New study highlights gaps in how diversity data are used in mental health research

Diversity characteristics recorded in electronic health records are rarely reported in published mental health research, despite being readily available to researchers, according to research by. King’s College London and South London and Maudsley NHS Foundation Trust.

Published in BMJ Open, the study reviewed 362 research articles that used data from the Clinical Record Interactive Search (CRIS) platform, one of the UK’s largest mental health research data resources. The researchers assessed how often diversity characteristics were reported in studies and compared this with the information available in the underlying electronic health records.

The team found that age (89.3%), sex (87.3%) and ethnicity (80.4%) were the most reported characteristics in published papers. However, other factors linked to health inequalities were used far less frequently, even when the data were available.

For example:

• Geographic location and experiences of homelessness were recorded for around 97.5% of patients in the database but were reported in fewer than 10% of publications
• Socioeconomic status was also highly available (97.5%) but was reported in only 42.5% of publications
• Pregnancy/maternity showed moderate availability but was reported in only 4.12% of publications
• Other characteristics showed low availability and lower reporting, including religion (14.3% available, 0.55% reported), and migration status (6.9% available, 0.8% reported).

This research suggests that the diversity of people represented in electronic health records is not always reflected in the evidence generated from those data. The authors call for greater attention to recording, accessing and reporting a wider range of diversity characteristics to ensure research findings are more inclusive and relevant to policy and practice.

The study also highlights opportunities for emerging technologies, including natural language processing, to make better use of information captured in clinical records, particularly for characteristics that are not routinely recorded in structured data fields.

The authors conclude that as electronic health record data play an increasingly important role in shaping mental health research and healthcare policy, improving the use of diversity data will be essential to supporting more equitable and representative research.