Assessing disability progression using the WHODAS 2.0 in multiple sclerosis: Investigating clinical and...

Identifying influences on disability accumulation in multiple sclerosis (MS), including modifiable factors other than the core features of disease itself, is vital for clinical care, but has often relied on instruments with acknowledged psychometric shortcomings. We model MS disability using the WHO Disability Assessment Schedule (WHODAS) 2.0, a validated measure based on the WHO's biopsychosocial model and sensitive to the breadth of disability-related domains important to people, to investigate the factors associated with its trajectory after diagnosis.

In a model allowing individual trajectories to vary around the population mean, we analysed the WHODAS 2.0, transforming it from ordinal raw scores to interval latent estimates via the Rasch Measurement Model (operational range 0 (low) to 128 (high disability)), across up to 6 surveys, from time of diagnosis to 24 years hence, by 5210 adults with MS from the UK's Trajectories of Outcome in Neurological Conditions-MS (TONiC-MS) longitudinal cohort study.

Whilst disability rose, on average, across the disease course, and was higher for people with progressive onset MS (mean difference (MD) 14.55, 95 % confidence interval (CI) 12.56, 16.54), there was considerable variability between people in both their level of disability at time of diagnosis (with the WHODAS 2.0 scores of 90 % of people with MS expected to lie between 1.63 to 61.90) and in their subsequent trajectories. We found higher education was associated with lower disability (MD -5.81, 95 % CI -7.46, -4.16) and also delayed disability progression. However, further analyses indicated that the effect of education may be partially-mediated by self-efficacy. Higher levels of self-efficacy were associated with lower disability, and this effect was greater for those with lower education. In addition, the WHODAS 2.0 is sensitive to facets affected by depression, including cognition and participation in community activities, and we found greater depression to be associated with higher disability (MD 7.98, 95 % CI 7.05, 8.91). Both a higher number of comorbidities, and smoking, were associated with greater disability, especially early in the disease course. Higher disability was also found in those: not in work (MD 15.42, 95 % CI 14.24, 16.59) and more fatigued (5.83, 95 % CI 5.46, 6.21 per 1SD increase on Neurological Fatigue Index MS). Our results also indicated within-individual variability was greatest in those recorded as relapsing remitting at study entry, compared to those recorded as having progressive MS.

The interplay between self-efficacy, education and disability reinforces the importance of interventions seeking to enhance self-efficacy. Our results additionally support early monitoring and targeting of comorbidities and smoking, which could reduce disability progression. The associations of time-varying depression and fatigue with disability also offer important opportunities for treatment. Characterising disability progression via validated measures provides information, targets for health promotion and indicators of groups warranting closer monitoring.