Categories
Nevin Manimala Statistics

Identifying underrepresented groups in oncology clinical trials using routinely collected data in an English academic trial setting

Trials. 2026 May 23. doi: 10.1186/s13063-026-09812-2. Online ahead of print.

ABSTRACT

BACKGROUND: To facilitate equitable access to novel treatments, cancer trial participants should represent as far as possible those that will receive the treatment in practice. We can identify groups who rarely participate in cancer trials by collecting demographic data from participants. In the UK, there is no standardised practice around demographic data capture, leading to inconsistent collection across trials. A lack of systematically collected and published quantitative data from participants in UK cancer trials may limit our ability to identify underrepresented groups.

METHODS: We reviewed availability and completeness of demographic data recorded from 2235 participants in six bladder and six head and neck cancer trials conducted by the Clinical Trials and Statistics Unit at the Institute of Cancer Research (ICR-CTSU) between 2001 and 2023. To assess the representativeness of trial populations, demographic data from trial participants were compared with published data (NHS Digital) from 260,350 people who were treated for these cancers between 2013 and 2022 in England, using chi-squared goodness-of-fit tests and one-sample tests of proportion. A survey was distributed to 12 clinical trials units conducting similar trials to establish which demographic data are routinely collected across the UK.

RESULTS: Data on ethnicity, postcode, smoking status, and co-morbidity burden were inconsistently captured across ICR-CTSU trials, with missing data. Amongst the overall trial population, people older than 80 (n = 486/2235, 22%), females (n = 466/2235, 21%), people living in the most deprived areas (n = 390/1447, 27%), and ethnic minority groups (n = 5/275, 2%) were underrepresented, with some differences by treatment modality. Responses from UK trial teams showed that aside from age and sex (routinely captured), smoking status was the most consistently captured (13/17 trials).

CONCLUSIONS: This study provides quantitative data on cancer trial participants examining several demographic factors and indicates potentially underrepresented groups in trials of the disease subtypes investigated. Missing data were likely observed as a result of data cleaning being focused on items directly addressing the research question. Collecting and analysing a broad range of demographic data with a focus on inclusivity can inform researchers of groups who may benefit from tailored interventions to increase accessibility to cancer trials in the future.

PMID:42177546 | DOI:10.1186/s13063-026-09812-2

By Nevin Manimala

Portfolio Website for Nevin Manimala