Premium
Mediator Variables in Headache Research: Methodological Critique and Exemplar Using Self‐Efficacy as a Mediator of the Relationship Between Headache Severity and Disability
Author(s) -
Peck Kelly R.,
Smitherman Todd A.
Publication year - 2015
Publication title -
headache: the journal of head and face pain
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.14
H-Index - 119
eISSN - 1526-4610
pISSN - 0017-8748
DOI - 10.1111/head.12633
Subject(s) - mediation , confidence interval , migraine , clinical psychology , psychology , physical therapy , medicine , psychiatry , political science , law
Background Despite advances in headache medicine, there remains little research on process‐related variables that mediate relations between headache and outcomes, as well as limited dissemination of optimal statistical methodology for conducting mediation analyses. The present paper thus aims to promote and demonstrate a contemporary approach to mediation analysis as applied to headache. Methods An overview of a contemporary path‐analytic approach to mediation analysis is presented, with an empirical exemplar for illustrative purposes. In the exemplar, headache management self‐efficacy ( HMSE ) was proposed as a mediator between headache severity and disability. The sample included 907 young adults (M age = 19.03 [SD = 2.26]; 70.8% female) with primary headache. Direct and indirect effects of headache severity on headache disability through HMSE were assessed using the espoused methods. Results Pain severity was positively associated with headache disability ( β = 2.91, 95% confidence interval [CI; 2.62, 3.19]) and negatively associated with HMSE ( β = −3.50, 95% CI [−4.24, −2.76]); HMSE was negatively associated with headache disability ( β = 0.07, 95% CI [−0.09, −0.04]). A positive indirect effect of pain severity on disability through HMSE was identified (point estimate = 0.24, 95% CI [0.14, 0.34]); thus, self‐efficacy mediated the association between pain severity and disability. The proposed mediation model accounted for 38% of total variance in disability ( P < .001). Conclusions There is a need for theory‐driven and statistically rigorous mediation analyses within the headache literature. In one exemplar application, self‐efficacy partially accounted for the disability resulting from headache. We advocate for increased attention to intervening variables in headache via dissemination of contemporary mediation analyses.