Predicting COVID-19 Outbreaks in Correctional Facilities Using Machine Learning.

Introduction. The risk of infectious disease transmission, including COVID-19, is disproportionately high in correctional facilities due to close living conditions, relatively low levels of vaccination, and reduced access to testing and treatment. While much progress has been made on describing and mitigating COVID-19 and other infectious disease risk in jails and prisons, there are open questions about which data can best predict future outbreaks. Methods. We used facility data and demographic and health data collected from 24 prison facilities in the Pennsylvania Department of Corrections from March 2020 to May 2021 to determine which sources of data best predict a coming COVID-19 outbreak in a prison facility. We used machine learning methods to cluster the prisons into groups based on similar facility-level characteristics, including size, rurality, and demographics of incarcerated people. We developed logistic regression classification models to predict for each cluster, before and after vaccine availability, whether there would be no cases, an outbreak defined as 2 or more cases, or a large outbreak, defined as 10 or more cases in the next 1, 2, and 3 d. We compared these predictions to data on outbreaks that occurred. Results. Facilities were divided into 8 clusters of sizes varying from 1 to 7 facilities per cluster. We trained 60 logistic regressions; 20 had test sets with between 35% and 65% of days with outbreaks detected. Of these, 8 logistic regressions correctly predicted the occurrence of an outbreak more than 55% of the time. The most common predictive feature was incident cases among the incarcerated population from 2 to 32 d prior. Other predictive features included the number of tests administered from 1 to 33 d prior, total population, test positivity rate, and county deaths, hospitalizations, and incident cases. Cumulative cases, vaccination rates, and race, ethnicity, or age statistics for incarcerated populations were generally not predictive. Conclusions. County-level measures of COVID-19, facility population, and test positivity rate appear as potential promising predictors of COVID-19 outbreaks in correctional facilities, suggesting that correctional facilities should monitor community transmission in addition to facility transmission to inform future outbreak response decisions. These efforts should not be limited to COVID-19 but should include any large-scale infectious disease outbreak that may involve institution-community transmission. Highlights: The risk of infectious disease transmission, including COVID-19, is disproportionately high in correctional facilities.We used machine learning methods with data collected from 24 prison facilities in the Pennsylvania Department of Corrections to determine which sources of data best predict a coming COVID-19 outbreak in a prison facility.Key predictors included county-level measures of COVID-19, facility population, and the test positivity rate in a facility.Fortifying correctional facilities with the ability to monitor local community rates of infection (e.g., though improved interagency collaboration and data sharing) along with continued testing of incarcerated people and staff can help correctional facilities better predict-and respond to-future infectious disease outbreaks.

When Is Mass Prophylaxis Cost-Effective for Epidemic Control? A Comparison of Decision Approaches.

BACKGROUND: For certain communicable disease outbreaks, mass prophylaxis of uninfected individuals can curtail new infections. When an outbreak emerges, decision makers could benefit from methods to quickly determine whether mass prophylaxis is cost-effective. We consider 2 approaches: a simple decision model and machine learning meta-models. The motivating example is plague in Madagascar. METHODS: We use a susceptible-exposed-infectious-removed (SEIR) epidemic model to derive a decision rule based on the fraction of the population infected, effective reproduction ratio, infection fatality rate, quality-adjusted life-year loss associated with death, prophylaxis effectiveness and cost, time horizon, and willingness-to-pay threshold. We also develop machine learning meta-models of a detailed model of plague in Madagascar using logistic regression, random forest, and neural network models. In numerical experiments, we compare results using the decision rule and the meta-models to results obtained using the simulation model. We vary the initial fraction of the population infected, the effective reproduction ratio, the intervention start date and duration, and the cost of prophylaxis. LIMITATIONS: We assume homogeneous mixing and no negative side effects due to antibiotic prophylaxis. RESULTS: The simple decision rule matched the SEIR model outcome in 85.4% of scenarios. Using data for a 2017 plague outbreak in Madagascar, the decision rule correctly indicated that mass prophylaxis was not cost-effective. The meta-models were significantly more accurate, with an accuracy of 92.8% for logistic regression, 95.8% for the neural network model, and 96.9% for the random forest model. CONCLUSIONS: A simple decision rule using minimal information about an outbreak can accurately evaluate the cost-effectiveness of mass prophylaxis for outbreak mitigation. Meta-models of a complex disease simulation can achieve higher accuracy but with greater computational and data requirements and less interpretability. HIGHLIGHTS: We use a susceptible-exposed-infectious-removed model and net monetary benefit to derive a simple decision rule to evaluate the cost-effectiveness of mass prophylaxis.We use the example of plague in Madagascar to compare the performance of the analytically derived decision rule to that of machine learning meta-models trained on a stochastic dynamic transmission model.We assess the accuracy of each approach for different combinations of disease dynamics and intervention scenarios.The machine learning meta-models are more accurate predictors of mass prophylaxis cost-effectiveness. However, the simple decision rule is also accurate and may be a preferred substitute in low-resource settings.

Modeling the Cost-Effectiveness of Interventions to Prevent Plague in Madagascar.

Plague (Yersinia pestis) remains endemic in certain parts of the world. We assessed the cost-effectiveness of plague control interventions recommended by the World Health Organization with particular consideration to intervention coverage and timing. We developed a dynamic model of the spread of plague between interacting populations of humans, rats, and fleas and performed a cost-effectiveness analysis calibrated to a 2017 Madagascar outbreak. We assessed three interventions alone and in combination: expanded access to antibiotic treatment with doxycycline, mass distribution of doxycycline prophylaxis, and mass distribution of malathion. We varied intervention timing and coverage levels. We calculated costs, quality-adjusted life years (QALYs), and incremental cost-effectiveness ratios from a healthcare perspective. The preferred intervention, using a cost-effectiveness threshold of $1350/QALY (GDP per capita in Madagascar), was expanded access to antibiotic treatment with doxycycline with 100% coverage starting immediately after the first reported case, gaining 543 QALYs at an incremental cost of $1023/QALY gained. Sensitivity analyses support expanded access to antibiotic treatment and leave open the possibility that mass distribution of doxycycline prophylaxis or mass distribution of malathion could be cost-effective. Our analysis highlights the potential for rapid expansion of access to doxycycline upon recognition of plague outbreaks to cost-effectively prevent future large-scale plague outbreaks and highlights the importance of intervention timing.

Predicting the Effectiveness of Endemic Infectious Disease Control Interventions: The Impact of Mass Action versus Network Model Structure.

BACKGROUND: Analyses of the effectiveness of infectious disease control interventions often rely on dynamic transmission models to simulate intervention effects. We aim to understand how the choice of network or compartmental model can influence estimates of intervention effectiveness in the short and long term for an endemic disease with susceptible and infected states in which infection, once contracted, is lifelong. METHODS: We consider 4 disease models with different permutations of socially connected network versus unstructured contact (mass-action mixing) model and heterogeneous versus homogeneous disease risk. The models have susceptible and infected populations calibrated to the same long-term equilibrium disease prevalence. We consider a simple intervention with varying levels of coverage and efficacy that reduces transmission probabilities. We measure the rate of prevalence decline over the first 365 d after the intervention, long-term equilibrium prevalence, and long-term effective reproduction ratio at equilibrium. RESULTS: Prevalence declined up to 10% faster in homogeneous risk models than heterogeneous risk models. When the disease was not eradicated, the long-term equilibrium disease prevalence was higher in mass-action mixing models than in network models by 40% or more. This difference in long-term equilibrium prevalence between network versus mass-action mixing models was greater than that of heterogeneous versus homogeneous risk models (less than 30%); network models tended to have higher effective reproduction ratios than mass-action mixing models for given combinations of intervention coverage and efficacy. CONCLUSIONS: For interventions with high efficacy and coverage, mass-action mixing models could provide a sufficient estimate of effectiveness, whereas for interventions with low efficacy and coverage, or interventions in which outcomes are measured over short time horizons, predictions from network and mass-action models diverge, highlighting the importance of sensitivity analyses on model structure. HIGHLIGHTS: • We calibrate 4 models-socially connected network versus unstructured contact (mass-action mixing) model and heterogeneous versus homogeneous disease risk-to 10% preintervention disease prevalence.• We measure the short- and long-term intervention effectiveness of all models using the rate of prevalence decline, long-term equilibrium disease prevalence, and effective reproduction ratio.• Generally, in the short term, prevalence declined faster in the homogeneous risk models than in the heterogeneous risk models.• Generally, in the long term, equilibrium disease prevalence was higher in the mass-action mixing models than in the network models, and the effective reproduction ratio was higher in network models than in the mass-action mixing models.

Effectiveness of interventions to reduce COVID-19 transmission in a large urban jail: a model-based analysis.

OBJECTIVES: We aim to estimate the impact of various mitigation strategies on COVID-19 transmission in a US jail beyond those offered in national guidelines. DESIGN: We developed a stochastic dynamic transmission model of COVID-19. SETTING: One anonymous large urban US jail. PARTICIPANTS: Several thousand staff and incarcerated individuals. INTERVENTIONS: There were four intervention phases during the outbreak: the start of the outbreak, depopulation of the jail, increased proportion of people in single cells and asymptomatic testing. These interventions were implemented incrementally and in concert with one another. PRIMARY AND SECONDARY OUTCOME MEASURES: The basic reproduction ratio, R0 , in each phase, as estimated using the next generation method. The fraction of new cases, hospitalisations and deaths averted by these interventions (along with the standard measures of sanitisation, masking and social distancing interventions). RESULTS: For the first outbreak phase, the estimated R0 was 8.44 (95% credible interval (CrI): 5.00 to 13.10), and for the subsequent phases, R0,phase 2 =3.64 (95% CrI: 2.43 to 5.11), R0,phase 3 =1.72 (95% CrI: 1.40 to 2.12) and R0,phase 4 =0.58 (95% CrI: 0.43 to 0.75). In total, the jail's interventions prevented approximately 83% of projected cases, hospitalisations and deaths over 83 days. CONCLUSIONS: Depopulation, single celling and asymptomatic testing within jails can be effective strategies to mitigate COVID-19 transmission in addition to standard public health measures. Decision makers should prioritise reductions in the jail population, single celling and testing asymptomatic populations as additional measures to manage COVID-19 within correctional settings.

Estimation of COVID-19 basic reproduction ratio in a large urban jail in the United States.

PURPOSE: To estimate the basic reproduction ratio () of SARS-CoV-2 inside a correctional facility early in the COVID-19 pandemic. METHODS: We developed a dynamic transmission model for a large, urban jail in the United States. We used the next generation method to determine the basic reproduction ratio We included anonymized data of incarcerated individuals and correctional staff with confirmed COVID-19 infections in our estimation of the basic reproduction ratio () of SARS-CoV-2. RESULTS: The estimated is 8.44 (95% Credible Interval (CrI): 5.00-13.13) for the entire jail. CONCLUSIONS: The high of SARS-CoV-2 in a large urban jail highlights the importance of including correctional facilities in public health strategies for COVID-19. In the absence of more aggressive mitigation strategies, correctional facilities will continue to contribute to community infections.