The Association Between Patient Reported Outcomes of Spinal Surgery and Societal Costs: A Register Based Study.

Study Design: Retrospective register-based study using Swedish registers and data prospectively collected in quality register Swespine.

Objective: Analyze the association of societal costs and spine surgery outcome in low back pain (LBP) patients based on patient reported outcome measures (PROMs).

Summary Of Background Data: Studies show that LBP has a substantial impact on societal cost. There are indications that the burden diverges over different patient groups, but little is known about cost patterns in relation to PROMs of LBP surgery.

Methods: We utilized a database with data from six registers. All lumbar spine surgery patients registered in Swespine 2000 to 2012 were identified. Swespine collects PROMs Global Assessment of pain improvement (GA), Oswestry Disability Index (ODI), Visual Analog Scale (VAS), and EuroQol five-dimension scale (EQ-5D). A literature search was conducted to identify threshold changes in ODI, VAS, and EQ-5D representing a significant improvement or deterioration as defined by the minimal clinically important difference (MCID). We categorized patients into groups by their GA response at 2-year follow-up and estimated mean changes in ODI, VAS, and EQ-5D for each group. These changes were compared with the MCID thresholds to determine a GA-anchored classification of surgical outcomes. Costs consisted of out/inpatient care, sick leave, early retirement, and pharmaceuticals.

Results: In total, 12,350 patients were included. GA 1-2 ("pain has disappeared"/"pain is much improved") were labeled successful surgery outcomes (67%), GA 3 ("pain somewhat improved"), undetermined (16%), and GA 4-5 ("no change in pain"/"pain has worsened") unsuccessful (17%). Costs of the unsuccessful and undetermined were higher than of the successful during the entire study period, with differences increasing markedly post-surgery. For the successful, a downward cost trend was observed; costs almost returned to the level observed 3 years pre-surgery. No such trend was observed in the other groups.

Conclusion: Identifying patients with higher probability of responding to surgery could lead to improved health and substantial societal cost savings.

Level Of Evidence: 3.