RTI uses cookies to offer you the best experience online. By clicking “accept” on this website, you opt in and you agree to the use of cookies. If you would like to know more about how RTI uses cookies and how to manage them please view our Privacy Policy here. You can “opt out” or change your mind by visiting: http://optout.aboutads.info/. Click “accept” to agree.
Sources of variation in estimates of Duchenne and Becker muscular dystrophy prevalence in the United States
Whitehead, N., Erickson, S. W., Cai, B., McDermott, S., Peay, H., Howard, J. F., Ouyang, L., & Muscular Dystrophy Surveillance, Tracking and Research Network (2023). Sources of variation in estimates of Duchenne and Becker muscular dystrophy prevalence in the United States. Orphanet Journal of Rare Diseases, 18(1), 65. https://doi.org/10.1186/s13023-023-02662-0
BACKGROUND: Direct estimates of rare disease prevalence from public health surveillance may only be available in a few catchment areas. Understanding variation among observed prevalence can inform estimates of prevalence in other locations. The Muscular Dystrophy Surveillance, Tracking, and Research Network (MD STARnet) conducts population-based surveillance of major muscular dystrophies in selected areas of the United States. We identified sources of variation in prevalence estimates of Duchenne and Becker muscular dystrophy (DBMD) within MD STARnet from published literature and a survey of MD STARnet investigators, then developed a logic model of the relationships between the sources of variation and estimated prevalence.
RESULTS: The 17 identified sources of variability fell into four categories: (1) inherent in surveillance systems, (2) particular to rare diseases, (3) particular to medical-records-based surveillance, and (4) resulting from extrapolation. For the sources of uncertainty measured by MD STARnet, we estimated each source's contribution to the total variance in DBMD prevalence. Based on the logic model we fit a multivariable Poisson regression model to 96 age-site-race/ethnicity strata. Age accounted for 74% of the variation between strata, surveillance site for 6%, race/ethnicity for 3%, and 17% remained unexplained.
CONCLUSION: Variation in estimates derived from a non-random sample of states or counties may not be explained by demographic differences alone. Applying these estimates to other populations requires caution.
