Childhood adverse life events and parental psychopathology as risk factors for bipolar disorder

Abstract

Childhood adverse events are risk factors for later bipolar disorder. We quantified the risks for a later diagnosis of bipolar disorder after exposure to adverse life events in children with and without parental psychopathology. This register-based population cohort study included all persons born in Denmark from 1980 to 1998 (980 554 persons). Adversities before age 15 years were: familial disruption; parental somatic illness; any parental psychopathology; parental labour market exclusion; parental imprisonment; placement in out-of-home care; and parental natural and unnatural death. We calculated risk estimates of each of these eight life events as single exposure and risk estimates for exposure to multiple life events. Main outcome variable was a diagnosis of bipolar disorder after the age of 15 years, analysed with Cox proportional hazard regression. Single exposure to most of the investigated adversities were associated with increased risk for bipolar disorder, exceptions were parental somatic illness and parental natural death. By far the strongest risk factor for bipolar disorder in our study was any mental disorder in the parent (hazard ratio 3.53; 95% confidence interval 2.73-4.53) and the additional effects of life events on bipolar risk were limited. An effect of early adverse life events on bipolar risk later in life was mainly observed in children without parental psychopathology. Our findings do not exclude early-life events as possible risk factors, but challenge the concept of adversities as important independent determinants of bipolar disorder in genetically vulnerable individuals.

Figure 1

Cumulative Incidence for Bipolar Disorder after age 15 years over Number of Early Adverse Events. The cumulative incidence at age 34 years for each life event were: no exposure 0.43 95% CI 0.39–0.47; Familial disruption 0.94 95% CI 0.85–1.03; Parental chronic somatic disorder 0.74 95% CI 0.64–0.85; Parental psychopathology 1.51 95% CI 1.31–1.73; Parental labour market affiliation 1.05 95% CI 0.89–1.23; Parental imprisonment 0.86 95% CI 0.72–1.03; Out-of-home care 1.47 95% CI 1.17–1.82; Parental loss to natural causes 0.44 95% CI 0.31–0.60; Parental loss to unnatural causes 1.23 95% CI 0.81–1.82. CI, confidence interval.

Figure 2

Bipolar risk after exposure to single and multiple early adverse life events. The blue bars show bipolar risk for each life event if it was combined with one or more other life events. The red bars indicate the independent risk of bipolar disorder for each of the eight life events. CI, confidence interval; HR, hazard ratio.