Creating value in spine surgery: using patient reported outcomes to compare the short-term impact of different orthopedic surgical procedures

Abstract

BACKGROUND CONTEXTSociety increasingly asks Medicine to create “value” for patients. As health-care costs rise, this question will become more important. Debate exists regarding the relative “value” of many surgical procedures, including spinal surgery. Comparison of the relative value that patients experience after different orthopedic procedures is theoretical, but informs the ongoing debate. METHODSThe Patient Reported Outcome Measurement Information System (PROMIS) assessments for Physical Function, Pain Interference, and Depression are routinely collected in our orthopedic clinics. Patients who underwent lumbar discectomy (DSC) or arthroscopic anterior cruciate ligament reconstruction (ACLR) were retrospectively identified. Data relating to PROMIS domains, patient demographics, and other relevant encounter details were extracted. The primary outcomes were (1) preoperative PROMIS domain scores, (2) scores at a minimum of 40 days postoperatively for DSC patients and 133 days postoperatively for ACLR patients, and (3) the change in scores with surgery. Propensity score matching identified age-, sex-, race-, and comorbidity-matched groups from each cohort. Chi-square tests and nonparametric Kruskal-Wallis tests compared the distribution of outcomes and characteristics. Multivariate linear regression models with interactions between the matched cohort and operative phase estimated the change in the outcomes scores between the two cohorts and controlled for the baseline differences between them. RESULTSBefore surgery, the DSC cohort had lower physical function, higher pain interference and higher depression scores as compared with the ACLR cohort. This pattern remained postoperatively, indicating less desirable outcomes for DSC patients. However, after controlling for their baseline scores, DSC patients experienced significantly greater improvements after surgery of 3.84 (95% CI 1.08–6.60; p=.01), −4.87 (95% CI −7.52 to −2.23; p<.001), and −2.95 (95% CI −5.70 to −0.21; p=.04) points in their physical function, pain interference, and depression scores, respectively, as compared with ACLR patients. CONCLUSIONSBased upon PROMIS assessments at short-term follow-up, DSC patients receive a larger benefit from surgery than ACLR despite the overall less desirable postoperative PROMIS scores in the DSC cohort. This result, while theoretical, informs the debate regarding the comparative value of DSC to patients.