Medical Decision Making

Even after a physician recommendation, many people remain unscreened for colorectal cancer (CRC). The proliferation of electronic health records (EHRs) and tethered online portals may afford new opportunities to embed patient-facing interventions within clinic workflows and engage patients following a physician recommendation for care. We evaluated the effectiveness of a patient-facing intervention designed to complement physician office-based recommendations for CRC screening.

Using a 2-arm pragmatic, randomized clinical trial, we evaluated the intervention’s effect on CRC screening use as documented in the EHR (primary outcome) and the extent to which the intervention reached the target population. Trial participants were insured, aged 50 to 75 y, with a physician recommendation for CRC screening. Typical EHR functionalities, including patient registries, health maintenance flags, best practice alerts, and secure messaging, were used to support research-related activities and deliver the intervention to enrolled patients.

A total of 1,825 adults consented to trial participation, of whom 78% completed a baseline survey and were exposed to the intervention. Most trial participants (>80%) indicated an intent to be screened on the baseline survey, and 65% were screened at follow-up, with no significant differences by study arm. One-third of eligible patients were sent a secure message. Among those, more than three-quarters accessed study material.

By leveraging common EHR functionalities, we integrated a patient-facing intervention within clinic workflows. Despite practice integration, the intervention did not improve screening use, likely in part due to portal-based interventions not reaching those for whom the intervention may be most effective.

Embedding patient-facing interventions within the EHR enabled practice integration but may minimize program effectiveness by missing important segments of the patient population.

This article describes the development of a system-based data platform for research developed to provide a detailed picture of the characteristics of the Urgent and Emergency Care system in 1 region of the United Kingdom.

CUREd is an integrated research data platform that describes the urgent and emergency care system in 1 region of the United Kingdom on almost 30 million patient contacts within the system. We describe regulatory approvals required, data acquisition, cleaning, and linkage.

The data platform covers 2011 to 2017 for 14 acute National Health Service (NHS) Hospital Trusts, 1 ambulance service, the national telephone advice service (NHS 111), and 19 emergency departments. We describe 3 analyses undertaken: 1) Analyzing triage patterns from the NHS 111 telephone helpline using routine data linked to other urgent care services, we found that the current triage algorithms have high rates of misclassifying calls. 2) Applying an algorithm to consistently identify avoidable attendances for pediatric patients, we identified 21% of pediatric attendances to the emergency department as avoidable. 3) Using complex systems analysis to examine patterns of frequent attendance in urgent care, we found that frequent attendance is stable over time but varies by individual patient. This implies that frequent attendance is more likely to be a function of the system overall.

We describe the processes necessary to produce research-ready data that link care across the components of the urgent and emergency care system. Making the use of routine data commonplace will require partnership between the collectors, owners, and guardians of the data and researchers and technical teams.

Electronic health records (EHRs) offer opportunities for comparative effectiveness research to inform decision making. However, to provide useful evidence, these studies must address confounding and treatment effect heterogeneity according to unmeasured prognostic factors. Local instrumental variable (LIV) methods can help studies address these challenges, but have yet to be applied to EHR data. This article critically examines a LIV approach to evaluate the cost-effectiveness of emergency surgery (ES) for common acute conditions from EHRs.

This article uses hospital episodes statistics (HES) data for emergency hospital admissions with acute appendicitis, diverticular disease, and abdominal wall hernia to 175 acute hospitals in England from 2010 to 2019. For each emergency admission, the instrumental variable for ES receipt was each hospital’s ES rate in the year preceding the emergency admission. The LIV approach provided individual-level estimates of the incremental quality-adjusted life-years, costs and net monetary benefit of ES, which were aggregated to the overall population and subpopulations of interest, and contrasted with those from traditional IV and risk-adjustment approaches.

The study included 268,144 (appendicitis), 138,869 (diverticular disease), and 106,432 (hernia) patients. The instrument was found to be strong and to minimize covariate imbalance. For diverticular disease, the results differed by method; although the traditional approaches reported that, overall, ES was not cost-effective, the LIV approach reported that ES was cost-effective but with wide statistical uncertainty. For all 3 conditions, the LIV approach found heterogeneity in the cost-effectiveness estimates across population subgroups: in particular, ES was not cost-effective for patients with severe levels of frailty.

EHRs can be combined with LIV methods to provide evidence on the cost-effectiveness of routinely provided interventions, while fully recognizing heterogeneity.

Electronic health records (EHRs) provide researchers with abundant sample sizes, detailed clinical data, and other advantages for performing high-quality observational health research on diverse populations. We review and demonstrate strategies for the design and analysis of cohort studies on neighborhood diversity and health, including evaluation of the effects of race, ethnicity, and neighborhood socioeconomic position on disease prevalence and health outcomes, using localized EHR data.

Design strategies include integrating and harmonizing EHR data across multiple local health systems and defining the population(s) of interest and cohort extraction procedures for a given analysis based on the goal(s) of the study. Analysis strategies address inferential goals, including the mechanistic study of social risks, statistical adjustment for differences in distributions of social and neighborhood-level characteristics between available EHR data and the underlying local population, and inference on individual neighborhoods. We provide analyses of local variation in mortality rates within Cuyahoga County, Ohio.

When the goal of the analysis is to adjust EHR samples to be more representative of local populations, sampling and weighting are effective. Causal mediation analysis can inform effects of racism (through racial residential segregation) on health outcomes. Spatial analysis is appealing for large-scale EHR data as a means for studying heterogeneity among neighborhoods even at a given level of overall neighborhood disadvantage.

The methods described are a starting point for robust EHR-derived cohort analysis of diverse populations. The methods offer opportunities for researchers to pursue detailed analyses of current and historical underlying circumstances of social policy and inequality. Investigators can employ combinations of these methods to achieve greater robustness of results.

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.

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.

We assume homogeneous mixing and no negative side effects due to antibiotic prophylaxis.

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.

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.

Policy makers are facing more complicated challenges to balance saving lives and economic development in the post-vaccination era during a pandemic. Epidemic simulation models and pandemic control methods are designed to tackle this problem. However, most of the existing approaches cannot be applied to real-world cases due to the lack of adaptability to new scenarios and micro representational ability (especially for system dynamics models), the huge computation demand, and the inefficient use of historical information.

We propose a novel Pandemic Control decision making framework via large-scale Agent-based modeling and deep Reinforcement learning (PaCAR) to search optimal control policies that can simultaneously minimize the spread of infection and the government restrictions. In the framework, we develop a new large-scale agent-based simulator with vaccine settings implemented to be calibrated and serve as a realistic environment for a city or a state. We also design a novel reinforcement learning architecture applicable to the pandemic control problem, with a reward carefully designed by the net monetary benefit framework and a sequence learning network to extract information from the sequential epidemiological observations, such as number of cases, vaccination, and so forth.

Our approach outperforms the baselines designed by experts or adopted by real-world governments and is flexible in dealing with different variants, such as Alpha and Delta in COVID-19. PaCAR succeeds in controlling the pandemic with the lowest economic costs and relatively short epidemic duration and few cases. We further conduct extensive experiments to analyze the reasoning behind the resulting policy sequence and try to conclude this as an informative reference for policy makers in the post-vaccination era of COVID-19 and beyond.

The modeling of economic costs, which are directly estimated by the level of government restrictions, is rather simple. This article mainly focuses on several specific control methods and single-wave pandemic control.

The proposed framework PaCAR can offer adaptive pandemic control recommendations on different variants and population sizes. Intelligent pandemic control empowered by artificial intelligence may help us make it through the current COVID-19 and other possible pandemics in the future with less cost both of lives and economy.

During the COVID-19 pandemic, the world witnessed a partisan segregation of beliefs toward the global health crisis and its management. Politically motivated reasoning, the tendency to interpret information in accordance with individual motives to protect valued beliefs rather than objectively considering the facts, could represent a key process involved in the polarization of attitudes. The objective of this study was to explore politically motivated reasoning when participants assess information regarding COVID-19.

We carried out a preregistered online experiment using a diverse sample (N = 1,500) from the United States. Both Republicans and Democrats assessed the same COVID-19–related information about the health effects of lockdowns, social distancing, vaccination, hydroxychloroquine, and wearing face masks.

At odds with our prestated hypothesis, we found no evidence in line with politically motivated reasoning when interpreting numerical information about COVID-19. Moreover, we found no evidence supporting the idea that numeric ability or cognitive sophistication bolster politically motivated reasoning in the case of COVID-19. Instead, our findings suggest that participants base their assessment on prior beliefs of the matter.

Our findings suggest that politically polarized attitudes toward COVID-19 are more likely to be driven by lack of reasoning than politically motivated reasoning—a finding that opens potential avenues for combating political polarization about important health care topics.