Predicting SF-6Dv2 utility scores for chronic low back pain using the Oswestry Disability Index and Roland-Morris
Thomas G Poder1,2, Nathalie Carrier3
1School of Public Health, Department of Management, Evaluation and Health Policy, University of Montreal , Montreal, QC, Canada.
Abstract:
Background: Generic preference-based measures are used to evaluate disability and health-related quality of life (HRQoL). Objective: To evaluate if Short Form Six-Dimensions (SF-6Dv2) is correlated with specific current questionnaires used in chronic low back pain (CLBP) and if a predictive equation of SF-6Dv2 could be established. Methods: Between October 2018 and January 2019, an online survey on CLBP was conducted. HRQoL was measured with two specific questionnaires, i.e. Oswestry Disability Index (ODI) and Roland-Morris Disability Questionnaire (RMDQ), and with the new version of the SF-6Dv2 as a generic preference-based measure. Results: 402 subjects completed at least two of the three HRQoL questionnaires. Mean (95% confidence interval) of SF-6Dv2, ODI, or RMDQ were, respectively, 0.561 (0.553-0.569), 43.7 (42.1-45.2), and 10.3 (9.8-10.8). SF-6Dv2 was moderately correlated with ODI and RMDQ (r = -0.635 and r = -0.542, p < 0.001). The best model to predict SF-6Dv2 explained 50.6% of variability and included ODI. The correlation between actual and predicted SF-6Dv2 was 0.71. Conclusion: This study demonstrated that SF-6Dv2 was moderately correlated with ODI and RMDQ and that ODI was a better predictor. There was a strong correlation between actual and predicted SF-6Dv2 from multivariate models. These results suggest that the model can be used in similar studies to estimate the SF-6Dv2 when it was not measured.
