Dynamic treatment selection and modification for personalised blood pressure therapy using a Markov decision process...

Area of Science:

• Cardiovascular Medicine
• Health Economics
• Computational Biology

Background:

• Personalized medicine aims to tailor treatments to individual patient characteristics and preferences.
• Current blood pressure treatment guidelines may not fully account for individual variations in medication response and side effect profiles.
• Optimizing hypertension management is crucial for reducing cardiovascular disease burden.

Purpose of the Study:

• To develop an automated strategy for selecting and modifying blood pressure treatments based on individual patient characteristics.
• To incorporate patient-specific likelihoods of benefiting from different medications and dosages.
• To account for the potential severity and impact of side effects in personalized treatment plans.

Main Methods:

• A Markov decision process (MDP) model was developed to estimate optimal blood pressure treatment strategies.
• Meta-analytic data and individual patient characteristics were incorporated into the MDP model.
• The MDP strategy was compared against current US blood pressure treatment guidelines (JNC8) and an intensive JNC8 variant using a microsimulation model.
• Patient demographics, cardiovascular disease risk factors, and side effect probabilities were sampled from the National Health and Nutrition Examination Survey (2003-2014).

Main Results:

• The MDP-based treatment (MDPT) strategy was found to be cost-saving from a societal perspective, with estimated discounted savings of US$1187 per capita.
• The MDPT strategy resulted in an estimated discounted gain of 0.06 quality-adjusted life-years (QALYs) per capita compared to the JNC8 guideline.
• QALY gains were primarily attributed to reductions in severe side effects associated with optimized treatment doses.
• The Intensive JNC8 strategy was dominated by the MDPT strategy, indicating superior performance.

Conclusions:

• An MDP-based approach offers a data-driven method for personalizing blood pressure treatment.
• This strategy can improve overall population health by optimizing medication selection and management.
• The findings suggest that personalized medicine approaches, like the developed MDP, can outperform current guideline-based strategies in hypertension management.