Australian general practitioners’ perceptions of the barriers and opportunities for community-based care of patients with viral hepatitis: a mixed methods study

Keywords: access to healthcare, culturally and linguistically diverse populations, engagement of primary care, general practitioners, mixed methods, models of care, primary health care, shared model of care, viral hepatitis.

Chronic hepatitis B virus (HBV) and hepatitis C virus (HCV) infections remain important health issues, especially in people from culturally and linguistically diverse (CALD) backgrounds.^1,2 The major contributors to the higher burden of chronic health complications from viral hepatitis (VH) in this group are delay in diagnosis, which is related to limited access to VH diagnostic and follow-up services,³ and socio-cultural barriers to engaging in the health system⁴ – often evidenced as failure to attend healthcare appointments and follow up. For example, in a nationwide study of HBV in Australia, in Brisbane in 2017, treatment uptake for HBV ranged from 2.0 to 17.0% and HCV from 12.0 to 38%,⁵ which is below the 2022 national target.^6,7 Although data on treatment uptake was not reported by country of birth, previous studies have shown that a large proportion of people living with HBV who accessed tertiary hospital liver services are clustered within specific geographic areas, in regions with a relatively high proportion of people born in Vietnam and China.⁸

In Australia, authorised general practitioners (GPs) can prescribe the treatment of HBV infection;⁹ however, the screening, diagnosis and treatment of HBV infection is not funded in primary care and there are cost implications for patients visiting their GP. Therefore, patients are referred to tertiary hospitals for specialty hepatology services. The treatment of HCV infection can be provided in primary care by GPs and nurses with experience and training in the treatment of VH. Liver fibrosis follow-up tests are not funded in primary care and there is a need for referral of patients to tertiary hospitals.^10,11

Bringing hepatology specialty services closer to community healthcare services for high-risk, hard-to-reach CALD communities is believed to improve equity of access, and clinic attendance rates.^6,12,13 General practitioner participation in VH care has been demonstrated to improve treatment rates and patient engagement in CALD populations.¹⁴ In response to this, a new model of care known as the Hepatology Education Research and Clinical Outreach (HEPREACH)¹⁵ programme has been trialled. The HEPREACH project aimed to overcome access barriers to hospital-based clinics by bringing specialty hepatology services closer to the most at-risk communities in the primary care setting through shared care of VH between hepatologists and GPs. While implementing HEPREACH, it was identified important that GPs’ engagement and opinion on the referral process, barriers and opportunities needed to be explored and better understood. This study explored the perception of GPs about the referral process and a shared, community-based model of care for patients with VH.

This was a mixed methods study where survey questionnaires were used concurrently with face-to-face, key informant interviews with GPs in Brisbane, Queensland, Australia. A semi-structured interview guide was used for key informant interviews.¹⁶ Our aim was to secure the advantages of each method and also to minimise the limitations of a single approach (triangulation) to enhance the validity of the responses in relation to our key research question.^17,18

Primary care practices with a high prevalence of HBV or HCV and low rate of uptake of VH treatment in the surrounding population were identified using data from the Viral Hepatitis Mapping Project.⁵ As of May 2018, there were 304 primary care practices in the Brisbane North Primary Health Network and 329 in the Brisbane South Primary Health Network.¹⁹ Interviews were focused among GP practices where a high caseload may be anticipated. Accordingly, six primary care practices from Brisbane North and 21 from Brisbane South Primary Health Network (PHN) were selected (Fig. 1).

Fig. 1.  Study participant selection and interview process. A total of 27 clinics were approached and 11 clinics agreed to participate in the study. Consent was obtained from the 23 GPs interviewed.

Fig. 1 presents the participant recruitment process. Practice managers at the purposively selected primary care practices were contacted via email and a follow-up telephone call was made to confirm the numbers of practising GPs in those practices and arrange interviews with GPs wishing to participate. All GPs from the selected practices were invited to participate in the survey, following consecutive sampling criteria.²⁰

The recruitment of GPs for key-informant interviews took place simultaneously with the sampling for the survey, but participants did not need to take part in both. The inclusion criteria were: both male and female GPs, a range of years of experience, having treated and referred a patient with VH in the past, and coverage of suburbs in the study area.²¹

The survey questionnaire was developed based on a review of relevant literature in the field.²² Initially, the content of the study instrument was reviewed by experts, including a hepatologist with experience in such studies. Then, two GPs who currently practice at the Mater Refugee Health and have experience of caring for patients with VH reviewed the survey tool. Quantitative data were collected by face-to-face interviews with GPs in the clinical setting and those interviews took about 15–20 min to complete.

An open-ended qualitative interview guide was used to explore the challenges and solutions to the VH patient referral process and community-based model of care using questions such as: ‘What barriers exist for you in the primary care setting to provide clinical care for viral hepatitis patients at your practice?’ with some prompts; for example, ‘patient-related barriers?’, ‘provider-related barriers?’, etc. Two researchers (BT and BL) conducted the key-informant interviews. They were not known to the participants before the data collection. To capture the discussion and verbatim responses from informants, notes were taken separately by the researchers during the interviews. Qualitative data collection was stopped when saturation of information was reached, and no new themes were emerging.

The rating scales for Likert scale questions were assigned a score of one to five as follows: 5 is most important, 4 is highly important, 3 is important, 2 is slightly important, and 1 is least important. For the variable ‘I would prefer my patients with viral hepatitis NOT to be seen by other GPs even with a special interest in viral hepatitis’, which shows a negative meaning, the scores were inverse coded. The ordinal perception measurements were presented with descriptive statistics. The internal consistency of the study instrument was tested using reliability analysis, and a Cronbach’s alpha score of ≥0.7 was considered acceptable.²³

Thematic analysis was used to analyse the qualitative data.²⁴ An inductive approach was used to develop themes from the content of the transcripts. Two researchers (BT and BL) analysed the transcripts. Initially, an immersion process was undertaken where transcripts were actively read and researchers became familiarised with the content.²⁵ The content was then divided into meaningful analytical units while coding the data,²⁵ and also identified key quotes from participants. The codes were organised into thematic categories based on their similarity.^26,27 Quantitative and qualitative data (data triangulation) were then combined to obtain a comprehensive understanding of the referral process.¹⁸

The Mater Misericordiae Ltd Human Research Ethics Committee gave the ethical clearance to conduct the study (review reference #HREC/MML/58831).

A total of 23 GPs responded to the survey questionnaires and 10 GPs responded to the qualitative interviews. A total of 27 primary care clinics were approached, with the expectation of an estimated 55 physicians (21 clinics in Brisbane South and six clinics in Brisbane North) and an anticipated 50% participation rate. A total of 23 GPs in 11 clinics (eight clinics in Brisbane South and three in Brisbane North) were interviewed for an average duration of 15–20 min. A total of 10 key-informant interviews were conducted each for an average 20 min. All the clinics were group practices, except one clinic in Brisbane South, which was a single GP practice. All the GPs reported having treated at least one VH or liver disease patient.

Of the 23 GPs who responded to the survey, most (86.4%) believed the treatment of VH patients using the community-based model of care is beneficial. Nearly two-thirds (63.7%) of GPs agreed that they would prefer to share the care of VH patients with a hepatology specialist. More than two-thirds (68.2%) of GPs believed the existing referral services for VH patients were effective, whereas nearly one-in-five (18.2%) disagreed about this (Table 1). The reliability analysis of the study tools can be found in Supplementary File S1.

Table 1.  General practitioners’ opinions about the community-based model of care and referral for viral hepatitis, Brisbane, Queensland, Australia, N = 23.

Patient loss to follow up (39.1%) was identified as the most important challenge of providing care for VH patients. More than one-in-five (21.7%) physicians described a lack of training for GPs about VH as the most important barrier to care for patients at the primary care followed by patient waiting time (17.4%) for an appointment at the hepatology specialty clinics after referral (Table 2).

Table 2.  Perceived barriers to care for patients with viral hepatitis, Brisbane, Queensland, Australia, N = 23.

In the qualitative exploration, the major barriers identified were patient loss to follow up, lack of bulk-billing facilities, and cost-effectiveness of treating VH. The most consistently cited concern in the care of VH patients for GPs was patient loss to follow up.

…Patient loss to follow up is a critical issue here. You don’t sometimes get patients returned even for a sustained virologic response test at the end of treatment [GP]. …at the primary care, if you are going to investigate and treat patients with viral hepatitis, investigations can be costly, causes more delays, patients don’t do it… and are loss to follow up [GP].

One of the challenges to VH care was the cost-effectiveness of diagnosis and treatment of VH in a busy clinic, given the number of tests needed to be done, the training required, and the options to treat other common diseases more effectively in a relatively shorter time. Treating VH patients is perceived to have taken a long time due to the extensive investigation process and initiating the treatment regimen. Similarly, GPs pointed out that many investigations for VH (such as transient elastography) are not bulk billed and reimbursed by Medicare. The likelihood of following up in the long run also played a crucial role in the engagement of GPs in VH care.

Another challenge is the need for long training period for GPs in the changing clinical management landscape. General practitioners needed to maintain a knowledge base throughout care and treatment of an adequate number of patients (this does not usually happen due to a low caseload of VH in their practice), requiring them to review management guidelines again and again, which, they described, was time-consuming, particularly in a time-pressured busy general practice. Due to the time constraints, they often preferred to refer patients with VH and liver diseases to specialty services rather than ‘reinventing the wheel’ each time.

…small number of (VH) patients are coming, but we need long time to manage…we have a busy schedule in the clinic, it is better to refer patients to hospitals than going through the guideline again and again…[GP].

Primary care physicians thought one of the reasons for not being motivated to attend training on VH care was the expected small caseload of VH they will handle after the training.

…I spend a day or two to attend a training on viral hepatitis but back here, there may be only a single case per week or less, which is not cost-effective… [GP].

Better access to diagnostic services for VH patients (69.6%) could ease the challenges of providing care as GPs. Bulk billing (full funding) VH diagnostic or treatment services in primary care was suggested by physicians (56.5%) to improve VH treatment and care uptake (Fig. 2).

Fig. 2.  Opportunities for better engagement and patient referral process for viral hepatitis patients. The opportunities that GPs believe will improve their engagement in VH care were access to diagnostics with a bulk billing, or cost reduction, and focused training.

The major strategies to overcome barriers to care for VH suggested by informants in the qualitative interviews were bulk billing VH and liver disease diagnostics, shared care of VH patients, and focused GP training. Informants felt that bulk billing tests, and if this were not possible, reducing the cost of diagnosis and treatment access, could improve testing, diagnosis and follow-up care for liver diseases.

…the tests required for viral hepatitis diagnosis can be costly and bulk billing improves testing rates for viral hepatitis. …It also improves follow-up to monitor progression to advanced liver disease… [GP].

General practitioners believe a shared model of care where specialist hepatologists are treating patients in primary care is an effective approach to solve patient- and provider-related barriers to care for VH. Specialists treating VH patients in primary care will bring specialist services closer to communities. It also helps GPs to consult with specialists and gain skills in caring for VH over time, leading to the capability to treat independently.

Focused training for GPs with a special interest or for those practising in a high-prevalence area was identified as a need. This was felt to be a way to increase the GP–GP referral confidence among GP colleagues, to develop community resource GPs who have the necessary skills to diagnose and treat VH. Such measures were identified to improve diagnosis and follow-up rates.

…without having trained in hepatitis treatment, it takes me a long time to deal with a single patient, I could have treated more patients with other illnesses (that) I am (more) familiar with the treatment…. [GP].

Training more GPs on the care of VH will offer the opportunity for follow up of patients in their nearby areas, where there is better access to free parking, better flexibility to get convenient clinic appointments and less waiting times, which could increase clinic attendance rates.

If Australia is to achieve its VH control goal (increase the proportion of people living with HCV who are initiated with direct-acting antiviral treatment to 65% and reduce HCV-related mortality overall by 65%, reduce the number of newly acquired HBV infections by 50% and increase the number of people living with HBV who are diagnosed to 80% by the end of 2022) as espoused in the National Strategy,^6,7 innovative changes in our models of care are needed. Understanding the perspective of GPs is critical to developing workable solutions. General practitioners felt their engagement in the care for VH was not satisfactory and described many challenges. One of the reasons for the lack of engagement is the high caseload of patients with other illnesses that GPs had to treat every day, and the complexity and time-consuming nature of treating a condition they were less familiar with. GPs were frustrated at the difficulties in handling such cases due to the lack of familiarity and need for training and skills required to treat VH patients translates into the need to take more time in a consult (eg to refer to the treatment guidelines), which reduces the cost-effectiveness per consult when compared with consulting for other diseases they are familiar with.²⁸

Elimination of VH would require focused hepatitis education to primary care providers through continuing medical education or periodic rounds of training,²⁹ and increasing the number of GPs able to treat VH.

The lack of bulk billing for VH diagnostic services such as transient elastography for liver fibrosis assessment and ultrasound, poses a significant barrier to GP involvement in the care for VH. These costs to patients make the overall care programme for the illness more expensive and prohibitive, compared to public hospitals where all investigations can be done at no expense to the patient. Providing a Medicare Benefits Scheme (MBS) rebate for elastography and increasing opportunities for bulk billing in community radiology for VH screening tests would ease patient accessibility and reduce costs.

Loss to follow up is a major barrier to care. Factors that improve retention of patients in care and follow up for monitoring are considered. If some part of VH care is to shift to the community and GPs, then ensuring focused training, information and support resources with secure linkage to existing services is necessary.³⁰ Language-specific reminders may help many individuals from non-English speaking CALD backgrounds. The existing referral process for VH between the primary care and hepatology specialist services is perceived by GPs as ineffective. Issues with referrals, combined with long patient waiting times at the specialty clinic, delays and difficulty in getting appointments in a reasonable time, and patient transport challenges indicate the existing system is broken. Better communication between primary care and specialty centres, feedback to referrers, and better information about what services are available to refer to, would improve tertiary level care independently of a more community-based model of care. Involvement of GPs in the referral of patients brings value and positive results to patient follow up.³¹

Community-based model of care for VH provides the opportunity for patients to have local consultation, reducing the challenges of attending a major hospital, which may be particularly difficult for patients with CALD backgrounds. Reducing transportation and parking costs and waiting times for patients may improve follow up and attendance.²² For GPs, potential benefits include the learning experience of sharing care and capacity building. Integrating specialty services into primary care provides better acceptability of tertiary-level services for patients.³² A more integrated type of care that delivers specialised services for people with VH from CALD backgrounds in the primary care setting benefits patients, promotes continuity of care and improves gains to the health system.^22,33

Although a community-based model of care offers exciting opportunities to improve access to VH care, several barriers may limit the effectiveness of its implementation.³⁴ Sustainable funding models for a community-based implementation of care is an essential element of success, and improving retention and follow up of both GPs and patients may need more time and funding than a narrow consultation-based remuneration model offered through MBS to GPs now.³²

Redistributing hepatology consultants into the community seems like a solution, but will be inevitably limited by available specialists, and it could pass on more costs to patients because of increased specialist travel time (while increasing convenience for the patient), and has logistical challenges when hospital health information systems are not always well-suited for sharing. The availability of consulting space in primary care and the need to recoup costs for consulting space in primary practice are also practical limitations to more a widespread adoption of community-based models.

In contrast to Australia where logistics for HCV care in the primary care still pose some challenges, New Zealand general practice has been successful in delivering community-based care for HCV through full funding of new HCV treatments.³⁵ Access to free fibroscans and the development of localised clinical and referral guidelines spanning primary and secondary care are integrated into GP patient management systems (HealthPathways).^36–38 New Zealand’s success offers support for the recommendations here to develop a shared care model with GP access to diagnostic services. The success stories from New Zealand could be applicable to the treatment of HCV in Australian primary care. The success of HCV treatment in New Zealand primary care could also be extrapolated to expand the treatment of HBV in Australia and New Zealand.

This study directly interviewed GPs with experience in VH care and referral to understand the limitations to specialty care than interviewing patients or specialists. However, the major limitation of the study was that it was conducted in an urban setting and may not be representative of the barriers and opportunities to VH care in non-metropolitan areas. There may be a potential for selection bias in this study as it is possible that GPs who agreed to participate might be more likely to be in favour of a community-based model of care than GPs who did not agree to participate.

Engagement of GPs to provide VH care is challenged by the time-consuming nature of VH management for many GPs. More training or provision of specialist–GP linkages may help to overcome this. Loss to follow up was a recurrent problem for GPs, who were not adequately resourced to follow up patients who did not attend their appointments. The lack of bulk billing of diagnostic services further limits the extension of care into the community. A broader and more profound capacity-building needs to occurs to allow community-based VH care to be sustainable.

A community-based model of care for VH was perceived as a practical approach by GPs, which brings in the opportunity of onsite consultation, experience-sharing with specialists and skill transfer. In Australia, ongoing focused VH training for GPs, bulk billing or cost reductions for diagnostics or consultations with hepatology specialists, could help to improve GPs' engagement with VH care.

All data collected and used to generate this manuscript were analysed and are reported in this manuscript.

The authors declare that they have no conflicts of interest.

There was no specific funding to the conduct of this study.

Supplementary material is available online.