Models for individual oral health promotion and their effectiveness: a systematic review

Authors


Dr Julie Satur
Head Oral Health Therapy
Senior Lecturer
Melbourne Dental School
The University of Melbourne
720 Swanston Street
Melbourne VIC 3010
Email: juliegs@unimelb.edu.au

Abstract

Background:  There is a recognized need to deliver oral health information to people during clinical encounters to enable them to develop personal skills in managing their own oral health. Traditional approaches to individual oral health education have been shown to be largely ineffective and new approaches are required to address personal motivations for preventive behaviour. This systematic review aims to identify and assess the effectiveness of behaviour models as a basis for individual oral health promotion.

Methods:  Electronic databases were searched for articles evaluating the effectiveness of health behaviour models in oral and general health between 2000 and 2007. Eighty-nine studies were retrieved and data were extracted from the 32 studies that met the inclusion criteria.

Results:  Thirty-two studies were identified in the fields of clinical prevention and health education, motivational interviewing (MI), counselling, and models based interventions. MI interventions were found to be the most effective method for altering health behaviours in a clinical setting.

Conclusions:  There is a need to develop an effective model for chairside oral health promotion that incorporates this evidence and allows oral health professionals to focus more on the underlying social determinants of oral disease during the clinical encounter. There is potential to further develop the MI approach within the oral health field.

Abbreviations and acronyms:
CHPPHF

Cochrane Health Promotion and Public Health Field

CVD

cardiovascular disease

IMB

Information Motivation Behavioural Model

MI

motivational interviewing

RCT

randomized controlled trial

Introduction

There is a recognized need to deliver oral health information to people during clinical encounters to enable them to develop personal skills in managing their own oral health. Dental professionals have traditionally applied the biomedical model of disease to target causes of illnesses through preventive and educational approaches. The underlying theory behind these approaches is that once individuals acquire the relevant knowledge and skills, they will then alter their behaviour to maintain optimal oral health.1 These approaches to individual oral health education based on paternalistic information giving have been shown to be largely ineffective.2 They overlook the broader context determining human behaviours, including factors such as the social, economic, political and environmental circumstances. Collectively, these factors are known as the social determinants of health.3,4

While oral ill health conditions are almost completely preventable, data on the prevalence of disease indicates that preventive approaches are often ineffective. This supports the view that the causes of oral diseases are grounded in the social determinants of health and the cultural and social circumstances in which people live.5 There is little doubt that oral health behaviours are inextricably connected to the other behaviours people apply to cope with their lives. These social and cultural contributors can be difficult to consider appropriately within the context of the clinical dental encounter. There is a need to advance this thinking to develop a model for effective chairside oral health promotion for use by oral health practitioners.

A number of research based health behaviour models exist to inform more evidence based approaches to developing personal skills and oral health literacy among individuals. These client or patient centred health behaviour models draw on psychological theories of self-efficacy, motivation, counselling and behaviour change.6 Such models have been utilized for smoking cessation, alcohol and substance abuse counselling, nutritional counselling and HIV/AIDS prevention with some success.7 Psychological models of behaviour change provide a framework for understanding the processes of change and the influence of social circumstances for individuals on their behaviours. Counselling has a range of meanings, but generally includes targeted and interactive information giving approaches that address individual behaviour, thereby meeting the specific needs of individuals.8 Motivational interviewing (MI) is an approach that uses collaborative and empathic interactions to develop a client’s internal and autonomous motivation to change.9

The aim of this study was to systematically review the literature to identify models for health behaviour change and evaluate evidence for their effectiveness. This work will inform the development of a model for oral health promotion in the clinical encounter.

Methods

Study selection

A search of the following electronic databases was used to identify relevant papers for inclusion in the review: Medline, PsychInfo, Cinahl, ERIC (CSA) and the Cochrane Library. Keywords included ‘behaviour modification’, ‘health behaviour modification’, ‘health education’, ‘health promotion’, ‘counselling’, ‘motivational counselling’, ‘motivational interviewing’, ‘oral health’, ‘general health’, ‘smoking cessation’, ‘diabetes prevention’, ‘HIV prevention’, ‘models, theories, interventions’, ‘effectiveness’ and ‘individual’. The reference lists from review papers on MI were also searched.10–12 This yielded 89 studies that were screened according to the inclusion criteria listed in Table 1, yielding the 32 studies which were included in this review. To validate the selection procedure a second researcher examined the 89 potentially relevant papers against the inclusion criteria, and any discrepancies were discussed until an agreed decision was reached.

Table 1.   Inclusion and exclusion criteria
CharacteristicsInclusion criteriaExclusion criteria
InterventionStudies evaluating the effectiveness of individual oral health promotion interventionsStudies assessing the effectiveness of dental materials, techniques and technology
Studies that report evaluations on context, process, impact and outcomes for individuals and communitiesStudies which have only provided treatment as their intervention
Studies measuring effectiveness over a minimum time-frame of one month
Studies evaluating interventions of common risk factor processes and messages guided by, and relevant to, the framework for oral health promotion, including those which do not have oral health as the primary focus
Studies originating from Australia, NZ, UK, Canada, USA and Europe
Articles where data on interventions, impact on health outcomes of people, government regulation (policy), or funding model are not reported.
Studies which only present observational data, i.e., an audit.
Study designSystematic reviews, meta-analyses and randomized control trials were sought initially. Other controlled trials and comparative studies were also considered.Narrative reviews, editorials, letters, articles identified as preliminary reports when results are published in later versions, articles in abstract form only.
PublicationArticles between 2000 and 2007Studies published prior to 2000
language(English language articles)(Non-English publications)

Quality assessment

The studies were assessed for quality against a combined schema incorporating type of evidence as designated by the Type of Evidence Schema13 (Table 2), the Health Gains Notation framework14 (Table 3) and the Cochrane Health Promotion and Public Health Field (CHPPHF) quality assessment screening questions for qualitative studies, quantitative studies and systematic reviews.15

Table 2.   The strength of all studies was evaluated according to the Type of Evidence Schema shown below
Type I evidenceat least one good systematic review (including at least one randomized controlled trial)
Type II evidenceat least one good randomized controlled trial 
Type III evidencewell-designed interventional studies without randomization
Type IV evidencewell-designed observational studies
Type V evidenceexpert opinion; influential reports and studies
Table 3.   The effectiveness of all studies was evaluated according to the Health Gains Notation shown below
1. “Beneficial”effectiveness clearly demonstrated 
2. “Likely to be beneficial”effectiveness not so firmly established
3. “Trade-off between beneficial and adverse effects”effects weighed according to individual circumstances
4. “Unknown”insufficient/inadequate for recommendation 
5. “Unlikely to be beneficial”effectiveness is not as clearly demonstrated as for 6
6. “Likely to be ineffective or harmful”ineffectiveness or harm clearly demonstrated

Quantitative studies were assessed for quality using the CHPPHF Quality Assessment Tool for Quantitative studies.15 This tool assessed for internal and external validity and rated the following criteria: selection bias, allocation bias, confounding, blinding, data collection methods, withdrawals and dropouts, statistical analysis and intervention integrity. Qualitative studies and systematic reviews were assessed and ranked for quality using the questions developed from the Critical Appraisal Skills Programme for the CHPPHF.15 Of the 32 studies, eight studies were evidence Type V and were not ranked further for quality. The studies were ranked for quality and classified as weak, moderate or strong evidence as outlined in Table 4.

Table 4.   Quality ranking criteria
 Reviewer quality rankingType of evidence 
Weak0–5 Poor or no quality ranking because of a lack of information or poorly designed study methods
Moderate6–7.0IV–VWell reported but weaker study designs or better studies lacking information. May also include multi-strategy programmes with poor attribution or process and impact evaluations but no outcomes reported
Strong7.1–10I–IIIWell-designed studies with good methods reporting including RCTs and systematic reviews

Results

The 32 studies yielded included nine studies of clinical prevention and health education, three studies of counselling, nine studies of models based interventions and 11 studies of motivational interviewing. Level of evidence was found to be strongest in studies on counselling interventions, followed by motivational interviewing interventions, models based interventions, and lastly, clinical prevention and health education models, as seen in Table 5.

Table 5.   Quality ranking – distribution of type of evidence
ModelWeakModerateStrong
Clinical prevention and health educationPalmer, 2004Chamberlain et al., 2005Watt, 2005
Shih, 2005Hudon et al., 2004
Plourde, 2006Vanobbergen et al., 2004
Phelan, 2006Ostberg, 2005
Motivational interviewingEmmons, 2001Martino et al., 2007Knight et al., 2006
Britt, 2004Kasil et al., 2006Dunn et al., 2001
Tappin et al., 2005
Burke et al., 2003
McCambridge et al., 2005
Channon et al., 2007
Kalishman et al., 2005
Counselling  Proper et al., 2004
Ebrahim et al., 2006
Hausen et al., 2007
Behaviour change modelsRise, 2004Laatikainen et al., 2007Eime et al., 2004
Bourbeau et al., 2004Buchaman et al., 2006Lumley et al., 2004
Carr et al., 2006
Renz et al., 2007
Thomas et al., 2007

Clinical prevention and health education

Of the nine studies found in the area of clinical prevention and health education, four studies did not meet the quality assessment criteria and were not included in this synthesis.

Health education for the purposes of this category means information and expert advice provision with a passive patient. There was strong evidence to suggest that clinical prevention, treatment and educational interventions are not the most effective approaches to reducing disease levels amongst individuals and communities. A qualitative study of preventive advice delivered by physicians in clinical encounters identified many barriers to integration of prevention into routine practice.16 Some of the barriers included patients’ motivation to carry out preventive measures, physicians’ workload and priorities, physicians’ lack of insight regarding patients’ risk behaviours, unclear recommendations, personal attitudes and beliefs. The participating physicians in the study also felt that they were not reaching the vulnerable populations who are at higher risk of disease. A study that applied a Child Advocacy Training Model of Prevention discovered similar challenges to integration of prevention into practice, including time limitations, varied motivation among practitioners and lack of available resources.17 Results of a qualitative study on adolescents’ view of oral health education indicated that oral health education provided in a clinical encounter is generally positively but vaguely remembered and not always applied into practice.18 The study showed that even participants who displayed knowledge of certain oral health topics did not always succeed in practising healthy habits.

Clinical prevention and health education interventions using standardized messages have failed to achieve sustainable improvements in oral health and, therefore, cannot be considered to significantly reduce caries prevalence of populations.1,19 These programmes have been described as palliative in nature, ignoring the underlying factors that create poor oral health.1

Counselling

Of the three studies found in the area of counselling, one was a systematic review and two were randomized controlled trials (RCTs). Counselling was defined as targeted and interactive information giving approaches that address individual behaviour, thereby meeting the specific needs of individuals.8 Only one of the three studies found the counselling approach to have significant effects on disease levels. A systematic review of multiple risk factor intervention comprised of counselling, education and drug therapy found the approach to be ineffective in achieving reductions in cardiovascular disease (CVD) mortality when used in general or workforce populations of middle-aged adults.20 This intervention’s results also stated that the counselling approach led to some lifestyle changes that resulted in small reductions in blood pressure, cholesterol, salt intake and weight loss. However, these changes were found to have little or no influence on risk of heart attack or death. One RCT of individual counselling showed no significant effect of this approach on sick leave levels of the participants and confirmed that more research is required in this area.8 Another RCT looked at effectiveness of oral hygiene and dietary counselling, and non-invasive preventive measures, on reducing DMFS among children with active initial caries.21 The approach was found to be successful in reducing total need for restorative care although lesion-specific results in regard to reversing active caries lesions were not reported in the study’s findings.21 Significantly less caries developed in the experiment group compared to the control group, with mean DMFS increments for the experimental and control groups being 2.56 (95% CI 2.07, 3.05) and 4.60 (3.99, 5.21), respectively (p < 0.0001): with a prevented fraction of 44.3 per cent (30.2 per cent, 56.4 per cent). The results indicate that this approach can significantly reduce dental decay among caries active children living in areas of low caries risk. However, the results also showed that frequent counselling sessions alone had little effect on oral health habits and dietary habits of the participants, other than increasing the use of xylitol and fluoride lozenges. The large number of fluoride and chlorhexedine varnish applications received by children in the experimental group undoubtedly also contributed to the reduction in DMFS increment obtained among this group. The three studies did not report their findings in regard to sustainability of effects, or the time and cost-effectiveness of the counselling approach.

Models based interventions

Of the nine studies found using interventions based on behaviour models, four studies were systematic reviews, one study was a RCT and two studies were observational. There were two studies found that did not meet the quality assessment criteria, thus these studies are not included in this report.

The studies yielded in this field found most interventions based on models of health behaviour change to be effective in reducing disease levels and risk behaviours in different settings. Two systematic reviews focused on Behaviour Change Models in Smoking Cessation. Results from a systematic review showed that smoking cessation programmes in pregnancy reduce the proportion of women who smoke, consequently reducing the incidence of low birth weight and preterm birth.22 Another systematic review examined the evidence for effectiveness of integrating behavioural interventions for tobacco use in dental services.23 This study concluded that tobacco cessation counselling interventions based on behaviour change models delivered by dental professionals may be effective in helping tobacco users to quit. A systematic review focusing on family based programmes and their effect on preventing adolescent smoking showed that children’s decisions to smoke are influenced by their family and friends, reinforcing the significance social circumstances and surroundings have on the health decision-making process.24

One systematic review found psychological models of behaviour change, such as the Health Belief Model, the Theory of Planned Behaviour, the Theory of Reasoned Action, the Locus of Control and the Protection Motivation Theory, to be effective in improving adherence to oral hygiene instructions amongst adults with periodontal disease.25 Psychological approaches to behaviour change resulted in improved plaque scores and was associated with enhanced self-reported brushing and flossing. The results of this study also showed improved self-efficacy beliefs in relation to flossing. However, no effect on dental knowledge or self-efficacy beliefs in relation to toothbrushing was found. A RCT applied ecological principles of behaviour change to a safety behaviour intervention on prevention of injuries in the recreational game of squash.26 Ecological principles of behaviour change provide a comprehensive perspective on intra-personal factors, policies and physical environmental influences on health-related behaviours, such as the use of protective eyewear during sport. The study concluded that behaviour change models provide comprehensive and a particularly relevant set of principles and guidelines for approaching safety initiatives in sports.

An observational study looking at diabetes prevention through lifestyle interventions found that this approach is feasible in primary health care settings, leading to reductions in risk factors.27 Another observational study searched for evidence of effectiveness of the Transtheoretical Model of Behaviour Change in reducing consumption of carbonated drinks by adolescents.28 The results showed that 45 per cent of adolescents in the sample reported some attempts to modify their behaviour. The study supported the Transtheoretical Model’s predictions in regards to the balance between the pros and cons (the positives and negatives of making the change) varied depending on which stage of change the individual was in. Thus, this study’s findings suggest that this model of behaviour change may be a useful framework through which more tailored health promotion interventions can be designed.

Motivational interviewing

Eleven studies that used the motivational interviewing (MI) approach to behaviour change were found. Of these studies, two were systematic reviews, one was a meta-analysis of controlled trials, five were RCTs and three were observational. Of the 11 studies found on MI, two did not meet the quality assessment criteria and were not included in this synthesis. Evidence shows that MI has been applied to a variety of fields including diabetes, asthma, hypertension, heart disease, substance abuse, smoking, HIV risk reduction, diet and exercise.10,11,29–32

Most studies displayed positive results for the effectiveness of MI in altering behaviours. One systematic review found positive results for the effects of MI on psychological, physiological and lifestyle change outcomes11 and concluded that MI has the potential to be an effective intervention in physical health care settings. Results of a systematic review indicated that 60 per cent of 29 studies yielded at least one significant behaviour change,29 confirming this model’s potential to positively influence individual behaviour. This study found MI to be potentially cost-effective, taking less time than comparison methods and only slightly adding to the total time of usual care. On the other hand, the results of one RCT suggested that MI alone was not cost-effective when applied to smoking cessation interventions in specific target groups, such as heavily addicted pregnant women who continue to smoke at maternity bookings.30

A meta-analysis of adaptations of MI found that 51 per cent of people who received MI treatment were improved at follow-up compared with 37 per cent receiving no treatment or treatment as usual.10 Regardless of comparison group, the effects of adaptations of MI did not appear to fade over time. Clients were prepared for change over a small number of sessions, and further sessions were organized to help clients to initiate and maintain change. This study also found that the MI interventions were shorter than alternative methods by an average of 180 minutes. The results of this study suggested that adaptations of MI could have positive consequences for a wide range of important life problems beyond target symptoms, concluding that MI impacts clients in broad and socially relevant ways, in addition to bringing about target symptom relief.10

The results of a RCT indicated that MI is an effective method of facilitating behaviour changes in teenagers with diabetes type 1, specifically in producing long-term improvements in glycaemic control, psychosocial well-being and quality of life.31 Another study (cluster randomized trial) showed that a single one-hour session of MI led to significant changes in drug use after three months. However, after 12 months the changes had largely, but not entirely, faded. The study concluded that deterioration of effect is the most probable explanation, however taking into consideration the reactivity to a three-month assessment, a late Hawthorne effect cannot be ruled out.33

The use of an adapted MI model (the Information Motivation Behavioural (IMB)) Model for HIV risk reduction counselling concluded that even a brief single exposure to this style of HIV prevention counselling could reduce HIV transmissions.32 The study found that motivational counselling demonstrated most positive outcomes for women in regard to HIV risk reduction. On the other hand, men who received full IMB sessions evidenced relatively greater use of risk-reduction behavioural skills and relatively lower rates of unprotected intercourse over six months follow-up, and had fewer sexually transmitted diseases. A study that taught medical students to apply brief motivational interviewing to promote client behaviour change within the time constraints imposed by a busy medical practice showed a positive response from the participants. The medical students were interested in the approach and were committed to incorporating MI into their future medical practice.12 Only one study was found that applied MI to oral health counselling. The results of this observational study suggest that MI and the Transtheoretical Model of Behaviour Change might be useful in constructing and focusing oral health counselling for school children that concentrates on personal dynamics of change.34

Discussion

In this review many approaches and models for health behaviour modification have been identified. Effectiveness of these approaches was evaluated and the studies were ranked for quality. All of the studies and models have confirmed the complexity of behaviour modification and the need to develop effective approaches to health promotion in the clinical context.

It is important to identify the client’s specific needs and concerns when attempting to alter a habit that is damaging to that person’s health. For individuals to change their behaviour they must learn to integrate new skills and knowledge into their everyday life. As the skills are applied to different situations, the individual develops a sense of self-efficacy and confidence in their ability to perform actions, manage challenges and overcome barriers to change. Self-efficacy has been found to play a significant role in determining which activities a person will perform or avoid,35 thus it must not be undervalued when designing a health promotion intervention. There is evidence to suggest that traditional approaches to health education based on information giving and expert advice are largely ineffective, with success rates of only 5 to 10 per cent.36,37 In addition to this, knowledge gain alone rarely leads to sustained changes in behaviour.38 Thus, there is a need for more effective approaches that focus on the broader context which determine patterns of behaviour.

The concept of readiness to change may help explain why simple provision of advice is limited in its effectiveness.39 Readiness for change can be understood as an individual’s current thoughts, feelings and attitudes regarding their intention to institute change in habits.40,41 It has been found that people who are asked to make radical changes to lifestyle vary over time in their readiness to change.42 For an individual to be ready to change, they must feel both confident in their ability to make changes and realize that change is important to them.43 Therefore, sensitivity to the client’s degree of readiness to change becomes an essential part of the communication and negotiation process. Interventions need to recognize the various stages clients may be in, in order to tailor appropriate measures to meet needs effectively.

The value of the Transtheoretical Model of Behaviour Change is to explain how individuals change their behaviour and to describe their readiness for change.34 There has been a considerable amount of research conducted using the Transtheoretical Model as a theoretical framework for behaviour change.44 This framework assumes that behaviour change is a dynamic, non-linear process that involves a number of distinct stages (pre-contemplation, contemplation, preparation, implementation and maintenance) through which an individual will pass as they adapt a new behaviour or alter a current behaviour. This model may assist the development of effective health promotion programmes by ensuring that interventions are designed to target the particular needs and beliefs of individuals and their readiness to implement and sustain change.

The Transtheoretical Model provides a framework for understanding the change process itself, while the MI approach provides a means of facilitating this change process.45 Motivational interviewing is an evidence-based, client-centred, practical and personalized counselling approach that is based on Prochaska and DiClemente’s Transtheoretical Model of Behaviour Change.9 The focus of this approach is also to prepare people for behaviour change through helping clients to explore and resolve ambivalence about change and make their own decisions about why and how to proceed.43,46 Thus, the person’s decisions about behaviour change are simply supported and guided by the health practitioner. MI helps to build trust, reduce the individual’s resistance to change and aims to alter how the client responds to problematic situations.

MI influences the decision-making process by actively engaging clients in an evaluation of their behaviour and the negative aspects of change.44,47 Studies on MI have found that clients feel listened to and understood by their health practitioner, while the health practitioners gain a greater sense of achievement from recognizing developments in the client’s readiness to change as important progress, rather than seeing the concrete behaviour change as the only goal.48

This approach appears to be consistent with a number of models of health behaviour, sharing constructs such as patient’s expectations about the consequences of engaging in the behaviour, the influence of a person’s perception of personal control over the behaviour and the social context of the behaviour. In contrast to traditional health education approaches where the professional often assumes the ‘expert’ role, MI places the client in the role of the expert; therefore the client decides how to interpret and integrate the information in the context of their own lives and social circumstances, and whether it is relevant.49 MI provides health practitioners with a means of tailoring their interventions to suit the patient’s needs and degree of readiness to change.48

The concepts of readiness, importance and confidence are useful to the extent that they allow the practitioner to understand the social context of the behaviour.50 Adopting a client-centred consulting style involves heightened sensitivity to the client’s social and environmental circumstances. This allows the practitioner to bring the underlying social determinants of health into the consultation and generate patient motivation more appropriately.

Addressing causes of oral disease in isolation from the clients’ life and social circumstances is ineffective in both the short and long term. There is a need to develop effective approaches for chairside oral heath promotion that allow oral health professionals to focus more on the underlying determinants of oral disease during the clinical encounter and to respect the expertise that patients have in their own lives. This allows dental practitioners to adjust provision of care appropriately to meet patients’ specific needs and support their skills, and ability to maintain their oral health. While this study will contribute to the development and testing of a model for chairside oral health promotion, it is also useful in informing the practice of individual clinicians. In understanding the broader context that determines behaviour, it is possible to be more effective in working with people to minimize disease-causing behaviours and harmful habits.

There is a considerable body of theory and research that suggests that MI may be effective for clinical areas beyond addiction, for which it was originally developed.51 Most of the studies included in this report found the approach to be cost-effective and beneficial to the clients. The results from this systematic review have shown that levels of MI training, MI skill and the optimal duration for MI interventions in health care settings remain unknown as often these data were not reported. There is an urgent need for more rigorous, good quality trials and research to assess the effectiveness of MI in domains outside addiction in order for its broader application to be considered.

Conclusions

This study has reviewed the evidence for health promotion and behaviour change targeted at individuals within a health practitioner consultation. Following quality assessment, 32 studies conducted between 2000 and 2007, were reviewed and strong evidence is synthesized in this report.

Clinical prevention and health education approaches alone have been found to be unsuccessful in achieving sustainable improvements in oral health. A conceptual movement away from the traditional biomedical ‘downstream’ and victim blaming approaches, to one addressing the ‘upstream’ underlying social determinants of oral health is necessary. Thus, there is a need for more supportive rather than judgemental approaches to oral health behaviour change. Motivational interviewing, based on the Transtheoretical Model, has been found to be one of the most effective approaches to altering clients’ behaviours. This approach has been successfully utilized in a variety of fields, including substance abuse, smoking, HIV risk reduction, diabetes and obesity. There is potential to develop this approach further within the oral health field.

Acknowledgements

The study was supported by a research grant from the Research Committee of the Melbourne Dental School, The University of Melbourne and the Victorian branch of the Australian and New Zealand Division of International Association for Dental Research (ANZIADR). We also acknowledge the contribution of Ms Cara Waller in the validation process.

Ancillary