Handling of missing items in the Oswestry disability index and the neck disability index. A study from Swespine, the National Swedish spine register.

Purpose: The Oswestry Disability Index (ODI) and the Neck Disability Index (NDI) scoring algorithms used by the Swedish spine register (Swespine) until April 2022 handled missing items somewhat differently than the original algorithms. The purpose of the current study was to evaluate possible differences in the ODI and NDI scores between the Swespine and the original scoring algorithms.

Methods: Patients surgically treated for degenerative conditions of the lumbar or cervical spine between 2003-2019 (lumbar) and 2006-2019 (cervical) were identified in Swespine. Preoperative and 1-year postoperative ODI/NDI data were used to evaluate differences between the Swespine and the original ODI/NDI algorithms with adjustment for at most 1 or 2 missing items using mean imputation.

Results: The preoperative as well as the 1-year postoperative ODI/NDI were approximately 1 unit out of 100 smaller for the Swespine algorithm, irrespective of adjustment model. The differences between preoperative and postoperative ODI/NDI scores were similar between the Swespine and the original scoring algorithms. There were occasional statistically significant differences between the preoperative-postoperative differences due to large sample sizes.

Conclusions: The Swespine algorithms, used until April 2022, underestimated the ODI and NDI by approximately 1 out of 100 units compared with the original algorithms. In addition, there were no statistically significant differences between the original algorithms when adjusting for at most 1 or 2 missing items. The algorithm has now been changed, also for historical data.