Population-based Genomic Screening Programs: The Need for Optimal Implementation to Ensure Health Equity

Population genomic screening of adults has emerged as a strategy to promote prevention of common diseases such as cancer and heart disease among persons with genetic conditions. This type of screening has the potential to identify millions of currently undetected people in the United States who are at risk of preventable diseases for which evidence-based recommendations already exist to reduce morbidity and mortality. However, as of today, many questions remain regarding screening acceptability, short- and long-term clinical outcomes, selection of genes for screening, and cost. Pilot population genetic screening programs are attempting to explore and fill these knowledge gaps.

A recent study explored the feasibility and effectiveness of a screening program in which potential participants with unknown disease risk were invited for genetic screening. Email invitations were sent to individuals who had previously received care at the University of Washington Medical Center, with oversampling of people from racial and ethnic minority groups. People who enrolled in the study were mailed a kit for at-home saliva collection. Once saliva samples were mailed back, DNA was extracted and tested for variants in 25 genes related to preventable conditions such as hereditary breast and ovarian cancer, Lynch syndrome, and familial hypercholesterolemia. If individuals were found to carry a variant associated with an increased risk for disease that can be modified through lifestyle or medical changes, a genetic counselor reached out to discuss the results and next steps with them. All other individuals received notice about their genetic screening results through a letter available on a secure online portal.

Overall, 40,857 people were emailed invitations for genetic screening, and 2,889 (7.1%) enrolled. This low enrollment rate varied across race and ethnicity groups, with the lowest acceptance of study invitations among African American individuals (3.3%) and the highest acceptance among Multiracial or Other Race individuals (13.0%). These race and ethnicity identifications were determined according to electronic health records. Study dropout was greatest just after email invitations were sent, suggesting that many people did not open their email invitation or did not click on the link in the invitation to learn more about the study.

Among the 2,864 study enrollees who received genetic screening results, 103 individuals (3.6%) had at least 1 actionable variant. Of these people, 31 reported already knowing about their genetic result from prior testing and 21 were likely to have met clinical diagnostic guidelines for genetic testing based on survey-reported personal and family health history. Overall, 74 (2.6%) new, actionable genetic findings were identified among enrollees in this study.

Results from Other Pilot Population Genetic Screening Programs

Findings from this study can be compared with results from other pilot population genetic screening programs such as Geisinger’s MyCode and the Healthy Nevada Project, which have both reported that population genetic screening can identify people with increased disease risk who may otherwise go undetected. In the case of MyCode, an observational study of electronic health records showed that 87% of participants found to have a pathogenic or likely pathogenic variant in genes associated with hereditary breast and ovarian cancer, Lynch syndrome, and familial hypercholesterolemia did not have a prior diagnosis of this result. Among Healthy Nevada Project participants, this value was 90%.

Enrollment in genetic screening has varied across research studies and programs depending on recruitment strategy. During recruitment of new participants for the BioMe Biobank, 93% of people who agreed to biobank participation consented to receive genetic screening results. When genetic screening was mediated by a primary care provider as part of the DNA10K program, 28% of those contacted about screening expressed interest, of whom 78% went through genetic screening. Enrollment in both studies was higher than what was seen in the study out of the University of Washington. High consent to receive genetic results in a biobank setting is not surprising, as people looking to participate in a biobank are likely already more amenable to research participation and receiving results. The higher enrollment reported by DNA10K may suggest that integrating genetic screening into existing health care and practices could increase participation.

Implications for Public Health

Differences in population genetic screening program design influence enrollment rates, particularly during the recruitment phase. In the absence of careful attention to how communities are engaged about genetic screening, population genetic screening may not reach all people, and existing health disparities could be widened. The effectiveness of population-wide screening to identify at-risk individuals will subsequently be diminished. Challenges in equitable reach of genetic screening should be an ongoing focus and consideration during pilot population genetic screening programs and evaluation.