Using Machine Learning to Examine Medication Adherence Thresholds and Risk of Hospitalization. Issue 8 (August 2015)

• Record Type: Journal Article
• Authors: Lo-Ciganic, Wei-Hsuan; Donohue, Julie M.; Thorpe, Joshua M.; Perera, Subashan; Thorpe, Carolyn T.; Marcum, Zachary A.; Gellad, Walid F.
• Is Part Of: Medical care. Volume 53:Issue 8(2015)
• Publication Date: 2015-08
• Subjects: Economics, Medical -- Periodicals; Insurance, Health -- Periodicals; Santé, Services de -- Administration -- Périodiques; Soins médicaux -- Périodiques; Medical economics -- Periodicals; Health insurance -- Periodicals; Medical economics -- United States -- Periodicals; Health insurance -- United States -- Periodicals; Comprehensive Health Care -- Periodicals; Personal Health Services -- Periodicals; Gezondheidszorg; Économie de la santé -- Périodiques; Santé, Services de -- Périodiques; Health insurance; Medical economics; United States; Periodicals; 362.10973
• DOI: 10.1097/MLR.0000000000000394
• Languages: English
• ISSNs: 0025-7079

Abstract

<abstract> <title> <x xml:space="preserve">Abstract</x> </title> <sec> <title>Background:</title> <p>Quality improvement efforts are frequently tied to patients achieving ≥80% medication adherence. However, there is little empirical evidence that this threshold optimally predicts important health outcomes.</p> </sec> <sec> <title>Objective:</title> <p>To apply machine learning to examine how adherence to oral hypoglycemic medications is associated with avoidance of hospitalizations, and to identify adherence thresholds for optimal discrimination of hospitalization risk.</p> </sec> <sec> <title>Methods:</title> <p>A retrospective cohort study of 33, 130 non–dual-eligible Medicaid enrollees with type 2 diabetes. We randomly selected 90% of the cohort (training sample) to develop the prediction algorithm and used the remaining (testing sample) for validation. We applied random survival forests to identify predictors for hospitalization and fit survival trees to empirically derive adherence thresholds that best discriminate hospitalization risk, using the proportion of days covered (PDC).</p> </sec> <sec> <title>Outcomes:</title> <p>Time to first all-cause and diabetes-related hospitalization.</p> </sec> <sec> <title>Results:</title> <p>The training and testing samples had similar characteristics (mean age, 48 y; 67% female; mean PDC=0.65). We identified 8 important predictors of all-cause hospitalizations (rank in order): prior hospitalizations/emergency department visit, number of prescriptions, diabetes complications, insulin use, PDC, number of prescribers, Elixhauser index, and eligibility category. The adherence thresholds most discriminating for risk of all-cause hospitalization varied from 46% to 94% according to patient health and medication complexity. PDC was not predictive of hospitalizations in the healthiest or most complex patient subgroups.</p> </sec> <sec> <title>Conclusions:</title> <p>Adherence thresholds most discriminating of hospitalization risk were not uniformly 80%. Machine-learning approaches may be valuable to identify appropriate patient-specific adherence thresholds for measuring quality of care and targeting nonadherent patients for intervention.</p> </sec> </abstract>