Medical Student Hackensack-Meridian School of Medicine Chatham, New Jersey, United States
Introduction: Real world evidence is needed in evaluating neurosurgical interventions effectiveness and safety. Patient reported outcomes are the main way in which the effectiveness of different interventions are assessed. Longer patient follow up times there can be difficulty in maintaining contact with patients outside of typical clinical follow-up windows. The objective of this study is to evaluate patient specific factors that may be associated with decreased extended follow up in Cervical Spondylotic Myelopathy (CSM) patients.
Methods: A retrospective analysis of the CSM quality outcomes database (QOD) of a high enrolling private practice in the QOD Study Group was performed. Patient baseline characteristics, socioeconomic descriptors, and patient reported outcome measures (PROMs) were collected. Patients were marked as “responders” if they had completed their 60 month follow up survey.
Results: From the single practice QOD 65 patients were identified who underwent an intervention for CSM who completed baseline questionnaires. At 60 months, 48 patients were available for follow up and 17 were deceased. Of these 48 patients, 30 completed their 60 month follow up questionnaires and 18 did not. There was no difference between response status by age and medical comorbidities, except non responders had higher rates of depression (27% v. 7%, p=0.04). No difference was observed between the number of levels treated, readmission, and reoperation. Non responders had significantly lower socioeconomic status than responders (p=0.001). There was no variation in PROMs at 12 or 24 months between response groups except for satisfaction. Non responders at 60 months were significantly less satisfied at 12 and 24 months (p=0.04 and p=0.0001 respectively).
Conclusion : Responders at 60 months saw disproportionately higher representation from higher socioeconomic status individuals. Non responders at 60 months had significantly less satisfaction with their procedures. Non responders to long term QOD follow up may introduce bias into data sets. A better understanding of these limitations is needed in different practice settings to minimize bias in long term follow up.