Strong integration between primary and secondary healthcare is fundamental for effective delivery of quality healthcare.1,2 Poor communication between healthcare sectors can lead to adverse events within both general practice and hospitals.3–5 The growing burden of chronic disease and the ageing population increases the need for effective integration between health sectors.1,6–8
General Practice Liaison (GPL), defined as being a process for communication and integration between general practitioners (GPs) and hospitals, is one of a range of approaches that has evolved in Australia to help better integrate primary and secondary care.7,9 GPL units are generally situated within hospitals, although some have a broader focus on liaison between acute and primary care.10 GPL staff undertake a range of activities to ‘improve communication and transfer of information between GPs and hospitals, for the ultimate benefit of patient care’.11,12 These include improving information flow, care processes and capacity across both sectors; for example, easing the process of GP referrals to non-GP specialist clinics, and optimising communication during hospital discharge.7,11
The previous Commonwealth Divisions of General Practice program (1992–2011) was the impetus for improving integration between general practice and hospitals in Australia, which provided the basis for a range of GPL services.11,13,14 GPL services have shown promise in improving information flow, communication between health services and coordination of care between primary and secondary care.10,11
In Victoria, GPL services have been incorporated into many healthcare services since the 1990s and were centrally coordinated under the then Victorian Department of Human Services (DHS) between 2007 and 2012.11 During this time, more than 20 health services in Victoria had a GPL unit, with most staffed by GPs or by staff with backgrounds in nursing, administration, community development and health promotion.11
Despite this activity, little is known of GP experiences of, or aspirations for, GPL services. The aim of this study was to 1) explore GPs’ experiences of interaction with a hospital health service and its GPL services and 2) investigate GPs’ perspectives on the shape of a potential expansion of the GPL role within this health service.
Methods
Design
This qualitative study used semi-structured, face-to-face and telephone interviews with GPs and was oriented toward a descriptive approach and sought to provide guidance on service improvement.15,16 The work was a sub-study of the South East Melbourne arm of IMPACT, a large international program exploring access to primary care for vulnerable communities.17
Setting
This study was set in the catchment area of a large metropolitan health service in Melbourne, Victoria, in late 2018 that was looking to explore the expansion of its existing GPL services. This multicultural and diverse region includes stable established communities, areas of rapid population growth and population groups disadvantaged in their ability to access services.18 The health service comprises three large research and teaching hospitals that serve over one-quarter of the city’s population (approximately 1.9 million people), alongside several community health service locations. GPL staff in this health service were available for contact via telephone and email, listed on a dedicated page of the health service website. The contact number was also documented on discharge summaries.
Participants
GP participants were drawn from a purposive sample of GPs recruited as part of the IMPACT program’s intervention within the region and provided written consent prior to being interviewed.19 IMPACT selection criteria were: GPs willing to accept new patients who lacked access to ongoing primary care, accredited under the National General Practice Accreditation scheme and not intending to leave the practice for at least two years from the time of recruitment. Additional details on IMPACT recruitment have been published elsewhere.17 For the current sub-study, the IMPACT GPs were asked to consent to answer extra interview questions about the GPL service, which were added to the IMPACT follow-up interview guide (Box 1).
| Box 1. General Practice Liaison (GPL) service–related interview questions |
- Can you tell me what contact you have with [Health service name] services?
- Were there any examples of contact with [Health service name] that stand out to you?
- If there was one thing you could change about liaison between primary care and [Health service name], what would it be?
- What suggestions do you have for improving communication between primary care and [Health service name] services?
- There is a GPL service at [Health service name]. This service intends to help general practitioners (GPs) to navigate hospital services and works to improve communication between hospitals and GPs and enable GP engagement and representation. What do you know about the [Health service name] GPL service?
- How do you think such a service could help you?
- What help would be most valuable to you around clinical areas that are relevant to your practice?
- Would you like to add anything else?
|
Data collection
Data was collected by semi-structured interviews performed in August and September 2018, mostly by SC (a psychology graduate and Master of Public Health candidate), with two interviews conducted by CA (a research fellow with a background in education). Both interviewers have prior experience in primary care research. There was no dependent relationship between the participants and researchers. Questions on participants’ knowledge, utilisation and perceived opportunities for the local GPL service were added to the follow-up interview guide for the IMPACT intervention. Additional questions explored the participants’ experience with referrals and communication with the local health service. Contact summary sheets were completed after each interview to document interviewer reflections.20 Interviews were continued until sufficient data saturation was achieved, which was determined when no new themes appeared. Audio files and transcripts were stored securely in password-protected computers, only accessible to members of the research team.
Data analysis
Interviews were audio-recorded and transcribed by a professional transcribing service. Transcripts were checked for accuracy then anonymised. NVivo 11 was used to manage the data and assist with thematic coding.21,22
Using thematic analysis, themes were identified from the interview guide and developed from the data. The analytical process was iterative. SC, then SW (an occupational therapy undergraduate student), read the contact summaries, then categorised the data into themes from the interview guide. Memos were created to record interesting observations and relevant points as transcripts were coded.
A check of inter-coder reliability between SC and SW was performed, which showed agreement of identified nodes and themes.23 A list of data-driven codes was then generated, and the pre-existing categories modified to accommodate new insights from the data by SC and SW. SC discussed emerging themes with MK (an experienced qualitative and primary healthcare researcher) and GR (an experienced primary healthcare researcher and GP). Interpretation and disagreements were resolved by consensus. Preliminary findings were reviewed with health service staff responsible for the GPL service.
Ethics approval was obtained from Monash University Human Research Ethics Committee (7572) and Monash Health Human Research Ethics Committee (17-0000-231L).
Results
Ten of the 16 GPs who were approached consented to interview. Six were female, four were male, and they had 10–40 years’ experience as a GP. Six were Australian graduates; the remainder had graduated in South Asia.
Interaction with the region’s public health service
GPs spoke of difficulties in helping their patients access services, and the lengthy time needed to navigate the hospital service system (Table 1). Many felt this to be a particular problem for patients on low incomes or without private health insurance who lacked alternatives outside the hospital sector. GPs found it difficult to access clear and accurate referral information on the hospital website. Discharge summaries were delayed, ambiguous or had a default option of ‘GP to follow up’ without clear guidance. One GP spoke of wanting reasons why a patient may ‘drop off’ the waitlist, while another spoke of a 2–3-month delay for a patient procedure due to lack of clarity regarding referral processes.
| Table 1. Themes and representative quotes regarding how general practitioners interacted with the health service, and their knowledge of and contact with the General Practice Liaison (GPL) service |
| Theme |
Representative quotes |
| Difficulty referring patients to hospital services |
‘… it is very hard to refer patients to [hospital] … the waiting time … is very long’ [P20101]. |
| Challenges with service navigation and communication |
‘... I had a lady who had an abnormal PAP smear that requires colposcopy. So I faxed to the colposcopy clinic and … that was knocked back … “no it has to be addressed to somebody else” … a couple of months have passed and then I get this fax referral back to me to say that “no it’s the wrong one, it’s the other lady that I was supposed to put on”. So two, three months have passed and she hasn’t had her colposcopy. So with this back and forth business, it’s very inefficient … it’s just poor communication and it can certainly be improved’ [P21217].
‘… I was trying to refer someone [to a clinic] and they had a … process which I found awkward … I had to … send the referral in to someone who would do the assessment and then decide on whether an appointment was appropriate … And although I’d liaised with the people … and they thought that the referral was appropriate … I still had to go through that argy-bargy and I found that unnecessary’ [P21320]. |
| Lack of timely discharge summaries |
‘They assume that I have received discharge summaries … which I don’t often have access to … and that’s a huge problem … we don’t know … the new medications … or what were the findings’ [P21217]. |
| Minimal knowledge and little contact with GPL service |
‘Nothing. I’m not sure if they’ve actually tried to communicate with GPs by saying what they offer, but I’m not aware of it, no’ [P20609].
‘I don’t know how does it work … sometimes I see that [website] and just know that … But not more than that …’ [P21216].
‘I’ve looked it up before … I tried to contact that person … I think I left a message, but nothing happened. So I guess I sort of gave up’ [P21217]. |
GPs had little awareness of the hospital GPL service, with some stating no knowledge at all. Of those who were aware of the service, few understood the GPL role or were clear about who staffed the service. Those who were aware of the service found out via the health service’s GPL service webpage, newsletters and from calling the telephone number on discharge summaries.
Given these responses, it was unsurprising that few GPs had had any contact with the GPL service. One GP had tried to contact the service to no avail, and other GPs were reluctant to call a service they knew little about. Only one GP mentioned receiving a proactive call from the GPL service several years prior.
Gaps potentially addressed by an active General Practice Liaison function
GPs were asked to identify ways in which the GPL service could help them, what clinical help would be most valuable to them and the desired qualifications of GPL staff (Table 2). While few had contact with the GPL service, most felt it had promise in addressing many of the challenges experienced.
| Table 2. Themes and representative quotes regarding gaps potentially addressed by an active General Practice Liaison (GPL) function |
| Theme |
Representative quotes |
| Assistance with waitlists and patient care |
‘… they should let us know if there is someone we really want to be seen, they will get earlier appointment … I think we should be given some kind of preference to look after very difficult patient(s)’ [P20101].
‘If you’ve got an issue with the patient and care, where that patient is complaining … you could liaise with someone so that … could be nipped in the bud before it developed into something worse’ [P21013].
‘[The GPL service] might be able to tell us about the waiting list … Why did they [patient] drop off?’ [P21013]. |
| Assistance with service navigation and referral pathways |
‘… if there was a service that you could … say … “I’ve got this patient with this, this, and this problems. What do you recommend?” And then they could … say, “… we have this, this, and this…These are the services we can provide this patient. Please refer” … the time it takes to try and work out what’s around … we’re all time poor’ [P20609].
‘… we need to know … how to access the services, because when you go online, it’s very basic, the information, and not all the clinics are listed there. It’s just a referral pathway I think we need to understand’ [P21217]. |
| Timely discharge summaries with a clear follow-up plan |
‘Discharge summary, we get sometimes but it’s not always … the content or what they have done in the hospital is not 100 per cent clear or what we have to do is not clear … They said “GP to follow up” but we don’t know what to follow up …’ [P21216].
‘[The importance of] getting discharge summaries in a timely manner cannot be stressed enough. Generally … it happens well. But once in a while, there’s a slip up … invariably the slip up will occur just when you really desperately need that discharge summary’ [P21320]. |
| Practice visits from GPL staff |
‘… when you’re able to put … a face to a name even though people might change, that doesn’t matter. But, that kind of thing, I think registers with you and you remember it. Rather than the flyer might come out or an email ...’ [P21320].
‘[They could] pay a visit to us and see what are our requirements, what type of patient we service, and which way they can be of help to us’ [P20101]. |
| A person who knows the hospital system to staff the GPL service |
‘… it doesn’t really matter as long as the person knows the system and be able to direct us … being a GP him or herself is helpful because you understand the challenges that we face. But it’s not necessary’ [P21217].
‘I mean, if they know the system … that’s really what matters. It doesn’t matter what qualifications they have’ [P21320].
‘I think you probably need a health professional … it’s because of a clinical reason. And that clinical reason needs to be understood by the person you’re dealing with’ [P21013]. |
GPs wanted to be informed when services and referral processes changed and to contact the GPL service to find out what is available for a patient with a particular diagnosis. They also wanted GPL service staff to assist with getting earlier appointments for patients who were vulnerable or needed to be seen urgently. They would value assistance with patient care issues before they escalate.
GPs had a strong preference for face-to-face contact with GPL service staff, particularly through practice visits. They felt it was important to put a face to a name and have a point of contact. They also wanted GPL service staff to explain their role and how the service could help them.
While most GPs did not have a strong preference for GPL service staff to be GPs, they preferred the service be staffed by a health professional, such as a nurse or nurse coordinator, rather than an administrator, to ensure clinical understanding. Their main requirement was that GPL staff have knowledge of the hospital service system and how to navigate it. It was also important that GPL service staff understood the challenges and needs of GPs and the types of patients they care for in their practices.
Discussion
This qualitative study explored the attitudes and perspectives of a sample of GPs concerning GPL in a health service where minimal formal GPL had been undertaken in previous years. Participants spoke of problems navigating referral pathways within the service and were frustrated with delays in receiving timely and clear discharge summaries and information about waitlists. Most GPs could see the value of an active GPL role and had insight into how a model could work for them.
While not the explicit focus of this study, it was clear that participants found it difficult to negotiate what for many was a complex hospital health service. The findings were consistent with other studies of GP attitudes towards large hospital health services.10 GPs have long complained about access to investigations conducted during hospitalisation and of receiving timely, accurate and legible discharge summaries.11,24 This study confirms that there are improvements to be made in communication and integration between hospitals and GPs in the community.
GPs in this study had limited contact with or knowledge of the GPL service but could clearly see its potential, particularly regarding assistance with referrals into the hospital and improving the quality of discharge summaries. Others have identified the lack of visibility of GPL services and common under-resourcing and understaffing of these services.11 Barriers to the impact of GPL on integration include lack of funding, insufficient staffing (most roles are part time, including in the hospital in this study), slow culture change in health services and health service lack of interest in GPs.13
GPs in this study wanted a health professional who knew the hospital system to staff the GPL service, without a strong preference that this person be a GP. Their acceptance of a non-GP in the role has significant budgetary implications and allows for other health professionals potentially staffing GPL services.13 Face-to-face contact with GPL staff was important for GPs to ensure GPL staff understand general practice and the challenges that GPs face in the variety of settings in which they work.
Interdisciplinary, sectoral and institutional ‘silos’ represent fundamental barriers to healthcare improvement. GPL services, in essence, contribute by improving collaboration across silos.9 Adequate staffing, resourcing and promotion of GPL services will improve communication and integration between primary and secondary care and ensure general practice and hospitals are working together to maximise patient care and safety.
Limitations
These results reflect the experience of GPs within the catchment area of a single urban Australian hospital health service and may differ from those in other areas. The study was limited by being a sub-study of the IMPACT study, with its specific eligibility criteria, and by the fact that most GPs had not used the GPL service. Nevertheless, the GPs recruited for the IMPACT study were all active in delivering primary care in the region and were interested in expanding their reach within the community.
Conclusion
There have been few peer-reviewed evaluations of GPL services in Australia. This study has provided insight into how GPs envision a successful GPL service and can inform development of GPL roles and staffing in Australian health services.
Implications for general practice
- Integration between primary care and hospitals is not always easy, with this study finding issues with communication, referrals and discharge summaries.
- GPL services are valuable resources to assist GPs to strengthen communication and integration with hospital services.
- GPL services may be an underused resource that could assist GPs with their most common frustrations in liaising with their local health services.