A modelling approach to estimate the prevalence of treatment-resistant schizophrenia in the United States.

• Main Authors: Michael Frank Mørup, Steven M Kymes, Daniel Oudin Åström
• Format: Article
• Language: English
• Published: Public Library of Science (PLoS) 2020-01-01
• Series: PLoS ONE
• Online Access: https://doi.org/10.1371/journal.pone.0234121

INTRODUCTION:Schizophrenia is a condition that places a significant burden on individuals with the condition, their family, and society. A large proportion of those treated for schizophrenia do not experience treatment response and are referred to as having "treatment-resistant schizophrenia" (TRS). Expert opinion has long held that the prevalence of TRS among individuals with schizophrenia is 30%, but the basis of this estimate is unclear. This article presents a model developed for estimating the prevalence of TRS in the United States 2014. METHODS:An incidence-prevalence-mortality model was developed to estimate the prevalence of TRS in the United States. The model was populated with data from public health agencies and published literature. Prevalence in 2014 was modelled using a Markov cohort simulation for each birth cohort between 1930 to 2014. RESULTS:Using different scenarios for baseline incidence, relative risks of mortality, it was estimated that approximately 22% of individuals with schizophrenia would be considered treatment-resistant in 2014. DISCUSSION:The results suggests that prevalence of TRS may be somewhat lower than the 30% often reported, however this is highly dependent on the definition of treatment resistance. Methods such as this may help answer epidemiological and health policy questions as well as test the influence of key underlying assumptions.