For each disorder, prevalence data were assembled by conducting systematic reviews of the published and gray literature to capture data on prevalence, incidence, remission, duration, and excess mortality [16,18,20,21,27–29]. DisMod-MR, a Bayesian meta-regression tool [11] developed specifically for GBD 2010, was then used to model prevalence by disorder type, age, gender, year, region and country. A generalized negative binomial model was estimated for all epidemiological data using super-region, region, and country random effects, as well as two sets of covariates: study level covariates that adjusted for systematic bias in the raw epidemiological data, and country level covariates that aided the predictive power of the model by adjusting for known ecological effects in the data, such as the effect of conflict or economic status on prevalence.
