Abstract
IMPORTANCE: Best practices for improving access to assessment of hereditary cancer risk in primary care are lacking. OBJECTIVE: To compare 2 population-based engagement strategies for identifying primary care patients with a family or personal history of cancer and offering eligible individuals genetic testing for cancer susceptibility. DESIGN, SETTING, AND PARTICIPANTS: The EDGE (Early Detection of Genetic Risk) clinical trial cluster-randomized 12 clinics from 2 health care systems in Montana, Wyoming, and Washington state to 1 of 2 engagement approaches for assessment of hereditary cancer risk in primary care. The study population included 95 623 English-speaking patients at least 25 years old with a primary care visit during the recruitment window between April 1, 2021, and March 31, 2022. INTERVENTION: The intervention comprised 2 risk assessment engagement approaches: (1) point of care (POC), conducted by staff immediately preceding clinical appointments, and (2) direct patient engagement (DPE), where letter and email outreach facilitated at-home completion. Patients who completed risk assessment and met prespecified criteria were offered genetic testing via a home-delivered saliva testing kit at no cost. MAIN OUTCOMES AND MEASURES: Primary outcomes were the proportion of patients with a visit who (1) completed the risk assessment and (2) completed genetic testing. Logistic regression models were used to compare the POC and DPE approaches, allowing for overdispersion and including clinic as a design factor. An intention-to-treat analysis was used to evaluate primary outcomes. RESULTS: Over a 12-month window, 95 623 patients had a primary care visit across the 12 clinics. Those who completed the risk assessment (n = 13 705) were predominately female (64.7%) and aged between 65 and 84 years (39.6%). The POC approach resulted in a higher proportion of patients completing risk assessment than the DPE approach (19.1% vs 8.7%; adjusted odds ratio [AOR], 2.68; 95% CI, 1.72-4.17; P < .001) but a similar proportion completing testing (1.5% vs 1.6%; AOR, 0.96; 95% CI, 0.64-1.46; P = .86). Among those eligible for testing, POC test completion was approximately half of that for the DPE approach (24.7% vs 44.7%; AOR, 0.49; 95% CI, 0.37-0.64; P < .001). The proportion of tested patients identified with an actionable pathogenic variant was significantly lower for the POC approach than the DPE approach (3.8% vs 6.6%; AOR, 0.61; 95% CI, 0.44-0.85; P = .003). CONCLUSIONS AND RELEVANCE: In this cluster randomized clinical trial of risk assessment delivery, POC engagement resulted in a higher rate of assessment of hereditary cancer risk than the DPE approach but a similar rate of genetic testing completion. Using a combination of engagement strategies may be the optimal approach for greater reach and impact. TRIAL REGISTRATION: ClinicalTrials.gov Identifier: NCT04746794.