A qualitative exploration of decisions about dental recall intervals - Part 1: attitudes of NHS general dental practitioners to NICE guideline CG19 on the interval between oral health reviews

Introduction The National Institute for Health and Care Excellence (NICE) Guideline CG19 recommends that the intervals between oral health reviews should be tailored to patients' disease risk. However, evidence suggests that most patients still attend at six-monthly intervals. Aim To explore facilitators and barriers to the implementation of CG19 in general dental practice. Methods Semi-structured telephone interviews were conducted with 25 NHS general dental practitioners (GDPs) in Wales, UK. Transcripts were thematically analysed. Results Dentists described integrating information on clinical risk, patients' social and dental history, and professional judgement when making decisions about recall interval. Although most GDPs reported routinely using risk-based recall intervals, a number of barriers exist to recall intervals at the extremes of the NICE recommendations. Many practitioners were unwilling to extend recall intervals to 24 months, even for the lowest-risk patients. Conversely, dentists described how it could be challenging to secure the agreement of high-risk patients to three-month recalls. In addition, time and workload pressures, the need to meet contractual obligations, pressure from contracting organisations and the fear of litigation also influenced the implementation of risk-based recalls. Conclusions Although awareness of the NICE Guideline CG19 was high, there is a need to explore how risk-based recalls may be best supported through contractual mechanisms.


Introduction
In 2004, the National Institute for Health and Care Excellence (NICE) published Clinical Guideline CG19 recommending that the interval between patients' oral health reviews should be determined according to their risk of dental disease. This includes consideration of the patient's medical, social and dietary history. NICE suggested that recall intervals should range from 3-24 months, depending on the outcome of the clinical assessment and that this should be discussed between dentist and patient. 1 Despite this guideline having been published over 15 years ago, evidence suggests that many patients still attend at 6-9-monthly intervals. 2 In 2012, a study reported that while 94% of responding dental practitioners were aware of the guideline, only 3% reported routinely recalling patients according to need. 3 The authors concluded that factors such as dentists' lack of understanding of the guideline, concerns about late diagnosis (poor outcome expectancy), as well as patient resistance to change, may be contributing to poor adherence rates.
Policymakers in the UK have prioritised the implementation of risk-based recall intervals. 2,4,5 It has been argued that the adoption of a risk-based approach to dental recall may foster the more efficient distribution of NHS dental resources, whereby patients are treated according to their dental needs and treatment requirements. 6 However, a recent paper reported that almost half of all dental practices 'rely on regular check-ups for those in good oral health to achieve [contractual] targets' . 7 This study, which is reported in two parts, sought to explore how decisions about dental recall intervals are made in general dental practice. Part one will describe the facilitators and barriers experienced by NHS general dental practitioners (GDPs) working in Wales towards the implementation of riskbased recall intervals. Part two will describe NHS patients' and GDPs' views regarding the potential role of shared decision-making (SDM) in a dental recall interval setting.

Methods
A qualitative semi-structured telephone interview study involving NHS GDPs in Wales, UK was conducted between March and September 2019.

Sampling and recruitment
GDPs were identified from a publicly-available database of NHS practices in Wales. The list was randomised and invitation letters sent until the required number of dentists had been recruited. In order to be eligible, dentists needed to spend at least 50% of their clinical time delivering NHS general dental care and not work at a practice yet involved in the Welsh NHS General Dental Service Reform Programme.
Based on Malterud et al.'s concept of information power, 8 it was anticipated that approximately 20-30 GDPs would be recruited, in order to reflect a range of experiences and practicing environments. Data collection, transcription and analysis were undertaken concurrently to allow follow up of interesting topics. Following the completion of 25 interviews, two investigators judged that the richness of data within a theme no longer appeared to be increasing. Thus, no further interviews were conducted.
The study was reviewed and given a favourable ethical opinion by the Proportionate Review Sub-Committee of the East of England -Cambridge Central Research Ethics Committee (REC ref: 19/EE/0031). All participants provided written consent to participate in the study and to have their data used as part of the research and confirmed this verbally before interview.

Data collection
Interview topic guides were prepared before data collection (Table 1; further details available on request). These were adapted over the course of the study as new findings emerged. Interviews were conducted by HS, a psychologist.

Analysis
The interviews were transcribed verbatim and analysed according to the Braun and Clarke's principles of thematic analysis, 9 supported by NVivo qualitative data analysis software (QSR International). Interviews were anonymised on transcription and checked against the original recording to ensure fidelity. A coding structure in which several 'codes' rested within a 'theme' was developed using a constant comparison method. In total, 20% of transcripts were double-coded by other members of the study team (FW and NJW) to test the fidelity of the coding structure. Any disagreements were resolved by discussion.

Results
Just over half of participating dentists were men (14/25) and the majority (15/25) described themselves as dental associates ( Table 2). The average time since qualification was 10.1 years (range 1-21 years). This was higher for practice owners (average 13.8 years) than associates (9.3 years). In total, 18 out of 25 dentists had qualified since the first publication of the NICE Guideline CG19 in 2004. Interviews generally lasted 20-30 minutes.
The themes relating to attitudes to the NICE Guideline CG19, its implementation and attitudes towards risk-based recalls, are described below. Themes relating to the use of SDM in discussions about dental recall interval are presented in part two.

NICE Guideline CG19 awareness and trust
All dentists were aware of NICE Guideline CG19. Although most were not familiar with the evidence that had informed their development, there was an implicit trust expressed in the NICE Guideline development process: Patients' desire to be involved in recall interval decisions; information need to be involved

Role in practice
Owner 8 Associate 15 Other 2

Date of primary dental qualification
Pre-publication of NICE Guideline CG19 in 2004 7 Post-publication of NICE Guideline CG19 in 2004 18

Agreement with CG19
While most dentists agreed that three-month recall intervals may be appropriate for patients at the highest risk of developing disease, most were uneasy extending the recall interval of low-risk, adult patients to the 24-month maximum recommended by the guideline. They believed that patients would be at risk of developing disease (particularly dental caries or oral cancer) and suffering adverse outcomes due to later diagnosis. As a result, many placed their lowest-risk patients on an annual or 18-month recall schedule: There were a number of practitioners who expressed concern that NICE's decision to recommend risk-based recalls may be politically or financially motivated. A small number of other practitioners perceived that CG19, along with guidelines in general, reduced clinical autonomy and typically did not consider 'frontline' implementation issues:

Implementation of CG19
The majority of dentists reported that clinical risk assessment informed their recall interval recommendations for all patients and only a small number did not consider this to be part of their routine practice: • 'We do literally use it every, every single day with every patient' (GDP20, male, other, qualified post-CG19 publication).
Practitioners described how establishing patients' recall interval involved the synthesis of clinical findings, dental and social history, together with practitioners' clinical experience and personal knowledge of the patient. Dentists highlighted that recall intervals were not a 'tick box' exercise but required clinical insight and was typically more challenging in new dentist-patient relationships: • 'I've been here 16

External barriers to implementationpatients' preferences
Dentists described tension between the implementation of CG19 and delivering care, which takes into account patient wishes about their recall interval. Most dentists believed that the majority of patients who regularly attended would prefer to come for check-ups more regularly than their risk profile may indicate. Many had experienced difficulty trying to change the recall of particular patients; usually when attempting to extend the recall of low-risk patients from 6-12-months. However, some acknowledged that reducing recall intervals to three months for high-risk patients also presented challenges and that these patients rarely attended at these intervals due to time, money and/or because irregular attendance patterns were more common among those with the highest rates of disease: •

External barriers to implementationcontractual factors, workload pressures and litigation
Dentists described how contractual pressures and a target-driven culture within NHS general dental practice influenced their implementation of risk-based recall intervals.
There was a recognition that regular six-month recalls for low-risk patients resulted in optimal generation of Units of Dental Activity (UDA). Some reported that placing higher-risk patients on three-or six-monthly recalls could also lead to additional Health Board-scrutiny. Several expressed concern that contract managers Time constraints were also identified as a potential barrier to the assessment of individual risk within clinical encounters. Some dentists were concerned this would likely be exacerbated by ongoing NHS general dental service reform in Wales, while others were concerned about potential litigation arising as a result of extending recall intervals: • 'To assess risk, you have to have had quite a bit of a conversation with a patient. The pressure of the practice is you don't have all the time in the world to do that. So you're relying on good assessment of risks with a small amount of information, which I suppose is risky in itself ' (GDP11, male, dental associate, qualified post-CG19 publication) • 'I know I'm the generation that's been brought up with six-month check-ups, but we're also getting sued left, right and centre. If I say a year and I'm wrong, I'm in trouble' (GDP13, male, practice owner, qualified pre-CG19 publication).

Discussion
In light of increasing emphasis on the implementation of risk-based dental recall intervals within dental policy, 2,4,5 this study sought to describe the attitudes of NHS GDPs working in Wales to the recommendations contained within the NICE Guideline CG19. 1 It also aimed to describe factors which may facilitate or impede their use in practice. Within the present study, practitioners were familiar with the guideline and most reported routinely basing decisions about patients' recall interval on their risk of disease. Clinical judgement and consideration of patients' social and dental history was perceived to be important when making decisions about recall interval and practitioners felt aggrieved when this was challenged by organisations responsible for contractual monitoring. A number of barriers were identified to the implementation of recall intervals at the extremes of the NICE recommendations (3 and 24 months). The vast majority of practitioners were concerned that 24-month intervals posed an unacceptable risk to patients' oral health. Conversely, dentists described how it could be challenging to secure patient compliance with three-month recalls due to perceived patient preferences. This study provides some insight into why, 15 years after the introduction of the NICE Guideline CG19, the majority of NHS dental patients still attend at six-monthly intervals. 2 Within medicine, a variety of factors have been described as influencing adherence to clinical guidelines. These include: awareness and familiarity; agreement; self-efficacy; outcome expectancy; ability to overcome the inertia of existing practice; and external barriers. 10 In the current study, awareness of CG19 was high and although practitioners were often unsure of the evidence underlying the recommendations, they were familiar with their content. The fact that risk-based recalls had been part of the clinical guideline landscape for over a decade and a half appeared to diminish the role of inertia of existing practice. In general, practitioners felt able to use a risk-assessment process to determine an appropriate recall interval (self-efficacy) and felt confident extending the recall intervals of lower-risk patients from 6-12 months without deleterious effects on oral health (outcome expectancy). Conversely, factors which represented barriers to the implementation of risk-based recalls related to agreement with recall intervals at either end of the NICErecommended timescale for adults (3 and 24 months), patient factors and features of the wider clinical environment, such as contractual monitoring, time and workload and clinical negligence liability.
The principal barrier to extension of recall intervals to 24 months for the lowest-risk patients were concerns about the development and/or progression of disease during this period. A recent Cochrane review reported that there was moderate-certainty evidence that there is likely to be little to no difference in terms of caries and gingival bleeding between 6-24-month recalls for adults over a fouryear period. 11 However, data to support this outcome came from one trial; while this trial was large and robustly conducted, it did not consider oral cancer development, 12 which was cited by several dentists in the present study as a concern related to long recall intervals. Future research is therefore indicated to explore the effects of dental recalls at different intervals on the diagnosis of pre-malignant or malignant lesions and size and stage of oral cancers at presentation.
The use of qualitative interviews in this study facilitated the detailed exploration of GDPs' attitudes towards risk-based recall guidelines. Guided by accepted theories of sample size determination in qualitative enquiry, 9 investigators judged that after 25 interviews further data collection was unlikely to yield substantial new findings. Nevertheless, it is not possible to state with certainty that the views of the participants of this study reflect the full spectrum of attitudes among NHS GDPs in Wales, or that these findings can be confidently generalised to other settings. It is not known whether the recall intervals of GDPs who participated in the study are indicative of the wider profession. It is plausible, however, that the barriers and facilitators to the implementation of riskbased dental recall guidelines reported in this study could contribute to improving the understanding of why so many patients still attend for oral health reviews at close to sixmonthly intervals.
If policymakers and bodies responsible for the organisation of dental services continue to place emphasis on the implementation of risk-based recall intervals and use these within contract monitoring arrangements, it should be with an awareness that there are patient factors external to the control of dental practitioners that act as barriers to compliance with short recall intervals. It has been recognised that there are financial disincentives within current NHS general dental service contracting arrangements against increasing the recall intervals of low-risk patients. 7 Therefore, to fully realise the potential of more equitable and effective distribution of NHS dental resources, commissioners will need to recognise and accommodate barriers to the use of the full range of recommended recall intervals.