Key points are not available for this paper at this time.
Abstract Background: Race and ethnicity (R/E) information is crucial for understanding health disparities and the efficacy/effectiveness of treatments across different patient groups. However, these data are often missing or inaccurate in EHR. Literature suggests that surveys can improve upon these gaps in EHR R/E data. FDA recently issued guidance on collecting and reporting R/E data using standardized terminology, which should be applied to demographic data in new clinical trials and studies. N-Power Medicine (NPM) developed a concise survey, incorporating FDA’s R/E recommendations, to collect SDoH data as part of its registry (Kaleido) enhanced oncology network platform for quality data collection. Here, we report the initial R/E findings from this survey. Methods: Patients participating in the Kaleido Registry at 2 community oncology practices across different US regions completed the survey during their clinic visits with assistance from NPM onsite staff. The survey allowed multiple race selections to include mixed races and a free-text option for “Other”. R/E data from the same patients were also extracted from EHRs. Completeness was reported as the percentage of informative answers: any race, other, and chose not to disclose. “Unknown” and “Not available” were missing data. Results: Between November 11, 2023 and April 30, 2024, 1109 patients completed the questionnaire with over 99% response rate. Both sites contributed equally and had similar patient demographics: about 61% female and 79% aged 60 or above. The large majority of patients had breast cancer followed by lung and colorectal cancers. Survey completeness was 96.7% for race and 99.6% for ethnicity, as compared with EHR completeness of 78.6% for race and 91.1% for ethnicity. 21% of EHR data for race was coded as "Unknown and Not Available. About 20% of the survey respondents were Hispanic or Latino. Both sites’ EHR data included expected R/E categories except for one site having 21 additional unique race entries. The majority of survey participants were White (W, 75.7%) followed by Asian (A, 3.6%), American Indian or Alaska Native (AI/AN, 3.2%), Black or African American (B/AA, 2.4%), Native Hawaiian or Pacific Islander (NH/PI, 1.0%) and multiple-selection (2.9%; Other and W (8), AI/AN (3), A (2); W and AI/AN (10), A (7); A and AI/AN (1), W and B (1)). 11% of responses were Other, which included various races (15% European/Caucasian), nationalities, ethnicities (49% Hispanic) and religions, hence, could be further disambiguated. We will report details regarding the survey respondent’s other SDoH. Conclusions: We successfully implemented a standardized questionnaire, incorporating FDA R/E collection recommendations, at two community oncology sites. The results demonstrate that the NPM R/E questionnaire has higher completeness compared to EHR data, allowing better assessment of the impact of R/E on clinical trial enrollment, genomic testing utilization, treatment patterns, and outcomes. This improvement is vital for understanding and addressing health disparities in oncology care. Citation Format: Michael Zachary Koontz, Ayse Tezcan, Julie Latham, Phyllis Ortiz, Joseph Nathaniel Paulson, Annette Campbell Fontaine. Improving Race and Ethnicity Data Completeness Using a Concise Survey Designed to Collect Social Determinants of Health (SDoH) Data: Results from Two Community Oncology Practices abstract. In: Proceedings of the 17th AACR Conference on the Science of Cancer Health Disparities in Racial/Ethnic Minorities and the Medically Underserved; 2024 Sep 21-24; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Epidemiol Biomarkers Prev 2024;33(9 Suppl):Abstract nr B051.
Koontz et al. (Sat,) studied this question.