Factors influencing uptake and sustained utility of HealthPathways in Australian general practice: a qualitative study

Theoretical framework

Bridging translational gaps between positive research outcomes for complex interventions, such as HealthPathways, and their practical implementation necessitates evidence rooted in robust theoretical frameworks from implementation science research (Grol and Grimshaw 2003; Woolf 2008; Eccles et al. 2009). Normalisation process theory (NPT) is a mid-range theory that enables reflection on how individuals and groups ‘normalise’ new ways of working in everyday practice (Murray et al. 2010). NPT offers an effective analytical framework for primary care research, allowing for the identification of contextual factors and work practices of individuals and groups that enable the normalisation of an intervention (May and Finch 2009).

NPT comprises four dynamic components: coherence (sense-making), cognitive participation (engagement), collective action (work efforts enabling the intervention) and reflexive monitoring (appraisal of intervention benefits and costs). In this study, NPT served as an analytical lens, allowing the research team to scrutinise HealthPathways implementation issues pertaining to its sustained value and utility in primary healthcare settings (Sturgiss and Barton 2023).

Study design and setting

Our qualitative study design used narrative inquiry set within a social constructivist paradigm (Creswell 2013). We used semi-structured interviews to gather subjective narrative data from a diverse group of participants engaged in the development, implementation and utilisation of HealthPathways. This approach allowed for a comprehensive understanding of individual and collective perspectives regarding the value and utility of HealthPathways across regional catchments.

This study was conducted in metropolitan Melbourne, Victoria, with participant recruitment occurring across three PHNs. For the purposes of this paper, the identity of the PHNs have not been disclosed to provide anonymity. Two of the PHNs actively promoted the use of HealthPathways, whereas the other provided practitioners in the region with access to PHN-designed care pathways. The study was overseen by management staff of one PHN and a reference group comprising GPs, PNs, PMs and PHN personnel.

Recruitment

Participants were recruited from two participant groups: (1) general practice clinicians (GPs and PNs) along with PMs, and (2) staff of PHNs, GP liaison officers within teaching hospitals and other key regional health service-related informants. Potential participants were identified in collaboration with PHN management and the reference group. Stratified purposive sampling, a qualitative non-probability sampling technique, was used to identify and select participants from each group to understand their differing views on the topic (Campbell et al. 2020). Additional participants were identified using a snowball sampling technique (Noy 2008).

Potential participants were initially contacted by email with an invitation to participate and a link to the study’s explanatory statement and consent form. The study was advertised through PHN communication channels and Monash University’s Practice Based Research and Education Network. Only the first participant group were offered a AUD50 e-gift voucher in recognition of their professional time, as many clinicians, particularly GPs, were forgoing paid working hours to participate in the study. Those interested, but unable to participate, were offered the opportunity to make a written submission.

Data collection

Semi-structured interviews were undertaken by members of the research team experienced in qualitative interviewing of primary care professionals and health service providers (SS, RL, SC). All interviews were conducted via Zoom videoconferencing between September and November 2022. The focus of the interviews varied according to the participant groups’ characteristics, and was informed by regular consultation with the reference group and by key themes that emerged from a preliminary evidence review and document analysis. Separate semi-structured interview guides were developed to reflect the differing roles and experiences of participants (see Supplementary material). GPs, PNs and PMs were asked about their direct use of HealthPathways in clinical/professional practice, whereas PHN staff and other regional informants were asked broader questions about HealthPathways implementation, and their perspectives on general practice engagement with clinical and referral pathways. Consent to participate was obtained using an online consent form and confirmed verbally at the start of each interview. Participants had the opportunity to ask questions about the study before, during and after the interview.

Data analysis

All interviews were audio-recorded, and particular interviews identified by the interviewers as crucial to analysis were transcribed verbatim. After each interview, a contact summary sheet was completed by the interviewers to summarise salient points, order unstructured participant accounts into a narrative and include questions that could be asked in subsequent interviews (Miles and Huberman 1994). These contact summary sheets aided the beginning of the data analysis process. All contact summary sheets and transcripts were uploaded to the NVivo software ver. R1(20; QSR International Pty Ltd 2023) to facilitate data organisation and management.

Data analysis occurred in two phases. Initially, an inductive content analysis approach was employed, following the guidance of Hsieh and Shannon (2005). Theme and sub-theme development was largely deductive, using a priori codes based on the interview questions, while remaining open to emergent themes. The second phase involved mapping these themes to the four mechanisms of NPT, using a coding framework developed by the researchers.

Rigour

To ensure the quality and trustworthiness of this study, we applied multiple strategies aligned with Lincoln and Guba’s (1985) evaluative criteria for rigour. Credibility was enhanced through investigator triangulation, with two of the three interviewers independently coding the data before meeting to discuss and resolve discrepancies, ensuring consistency and reliability in the coding process. Weekly peer debriefing sessions were conducted with the broader research team to discuss emerging codes and themes, interpret key findings and explore new avenues of enquiry. The research team brought together diverse expertise, with most researchers having a background in primary care research, and we sought to maintain a reflexive stance by actively considering how our shared experiences could influence data interpretation. The senior author’s role as a GP provided a critical clinical perspective that was central to shaping the study’s design and analysis. The diligent use of contact summary sheets (Miles and Huberman 1994) offered a transparent ‘audit trail’ that documented the progression from raw data to the generation of findings, promoting dependability through systematic data analysis. Findings were further contextualised within the evolving landscape of Australian primary care through close collaboration with the reference group and PHN management, enhancing dependability. Finally, we aimed to improve transferability by providing thick descriptions of the data and offering rich contextual detail, allowing readers to assess the relevance of our findings to other settings and contexts.

Ethics approval

Ethical approval for this study was granted by the Monash University Human Research Ethics Committee (ID: 23950).

Participant demographics

A total of 35 interviews were conducted with 43 key informants across three PHN regions in Melbourne, with some interviews conducted with two participants. Some interviewees held multiple roles. We interviewed 18 GPs, 14 PHN staff members, six PMs, four GP liaison/hospital employees, one PN, one PHN board member and one health service partnership employee. Of the 18 GPs, 11 were female, three had worked in general practice for <5 years and 11 for ≥20 years. No written submissions were received.

Four themes emerged from the process of analysis: understanding HealthPathways’ purpose and relevance, receptivity to adopting HealthPathways, current HealthPathways utilisation and impact, and reflections on HealthPathways. Themes were interrelated and not always linear, and each theme mapped predominantly to one construct of the NPT.

Understanding HealthPathways’ purpose and relevance (coherence)

This theme explores participants’ understanding of the purpose of HealthPathways, their perceptions of its individual and shared value in clinical practice, and how they distinguish it from other clinical decision-making tools and referral information sources.

More than half of the GP participants had previously or were currently using HealthPathways. This included GPs within the PHN catchment that did not use HealthPathways, but still had access due to working across multiple PHN regions or through shared login details from peers.

When GP participants were asked about their decision-making processes or tools for clinical management or referrals, they reported actively seeking clinical advice from various sources, both Australian and international. The Electronic Therapeutic Guidelines (ETG) and the guidelines supported by the Royal Australian College of General Practitioners (RACGP) were the most frequently cited resources for assisting in prescribing and clinical diagnosis. For referrals, GPs preferred to rely on multiple resources, including HealthPathways, to make informed clinical decisions.

I have three main sources of information: ETG, electronic therapeutic guidelines, Australian Medicines Handbook and HealthPathways, which is quite common. Oh, and otherwise, there’s always the RACGP research articles and guidelines. And if all those don’t work, I ask for help from another doctor. (GP Registrar, ID 24)

Understanding of HealthPathways’ purpose varied among participants. Those who had used it regularly understood its distinction from clinical guidelines, such as ETG. In contrast, those unfamiliar with it found it difficult to differentiate it from other clinical guidelines or referral networks.

So, treatment guidelines I just look on the web, and that could be anywhere. If it’s referral guideline, we’re often stuck with the hospital network that we’re referring to. I don’t know what that pathway thing (sic) has to offer and what would make me engage with it? (GP, ≥20 years of experience, ID 29)

The perceived value of HealthPathways was closely related to the GP’s degree of clinical general practice experience. Doctors who were new to the Australian healthcare system or those who had recently joined practices used it to familiarise themselves with the local healthcare landscape, aiding in appropriate referrals and accessing relevant resources. GP registrars and trainees used the clinical management and assessment components of HealthPathways for decision-making and as a learning aid.

I use HealthPathways religiously. When you’re a new GP Reg, you don’t know anything, right? You start off being completely new, completely green, completely impressionable, and you need a lot of support because you don’t know what to do, but also to kind of back yourself legally as well, medico legally. So, I will write in my notes, did this as per Melbourne Health Pathways, or as per ETG, I will write in my notes where I got the reference for my decision. (GP Registrar, ID 19)

Conversely, many experienced doctors found HealthPathways to be less relevant in their daily practice.

So, with experience, you get to know your local hospital networks, your private hospitals, your public hospitals. These are my clinical pathways, going on to websites of different hospitals, and finding specialists for whatever problem the person has. Yeah, so that’s how I’ve done it, without having access to these clinical pathways that you mentioned. (GP and Practice Owner, ≥20 years of experience, ID 5)

However, some experienced GPs did use HealthPathways, and viewed it as a guiding tool that makes clinical practice easier without taking over clinical judgment. They highlighted the value of HealthPathways for managing atypical and complex presentations, and for staying current with the expanding breadth of medical knowledge through its provision of synthesised, evidence-based information at the point of care.

As general practitioners, we see everyone from day zero to 100 years old, and its head to toe. There is nothing that is not seen on a daily basis, and you don’t know what walks through your doors on a particular day, you can’t remember everything as a human being. So, these pathways do make life easy for everybody. (GP and Practice Owner, ≥20 years of experience, ID 6)

Receptivity to adopting HealthPathways (cognitive participation)

This theme delves into the motivations and inhibitions behind initiating and continuing to use HealthPathways, and how key stakeholders interact with each other to influence use.

Participants recognised GPs as the primary users of HealthPathways. Peer interaction within informal general practice networks emerged as the predominant method for encouraging HealthPathways adoption. GPs who used HealthPathways were often introduced to it by colleagues who found it valuable. PMs, who engage more with PHN provider support staff than GPs, seemed to have a role in promoting resources, such as HealthPathways, to GPs within clinics.

It [HealthPathways] was recommended to me by another peer who used it. She was using it at the time, and I’d walked into the room, and I was like, Oh, what’s that? So, she showed me. And then she said it was another doctor in the practice that had told her. (GP, ≥5 years of experience, ID 32)

Most of the GPs in my clinic don’t know about those pathways, and I try to bring it up in our clinical meetings. (Practice Manager, ID 11)

The preference for referring patients to professionals known and trusted by GPs was highlighted, as it provides a sense of reassurance for both them and their patients. Along with this preference, experienced doctors familiar with the healthcare system exhibited resistance to using HealthPathways due to a perceived lack of necessity. They believed their established relationships and knowledge were sufficient, and they valued the efficiency of their current practices.

I suppose you got to have a need for something, and then you got to go and search it out. Now, because I don’t really know what’s there, I haven’t looked at it, and because I seem to muddle through what I need to do, just through experience, I seem to get there. Okay, that may not necessarily be the right way to go, but that’s what’s happened. (GP and Practice Owner, ≥20 years of experience, ID 5)

I thought it [HealthPathways] looked interesting. But I haven’t spent much time on it – because time efficient wise, it’s not more efficient or quicker than what I do currently. I’ve been doing it for a long time. And I know who’s who and what’s what for me to do this. And it’s probably still quicker for me to do a referral letter, print off the pages and do it that way than it is to try and do an electronic referral, at that point. So, it’s probably just partly a me problem. (GP, ≥20 years of experience, ID 33)

If the information sought was not readily and quickly accessible, participants perceived that GPs might abandon HealthPathways after a few unsuccessful attempts. Additionally, if the platform was perceived as difficult to learn, practitioners were likely to revert to their established practices that have proven effective for them.

If they have tried once or twice, and not found what they are looking for, they will not go back to it, or it takes too long to access the appropriate information. (PHN staff, ID 7)

GPs’ engagement with HealthPathways was further influenced by time constraints and the overwhelming amount of available information, which made regular use of the platform challenging. The availability of other resources and time needed to address the day-to-day complexities of clinical care were added considerations.

Furthermore, the fragmentation of information sources and the challenge of navigating different resources posed difficulties for GPs. They expressed concerns about how to choose the appropriate clinical decision support tool.

When we start trying to direct GPs, ‘There’s some information over here, and there’s some over there. And also, over here if you open this database. And oh, don’t forget that one,’ and they’ve got to piece it all together. There’s the potential that the messaging across all of those is slightly different, all we’re going to do is confuse GPs. (PHN staff, ID 8)

Current HealthPathways utilisation and impact (collective action)

This theme explores the integration of HealthPathways into daily practice, examining user interactions, its fit within organisational practices and the broader health system, and the factors that make it easy or challenging to use.

Most GPs and PMs who did not use HealthPathways in their clinics were either unaware of the platform or lacked knowledge on how to use it. However, when informed about its potential benefits, both GPs and PMs expressed a positive outlook on its utility, provided that HealthPathways was effectively promoted and advertised. Adequate training and education for GPs by PHN provider support staff were identified as crucial for effective use, with in-person demonstrations highlighting how HealthPathways can enhance GP workflow.

I would need someone to sell it to me, you know, tell me, ‘Gee if you log into this, you’re going to get a whole lot of support and ideas and advice’. And there may be value in it. I just don’t really know. (GP and Practice Owner, ≥20 years of experience, ID 5)

People need to know about it. Even I didn’t know the PHN had all these pathways, so I think promoting it at the general practice level, to say there are things available will be important. (GP and Practice Owner, ≥20 years of experience, ID 6)

HealthPathways was recognised for its potential to improve referral efficiency by preventing rejections, ensuring necessary tests are conducted and avoiding unnecessary delays in patient care. Engaging GPs in the development and implementation of pathways was emphasised as crucial for their success.

If they [GPs] work well in helping us develop that pathway, the more engaged they are, the likelihood of increased accuracy or appropriateness of referrals is increased. (PHN staff, ID 8)

A significant challenge identified was the limited interoperability of HealthPathways with other medical software or IT systems. Participants expressed the need for systems to communicate with each other, be linked, accessible and integrated to realise the full value of HealthPathways.

You need to combine it with other pieces of work to make it realise its value. (Health Service Partnership staff, ID 21)

Some participants stressed the importance of ensuring consistency and standardisation of HealthPathways programs across Australia to mitigate the fragmentation of information and referral processes. Consistent pathways were seen as essential for enabling GPs to provide optimal care.

We have different pathways by catchment. Madness. Duplicating…Sometimes you need a bespoke pathway, you really do, because there’s a particular population who’s in your catchment. But for the main part, you would think about 80% of pathways should be replicable, repeatable across the entire Australian population. (PHN staff, ID 4)

Strong partnerships with health services and the alignment of HealthPathways with the state-wide referral criteria being implemented by the Victorian Department of Health were emphasised as key for improving referral efficiency to specialist clinics. The integration of HealthPathways with these criteria in Victoria has strong support from the state government, and participants viewed this as a positive step towards enhancing the quality, consistency and appropriateness of referrals, ultimately improving the effectiveness of patient care.

Now that the [Victorian] Department particularly are using the state-wide referral criteria. That’s the way of the future. (GP Liaison, ID 28)

Reflection on HealthPathways (reflexive monitoring)

Under this theme, findings relate to the appraisal of HealthPathways’ value and impact on both the health system and individual clinical work.

The HealthPathways platform currently does not utilise person-specific logins, resulting in limited data and feedback on its utilisation and impact. PHN staff noted that, although the concept of HealthPathways is perceived as potentially useful by GPs and PMs, the absence of quantitative monitoring hampers accurate assessment of its accuracy and effectiveness.

I keep using this lax language about ‘may use it, may not’, because actually nobody knows. Why don’t we know? Because the system that it sits on isn’t able to give us any information about how many individual users there are, what pathways they’re accessing and even whether or not the action they take with their patient is based on that pathway. (PHN staff, ID 4)

You don’t know whether it’s one person in that practice who’s an extreme die-hard fan using it so many times a day, or whether there’s 10 GPs all using it really regularly. (PHN staff, ID 8)

GPs reflected that receiving feedback on the appropriateness and accuracy of their referrals would help them identify areas for improvement. Such feedback was seen as essential for enhancing clinical practice and avoiding the entrenchment of incorrect habits.

Can be useful to find data on how the GPs are working – i.e. are they doing the right referrals, do the referrals fit the pathways, how many referrals were not appropriate? This feedback can be beneficial for GPs to know what to improve on as it can otherwise become habitual. (GP, ≥10 years of experience, ID 31)

Harder question is [the] other side of that – has that referral been taken up? Specialists and psychologists will write back, but that does not always happen. (Psychologist, ID 16)