Mental/Behavioral Health as a Predictor of Initial COVID-19 Diagnosis: Results from the Colorado All Payer Claims Data Set to June 30, 2020.

Abstract

Background: Mental health conditions have not been sufficiently considered as a pre-existing condition for COVID-19. To address this, an analysis was completed from the Colorado All Payer Claims Data through June 30, 2020, using a matched design to non-COVID-19 patients. \n \nMethods: The study was based on individuals listed in the data set with an insurance claim with a date of service from March 1, 2020 to June 30, 2020. (COVID-19 patients were identified based on a claim with a principal diagnosis of COVID-19 between March-June 2020. Propensity score matching (PSM) was used to create the reference group based on the calendar month of each COVID-19 event date and a claim dated in the same month for the reference group. Additional criteria for matching included: age band (0-29 years, 30-64 years, 65+ years), sex, estimated number of months of insurance coverage and quartiles of health care expenditures based on the 12 months prior to the event date. \n \nVariables examined in prediction models for COVID-19 incidence included primary diagnosis of a mental/behavioral health (MBH) condition, hypertension, diabetes, chronic obstructive lung disease, asthma, and chronic kidney disease in the 12 months prior to the event date and interaction terms between MBH and each of the five physical health conditions. \n \nResults: The study comprised n=7,599 in both the COVID-19 group and the reference. A prior mental/behavioral health diagnosis was an independent predictor among people with a COVID-19 diagnosis. Mental/behavioral health conditions with especially high prevalence differences between COVID-19 patients and non-COVID-19 patients included schizophrenia/other psychotic disorders and depressive disorders. Previously recognized risk factors including hypertension, diabetes, chronic obstructive pulmonary disease, chronic kidney disease and asthma were also confirmed as independent predictors of COVID-19 in this population. Most significantly, when a diagnosis of mental/behavioral health appeared in conjunction with one of these physical diseases, it acted as a multiplier for predicting COVID-19 (the multiplier effect ranges from 2.28 to 4.29 over individuals with only one of the major medical diseases previously linked to COVID-19 risk). \n \nRecommendation: Current immunization strategies and outreach should consider putting Individuals with prior mental/behavioral health history in high priority groups. As vaccine becomes more generally available consideration should be given monitoring vaccine compliance among the MBH population.