8:36 General versus spine-specific health surveys: do comorbidities influence baseline results?

Abstract

Purpose of study: Health status instruments have become important tools for understanding the impact of disease and the effectiveness of interventions. Two types of instruments are in current use. General-health surveys, such as the Short Form (SF)-36, attempt to measure the overall health status of patients, and may reflect multiple diseases or disabilities. Condition-specific surveys, such as the Oswestry, attempt to isolate the specific symptoms related to one condition and should not reflect the presence of unrelated comorbid conditions. The objectives of this study are to assess whether the SF-36 is more sensitive to the number of self-reported comorbidities than the Oswestry and to determine which comorbidities have the greatest impact on baseline health.Methods used: This is an observational study of 26,736 patients from the National Spine Network. Patients were grouped according to the number of self-reported comorbidities (0 to 5). Baseline SF-36 scores were computed for the Physical Component Summary (PCS), Bodily Pain (BP) and Physical Function (PF) scales. The Oswestry Disability Index (ODI, MODEMS version) was scored in reverse order (0 = severed disability, 100 = no disability) for consistency with the SF-36 scales. Analysis of variance was used to assess the difference in mean health status across comorbidity groups. Multiple linear regression was used to identify the most influential comorbidities.of findings: Baseline SF-36 and Oswestry scores were significantly lower for patients with a greater number of comorbidities. For patients with zero to five comorbidities, ODI scores ranged from 62.2 to 48.8, PCS scores from 33.3 to 26.0, PF scores from 50.4 to 29.1 and BP scores from 33.7 to 23.0 (p<.001 for all scales). Comorbidities most adversely impacting the ODI scores included depression, smoking, osteoporosis, osteoarthritis, blood disorders and headaches. Comorbidities most negatively impacting PCS scores include osteoarthritis, smoking, blood and lung disorders. For the BP and PF scales, the greatest negative impact came from depression, smoking, osteoporosis and osteoarthritis. Additional factors associated with low ODI and SF-36 scores included obesity, poor self-rated health, compensation and low education (factors significant at p<.01).Relationship between findings and existing knowledge: Contrary to current assumptions regarding condition-specific surveys, comorbidities are associated with lower ODI scores. The association between lower baseline ODI scores and a higher number of self-reported comorbidities may relate to worse underlying spine disease in patients with these comorbidities, but it also may reflect lack of true disease specificity for the ODI.Overall significance of findings: Clinicians and researchers should understand and consider the effect that comorbidities have on health-status measures. Further research may be necessary in order to develop a truly "condition-specific" instrument for spine patients.Disclosures: No disclosures.Conflict of interest: No conflicts.