Hepatitis C Infection Confirmation and Fewer Visits Contribute to Improving Treatment Rates in a Predominantly African American Health Center

Abstract

Background/Objectives: The approval of direct-acting antiviral (DAA) therapy for hepatitis C (HCV) resulted in a highly effective oral treatment for patients. The primary objective of this study was to identify reasons that patients were not treated versus why patients were treated. Identifying potential reasons for the failure to treat can provide a pathway to interventions using evidence-based data. Methods: The electronic medical records in an urban predominately African American (AA) population were searched for all patients with HCV seen at least once in a Gastroenterology or Infectious Disease clinic in 2019. Data collected included demographics, treatment visits, laboratory data, insurance and ZIP codes for median income. Results: Of the 441 patients who were not yet treated at the first 2019 visit, only 43% were treated by July 2020. Insurance and median income were not factors in failure to treat. Patients with an average of four visits were more likely to be treated than those with two or less, suggesting that failure to follow up was a significant factor for patient treatment (42% vs. 8% p < 0.0001). Confirmation of viral infection at first visit was an important factor with respect to treatment (treated 38% vs. not treated 25% p < 0.02). Conclusions: Significant numbers of our patients (57%) failed to be treated after at least one clinic visit. The two critical factors were PCR confirmation prior to the initial visit and the requirement for multiple visits before the initiation of treatment. Since the degree of fibrosis had no impact on treatment, initiating treatment immediately after confirming infection with HCV should improve patient treatment rates and outcomes.

1. Introduction

An estimated 2.2 million individuals in the United States are living with hepatitis C (HCV), and the prevalence has remained largely unchanged since the 2013–2016 time period [1]. The emergence of Interferon-free direct-acting antivirals (DAAs) provided an efficacious and well-tolerated treatment for HCV. After 8–12 weeks of treatment, DAA combinations clinically cure HCV for more than 95 percent of individuals with chronic HCV infection. Despite this, data show that less than 50% of patients diagnosed with HCV were treated with DAAs from 2014 to 2020 [2,3,4,5,6]. Furthermore, a meta-analysis of 65 studies reported an overall treatment rate for HCV-RNA-positive patients of 41% (range 20–51%), with 75% of the treatment studies performed in United States populations [7]. With respect to improvements over time, Jaing et al., in 2023, reported improvement from 3% in 2011 (Interferon era) to 54% in 2018 due primarily to universal coverage in Canada for all DAAs [8,9]. Thus, based on the literature, it is apparent that despite the widespread availability of DAAs, achieving treatment of more than half of the individuals who are identified as HCV-infected (polymerase chain reaction positive for virus (PCR positive)) remains an elusive goal.

The literature suggests multiple impediments to care. One of the most effective is the use of a multifaceted team approach [10,11,12,13,14,15]. This would include enhanced personal interventions coupled with the use of electronic medical records to identify and contact potential candidates for therapy. In one study, these resource-intensive interventions mediated primarily through primary care settings resulted in an improvement from 65% to 76% [12]. An even more complex issue that is present in the literature is that many individuals with HCV have socioeconomic barriers such as drug addictions, engagement in opioid agonist therapy, incarceration, poor or no insurance, flexible living conditions and complications related to reliable communication. While these issues are clearly present and addressable, there is a need to address the more critical issue of treatment after the first visit to either a Gastroenterology or Infectious Disease clinic with a diagnosis of suspected HCV.

To achieve a more effective elimination of HCV, evidence-based information is needed to identify areas where intervention may be implemented to increase the treatment rates to above 50%. Given the fact that HCV is twice as likely to be in African American as non-AA patients, studies in clinics such as ours that see predominately AA patients should be useful. In addition, the location of our clinics in an urban medical center that serves a population of socioeconomically challenged individuals may also provide clues to strategies for improved treatment rates. The primary objective of this study was to identify patients who were not treated after a visit to an urban medical center clinic and compare them to patients who were treated. The hypothesis was that after identifying potential reasons for the failure to treat, interventions could then be made using evidence-based data. The 2019 patients were seen at least once in Gastroenterology (GI) or Infectious Disease (ID) clinics. They were evaluated for characteristics that were different between HCV patients who were treated and those not treated.

Thus, this study’s main aim was to identify reasons that patients were not treated versus reasons patients were treated. Identifying potential reasons for the failure to treat can provide a pathway to interventions using evidence-based data.

2. Methods

The electronic medical records of a large urban and predominately African American clinic population were searched for an ICD-10 billing code of hepatitis C virus visits in 2019. Data were collected under an institutional IRB approved project from 2019 HCV patient electronic medical records (EMRs) including demographics, laboratory studies and treatment history. If patients were seen in the last months of 2019, we used the 2020 records to assess 6-month follow-up visits for all patients to establish a baseline time frame for treatment of at least 6 months. Fibrosis at the initial visit was based on laboratory data and online calculators using both the Aspartate Transaminase (AST) to Platelet Ratio Index (APRI) and Fibrosis-4 (FIB-4; which includes AST, Alanine Transaminase (ALT), platelets and age of the patient). ZIP codes were used to provide median income data for patients based on data for 2020 income via searching INCOME BY ZIPCODE by CUBIT Planning Inc information on their website. Insurance information at the time of the initial visit was obtained from the medical records and classified as Medicare, Medicaid (if supplementing Medicare in the context of Medicaid advantage plans, it was defined as Medicaid) or Commercial. We also used the EMRs to determine the number of 2019 visits each patient had prior to starting treatment. We did not assess whether treated patients had previous visits in 2018 and thus may have underestimated the number of visits prior to treatment for some patients. With respect to treatment, we defined success as treatment initiation within 6 months of the initial visit (i.e., through July 2020 for patients who had their first visit in December 2019). The analysis was performed using SAS JMP statistical software (v17.2.0). We compared the characteristics of patients who were treated with patients who were not treated using Student’s t-test for numeric values and a Contingency Analysis with Pearson Chi-Squared and Odds Ratio analysis.

3. Results

There were 587 patients with at least one visit in 2019. Treatment had been initiated in 140 patients prior to 2019, and an SVR was achieved in all but 6 of the 140 patients who initiated treatment prior to the 2019 visit (Figure 1). The six patients seen in 2019 who had previously failed to achieve an SVR were not included in the untreated patient group. Of the 441 patients who were not yet treated at the first 2019 visit (441/587 = 75%), only 189 (189/441 = 43%) were treated by July 2020. As shown in

Table 1. Number of patients seen per month in 2019 and treatment efficiency.

Month	Total Patients	Treatment Eligible	Treatment Eligible (% of Total)	Treated	Not Treated	Treated (%)	Months to Treatment
January	114	87	76%	56	31	64%	5.6
February	79	60	76%	32	28	53%	4.5
March	62	37	60%	19	18	51%	4.4
April	67	41	61%	12	29	29%	5.3
May	29	23	79%	10	13	43%	4.4
June	40	34	85%	11	23	32%	5.7
July	42	35	83%	15	20	43%	3.8
August	38	32	84%	17	15	53%	4.1
September	27	23	85%	4	19	17%	2.7
October	38	32	84%	6	26	19%	3.6
November	26	20	77%	5	15	25%	4.8
December	25	17	68%	2	15	12%	7.5
Year	587	441	75%	189	252	43%	4.7

and Figure 2, the number of total patients seen each month in 2019 declined from 114 in January to 25 in December. Also noted was a variation in the monthly percentage of patients who were eventually treated, although the possibility that patients were seen for multiple visits prior to 2019 was not evaluated.

Neither race nor gender was associated with failure to treat (Figure 1,

Table 2. Comparison between treated and not-treated HCV patients.

	Treated (n = 189)	Not Treated (n = 252)	Odds Ratio	p Value (t-Test or Odds Ratio)
African American (AA)	87%	84%	0.77	p > 0.05
Male	67%	63%	1.16	p > 0.05
Average Age (years)	61	60		p > 0.05
Seen in Gastroenterology (GI) orInfectious Disease (ID) Clinics	74%	74%	0.99	p > 0.05
APRI (Average) *	0.59	0.69		p > 0.05
FIB-4 Average) **	2.20	2.59		p > 0.05
Medicaid Insurance	55%	55%		p > 0.05 ***
Medicare Insurance	41%	37%
Commercial Insurance	2%	3%
Average Income ($)	$38,781	$38,414		p > 0.05
Number of 2019 Visits (Average)	4	2		p < 0.0001
Only 1 Visit in 2019	5%	45%	0.08	p < 0.0001
Viral Confirmation at First Visit	38%	27%	1.63	p< 0.018

* Aspartate Aminotransferase to Platelet Ratio Index (APRI). ** Fibrosis-4 Index (Fib-4). *** No Odds Ratio for multiple comparison analysis.

).

Table 3. Reason for no treatment by race and gender.

	Patient Follow Up	Co-Morbidity	Drug/Alcohol	Insurance	Low Fibrosis	Other	Reluctant
AA-Female	76%	14%	3%	2%	2%	0%	3%
AA Male	76%	8%	4%	7%	3%	2%	0%
Non-AA Female	65%	7%	0%	14%	7%	7%	0%
Non-AA Male	78%	4%	7%	7%	4%	0%	0%
All Patients	76%	9%	4%	6%	3%	1%	1%

presents the reasons that patients were not treated. The dominant association with not being treated was a lack of patient follow-up. With respect to comparing HCV patients who were treated or not treated, Table 2 presents the data for a wide range of potential factors that could impact treatment. Age for treated (61 years) vs. age for not treated (60 years) was not significantly different (p = 0.22). The primary association with not treating was patient failure to follow up (n = 93; 76%). Insurance issues when mentioned in the EMRs played a significantly lower part in failure to treat (15/252 = 6%). Other associations with not treating included drug or alcohol issues (n = 9) and low fibrosis scores with failure to return for treatment (n = 9). There were also two patients who were reluctant to agree to treatment. The dominance of follow-up failures was true for both gender and race (Figure 2, Table 2). Most of our patients had Medicaid (n = 133; 53%) as their primary insurance, consistent with their being socioeconomically challenged. Medicare was the second most likely category of insurance (n = 94; 37%) either due to age-based retirement or disability. There was no relationship between the categories of insurance and the reason for not being treated. Treated and not-treated HCV patients were also not different with respect to the clinic where they were seen (GI vs. ID), the type of insurance or median income as defined by ZIP code. Somewhat unexpected was that of the 252 patients, 123 (49%) had more than one visit and yet were not treated. The degree of fibrosis (assessed by the APRI and FIB-4) was not different between treated and not-treated patients. This suggests that fibrosis assessment was not impacting treatment decisions and that delaying treatment for fibrosis assessment was not needed. Patients with an average of four visits were more likely to be treated than those with two, and having one visit with no follow-up was the most dramatic factor for patient treatment (42% vs. 8% p < 0.0001). Viral infection confirmation available at first visit was an important factor with respect to treatment (treated 38% vs. not treated 25% p < 0.02). Patients with a first visit early in 2019 were more likely to be subsequently treated than patients seen later despite the criteria of treatment being extended through the first 6 months of 2020.

4. Discussion

Consistent with results reported in the literature, significant numbers of our patients (57%) failed to be treated after a clinic visit. This low rate of treatment continues to be a reason that the elimination of HCV remains elusive [9,10,11,12,13,14,15]. Perhaps more striking is the Center for Disease Control (CDC) report that there was a decline by half in the numbers of HCV patients treated between 2016 and 2020. While many of our patients were socioeconomically challenged as defined by Medicaid insurance, insurance was not a significant barrier to treatment, even though in 2019, many of the Medicaid-based insurance plans continued to require prior authorization for treatment. The primary reason patients were not treated was that multiple visits were needed before treatment was initiated, and many patients failed to return for treatment after an initial visit [14,15]. The other reason for failure to treat was the lack of viral infection confirmation by PCR at first visit. While there have been multiple studies identifying psychosocial issues, the major confounding issue remains the need for multiple visits prior to treatment and the failure of eligible patients to keep appointments. While issues such as substance abuse may be contributing factors, others include more seemingly mundane issues, such as transportation and reliable methods for communication with respect to reminders of appointments, which continue to plague patients.

With respect to the issue of whether GI or ID doctors should be a primary provider of HCV treatment, our study clearly demonstrates that in 2019, at our institution, most patients (75%) were seen by a GI doctor, with the remaining 25% being seen and treated by an ID doctor, and both providers were equally efficient. No patients were treated outside of the two specialties. Studies have suggested that an increase in treatment rates could be achieved by treatment in primary care settings, but data supporting that remains controversial, especially in settings like the US where multiple insurance payment models can create barriers to non-specialist care [11,12,13]. It is important to note that our study does not focus on the issue of identifying and linking patients to HCV treatment evaluation since most patients come to the clinic via a referral from a primary care provider. While the issue of linkage to care as defined by a visit to a GI or ID doctor was not addressed, this study demonstrates that many patients who have been identified and subsequently linked to a specialist for care fail to be treated for their HCV.

While there is speculation about the possibility that poor health literacy and/or the presence of social determinants of health issues (SDOHs) may play a role, studies to prospectively assess these issues with validated questionnaires are needed. The decline in patient visits from the beginning to end of 2019 and the fact that the number of patients treated decreased was unexpected and warrants further evaluation. With respect to this study, the patients had their first visits in 2019, which is prior to the beginning of the COVID-19 period in 2020. Studies to assess the issue of treatment in 2023, which would be post-COVID-19, are clearly warranted to obtain a clearer picture of treatment rates in the era of universal treatment recommendation.

The primary limitation to this study is that the impact of the COVID-19 pandemic, which began to impact medical clinic operations in the spring of 2020, cannot be assessed. While this impacts the number of patients, it does not negate the important observation with respect to treatment improvement with early viral confirmation and fewer visits. The second limitation is that insurance and ZIP codes for many of the patients are fluid and may change over a 1-year period. Also, while ZIP codes can provide an approximation of income, the results are based on averages and may not reflect accurately the income of the specific individuals. A third limitation is the issue of causative vs. correlative observation in retrospective studies. While these “reasons” may not be validated by questionnaires, the observation still provides important information with respect to a potential way to improve treatment rates in a primarily urban African-American-dominant clinic setting.

With the development of highly effective and short-term therapies for hepatitis C, a continuing problem remains in translating these therapies into real-world practice. This study clearly demonstrates that health centers should focus on having confirmation of viral infection as early as possible, because it will significantly improve the treatment rate. Thus, encouraging primary care providers to perform a viral infection confirmation prior to a specialty provider visit should be encouraged. The study also points out the value of health centers evaluating ways to initiate treatment as soon as possible, because fewer visits results in an improved treatment rate.

In conclusion, two important ways that treatment rates can be improved were identified in this study. Since the degree of fibrosis has no impact on treatment, initiating treatment immediately after confirming infection with HCV should improve patient outcomes and may be the most successful way to improve treatment rates. The failure to treat patients with multiple visits does need further investigation since the second visit should clearly have defined the patient infection by PCR. This study also suggests that a policy to initiate treatment immediately upon confirmation of infection via PCR would significantly improve linkage to care.