Journal of Medical Internet Research - International Scientific Journal for Medical Research, Information and Communication on the Internet

Machine learning has become an increasingly powerful tool for solving complex problems, and its application in public health has been underutilized. The objective of this study is to test the efficacy of a machine-learned model of foodborne illness detection in a real-world setting. To this end, we built FINDER, a machine-learned model for real-time detection of foodborne illness using anonymous and aggregated web search and location data. We computed the fraction of people who visited a particular restaurant and later searched for terms indicative of food poisoning to identify potentially unsafe restaurants. We used this information to focus restaurant inspections in two cities and demonstrated that FINDER improves the accuracy of health inspections; restaurants identified by FINDER are 3.1 times as likely to be deemed unsafe during the inspection as restaurants identified by existing methods. Additionally, FINDER enables us to ascertain previously intractable epidemiological information, for example, in 38% of cases the restaurant potentially causing food poisoning was not the last one visited, which may explain the lower precision of complaint-based inspections. We found that FINDER is able to reliably identify restaurants that have an active lapse in food safety, allowing for implementation of corrective actions that would prevent the potential spread of foodborne illness.

Context: Foodborne illness affects 1 in 4 US residents each year. Few of those sickened seek medical care or report the illness to public health authorities, complicating prevention efforts. Citizens who report illness identify food establishments with more serious and critical violations than found by regular inspections. New media sources, including online restaurant reviews and social media postings, have the potential to improve reporting. Objective: We implemented a Web-based Dashboard (HealthMap Foodborne Dashboard) to identify and respond to tweets about food poisoning from St Louis City residents. Design and Setting: This report examines the performance of the Dashboard in its first 7 months after implementation in the City of St Louis Department of Health. Main Outcome Measures: We examined the number of relevant tweets captured and replied to, the number of foodborne illness reports received as a result of the new process, and the results of restaurant inspections following each report. Results: In its first 7 months (October 2015-May 2016), the Dashboard captured 193 relevant tweets. Our replies to relevant tweets resulted in more filed reports than several previously existing foodborne illness reporting mechanisms in St Louis during the same time frame. The proportion of restaurants with food safety violations was not statistically different (P = .60) in restaurants inspected after reports from the Dashboard compared with those inspected following reports through other mechanisms. Conclusion: The Dashboard differs from other citizen engagement mechanisms in its use of current data, allowing direct interaction with constituents on issues when relevant to the constituent to provide time-sensitive education and mobilizing information. In doing so, the Dashboard technology has potential for improving foodborne illness reporting and can be implemented in other areas to improve response to public health issues such as suicidality, spread of Zika virus infection, and hospital quality.

How does a machine learning model compare with logistic regression in predicting childhood lead poisoning during infancy? This prognostic study of 6812 children in a Women, Infants, and Children cohort in Chicago, Illinois, used blood lead level surveillance data, housing characteristics, and demographic characteristics to predict lead poisoning risk. Using predictors from a range of spatiotemporal scales, a random forest model significantly outperformed a parsimonious logistic regression. These findings suggest that machine learning can be used to more precisely predict risk of elevated blood lead levels to guide allocation of prevention resources to children most at risk. Childhood lead poisoning causes irreversible neurobehavioral deficits, but current practice is secondary prevention. To validate a machine learning (random forest) prediction model of elevated blood lead levels (EBLLs) by comparison with a parsimonious logistic regression. This prognostic study for temporal validation of multivariable prediction models used data from the Women, Infants, and Children (WIC) program of the Chicago Department of Public Health. Participants included a development cohort of children born from January 1, 2007, to December 31, 2012, and a validation WIC cohort born from January 1 to December 31, 2013. Blood lead levels were measured until December 31, 2018. Data were analyzed from January 1 to October 31, 2019. Blood lead level test results; lead investigation findings; housing characteristics, permits, and violations; and demographic variables. Incident EBLL (≥6 μg/dL). Models were assessed using the area under the receiver operating characteristic curve (AUC) and confusion matrix metrics (positive predictive value, sensitivity, and specificity) at various thresholds. Among 6812 children in the WIC validation cohort, 3451 (50.7%) were female, 3057 (44.9%) were Hispanic, 2804 (41.2%) were non-Hispanic Black, 458 (6.7%) were non-Hispanic White, and 442 (6.5%) were Asian (mean [SD] age, 5.5 [0.3] years). The median year of housing construction was 1919 (interquartile range, 1903-1948). Random forest AUC was 0.69 compared with 0.64 for logistic regression (difference, 0.05; 95% CI, 0.02-0.08). When predicting the 5% of children at highest risk to have EBLLs, random forest and logistic regression models had positive predictive values of 15.5% and 7.8%, respectively (difference, 7.7%; 95% CI, 3.7%-11.3%), sensitivity of 16.2% and 8.1%, respectively (difference, 8.1%; 95% CI, 3.9%-11.7%), and specificity of 95.5% and 95.1% (difference, 0.4%; 95% CI, 0.0%-0.7%). The machine learning model outperformed regression in predicting childhood lead poisoning, especially in identifying children at highest risk. Such a model could be used to target the allocation of lead poisoning prevention resources to these children.

Lead poisoning is a major public health problem that affects hundreds of thousands of children in the United States every year. A common approach to identifying lead hazards is to test all children for elevated blood lead levels and then investigate and remediate the homes of children with elevated tests. This can prevent exposure to lead of future residents, but only after a child has been poisoned. This paper describes joint work with the Chicago Department of Public Health (CDPH) in which we build a model that predicts the risk of a child to being poisoned so that an intervention can take place before that happens. Using two decades of blood lead level tests, home lead inspections, property value assessments, and census data, our model allows inspectors to prioritize houses on an intractably long list of potential hazards and identify children who are at the highest risk. This work has been described by CDPH as pioneering in the use of machine learning and predictive analytics in public health and has the potential to have a significant impact on both health and economic outcomes for communities across the US.