Interprofessional diabetes and oral health management: what do primary healthcare professionals think?

Background: Diabetes and periodontitis have a bi-directional relationship. And yet, collaborations between primary healthcare practitioners in diabetes and oral health care are minimal. This study explored the views of general practice and oral health professionals on the link between diabetes and periodontitis, and interprofessional diabetes and oral health management. Methods: A sequential mixed-methods exploratory research design was used. General practice and oral health professionals were recruited from four community health centres in Melbourne. Quantitative surveys explored participants’ experiences, attitudes and knowledge of diabetes and oral health management and interprofessional collaboration; qualitative follow-up interviews explored survey responses with selected participants. Results: 58 participants completed the online surveys; 22 then participated in semi-structured interviews. Participants generally had strong intentions to collaborate interprofessionally in diabetes and oral health management. Most general practice and oral health professional participants were willing to perform simple screening for periodontitis or diabetes respectively. Themes from the interviews were grouped under three domains: ‘a ttitude towards diabetes and oral health management’, ‘subjective norms’ and ‘perceived behavioural control’; and an overarching domain to describe participants’ ‘current practice’. Existing siloed primary healthcare practices and lack of formal referral pathways contribute to poor interprofessional collaboration. Most participants were unsure of each other’s responsibilities and roles. Their lack of training in the relationship between general and oral health, compounded by systemic barriers including time constraint, high dental costs, long public dental waiting list and unintegrated health information systems, also impeded interprofessional care. Conclusions: The diabetes and oral health link is not properly recognised or managed collaboratively by relevant primary healthcare professionals in Australia. There is, nonetheless, strong intentions to engage in interprofessional diabetes and oral health care to contribute to improved patient outcomes. Primary healthcare professionals need dedicated and accredited interprofessional training and competencies, formal referral systems and sustainable health policies to facilitate collaboration.


Introduction
Current evidence shows a bidirectional link between diabetes and chronic periodontal disease (periodontitis). Diabetes is associated with increased risk of an inflammatory response to periodontal micro-biota. Severe periodontitis is three-to four-times more prevalent in people with diabetes. Periodontitis on the other hand seems to affect blood glucose levels in patients with diabetes. Severity of periodontitis may be associated with increased diabetes episodes requiring hospitalisation [1][2][3] . People with both diabetes and periodontitis have increased risk of premature tooth loss, poorer diet, poorer diabetes control and more cardiovascular complications. In the 1990s, chronic periodontitis was added as the sixth complication of diabetes mellitus 2 .
Both Australian and international guidelines recommend that diabetes care providers should undertake oral health reviews and provide dental referrals if required. The Royal Australian College of General Practitioners (RACGP) recommends to assess the oral health of patients with diabetes 4 . The International Diabetes Federation (IDF) recommends the strengthening of interdisciplinary collaboration to improve general patient outcomes and as a primary means to prevent periodontitis for patients with diabetes 5 . Despite these recommendations, the potential for screening patients with diabetes for early management of gum problems is often overlooked in primary care. There are siloes in practice and a lack of collaboration between general practice and oral health professionals [6][7][8] . Consistent with current guidelines, general practitioners (GPs) usually prescribe short-term pain relief and/or antibiotics for teeth and gum issues and may advise patients to see a dentist. This is usually the extent of their involvement with oral health management. Similarly, diabetes screening is rarely performed by OHPs despite research showing significant proportions of dental patients have undiagnosed diabetes or pre-diabetes risks 3,9,10 .
Much of the current literature nonetheless focuses on assessment of diabetes screening in dental settings or the evolution of the dental profession 3,9,11 . Some have also explored the views of healthcare professionals on integration of diabetes and periodontitis management including some recent work conducted at the Centre for Oral Health Outcomes, Research Translation and Evaluation at Western Sydney University in New South Wales [12][13][14][15][16][17] . However, limited research on interprofessional diabetes and oral health care has been conducted in Victoria or focused on public community health service (CHS) setting.
This study aimed to explore the knowledge, practice and attitude of Victorian general practice professionals (GPPs) [including GPs, primary health care nurses (PCNs), diabetes educators (DEs)] and oral health professionals (OHPs) [including dentists (Ds), dental hygienists (DHs), oral health therapists (OHTs), dental therapists (DTs) and dental assistants (DAs)] in managing diabetes and periodontitis and their views on interprofessional care in CHS setting.

Ethics approval
This research was approved by human research ethics committees at University of Melbourne (ID 1750835), Deakin University (ID 2018-190) and La Trobe University (ID 1750835).
Written informed consent from the participants for the publication of findings from this research was obtained. In accordance with the requirements of the ethics committee and the approved research protocol, details that would potentially identify participants due to the small sample size have been removed or replaced with codes in this publication.

Study design
This is a sequential mixed-methods exploratory study. Quantitative online surveys and qualitative semi-structured interviews were conducted with healthcare professionals. Quantitative analysis reported descriptive statistics only. Qualitative analysis used a mixed inductive and deductive approach to explore the experiences of healthcare professionals and reporting was guided by the consolidated criteria for reporting qualitative research checklist (COREQ) 18 .

Advisory group
An advisory group guided the implementation of the study. It consisted of representatives from consumers, practitioners (GP, DE, general dentist, periodontist, oral health therapist) and managers of the CHSs involved.
Research team PL is an academic primary care researcher; EB is a diabetes nurse educator; HC is an academic dental public health researcher; MC and AT were honours research students and RM was a public general dentist at the time of the project. The team has an interest in promoting interprofessional primary healthcare.

Setting and participant recruitment
GPPs and OHPs were recruited purposively from four CHSs in Victoria. The project was presented by the research team to eligible staff at two CHSs; email invitations with a short introduction video were sent to eligible staff via their management teams at the other two. Staff were assured that participation was voluntary. Project description and an anonymous survey link were given to all participants.
On completion of the survey, participants' contact details were sought if they opted to participate in follow-up interviews. A matrix (gender, age, professional role) was used to select, for the interviews, a broad representation of those who provided contact details to ensure maximum variation.

Quantitative surveys
Two online surveys (one for GPPs; one for OHPs) were conducted via the REDCap electronic data capture tool hosted at The University of Melbourne platform 19 . The questions (Table 1  and Table 2) were developed based on a review of the literature and guided by our advisory group. They were piloted with GP registrars, academic nurses and dentists in the research team's network before the surveys were rolled out.
Likert scales gauged participants' agreement with statements relating to confidence, current practice and interprofessional collaboration in oral health or diabetes management and perceived feasibility of screening for periodontitis or diabetes within routine practice. Data were analyzed in Microsoft Excel (2017) (RRID:SCR_016137) to produce descriptive statistics. Google Sheets (RRID:SCR_017679) is a free alternative.

Qualitative interviews
Participants were asked to opt into interviews to explain their survey responses, identify barriers to diabetes and oral management, and suggest ways to improve interprofessional diabetes and oral health management (Table 3 and Table 4). Author AT interviewed GPPs whilst author MC interviewed OHPs either by phone or in-person at the participants' practice. Both were trained by author PL in interview techniques and did practice interviews with authors PL and HC. Questions were pilot tested with students in the Department of General Practice Honours student cohort prior to conducting the interviews.
Interviews were audio-recorded and transcribed, and field notes were taken. Transcripts were offered to participants for review before being imported into QSR International's NVivo 12 qualitative analysis software (RRID:SCR_014802) 20 . RQDA package for R (RRID:SCR_001905) is an open-source alternative. AT coded all GPP interviews, MC coded all OHP interviews while the rest of the team (PL, EB, RM and HC) coded up to six interviews each, ensuring every transcript was coded by at least two researchers. Transcripts were first inductively coded separately and then collectively by the research team. Following several iterative meetings to reach consensus in coding and categorising differences, the research teams decides that the Theory of Planned Behavior model (TPB) which outlines three domains affecting intention to perform a behavior: attitude towards the behavior (or beliefs which influence an individual to perform a behavior), subjective norms (or perceived external pressures as influenced by judgement of others) and perceived behavioral control (or ease or difficulty in performing the behavior as determined by external factors) is congruent with patterns emerging 21 . Deductive analysis using a framework analysis approach then followed using the TPB to identify patterns and elicit themes 22 . The team continued to meet to discuss the themes elicited until agreement was reached.

Survey participant demographics
A total of 58 participants completed the survey between April and July 2018: 20 from general practice (eight GPs, nine PCNs and three DEs) and 38 from dental practice (18 Ds, four DHs, six OHTs, five DTs and five DAs). (Table 5) Table 6 shows the survey results. Most GPPs (75%) had no oral health training in their professional education. The majority rarely or never assessed the mouths of patients (70%) and were not confident in identifying oral disease (60%), discussing oral health with their patients (55%) or managing oral health in patients with diabetes (80%).

Survey results
In contrast, most OHPs (74%) learnt the relationship between oral health and diabetes in their professional training. The majority were confident in identifying risk factors of type 2 diabetes (66%) and discussing diabetes with their patients (82%) and managing patients with both diabetes and periodontal disease (82%). However, most rarely or never consult GPs (69%). Most GPPs (55%) occasionally referred patients to OHPs while most OHPs rarely or never referred patients to GPPs.
All GPPs agreed that oral health screening was within their role (100%) and most were comfortable to perform simple oral health screening (80%). All thought that oral health screening was feasible in practice (100%) but most thought that it would be welcome by their colleagues (80%). Almost all (95%) welcomed oral health training specifically in diabetes management. All agreed that OHPs should screen patients with periodontitis for diabetes (100%) and almost all thought that better interprofessional collaboration would benefit patients (95%). These results are similar to those from corresponding statements for the OHPs. However, only 65% of GPPs said they would welcome the opportunity for continuing education/ training in oral health, compared with 95% of OHPs who said they would welcome continuing education/training in diabetes.

Interview participants and themes
Five GPPs (four PCNs, three DEs) and 10 OHPs (four Ds, two DHs, two OHTs, one DT and one DA) were further interviewed. Interviews lasted 20 minutes on average. One participant declined to be audio-recorded; none took up the offer to review their transcripts or offer additional feedback.
Data saturation was determined to have been reached. Ten themes were grouped under the three TPB domains and an additional overarching domain to describe participants' current practice.

Domain 1: current practice Theme 1: separate diabetes and oral health management
Most GPP acknowledged that they did not routinely assess the mouth of their patients with diabetes.
"I don't usually do it routinely unless there is a particular symptom that they complained of or as I am talking to them I can see that they have got an oral health issue" GP2, female, 51-60 years old, worked 11-20 hours per week, 25-30 years' experience OHPs on the other hand often discussed the diabetes and oral health link during initial patient examination.
"…if the patient says they have diabetes or has maternal or paternal history of diabetes, I discuss with my patients the risk he and she can have. If he or she has already been diagnosed with gum disease, I inform them about why it's so important that (their diabetes) should be controlled." D2, female, 31-40 years • Prompt: What are some barriers you have experienced when broaching the subject of oral health with a patient with type 2 diabetes or risks of diabetes?
2. In the survey, you indicated that you always/often/occasionally/rarely/never conduct oral investigations on patients with type 2 diabetes or risks of diabetes and suspected oral health conditions. Please explain more.
3. How confident do you feel about managing a patient with type 2 diabetes or risks of diabetes and oral health problems in your practice? Please explain more.
• Prompt: What are some barriers you have experienced when managing a patient with diabetes or risks of diabetes and oral health problems?
4. You have indicated that with patients with diabetes or risks of diabetes who have oral health problems, you always/often/ sometimes/rarely/never refer them to an OHP. Please tell me more about why this is so.
5. In the survey, you strongly agree/agree/neutral/disagree/strongly disagree that better collaboration between GPs, PNs & OHPs would benefit patients with type 2 diabetes or risks of diabetes and oral health problems. Would you please explain?
• Prompt: What are your thoughts on possible ways to improve the collaboration between GPs/PNs and OHPs regarding the management of diabetes and periodontal disease?
6. What other suggestions do you have that would enable a GP to better manage patients with diabetes or risks of diabetes and oral health problems?
7. In the survey, an oral health screening tool was mentioned, which involves a visual non-invasive inspection with a torch and a series of approximately 5 screening questions. What concerns would you have regarding the feasibility of implementing this screening tool to patients with type 2 diabetes or risks of diabetes in your practice?
8. What are your concerns regarding the acceptability of your staff in implementing this screening tool to patients with diabetes or risks of diabetes in your practice?
9. How much time would you be willing to devote to training and education related to the impact of oral health on the management of diabetes and with regards to the implementation of the proposed oral health screening tool?
10. What do you think is the role for Oral Health Practitioners in assessing the type 2 diabetes risk status of their dental patients and then referring for management by a GP? Why do you say that? • Prompt: What are some barriers you have experienced when broaching the subject of the risk of Type 2 diabetes with a patient who has periodontal disease? 2. In the survey, you indicated that you are very confident/confident/not confident/not confident at all at managing a dental patient with risk factors for diabetes, such as periodontal disease. Please explain more.
• Prompts: What are some barriers you have experienced when managing a dental patient with risk factors for diabetes (including periodontitis) in your practice?
3. In the survey, you indicated that with patients who have periodontal disease and suspected diabetes, you always/often/ sometimes/rarely/never refer them to a GP. Please tell me more about why this is so.
4. In the survey, you considered it within/not within your role as an OHP to undertake diabetes screening for patients with periodontal conditions. Please explain more.

5.
In the survey, you strongly agree/agree/neutral/disagree/strongly disagree that better collaboration between GPs, PNs and OHPs would benefit patients with risk factors for diabetes? Would you please explain?
• Prompt: What are your thoughts on possible ways to improve the collaboration between GPs/PNs and OHPs regarding the management of diabetes and periodontal disease?
6. What other suggestions do you have that would enable an OHP to better manage patients with risk factors for diabetes?
7. In the survey, you indicated your familiarity/unfamiliarity with the AUSDRISK Type 2 diabetes screening tool. What concerns would you have regarding the feasibility of implementing this screening tool to patients with periodontitis in your practice?
8. What are your concerns regarding the acceptability of your staff in implementing the AUSDRISK Tool to patients with periodontal disease in your practice?
9. How much time would you be willing to devote to training and education with regards to the impact of Type 2 diabetes on the management of patients with periodontal disease, as well as the implementation of the AUSDRISK tool in your practice?
10. What do you think is the role for general practice staff in assessing the oral health of their patients with type 2 diabetes and then referring for management by an OHP? Why do you say that?   old, worked 31-40 hours per week, 6-10 years experience

Theme 2: poor interprofessional communication or collaboration
Even where medical and dental services were co-located, they were siloed in practice.
"I have dentists on-site here, but we only really get called when someone is feeling faint. There is little two-way communication." GP5, female, 51-60 years old, worked 21-30 hours per week, 25-30 years' experience

Theme 3: lack of formal referral process
Most participants tended to refer patients to each other informally.
"I would just ask them if they have seen the dentist. Then they would say yes or no. If they haven't then I would urge them to go (and) make an appointment with the dentist." GP2, female, 51-60 years old, worked 11-20 hours per week, 25-30 years' experience "So I haven't referred any patients to a GP directly to get it (diabetes) screened, but I have requested them to see a GP to make sure that their diabetes is under control so I can go ahead with my treatment plan." D2, female, 31-40 years old, worked 31-40 hours per week, 6-10 years' experience GPPs noted that they received little feedback from OHPs following 'referral'.
"When I refer patients to a physiotherapist or a psychologist, or a cardiologist, I get a letter back. I don't get anything back from our dental services." GP1, male, >60 years old, worked 21-30 hours per week, >30 years' experience Formal referrals from OHPs to non-GP health professionals were more common.
"I have never referred to a GP for diabetes. We do have diabetes educators… and I would refer for that." D1, female, 31-40 years old, worked 31-40 hours per week, 11-15 years' experience

Domain 2: attitude towards diabetes and oral health management Theme 4: responsibilities and roles
Many GPPs admitted that oral health was generally overlooked. Many did not think oral health should be their responsibility.
"I don't think we really know what to do, I think we really leave that to our dental colleagues" GP3, male, 31-40 years old, worked 31-40 hours per week, <5 years' experience In contrast, most OHPs thought they should have a role in diabetes screening.
"I think it should be (within our responsibilities). It isn't though, at the moment." OHT2, female, <30 years old, worked 31-40 hours per week, <5 years' experience However, two dentists expressed apprehension about the 'unfamiliar territory' of the Australian Diabetes Risk Assessment (AUSDRISK) tool.
"Another thing is the waist measurement. I don't know about that. It's also not really in our place to do so." D4, female, <30 years old, worked >40 hours per week, <5 years' experience GPPs generally agreed that diabetes risk screening is viable in the dental setting.
"…(screening) for diabetes is so simple these days it doesn't even require a fasting blood test, let alone a glucose tolerance test" GP1, male, >60 years old, worked 21-30 hours per week, >30 years' experience Most OHPs also felt GPs and nurses could conduct simple oral health screening and prevention. However, some opposed the idea. High dental costs and long public dental waiting list were the most common reasons that GPs, PCNs and DEs gave for their reluctance to refer patients to OHPs.
"Another barrier is cost… (Patients) are so used to bulk-billing and they thought that if medical bulk bills, why not dental as well." PCN2, female, 41-50 years old, worked 21-30 hours per week, <5 years' experience "…even the minor delay of even a week or two is sufficient for the patient to scurry away and say I'll do it another time, and then the opportunity is lost." GP1, male, >60 years old, worked 21-30 hours per week, >30 years' experience

Discussion
Our research aims align with the National Oral Health Plan's recommendations for greater collaboration of OHPs with the broader health workforce 23 . Our findings contribute to a growing evidence base for interprofessional collaboration between medical and oral health professionals and will help support the RACGP guidelines on diabetes management and IDF guidelines on interprofessional collaboration 9,10 . This corresponds with the recommendations from a UK study to develop initiatives and policies to promote and embed oral health management as part of diabetes care 24 .
The TPB model provided the framework to explain the key factors influencing healthcare professionals' consideration of interprofessional care of diabetes and periodontitis 21 . Several attitudinal beliefs and societal normative influences strongly impact their collaborative behavioral patterns. Our results are similar to those from studies that have found many nonoral healthcare professionals do not manage the oral health of patients with diabetes 25,26 .
Like our study, a German study also reported a lack of collaboration from OHPs which was likely a result of the informal nature of 'verbal referrals' usually directed at OHPs 27 . Other research shows that OHPs supported diabetes screening becoming part of oral health professionals' standard care but the convoluted referral system dissuaded them from providing formal referrals 8 . It is important that a simple and structured referral system, like the one between medical specialists, be developed between medical/nursing practitioners and OHPs to promote effective interprofessional collaboration.
Currently, Australian medical and dental practices use completely different information systems that are not integrated. This compounds service fragmentation and suboptimal clinical outcomes. Appropriate policies are required to incorporate information sharing in health systems to support interprofessional collaborative relationships 28 .
It was not surprising that time constraint was a barrier particularly for GPs and dentists. They may be more suited to be involved after the initial primary prevention strategies. The barrier of the healthcare profession 'silo' culture is well-known and is also reflected in Marshall and Spencer's paper which cites a "separateness" between Australian medical and dental practices 29 . However, improved management of periodontitis would potentially improve blood glucose control, which would in turn further improve periodontal health resulting in longer-term fewer visits to GP and dental clinics and ultimately save time and resources.
Further training in diabetes and oral health management would increase healthcare professionals' knowledge and confidence 30 . Ward et al. found that nurses who were confident with their oral health education were more likely to screen patients with diabetes for periodontitis 25 . The importance of interdisciplinary training is consistent with Lamster and Eaves' push for greater interprofessional collaboration and emphasis on respecting all health disciplines, increasing the understanding of each profession's role, providing more effective communication and maximizing safety, efficiency and effectiveness 11 . Currently there are minimal interprofessional training opportunities. Development of future training should have an interprofessional focus, be as conducive as possible and be accredited for CPD.

Strengths and limitations
Our mixed-methods approach allowed an in-depth exploration of participants' views. Although the sample size was small, the wide range of healthcare professionals provided broad perspectives. Unequal representation from different professional groups may impede the generalizability of the findings even though data saturation was reached. Our focus on CHSs with co-located general practice and dental services may have limited the extrapolation of our findings to other settings.

Conclusion
Primary healthcare professionals generally recognized the importance and have strong intentions to engage in interprofessional diabetes and oral health management. Accredited interprofessional training should bridge the divide between medicine and dentistry. Formal referral processes are necessary to improve interprofessional feedback and communication.
Health policies and advocacies need to target dental costs and public dental waiting lists to motivate referrals. An effective and feasible interprofessional collaborative diabetes and oral health care model would contribute to improved patient outcomes. Future studies should include the views of patients, policy makers and other stakeholders.

Data availability
Underlying data Deidentified data of this research will only be provided on request. Reviewers or other researchers intending to reproduce the study may make this request by emailing the corresponding author. This conditional withholding of data is necessary to protect the privacy and confidentiality of the participants who were sourced from a small number of community health services and the final sample size was small.

Author contributions
Phyllis Lau (PL), Hanny Calache (HC) and Rachel Martin (RM) contributed to the conceptualisation and design of the study, funding acquisition, recruitment of participants, collection of the raw data and data analysis. Evelyn Boyce (EB) managed the project administration and assisted with recruitment of participants, data collection and analysis. Anthony Tran (AT) and Matthew Chen (MC) were Honours research students at the time and, under the supervision of PL and HC, conducted the surveys and interviews, and analysed the data as part of the research team. All authors contributed to the drafting, revisions and final approval of the version to be published.

Open Peer Review
Yes