I delivered this presentation as an invited speaker at the 2019 Centre for Primary Health Care and Equity (CPHCE) Annual Forum, University of NSW, Sydney, 14 August 2019. [Slides (PDF) : Transcript (DOCX)]
YouTube presentation:
Tam CWM. ED2GP – integrating general practice and emergency care. Centre for Primary Health Care and Equity Annual Forum. University of New South Wales, Sydney. 14 August 2019.
Transcript
Hello, I’m Michael Tam, a GP Staff Specialist at the Academic Primary and Integrated Care Unit of the South Western Sydney Local Health District, and a conjoint academic at the School of Public Health and Community Medicine of the University of New South Wales.
This presentation was first delivered at the Centre for Primary Health Care and Equity Annual Forum, on 14th August 2019, at the University of New South Wales.
I am very pleased to be able to tell you a story about three of our projects undertaken in partnership with the Fairfield Hospital Emergency Department, which fits under the forum theme of integrated care.
First, a little bit about Fairfield NSW to set the context. If you look at a map of Sydney, Fairfield can seem like it’s in the middle of nowhere. However, I like to think that it is in the middle of everywhere as it is a fascinating place! The Fairfield Local Government Area, or LGA, is one of the most culturally diverse in Australia and is home to many migrants who have escaped the war in Syria and Northern Iraq. It is an area of low socioeconomic status, with its residents having one of the lowest median incomes in the country, and with one of the highest per capita expenditure on gambling. Fairfield LGA is number 4 in the country for antibiotic prescribing, and we’re number 1 for amoxicillin! Reflecting on these snippets of information, where do you think Fairfield is for opioid prescribing? Above median or below median? Instinct might make you think that it will be above median, but Fairfield LGA is in the lowest decile of opioid prescriptions dispensed in the Australian Atlas of Healthcare Variation. As before, Fairfield is a fascinating place where we need to challenge our assumptions.
Our story starts just before 2016 with the ED2GP study. Some unexpected findings drove the ED2GP for women study in 2018, and observations from both these studies are the topic of interest for the “Do you understand me?” study that is running currently in 2019.
The ED2GP study began with an idea from one of our excellent medical students at UNSW, Belinda Watson. She was driven by a question from her experiences of clinical rotations. What happened to older people after they were discharged from hospital? Together, we developed a protocol for a relatively small student study proposal for her application for one of the GP Synergy Medical Student Scholarships. GP Synergy is the vocational GP training organisation in NSW and the ACT, and they offer these scholarships to encourage and build GP research. Belinda was successful!
The other coinvestigators were from the ED team – the nursing unit manager, deputy director, and director.
The main aim of the ED2GP study was to quantify the proportion of older people who’d been seen in the Fairfield ED, had followed up with their GP as instructed, within a reasonable time period. As this was an observational study, we also wanted to describe the people who had attended follow up, as well as those who had not.
The primary outcome was the proportion of participants who had attended GP follow-up by day 7 after discharge.
Our inclusion criteria were those aged 65 years and older, and were directly discharged back into the community from the ED. They also needed to have received explicit instructions for GP follow-up in the discharge letter to be included. Prospective participants were those meeting the inclusion criteria over a two-week period in 2016. All these patients were contacted by telephone between day 8 and 14 after discharge from the ED, with data collected from those who consented using a structured interview and from the ED electronic record. Questions included whether they had followed up with a GP, and if so, when, and with whom.
We managed to recruit 50 out of a total pool of 70 prospective participants. The demographics are as on the slide. Most of the participants were Australian Triage Score of 3 or 4. Of note, the median distance from the patient’s home to the GP was 3.8 km. This is likely beyond walking distance for most, as almost half used some walking aid. On the 6-item Lubben Social Network Scale for social isolation, a fifth of the participants were socially isolated.
Around three-quarters of older people had attended GP follow up by day 7 after discharge – with half of these people having attended by day 3. This points towards the imperative of having discharge communication ready at the time of discharge for clinical handover to occur. Interestingly, almost a fifth participants who attended follow up consulted with a GP who was not the addressee of the discharge summary.
The exploratory analysis was very interesting from a hypothesis generation and descriptive perspective. Firstly, a lower proportion of women seemed to have followed up. Whereas almost all men had attended follow up, only two-thirds of women had. Why was this the case? There was some indication that sicker people were less likely to follow up, insofar that those who did not follow up took more regular medications on average and had a longer ED admission. Lastly, although social isolation as measured using the Lubben Social Network Score wasn’t associated with 7-day follow up, one of the items was strongly associated – “how many relatives do you feel close to such that you can call on them for help?”. Those who did not follow-up reported they had fewer relatives that they felt close to.
What does this all mean? I’m more a qualitative researcher than quantitative one and we were particularly interested in this first finding of the apparent stark gender difference. My naïve intuition would have predicted perhaps the converse – women typically have higher rates of GP visits as compared to men. In the quantitative descriptions, there were shadows of possibilities – was it related to access to car transport? However, there was no way of knowing without asking the people.
Shaddy Hanna, another fantastic medical student researcher, undertook the ED2GP for women study as his ILP, or independent learning project in 2018. He too, was a GP Synergy Medical Student Scholar. Our coinvestigator team – larger now – including my colleague and Acting Director of our Unit, another medical student researcher, and the ED team – deputy director, nursing unit manager, and director.
The primary aim was to describe how older women access GP follow-up after an ED visit. We wanted to measure and confirm the previous two-thirds follow up finding. There were some other secondary aims also, specifically measuring GP follow-up intentions in the ED waiting room and comparing it with the information in the discharge letter, and actual behaviour. We wanted to understand the previous finding of one-fifth of patients seeing a GP who was not the address of the discharge letter. This study was undertaken under a critical realist perspective, with the qualitative analysis using Grounded Theory approach, as per Corbin and Strauss.
Inclusion criteria is like the ED2GP study, though women only. An initial structured interview was administered in the ED, specifically, about GP follow-up intentions – were they planning to, and with whom. Some demographic data, and the discharge letter was obtained from the ED electronic record. At day 8 to 14 after discharge, the women were contacted by telephone for a second interview, which included structured and semi-structured questions. We wanted to know not only what their follow-up behaviours were, but their beliefs and attitudes around GP follow-up, especially considering their recent lived experience.
We consented 176 older women in the ED, of which 49 were excluded as there were admitted as an in-patient, and a further 20 as they their discharge letter did not include an instruction for GP follow-up. Of the 107 eligible participants, 7 dropped out at the telephone interview stage resulting in a round 100 participants in the study.
The mean age of our participants was 77 years, about half were widowed, most spoke a language other than English at home, and most did not have a diving license. Of those that did, only half drove daily.
This model is the final output of the analysis. Earlier in the analysis, we had thought that we would describe it in terms of barriers and facilitators, but what emerged from the data was that there seemed to be a mental calculus of the perceived costs, and the perceived benefit or value, of attending GP follow up. For both costs and benefits, there were a range of factors that could be lessened or increased. The model might help explain some of the previous findings – for instance, the challenge of access to physical transport was an important perceived cost. Some specific insights included the cost of the perceived inconvenience to others. Some of the older women indicated that they relied on their family, often an adult daughter or son, to transport and support them to health appointments, and they didn’t want to be, or be seen to be, a burden. Another insight related to the importance of ED messaging. How the ED clinicians relayed the importance of follow up carries weight.
We confirmed that around two-thirds of older women in this context attended GP follow-up by day 7 with a very similar proportion to the ED2GP study.
Of some of the other secondary aims, of the women who had attended follow-up, the majority consulted with the GP, or the same clinic, they had indicated in the structured interview in the ED. We saw the same discrepancy between the discharge letter addressee in this study – this is related to some sort of data process issue in the ED.
Both ED2GP studies identified the high proportion of people who spoke a language other than English at home. In fact, this resulted in recruitment and data gathering challenges as those studies were undertaken in English. The single most common language spoken, after English, was Arabic.
Jehan Karem is our fantastic medical student researcher this year undertaking her ILP in the “Do you understand me?” study. She is fluent in both Arabic and English. With the track record of the partnership between our Unit and the ED, we were successful in obtaining one of the South West Sydney Research Grants with a specific focus on disadvantaged populations.
The team now includes not only the APICU and the ED clinicians, but also Ms El-Khair, the Consumer Participation Manager at Fairfield Hospital, whose expertise with the local population helped ensure that our translations (verbal and written) were appropriate, and also Dr MacIsaac, an emerging community GP leader with significant previous rural ED experience.
The main goal of this study is to measure the concordance in health communication between Arabic-speaking patients and doctors, with a focus of how this is associated with spoken English-language proficiency. However, as this study is observational of a largely poorly studied and reported topic, we are also gathering deeper descriptive data on the languages spoken by this population, the perceptions of communication adequacy, and what methods of translation and interpreting were used.
This study recruited patient and ED doctors as participant pairs. The patient participants were all adults, were discharged directly back to the community, and must identify as Arabic-speaking. For the main outcome of concordance, three specific questions – what was the presenting medical problem, what was the diagnosis, and what were the discharge instructions – were asked and recorded for the doctor, and the patient in structured interviews in the ED, immediately at the point of discharge. Interviews with patient participants were in either English, or Arabic, or both, depending on their preference. The patient and doctor responses are then compared – concordance is decided by the clinician investigators by consensus.
Data collection was complete a few weeks ago in June 2019 and we are now at the point of analysis. 172 patient-doctor participant pairs were recruited. Already, some descriptive findings challenge our expectations and point towards possible issues of structural inequity in investment. For instance, of the data representing 172 patients cared for by the ED team, some of whom have limited to no spoken-English language proficiency, it was observed that a professional health interpreter was not, and possibly, could not, be used within the episode of care even once.
This series of three studies have helped build the sense of partnership and research capacity at a very important interface of the health system. This is particularly so for disadvantaged communities in a disadvantaged part of Sydney. With the track record, we were able to obtain a small competitive grant this year – this is very welcome as it will allow us to disseminate some of these findings back to the local community through a community engagement forum.
And further, I’m very chuffed that if you go to the GP Synergy website and search for the medical student scholarship, you’ll find that Belinda Watson’s experience is used as a case study for the successful outcome of the scholarship.
The first study is published, available here. The second study is being written up, and the of course the third is still underway. Apart from publications, the data from these studies will help inform on-the-ground initiatives and service implementations, that should benefit the local community.
Thank you!
This presentation would not have been possible without firstly all the generous participation of the patients and staff of the Fairfield ED, the administration staff at the Academic Primary and Integrated Care Unit, and of course, my medical student researchers Belinda, Shaddy, and Jehan. Special thanks to GP Synergy for the scholarships to the students, and South West Sydney Research for the grant.
