Large-Scale, Primary Care–Based Hepatitis C Treatment in an Urban, Medically Underserved Patient Population

Abstract

Hepatitis C virus (HCV) infection is a critical public health concern in the United States. HCV is highly curable, but access to care is limited for many patients. Primary care models can expand access to HCV care. The Grady Liver Clinic (GLC) is a primary care–based HCV clinic founded in 2002. During 20 years, using a multidisciplinary team, the GLC expanded its operations in response to advances in HCV screening and treatment. We describe the clinic model, patient population, and treatment outcomes of the clinic from 2015 through 2019. During this period, 2689 patients were seen in the GLC, and 77% (n = 2083) initiated treatment. Eighty-five percent (1779 of 2083) of patients who started treatment completed treatment and were tested for cure, and 1723 (83% of the total treated cohort, 97% of those tested for cure) were cured. Building on a successful primary care–based treatment model, the GLC dynamically responded to the changes in HCV screening and treatment guidelines, continually increasing access to HCV care. The GLC serves as a model of primary care–based HCV care that aims to achieve HCV microelimination in a safety-net health system. Our findings support the notion that for the United States to achieve elimination of HCV by 2030, generalists can and should provide HCV care, particularly in medically underserved patient populations.

Hepatitis C virus (HCV) infection is a common cause of chronic liver disease worldwide. In the United States, HCV accounted for 14 242 deaths in 2019 and is a leading cause of morbidity and mortality, primarily through the development of cirrhosis and liver cancer. ¹ In the past decade, the number of annual cases of acute HCV infections increased almost 5-fold in the United States because of the nation’s opioid crisis. ¹ As such, HCV infection is a critical public health concern in the United States. The World Health Organization set a target of HCV elimination by 2030 ² ; however, the United States is not on track to achieve this goal. The COVID-19 pandemic contributed further delays to HCV elimination goals, with notable decreases in HCV screening since the onset of the pandemic. ³

HCV is highly curable given the advent of second-generation direct-acting antiviral (DAA) medications in 2013, with cure rates of 90% to 95%. ⁴ Despite the availability of effective treatment for HCV, the number of patients with HCV who initiated treatment in the United States declined from 164 247 in 2015 to 83 740 in 2020. ⁵ Access to HCV care is still limited for many patients. Often, patients do not have access to academic centers where specialists practice. In addition, lack of health insurance and long wait times for specialist appointments add to the long list of hindrances to optimal HCV treatment. Superimposed on these deterrents, groups that tend to have limited access to HCV treatment, such as African American and Hispanic people, tend to have higher prevalence and all-cause mortality associated with HCV infection. ⁶ Nonspecialist HCV treatment is safe and effective and, compared with specialist care, may even result in higher treatment and cure rates.^7,8 Thus, primary care models can be successfully used to expand access to HCV care.

Given the increased need for access to HCV treatment for medically underserved populations, a group of general internists founded The Grady Liver Clinic (GLC) at Grady Memorial Hospital in the Grady Health System (GHS) in Atlanta, Georgia, in 2002. The goal was to improve access to HCV care for Grady’s medically underserved population who have a high prevalence of hepatitis C and, thereby, decrease the burden of the disease among those disproportionately affected. In 2012, we detailed the GLC model and described patient demographic characteristics and treatment referral patterns for the first 5 years of the clinic’s operations (2002-2007) to demonstrate that primary care–based HCV management in a medically underserved urban area can be successfully implemented. ⁹ Since that report, the field of HCV treatment has changed dramatically, largely because of advances in antiviral treatment as well as implementation of routine HCV screening. Thus, during the last 2 decades, the GLC has undergone a transformation in patient population, treatment access, and HCV cure rates.

Purpose

The purpose of this study was to describe the clinic model, patient population, and treatment outcomes of a primary care–based hepatitis C clinic in a medically underserved population in an urban area to demonstrate how this model can contribute to the achievement of HCV elimination.

Methods

Setting and Program Description

Patient population

The GLC serves the primarily African American, low-income, medically uninsured and underinsured patient population of the GHS, a safety-net health system in Atlanta, Georgia. About 90% of patients report an annual household income <$20 000, and 57% of patients read below the eighth-grade level. ¹⁰ A large proportion (37%) of patients in the health system lack health insurance; of those who have health insurance, 26% have Medicare, 22% Medicaid, and 15% commercial health insurance. ¹¹ This study was reviewed and approved by the Emory University Institutional Review Board (IRB no. 00008840).

Infrastructure and staffing

The GLC is based in the Primary Care Center of Grady Memorial Hospital. Clinical spaces (administrative and patient care rooms) are allocated and dedicated for GLC use. Since its founding in 2002, the GLC has grown from operating 1 day per week to 4 days per week. It is run and staffed by general internists with expertise in HCV management and treatment. Clinical staffing consists of 6 general internists from the Emory University Division of General Internal Medicine (who see patients 2 days per week) and a nurse practitioner (who sees patients 4 days per week). The GLC team also includes a clinical pharmacist, a pharmacy technician, a patient navigator, and a program manager.

The Emory University Division of General Internal Medicine supports GLC faculty by counting GLC shifts toward overall clinical faculty time equivalent. One faculty member serves as medical director and another serves as associate medical director for the GLC (without additional protected time for these roles). As new faculty join the GLC, they are trained by experienced faculty. Additional GLC staff members are GHS employees whose time is dedicated to the GLC. Patient access representatives, nursing staff, and medical assistant staff at the Primary Care Center also support the GLC. The Primary Care Center clinical manager is responsible for administrative management of the GLC.

The expansion of the GLC team was made possible in part by GHS participation in the Health Resources and Services Administration’s 340B program, ¹² which requires pharmaceutical manufacturers participating in Medicaid to sell outpatient drugs at substantially reduced prices to qualifying health care organizations that care for patients who have a low income and/or lack health insurance. Specifically, GHS was able to purchase DAA medications at discounted prices. The money saved on these DAA medications was reinvested in the GLC to expand services by growing the multidisciplinary team, which allows for the provision of comprehensive HCV services from screening to treatment.

Screening and linkage-to-care programs

Our team directed the implementation of routine HCV screening and linkage to care based on Centers for Disease Control and Prevention and US Preventive Services Task Force guidelines starting in 2012. We started by screening those born from 1945 through 1965 on a small scale in our Primary Care Center, and over time we expanded the screening program throughout GHS to include GHS neighborhood health centers, the emergency department, the walk-in clinic, the obstetrics clinics, and the inpatient service.^13,14 We now perform universal screening of all adults as recommended in updated 2020 screening guidelines.^15,16 Hallmarks of the screening program include electronic health record–based screening algorithms and prompts to identify patients eligible for screening and patient navigators to disclose results and link patients to care (the majority at the GLC). These efforts from 2012 through 2019 resulted in the screening of more than 42 000 patients, the revealing of a ~6.5% prevalence of HCV exposure among outpatients screened and a 14% prevalence among inpatients screened, the identification of more than 1800 patients with chronic HCV infection, and linkage to care of more than 60% of eligible patients.^13,14

Treatment programs

For patients diagnosed with HCV, the GLC is the primary site for HCV care at GHS. The pretreatment program includes patient education, liver fibrosis staging, and preventive management, including vaccinations for hepatitis A and B. Following national HCV treatment guidelines, we prescribe HCV medications that are covered through public or private health insurance or, for patients who lack health insurance, patient assistance programs offered by pharmaceutical companies. The patient navigator is instrumental in patient education and support at all stages of the care cascade, from linking referred patients to care, to providing one-on-one education in clinic using a customized information booklet, to navigating barriers to follow-up. The pharmacy technician handles the authorization process for medication access by completing applications for prior authorization or patient assistance programs. During HCV treatment, patients are monitored monthly at in-person or telehealth visits by a clinical pharmacist with the support of the patient navigator. Most patients receive DAA medications that are filled at the GHS pharmacy and either hand delivered at in-person visits or couriered to their homes after telehealth visits. Patients with cirrhosis continue to be followed at the GLC after HCV cure and are enrolled in a cirrhosis management program that includes biannual clinical assessment and hepatocellular carcinoma screening.

Data Collection and Analysis

We performed a retrospective medical record review of all patients who had at least 1 visit in the GLC from January 1, 2015, through December 31, 2019. We extracted data on demographic characteristics, laboratory tests, fibrosis staging, and details of HCV treatment.

For demographic characteristics, we extracted data on age, sex (male or female), and race. Because racial and ethnic minority populations are disproportionately affected by social determinants of health and we wanted to demonstrate how our patient population is disproportionately affected by social determinants of health, we included race in our analysis. We used data on race as it was listed in GHS medical records; these options are Black, White, Hispanic, and Asian/Pacific Islander. For analysis, we combined the categories of Hispanic and Asian/Pacific Islander into “Other.”

For laboratory tests, we collected data on aspartate aminotransferase (AST), alaninaminotransferase (ALT), platelet count, and HCV genotype, the strain of the virus with which the patient was infected, categorized as genotype 1, genotype 2, genotype 3, or other. For fibrosis staging, we collected Fibrosis-4 (FIB-4) measurement (calculated with age, AST, ALT, and platelet count) and/or transient elastography scores from a FibroScan (a specialized ultrasound machine for scanning the liver). For determination of cirrhosis, we used the clinician’s determination as documented in the medical record. For details of HCV antiviral therapy, we collected the treatment regimen, treatment dates, and outcomes of testing for cure.

We counted the number of patients who started HCV treatment and the number who did not start treatment, and we compared the characteristics of these 2 groups. We did not systematically collect data on reasons for nonreferral for treatment; however, we did review the medical records of those who did not start treatment to determine if a reason for nonreferral could be gleaned. We also collected data on the number of patients who did not start treatment after referral and the reasons they did not start treatment, and the number of patients who completed treatment and were tested for cure, were cured, failed treatment, awaited testing for cure, and discontinued treatment. We defined cure as sustained virologic response (undetectable virus 12 weeks after completion of treatment).

Unless otherwise stated, we described data as number (percentage) and mean (SD). We compared findings by using Pearson χ² tests and unpaired t tests. P values ≤.05 were considered significant. Multivariable regression analysis explored the association of age, sex, race, HCV genotype, and cirrhosis status with cure. We used R version 3.4.1 (R Foundation for Statistical Computing) to perform all analyses.

Outcomes

From 2015 through 2019, the GLC saw 2689 patients. The mean age of the patients was 64.3 years; 1720 (64.0%) were male, 2251 (83.7%) were Black, and 617 (23.0%) had cirrhosis (Table 1). During the study period, 2083 (77.5%) patients started HCV treatment. Of the 606 (22.5%) who did not start treatment, 482 (17.9% of 2689) patients were evaluated in the GLC but never referred for treatment. Common reasons for nonreferral included loss to follow-up, patient preference, and provider determination of ineligibility for treatment (short life expectancy or comorbid medical illness, such as active cancer). The remaining 124 (4.6% of 2689) patients were referred for treatment but never started treatment. The reasons for nontreatment after referral included unable to contact (n = 38), did not keep treatment start appointments (n = 30), unknown (n = 21), health insurance issue (n = 10), and patient declined treatment after referral (n = 11). Fourteen patients had additional reasons, such as being treated elsewhere, having relocated, being hospitalized, or being incarcerated.

Table 1.

Hepatitis C–related characteristics of the Grady Liver Clinic population, Atlanta, Georgia, 2015-2019 ^a

Characteristic	Total no. of patients (N = 2689)	Started treatment
		No (n = 606)	Yes (n = 2083)	P value b
Age, mean (SD), y	64.3 (9.0)	63.3 (10.2)	64.6 (8.5)	.001
Male sex	1720 (64.0)	400 (66.0)	1320 (63.4)	.25
Race				.001
Black	2251 (83.7)	478 (78.9)	1773 (85.1)
White	354 (13.2)	108 (17.8)	246 (11.8)
Other c	84 (3.1)	20 (3.3)	64 (3.1)
Genotype				.03
1	2385 (89.7)	502 (86.9)	1883 (90.5)
2	132 (5.0)	31 (5.4)	101 (4.9)
3	60 (2.3)	19 (3.3)	41 (2.0)
Other	82 (3.1)	26 (4.5)	56 (2.7)
Cirrhosis				<.001
Yes	617 (23.0)	149 (24.9)	468 (22.5)
No	2039 (76.1)	424 (70.8)	1615 (77.5)
Unknown	25 (0.9)	26 (4.3)	0

^aData source: medical records of Grady Liver Clinic. All values are number (percentage) unless otherwise indicated. Not all categories sum to the total number because of missing data; participants with missing data were excluded from percentages.

^bDetermined by Pearson χ² and unpaired t tests. P ≤ .05 was considered significant.

^cIncludes Hispanic and Asian/Pacific Islander; the options in the medical record were Black, White, Hispanic, and Asian/Pacific Islander.

The patients who were treated were significantly older (64.6 y) than those who were not treated (63.3 y) (P < .001); we found no differences by patient sex. A significantly smaller percentage of patients who were treated had cirrhosis (22.5%) compared with those who were not treated (24.9%) (P < .001) (Table 1). All patients who received treatment were treated with a DAA medication. Of the 2083 patients who started treatment, 1779 (85.4%) completed treatment and were tested for cure. Of these patients, 1723 (82.7% of 2083 and 96.9% of 1779) were cured, and 56 (2.7%) failed treatment. Of the remaining 304 patients who started treatment, 236 (11.3% of 2083) were not tested for cure, and 68 (3.3% of 2083) discontinued treatment. Multivariable regression analysis of data on patients who completed treatment and were tested for cure showed that only cirrhosis status was significantly associated with cure; specifically, that patients with cirrhosis were less likely than patients without cirrhosis to be cured (Table 2).

Table 2.

Per-protocol multivariable regression analysis of characteristics associated with sustained virologic response among patients with hepatitis C in the Grady Liver Clinic population, Atlanta, Georgia, 2015-2019 ^a

Characteristic	Odds ratio (95% CI)	P value b
Age	1.11 (0.79-1.46)	.51
Sex
Female	1 [Reference]
Male	0.63 (0.33-1.15)	.14
Race
Black	0.53 (0.11-1.49)	.28
White	1 [Reference]
Other c	— d	— d
Genotype
1	1 [Reference]
2	0.59 (0.20-2.56)	.40
3	0.45 (0.08-8.62)	.46
Cirrhosis
No	1 [Reference]
Yes	0.31 (0.17-0.54)	<.001

^aData source: medical records of Grady Liver Clinic.

^bDetermined by Pearson χ² test. P ≤ .05 was considered significant.

^cIncludes Hispanic and Asian/Pacific Islander; the options in the medical record were Black, White, Hispanic, and Asian/Pacific Islander.

^dNot calculated.

Lessons Learned

The outcomes of our primary care–based HCV treatment program support the results of other studies that demonstrated that HCV treatment administered by nonspecialist providers is as effective as that provided by specialists. ⁸ For patients in our study who completed treatment and were tested for cure, cure rates matched those of clinical trials. ⁴ Our findings underscore the critical need for nonspecialist providers in the United States to provide HCV management and treatment. Expansion of access to HCV care beyond specialty providers is crucial if the United States is to achieve the World Health Organization’s target of HCV elimination by 2030. ²

Our primary care–based HCV clinic has existed since 2002 and has grown and adapted to extensive changes in HCV management and treatment. In particular, the GLC has achieved substantial growth in treatment capacity. Since 2002, several primary care–based HCV programs have been described.^7,8,17-22 Compared with these programs, the GLC is one of the oldest and largest single-site primary care–based HCV treatment programs in the United States. Several factors played a role in the longevity and growth of the GLC. The continued support of faculty time dedicated to the GLC by the Emory University Division of General Internal Medicine leadership has been critical. The GLC also receives support from the GHS administration, including use of examination rooms and triage staff of the Primary Care Center. GHS appropriately applied savings from its participation in the 340B program to fund additional staff, which led to GLC expansion in 2017. Finally, as HCV screening guidelines expanded, we implemented large-scale HCV screening across GHS, which uncovered a high prevalence of HCV exposure among outpatients screened (~6.5%) and inpatients screened (14%) in our health system.^13,14 Systematic screening, coupled with linkage to care, which uses patient navigation and program management, fueled growth in the GLC patient population.

Several opportunities exist for us to build on our successful primary care–based HCV treatment model. For example, when we identified a gap in care among our patients with cirrhosis, many of whom were not receiving the appropriate monitoring for complications of cirrhosis, such as screening for hepatocellular carcinoma, we added a cirrhosis management service. In addition, we are beginning to add treatment programs for other common chronic comorbid conditions in our patient population, specifically alcohol use disorder and nonalcoholic fatty liver disease, the management of which is also well suited to primary care providers.

Our model leverages the resources of an academic, hospital-based primary care clinic, which contributes to its success. However, we believe that nonacademic– and non–hospital-based primary care settings can also leverage their resources to achieve similar outcomes to our own. Several studies have described effective HCV management by primary care providers in various settings.^7-9,17-22 As such, in settings where there are leadership champions for elimination of HCV, various strategies can be used to identify, link, and navigate patients to cure of HCV infection.

Another key to our clinic’s success has been our ability to access DAA medications for patients with various payor sources, including a substantial proportion without health insurance. We acknowledge that the availability and ease of accessing DAA medications vary not only between practice settings but also across states. Patients in states without Medicaid expansion (eg, Georgia) and patients insured by Medicaid in states whose Medicaid coverage is limited by fibrosis or sobriety status have special challenges in obtaining DAA medications. Our success in treating a high proportion of uninsured and Medicaid-insured patients is a testament to the fact that treatment is accessible but may require an investment in resources, such as our pharmacy technician, who serves as a medication access coordinator.

We have demonstrated the longevity and growth of the GLC, a primary care–based HCV treatment program, during 20 years. The GLC has successfully provided care to a patient population with numerous barriers to care by expanding our screening and treatment capacity and utilizing a multidisciplinary team. Our study adds to the emerging body of literature that suggests that for the United States to achieve elimination of HCV by 2030, generalists can and should provide HCV care, particularly in medically underserved patient populations.