In biomedical and health research, groups identified as “vulnerable” are often excluded from research resulting in lack of evidence and downstream health disparities. This research aims to map out the landscape of published work about vulnerable populations by analyzing (1) groups that are considered vulnerable in published health research, (2) methods that are used to study them, and (3) external factors (country and dichotomized gender of author) that impact research on vulnerable populations.

Based on data sourced from PubMed and Web of Science, we used a cross-sectional bibliometric assessment to identify 40,762 publications from 2003 to 2023 that mentioned “vulnerable population” in the abstract, title or keywords. Main outcome measures included the relative frequency of Medical Subject Headings terms in articles about vulnerable populations and all PubMed articles, and distribution of authorships by gender and nationality.

Compared to all PubMed articles, manuscripts featuring social and environmental factors (poverty, immigration, health disparities) were over twenty times more likely to be associated with vulnerable populations. Additionally, community-based and qualitative approaches were more than five times more likely to be associated with vulnerable populations compared to all PubMed articles. Women served as first authors in 45.2% of all publications in PubMed, while they authored 51.3% of the publications on vulnerable populations. The countries contributing the most to vulnerable population research include the U.S. (39.4%), Canada (7.1%), Australia (6.5%), and England (5.3%).

By exposing the construct of vulnerability in health research, this research helps to understand research gaps and biases.