Patient Access to Hepatitis C Treatment After Incorporation of Pharmacists in a Hepatology Clinic

Abstract

Background: Modern hepatitis C virus (HCV) treatment regimens yield cure rates greater than 90%. However, obtaining approval for treatment through the prior authorization (PA) process can be time consuming and require extensive documentation. Lack of experience with this complex process can delay HCV medication approval, ultimately increasing the amount of time before patients start treatment and in some cases, prevent treatment altogether. Objectives: Assess the impact of incorporating clinical pharmacists into specialty pharmacy and hepatology clinic services on medication access, patient adherence, and outcomes in patients being treated for HCV. Methods: We performed a retrospective cohort exploratory study of patients seen in an academic medical center hepatology clinic who had HCV prescriptions filled between 8/1/15 and 7/31/17. Patients were categorized by whether they filled prescriptions prior to (Pre-Group) or after (Post-Group) the implementation of a pharmacist in clinic. The Post-Group was further divided according to whether the patient was seen by a pharmacist in clinic (Post-Group 2) or if the patient was not seen by the pharmacist, but had their HCV therapy evaluated by the pharmacist before seeking insurance approval (Post-Group 1). Results: The mean time from the prescription being ordered to being dispensed was longer in the Pre-Group (50.8 ± 66.5 days) compared to both Post-Groups (22.2 ± 27.8 days in Post-Group 1 vs 18.9 ± 17.7 days in Post-Group 2; P < .05). The mean time from when the prescription was ordered to when the PA was submitted was longer in the Pre-Group (41.6 ± 71.9 days) compared to both Post-Groups (6.3 ± 16 in Post-Group 1 vs 4.1 ± 9.7 in Post-Group 2; P < .05). Rates of medication adherence and sustained virologic response were similar between all groups. Conclusion: Incorporation of clinical pharmacists into a hepatology clinic significantly reduced the time patients waited to start HCV treatment. In addition to improving access to medications, implementation of the model helped to maintain excellent medication adherence and cure rates.

Introduction

Hepatitis C virus (HCV) is a blood borne virus that affects nearly 3.5 million people in the United States. ¹ After an acute infection, 75% to 85% of patients progress to chronic HCV. ² Within 30 years of initial infection, up to 25% of patients with chronic HCV will progress to cirrhosis and an additional 6% will develop hepatocellular carcinoma. ³ The goal of HCV treatment is to achieve cure in order to halt progression of liver disease. ¹ Modern treatment regimens for HCV yield cure rates greater than 90%, ultimately reducing the risk of cirrhosis and death. Even so, lack of access to these medications negatively impacts the number of patients treated and cured. In 2015, 20% of the global population with chronic HCV knew of their diagnosis, but only 7.4% of those diagnosed received treatment. ⁴

The cost of an HCV treatment course ranges from $26 400 to more than $100 000 per patient. As a result, insurance payers have focused on ways to improve patient outcomes while controlling cost.^5,6 One example is a requirement by some insurance payers to limit approval to patients meeting designated criteria, such as having a certain level of liver fibrosis. Obtaining insurance approval through the prior authorization (PA) process can be time consuming and require extensive documentation. Lack of understanding of this complex process can delay medication approval, ultimately increasing the amount of time before patients start treatment. In some cases this can avert treatment altogether.

Specialty pharmacies offer a variety of services not found in the traditional retail pharmacy setting. These services include clinical assessment of medication regimens, robust patient education, adherence support, management of adverse effects, outcomes evaluation, and cost and insurance assessment. ⁶ Together, the goals of these services are to increase medication access, promote adherence to medications, and improve patient outcomes.

The health system specialty pharmacy (HSSP) model (Figure 1) can provide additional benefits. This model consists of 4 critical pieces with the first component being the patients themselves. When put all together, this model aims to improve communication between patients, health care providers, pharmacies, and insurance payers to increase access and improve outcomes for patients on specialty medications.

Figure 1.

Health system specialty pharmacy model.

The second component of this model is the Pharmacy Patient Advocate (PPA). The PPA is a dedicated specialty pharmacy technician that serves as a liaison between patients, payers, ambulatory clinical pharmacists, and the specialty pharmacy to ensure access to medications. Responsibilities include determining benefit eligibility, obtaining insurance approval by completing PAs and appeals, and locating financial assistance through medication assistance programs (MAPs) such as manufacturer co-pay assistance cards and grants. The health system has multiple PPAs who specialize in specific disease states. By narrowing their focus, the PPAs become knowledgeable about the PA process for a select number of medications which increases their ability to process PAs efficiently and effectively. The utilization of PPAs reduces the need for ambulatory clinical pharmacists to complete PAs, thus providing pharmacists more time to participate in direct patient care.

The next component of the model is the ambulatory clinical pharmacist. These pharmacists work closely with the HSSP while being embedded in various ambulatory clinics throughout the health system. Pharmacists collaborate with providers to form patient care plans and determine treatment regimens. Before a patient begins a new specialty medication, an ambulatory clinical pharmacist completes an initial medication assessment to evaluate appropriateness. Additionally, the pharmacist completes education, medication reassessments and follow up through phone calls or in person visits for their respective clinic patients. Throughout the entire process the ambulatory clinical pharmacist works closely with PPAs to ensure patients have access to medications.

The fourth and final component of the model is the specialty pharmacy. The ambulatory clinical pharmacist and PPAs function as part of the specialty pharmacy, however the specialty pharmacy is also responsible for dispensing the medication to the patient. Furthermore, to enhance medication adherence, the specialty pharmacy contacts patients on a routine basis to manage refill requests and avoid lapses in therapy. It is important to note that the specialty pharmacy refill service used in this study was not automated. A member of the specialty pharmacy team would contact patients by telephone and confirm a refill was needed for a medication to be shipped.

One such example of this model is the collaboration between the pharmacy department and the hepatology program. Pharmacists in the hepatology clinic work closely alongside providers seeing patients with chronic HCV and providing patient education, evaluating labs and determining treatment regimens with the provider. Once a treatment plan is selected, the pharmacist works with the PPA to ensure patients receive their medications in a timely manner and at an affordable price and counsels the patient and/or caregiver.

The intent of the HSSP model is to increase medication access and adherence while maintaining or improving patient outcomes. Similar models have been established at other health systems within the United States and have demonstrated improved patient outcomes.^7,8 The purpose of this study was to assess the impact of this model on medication access and outcomes in patients being treated for HCV in a hepatology clinic.

Methods

A single center, retrospective cohort exploratory study of patients who had prescriptions for HCV treatment filled at the health system’s specialty pharmacy between August 1, 2015 and July 31, 2017 was performed. Adult patients who were seen in the hepatology clinic were included in the study. Patients were excluded if their insurance did not allow them to fill at the health system’s specialty pharmacy or they did not start on treatment. Due to transmission issues with the prescribing system resulting in the inability to gather accurate reporting data, patients prescribed sofosbuvir/velpatasvir between August 1, 2015 and February 1, 2016 were excluded.

Patients who filled prescriptions prior to implementation of a pharmacist in clinic (8/1/2015-2/1/2016) were included in the pre-implementation group (Pre-Group). Patients with a prescription between 2/2/2016 and 5/31/2016 were excluded as this was the time period when the pharmacist was being established in clinic. The Pre-Group was compared to the post-implementation group (Post-Group), which was comprised of patients whose prescriptions were filled after implementation of a pharmacist in the hepatology clinic (6/1/2016-7/31/2017). The Post-Group was further divided according to whether the patient was seen by a pharmacist in clinic. The pharmacist assessed the appropriateness of the prescribed HCV therapy for all patients in the Post-Group. The pharmacist assessed and collaborated on prescribing the therapy for patients in Post-Group 1 but did not see them in person due to the high volume of patients. Post-Group 2 comprised patients followed by a single advanced practice registered nurse in the hepatology clinic. The pharmacist would visit with all these patients in person and assist in determining appropriate HCV treatment regimens.

The primary endpoint of this study was the mean time to fill of HCV medications. The time to fill was defined as the time from the prescription being ordered to being dispensed. Secondary endpoints included mean time from prescription written to PA submitted and mean time from prescription written to PA approval. Because of the short duration of treatment, traditional adherence metrics such as proportion of days covered are not recommended to assess HCV treatment. Thus, we utilized the definitions recommended by Pharmacy Quality Alliance, which is the percentage of individuals 18 years and older who initiate and complete the minimum intended duration of treatment with no significant gaps in therapy. A significant gap in therapy is defined as greater than 15 days of cumulative gap between first and last fill of an HCV regimen. ⁹

To show a 50% reduction in time to fill with an alpha of .05 and 80% power from the Pre-Group to either Post-Group, 114 patients would be required in each group (228 total). Non-continuous variables were analyzed using a chi square test or Fisher’s exact test and continuous variables were analyzed using t-tests or analysis of variance tests. A P-value <.05 was considered statistically significant. All analysis was completed with SPSS v.24, Armonk, NY. This study was approved by the health system’s Institutional Review Board.

Results

Of the 433 patients prescribed HCV treatment during the study period, 121 met inclusion criteria (Figure 2). A total of 312 patients were excluded, most often due their insurance requiring their prescription be filled at a different specialty pharmacy. Ten patients were excluded from the Post-Group because their insurance denied coverage following an appeal. Additional options such as medication assistance from the manufacturer were being evaluated for these patients. As there was no resolution they were unable to be included in the analysis. The 38 patients that did not start HCV treatment during the study period were for reasons such as lack of follow-up, pregnancy, or breastfeeding, resulting in a delay of treatment initiation.

Figure 2.

Patient selection.

Of the 121 patients included, 22 were in the Pre-Group, 43 were in Post-Group 1, and 56 were in Post-Group 2. Baseline characteristics were generally similar between groups (Table 1). Post-Group 1 was the only study arm to contain liver transplant recipients. The most common HCV genotype among the study population was genotype 1a. No patients in the study were found to have genotype 4, 5, or 6. Across the 3 groups, 8 different HCV treatment regimens were utilized, and the most common treatments included ledipasvir/sofosbuvir, sofosbuvir/velpatasvir, or elbasvir/grazoprevir-based regimens (Table 1).

Table 1.

Patient Demographics (Total, N = 121).

Characteristics	Pre-Group (N = 22)	Post-Group 1 (N = 43)	Post-Group 2 (N = 56)	P-value
Male gender, % (n)	55 (22)	58.1 (25)	48.2 (27)	NS
Age (years), mean (SD)	59.8 (9)	59.7 (10.8)	61.1 (7.8)	<.05 a
Treatment naïve, % (n)	77.3 (17)	76.3 (29)	76.7 (33)	NS
Cirrhosis, % (n)	40.9 (9)	39.5 (17)	25 (14)	NS
Post-liver transplant, % (n)	0 (0)	9.3 (4)	0 (0)	NS
Hepatitis C virus genotype, % (n)
1 a	63.6 (14)	53.5 (23)	58.9 (33)	NA
2	4.5 (1)	16.3 (7)	8.9 (5)
3	9.1 (2)	14 (6)	10.7 (6)
Hepatitis C virus treatment, % (n)
Ledipasvir/sofosbuvir	81.8 (18)	20.9 (9)	37.5 (21)	NA
Ledipasvir/sofosbuvir + ribavirin	4.5 (1)	16.3 (7)	1.8 (1)
Sofosbuvir/velpatasvir	N/A	27.9 (12)	19.6 (11)
Sofosbuvir/velpatasvir + ribavirin	N/A	14 (6)	8.9 (5)
Elbasvir/grazoprevir	0 (0)	18.6 (8)	28.6 (16)
Elbasvir/grazoprevir + ribavirin	0 (0)	0 (0)	1.8 (1)
Sofosbuvir + daclatasvir	8.1 (2)	2.3 (1)	0 (0)
Sofosbuvir + daclatasvir + ribavirin	4.5 (1)	0 (0)	1.8 (1)

Note. NS = not significant; SD = standard deviation; NA = not available.

^aBetween Pre-Group and Post-Group.

The mean time to fill was longer in the Pre-Group (50.8 ± 66.5 days) compared to both Post-Groups (22.2 ± 27.8 days in Post-Group 1 and 18.9 ± 17.7 days in Post-Group 2; P < .05). No difference was detected between Post-Group 1 and 2 (Figure 3). Compared to the Pre-Group, the mean time to fill was reduced by 56% in Post-Group 1 and by 63% in Post-Group 2.

Figure 3.

Access to treatment.

The mean time from when the prescription was ordered to when the initial PA was submitted was significantly longer in the Pre-Group (41.6 ± 71.9 days) compared to both Post-Groups (6.3 ± 16 days in Post-Group 1 and 4.1 ± 9.7 days in Post-Group 2; P < .05); with no significant difference between the Post-Groups. Furthermore, the mean time from when the prescription was ordered to when the PA was approved was longer in the Pre-Group (43.7 ± 104.4 days) compared to Post-Group 1 (16.6 ± 26.1 days; P = .1) and Post-Group 2 (13.7 ± 17.2 days; P = .04). Compared to the Pre-Group, the mean time to PA approval was reduced by 62% in Post-Group 1 and 69% in Post-Group 2.

Of the study population, 1 patient did not complete therapy due to passing away during treatment. Of the patients that completed treatment, no patient had a significant gap in therapy. Among patients who had an HCV viral load assessed at least 12 weeks after completing HCV treatment, the rate of sustained virologic response (SVR) was no different between groups (Table 2).

Table 2.

Outcomes and Cost of Care.

Characteristics	Pre-Group (n = 22)	Post-Group 1 (n = 43)	Post-Group 2 (n = 56)	P-value
Patient outcomes
Sustained virologic response rate a % (n/total)	95.2 (20/22)	94.6 (35/37)	96 (24/25)	NS
Financial coverage
Payer % (n)
Medicare	22.7 (5)	55.8 (24)	60.7 (34)	<.05b,c
Medicaid	18.2 (4)	23.3 (10)	10.7 (6)
Commercial	59.1 (13)	20.9 (9)	28.6 (16)
Medication assistance program (MAP) % (n)	27.3 (6)	46.5 (20)	42.9 (24)	NS
MAP method % (n)
Grant	16.7 (1)	27.9 (12)	23.2 (13)	NS
Copay card	83.3 (5)	18.6 (8)	17.9 (10)
MAP amount received per patient ($) mean ± SD	$6750 ± 7728	$3454 ± 2929	$4378 ± 478	NS
Copay ($) median	$0	$0	$3	NS
$0 copay for course of treatment % (n)	68.2 (15)	62.8 (27)	50 (28)	NS

Note. NS = not significant; SD = standard deviation.

^aDefined as an undetectable HCV viral load at least 12 weeks after completion of HCV treatment

^bBetween Pre-Group and Post-Group 2.

^cBetween Pre-Group and Post-Group 1.

Patients in the Pre-Group were more likely to have commercial insurance compared to those in the Post-Groups, who were more likely to have Medicare (Table 2). Overall, 41% of patients received additional funding through MAPs with no difference in the mean MAP dollars each patient received between groups. Furthermore, the median copay was $0 in all 3 groups with 61% of patients having zero out-of-pocket cost.

Discussion

High-cost specialty medications such as those used to treat HCV can be challenging for patients to access. Insurance payers commonly require a PA for approval and some restrict approval to patients meeting certain criteria. Implementation of the HSSP model within the hepatology clinic improved access to therapy through decreased time to approval and filling of HCV treatment with a greater impact when the pharmacist met with the patient and provider in clinic.

Herein we saw that patients who had a pharmacist involved during their hepatology clinic visit had an even quicker time to start medication. For the 3 treatment groups, the only variation in the HSSP model was the level of ambulatory clinic pharmacist involvement as the PPAs and specialty pharmacy involvement was consistent across the 3 groups. This reduction in time to start medication is likely a result of the pharmacist evaluation of lab work and collaboration in determining a regimen during the visit. During each visit, the pharmacist would evaluate whether the patient had the necessary laboratory work and would work with the provider to determine the most appropriate treatment regimen. If laboratory tests were needed, the pharmacist would work with the clinic staff to ensure tests were completed during the visit. For patients not seen by a pharmacist, the pharmacist reviewed the patient’s chart after the medication was ordered and if laboratory results were missing, the clinic staff and patient were contacted to complete additional tests, many times resulting in a delay in submitting the PA and starting treatment. Additionally, if the medication regimen that was ordered was not optimal delays in therapy could occur.

Bagwell et al reported the positive impact of an integrated specialty pharmacy services model on HCV patient care. The Vanderbilt University Medical Center (VUMC) model described was similar to ours and included patients being treated for HCV in an infectious disease (ID) clinic. After implementation of the model, the authors found a 78% reduction in median time to medication approval. Additionally, a 68% reduction in mean time to medication initiation was noted after the patient’s initial clinic visit. ⁷ This data is comparable to our findings with a 69% reduction in time to medication approval and 63% reduction in time to fill.

In addition to obtaining approval for treatment, patient cost can be a barrier to treatment. Research suggests that patients are less likely to take a medication when they have to pay more than $100 for a prescription, concluding that high out of pocket cost can lead to reduced adherence rates.^10,11 Zhu et al described a partnership between a specialty pharmacy and a hepatology clinic where the pharmacy sought financial assistance for patients receiving HCV treatment. Through coordinated care between the clinic and the specialty pharmacy, 98.1% of patients being treated for HCV had an out-of-pocket cost of $20 or less per month. ¹² Bagwell et al reported that after implementation of the integrated specialty pharmacy service model in the VUMC ID clinic, 53% of patients had no out of pocket cost for their HCV medications. ⁷ In our study, there was a significant difference between the types of insurance coverage between the Pre and Post-Groups. The majority of patients in the Pre-Group had commercial insurance while the Post-Groups had Medicare or Medicaid. This difference is likely due to government-sponsored plans expanding coverage over time. Regardless of type of insurance coverage, patients in all 3 groups had PPA assistance to reduce medications costs and 98.8% of patients had an out-of-pocket cost of $20 or less and over half of patients had no out of pocket cost. Additionally, the median patient copay was $0 for all treatment arms. Ultimately, if patients are able to afford their medications, they are more likely to be adherent with their regimen, eventually translating to improved outcomes.

One aspect of optimizing the chance of achieving SVR is medication adherence. With only one patient not completing treatment, the high level of adherence seen across all three treatment arms can be contributed to the refill adherence program supported by our HSSP model, allowing close follow up and ensuring timely refills to prevent lapses in therapy. The SVR rates observed in the presented study were 90% or higher in all three groups, which is consistent with real word data showing an overall SVR rate of 93.7% with direct acting antiviral therapy. ¹³ Similar studies of patients treated for HCV in a pharmacist-driven treatment model yielded SVR rates of 93% and 95.1%.^14,15 While our small sample size likely contributed to an inability to show a significant difference in cure rates between groups, because cure rates were so high, we would not expect to see a difference in patient outcomes between groups even with the addition of more patients. Due to lack of patient follow up, SVR data was not available for all patients in the Post-Groups. Even though the SVR rates were similar across all groups, the significant reduction in mean time to fill of HCV medications in the Post-Groups could be clinically relevant as it may lead to a faster time to cure, potentially reducing the risk of patients progressing from chronic HCV. Moreover, decreasing the time to start medication decreases the overall burden on the healthcare system and increases patient satisfaction.

There are several limitations of our study. First, the retrospective nature of our study carries the inherent limitations of any retrospective study, such as incomplete patient data and the inability to control for confounders. One confounder of the study may have been knowledge of the PA process. As health care providers, clinic staff, and insurance payers handle more HCV-related PAs, they may become more efficient at completing and processing the required documents. This could have contributed to reduced time to fill in the Post-Groups. However, results of this study show that compared to the Pre-Group, mean time from prescription written to PA approval was significantly reduced in Post-Group 2, but not Post-Group 1, which helps reduce the likelihood of this confounder and supports the benefit of pharmacist seeing patients directly in clinic. Second, time itself could have impacted study results. Even though the study period was only 2 years, insurance formularies and PA requirements likely changed over the time period, increasing access to these medications.

Third, the sample size may have been too small to show a difference between groups, impacting the significance of the results. Forty-eight patients screened for inclusion in the Post-Group were excluded due to insurance denial of prescription coverage or lack of appointment follow up leading to inability to initiate HCV treatment. Although this accounted for 24% of patients screened, a cohort study by Zuckerman et al reported 40% of patients who were referred to the VUMC ID clinic did not begin HCV treatment. ¹⁵ Similar to Zuckerman A et al, lack of appointment adherence and/or incomplete patient workup were the most frequent reasons why patients did not proceed to treatment. Since clinic pharmacists keep track of HCV patients seen in clinic, there is opportunity to collaborate with clinic support staff to provide greater outreach to HCV positive patients and help increase appointment adherence. Additionally, with the implantation of a second pharmacist in the hepatology clinic, more HCV patients can be seen by pharmacists in clinic which can help ensure laboratory workup is completed during patient appointments.

Ten patients in the presented study were prescribed HCV therapy, but were denied coverage by their insurance payer despite appeal attempts. For those patients, the ambulatory care clinical pharmacist and PPA(s) will assess their eligibility for patient assistance programs. Some of these patients excluded from the study may have qualified for a program and started treatment, however this study did not follow these patients beyond the designated time frame. While we provide support for patients whose insurance mandates they fill at an outside specialty pharmacy and patients without insurance who receive medication through a manufacturer patient assistance program, we did not include those patients in this analysis. Because we can have a larger impact on patients that fill with our health system’s specialty pharmacy, these results would likely be impacted in those patients and limits the generalizability of our results.

Conclusion

The HSSP model is a collaborative effort between the specialty pharmacy, ambulatory clinical pharmacist, multidisciplinary team, patient, and insurance payer. This study demonstrated that enhancement of specialty pharmacy services through the addition of a pharmacist in clinic decreased the time it took for patients to start HCV treatment. In addition to improving access to medications, implementation of the HSSP model maintained excellent medication adherence and SVR rates in patients being treated for HCV.