Abstract
BACKGROUND: As access to cannabis has increased, there has been a rise in a condition called cannabinoid hyperemesis syndrome (CHS). This study estimates annual prevalence of suspected CHS at emergency department visits (ED) over an 11-year period in Northern California. METHODS: This retrospective observational cohort study used electronic health records from Kaiser Permanente Northern California. Two CHS case definitions were used to construct two cohorts of adults (18+) with ≥1 CHS visits from 2009 to 2019. The primary definition used a narrow definition based on past studies (CHS group 1) and an exploratory definition allowed for a broader range of codes (CHS group 2); both definitions required a primary diagnosis of vomiting. Annual prevalence of CHS and annual rates of counts of CHS visits estimated using a log-link Poisson model are reported per group. FINDINGS: There were 57,227 patients with ≥1 CHS visits included in CHS group 1 and 65,645 patients included in CHS group 2. Over eleven years, CHS increased across groups with the fastest rise in CHS group 1 (prevalence ratio = 2.75, 95 % confidence interval [CI] 2.65-2.85, p<.0001 from 2009 to 2019 vs. prevalence ratio = 2.34, 95 % CI 2.27-2.43). CHS group 1 also exhibited the largest increase in ED visits (rate ratio = 2.35, 95 % CI 2.27-2.43, p<.0001). CONCLUSION: In a large California population, suspected CHS increased over time across definitions. Annual prevalence increased by 134-175 %, depending on CHS definition. CHS group 2's definition may have been too broad and changes in ICD-10-CM coding may have impacted estimates.