1004 Variability in Non-Response to Treatment Among Hospitals and Ambulatory Surgical Centers: A Michigan Spine Surgery Improvement Collaborative (MSSIC) Study

View at publisher Summarise with VIRAT
Authors: Geoffrey D. Kahn, Anisse N. Chaker, Heegook Yeo, Jianhui Hu, Hsueh-Han H. Yeh, Nicole Schaub, Kari Jarabek, Jawad A. Saad, Saleh Al-Juburi, Michael Melhem, Noojan J. Kazemi, Daniel Park
+8 more Clifford Houseman, John A. Campbell, Matthew R. Karek, Miguelangelo Perez-Cruet, Michael J. Karsy, Muwaffak M. Abdulhak, David R. Nerenz, Victor W. Chang
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

INTRODUCTION: Spine surgery is typically performed to relieve pain or improve physical function, however some patients do not improve on either dimension following surgery. We have developed a definition of this “treatment non-response” and have identified a set of variables that act either as risk factors or protective factors for non-response. It is not known how variable the rate of non-response is from hospital to hospital. METHODS: The Michigan Spine Surgery Improvement Collaborative (MSSIC) registry was queried for patients undergoing elective lumbar decompression or fusion surgery from January 2017 to June 2024. Treatment non-response was determined based on PROMIS-PF scores and a combination of 90-day leg and back pain NRSs. Patients who failed to achieve minimal clinically important difference (MCID) or Patient Acceptable Symptom State (PASS) for both back and leg pain as well as for PROMIS-PF were considered treatment non-responders. Patients who achieved MCID or PASS on either PROMIS-PF or pain scores but were not highly satisfied with surgery (NASS Satisfaction index 1) were also considered treatment non-responders. RESULTS: 27,218 patients were analyzed with a 90-day treatment non-responder rate of 27.5%. Substantial between-site variation was observed ranging from 12% to 35%. When rates were adjusted for a range of risk or protective factors associated with non-response, top-performing sites had rates up to 5.7% lower than expected and lowest-performing sites had rates up to 8.6% higher than expected. For eight sites, Chi-squared tests showed observed failure rates were significantly different than expected. CONCLUSIONS: When comparing non-response to treatment rates between over 30 different hospitals or ASCs, we found significant variability in rates by site. This suggests a potential target for quality improvement if the underlying factors separating high- and low-performing sites can be identified.

Fields of Study & Subjects

Spine and Intervertebral Disc Pathology Musculoskeletal pain and rehabilitation Healthcare Quality and Management

Publication details

Journal / source
Neurosurgery
Volume
72
Issue
Supplement_1
Pages
173–174
Publication date
26 March 2026
Language
EN