Don’t Put Off Until Tomorrow What You Can Do Today: Hospital Admissions as an Opportunity to Treat Hepatitis C

(See the Major Article by Midgard et al. on pages 582–90.)

Progress on hepatitis C virus (HCV) elimination in the United States is well behind the pace needed to realize the 2030 goals set forth in 2016 [1]. Based on the most recent viral hepatitis report from the Centers for Disease Control and Prevention [2], data continue to trend in the wrong direction, with nearly 70 000 new infections estimated in 2021, representing a 5% increase from 2020 and continuing a trend of year-over-year increases in incidence since 2010. In addition, despite the introduction of highly effective direct-acting antivirals (DAAs) for HCV and treatment of >1.2 million persons since 2014, the prevalence of HCV infection in the United States (approximately 2.2 million HCV RNA–positive persons) has remained stable since 2016, reflecting a high rate of ongoing transmission [3, 4]. On all fronts we are losing the battle to eliminate HCV in the United States; business as usual is clearly not enough. To turn the tide, a multifaceted approach to HCV diagnosis and treatment delivery is required.

Persons who inject drugs are a priority population for treatment because of high HCV prevalence and incidence, providing an opportunity not only to improve individual health but also to prevent transmission. This population also experiences disproportionate hospitalization rates, which have increased dramatically over the last decade, owing to sequelae of injection drug use [5, 6]. Hospitalization represents a potential intervention point for HCV diagnosis, education, disease assessment, and treatment initiation, particularly in light of simplified treatment approaches for HCV [7, 8]. Similar interventions have proved successful for initiation and retention with medical treatment for substance use disorder [9, 10].

The article by Midgard et al [11] in this issue of Clinical Infectious Diseases seeks to expand on the accepted options for HCV assessment and treatment by examining outcomes related to a program to expedite delivery of HCV treatment in hospitalized persons, many of whom had recent injection drug use. The randomized intervention was rolled out in step-wedge fashion across 7 clinical departments encompassing medical, addiction, and psychiatry inpatient units across Oslo, Norway. The setting is important, as discussed in further detail, because it significantly affects feasibility and impact of such an intervention. Follow-up was individualized according to the treating physician, in line with the pragmatic design of the study, but this likely adversely affected retention rates.

As could be expected, and part of the rationale for the stepwise cluster randomized trial design, DAA starts and completion within 6 months (primary end point) were significantly higher in the intervention group. However, perhaps counterintuitively there was no difference in measured sustained virologic response (SVR) between groups at the end of the follow-up period (“data lock”). On the surface this may seem to be a failure of the intervention, yet it is important consider the reasons for non-SVR among the intervention and control groups. Missing data accounted for lack of SVR in the intervention group (21% vs 4% in the control group), while the primary reason for not achieving SVR in control group was never starting medication (29% vs 11% in the intervention group). While higher rates would be ideal, the lower absolute SVR in the intervention group should not dissuade from consideration of such an approach, given that treatment completion rates were relatively high. It has been repeatedly demonstrated that an SVR at 12 weeks (SVR12) of >90% will be realized, including populations with recent or ongoing drug use, even in the setting of suboptimal medication adherence [12, 13]. In all likelihood the intervention group realized higher cure rates despite loss to follow-up.

Interestingly, when the time frame was extended to data lock, the difference in treatment initiation narrowed with overlapping confidence intervals. While loss to follow-up for treatment initiation was higher in the control group; among those who completed therapy, loss to follow-up for SVR assessment was higher in the intervention group. These findings suggest some benefit of starting medication in keeping persons who inject drugs engaged temporarily; conversely, those in the control group who were able navigate outpatient treatment follow-up to start DAAs were more likely to stay engaged. It would be of interest to know whether loss to follow-up in the intervention group was higher in those not started on DAAs while hospitalized. Regardless, inpatient starts are only one piece. The significant loss to follow-up in the intervention group points to the need for additional services/interventions after discharge to keep marginalized populations engaged in care. This mirrors rapid human immunodeficiency virus (HIV) treatment and preexposure prophylaxis, in the sense that while upfront treatment initiation has a big impact, ongoing support for engagement in care and follow-up is still critical [14, 15].

Some limitations and unique aspects of the study deserve mention. Importantly, these are not critiques but rather limitations inherent in rolling out a pragmatic study at multiple locations while balancing broad applicability with scientific rigor. Many of these limitations could be overcome during local clinical implementation.

The relatively slow turnaround time for HCV RNA in some locations resulted in 7% in the intervention arm not starting therapy in house. In addition, 40% of the RNA positive population were discharged before enrollment despite a median stay of 5 days in the intervention group. Was this group fundamentally different (eg, did they have self-directed discharge)? This suggests that other methods will be needed to engage a significant portion of the at-risk population. Rapid point-of-care HCV RNA assays reduce the time for results to 60–90 minutes and could limit lost opportunities for in-hospital DAA initiation [16].

SVR as a secondary end point is another potential limitation, though we agree with the investigators' approach in selecting DAA completion as the primary end point. As noted, less-than-perfect adherence is well tolerated with DAA therapies, and if the full course of medication is taken (with reasonable adherence) SVR can generally be assumed. It is also worth noting that SVR in the current study was assessed at 4 weeks (SVR4). While SVR12 is the accepted standard, we agree it was appropriate to assess SVR4 in this study, given the high rates of homelessness and often transient housing among persons who inject drugs. In addition, most HCV recurrences occur within 4 weeks of treatment completion, and 98% of persons attaining SVR4 also achieve SVR12 [17, 18].

The major limitation of the OPPORTUNI-C study, at least for those in the United States, is reproducibility. When reviewing a study and judging its impact, clinicians are faced with the questions of how the study findings apply to their patients and whether these findings could reasonably be expected to be replicated where the clinicians practice. Unfortunately, the harsh reality of the fragmented US healthcare system in this regard would provide a resounding answer of “no” to the question of reproducibility, at least not without major changes to current Medicaid policies around the country. Specifically, Medicaid expansion is needed to cover those who are uninsured in nonexpansion states, prior authorization restrictions must be eliminated so that all persons with HCV can access treatment in accordance with national guidelines, and barriers from discrepant systems for inpatient versus outpatient medication pricing and coverage should be addressed [19]. “Carve outs” for HCV medications, as done under some Medicaid programs for HIV antiretrovirals could help, provided that an agreement is in place to cover the mediations at a national level.

Perhaps the approximately $12.3 billion currently occupying President Biden's 2024 budget can provide the capital needed to ensure medication access in an unrestricted fashion for those on public insurance programs across the United States [20]. Importantly within this program are provisions for a “national prescription model” to provide DAAs, presumably free of charge, to those with Medicaid/Medicare. This alone would remove a significant barrier to implementing programs, such as the one explored in the OPPORTUNI-C study. If the United States is truly serious about eliminating HCV, insurance coverage and reimbursement processes need to change.

What are other potential solutions? First, systems to outreach patients with HCV when admitted to the hospital or the emergency department can be useful for connecting patients to outpatient HCV treatment appointments [21]. Emergency departments may be another clinical venue to explore opportunistic HCV treatment, and primary care–based treatment is a well-established patient-centered approach to increasing treatment rates [22]. HCV treatment is now the standard of care in many prisons around the country, while jail-based treatment is less common though potentially feasible and impactful. Partnerships with nonclinical organizations, such as syringe service programs, have also been shown to be effective for offering a range of HCV services [23]. On-site treatment is often preferred, though treatment in mobile units or through telehealth has emerged as a practical solution to the logistical challenges of providing care outside medical settings. Finally, simply offering drop-in visits acknowledges the competing priorities and challenges our patients often face and shows them that offering HCV treatment when it fits into their busy lives is our shared priority.

Finally, let us not forget that while HCV DAA treatment has always been cost-effective, recent data suggest that treatment provides significant cost savings, which are realized in just a few years [24]. It is time for the United States to finally make progress on HCV elimination by putting our money where our HCV is.

Author notes

Potential conflicts of interest. S. E. R. reports grants from the National Institute on Drug Abuse, National Institutes of Health (grants 1R01DA042982-01A1 and 3R01DA42982-01A1S1); honoraria for CME lectures from Clinical Care Options and Integritas Communications; honoraria for participation in the National PrEP Curriculum Advisory Board from the University of Washington; and leadership roles on the Board of Trustees for the Boulder County AIDS Project and the Denver Harm Reduction Action Center. D. L. W. reports institutional research grants from Gilead Sciences; personal royalties from UpToDate; personal consulting fees and travel support for participation on advisory boards for Gilead Sciences and AbbVie; and payment for a lecture on hepatitis D in human immunodeficiency virus from Gilead Sciences.

Both authors have submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Conflicts that the editors consider relevant to the content of the manuscript have been disclosed.