The optimal allocation of Covid-19 vaccines.

We develop a simple model of vaccine prioritization for a potential pandemic. We illustrate how the model applies to the case of Covid-19, using an early 2020 primitive estimate of occupation-based exposure risks and age-based infection fatality rates. Even based on primitive estimates the vaccine distribution strongly emphasizes age-based mortality risk rather than occupation-based exposure risk. Among others, our result suggests that 50-year-old food-processing workers and 60-year-old financial advisors should have been equally prioritized. We also find that the priorities minimally change when certain populations' exposure risks are altered by targeted stay-at-home orders or call-up of essential workers.

Solving Homelessness from a Complex Systems Perspective: Insights for Prevention Responses.

Homelessness represents an enduring public health threat facing communities across the developed world. Children, families, and marginalized adults face life course implications of housing insecurity, while communities struggle to address the extensive array of needs within heterogeneous homeless populations. Trends in homelessness remain stubbornly high despite policy initiatives to end homelessness. A complex systems perspective provides insights into the dynamics underlying coordinated responses to homelessness. A constant demand for housing assistance strains service delivery, while prevention efforts remain inconsistently implemented in most countries. Feedback processes challenge efficient service delivery. A system dynamics model tests assumptions of policy interventions for ending homelessness. Simulations suggest that prevention provides a leverage point within the system; small efficiencies in keeping people housed yield disproportionately large reductions in homelessness. A need exists for policies that ensure reliable delivery of coordinated prevention efforts. A complex systems approach identifies capacities and constraints for sustainably solving homelessness.

Automatic detection of eating disorder-related social media posts that could benefit from a mental health intervention.

OBJECTIVE: Online forums allow people to semi-anonymously discuss their struggles, often leading to greater honesty. This characteristic makes forums valuable for identifying users in need of immediate help from mental health professionals. Because it would be impractical to manually review every post on a forum to identify users in need of urgent help, there may be value to developing algorithms for automatically detecting posts reflecting a heightened risk of imminent plans to engage in disordered behaviors. METHOD: Five natural language processing techniques (tools to perform computational text analysis) were used on a data set of 4,812 posts obtained from six eating disorder-related subreddits. Two licensed clinical psychologists labeled 53 of these posts, deciding whether or not the content of the post indicated that its author needed immediate professional help. The remaining 4,759 posts were unlabeled. RESULTS: Each of the five techniques ranked the 50 posts most likely to be intervention-worthy (the "top-50"). The two most accurate detection techniques had an error rate of 4% for their respective top-50. DISCUSSION: This article demonstrates the feasibility of automatically detecting-with only a few dozen labeled examples-the posts of individuals in need of immediate mental health support for an eating disorder.

Clinical risk prediction using language models: benefits and considerations.

OBJECTIVE: The use of electronic health records (EHRs) for clinical risk prediction is on the rise. However, in many practical settings, the limited availability of task-specific EHR data can restrict the application of standard machine learning pipelines. In this study, we investigate the potential of leveraging language models (LMs) as a means to incorporate supplementary domain knowledge for improving the performance of various EHR-based risk prediction tasks. METHODS: We propose two novel LM-based methods, namely "LLaMA2-EHR" and "Sent-e-Med." Our focus is on utilizing the textual descriptions within structured EHRs to make risk predictions about future diagnoses. We conduct a comprehensive comparison with previous approaches across various data types and sizes. RESULTS: Experiments across 6 different methods and 3 separate risk prediction tasks reveal that employing LMs to represent structured EHRs, such as diagnostic histories, results in significant performance improvements when evaluated using standard metrics such as area under the receiver operating characteristic (ROC) curve and precision-recall (PR) curve. Additionally, they offer benefits such as few-shot learning, the ability to handle previously unseen medical concepts, and adaptability to various medical vocabularies. However, it is noteworthy that outcomes may exhibit sensitivity to a specific prompt. CONCLUSION: LMs encompass extensive embedded knowledge, making them valuable for the analysis of EHRs in the context of risk prediction. Nevertheless, it is important to exercise caution in their application, as ongoing safety concerns related to LMs persist and require continuous consideration.

Community- and data-driven homelessness prevention and service delivery: optimizing for equity.

OBJECTIVE: The study tests a community- and data-driven approach to homelessness prevention. Federal policies call for efficient and equitable local responses to homelessness. However, the overwhelming demand for limited homeless assistance is challenging without empirically supported decision-making tools and raises questions of whom to serve with scarce resources. MATERIALS AND METHODS: System-wide administrative records capture the delivery of an array of homeless services (prevention, shelter, short-term housing, supportive housing) and whether households reenter the system within 2 years. Counterfactual machine learning identifies which service most likely prevents reentry for each household. Based on community input, predictions are aggregated for subpopulations of interest (race/ethnicity, gender, families, youth, and health conditions) to generate transparent prioritization rules for whom to serve first. Simulations of households entering the system during the study period evaluate whether reallocating services based on prioritization rules compared with services-as-usual. RESULTS: Homelessness prevention benefited households who could access it, while differential effects exist for homeless households that partially align with community interests. Households with comorbid health conditions avoid homelessness most when provided longer-term supportive housing, and families with children fare best in short-term rentals. No additional differential effects existed for intersectional subgroups. Prioritization rules reduce community-wide homelessness in simulations. Moreover, prioritization mitigated observed reentry disparities for female and unaccompanied youth without excluding Black and families with children. DISCUSSION: Leveraging administrative records with machine learning supplements local decision-making and enables ongoing evaluation of data- and equity-driven homeless services. CONCLUSIONS: Community- and data-driven prioritization rules more equitably target scarce homeless resources.

Identifying relevant data for a biological database: handcrafted rules versus machine learning.

With well over 1,000 specialized biological databases in use today, the task of automatically identifying novel, relevant data for such databases is increasingly important. In this paper, we describe practical machine learning approaches for identifying MEDLINE documents and Swiss-Prot/TrEMBL protein records, for incorporation into a specialized biological database of transport proteins named TCDB. We show that both learning approaches outperform rules created by hand by a human expert. As one of the first case studies involving two different approaches to updating a deployed database, both the methods compared and the results will be of interest to curators of many specialized databases.