Recruitment bias in chronic pain research: whiplash as a model

Abstract

In science findings which cannot be extrapolated to other settings are of little value. Recruitment methods vary widely across chronic whiplash studies, but it remains unclear whether this generates recruitment bias. The present study aimed to examine whether the recruitment method accounts for differences in health status, social support, and personality traits in patients with chronic whiplash-associated disorders (WAD). Two different recruitment methods were compared: recruiting patients through a local whiplash patient support group (group 1) and local hospital emergency department (group 2). The participants (n=118) filled in a set of questionnaires: the Neck Disability Index, Medical Outcome Study Short-Form General Health Survey, Anamnestic Comparative Self-Assessment measure of overall wellbeing, Symptom Checklist-90, Dutch Personality Questionnaire, and the Social Support List. The recruitment method (either through the local emergency department or patient support group) accounted for the differences in insufficiency, somatization, disability, quality of life, self-satisfaction, and dominance (all p values <.01). The recruitment methods generated chronic WAD patients comparable for psychoneurotism, social support, self-sufficiency, (social) inadequacy, rigidity, and resentment (p>.01). The recruitment of chronic WAD patients solely through patient support groups generates bias with respect to the various aspects of health status and personality, but not social support. In order to enhance the external validity of study findings, chronic WAD studies should combine a variety of recruitment procedures.