Purpose: Surgery for degenerative lumbar spinal conditions offers tremendous benefit for patients with moderate/severe symptoms failing non-operative treatment. There is little appreciation among referring family physicians (FPs) on factors that identify the ideal surgical candidate. Differences in preferences between patients and physicians leads to wide variation in referrals and impedes the shared decision-making process. Our purpose was to identify the dominant clinical factors influencing patient, FP, and surgeon preferences for lumbar spinal surgery.

Method: We used conjoint analysis, a rigorous method for eliciting preferences, in a mailed survey to all orthopaedic and neurosurgeons, a random sample of FPs, and patients in Ontario to determine the importance that respondents place on decisions for lumbar spinal surgery. We identified six clinical factors (walking tolerance, duration of pain, pain severity, neurological symptoms, typical onset, and dominant location of pain) and presented 16 hypothetical vignettes to participants who rated, on a six-point-scale, their preference for surgery. Data were analyzed using random-effects ordered probit regression models and relative importance of each clinical factor was reported.

Results: We obtained responses from 131 surgeons, 202 FPs, and 164 patients. We demonstrated that despite wide variations in overall responses, all six clinical factors were highly associated with surgical preference (p< 0.01). Surgeons placed the highest importance on the location of pain (34%), followed by pain severity (19%) and walking tolerance (19%). FPs considered neurological symptoms (23%), walking tolerance (20%), pain severity (20%), and typical onset (16%) to all be of similar importance. Pain severity (29%), walking tolerance (29%), and duration of pain (28%) were the most important factors for patients in deciding for surgery. Orthopaedic (over neurosurgical) specialty was statistically associated with a lower preference for surgery (p< 0.047). Older patient age (p< 0.03) and previous surgical consultation (p< 0.03) were both associated with a greater patient preference for considering surgery.

Conclusion: Different preferences for surgery exist between surgeons, FPs and patients. FPs may reduce over- and under-referrals by appreciating surgeons’ importance on location of pain (leg versus back). Surgeons and FPs may improve the shared decision making process by understanding that patients place high importance on duration, severity, and walking tolerance.