Is toothbrushing behaviour habitual? Cues, context, motivators and patient narratives

Objectives: Toothbrushing is generally considered as a key self- care behaviour necessary for maintaining good oral health. Although twice- daily brushing is widely recommended as beneficial since this provides both biofilm removal and a regular ap-plication of fluoride, some people such as those with low socio- economic (SES) backgrounds often brush less regularly. Habit theory identifies that behaviours become habitual when they are undertaken repeatedly in response to a particular cue within a stable context. Once behaviour becomes automatic, long- term maintenance is more likely even if motivation wanes. Establishing toothbrushing as a habitual behaviour is therefore an intervention strategy which may help reduce health inequalities. Therefore, the objective was to more fully understand the nature of toothbrushing behaviour in adults and what prompts its Twenty- nine semi- structured interviews were undertaken North West England. Thematic analysis using a framework method was used to code the data into theoretically derived categories. Results: Morning toothbrushing was found to be strongly integrated into cleansing routines and was identified as a behaviour predominantly initiated in response to visual cues. Some toothbrushing behaviour was prompted by internal cues, expressed as strong feelings or urges. These were more related to night- time toothbrushing habits. Common morning motivators aesthetics and social acceptability. Evening motivators relatively idiosyncratic. of the mouth a hard, manual working day emerged as a strong driver for low SES males.


| INTRODUC TI ON
Regular toothbrushing with a fluoridated toothpaste has been repeatedly demonstrated as critical to preventing dental caries and maintaining good periodontal health. [1][2][3][4][5] It simultaneously removes harmful oral biofilm while delivering fluoride to the oral environment to help resist demineralization following sugar consumption. 6,7 However, although toothbrushing is considered a relatively simple and low-cost behaviour, many people brush less regularly than is recommended, and this is associated with poorer oral health. 8,9 Previous approaches to interventions aimed at improving toothbrushing behaviour have predominantly focused on targeting behavioural change components, such as motivation (eg motivational interviewing), capability (eg self-efficacy interventions) and opportunity (eg supply of toothbrushes, toothpastes), necessary to instigate better behaviours. So far there has been relatively limited consideration of behaviour maintenance aspects of toothbrushing interventions, [10][11][12] and to what extent any intervention effects diminish over time. 13,14 Habit formation theory pays more attention to the behaviour maintenance aspect of behavioural change interventions. Habits are defined as automatic behavioural responses initiated by individualized, salient cues (stimuli). 15,16 Cues can be either time (initiating behaviour at a certain time of day), activity (initiating behaviour at the end of a predetermined activity) or event-based (initiating behaviour by something in the environment). 17,18 Habits are established by the constant repetition of behaviour following encounter of a cue in a stable context. Over time, when the salient cue is encountered, an automatic 'stimulus generates impulse to act as a result of a learned stimulus-response association' occurs. 16 Automaticity, that is the automatic initiation and/or execution of behaviour following encounter of a cue, is therefore the 'active ingredient' or defining feature of a habit. Automaticity is defined by four distinct features: (a) absence of deliberation; (b) absence of awareness; (c) absence of mental effort and (d) absence of conscious control. 19 It is the automatic nature of habits which has thought to determine long-term behavioural maintenance.
Although the word 'habit' is often used to refer to frequent, persistent and customary behaviours, there is some confusion in the way the term is used, even within the psychological literature. 15 This is reflected in the dental literature where toothbrushing routines have been studied in relation to frequency and routines, but not necessarily in relation to mental context-behaviour associations or impulses to act which are generated by sub-conscious or automatic processes. A review by Gardner 15 recommends that habits are defined as 'a process by which a stimulus generates an impulse to act as a result of a learned stimulus-response' association, so automaticity is a key defining feature in studies aiming to link the investigation of routines to underlying habit theory.
Consideration of toothbrushing behaviour as a 'habit' is starting to emerge in the literature, although this has been mainly a theoretical proposition with limited empirical work. 20,21 Aunger (2007), for example, identified that toothbrushing is commonly routinized and performed in such a recurring, standard sequence of actions that the behaviour becomes largely sub-conscious. 22 He suggests the existence of scripts within people's routines, which chunk sequences of behaviour into sub-scripts where behaviour follows such a standardized pattern, that these are then done without intruding on working memory availability, thus enhancing cognitive economy. This proposition, however, does not appear to be supported by the little previous empirical work which has been done on toothbrushing routines in adults, which found that those with less regular and more flexible routines, brushed their teeth more frequently. 23 On the other hand, a recent study of children's toothbrushing habits and routines, which is the first to investigate toothbrushing routines using a measure of habit strength, found that after controlling for other factors, each increase Self-Report Habit Index was associated with a 21% decrease in missed evening brushing. 24 Having a more stable day-to-day routine was associated with stronger brushing habits both in the morning and evening. 24 Since toothbrushing behaviour in children is closely associated with the process of socialization, and with maternal influence, 25,26 more studies of toothbrushing behaviour in adults are needed. This study therefore aimed to investigate the nature of toothbrushing behaviour in adults, especially considering what prompts its instigation (deciding to do) and execution (carrying out the behaviour) in the morning and evening and to investigate whether processes (once established) are automatically or cognitively driven.

| Data collection
A qualitative descriptive approach was taken to explore the nature of toothbrushing behaviour in adults. Ethical approval was obtained by the NHS Research Ethics Committee (15/EE/0053), and NHS research governance approvals were obtained prior to the study. Patients were approached when attending their urgent dental care appointment and were offered a choice their location for their semi-structured interviews, with the offer of being at home, in a café or at the UDC. All participants, however, wished to have this conducted at the UDC, while they waited for their dental appointment (or afterwards in some cases). Written informed consent was obtained from all participants.
Twenty-nine semi-structured interviews, lasting up to 30 minutes, were conducted by a single researcher (HR) using a topic guide.
The topic guide explored people's dental experience, cues (stimuli), detailed description of daily routines including toothbrushing (both morning and evening) and motivators (motivation for behaviour repetition) for toothbrushing behaviour (Appendix 1 online). The guide was flexible to allow for the questioning of emerging data. In addition, after the first few interviews, the wider research team (including experienced supervisors) reviewed the transcripts of each of the interview to determine areas within the interview where further probing would have been appropriate or question phrasing could have been altered. All interviews were audio-taped and transcribed for analysis. Patient transcripts were anonymized using codes P1-P29.
In addition, field notes were made during and after the interviews as memos to capture additional observational data and the interviewer's reflections. Reflexivity is also important to ensure quality control. 27 Since the researcher who undertook the interviews and led data analysis had a dental background and was aware of theory relating to habit formation, this may have influenced perceptions and judgements about emerging themes. The interviewer who had been trained by undertaking a course on interviewing techniques was supported by two supervisors (one with a dental and one with a nondental background), who gave feedback on interview technique and the emerging analysis. Analysis therefore required that there was a constant critical reflection of the researcher's and wider research team's positionality, in terms of toothbrushing behaviour, values and personal routines, and how these might have impacted upon the knowledge formed via co-production (ie between the researchers and the participants).

| Sample and recruitment
Initially, purposive sampling was undertaken to ensure a mixture of participant ages, socio-economic status (SES) and daily toothbrushing frequencies (see Table 1). Participant SES was determined by translating their home postcode into an Index of Multiple Deprivation (IMD) quintile-the fifth quintile being the most deprived. In addition, the individual's occupations were also recorded as a measure of SES. The principle under-pinning purposive sampling in qualitative methods is the enabling of selection of 'information-rich' cases to study. 28,29 It was important to have a range participant's SES as establishment of habitual behaviour, via routines, may be more challenging in low socio-economic groups due to a more 'hectic' nature of their lived experience. 30 As too was having a sample with a mixture of toothbrushing frequencies to help determine toothbrushing habits at different times of the day. It was therefore important to understand the lived experience from a range of groups, to obtain an in-depth understanding of the phenomena through this lens.
However, when morning and evening toothbrushing behaviour emerged as being different in nature, the sampling method shifted to include a theoretical sampling dimension to ensure we had sufficiently diverse types of morning and evening routines to enable us to understand this more fully. Theoretical sampling followed from the emerging data analysis when initial exploratory interviews gave way to confirmatory fieldwork which sought to gather confirming as well as disconfirming cases (patients). 28 Recruitment finished when data saturation had been reached, and further cases did not add to the analysis.
A dental centre providing NHS urgent dental care in a deprived area of North West England was chosen as the recruitment site because patients attend from both high and low SES backgrounds.

| Data analysis
Our enquiry used both deductive and inductive approaches with an interplay between the two. [31][32][33] The enquiry set out using deductive reasoning using key components of habit theory such as unconscious behaviour and the importance of stable contexts and cues as useful theoretical approaches to help guide exploration of the toothbrushing phenomenon. Using deductive reasoning, we sought to explore how habit theory might be expressed across a range of individuals' circumstances and routines, such as shift work, while being open to finding that the theory was not confirmed by the data. This meant that as the enquiry progressed, we progressively used inductive progresses, keeping an open mind to the possibilities of the data and the perspectives of participants. 33 As themes emerged, we reverted to deductive processes to seek out cases to verify and elucidate what appeared to be emerging until data saturation was reached.
Framework analysis was used to analyse data. 34 NVIVO was used to organize data into broad themes initially developed from the theoretical literature on habitual theory behaviours, such as 'cues' and 'motivators'. Data collection and analysis occurred simultaneously so that the researcher could iteratively develop and refine the initial set of codes (codes and categories being developed or confirmed from the data rather from purely the initial theoretically proposed categories). Emerging themes were nested in subsequent interviews; additional interview data did not add to the analysis.
To strengthen the trustworthiness of the research, an audit trail of research decisions and how data were collected, recorded, and analysed was kept enabling cross-checking and confirmability. Line-by-line coding of a selection of transcripts was undertaken by the wider analytic team, including a nondentist, to ensure consensus of coding and understanding of the data, and interpretations were subject to peer debriefing in presentations of findings to neutral colleagues. 35 Participants are identified with codes P1-29 to preserve anonymity. In dialogue, Int = interviewer and P = participant.

| RE SULTS
Twenty-nine participants were interviewed. The most frequently

| Toothbrushing is initiated by stimuli embedded in routines
When participants described their daily routines, they referred to specific stimuli which prompted toothbrushing behaviour in the morning and evening, which was evocative of the importance of cues in initiating the enactment of habitual behaviour. 16

| Toothbrushing motivators vary between people and by time of day
Toothbrushing motivators were found to vary between participants, with differences found between morning and evening behaviour.
Evening toothbrushing motivators appeared to be more idiosyncratic than morning ones. Two important motivators were included a) perceived aesthetics and b) cleansing.

| Perceived aesthetics
A common morning motivator described was the perceived aesthetic benefit of brushing teeth, increasing social acceptability.
Participants reported wanting to make themselves presentable to others at the start of their day.
I try to because when you start work the first thing is people look at your outer skin, your oral health and stuff like that, especially when you are engaging with people you have got to do all this.

| Cleansing
One important motivator, especially reported by people with low SES backgrounds, was the cleaning effect of brushing. Participants felt motivated to remove the contaminants from the day (such as smoking and unhealthy eating) and restore their mouth to a fresh, unspoiled state.
Well I love the feeling of just like having a clean mouth after a long day after eating all my junk food and whatever it's just nice to feel and the taste of them, I like the taste of them knowing its fresh.
(P16, Female) The 'cleansing' motivator was described as not only a physical, but a psychological cleansing. 'Cleansing' the mouth following a hard, manual working day, was sometimes a motivator to brush at an additional, third time in the day (once home from work). Interestingly, this additional toothbrushing behaviour was still strongly linked to bodily cleansing and cued by taking a shower or having a bath.
Just to be clean and know that I have got the day gone out of my mouth.

| D ISCUSS I ON
This study helps in developing our understanding of the nature of morning and evening toothbrushing behaviours, which in turn aids the effective design of future interventions aiming to establish twice-daily toothbrushing. In summary, toothbrushing behaviour in adults appears to be automatically performed in individuals who brush regularly. Indeed, toothbrushing (principally morning) behaviour is predominantly integrated into personal daily sequences of behaviour, cued by several different salient stimuli. Finding toothbrushing to be an automatic behaviour built into routines is consistent with the theory outlined earlier by Aunger. 22 Toothbrushing in the evening was found to be relatively less habitual, perhaps due to a lack of a strongly routinized evening brushing routine, resulting in some inconsistent toothbrushing prior to bed.

This mirrors a recent study into children's toothbrushing behaviours
where the lack of an evening routine was found to be related to poor brushing behaviours. 25 Therefore, future habitual interventions would need to include a focus on establishing a stable evening routine prior to bed and then including toothbrushing within this sequence of events. Alternatively, toothbrushing could be attached to an already stable evening behaviour. For example, finishing the evening meal might be identified as an appropriate external cue to brush, and habitual brushing established around the end of the meal. This approach would require caution and future exploration to ensure that the efficacy of brushing was not affected by other activities such as snacking on cariogenic foods or drinks after brushing.
Motivators for behaviour repetition are important to habit establishment, 37 although these can diminish over time without impacting upon habitual behaviour, since action (eg brushing) eventually becomes automatically initiated by cues without conscious processing. 38 This study suggests that motivating components of future interventions should include a personalized, and not necessarily scientific reason for toothbrushing. This may be particularly useful when developing toothbrushing interventions for low SES groups.
Examples may include psychological as well as physical cleansing from the burdens of the day. There is good evidence, from outside dentistry, that mouthwashing results in psychological benefits for individuals because it is effective as a moral cleansing activity. 39 This study only focused on eliciting patient narratives around toothbrushing behaviour, although it should be acknowledged that other aspects of daily routines such as habits relating to sugar consumption, interdental cleaning, dental visiting, smoking and alcohol consumption also impact oral health, although these were beyond the scope of this study. However, given that there have been so few studies concerning the automatic nature of oral health behaviours and that we know that complex behaviours achieve lower levels of automaticity than simple behaviours, 29 we decided to limit this investigation to toothbrushing since this was the simplest of these behaviours to study, and also usually frequently undertaken. It is likely that more complex behaviour such as sugar consumption behaviours would require design of more complex interventions, and perhaps habit theory may only apply to a proportion of the behaviour sequences involved. Nevertheless, these other behaviours are still important future areas of study in this light. For example, a systematic review for healthcare seeking, for example, identifies that there are very few studies in this area which takes this perspective. 46 Our study demonstrates that adults indeed, do engage in toothbrushing as an automaticity driven behaviour. It finds that although the environmental context is important (the presence of visual cues, eg of a toothbrush in the bathroom), although the behaviour may be so ingrained in scripts of personal routines that it is sustained when travelling, or during shift work-when the wider environment has shifted. The interplay between the social and environmental context and personal routines is an important area for future study-especially since a previous study of adults found that it was those with more regular and less flexible routines, who brushed less often. 23 This will be important when considering how oral health promotion programmes might help mitigate the effects of socio-economic differences and reduce inequalities in health. It also remains to be seen whether interventions which explicitly treat toothbrushing as routine are more effective than those who do not. 22 These might involve inserting a toothbrushing 'scene' into a nightly routine ,for example, but would probably also involve using visual cues (eg a note on the fridge), and motivational components (eg cleansing away the day).
In conclusion, this study helps understand the nature of toothbrushing behaviour which appears, when established, to be performed in a habitual manner. However, there is a need, perhaps, to start to consider toothbrushing behaviour in the morning separately from the evening when designing future habitual interventions. Due to the distinct differences in cues to initiate behaviour and motivators to drive the behaviour repetition identified, future design of interventions would require tailoring to the specific target behaviour.
The important components would be to identify and consider salient cues to initiate toothbrushing and ensure appropriate motivators for behaviour repetition are considered.

ACK N OWLED G EM ENTS
This work was supported by the Oral and Dental Research Trust under the GlaxoSmithKline Research Grants Programme, 2014.

CO N FLI C T O F I NTE R E S T
No conflict of interest to report.

AUTH O R CO NTR I B UTI O N S
Dr Heather Raison contributed to conception, design, data acquisition and interpretation; performed analyses; drafted; and critically revised the manuscript. Professor Rhiannon Corcoran contributed to conception, design, data interpretation; drafted; and critically revised the manuscript. Professor Rebecca Harris contributed to conception, design and data interpretation and analysis; drafted; and critically revised the manuscript.

DATA AVA I L A B I L I T Y S TAT E M E N T
Research data not shared.