Differential Vulnerability to Early-Life Parental Death: The Moderating Effects of Family Suicide History on Risks for Major Depression and Substance Abuse in Later Life

Abstract

Only a portion of those individuals exposed to parental death in early life (PDE) develop behavioral health disorders. We utilized demographic pedigree data from the Utah Population Database to test for differential vulnerability to PDE by creating a risk score of familial susceptibility to suicide (FS) at the population level. Using logistic panel regression models, we tested for multiplicative interactions between PDE and FS on the risks of major depressive disorder (MDD) and substance abuse (SA), measured using Medicare claims, after age 65. The final sample included 155,983 individuals (born 1886–1944), yielding 1,431,060 person-years at risk (1992–2009). Net of several potential confounders, including probability of survival to age 65, we found an FS × PDE interaction for females, in which PDE and FS as main effects had no impact but jointly increased MDD risk. No statistically significant main or interactive effects were found for SA among females or for either phenotype among males. Our findings are consistent with a differential vulnerability model for MDD in females, in which early-life stress increases the risk for poor behavioral health only among the vulnerable. Furthermore, we demonstrate how demographic and pedigree data might serve as tools for investigating differential vulnerability hypotheses.

Acknowledgments

We wish to thank the Huntsman Cancer Foundation for database support provided to the Pedigree and Population Resource of the HCI, University of Utah. We also thank Alison Fraser and Diana Lane Reed for assistance in managing the data; and Rebecca L. Utz and Ming Wen for their suggestions. A similar version of this work appeared previously as part of the primary author’s doctoral dissertation.

Funding

This work was supported by National Institutes of Health grant AG022095 (Early Life Conditions, Survival and Health; Smith PI). Partial support for all datasets within the UPDB was provided by Huntsman Cancer Institute (HCI) Cancer Support Grant P30 CA42014 from the National Cancer Institute.

Additional information

Funding

This work was supported by National Institutes of Health grant AG022095 (Early Life Conditions, Survival and Health; Smith PI). Partial support for all datasets within the UPDB was provided by Huntsman Cancer Institute (HCI) Cancer Support Grant P30 CA42014 from the National Cancer Institute.