Does dysphoric mood really predict the outcome of lumbar surgery? Methodological pitfalls in psychological research

This study investigates the relationship between depression and continuous pain after lumbar surgery by means of the Beck Depression Inventory (BDI). To assess the possibility that some somatic symptoms are confounded with pain, the items of the inventory were divided into a cognitive-affective and somatic subscale. Data analysis is based on two assumptions: (1) continuous pain after surgery is more closely related to the somatic subscale than to the cognitive-affective subscale of the BDI; and (2) postsurgical pain can be better predicted by the somatic than by the cognitive-affective subscale presurgery. The sample consisted of 61 patients with lumbar nucleotomy and 107 patients with lumbar spondylodesis. Treatment outcome was evaluated 6 months after surgery in the nucleotomy patients and 9 months after surgery in the spondylodesis patients. At follow-up, subjects conducted a pain diary over a period of 2 weeks. Correlational methodology was applied. Data analysis of both samples confirms the first assumption, whereas the second assumption is only partly confirmed. Only in the spondylodesis sample was pain intensity at follow-up predicted by the total BDI score. Regarding the subscales, the cognitive-affective scale, but not the somatic scale, was related to surgical outcome in this sample. The relationship between presurgical depression and pain at follow-up failed to reach statistical significance in the nucleotomy sample. It is concluded that lack of awareness of the confounding effects of somatic items in questionnaires for the assessment of mood may contribute to erroneous conclusions drawn from studies reported in the literature.