Using Computational Approaches to Improve Risk-Stratified Patient Management: Rationale and Methods

Abstract

BackgroundChronic diseases affect 52% of Americans and consume 86% of health care costs. A small portion of patients consume most health care resources and costs. More intensive patient management strategies, such as case management, are usually more effective at improving health outcomes, but are also more expensive. To use limited resources efficiently, risk stratification is commonly used in managing patients with chronic diseases, such as asthma, chronic obstructive pulmonary disease, diabetes, and heart disease. Patients are stratified based on predicted risk with patients at higher risk given more intensive care. The current risk-stratified patient management approach has 3 limitations resulting in many patients not receiving the most appropriate care, unnecessarily increased costs, and suboptimal health outcomes. First, using predictive models for health outcomes and costs is currently the best method for forecasting individual patient’s risk. Yet, accuracy of predictive models remains poor causing many patients to be misstratified. If an existing model were used to identify candidate patients for case management, enrollment would miss more than half of those who would benefit most, but include others unlikely to benefit, wasting limited resources. Existing models have been developed under the assumption that patient characteristics primarily influence outcomes and costs, leaving physician characteristics out of the models. In reality, both characteristics have an impact. Second, existing models usually give neither an explanation why a particular patient is predicted to be at high risk nor suggestions on interventions tailored to the patient’s specific case. As a result, many high-risk patients miss some suitable interventions. Third, thresholds for risk strata are suboptimal and determined heuristically with no quality guarantee.ObjectiveThe purpose of this study is to improve risk-stratified patient management so that more patients will receive the most appropriate care.MethodsThis study will (1) combine patient, physician profile, and environmental variable features to improve prediction accuracy of individual patient health outcomes and costs; (2) develop the first algorithm to explain prediction results and suggest tailored interventions; (3) develop the first algorithm to compute optimal thresholds for risk strata; and (4) conduct simulations to estimate outcomes of risk-stratified patient management for various configurations. The proposed techniques will be demonstrated on a test case of asthma patients.ResultsWe are currently in the process of extracting clinical and administrative data from an integrated health care system’s enterprise data warehouse. We plan to complete this study in approximately 5 years.ConclusionsMethods developed in this study will help transform risk-stratified patient management for better clinical outcomes, higher patient satisfaction and quality of life, reduced health care use, and lower costs.

Automatically explaining machine learning prediction results: a demonstration on type 2 diabetes risk prediction

“…We first review the draft automatic explanation method for machine learning prediction results outlined in our paper [2].…”

“…Our presentation is more general than that in our paper [2], which focuses on asthma outcomes. Our presentation also includes many additional details, insights, and rationales beyond those described in our paper [2]. We will optimize and complete the method in the "Optimizations and refinement" section.…”

“…Usually, a large number of association rules are learned from historical data even after removing redundant rules [19][20][21]24]. To avoid overwhelming the users of the predictive modeling tool for the healthcare application, our previous paper [2] uses three additional techniques to further prune rules. First, only features that the first model uses to make predictions are considered in rule construction.…”

“…Our previous paper [2] only outlined the main ideas of our automatic explanation method for machine learning prediction results, but did not complete the method. Without additional optimization and refinement, this method will not perform satisfactorily.…”

“…At present, several explanation methods for machine learning predictive models exist, but they are model specific and decrease accuracy [7,11,12]. Recently, we proposed the first draft method for automatically explaining results for any machine learning predictive model without degrading accuracy [2]. Our work [2] outlined the method's main ideas, but neither completed the method nor conducted a computer coding implementation.…”