Spine Surgeons Accurately Estimate the Probability of Favorable One-Year Postoperative Lumbar Surgery Outcomes

Abstract

It is important that surgeons be able to make accurate probability estimates as a basis for offering good advice to lumbar surgical candidates. The purpose of this study was to evaluate the accuracy of spine surgeons' preoperative estimates of the probability of improvement in pain or functional outcomes after lumbar surgery. The study design was that of a prospective cohort study in a group practice setting. Patients aged 18–65, presenting as candidates for surgery with a diagnosis of lumbar radiculopathy and no history of low-back surgery, were asked to participate in the study. Data from 149 surgical patients were included in the analysis. Change in the North American Spine Society (NASS) pain and function instrument was the outcome measure. For patients who were offered surgery, the surgeon estimated the probability that the patient would experience an improvement in pain or function one year after surgery. Patients were surveyed again at 3 and 12 months postsurgery. If the patient's NASS score improved, the patient was considered to have experienced improvement in pain and function. Surgeons predicted that 33% of patients had greater than a 90% chance of improvement; 38% of patients had a 81–90% chance of improvement; and 23% of patients had a 71–80% chance of improvement. Only 6% of patients were given less than a 70% chance of improvement. Ninety-four percent of patients that had been predicted to have a 91% or higher probability of improvement actually did achieve an improvement. Of those predicted to have an 81–90% chance of improvement, 93% actually improved. Of those predicted to have a 71–80% probability of improvement, 79% actually improved. Overall, surgeons predicted that 88% of patients would improve, and 90% patients actually reported improvement. For patients who were offered surgery, the surgeon's estimate of probability of success at the time of surgery recommendation was reasonably accurate. Surgeons correctly assessed probability of success for those given a 71–80% chance of improvement and those given a 91% or higher chance of improvement. Surgeons were slightly pessimistic in their probability estimates when they estimated probability of improvement in the 81–90% range.