The Potential for a Benefit-Estimating Algorithm to Improve Recommendations for Preventive Services: Comparing...

Area of Science:

• Clinical Informatics
• Preventive Medicine
• Health Services Research

Background:

• Prioritizing preventive services is crucial for patient health outcomes.
• Individualized approaches can enhance the effectiveness of preventive care.
• Primary care providers (PCPs) face time constraints in clinical decision-making.

Purpose of the Study:

• To compare an algorithm for Individualized Precision Prevention (IPP) with PCPs' prioritization of preventive services.
• To assess the concordance between algorithmic recommendations and expert clinical judgment.
• To evaluate the potential of IPP to support PCP decision-making in preventive care.

Main Methods:

• Developed 12 patient care scenarios focused on preventive services.
• Recruited 40 PCPs to rank preventive service priorities for each scenario.
• Compared algorithm-generated rankings with PCP rankings using length-dependent rank-biased overlap (LDRBO).

Main Results:

• A moderate concordance (mean LDRBO = 0.45) was found between the IPP algorithm's rankings and PCPs' consensus rankings.
• The algorithm suggested preventive services not consistently ranked in the PCPs' top 3 for half of the scenarios.
• This indicates potential for the algorithm to highlight beneficial services PCPs may overlook.

Conclusions:

• The IPP algorithm aligns with clinical expert judgment and can assist PCPs in prioritizing preventive services.
• The moderate concordance suggests the algorithm can offer valuable, data-driven advice for preventive care.
• IPP has the potential to improve shared decision-making between providers and patients regarding preventive services.