 1980 and 1996, minus 440 who could not be matched through the population registers with their parents at the beginning of the year that they died because the children did not live with their parents. Table 2 shows the baseline characteristics of the parents in the exposed (n=21 062) and unexposed (n=293 745) cohorts. These 314 807 individuals yielded 3 316 649 person-years, with a median follow-up time of 11·0 years each.

Discussion

Our data indicate that the death of a child is associated with an overall increased mortality in mothers, and a slightly increased early mortality from unnatural causes in fathers.

Psychological stress after the death of a child could affect health outcomes in several ways. The most intense reactions are often seen shortly after the bereavement, with a short-term peak of mortality from unnatural causes.3, 4, 5, 6, 7 In the long term, various pathophysiological changes related to stress could

References (29)

  • LeistikowBN et al.

    Smoking as a risk factor for injury death: a meta-analysis of cohort studies

    Prev Med

    (1998)
  • American Psychiatric Association

    DSM-III-R: diagnostic and statistical manual of mental disorders

    (1987)
  • LevavI et al.

    Israeli PERI life events scale: ratings of events by a community sample

    Isr J Med Sci

    (1981)
  • OsterweisM et al.

    Bereavement: reactions, consequences, and care

    (1984)
  • ReesWD et al.

    Mortality of bereavement

    BMJ

    (1967)
  • SchaeferC et al.

    Mortality following conjugal bereavement and the effects of a shared environment

    Am J Epidemiol

    (1995)
  • MartikainenP et al.

    Mortality after the death of a spouse: rates and causes of death in a large Finnish cohort

    Am J Public Health

    (1996)
  • LichtensteinP et al.

    A twin study of mortality after spousal bereavement

    Psychol Med

    (1998)
  • DyrgrovenA

    Parental reactions to the loss of an infant child: a review

    Scand J Psychol

    (1990)
  • LevavI et al.

    An epidemiologic study of mortality among bereaved parents

    N Engl J Med

    (1988)
  • JohansenC et al.

    Psychological stress, cancer incidence and mortality from non-malignant diseases

    Br J Cancer

    (1997)
  • LevavI et al.

    Cancer incidence and survival following bereavement

    Am J Public Health

    (2000)
  • McEwenBS

    Protective and damaging effects of the stress mediators

    N Engl J Med

    (1998)
  • GoodkinK et al.

    Physiological effects of bereavement and bereavement support group interventions

  • Cited by (380)

    • Palliative Care Family Support in Neonatology

      2023, Principles of Neonatology
    • What Are Special Considerations for Pediatric Palliative Care?

      2023, Evidence-Based Practice of Palliative Medicine, Second Edition
    View all citing articles on Scopus
    View full text