Presenting patients with information on their oral health risk: the PREFER three-arm RCT and ethnography

Abstract

BackgroundA new NHS dental practice contract is being tested using a traffic light (TL) system that categorises patients as being at red (high), amber (medium) or green (low) risk of poor oral health. This is intended to increase the emphasis on preventative dentistry, including giving advice on ways patients can improve their oral health. Quantitative Light-Induced Fluorescence (QLF™) cameras (Inspektor Research Systems BV, Amsterdam, the Netherlands) also potentially offer a vivid portrayal of information on patients’ oral health.MethodsSystematic review– objective: to investigate how patients value and respond to different forms of information on health risks. Methods: electronic searches of nine databases, hand-searching of eight specialist journals and backwards and forwards citation-chasing followed by duplicate title, abstract- and paper-screening and data-extraction. Inclusion criteria limited studies to personalised information on risk given to patients as part of their health care.Randomised controlled trial (RCT)– setting: NHS dental practice. Objective: to investigate patients’ preferences for and response to different forms of information about risk given at check-ups. Design: a pragmatic, multicentred, three-arm, parallel-group, patient RCT. Participants: adults with a high/medium risk of poor oral health attending NHS dental practices. Interventions: (1) information given verbally supported by a card showing the patient’s TL risk category; (2) information given verbally supported by a QLF photograph of the patient’s mouth. The control was verbal information only (usual care). Main outcome measures: primary outcome – median valuation for the three forms of information measured by willingness to pay (WTP). Secondary outcomes included toothbrushing frequency and duration, dietary sugar intake, smoking status, self-rated oral health, a basic periodontal examination, Plaque Percentage Index and the number of tooth surfaces affected by caries (as measured by QLF).Qualitative study– an ethnography involving observations of 368 dental appointments and interviews with patients and dental teams.ResultsSystematic review – the review identified 12 papers (nine of which were RCTs). Eight studies involved the use of computerised risk assessments in primary care. Intervention effects were generally modest, even with respect to modifying risk perceptions rather than altering behaviour or clinical outcomes. RCT – the trial found that 51% of patients identified verbal information as their most preferred form, 35% identified QLF as most preferred and 14% identified TL information as most preferred. The median WTP for TL was about half that for verbal information alone. Although at 6 and 12 months patients reported taking less sugar in drinks, and at 12 months patients reported longer toothbrushing, there was no difference by information group. Qualitative study – there was very little explicit risk talk. Lifestyle discussions were often cursory to avoid causing shame or embarrassment to patients.LimitationsOnly 45% of patients were retained in the trial at 6 months and 31% were retained at 12 months. The trial was conducted in four dental practices, and five dental practices were involved in the qualitative work.ConclusionsPatients prefer personal, detailed verbal advice on oral health at their check-up. A new NHS dental practice contract using TL categorisation might make this less likely.Future workResearch on how to deliver, within time constraints, effective advice to patients on preventing poor oral health. More research on ‘risk work’ in wider clinical settings is also needed.Trial registrationCurrent Controlled Trials ISRCTN71242343.FundingThis project was funded by the National Institute for Health Research (NIHR) Health Services and Delivery Research programme and will be published in full inHealth Services and Delivery Research; Vol. 8, No. 3. See the NIHR Journals Library website for further project information.