Background: Methodological issues have been identified when quantifying exposure to adversity and abuse. To address a single type may obscure covarying effects. To sum multiple exposures gives equal weight to each. Latent class analysis (LCA) addresses this by identifying homogenous subpopulations. Most studies using LCA have pooled gender data in spite of evidence that the nature and frequency of exposure differs by gender. Males report more interpersonal abuse, females report more of other exposures, particularly sexual.
Objective: This study aimed to identify if stratifying data by gender resulted in different profiles of adversity/abuse Participants and setting: The National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) wave II, a large community-based survey, nationally representative of the US population.
Method: This study used 14 indicators of childhood adversity as the basis for LCA.
Results: The number and nature of classes differed by gender. The best solution for females was 4-class: a low risk class; a class at higher risk of sexual abuse; a class at higher risk of physical abuse; a class at higher risk of combined physical and sexual abuse. The best solution for males had only 3-classes; a low risk class, a class at higher risk of sexual abuse; a class at higher risk of physical abuse. The combined dataset resulted in a solution similar to the female solution.
Conclusion: The importance of developing models for males and females separately was evidenced by the male and female classes being differentially associated with mental health variables.
Keywords: Abuse; Adversity; Childhood trauma; Gender; Latent class analysis.
Copyright © 2019 Elsevier Ltd. All rights reserved.