Developing new models of shared primary and specialist HIV care in the UK: a survey of current practice

Abstract

HIV care in the UK is led by hospital-based specialists with limited general practitioner (GP) involvement. As GPs have expertise in some non-microbial HIV-associated co-morbidities (e.g. cardiovascular disease), and more people are disclosing their HIV status to their GPs, there could be benefits in sharing HIV care. We describe contemporary models of shared HIV care in relevant developed world settings to inform future shared HIV care in the UK. An interview survey of key informants was used to explore experiences and models of shared care, and identify promoting and inhibiting factors. We interviewed ten key informants from six shared care models. There were three broad categories of shared care, with varying degrees of GP involvement. Strong clinical leadership in primary care, good professional relationships and communication, and tailored GP training were facilitators. Barriers included stigma, confidentiality concerns, and low prevalence of HIV outside major conurbations. Contemporary shared HIV care models have emerged organically and seem to work when grounded in good collaboration between a small number of dedicated GPs and specialist units. We propose two models for further study which may only be feasible in high HIV caseload practices. User acceptability, clinical and cost effectiveness must be considered.

Keywords

HIV AIDS primary care shared care models

Introduction

The use of highly active antiretroviral therapy has transformed HIV infection into a chronic condition, in which non-microbial co-morbidities such as cardiovascular disease and mental health issues create an increasingly important health burden.¹

In the UK, HIV in- and out-patient medical care is usually provided by hospital-based specialist physicians. Clinical outcomes are among the best world-wide.² However, stigma has meant that over a quarter of HIV-positive patients in England have sought care away from their place of residence³ and many have been reluctant to disclose their diagnosis to other healthcare professionals, including their general practitioners (GPs). As a result, many patients have been cared for in relative isolation with specialist physicians providing primary as well as specialist care, and this has been popular with many patients.⁴ More recently, in line with national recommendations⁵ this is beginning to change.⁶ It is increasingly difficult for specialists to prescribe for conditions that should be managed in primary care (e.g. hypertension). In addition, many patients now disclose their HIV status to their GPs, enabling HIV specialists to update GPs on their patients’ clinical status and medication.

In England, there is considerable variation in the prevalence of HIV and most HIV-positive people live in major cities, often in concentrated areas of these cities.⁷ Elsewhere, the absolute number of HIV-positive patients is low, meaning that many GPs will have few HIV-positive patients registered in their practices, little experience in HIV medicine and may lack the knowledge, skills and confidence for high-quality HIV care. However, GPs have considerable expertise in managing some of the increasingly important non-microbial HIV co-morbidities. Ongoing reforms in the National Health Service (NHS) provide opportunities for optimising HIV care, which could include increased shared care with GPs. However, change could threaten clinical outcomes if service developments are poorly considered and implemented. This study aims to describe contemporary models of shared primary and specialist care for people with HIV, including barriers to implementation and factors influencing their success, to inform future HIV care provision in the UK setting.

Methods

Definitions

In the absence of accepted definitions, we created the following terms:

Shared HIV care: provision of care in which HIV-positive patients receive their health care in both primary care settings (for the most part, general practice) and from HIV specialists, and where the GP is aware of the patient’s HIV status. The care provided in primary care might include primary care appropriately tailored to the patients’ HIV status (appropriately tailored HIV primary care, see below); and/or some elements of specialized care normally only provided by specialists in the UK, such as monitoring of CD4 count; and include formal and informal shared care arrangements which are not necessarily overtly coordinated and may not have clear specifications of which elements of care are provided in which setting.

Appropriately tailored HIV primary care: health care provided in the primary care setting, which includes interventions such as cervical screening and vaccination for which the recommendations for HIV-positive people^8,9 differ from those of the general population.

Phase 1 – Scoping exercise to derive our sample of key informants

We used our systematic literature review¹⁰ of current practice in shared HIV care together with informal discussions with key stakeholders (HIV specialists, GPs involved in shared care, key figures in the British HIV Association and HIV commissioners) to provide an overview of the status quo, identify further shared care exemplars and identify our informants.

Phase 2 – Development and administration of survey

Informed by findings from Phase 1, we developed a survey questionnaire (available from authors on request), with the aim of exploring in detail the models of shared care we had identified with the GPs and specialists (informants) involved in each model.

We approached 12 potential informants by email using a standard letter informing them of the purpose of the study and inviting them to participate in either a face-to-face or telephone interview using the survey questionnaire, which consisted of a brief survey of practice and open-ended questions. We arranged a suitable date and time with those willing to participate and gained consent for audiotaping the interview.

We explored the following areas: (1) context in which the key informant works; (2) commissioning and funding; (3) care pathways including the division of care between primary care and specialists and the movement of patients between primary and specialist care; (4) eligibility criteria for GPs and their staff to be involved in shared HIV care including training and supervision issues; (5) patient involvement in the setting up and/or evaluation of the care model; (6) feasibility, to identify reasons for success, perceived barriers, concerns about the practicality of the arrangements and suggested improvements.

Phase 3 – Data analysis

The interviews were transcribed verbatim. We categorised the interview data according to the areas explored. The interviewer discussed any points needing clarification with the informant by email. We then synthesised published elements of the care pathways with interview data to summarise the models of shared HIV care (Table 1; see online supplemental tables 1 to 3 for additional details) and identify factors key to their success as well as potential barriers to their wider use.

Table 1.

Overview of six models of shared HIV care.

Geographical location	Setting of model	Key informant(s)	Approximate number of HIV-positive patients; main risk group	Brief description of model and elements of care delivered in primary care
Category 1: Primary care clinicians provide elements of specialised HIV care alongside appropriate HIV primary care
Model 1: Manchester, UK	One city-centre general practice; local HIV specialist clinics	One GP and one HIV specialist	>220; MSM	Regular informal contact between primary care and HIV specialist clinicians. Practice-based, externally commissioned HIV specialist nurse co-ordinates care pathways. Practice provides some specialist care for newly diagnosed patients and monitoring of asymptomatic patients (e.g. CD4 counts, viral loads, cryptococcal antigen), as well as appropriate primary care for HIV-positive patients
Model 2: Zurich, Switzerland	Inner-city GP who is also an HIV specialist; specialists at a University-based HIV outpatient clinic	One HIV specialist GP	>130; MSM	No formal co-ordination between GP and hospital-based specialists. Patients can choose where they receive different elements of care. GP provides many elements of ‘specialist’ care (e.g. ARV prescribing, TDM, viral resistance testing) but refers complex cases to hospital-based specialists (e.g. primary HIV infection, treatment failure with emerging ARV drug resistance)
Model 3: Sydney, Australia	Urban private general practice staffed by ‘HIV specialist’ GPs; university hospital-based HIV specialists	One HIV specialist GP	>200; MSM	No formal co-ordination between primary and secondary care. Patients can choose where they receive different elements of care. Some HIV-positive patients managed holistically by GPs (includes both specialist HIV and sexual health care and primary care) and some patients receive only primary care in the practice
Category 2: Primary care clinicians address only the primary care needs of HIV-positive patients
Model 4: Brighton and Hove, UK	28 inner-city general practices; one HIV specialist clinic	Two GPs (one with special interest in HIV) from two different practices	The practices of the two GPs had <50 and 140 patients, respectively; MSM	Elements of care defined in and funded through HIV LES commissioned by PCT¹³ and include: producing care plan and annual review of each HIV-positive patient; recording ARVs and monitoring adherence; recommended vaccinations; annual cervical cytology; cardiovascular risk assessment; management of abnormal lipids; mental health management
Model 5: South East London, UK	Inner-city general practice; multiple HIV specialist clinics across London and beyond	One GP	51–100; Black African heterosexuals & MSM	No formal co-ordination between GP and hospital-based specialists. Interested GP ensures provision of appropriate primary care for known HIV-positive patients in the practice. This includes: recommended vaccinations; annual cervical cytology; monitoring cardiovascular risk and abnormal lipids; mental health management
Category 3: Primary care clinicians provide specialised care opportunistically to socially vulnerable HIV-positive patients alongside appropriate HIV primary care
Model 6: North East London, UK	Several inner-city general practices; one HIV specialist clinic	Two GPs from one general practice and one HIV specialist	6–50; injecting drug users (vulnerable group); MSM & Black African heterosexuals	Informal arrangement between HIV physician and specific GPs largely in relation to vulnerable group of patients who fail to attend specialist service and HIV-positive women living in local homeless hostel. HIV specialist provides individualised care plans for GPs to use opportunistically for specific patients (includes specialised blood tests e.g. CD4 count as required)

ARV: antiretroviral; LES: locally enhanced service; MSM: men who have sex with men; PCT: Primary Care Trust; TDM: therapeutic drug monitoring.

Ethical approval was obtained from the Queen Mary Research Ethics Committee (QMREC2011/65).

Results

The literature review¹⁰ and scoping exercise identified six relevant models of shared HIV care from Australia, Switzerland and UK.

Ten key informants (two HIV specialists, two HIV specialist GPs, one GP with a special interest in HIV, five GPs – see Table 1) agreed to participate and underwent a telephone or face-to-face questionnaire-based interview with JH between January and May 2012. Interviews lasted between 25 and 70 min.

The six models could be grouped into three broad categories based on the elements of HIV care provided in primary care (see Table 1): (1) GPs provided some elements of specialised HIV care alongside appropriately tailored primary care (see Table 1 and online supplementary Table 1); (2) GPs provided appropriately tailored primary care only (see Table 1 and online supplementary Table 2); (3) GPs provided specialised HIV care opportunistically to some socially vulnerable HIV-positive patients who do not attend specialist care, as well as providing appropriately tailored primary care to other known HIV-positive patients (see Table 1 and online supplementary Table 3).

With the exception of the Swiss model,¹¹ there were no surveys of patient satisfaction or clinical outcomes reported. In this study, clinicians commented on what they believed their patients felt about the care received.

Factors important for success of shared care models formed five main themes (within which participants also suggested some areas for improvement): good clinical leadership in the practices involved; good professional relationships and two-way communication between specialists and primary care; development of ‘HIV-friendly’ practices involving both clinical and non-clinical staff; treating HIV infection like other chronic diseases and availability of up-to-date training in HIV medicine (Table 2).

Table 2.

Facilitators and barriers to successful HIV shared care models.


Theme	Respondents’ views
Facilitators
Good clinical leadership	Motivated, up-to-date primary care clinicians championing the process and providing clinical leadership in the practices involved.
Good professional relationships and communication	Good relationships between specialists and primary care facilitated shared care. Important for GPs to have known specialists who were accessible for advice and supportive of primary care involvement. This can be difficult to achieve^a. Two-way communication important – primary care clinicians need access to clinical information about patients, especially in relation to antiretroviral prescriptions, via regular clinical updates from specialists. Improvements could be made by primary care clinicians in informing specialists about new diagnoses made and medications started in primary care.
Development of ‘HIV-friendly’ practices	The involvement of the whole general practice team, including non-clinical staff (e.g. receptionists) important particularly in relation to confidentiality. Key issues included: appropriate disclosure within a practice (e.g. reception staff having access to records for appropriate administration only); avoiding inadvertent disclosures due to audible conversations at reception or visible diagnoses on computer screens; and reassuring patients by, for example, displaying notices about confidentiality in general practices.
Treating HIV infection like any other chronic disease	HIV is a chronic condition and GPs should provide good quality, appropriately tailored primary care for patients with HIV infection as they would for patients with other chronic conditions.
Training in HIV medicine	Availability of up-to-date education in HIV medicine tailored to the primary care context, especially in relation to key antiretroviral drug interactions. Other areas for training included: caring for sero-discordant couples; confidentiality and disclosure of HIV status to third parties; managing patients who have been tested for HIV but decline to know their result; managing patients who do not access specialist care.
Barriers
Lack of financial incentive for provision of HIV care in primary care	Some informants noted that the informal arrangements described in London are very dependent on the motivation of individual GPs who have recognised the needs of HIV-positive patients and developed appropriate care pathways without any specific financial reward attached to this work.
Inaccurate perceptions about HIV infection	HIV is still considered to be a specialist condition by some patients and non-specialist doctors. GPs felt that many long term patients have been used to a gold standard service in specialist units, all aspects of which may not be replicable in primary care. Limited experience of looking after HIV-positive patients means that some GPs lack confidence to manage even the primary care aspects of HIV infection.
Barriers (continued)
The impact of stigma	HIV infection remains a stigmatised condition and, consequently, GPs can remain unaware that patients registered with them are HIV-positive. One GP gave an example of how stigma can also unnecessarily protract normally straightforward tasks such as writing a housing letter, where consent for HIV disclosure may need to be sought before writing the letter, which would not normally be the case for other chronic conditions.
Need for minimum caseload	GP informants felt that in practices with low numbers of HIV-positive patients, the time investment needed by primary care clinicians for training and the difficulty of developing and maintaining adequate skills without a minimum caseload call into question the viability of providing anything more than straightforward but appropriate primary care.
Lack of formal framework supporting the provision of HIV care in primary care	Some informants suggested that a more formalised framework might facilitate wider use of shared HIV care models. Agreeing detailed plans of which elements of care are to be provided in the different settings is also key to avoiding omission or duplication. Some GPs reported that filling out a mandatory care plan can be seen as intrusive or unnecessary by some patients and also as duplication of what happens in the specialist centre with patients wondering why they need to come for a health check with their GP if they are well.
Infrastructure issues	None of the models had a shared information technology system across specialist and primary care. Such a system could facilitate communication between primary and specialist care including viewing of test results done in both settings. This would reduce unnecessary duplication of investigations and aid diagnostic processes

One GP informant reported that between them, the known HIV-positive patients in his practice were attending every HIV unit in London and some outside London.

Six main factors emerged as barriers to wider use of shared HIV-care models: lack of financial incentive for provision of HIV care in primary care; inaccurate perceptions about HIV infection by some patients and clinicians; impact of stigma; need for minimum caseload to justify time investment in training and maintaining skills; lack of a formal framework supporting the provision of HIV care in primary care and lack of shared information technology systems (Table 2).

Discussion

Several shared care models exist in areas of high HIV prevalence, which have tended to develop organically within a specific locality in response to a particular need. We used a deliberately broad definition of shared HIV care to try to capture what is happening in each of the differing models and to learn from the experiences of all of them, including the very informal arrangements in London. On the whole, they were not centrally planned or commissioned and so do not fit Hickman et al.’s¹² definition of shared care for patients with a chronic condition as including the ‘planned delivery of care’.

Mostly they entail either: specialists providing HIV-related monitoring and prescribing with general practice providing appropriately tailored HIV primary care; or general practice undertaking some elements of specialist HIV care such as CD4 and viral load monitoring for stable HIV-positive patients in addition to appropriately tailored HIV primary care, whilst antiretroviral prescribing and other HIV-related care remains in the specialist unit. In most of the models, patients belonged to a single HIV risk group, namely men who have sex with men, and attended only one or two local specialist centres.

The perceived success of the models seems highly linked to good clinical leadership and effective collaboration between interested GPs interacting with only a small number of specialists or specialist units. This facilitates the development of good inter-professional relationships, communication and support. A clear understanding of confidentiality; refinement of the specifications to define exactly which elements of care are provided in each setting and appropriate incentivisation for general practices also appears important. The involvement of non-clinical, particularly reception staff, as well as clinical general practice staff was important in building patient trust in the service. However, it should be noted that only one of the published models¹¹ included clinical and patient acceptability outcomes and none have done any health economics evaluation.

Many people with HIV in the UK are highly satisfied with their current model of care, which for the majority is within specialist services with limited GP input,⁴ although some report feelings of stigma.⁴ This specialist-based model has facilitated exemplary clinical outcomes for people with HIV in the UK.² Our findings suggest that subjectively at least, shared care could provide an alternative option for some patients, especially as more people appear willing to disclose their status to their GP, in line with national recommendations,⁸ and could offer health benefits as GPs have considerable expertise in managing the non-microbial HIV-associated co-morbidities such as cardiovascular disease and mental health issues. In the context of stretched healthcare resources and rising numbers of people living with HIV, it is important to consider new models of care which could confer health and cost benefit.

To our knowledge, this is the first study of models of shared HIV care in developed countries, which attempts to gain deeper understanding of factors affecting feasibility and perceived success of the models of care from the clinicians involved. The design enabled us to capture operational details of the models in a readily comparable form, whilst leaving room for informants to elaborate on promoting and inhibiting factors. We aimed to interview one GP and one specialist from each model to better understand the issues from both settings. However, we did not succeed in interviewing HIV specialists involved in all the models because of lack of response and time constraints.

Our findings are unlikely to be generalisable to all GP and specialist settings: the models described in this study operate within general practices with a high level of interest in HIV medicine, which cannot be assumed to be the case in all practices, especially those in areas of low HIV prevalence.

Whilst there is little published evidence for clinical and cost effectiveness for shared HIV care, these findings broadly support the consideration of two types of shared HIV care model. A chronic disease model could entail specialist HIV care provision from the specialist unit and appropriately tailored HIV primary care provided in general practice. A stable patient model for people with well-controlled HIV disease could involve the specialist unit overseeing antiretroviral therapy with general practice taking on some elements of care, traditionally provided by specialists, such as CD4 and viral load monitoring, in addition to appropriately tailored primary HIV care. Central to both models is effective communication, clarity of provision and a ‘whole practice’ approach that may need financial incentivisation to ensure widespread adoption in the UK. The geographical setting could be a major factor in determining success: shared care may not be appropriate for GPs with very few HIV-positive patients as the investment in training may be disproportionate to the opportunities to use it. London, where most people with HIV reside, could provide challenges to the development of key working relationships between primary and secondary care as patients in any one general practice may be attending several different specialist units.

Shared HIV care warrants further study but changes in service provision for people with HIV must be carefully developed with appropriate user involvement and robustly evaluated for evidence of clinical and cost effectiveness, and user acceptability if we are to maintain the UK’s long history of excellence in HIV-related clinical outcomes and user satisfaction.

Footnotes

Authors’ contributions

All authors contributed to preparation of the manuscript. CSE had the original idea for the study and assisted with study design and interpretation of findings. JH assisted with study design, conducted all interviews, analysed and interpreted findings. LJS assisted with study design and analysis and interpretation of findings. AW assisted with literature searching, and identification of key informants.

Declaration of conflicting interest

The authors declared the following potential conflicts of interest with respect to the research, authorship, and/or publication of this article: CSE and AJW are practising HIV specialist physicians; JH has previously received funding from Gilead and was funded for this study through the unrestricted education grant from Gilead Sciences Ltd, UK.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This study was funded by an unrestricted education grant from Gilead Sciences Ltd., UK.

References

Kearney

Moore

. The ageing of HIV: implications for geriatric medicine. Age Ageing 2010; 35: 536–541.

British HIV Association. Standards of care for people living with HIV 2013, London: Mediscript Ltd, 2012. www.bhiva.org (accessed 5 June 2015).

Huntington

Chadborn

Rice

. Travel for HIV care in England: a choice or a necessity. HIV Med 2011; 12: 361–366.

Weatherburn

Keogh

Reid

. My care, I care: a study of what people with HIV value about NHS HIV services across London, London: Sigma Research@London School of Hygiene and Tropical Medicine, 2013.

BHIVA position statement. The future role of primary and community care in HIV, www.bhiva.org (2011, accessed 5 June 2015).

Desai

Field

Crompton

. Information for action: a method to inform HIV shared care planning in primary care at the PCT level. Sex Transm Infect 2011; 87: 295–295.

Field N, Desai M, Crompton J, et al. Mapping HIV at a primary care level: a method to inform shared-care planning. In: HPA conference oral abstract presentation: Warwick, UK, 13–14 September 2011.

British HIV Association Guidelines for the routine investigation and monitoring of adult HIV-1-infected individuals, 2011, www.bhiva.org (2011, accessed 5 June 2015).

British HIV Association Guidelines for immunisation of HIV-infected individuals 2008, www.bhiva.org (2008, accessed 5 June 2015).

10.

Mapp F, Hutchinson J and Estcourt CS. A systematic review of contemporary models of shared HIV care and HIV in primary care in high-income settings. Int J STD AIDS. 2015; 26: 991–997.

11.

Page

Weber

Somaini

. Quality of generalist vs. specialty care for people with HIV on antiretroviral treatment: a prospective cohort study. HIV Med 2003; 4: 276–286.

12.

Hickman

Drummond

Grimshaw

. A taxonomy of shared care for chronic disease. J Pub Health 1994; 16: 447–454.

13.

NHS Sussex. New general medical services contract – specification for the provision of an enhanced service for HIV in primary care: 2010 NHS Sussex, www.sussex.nhs.uk (2010, accessed 5 June 2015).

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.29 MB