Use of machine learning to predict hypertension-related complication outcomes of varying severity

ObjectiveA challenge in hypertension-related risk management is identifying which people are likely to develop future complications. To address this, we present administrative-claims based predictive models for hypertension-related complications. Materials and MethodsWe used a national database to select 1,767,559 people with hypertension and extracted 112 features from past claims data based on their ability to predict hypertension complications in the next year. Complications affecting kidney, brain, and heart were grouped by clinical severity into three stages. Extreme gradient boosting binary classifiers for each stage were trained and tuned on 75% of the data, and performance on predicting outcomes for the remaining data and an independent dataset was evaluated. ResultsIn the cohort under study, 6%, 17%, and 7% of people experienced a hypertension-related complication of stage 1, stage 2, or stage 3 severity, respectively. On an independent dataset, models for all three stages performed competitively with other published algorithms by the most commonly reported metric, area under the receiver operating characteristic curve, which ranged from 0.82-0.89. Features that were important across all models for predictions included total medical cost, cost related to hypertension, age, and number of outpatient visits. DiscussionThe model for stage 1 complications, such as left ventricular hypertrophy and retinopathy, is in contrast to other offerings in the field, which focus on more serious issues such as heart failure and stroke, and affords unique opportunities to intervene during earlier stages. ConclusionPredictive analytics for hypertension outcomes can be leveraged to help mitigate the immense healthcare burden of uncontrolled hypertension. LAY SUMMARYAs the leading preventable risk factor for morbidity and mortality in the world, identifying which people with hypertension are likely to exacerbate is critically important for development of effective intervention strategies. Here we present a suite of predictive models that can predict future risk of development of hypertension-related complications. To have utility for triaging as well as identifying mild cases before they progress to critical end phases, the models predict three different stages of severity of hypertension-related complications. Our algorithms utilize variables calculated for the most recent 12 months, and predict probability of a hypertension-related complication for the next 12 months using administrative claims as the data source. Because the types of complications that have been analyzed can also result from comorbidities besides hypertension, such as diabetes and hyperlipidemia, these diagnoses are included as variables. Other variables pertain to demographic characteristics, prescription information, relevant procedures, and utilization patterns. Overall, all three models exhibited strong predictive performance. The ability to use straightforward variables found in claims data to predict future risk of disease-related complications, complemented with targeted clinical intervention strategies, has the potential to reduce cost of care and improve health outcomes for the many people living with hypertension.