Evaluation of a community dental clinic providing care to people experiencing homelessness: A mixed methods approach

Abstract Background People who experience homelessness have higher dental treatment needs compared to the general population. However, their utilization of dental services and levels of treatment completion are low. Peninsula Dental Social Enterprise, a not‐for‐profit organization in the United Kingdom, established a community dental clinic to improve access to dental care for this population. Objectives To evaluate the impact and acceptability of the community dental service for patients and examine the barriers and enablers to using and providing the service. Methods The evaluation included a retrospective assessment of anonymous patient data and thematic analysis of semi‐structured interviews with patients, support staff and service providers. The interviews were thematically analysed. A cost analysis of the dental service was also conducted. Results By 18 February 2020, 89 patients had attended the clinic. These included 62 males (70%) and 27 females (30%), aged 38.43 years on average (SD ± 11.07). Of these, 42 (47%) patients have completed their treatment, 23 (26%) are in active treatment and 24 (27%) left treatment. In total, 684 appointments (541.5 hours clinical time) were given. Of these, 82% (562) of appointments were attended (452.5 hours clinical time). The 22 interviews that were conducted identified flexibility, close collaboration with support services and health‐care team attitudes as key factors influencing service utilization and continuity of care. Conclusions This study provides details of a highly acceptable and accessible dental care model for people experiencing homelessness, with recommendations at research, practice and commissioning levels.


| INTRODUC TI ON
Homelessness is associated with increased morbidity and mortality. 1 Dental problems are among the most common health concerns affecting people experiencing homelessness, 2,3 with higher levels of untreated dental disease and more missing teeth than the general population, 4,5 causing poorer oral health-related quality of life. 6 Severely limited access to dental care is compounded by high levels of non-attendance and low levels of treatment completion. [7][8][9] Both the lived experience of homelessness and characteristics of the health-care system contribute to the low uptake of dental services. 10 Disproportionate differences in oral health between population groups are due to an interaction of a number of factors (eg socioeconomic and political environment), many beyond an individual's control. 11,12 Dental service utilization contributes to oral health inequalities. 13 Watt and colleagues state that addressing this requires 'coordinated strategic action at both clinical and population levels'. 12 Freeman and colleagues 14 have developed a theoretical framework for 'inclusion oral health' focusing on innovative solutions to tackle inequalities associated with poor oral health in individuals experiencing social exclusion. Their action plan addressing oral health services, research and dental education can make dentistry a powerful catalyst to reduce inequalities. 14 Clearly, dental teams and services have a key role in improving 'access and the quality of dental care for vulnerable groups', acting as 'advocates for policy change' to reduce oral health inequalities. 12 This is reflected in the UK National Health Service (NHS) long-term plan, which prioritises the health care of those with additional needs. 15 This plan also highlights the important role social enterprises play in addressing health-care needs, 15 since they can respond more flexibly to patients than other NHS bodies. The need for flexible service provision accommodating the complex needs of people experiencing homelessness is consistently highlighted in existing literature as strongly influencing utilization. 7,9,16 Peninsula Dental Social Enterprise (PDSE) is a not-for-profit organization responsible for running the dental education clinics of the Peninsula Dental School, University of Plymouth. It is committed to improving oral health and reducing inequalities in the South West of the UK, through education, community engagement, training and treatment. 17 One of its main aims is to ensure access to dental care for all, particularly those excluded from mainstream dentistry, 18 including the homeless community. In response to the significant NHS dental waiting list in Plymouth city (over 14 000 people) 19 and repeated calls for improved access, 20 PDSE established a community dental clinic in January 2018 for those experiencing homelessness.
PDSE's approach lies within Freeman and colleagues' inclusion oral health framework, which suggests that 'dentistry could act as an agent for social inclusion as a more responsive, all-encompassing form of oral healthcare and delivery'. 14 Despite the acknowledged importance of such clinics, research exploring their impact and/or effectiveness has been limited, and mostly descriptive or quantitative. An important missing element is exploration of care models and processes that support or inhibit the delivery and use of dental clinics for individuals experiencing homelessness. 7,9,21

| Aim
This research aims to describe a care model developed for people who experience homelessness, evaluate its impact and acceptability from a patient perspective and examine the barriers and enablers to providing and using the service.

| Description of model
The PDSE community dental clinic is located at the Dental Education Facility in Devonport, one of the most deprived areas in Plymouth. 22 It is currently a pro bono contribution to the local community.
Initially, the clinic treated people experiencing homelessness, expanding within the last year to include individuals using drug and alcohol services, as well as vulnerable women who risk of having multiple children removed from their care.
The first patients were triaged through a student project conducted in a residential homeless centre. 23 Later, referrals were made through the 'Teeth Matter' oral health intervention project, 24 the research findings of which have informed the development and running of the service. 24,25 Thereafter, referrals have been made primarily through the lead volunteer of the homeless centre and through support workers based in the other organizations that PDSE collaborates with. The lead volunteer, acting as a link worker, also facilitates referrals from a GP outreach service. She has 10 years of experience in the homeless sector, concentrating on health issues.
The community clinic began operating in January 2018, for half a day per week. This increased to a full day in August 2018 and then, due to high demand and success, two days per week in September 2019. A salaried dentist provides both routine and urgent treatment. Subject to patients' consent, appointments are arranged in coordination with the lead volunteer or a support worker, who also provide appointment reminders, transport to the clinic and chaperoning during treatment, as needed. The model of care is presented in Figure 1.

Patient or Public Contribution
• Potential patients, peer advocates with lived experience of homelessness and community care-givers were involved in the design of the service evaluated in this paper.
• Patients and community care-givers were interviewed as part of this study.
• A community care-giver also contributed to the interpretation of data, as part of critically revising the manuscript.

| ME THODS
The evaluation included a retrospective assessment of anonymized patient data and semi-structured interviews conducted with patients, support staff (a support worker and a volunteer), as well as service providers. A basic cost analysis of the service was also conducted.

| Retrospective data analysis
The retrospective analysis focused on patient demographics, attendance figures, number and type of treatments provided and treatment status (complete/incomplete).

| Theoretical approach
A phenomenological approach 26 was adopted to study the lived experiences of patients, care providers and support staff in receiving, providing or facilitating care at the clinic.

| Design
This was a qualitative research study.

| Recruitment
Participants recruited using purposeful sampling were approached through a gatekeeper. Other stakeholders including support staff and service providers (managers, administrators and clinicians) were invited via email. Participants were given the opportunity to ask questions and signed consent forms prior to being interviewed.

| Setting
Patients and support staff were interviewed at the residential homeless centre. Service providers were interviewed at PDSE's clinic premises.

| Data collection
Semi-structured, face-to-face interviews were conducted from September 2018 to February 2020, until data saturation was achieved. Topic guides were used (Appendix S1) with the opportunity for participants to expand on issues important to them. The guides were informed by the findings of a systematic review and primary research conducted earlier by the research team. 10

| Data analysis
A descriptive analysis of demographic and clinical data was conducted using IBM SPSS (version 24). Interview transcripts were uploaded onto NVivo 12 software (QSR International Pty Ltd. version 11,2015), then thematically analysed [by RB] 27 using an inductive approach. Reflective thematic analysis was chosen for its flexibility, and because it allows researchers to identify and interpret themes/ patterns in a data set across different groups of people, leading to greater insight. 27,28 Following the six steps described by Braun and Clarke, 29 a researcher [RB] immersed herself in the data, identified initial codes using a line-by-line approach, grouped the codes into themes and then reviewed the themes. The researcher then defined and named the themes and produced the report. To ensure rigour in the analysis rather than 'correctness' in the coding, a second experienced researcher [MP] reviewed the analysis and questioned how the data were coded, assumptions made, and the rationale for decisions.

| Cost analysis
This is based on the NHS publicly funded model and includes the operating costs of the clinic to PDSE (based on pay and non-pay costs) and the cost per case, with a comparison to NHS funding where the service to be formally commissioned. Community inputs, currently provided on a voluntary basis, were not costed.

| Ethical approval
The study was reviewed and approved by the Faculty of Health   Table 1.   The operating cost of the clinic to PDSE of £152.59 per hour includes pay costs (dentist and dental assistant) and non-pay costs (clinic overheads, consumables, dental materials and laboratory costs). The average cost per course of treatment is £854.50. This compares unfavourably with the funding available (£300) from the NHS if the service were state funded only.

| Qualitative research
From 22 interviews (nine PDSE staff members, 11 patients, one support worker, and one volunteer), key themes were identified and grouped within the following domains: barriers to accessing dental care in general; barriers to accessing and delivering the clinic; respective enablers; impacts of the clinic; and suggested improvements. The themes and sub-themes are discussed below supported by verbatim interview extracts.

| Barriers to accessing dental care in general
Patients and staff members identified a number of barriers including previous dental care experiences that 'create a fear based around

| Barriers to delivering the clinic
Of the few barriers identified, most related to challenging behaviours attributed to 'severe mental illness', addiction and/or aggression.

Staff
The non-judgemental, empathetic, 'friendly and helpful' (patient, par- identified as beneficial as people were less likely to forget.

Student and research involvement, clinic location and environment
Many participants described the initial contact with students and research staff at the residential homeless centre, the clinic's location

Continuity provided by 'appointments always being on a Monday'
(staff, participant 11) was seen as beneficial by both patients and staff. Use of a private surgery rather than bays in a larger multiple-patient teaching surgery was appreciated by patients. It offered them the possibility of sharing their oral health history in private, recounting for example loss of teeth through a violent domestic abuse assault, or act of 'self-harm' as disclosed by one participant.

Flexibility
Flexibility was identified as most influential enabler. This included allowing another patient to step in if the original patient could no longer attend, responding to patient circumstances that affected attendance in an understanding and supportive way, providing greater allowance for missed appointments than usually permitted and allowing patients to acclimatize to the clinic before any dental work starts.

Funding
The clinic's funding structure was identified as

| Impact of the clinic
Participants identified a number of patient benefits (Table 4).

| D ISCUSS I ON
Our study has shown that the community dental clinic is highly successful in terms of uptake of care and subsequent attendance. It is positively perceived by patients, support staff and health-care providers alike and has a significant positive impact on patients who demonstrate willingness to engage in treatment. Flexibility, close collaboration with support services and attitudes of the health-care team strongly influence the utilization of the service, continuity of care and attendance rates.
We found that 42 out of 89 (47%) of patients had completed a treatment plan, while 27% failed to return for treatment completion.
In contrast, previous research evaluating services for people experiencing homelessness 7,9,16,21  the dental team to take patients' circumstances into account and provide a truly patient-centred service. Link workers can also assess whether a patient needs additional support to attend or to, for example, complete a medical history questionnaire. Moreover, they can notify the clinic of last minute cancellations due to a medical concern or other unforeseen issue and identify other patients who can make use of the appointment, avoiding lost clinical time. Thus, link workers can support service sustainability and patient satisfaction. Ideally, they should have experience in working with patients with complex needs and understand the importance of oral health to be able to motivate patients to seek treatment and help them keep their appointments.
For the approach to be fully effective, flexibility in service provision is essential, reflecting findings of previous studies. 9,16 High failure-to-attend among this population may be due to the inability of services to accommodate chaotic lifestyles. 16 Thus, adapting to patients' diverse needs is paramount in promoting uptake.
People with experience of homelessness commonly have a history of marginalization, at societal and health-care service levels, 8 compounded by a perceived stigma from health-care teams 10 which may exacerbate anxiety. 6 Our findings demonstrate the importance of the dental team's approach to patients' dental journeys, with the attitudes and professionalism of both reception clinical staff being highly valued by patients. With regard to the dentist, the patients acknowledged how important it was for them to be able to discuss treatment plans and options. This approach helps patients feel empowered and actively involved in their treatment.

| Implications
For patients with multiple and complex care needs, a reductionist approach is unlikely to work well, suggesting that expanding the pro-

| Strengths and weaknesses of the study
The study explored the views of patients, support staff and providers, offering insight into the views and experiences of all those involved. The use of thematic analysis enabled systematic analysis of the data. In order to ensure its trustworthiness, 28 it is important to ensure the credibility, confirmability, dependability and transferability of the findings. Involvement of a second experienced researcher in the analysis of data ensured credibility. To attain confirmability, the narrative descriptions were supported by the relevant context and quotes so that findings could be trusted. To ensure dependability, the research and thematic analysis process was clearly documented.

| Unanswered questions and future research
Identifying elements of a successful pathway that allows an increased integration of dentistry with other services is important and can lead to improved patient outcomes. For example, considering the high prevalence of tobacco use and alcohol consumption among the homeless community, 6,8 investigating the acceptability, feasibility and effectiveness of providing smoking and alcohol advice at a dental setting, is warranted.
Studies exploring the impact of 'peer support' by those who completed their dental treatment on encouraging uptake and maintenance of dental service use among other people experiencing homelessness are needed. This could help identify attributes sought in a 'peer' or 'link worker' to help promote uptake of care.
Conducting interviews with commissioners, to explore their views and attitudes, as well as challenges, towards flexible commissioning services for vulnerable groups, is recommended.

| CON CLUS IONS
This study provides details of a highly successful, acceptable and accessible dental care model for people experiencing homelessness that could be implemented in other locations. It highlights the paramount importance of delivering a flexible service that accommodates the complex needs of this patient group, working closely with community services, treating patients with compassion and providing traumainformed care. Although successful in terms of patient and provider acceptability, it would be preferable in the interests of sustainability that future services be funded through flexible commissioning by the NHS.

ACK N OWLED G EM ENTS
We would like to thank all participants who took part in this study.
We also thank Ms Alex Gude for transcribing the interviews.

Robert Witton is the Chief Executive Officer of Peninsula Dental
Social Enterprise. He did not participate in data analysis and interpretation. The other authors declare no conflict of interest.

AUTH O R CO NTR I B UTI O N S
MP: Made substantial contributions to study conception and design, and acquisition, analysis and interpretation of data.

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available from the corresponding author upon reasonable request.