Mind the Gap: Evaluating the Representativeness of Quantitative Medical Language Reasoning LLM Benchmarks for African Disease Burdens

TL;DR

This paper evaluates the representativeness of medical language model benchmarks for African disease burdens, highlighting significant underrepresentation of diseases prevalent in Africa and proposing a guideline-based benchmark to improve evaluation relevance.

Contribution

It introduces Alama Health QA, a regionally curated benchmark aligned with African clinical guidelines, addressing the gap in existing benchmarks.

Findings

Alama Health QA captures over 40% of NTD mentions in evaluated corpora.

Global benchmarks lack representation of key African diseases like sickle cell.

Existing benchmarks often do not align with regional clinical guidelines.

Abstract

Introduction: Existing medical LLM benchmarks largely reflect examination syllabi and disease profiles from high income settings, raising questions about their validity for African deployment where malaria, HIV, TB, sickle cell disease and other neglected tropical diseases (NTDs) dominate burden and national guidelines drive care. Methodology: We systematically reviewed 31 quantitative LLM evaluation papers (Jan 2019 May 2025) identifying 19 English medical QA benchmarks. Alama Health QA was developed using a retrieval augmented generation framework anchored on the Kenyan Clinical Practice Guidelines. Six widely used sets (AfriMedQA, MMLUMedical, PubMedQA, MedMCQA, MedQAUSMLE, and guideline grounded Alama Health QA) underwent harmonized semantic profiling (NTD proportion, recency, readability, lexical diversity metrics) and blinded expert rating across five dimensions: clinical relevance, guideline alignment, clarity, distractor plausibility, and language/cultural fit. Results: Alama Health QA captured >40% of all NTD mentions across corpora and the highest within set frequencies for malaria (7.7%), HIV (4.1%), and TB (5.2%); AfriMedQA ranked second but lacked formal guideline linkage. Global benchmarks showed minimal representation (e.g., sickle cell disease absent in three sets) despite large scale. Qualitatively, Alama scored highest for relevance and guideline alignment; PubMedQA lowest for clinical utility. Discussion: Quantitative medical LLM benchmarks widely used in the literature underrepresent African disease burdens and regulatory contexts, risking misleading performance claims. Guideline anchored, regionally curated resources such as Alama Health QA and expanded disease specific derivatives are essential for safe, equitable model evaluation and deployment across African health systems.