The mediating role of posttraumatic stress reactions in the relationship between child abuse and physical health complaints in adolescence and young adulthood

Background: Child abuse is associated with physical health problems in adolescence and adulthood, but the mechanisms involved are unclear. A possible mediating effect of posttraumatic stress reactions (PTSR) linking child abuse with later physical health complaints has not been fully investigated. Objective: The current study investigated whether PTSR could be a potential mediator in the relationship between child abuse and physical health complaints in adolescents and young adults. If so, we also investigated whether this was the case for different child abuse types alone or in combination. Method: The study sample comprised 506 adolescents and young adult victims of child abuse and 504 unexposed matched controls aged 16–33 from a community sample. We measured child abuse retrospectively and current PTSR at wave 1 (2013), and current physical health complaints at wave 2 (2014/15). We tested a model of PTSR as a possible mediator between child abuse and physical health complaints and conducted causal mediation analysis to estimate direct and indirect pathways. Each child abuse type was studied in isolation and in combination with other abuse types. Results: PTSR had a notable, significant mediating effect on the relationship between child abuse and physical health complaints in our overall model (average causal mediation effect; ACME = 0.14, p < 0.001), accounting for 85% of the total effect. The mediated pathway was also significant in analyses of the different child abuse types. The mediating effect of PTSR was most prominent in individuals reporting exposure to more than one child abuse type. Conclusions: The current study indicates that PTSR may be an important mediator in the relationship between child abuse and physical health complaints. Health professionals should be aware of the important role that PTSR may have in maintaining or exacerbating physical health problems in child abuse victims. However, a reverse model could not be tested in this study and the results need confirmation in future prospective studies.