Activities for raising public awareness of HCV included: advertisements on billboards, public transportation and local radio; and interviews on radio, print and television. Materials used included those from the CDC’s ‘Know More Hepatitis™’ campaign²⁷ and additional materials developed by HepCCATT. Messaging was primarily targeted to persons born between 1945 and 1965 and other risk groups, focusing on the benefits of HCV testing and on the reduction of stigma associated with an HCV diagnosis. Radio interviews also mentioned other risk factors, such as injection drug use. Evaluation data were not collected for this activity.

Workforce development training based on the ECHO²⁶ model used by ECHO-Chicago²⁸ was provided to the health centers’ PCPs and CM teams. Safety net health providers were actively recruited, though all interested could participate. Curricula were developed by the ECHO-Chicago multidisciplinary subject matter expert (SME) team: a hepatologist, pharmacist, case manager and addiction specialist.

Curricula evolved over time, adapting to new treatments and policies. As treatment simplified, focus shifted to streamlining treatment and monitoring protocols, navigating authorization processes and screening emerging at-risk populations. Participants completed the 10-session training in cohorts of 10–25 persons by joining weekly, hour-long sessions via videoconferencing. Sessions were facilitated by the SME team and were comprised of a short lecture followed by at least two participant-led new case presentations. Discussion of prior patient cases occurred as needed. Protected health information was not shared.

The hepatologist and other members of the SME team facilitated DAA prescriptions by confirming specialist consultations and by providing support in navigating barriers.

Case managers were hired through the Community Outreach Intervention Program (COIP) at the University of Illinois School of Public Health. Described elsewhere,²⁹ direct CM was provided by a team of case managers to 181 HCV chronically infected clients receiving care at a HepCCATT-participating site.

Over 5 years, HepCCATT partners engaged the local and state policymakers to evaluate changes to Medicaid requirements, participated in statewide HCV coalitions, connected legal teams to patients’ providers and raised awareness of the common barriers to HCV care through local media. Resources did not allow us to collect the necessary data to evaluate the impact of this work.

In partnership with the Illinois and Chicago Departments of Public Health (IDPH and CDPH), a HCV registry (Registry) was developed to track progression through the care cascade, which was defined as moving from diagnosis to engagement in HCV-specific care (defined below), treatment and cure. The Registry brought together existing state-collected HCV surveillance data (laboratory data and provider reports) with electronic health record (EHR) data from 108 participating clinical sites (90% FQHCs) from 13 health centers. Prescription information from two major pharmacy chains was included. Data were collected and linked as part of the state’s Medical Study Act³⁰ for public health surveillance and were kept in a state-sponsored public health node.

Data were collected for the year prior to the start of HepCCATT (baseline) and then for project years (PYs) 1–5. The Registry code book was based on the requirements set forth by the CDC’s National Center for HIV/AIDS, Viral Hepatitis, STD, and TB Prevention and was consistent with the IDPH requirements. In addition to analyzing data for surveillance purposes, starting in PY4, the Registry generated monthly alerts to HepCCATT sites about their HCV patients who had either fallen out of, or failed to initiate, the next level of care.

Individuals were added to the Registry if they received a positive HCV antibody, RNA, or genotype test during the baseline or PYs of HepCCATT. Negative antibody screening was only provided in the aggregate by PY.

A de-identified restricted dataset from the Registry was provided for analysis that excluded patients with negative antibodies, health center identifiers (including FQHC status), patient identifiers, PCP identifiers (including ECHO–HCV training status) and masked sites of care and dates of service. Instead of dates, there was an indicator for the number of weeks between the initial diagnosis and each care event (e.g. treatment initiation). The PY of entry into the Registry was included, as were gender (male, female and any transgender as ‘other’), birth year, race, ethnicity, zip code and insurance (Medicare, Medicaid, private or uninsured). Sustained viral response (SVR) was defined as undetectable viral load documented at least 3 months after the completion of therapy.

Baseline was defined as 29 September 2013 to 30 September 2014, with each subsequent year (PY1–5) beginning on 30 September. PY5 (30 September 2018–29 September 2019) was only used for follow-up data.

This analysis is restricted to patients with current HCV infection as defined by a positive RNA or genotype test (‘RNA+’) and includes descriptive comparisons based on the birth cohort. Birth cohorts were originally defined as born ‘before 1945’, from ‘1945 through 1965’ and ‘after 1965’. Because of low numbers for the oldest group, the two older cohorts were combined as born ‘1965 or earlier’.

‘Engaged in care’ was defined as receiving any of the following: a genotype test, any fibrosis staging test, or HCV-specific treatment. Likely cirrhosis was defined as an AST to Platelets Ratio Index (APRI) score of 1.0 or above.³¹ Numbers of RNA+ people by zip code were categorized into five levels (created by ArcGIS and based on natural groupings within the data³²) for each separate PY

For this analysis, participants were limited to those from clinics within Chicago and were presented in five levels (again created by ArcGIS and based on natural groupings within the data³²) cumulatively, as HCV knowledge following training was assumed to persist over time. Providers’ health center types were categorized as (i) safety net organizations defined as FQHCs, safety net hospitals (either by state determination³³ or by being in the top 25% of Medicaid inpatient and charitable care in the state³⁴), or FCCs; or (ii) health centers serving populations with increased HCV risk (adults 65+, dual eligible patients on Medicaid/Medicare, or patients with substance abuse disorder) and (iii) all others.

Descriptive statistics were calculated in STATA 15.1 (College Station, TX), and maps were made in ArcGIS ArcMap 10.7 (Redlands, CA).

The number of individuals with HCV antibody testing was 13 278 in the baseline year, 17 778 in PY1, 22 095 in PY2, 31 474 in PY3 and 22 296 in PY4. Antibody testing increased 137% from baseline to PY3 before leveling off in PY4.

Table 1 shows RNA+ patients were mainly male (64%), born in 1965 or earlier (68%), and non-Hispanic African American (44%), though race/ethnicity was unknown for 25%. Medicaid was the most common insurance (33%), with 32% missing. Engagement in care was 74%, likely cirrhosis was 17% and treatment initiation was 11% (559 among 5307 engaged in care).

Total RNA+ diagnoses increased to 61% from baseline to PY3 before leveling off in PY4. The majority of newly diagnosed cases were born between 1945 and 1965 in all PYs. However, diagnoses among the younger cohort (born after 1965) increased 84% from baseline to PY4. Diagnoses in those born on or after 1986 increased to 180% from 56 diagnoses in the baseline period to 157 in PY4.

Demography varied by birth cohort. In contrast to those born after 1965, persons born 1965 or earlier were notably more non-Hispanic African-American (51 versus 31%). Those born after 1965 had higher percentages Hispanic (23 versus 12%) and non-Hispanic white (24 versus 10%). Insurance also varied by birth cohort, with being uninsured higher among those born after 1965 (16 versus 10%); there were some expected differences, such as higher levels of Medicare in patients born 1965 or earlier (22 versus 3%). Among those born 1965 or earlier, both engagement in care (75 versus 70%) and likely cirrhosis (18 versus 14%) were slightly higher, but treatment and SVR levels were similar.

The HCV care cascade shows a high proportion engaged in care (74%), but lower percent for those treated (11%), Fig. 2. Among those treated, known SVR is 65%.

Over the 5-year project, 215 PCPs with prescriptive capacity and 56 other health care professionals in Chicago were trained. Providers working in the safety net comprised 86% of participants (68% FQHC, 17% safety net hospital and 1% FCC). An additional 10% of participants served groups at increased HCV risk, and 4% were from pharmacies and smaller clinics.

HCV RNA+ diagnoses by zip code and participants are shown in Fig. 3. The zip code color darkens with more cases, and participants are shown in increasing size proportional with persons trained. The cumulative map (right) shows health centers by type. The RNA+ diagnoses are from HepCCATT participating sites only and account for 37% of the total cases in Chicago.

The HepCCATT project was associated with an increase in testing, diagnosis and treatment of patients with HCV across the safety net sites in Chicago.

Using the ECHO-Chicago program, HCV training was completed with many safety net providers. HepCCATT successfully reached the most vulnerable populations in Chicago, as seen in Fig. 3.

Providers showed significantly increased self-efficacy in providing hepatitis C care (data not shown). This is a relatively low-cost approach to rapidly expanding the number of providers over a geographic area capable of providing HCV care.

During the course of HepCCATT, efforts shifted from providing direct CM to increasing capacity as recognition grew that the need for CM services far exceeded the capacity of our CM team. Interest in CM training resulted in 14 of 36 (39%) participating organizations participating in this additional voluntary training.

Illinois Medicaid restrictions proved a challenge in treating patients. In November of 2018, Illinois removed the advanced stage fibrosis and substance use requirements for Medicaid patients.²⁰ While these policy changes cannot be attributed solely to HepCCATT efforts, our input was often requested in meetings with the state to loosen DAA restrictions. Partnering with community organizations, HepCCATT shared policies from other states as well as the cost and outcome studies documenting the value of DAA treatment and provider and patient testimonials.

The Registry documented HCV at 109 sites of care and included 37% of all known Chicago cases, confirming higher prevalence in men and African Americans. Antibody testing rose each year at HepCCATT sites until likely reaching saturation in PY4. We believe this was an effect of the project, but without additional data, we cannot be sure. However, nearby large cities did not show the same trend. In the approximate time periods of HepCCATT, 2014–17, there are data available for St. Louis and Milwaukee Counties. HepCCATT sites showed a 61% increase in cases, while St. Louis and Milwaukee Counties both reported a decline.^35,36

The elimination of HCV is possible via DAAs. However, there are not enough specialists to treat all infected individuals, so they must be able to receive disease counseling, screening and care from PCPs.^23–25

HepCCATT adds a rapid and scalable method to train PCPs in treating HCV patients, especially providers that serve high-risk populations. By merging HCV-relevant information from different EHRs with the pharmacy and public health data, HepCCATT made determinations of engagement in care, treatment initiation and cure possible in difficult-to-treat populations.

HepCCATT was designed as a Chicago-specific public health intervention and not as a research study. It coincided with new treatments, health trends and initiatives. HepCCATT may not be fully generalizable to other settings. However, the lessons learned in training PCPs and CM and in establishing a multi-organization disease Registry can serve future efforts for HCV elimination.

Our data do not allow us to definitively document whether treatment was missing or not prescribed for 92% of the RNA+ population. However, Illinois reports that the overwhelming majority of Medicaid patients obtain their medications through the pharmacy chains that submitted data to the Registry.

Only 65% of those treated were documented to achieve SVR, most likely due to the lack of follow-up RNA testing at the site of care. Treatment completion data were not available.

With dataset restrictions, we could not link ECHO participants with patient diagnoses. The analytic dataset contained only positive antibody tests; total screening numbers were only provided in the aggregate by PY, limiting the analysis. The reasons for the cohort differences between those born before and after 1965 could be related to the differences in risk factors (e.g. opiate use) or influenced by demographic characteristics.

We were unable to collect the data required to directly measure the impact of the public education campaigns or of increased awareness among PCPs.

Some of the missing staging information is attributable to providers’ EHR practices. Clinical notes and free text fields were not included; only information in discrete fields was delivered.