220. Do favorable short-term patient reported outcomes after spine surgery predict loss to one- and two-year follow-up?

Abstract

<h3>BACKGROUND CONTEXT</h3> Because many spine surgery patients are lost to long-term postoperative follow-up, retrospective patient outcome studies are subjected to attritional bias. Controversy ensues in the literature on whether long-term follow-up is associated with adverse clinical outcomes. Furthermore, the factors associated with loss to follow-up after spine surgery are incompletely elucidated. <h3>PURPOSE</h3> To determine if spine surgery patients with greater improvement in patient reported outcomes measures (PROM) at early postoperative follow-up are more likely to be lost to follow-up at the one- year and two- year postoperative visits. <h3>STUDY DESIGN/SETTING</h3> Retrospective cohort analysis. <h3>PATIENT SAMPLE</h3> All patients over age 18 who underwent any primary or revision decompression or fusion surgery for degenerative spinal pathologies at a single academic institution were retrospectively identified. All included patients attended a short-term (3 month) postoperative follow-up visit. Patients with a traumatic injury, infection, malignancy, or incomplete medical records or outcome measures were excluded. <h3>OUTCOME MEASURES</h3> Postoperative short- and long- term PROMs: Neck Disability Index (NDI), Oswestry Disability Index (ODI), Short Form-12 (SF-12), VAS Neck, VAS Arm, VAS Back, VAS Leg, and PROMIS pain scores. Delta outcome score. Recovery ratio. <h3>METHODS</h3> Patient surgeries were stratified by anatomic region, revision status, and number of levels fused. Distance from patient residence to postoperative clinic site was determined. Patients lost to one- and two-year postoperative follow-up visits were identified. Univariate analysis compared patient demographics, surgical factors, distances traveled, and PROMs across groups with and without one- and two- year follow-up, respectively. Logistic regression isolated predictors of loss to follow-up. Multivariate regression compared PROMs based on long-term follow-up status. <h3>RESULTS</h3> A total of 1,412 patients met the inclusion criteria. In the total cohort, 25% of patients were lost to one- year follow-up and 66% were lost to two-year follow-up. Change in patient outcome measures and recovery ratios did not meaningfully differ between patient groups with and without one- and two-year follow-up on univariate nor multivariate analyses. Multiple patient factors were found to be associated with failure to follow up at one year after spine surgery including: decreased medical comorbidities, primary and single level surgeries, private insurance status, and longer distance from residence to clinic. Additionally, younger age and female sex were associated with failure to follow up two years after index surgery. <h3>CONCLUSIONS</h3> Patient outcomes were not found to impact loss to long-term follow-up after spine surgery. While a multitude of factors may predict patient propensity to follow-up, the lack of association between postoperative follow-up status and clinical outcome may limit bias introduced in retrospective PROM studies. <h3>FDA DEVICE/DRUG STATUS</h3> This abstract does not discuss or include any applicable devices or drugs.