McKay MT, Cannon M, Healy C, Syer S, O'Donnell L, Clarke MC. A meta-analysis of the relationship between parental death in childhood and subsequent psychiatric disorder.
Acta Psychiatr Scand 2021;
143:472-486. [PMID:
33604893 DOI:
10.1111/acps.13289]
[Citation(s) in RCA: 10] [Impact Index Per Article: 3.3] [Reference Citation Analysis] [Abstract] [Key Words] [Track Full Text] [Journal Information] [Submit a Manuscript] [Subscribe] [Scholar Register] [Received: 07/29/2020] [Revised: 01/22/2021] [Accepted: 02/10/2021] [Indexed: 01/27/2023]
Abstract
OBJECTIVE
To systematically review evidence for an association between parental death in childhood, and the subsequent development of an anxiety, affective or psychotic disorder.
METHODS
Electronic databases (Scopus, Medline (for Ovid), EMBASE and PsychINFO) were searched for peer-reviewed, cohort studies in the English language. Meta-analyses were performed for studies reporting hazard ratios, incidence rate ratios and odds ratios. Two studies reported risk ratios, and these were included in an overall pool of odds, risk and incidence rate ratios. Sensitivity analyses were performed (removal of one study at a time) for all meta-analyses, and study quality assessed using the Newcastle-Ottawa Scale.
RESULTS
Fifteen studies were retained, and where required, data were averaged in advance of pooling. Significant results were observed in studies reporting hazard ratios (k = 4, 1.48 [95% CI = 1.32-1.66]), incidence rate ratios (k = 3, 1.37 [95% CI = 1.01-1.85]), but not odds ratios (k = 4, 0.87 [95% CI = 0.72, 1.05]). However, the overall pooled effect (using odds, incidence rate and risk ratios) was statistically significant (k = 9, 1.22 [95% CI = 1.03-1.44]).
CONCLUSION
Overall, the evidence suggests that there is a positive association between the death of a parent before age 18, and the subsequent development of an anxiety, affective or psychotic disorder. The lack of a significant pooled effect in studies reporting results as odds ratios is likely an artefact of study design.
LIMITATIONS
Data were clustered in four countries making generalizability uncertain. Studies adjusted for a variety of possible confounders, and follow-up after death varied considerably.
Collapse