Predicting Coronavirus Disease 2019 Infection Risk and Related Risk Drivers in Nursing Homes: A Machine Learning Approach.

OBJECTIVE: Inform coronavirus disease 2019 (COVID-19) infection prevention measures by identifying and assessing risk and possible vectors of infection in nursing homes (NHs) using a machine-learning approach. DESIGN: This retrospective cohort study used a gradient boosting algorithm to evaluate risk of COVID-19 infection (ie, presence of at least 1 confirmed COVID-19 resident) in NHs. SETTING AND PARTICIPANTS: The model was trained on outcomes from 1146 NHs in Massachusetts, Georgia, and New Jersey, reporting COVID-19 case data on April 20, 2020. Risk indices generated from the model using data from May 4 were prospectively validated against outcomes reported on May 11 from 1021 NHs in California. METHODS: Model features, pertaining to facility and community characteristics, were obtained from a self-constructed dataset based on multiple public and private sources. The model was assessed via out-of-sample area under the receiver operating characteristic curve (AUC), sensitivity, and specificity in the training (via 10-fold cross-validation) and validation datasets. RESULTS: The mean AUC, sensitivity, and specificity of the model over 10-fold cross-validation were 0.729 [95% confidence interval (CI) 0.690‒0.767], 0.670 (95% CI 0.477‒0.862), and 0.611 (95% CI 0.412‒0.809), respectively. Prospective out-of-sample validation yielded similar performance measures (AUC 0.721; sensitivity 0.622; specificity 0.713). The strongest predictors of COVID-19 infection were identified as the NH's county's infection rate and the number of separate units in the NH; other predictors included the county's population density, historical Centers of Medicare and Medicaid Services cited health deficiencies, and the NH's resident density (in persons per 1000 square feet). In addition, the NH's historical percentage of non-Hispanic white residents was identified as a protective factor. CONCLUSIONS AND IMPLICATIONS: A machine-learning model can help quantify and predict NH infection risk. The identified risk factors support the early identification and management of presymptomatic and asymptomatic individuals (eg, staff) entering the NH from the surrounding community and the development of financially sustainable staff testing initiatives in preventing COVID-19 infection.

Stemming the Tide of COVID-19 Infections in Massachusetts Nursing Homes.

BACKGROUND/OBJECTIVES: In April 2020, Massachusetts nursing homes (NHs) became a hotspot for COVID-19 infections and associated deaths. In response, Governor Charles Baker allocated $130 million in additional funding for 2 months contingent on compliance with a new set of care criteria including mandatory testing of all residents and staff, and a 28-point infection control checklist. We aimed to describe the Massachusetts effort and associated outcomes. DESIGN: Longitudinal cohort study. SETTING: A total of 360 Massachusetts NHs. PARTICIPANTS: The Massachusetts Senior Care Association and Hebrew SeniorLife rapidly organized a Central Command team, targeted 123 "special focus" facilities with infection control deficiencies for on-site and virtual consultations, and offered all 360 facilities weekly webinars and answers to questions regarding infection control procedures. The facilities were also informed of resources for the acquisition of personal protective equipment (PPE), backup staff, and SARS-CoV-2 testing. MEASUREMENTS: We used two data sources: (1) four state audits of all NHs, and (2) weekly NH reports to the Massachusetts Center for Health Information and Analysis. Primary independent process measures were the checklist scores and adherence to each of its six core competencies. Primary outcomes were the average weekly rates of new infections, hospitalizations, and deaths in residents and staff. We used a hurdle mixed effects model adjusted for county COVID-19 prevalence to estimate relationships between infection control process measures and rates of new infections or deaths. RESULTS: Both resident and staff infection rates started higher in special focus facilities, then rapidly declined to the same low level in both groups. Adherence to infection control processes, especially proper wearing of PPE and cohorting, was significantly associated with declines in weekly infection and mortality rates. CONCLUSION: This statewide effort could serve as a national model for other states to prevent the devastating effects of pandemics such as COVID-19 in frail NH residents.