Predicting Adverse Outcomes in At-Risk Populations with Machine Learning Methods - PhD thesis summary

Aug 5, 2026

Upload by Quinn

Canada’s publicly funded health system serves a heterogeneous population, including subgroups that generate disproportionate utilization and adverse outcomes. This PhD evaluates machine-learning approaches to identify high-risk individuals within prescription opioid users, older adults taking benzodiazepines, and people with heart failure using administrative health data. The work assesses prediction performance and decision value for health system planners, focusing on admissions and deaths and estimating potential cost savings from ML-assisted programs. Key results compare ML against guideline- or history-based methods, highlight limits when predictive signal is weak, and discuss data and governance needs.

LanguageEnglish

CategoryResearch & Report

FormatPDF

Pages238

Reading Time10 h 0 min

AccessFree to download

Abstract

Thesis objectives and studied at-risk groups

Primary outcomes and evaluation goals

Main findings by condition

Opioid 30-day adverse outcomes

Heart failure readmissions

Benzodiazepine risk in older adults

Opioid-related cost and intervention implications

Conclusion and future research directions

Data sources, bias, and ML transparency

Predicting Adverse Outcomes in At-Risk Populations with Machine Learning Methods by Vishal Sharma A thesis submitted in partial fulfillment of the requirements for the degree of Doctor of Philosophy in Epidemiology School of Public Health University of Alberta © Vishal Sharma, 2023 Abstract Canada has a publicly funded health system and heterogeneous population. There are segments of this population which account for substantial health care utilization and adverse outcomes. Machine learning (ML) approaches can assist public health intervention programs to mitigate health system costs and improve patient outcomes, particularly for segments within Canadian society that qualify as at-risk. The specific ones to be studied in this PhD are prescription opioid users, older adults taking benzodiazepines, and people with heart failure (HF) . Identifying high risk individuals within these segments using ML methods trained on administrative health data as well as assessing prediction performance and value to inform health system planners are the objectives of this PhD program. This PhD studied outcomes related to admissions and deaths and presented findings on potential cost savings of ML assisted programs. The main findings in this thesis were: 1. Machine-learning classifiers, especially incorporating hospitalization and physician claims data, have better predictive performance compared to guideline or prescription history only approaches when predicting 30-day risk of adverse outcomes pursuant to an opioid dispensation. Prescription monitoring programs and health departments with access to administrative data can use ML classifiers to effectively identify those at higher risk compared to current guideline-based approaches, 2. Despite predicting readmissions in patients with HF better than the LaCE, even the best ML model trained on administrative health data (XGBoost) did not provide substantially informative prediction performance as it only generated a moderate shift from pre to post-test probability. Health systems wishing to deploy such a tool should consider training ML models with additional data. Adding other techniques like Natural Language Processing, along with ML, to use other clinical information (like chart notes) might improve prediction performance, 3. Developing ML models using only administrative health data may not provide health regulators with sufficient informative predictions to use as decision aids for potential interventions, especially if considering daily or quarterly classifications of benzodiazepine risks in older adults. ML models may be informative for this context if yearly classifications are preferred. Health regulators should have access to other types of data to improve ML prediction, and 4. Prescription drug monitoring programs can use ML classifiers to identify patients at risk of adverse outcomes from opioids and potentially reduce health-care costs by intervening on high-ranked predictions. Better access to available administrative and clinical data could improve the prediction performance of ML classifiers, especially if probability thresholds are important, and thus expand opioid stewardship efforts and further reduce costs. In conclusion, the findings suggest that ML methods may demonstrate value in opioid stewardship programs with limited benefits in predicting adverse outcomes in older adults taking benzodiazepines and readmissions in people with HF. Health systems wishing to integrate ML into their program planning may benefit from additional sources of data to train ML models. Data governance, bias and ML transparency are key issues requiring future research. Preface This thesis is an original work by Vishal Sharma. The studies presented in this thesis received research ethics approval from the University of Alberta Research Ethics Board under the following approvals: Pro00083807_AME1, Pro00097809, Pro000838807_AME6 and Pro00083807 . A version of Ch

FAQs

Which at-risk populations are studied in this thesis?

The thesis studies prescription opioid users, older adults taking benzodiazepines, and people with heart failure (HF).

What data sources are used to train machine-learning models?

Models are trained on administrative health data, and the thesis discusses the need for additional data when predictive performance is limited.

How do ML models perform for predicting adverse outcomes after opioid dispensing?

Machine-learning classifiers incorporating hospitalization and physician claims data show better predictive performance than guideline- or prescription-history-only approaches for the 30-day risk of adverse outcomes.

What limitations are observed for heart failure readmission prediction?

Even the best model (XGBoost) shows only moderate shifts in prediction probabilities, and improving performance may require training with additional data and potentially incorporating other techniques such as NLP for clinical notes.

Aug 5, 2026

Upload by Quinn