Changes in the ascertainment of multiple sclerosis.

OBJECTIVE

With diagnostic criteria alterations, increased MRI availability, and awareness of therapies, temporal changes in incidence and prevalence rates may occur, with an increase in the proportion of mildly affected persons diagnosed with multiple sclerosis (MS). The authors assessed temporal trends in the delay from symptom onset to diagnosis (DONDX), and determined whether the degree of disability at diagnosis differs by year of symptom onset (YONSET), using the NARCOMS Registry.

METHODS

The authors selected US participants with an age at symptom onset of 10 to 60 years, and YONSET > or = 1980 (n = 16,581). The authors divided YONSET into 5-year groups and compared DONDX between groups using multivariate Cox regression. The authors classified participants enrolled within 2 years of diagnosis (n = 5,548) as having mild, moderate, or severe disability using Patient Determined Disease Steps, and assessed the association of disability with YONSET using polytomous logistic regression.

RESULTS

DONDX decreased with later YONSET (r = -0.43, p < 0.0001). This association remained after adjustment for demographic factors in a multivariate Cox model. Later YONSET was associated with increased odds of having mild disability at diagnosis as compared to severe disability (OR = 1.10 per year; 1.09 to 1.11).

CONCLUSION

The delay from symptom onset to diagnosis is steadily decreasing in MS. An increasing proportion of patients with MS have mild disability at diagnosis after accounting for confounders. As the effectiveness of therapies is influenced by disease duration, this has implications for comparison of treatment effects in modern clinical trials to earlier study results.