VA Data Could Prove Key in Assessing Real-World Efficacy of Hepatitis C Drugs

Investigators have targeted a new source for hepatitis C (HCV) treatment analysis—a source with relevant data in the era of direct-acting antiviral (DAA) therapy.

The national Veterans Health Administration (VA) Corporate Data Warehouse (CDW) is a resource with accurate data on DAA use for chronic HCV treatment, making this data source a reliable base from which researchers can create retrospective analyses, according to a new study.

As a result, this observational data could serve an important role in pharmacoepidemiologic studies that would assess the overall benefit and risks of modern HCV therapies, investigators, led by Christopher T. Rentsch, MPH, noted. The CDW is a national source of documented prescriptions and pharmacy fills. In contrast, electronic data from each VA facility going through multiple extracts, transforms, and loads—meaning original data validity could be compromised.

Investigators retrospectively assessed HCV-infected patients treated with DAAs who were included in the Veterans Aging Cohort Study (VACS) at 5 different VA treatment sites (n = 501). All patients had pharmacy fill data available for analysis. The range of DAA agents or regimens prescribed in the cohort were as follows:

• Boceprevir
• Telaprevir
• Simeprevir
• Sofosbuvir
• Daclatasvir
• Ledipasvir/ sofosbuvir
• Ombitasvir/ paritaprevir/ ritonavir
• Ombitasvir/ paritaprevir/ ritonavir/ dasabuvir
• Elbasvir/ grazoprevir
• Sofosbuvir/ velpatasvir

The researchers also randomly selected patients with chronic HCV infection who were not treated for HCV during the 2-year study period. These patients underwent chart review. Study investigators calculated the positive predictive and negative predictive values overall and by VA treatment site.

Out of 2416 prescriptions recorded for the overall cohort, only 494 were validated based on data from the CDW approximately 6 months following the study period. This validation resulted in a positive predictive value of 98.6% (95% CI: 97.6 - 99.6). Across all 5 treatment sites, the positive predictive values featured a range of 96.7% to 100% (P = .60).

In 100 HCV patients in the CDW who did not receive HCV treatment, approximately 97% (n = 97) had confirmation of non-treatment, whereas 3% (n = 3) did receive HCV treatment during the 2-year study period. These findings resulted in an overall negative predictive value of 97.0% (95% CI: 93.7 - 100), which were comparable across all treatment sites (P = .51).

In data extraction 6 months after the main study period, only 1 out of 22 prescriptions were found for 3 patients who had no records of treatment recorded in the CDW. This finding yielded an updated positive predictive value of 98.6% (95% CI: 97.6 - 99.6) as well as a negative predictive value of 99.0% (95% CI: 97.1 - 100).

The analysis of patients attending only 5 VA sites may have resulted in findings that lack generalizability across all VA centers. Additionally, the retrospective nature of the study was a potential limitation of the analysis.

That said, observational data could still play an important role in safety- and efficacy-based assessments of modern HCV therapies.

“These findings provide assurance to researchers who use national VA CDW data for retrospective cohort studies that the CDW contains accurate information, particularly after a 6-month lag for data extraction, on HCV therapies in the modern treatment era, even for VA patients who obtained treatment from non-VA providers,” investigators wrote.

Assessing comparative effectiveness of several different DAA regimens, the importance of adverse effects (eg, acute liver injury or HBV reactivation), and the effect of alcohol abuse on DAA receipt with treatment outcomes may be an important future area of research, investigators concluded.

The study, "Provider verification of electronic health record receipt and nonreceipt of direct-acting antivirals for the treatment of hepatitis C virus infection," was published online in the Annals of Epidemiology.

This article was originally published by MD Magazine.