A Predictive Model for Severe Covid-19 in the Medicare Population: A Tool for Prioritizing Scarce Vaccine Supply

Background Recommendations for prioritizing populations for COVID-19 vaccination have focused on front-line health care personnel and residents in long term care, followed by other individuals at higher risk for severe disease. Existing models for identifying higher risk individuals including those over age 65 lack the needed integration of socio-demographic and clinical risk factors to ensure equitable vaccine allocation. Methods We developed a predictive model for severe COVID-19 using clinical data from de-identified Medicare claims for 16 million Medicare fee-for-service beneficiaries, including 1 million COVID-19 cases, and socio-economic data from the CDC Social Vulnerability Index. To identify risk factors for severe COVID-19, we used multivariate logistic regression and random forest modeling. Predicted individual probabilities of COVID-19 hospitalization were then calculated for population risk stratification and COVID-19 vaccine prioritization, and for mapping of population risk levels at the county and zip code levels on a nationwide dashboard. Results The leading Covid-19 hospitalization risk factors driving the risk model were: Non-white ethnicity (particularly North American Native, Black, and Hispanic), end-stage renal disease, advanced age (particularly age over 85), prior hospitalization, leukemia, morbid obesity, chronic kidney disease, lung cancer, chronic liver disease, pulmonary fibrosis or pulmonary hypertension, and chemotherapy. However, previously reported risk factors such as chronic obstructive pulmonary disease and diabetes conferred modest hospitalization risk. Among all social vulnerability factors analyzed, residence in a low-income zip code was the only risk factor independently predicting Covid-19 hospitalization. The mapped hospitalization risk levels showed significant correlations with the cumulative COVID-19 case hospitalization rates at the zip code level in the fifteen most populous U.S. major metropolitan areas. Conclusion This multi-factor risk model which predicts severe Covid-19and its population risk dashboard can be used to optimize Covid-19 vaccine allocation in the higher risk Medicare population where socio-demographic and comorbidity risk factors need to be considered concurrently. ### Competing Interest Statement The authors have declared no competing interest. ### Funding Statement This work was funded by the Johns Hopkins University Applied Physics Laboratory (JHU-APL) under a prime contract with the Department of Defense Joint Artificial Intelligence Center (JAIC) for Project Salus. Project Salus is a JAIC project in support of the National Guard and other military personnel to provide needed supplies and personnel to health care facilities impacted by the Covid-19 pandemic. ### Author Declarations I confirm all relevant ethical guidelines have been followed, and any necessary IRB and/or ethics committee approvals have been obtained. Yes The details of the IRB/oversight body that provided approval or exemption for the research described are given below: This project was carried out under the supervision of the Department of Defense Joint Intelligence Center under a data use agreement with CMS. All necessary patient/participant consent has been obtained and the appropriate institutional forms have been archived. Yes I understand that all clinical trials and any other prospective interventional studies must be registered with an ICMJE-approved registry, such as ClinicalTrials.gov. I confirm that any such study reported in the manuscript has been registered and the trial registration ID is provided (note: if posting a prospective study registered retrospectively, please provide a statement in the trial ID field explaining why the study was not registered in advance). Yes I have followed all appropriate research reporting guidelines and uploaded the relevant EQUATOR Network research reporting checklist(s) and other pertinent material as supplementary files, if applicable. Yes The Medicare claim data files used in this research is protected health information as defined by the Health Insurance Portability and Accountability Act (HIPAA). Access to these data sets require CMS approval. The 2018 CDC Social Vulnerability Index data can be downloaded from the CDC website. [https://www.atsdr.cdc.gov/placeandhealth/svi/data\_documentation\_download.html][1] [1]: https://www.atsdr.cdc.gov/placeandhealth/svi/data_documentation_download.html