To Örebro University

Background and aims: Loss of a first-degree family member in childhood constitutes a major psychological stressor, and is associated both with subsequent psychiatric and somatic morbidity. The potential influence on type 1 diabetes risk has however not yet been fully elucidated. In this study we therefore aimed to investigate the impact of childhood bereavement on type 1 diabetes risk.

Materials and methods: We conducted a population-based study in Sweden, encompassing 2,321,318 children born in Sweden January 1 1990 to December 31 2012. The follow up ended December 31 2013, at death of the child, type 1 diabetes diagnosis, emigration or when the child turned 19 years. All children were followed from the age of one, with exposure defined as death of a mother, father, or sibling. Type 1 diabetes diagnoses were extracted from the National Patient Register. We applied Cox proportional hazards models with attained age as time scale and loss of family member as a time varying variable, adjusting for potential confounders including parental type 1 diabetes, parental country of birth, and region of residence. We further categorized child age at bereavement as pre-school (1-6 years), school age (7-12 years) and teenage (13-18 years).

Results: During follow-up (median 10.8 years), 50,253 (2.2%) children experienced loss of a family member. Median age at loss was 9.6 years, and 32% of all deaths were categorized as traumatic (accident, suicide, violence, or other sudden unnatural deaths). In total 10,965 children were diagnosed with type 1 diabetes during follow-up and median age at diagnosis was 8.5 years. We observed no overall association between childhood bereavement and type 1 diabetes risk (crude HR 1.00, 95% CI 0.86-1.18, adjusted HR 0.96, 95% CI 0.82 -1.13). The risk was not influenced by sex of the child, cause of death of family member, or familial relationship to the deceased. However, we noted an association when the exposure occurred during the teenage years (adjusted HR 1.67, 95% CI 1.15-2.43).

Conclusion: Overall, childhood bereavement was not associated with the risk of type 1 diabetes, but the impact of childhood loss on type 1 diabetes may be modified by age at bereavement.