Association for Public Policy Analysis & Management

Approximately 2.7 million Americans are chronically infected with the hepatitis C virus (HCV).  HCV infection is largely asymptomatic for decades, after which approximately one third of infected patients develop decompensated cirrhosis, hepatocellular carcinoma, and die prematurely. Approximately 80% of HCV-infected Americans were born during the years 1945–1965. Given this group’s high infection prevalence, in 2012 and 2013, the Centers for Disease Control and Prevention (CDC) and the U.S Preventive Services Task Force recommended a one-time HCV antibody test for all adults born during 1945-1965. CDC’s Birth-cohort Evaluation to Advance Screening and Testing for Hepatitis C (Best-C) was designed to assess the impact and cost-effectiveness of implementing birth cohort (BC) testing in primary care settings. To assess the impact of site-specific testing interventions on the probability of HCV testing among BC patients as compared to usual care and assess patient characteristics associated with testing, and the incremental costs per person tested and per case identified at each site.   

From December 2012-March 2014, three health systems implemented independent, tailored testing interventions using randomized designs to compare intervention testing and identification rates to usual care. Site 1 mailed pre-registered lab test orders and repeated reminders to a randomly selected list of active patients compared to a second list who received no mailings. Site 2 created an electronic health record best practice alert (BPA) implemented or not implemented based on cluster randomized design. Site 3 directly recruited patients following a scheduled primary care visit and used a cluster randomized crossover design. Multilevel multivariable regression was used to estimate the risk ratio for HCV testing; activity-based costing was used to estimate costs. 

HCV testing was significantly more common for all interventions compared to controls; adjusted risk ratio (aRR) was 19.2, (95% CI, 9.7 – 38.2), 13.2 (95% CI, 3.6 – 48.6), and 35.0 (95% CI 19.5 –62.9) for sites 1, 2, and 3, respectively. Across sites, patients who were non-Hispanic, born before 1950, and publicly insured were more likely to be tested. Non-intervention testing ranged from $19-$26 per person tested, while the mail intervention was $63 per person tested, direct recruitment was $53 per person tested, and BPA was $45 per person tested when including start-up costs, and BPA was $23 per person tested when excluding startup costs.  Over the study period the cost per new case identified under usual care ranged from $644-$1,148 for the three sites. Costs for direct recruitment and the BPA (including startup costs) ranged from $4,230-$4,583 per newly identified case, respectively. After excluding startup costs, BPA was $2,336 per new case identified. The mail intervention was $7,005 per new case identified.

All interventions increased HCV testing among the BC compared to usual care, but also increased the aggregate and per person costs of testing. The cost per case identified excluding start-up costs was lowest for the BPA intervention, suggesting that integrating BC testing into usual care is likely to be more cost-effective than instituting an intervention in addition to usual care, e.g., repeated mailings and direct recruitment.