The main oral diseases and conditions have common risk factors with many systemic diseases (Petersen 2003; Sheiham 2005), which in turn can directly or indirectly compromise oral tissues (Dyer 2006). In response to the evidence linking oral health and general health, recommendations have been made to staff in dental healthcare settings to adopt a common risk factor approach (Petersen 2008; Sheiham 2000). The key principle of this integrated approach is that general health promotion in dental healthcare favours the control of many social and behavioural risk factors and has more of an impact on a large number of diseases than disease-specific approaches (Petersen 2003; Watt 2005). Risks factors for oral and general health tend to co-occur and to cluster among population subgroups at the lower levels of the social gradient (Sheiham 2011). Thus, multiple health behaviour change interventions, which are activities targeting two or more health behaviours, either simultaneously or sequentially, within a limited time period (Prochaska 2010), are likely to have greater impact on public health than single behaviour interventions (Prochaska 2008).
Description of the condition
In this review, health promotion is defined as the planned change of lifestyle and life conditions having an impact on health, using a variety of specific strategies including health education, social marketing and mass communication on the individual side, as well as political action, community organization and organizational development on the collective side, in order to improve the health and well-being of populations (O'Neill 2006), and to reduce social health inequalities (Ridde 2007). This definition is more practical and holistic because it considers individual as well as collective interventions.
Dental healthcare settings may provide valuable opportunities to adopt health promotion practices, as many patterns of disease are not strictly limited to oral health. Through health promotion practices, dental healthcare providers can address both systemic and oral diseases and conditions (Greenberg 2010; Greenberg 2012a; Reznik 2012; Tremblay 2011; Vernillio 2003), including among others, dental caries and sports-related craniofacial injuries (U.S. Community Preventive Services Task Force 2002), periodontal diseases (Petersen 2005), diabetes mellitus (Esmeili 2010; Lamster 2012), adverse pregnancy outcomes (Xiong 2006), oral cancer (Lancet Editorial 2009; Mashberg 1989), oral lesions associated with human papillomavirus (Cleveland 2011; Laronde 2008), and HIV/AIDS (Petersen 2006; Zakrzewska 1999).
Health promotion practices may affect the utilization of other health services; reduce treatment expenses; improve quality of life (Sheiham 2005) for the patients, their family and society; and reduce health inequalities (Lévesque 2009; Sheiham 2011; Vibeke 2011; Watt 2007).
Description of the intervention
Interventions designed to increase health promotion practices in dental healthcare settings come in varied forms. Given that dental healthcare providers work in an organizational context, these interventions can be addressed towards the individual clinician, as well as towards the organizational structure of practice.
Strategies for increasing health promotion practices may include theory-driven interventions (Bonetti 2006; Godin 2008), and educational meetings. For example, training dental healthcare providers through education may allow staff to keep abreast of the latest developments in dentistry and update their knowledge. Training dental healthcare providers enhances their knowledge and skills, and increases their self efficacy to perform health promotion practices (Amit 2011; Reena 2012). For instance, given that a compromised dentition can impede the achievement of dietary goals (Moynihan 2004; Moynihan 2007), so tailored interventions such as nutritional counselling by dental healthcare providers may enhance quality of life among edentulous (toothless) people by facilitating increased consumption of fruits and vegetables (Bradbury 2006).
Interventions that aim to increase health promotion practices are also organizational interventions involving all members of the dental healthcare team, which could promote discussion between dental staff and could reinforce their knowledge, awareness and skills (Rosseel 2011). This could lead to the adoption of agreements by dental organizations, which could increase the promotion of best evidence practices, such as the use of sealants (Clarkson 2008).
How the intervention might work
Due to an ageing population, and increasing rates of systemic diseases and conditions, dental healthcare providers need to adapt their practices. Research examining behaviour change among healthcare professionals (Godin 2008), and complex behaviour change processes (Michie 2009), allows us to make certain hypotheses regarding the mechanisms by which interventions might support health promotion practices. For example, the distribution of printed educational materials may improve the attitudes and skills of dental healthcare providers concerning the adoption of guidelines for tobacco cessation (Albert 2005). Continuing professional development activities can increase the clinical performance of dental healthcare providers and their compliance with infection control practices such as use of gloves, masks and protective eyewear (Cohen 1996; McCarthy 1996).
Considering that individual factors as well as social, physical and environmental conditions influence behaviour (Bartholomew 2011; Kitson 2009; McLeroy 1988; Richard 2011; Rycroft-Malone 2008; Spallek 2010), the social ecological model provides a useful framework to consider the complex interplay between individual, and physical and social influences from the wider environment over time (McLeroy 1988). The ecological approach integrates environmental and individual targets of intervention across a variety of settings including organizations, communities, societies and supranational systems (Richard 1996). In order to capture the mediating mechanisms at multiple levels of influence and to understand how interventions work in dental healthcare settings, the application of the ecological perspective represents a comprehensive approach to identifying and analysing interventions that target multiple influences on dental healthcare providers' behaviours (Bartholomew 2011) (Figure 1 from Richard 1996).
|Figure 1. Ecological model adapted from Richard 1996|
Why it is important to do this review
Dentistry is a dynamic profession and a complex socio-technical domain (Song 2010), continually reshaped by new scientific knowledge, devices, techniques and materials (Beauchamp 2008). There is an increasing body of knowledge of the aetiology of oral diseases and conditions, and the relationship between oral and systemic health. Oral signs and symptoms can be precursors to systemic diseases and conditions. Moreover, the ageing population and the growing prevalence of chronic diseases and conditions, as well as the large number of infectious diseases are all factors that could favour the involvement of dental healthcare providers in general health promotion. Given that certain interventions of general health promotion are not directly within the scope of dental practice, oral health promotion could be a strategy to improve overall health.
Systematic reviews have addressed oral health promotion in dental settings (Satur 2010; Victorian Department of Health 2011), among specific people (McGrath 2009), or using behaviour change models (Yevlahova 2009). However, some reviews in the oral health education and promotion literature conducted in certain contexts have identified the limitations of many interventions to produce sustained improvements in oral health (Brown 1994; Kay 1998).
Previous systematic reviews have addressed the effectiveness of health promotion interventions in specific topics in dental settings, such as tobacco cessation (Carr 2012; Needleman 2010; Studts 2009; Warnakulasuriya 2002), obesity (Suvan 2011), dietary counselling (Harris 2012), and alcohol consumption counselling (McAuley 2011). Only one previous systematic review has considered simultaneously different health promotion practices (Dyer 2006). Despite its relevance, this review only included circulatory disease and cancer and the participants were different healthcare workers and not only dental healthcare providers. Consequently, our review will fill some of the gaps in the current knowledge by considering health promotion practices broadly for all oral and systemic diseases and conditions, and by focusing on dental healthcare providers only.
Furthermore, there is a reported willingness on the part of dental healthcare providers to become more involved in health promotion and prevention programmes, and on the part of patients to receive these interventions (Barasch 2012; Campbell 1999; Greenberg 2012b; Jontell 2009; McAuley 2011; Victoroff 2006). Dental healthcare settings offer a promising venue to perform general health promotion by dental healthcare providers and an opportunity to improve social costs related to oral diseases and systemic conditions. Adopting an integrated approach to health promotion may be beneficial through an impact on the trade-off between resource inputs and resource consequences. For instance, dental healthcare providers' investment in resources (physical, human) could result in beneficial effects, such as more preventive care (Conrad 2010; Pourat 2009), more screening and enhanced quality of life for people with systemic or oral diseases and conditions, or both. Thereby, identifying which interventions can help dental healthcare providers in their work to promote the general and dental health of their patients is expected to improve patient outcomes.
To assess the effectiveness of interventions aimed at increasing general health promotion practices by any dental healthcare provider in dental healthcare settings.
Criteria for considering studies for this review
Types of studies
We will include randomized controlled trials (RCTs), non-randomized controlled trials (NRCTs), controlled before-and-after studies with a minimum of two control groups, and interrupted time series (ITS) studies (where there is a clearly defined point in time of when the intervention occurred and at least three data points before and three data points after the intervention) (EPOC 2012a). For health economics studies, the review will consider relevant studies based upon data sourced from effectiveness studies that meet eligibility criteria for the effectiveness component of the review. We will then include economic evaluation studies, including cost-effectiveness analysis (CEA), cost-utility analysis (CUA) and cost-benefit analysis (CBA) of interventions, and of interventions versus comparators. Moreover, we will consider randomized trials reporting more limited information, such as estimates of resource use or costs associated with interventions, and of interventions versus comparators. We will include published studies in all languages.
Types of participants
Participants will be defined as dental healthcare providers. They will refer to all people in dental healthcare settings directly involved in patient care including dentists, dental hygienists, preventive auxiliaries, dental nurses and dental assistants (Brocklehurst 2012a; NHS 2006; NIDCR 2011). We will consider specific titles in each country in the search strategy. We will exclude undergraduate students and dental technicians.
We will place no restrictions on the setting in which the intervention is offered such as dental offices, hospitals, community settings, workplaces, schools, outreach services, special housing for older people or residential care settings.
Types of interventions
We will include any intervention involving dentists or other providers of the dental team, or both, and which aims to improve health promotion practices and patient outcomes in dental healthcare settings. Interventions may also target social networks of dental healthcare providers, their professional organizations and the communities in which they work. We will differentiate between these in the analyses if possible. We will classify interventions according to the Cochrane Effective Practice and Organisation of Care (EPOC) Group taxonomy (EPOC 2012b).
We will include studies that assess the effects of a single intervention or multiple interventions and strategies (professional, regulatory, mediated by patients, theory-based, tailored, organizational and financial) that aim to improve professional practice, and patient outcomes. We will include:
- all interventions aimed at dental healthcare providers versus no interventions;
- all interventions aimed at a particular group of dental healthcare providers versus the same interventions aimed at other dental healthcare providers; and
- all interventions aimed at dental healthcare providers versus the same interventions aimed at other healthcare professionals, such as physicians, nurses or respiratory therapists.
Professional interventions will include educational interventions, local opinion leaders' interventions, patient-mediated interventions, audit and feedback, reminders, marketing and mass media campaigns (Grimshaw 2008; McAuley 2002). Educational interventions will include printed educational materials used by dental healthcare providers, didactic or interactive educational meetings and educational outreach visits (a personal visit to a dental healthcare provider by a trained person in their own setting) that aim to change the behaviour of dental healthcare providers in order to improve patient outcomes. Formal and informal local opinion leaders are those who could target and change the knowledge, attitudes, and social norms of their peer group. Reminders are the manual or computerized interventions that prompt the dental healthcare provider to perform patient care.
Regulatory interventions will include interventions that aim to change health services, delivery or costs by regulation or law (McAuley 2002), such as changes in medical liability, management of patient complaints, peer review and licensure (EPOC 2012c).
Patient-mediated interventions will be those aimed at changing healthcare professionals' behaviour through interactions with patients (Entwistle 1995).
A theory-based intervention is conducted within an underlying conceptual framework (Grimshaw 2008).
Tailored interventions are interventions planned to improve professional practice by altering barriers to change (Baker 2010).
Organizational interventions are those that involve a change in the structure or delivery of healthcare (McAuley 2002).
Financial interventions include changes in how professionals are reimbursed, and related incentives and penalties (EPOC 2012c). Given that financial interventions are the subject of another review (Brocklehurst 2012b), we will exclude such interventions from this review.
Types of outcome measures
We will include self reported behaviour outcomes if they are objective or measured with a validated scale. Interventions will concern children, adolescents, adults, older people, specific populations (e.g. ethnic groups, vulnerable populations, migrants, homeless people, incarcerated individuals), and people with specific disabilities and systemic diseases. We plan to present the main information concerning the quality of evidence, the magnitude of effect of the interventions examined and the sum of available data for the primary outcomes for a given comparison in a 'Summary of findings' table. Given that we plan to consider more than one main comparison, separate 'Summary of findings' tables for each comparison will be required (Schünemann 2011a). We will use the GRADE approach for preparing 'Summary of findings' tables and to evaluate the quality of evidence of each primary outcome (Schünemann 2011b).
1. Professional outcomes
- Objective measures of dental healthcare providers' behaviours.
- Objective measures of dental healthcare providers' process outcomes such as quality of dental care and time management or workload.
2. Patient outcome measures
- Objective patient outcome measures: mortality, morbidity, utilization of health services, access to health services, medication adherence, blood pressure, plasma cholesterol concentration, plasma glucose concentration, body weight reduction, blood glucose monitoring, blood via finger prick, oral fluids for salivary diagnostics, gingival inflammation and urine testing.
- Self reported psychosocial measures of health status, quality of life, well-being and measures of disability or functional status.
- Self reported behavioural outcome measures: tobacco or smoking prevention, reduction of tobacco use, tobacco cessation, physical activity, consumption of fruits and vegetables, advice on alcohol consumption, dietary counselling, eating disorders, blood pressure monitoring and advice on skin cancer prevention.
- Self reported oral health-related behaviours: oral hygiene, toothbrushing, toothpaste use, use of dental floss, use of fluoride toothpaste, mouthwash, mouthrinse, chlorhexidine, interdental brushes and woodsticks; presence of pit and fissure sealants, denture cleaning practices, use of helmets, facemasks, annual dental visits and mouth guards in sports.
1. Professional outcomes
- Self reported barriers and facilitators to behaviour adoption.
- Attitude, motivation, intention, awareness, uptake of clinical research, satisfaction and implementation of knowledge.
- Increased productivity and revenue.
2. Patient-reported outcome measures
- Self reported oral health-related cognitions: beliefs and attitudes towards oral health-related behaviour.
- Satisfaction or experience with the process of care, acknowledgement of patient.
- Investment in terms of time and transportation to attend the intervention, and additional tests and visits to medical facilities.
- Patient out-of-pocket expenses.
3. Resource use and clinical activities
3.1 Resource use
- Healthcare resources (e.g. re-organization of office space, renting office space, laboratory costs, purchasing equipment and supplies, training, dental team's skills in counselling, recruitment and retention of dental hygienists).
- Indirect medical resource use (e.g. printed educational materials, staff time or number of hours worked).
3.2 Clinical activities
- For dental healthcare providers, for example, changes in waiting time for an appointment; changes in number of dental visits.
- For patients, for example, self reported discomfort-free days; self reported number of days off work.
We will assess the resources invested in health promotion interventions by dental healthcare providers as well as the economic impact of these interventions using cost analysis, CEA, CUA and CBA (Shemilt 2011a). Considering that estimates of the joint distribution of incremental costs and effects accruing from an intervention compared with relevant alternatives are sensitive to variability and values attached to specific health-related outcomes (Shemilt 2006), we will report the relevant data provided by authors. In case of alternative interventions, we will choose data objectively assessed. If possible, we will not choose surrogate or interim outcomes.
We will consider the perspectives of dental healthcare providers, patients and the health system, as well as different time horizons (short, medium and long term) adopted by authors.
For instance, tobacco use has numerous adverse effects on oral health. The short-term effects are aesthetic considerations, such as halitosis, staining and discolouration of the teeth and dental restorations (Bouquot 1992). Medium- and long-term effects vary and include tooth decay, dental abrasion, dental erosion, implant failure, periodontal disease, and tooth loss (Bouquot 1992; Sham 2003), alteration of local physiology and microbiology (Rad 2011), and oral mucosal changes (Bouquot 1992).
The potential benefits of smoking cessation can be observed at microbiological and immunological levels in the case of chronic periodontitis (Preshaw 2005). Thus, authors can report these effects at different time points of clinical measurement. Smoking cessation could be assessed at short-term (seven days), medium-term (6 to 12 months) and long-term (24 months) intervals (Ockene 2000). Thereby, we will consider level effects in function of the purpose of the studies and the reported time by authors.
Frequency of dental visits provides another relevant example. In the literature, a consensus has not been reached on the optimum periodicity for oral examination by the dentist, thus preventive dental care could be varied across time points. However, at least one contact within the past year is recommended (US Department of Health and Human Services 2000). As a result, we will consider the level effects at 6, 12 and 24 months in the case of studies on dental visits.
We will report adverse outcomes such as dental fluorosis that may result from prescription of fluoride products by dental healthcare providers (Fomon 2000). We will also report other outcomes, such as significant attention given to disadvantaged people (Evans 2003; Oliver 2008), induced by health promotion practices in dental healthcare settings.
Search methods for identification of studies
We will identify primary studies using the following bibliographic databases, sources and methods. We will search the Cochrane Database of Systematic Reviews, the Database of Abstracts of Reviews of Effectiveness (DARE), and the databases listed below in order to identify related systematic reviews. M. Fiander, Trials Search Co-ordinator of the EPOC Group wrote the MEDLINE search strategy in consultation with the authors (Appendix 1). M. Fiander will adapt the MEDLINE strategy for other databases.
We will search the following databases for primary studies.
- EPOC Group, Specialized Registers, Reference Manager.
- Cochrane Central Register of Controlled Trials (CENTRAL).
- MEDLINE, and MEDLINE In-Process and other non-indexed citations, OvidSP.
- EMBASE OvidSP.
- Cochrane Database of Systematic Reviews.
- CINAHL (Cumulative Index to Nursing and Allied Health Literature), EBSCOhost.
- African Index Medicus (AIM), Virtual Health Library (VHL).
- ASSIA - Applied Social Science Index and Abstracts , Proquest.
- Science Citation Index and Social Science Citation Index, ISI Web Knowledge.
- Dissertation and Theses Database, Proquest.
- ERIC - Education Resources Information Centre, Proquest.
- Global Health Database - CABI direct, Ovid.
- IBSS - International Bibliography of Social Sciences, Proquest.
- Database of Educational Research, Proquest.
- Pan American Health Organization database (PAHO), Virtual Health Library (VHL).
- World Health Organization Library Information System (WHOLIS), Virtual Health Library (VHL).
- NHS Economic Evaluation Database (NHS EED): www.crd.york.ac.uk/CRDWeb/AboutNHSEED.asp.
- Health Economic Evaluations Database (HEED): heed.wiley.com/ohe/autolog.asp.
- Cost-Effectiveness Analysis (CEA) Registry: https://research.tufts-nemc.org/cear4/
- Paediatric Economic Database Evaluation: pede.ccb.sickkids.ca/pede/index.jsp.
The MEDLINE search strategy combines the concepts of health promotion practices with dental healthcare settings and dental healthcare providers. Health promotion practices are a broad concept, consequently the approach supports the definition used in this review.
We have used the Cochrane Highly Sensitive Search Strategy (sensitivity- and precision-maximizing version - 2011 revision) (Lefebvre 2011), to identify randomized trials, and an EPOC methodology filter to identify NRCT designs.
Searching other resources
We will conduct a grey literature search to identify studies not indexed in the databases listed above. We will include the sites listed below (see Appendix 2). If we find any additional sources, we will document them in the review.
We will search the following Registries.
- International Clinical Trials Registry Platform (ICTRP), World Health Organization (WHO): www.who.int/ictrp/en/.
- ClinicalTrials.gov, US National Institutes of Health (NIH): clinicaltrials.gov/.
We will also:
- screen individual journals and conference proceedings (e.g. handsearching);
- review reference lists of all included studies, relevant systematic reviews, primary studies or other publications;
- contact authors of relevant studies or reviews to clarify reported published information or seek unpublished results/data;
- contact researchers with expertise relevant to the review topic or EPOC interventions; and
- conduct cited reference searches for all included studies in ISI Web of Science/Web of Knowledge.
We will perform handsearching in individual journals with a specific focus on dentistry and health promotion. We will search: Journal of Dental Research, Journal of Public Health Dentistry, Community Dental Health, British Dental Journal, Primary Dental Care, Swedish Dental Journal, Oral Health and Preventive Dentistry.
Data collection and analysis
Selection of studies
Two independent review authors will screen titles or abstracts, or both of search results in light of the inclusion and exclusion criteria. We will exclude irrelevant references. We will obtain copies of the full-text of potentially relevant references. When there are insufficient data in the title or abstract, or both to make a clear decision regarding eligibility of studies, we will obtain full reports. We will record this process in a PRISMA flow chart (Liberati 2009). Two review authors will independently assess the eligibility of retrieved papers. We will resolve any disagreements between review authors through discussion. We will remove studies not matching the inclusion criteria and we will note the reasons for their exclusion in the 'Characteristics of excluded studies' table.
Data extraction and management
We will develop and test a data extraction form based on the EPOC template. Two independent review authors will review each paper or extract, and report any relevant data. We will extract data from the studies as recommended in the EPOC data collection checklist (EPOC 2009). Where possible, we will note the underlying theory of the intervention and theoretical constructs, including contextual factors. For health economic studies, we will adapt the data extraction form from the format and guidelines used to produce structured abstracts of full economic evaluations for inclusion in the National Health Service Economic Evaluation Database (NHS EED) (Shemilt 2011b). We will resolve any disagreements between review authors through discussion. We will contact the authors of individual studies for clarification or missing data.
Assessment of risk of bias in included studies
We will investigate assessment of risk of bias based on EPOC guidelines. Each review author will grade the studies using a modified version of the 'Risk of bias' table that is designed to replace quality criteria in the EPOC data collection checklist (EPOC 2009). Each author will group the included studies into: (1) low risk of bias (studies with low risk of bias for all items, or with unclear risk of bias in some items but that has no impact on our confidence in the results); (2) moderate risk of bias (studies with unclear risk of bias in several items that has a potential impact on our confidence in the results) and (3) high risk of bias (studies with high risk of bias in one or more of the items that is likely to affect our confidence in the results. We will compare the evaluations, discuss and resolve any inconsistencies between review authors in the interpretation of inclusion criteria and their significance to the selected studies. We will resolve any disagreements in risk of bias ratings through discussion.
We will assess risk of bias according to the nature of the study reviewed and in accordance with The Cochrane Collaboration's 'Risk of bias' tool (Higgins 2011a): (a) random sequence generation; (b) allocation concealment; (c) blinding of participants and personnel; (d) blinding of outcome assessors; (e) incomplete outcome data; (f) selective outcome reporting and (g) other sources of bias (such as carry-over in cross-over designs or recruitment bias in NRCTs). For non-randomized studies, we will assess the risk of bias with a similar process used for randomized studies, although attention must be paid to the possibility of selection bias (Reeves 2011).
We will assess the following domains for studies with a separate control group (RCTs, NRCTs, controlled before-and-after studies).
- Was the allocation sequence adequately generated?
- Was the allocation adequately concealed?
- Were baseline outcome measurements similar?
- Were baseline characteristics similar?
- Were incomplete outcome data adequately addressed?
- Was knowledge of the allocated interventions adequately prevented during the study?
- Was the study adequately protected against contamination?
- Was the study free from selective outcome reporting?
- Was the study free from other risks of bias?
For ITS studies, we will assess the following domains.
- Was the intervention independent from other changes?
- Was the shape of the intervention effect pre-specified?
- Was the intervention unlikely to affect data collection?
- Was the knowledge of the allocated interventions adequately prevented during the study?
- Were incomplete outcome data adequately addressed?
- Was the study free from selective outcome reporting?
- Was the study free from other risk of bias?
For the cost or resources use outcomes, we will classify the included studies by study design. To assess their methodological quality, we will use the Consensus on Health Economic Criteria (CHEC-list), a 19-item assessment tool developed through a Delphi procedure (CCEMG 2012; Evers 2005). The CHEC-list addresses internal and external validity aspects of individual economic evaluation studies. Of these criteria, 18 relate to internal validity issues. Of these, five pertain to study design (the description of study population; description of the alternatives that are being compared; the research question; the research design and the time horizon. Eleven criteria relate to the conduct of the economic evaluation (selection and justification of the perspective for the analysis; identification, measurement and valuation of resource use (e.g. costs); identification, measurement and valuation of outcomes (e.g. consequences); adjustment for costs and outcomes occurring in the future (e.g. discounting); congruency between the presented data and conclusions; independence of the investigators; and a discussion of ethical and distributional issues. The remaining two criteria address the issue of analysis, that is, whether an incremental analysis of costs and outcomes and a sensitivity analysis were performed. The former refers to an analysis involving a joint comparison of the difference in costs and difference in outcomes between the two interventions and the latter to a testing of assumptions made in the main analysis. Finally, one criterion addresses external validity, namely, the generalizability of results. In addition, the assessments should consider the biases that have a notable impact on the results or conclusions of the trial, recognizing subjectivity involved (Higgins 2011a).
Measures of treatment effect
We will use Review Manager 5 to report quantitative data (RevMan 2012). For RCTs, we will analyze data by determining the unadjusted absolute difference and 95% confidence intervals, while for NRCTs and controlled before-and-after studies, we will determine the adjusted absolute difference between intervention and control groups and 95% confidence intervals for dichotomous data. We will determine the standardized mean difference (SMD) and 95% confidence intervals for continuous data. For ITS design studies, we will calculate the difference in slope and the difference in level of intervention. We will derive precise P values for all comparisons, where possible.
Unit of analysis issues
Comparisons that randomize or allocate clusters but do not account for clustering during analysis have potential unit of analysis errors resulting in artificially extreme P values and over-narrow confidence intervals (Ukoumunne 1999). We will attempt to re-analyze studies with potential unit of analysis errors where possible and include them within the meta-analysis. We will undertake appropriate corrections if possible, by contacting primary authors to obtain missing information. If data are not available within the correct unit of analysis, we will re-assess results and estimate a new P value.
If re-analysis is not possible, we will exclude the studies from the meta-analysis and report the effect size of the studies separately (without P values or confidence intervals) in tabular form.
For ITS design studies, we will abstract the difference in slope and the difference in level pre- to post-intervention (Ramsay 2003). We will analyze the post- versus pre-intervention difference at specific time points (three months, six months and yearly thereafter). For instance, if the clinical measure has been reported at six months, we will report the level effect at this same time.
If the differences are not available in the primary reports, we will attempt re-analysis using data from graphs or tables using time series regression techniques. We will estimate the best fit pre-intervention and post-intervention lines using linear regression and autocorrelation adjusted where appropriate (Draper 1981). We will test first-order autocorrelations statistically using the Durbin-Watson statistic and investigate higher-order autocorrelations using the autocorrelation and partial autocorrelation function (Draper 1981).
Dealing with missing data
We will contact authors of reviews if incomplete methodological and statistical information were mentioned or remarked on in studies. In the cases of unsuccessful attempts, we will provide a narrative synthesis of the data as reported. We will document the loss to follow-up information and assess it as a source of potential bias (EPOC 2009).
We will capture unobtainable methodological data in our risk of bias assessment of studies and tabulate them in our report. At the same time, we will assess and manage unobtainable statistical data following the guidance provided in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011b). We will conduct sensitivity analyses on our missing data management decisions and take into account the possible impact of missing data on our findings within the 'Discussion' section of our review.
Assessment of heterogeneity
We are expecting to find content and methodological diversity arising from different types of interventions, outcome measurement and study designs. Prior to data synthesis, we will categorize the studies by type (tailored, theory-based, professional, etc.), targets (policy, community, organizational, inter-/intra-personal), focus of interventions (oral health, general health, professional behaviour change, increasing access and equity of health care), setting (dental office, community based), and study design (RCT versus NRCT), controlling for potential sources of heterogeneity.
In addition, we will group together studies that appear to be methodologically homogenous (similar type, targets, and focus of intervention, research design, outcome measure and time to outcome). For each study grouping that has a sufficient number of studies to undertake a random-effects meta-analysis of the combined data according to the methods outlined in the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2011), we will produce forest plots, calculate the Chi
If meta-analysis is not possible due to substantial heterogeneity, we will present the results of included studies in a tabular form and undertake a qualitative analysis based upon the study quality, and the size and direction of effect observed. We will also present the statistical significance of the studies.
If heterogeneity exists in any of our study groupings, we will explore this heterogeneity by preparing bubble plots and tables that represent group comparisons relative to each of these variables in relationship to the size of the effect. We will investigate it further using subgroup analysis or meta-regression, or both. If an indication of substantial heterogeneity is determined, we will not produce a pooled estimate and will present a narrative summary of our findings.
Assessment of reporting biases
We plan to explore publication bias graphically by using funnels plots and the Eggers' test methods (Egger 1997), to formally test for the funnel plot asymmetry. However, asymmetric funnel plots are not necessarily caused by publication bias (Sterne 2004). For instance, inadequate methodological approaches in controlled trials, particularly in NRCTs could be associated with reporting bias (Egger 2003; Schulz 1995).
For each study, we will report the main results in natural units. We will perform a standard method of presentation of results for all types of comparisons. For RCTs, NRCTs, controlled before-and-after and ITS studies, we will report comparisons separately for each study design and depending on the nature of data (dichotomous or continuous) as follows.
- The number of comparisons showing a positive direction of effect.
- The number of comparisons showing statistically significant effects.
- The median effect size across all comparisons.
- The median effect size across all comparisons without unit of analysis errors.
In cases where there are insufficient data to calculate the mean effect sizes, we will report study results in tabular form and perform a qualitative assessment based upon the quality, the size and the direction of the effect observed and the statistical significance of the studies.
Where baseline results are available from RCTs, NRCTs, controlled before-and-after and ITS studies, we will present results as pre-interventions means, post-interventions means or proportions for both control and study groups. When the baseline data are not available, we will present the relative percentage change (the difference between post-intervention values in the study and control groups expressed as a percentage of post-intervention values in the control group).
We will analyze the included studies to investigate the similarity of several characteristics such as the level, type, focus and targets of interventions, study design, dental healthcare settings, and dental healthcare providers' and patients' characteristics. If studies are too heterogeneous that statistical combination is not possible, we will present a descriptive synthesis of included studies using a narrative summary in tables. In addition, we will also report the range of effects.
We will conduct the economic component of this review according to current guidance on the use of economic methods in the preparation and maintenance of Cochrane reviews (Shemilt 2011a). We will summarize characteristics and results of included economic evaluations using additional tables, supplemented by a narrative summary that will compare and evaluate the methods used and the principal results between studies. If unit cost data are available, we will also tabulate them. We will report the currency and price year that applies to measures of costs in each original study alongside measures of costs and incremental analysis of costs.
Subgroup analysis and investigation of heterogeneity
If there are enough studies, we will perform the following subgroup analyses.
- Type of intervention (tailored, theory-based, community-based, multiple health behaviour change versus single-behaviour interventions).
- Targets of intervention (policy, community, organizational, inter-/intra-personal).
- Focus of intervention (oral health, general health, professional change behaviour and others such as continuing professional development, increasing access and equity of health care).
- Study design.
- Main potential co-founders:
- subsets of studies (urban or rural settings, private or public settings);
- subsets of participants (e.g. dentists and no dentists, years in practice, gender, speciality, solo or group practice);
- subsets of patients (age, chronic disease, comorbidity, disability, smoker or not, disadvantaged people or not, etc.).
If possible, we will conduct meta-analyses within subgroups and meta-analyses that combine several subgroups using standard interaction tests.
We will conduct separate meta-analyses by type of outcomes (professional and patient outcomes). For outcomes measured at several times:
- we will first attempt to obtain individual data and perform an analysis that uses the follow-up for each participant. If this is not available, we will compute an effect for each individual participant that incorporates all times, such as an overall mean or median;
- we will define several different outcomes based on different periods of follow-up, and perform separate analyses if individual data are not available. For instance, time frames might be defined to reflect long-term, medium-term and short-term follow-up; and
- we will perform data analysis at a single time point that we will select. In this case, we will only consider studies in which it is presented.
In the meta-analysis, the rationale for pooling different types of interventions is usually performed to determine the impact of heterogeneity among interventions rather than to estimate the effects of homogeneous interventions (Kawamoto 2005).
If adequate data are available, we will perform sensitivity analyses to assess the effects of study quality (excluding studies identified as being of low methodological quality) on effect size and to examine the findings in light of study characteristics and decisions made during the review. Thus, we will assess the impact of:
- exclusion of studies with a high risk of bias or trials randomized by cluster;
- change review data processing decisions such as dealing with missing data; and
- exclusion of unpublished studies.
We will use a tabular form to present the results.
The review authors would like to thank Christina Godfrey, Margaret Harrison and Ian Graham for their helpful suggestions in the review preparation and the reading of the draft version of the protocol; Stéphane Turcotte for statistical advice and Michelle Fiander for the search strategy and the search results. We thank Craig Ramsay, Luke Vale, Tomas Pantoja and Lisa Bero as Effective Practice and Organisation of Care (EPOC) editors, Clare Dooley as Managing Editor and also Lucy O'Malley and Paul Brocklehurst for their peer review comments.
Appendix 1. MEDLINE strategy
Database: Ovid MEDLINE(R) In-Process and Other Non-Indexed Citations and Ovid MEDLINE(R) <1946 to Present>
1 Public Health Dentistry/ or Community Dentistry/ (3466)
2 ((dentist$ or dental) and (health adj2 (general or public or promot$))).ti. (978)
3 ((dentist$ or dental) adj4 (early intervention? or health promotion$ or early diagnos$)).ab. (143)
4 or/1-3 [Public Health Dentistry] (3978)
5 exp Dentists/ or exp Dental Auxiliaries/ or dental Staff/ or "dental staff, hospital"/ or Dentist practice patterns/ (27,104)
6 exp Dental Care/ or exp Preventive Dentistry/ (48,766)
7 Dental clinics/ or Dental offices/ (5287)
8 (dentist? or dentistry or hygienist?).ti. or dentist?.ab. (43,620)
9 ((dental or dentistry) adj2 (assistant? or auxiliar$ or clinic? or clinician? or doctor? or hygienist? or nurse? or office? or personnel or practice? or practitioner? or professional? or provider? or service or services or staff or surgeon? or technician? or technologist? or therapist?)).ti,ab. (26,404)
10 General Practice, Dental/ or partnership practice, dental/ or Group Practice, Dental/ or Practice management, dental/ (14,567)
11 Dental Health Services/ (3305)
12 "education, dental, continuing"/ (2996)
13 (dentist$ adj4 (continuing adj2 education$)).ab. or (dentist$ and ((continuing adj2 education$) or CME)).ti. (246)
14 Dentist-Patient Relations/ (7180)
15 Dentistry/ or exp Specialties, Dental/ or Economics, Dental/ or Dental Facilities/ or Dental Audit/ (57,013)
16 or/5-15 [Dentists, Dentistry] (161,415)
17 exp Health Promotion/ (46,909)
18 Public Health/ or Preventive Medicine/ or Primary Prevention/ or "Early Intervention (Education)"/ (77,525)
19 exp Preventive Health Services/ or Early Medical Intervention/ or Consumer Health Information/ or Health Literacy/ or Health Fairs/ or Patient Education as Topic/ (388,517)
20 Early Diagnosis/ or "Early Detection of Cancer"/ (13725)
21 exp mouth neoplasms/di, pc or gingival neoplasms/di, pc or exp leukoplakia, oral/di, pc or lip neoplasms/di, pc or palatal neoplasms/di, pc or exp salivary gland neoplasms/di, pc or tongue neoplasms/di, pc (8273)
22 (exp mouth neoplasms/ or gingival neoplasms/ or exp leukoplakia, oral/ or lip neoplasms/ or palatal neoplasms/ or exp salivary gland neoplasms/ or tongue neoplasms/) and (diagnos$.ti,hw. or detect$.ti.) (7217)
23 "Tobacco use cessation"/ or Smoking cessation/ (17,888)
24 Smoking/ or Alcohol Drinking/pc (110,892)
25 Mass screening/ or Anonymous Testing/ or Multiphasic Screening/ or "Diagnostic Tests, Routine"/ (81,676)
26 screen$.ti. or (testing adj2 (genetic or rapid or anonymous or HIV or diabet$)).ti,ab. (117,633)
27 (counsel$ or advise or advice or advising).ti. (19,229)
28 (health$ adj2 promo$).ti,ab. (26,372)
29 (primary prevention$ or prevent$ medicine or preventive practice? or (disease? adj2 prevention$)).ti,ab. (25,180)
30 (public health or consumer? health or (health adj2 communit$) or ((preventative or preventive or program$) adj2 health$) or ((community or consumer? or patient?) adj2 (educat$ or knowledg$ or awareness)) or health literacy or health fair?).ti,ab. (176,982)
31 ((diabetic? or patient? or smoking or smoker? or tobacco or (weight adj2 (reduc$ or loss or management?))) adj5 (advice or advis$ or counselling or counsel? or program or programme? or referral or referring or screen$)).ti,ab. (74,716)
32 ((assess$ or monitor$) adj3 (blood pressur$ or chronic or diabet$ or (general adj2 health) or glucose or obesit$ or obese$ or overweight or sugar or (weight adj2 (gain? or loss or losses or increas$)))).ti,ab. (27,145)
33 (early adj2 (detection or diagnos$)).ti,ab. (90,048)
34 (((tobacco or smoking) adj3 (cessation or program$ or quit$ or detect$)) or ((alcholic? or alcoholism or alcohol$ abuse or excess$ drink$ or heavy drink$) and (detect$ or assess$ or counsel$ or advise? or advice))).ti,ab. (26,781)
35 ((cancer? or chronic or condition? or early hypertension or hyper-tension or hypertensive or hyper-tensive or hypoglyc?em$ or hypo-glyc?em$ or hyperglyc?em$ or hyper-glyc?em$ or neoplasM?) adj3 (detect$ or diagnos$)).ti,ab. (93,072)
36 Family Health/ or general practice/ or family practice/ or general practitioners/ or physicians, family/ or physicians, primary care/ or primary health care/ or "continuity of patient care"/ or patient-centered care/ (151,240)
37 ((primary adj2 (care or healthcare)) or (family health or healthy people or (healthy adj2 program$) or (family adj2 (doctor? or physician? or practitioner? or practice)))).ti,ab. (100678)
38 or/17-37 [Health Promotion & related concepts] (1,170,335)
39 Professional role/ (6687)
40 (role? adj2 (new or chang$)).ti,ab. (10,360)
41 practice pattern?.ti,ab,hw. or ((dentist? or provider? or practitioner? or physician?) adj3 (behavio?r? or pattern?)).ti,ab. (42,890)
42 ((service? or service level? or skill?) adj3 (mix$ or combine? or combination?)).ti,ab. (1728)
43 (reskill$ or new skill? or upskill$ or "up-skill$" or ((advanc$ or develop$) adj2 skill?)).ti,ab. (5476)
44 ((service or services) adj2 (new or newly or chang$ or improv$)).ti,ab. (7776)
45 ((chang$ or exchang$ or expand$ or extend$ or increase? or reduc$ or replac$ or limit? or focus$ or redistribut$) adj3 (task? or job? or work? or responsib$ or role? or duty or duties or job function? or service? or care)).ti,ab. (106,052)
46 ((chang$ or design? or organi?e? or creat$) adj2 (program? or programme?)).ti,ab. (6031)
47 or/39-46 [Professional Role/Services] (178,072)
48 (randomized controlled trial or controlled clinical trial).pt. or randomized.ab. or placebo.ab. or clinical trials as topic.sh. or randomly.ab. or trial.ti. (792,382)
49 exp animals/ not humans.sh. (3,736,637)
50 48 not 49 [Cochrane RCT Filter 6.4.d Sens/Precision Maximizing] (732,542)
51 intervention?.ti. or (intervention? adj6 (clinician? or collaborat$ or community or complex or DESIGN$ or doctor? or educational or family doctor? or family physician? or family practitioner? or financial or GP or general practice? or hospital? or impact? or improv$ or individuali?e? or individuali?ing or interdisciplin$ or multicomponent or multi-component or multidisciplin$ or multi-disciplin$ or multifacet$ or multi-facet$ or multimodal$ or multi-modal$ or personali?e? or personali?ing or pharmacies or pharmacist? or pharmacy or physician? or practitioner? or prescrib$ or prescription? or primary care or professional$ or provider? or regulatory or regulatory or tailor$ or target$ or team$ or usual care)).ab. (129,583)
52 (pre-intervention? or preintervention? or "pre intervention?" or post-intervention? or postintervention? or "post intervention?").ti,ab. [added 2.4] (7458)
53 (hospital$ or patient?).hw. and (study or studies or care or health$ or practitioner? or provider? or physician? or nurse? or nursing or doctor?).ti,hw. (652,486)
54 demonstration project?.ti,ab. (1765)
55 (pre-post or "pre test$" or pretest$ or posttest$ or "post test$" or (pre adj5 post)).ti,ab. (53,415)
56 (pre-workshop or post-workshop or (before adj3 workshop) or (after adj3 workshop)).ti,ab. (482)
57 trial.ti. or ((study adj3 aim?) or "our study").ab. (504,720)
58 (before adj10 (after or during)).ti,ab. (317,984)
59 ("quasi-experiment$" or quasiexperiment$ or "quasi random$" or quasirandom$ or "quasi control$" or quasicontrol$ or ((quasi$ or experimental) adj3 (method$ or study or trial or design$))).ti,ab,hw. [ML] (88,635)
60 ("time series" adj2 interrupt$).ti,ab,hw. [ML] (720)
61 (time points adj3 (over or multiple or three or four or five or six or seven or eight or nine or ten or eleven or twelve or month$ or hour? or day? or "more than")).ab. (7024)
62 pilot.ti. (32,601)
63 Pilot projects/ [ML] (71,559)
64 (clinical trial or controlled clinical trial or multicenter study).pt. [ML] (582,140)
65 (multicentre or multicenter or multi-centre or multi-center).ti. (24,259)
66 random$.ti,ab. or controlled.ti. (643,301)
67 (control adj3 (area or cohort? or compare? or condition or design or group? or intervention? or participant? or study)).ab. not (controlled clinical trial or randomized controlled trial).pt. [ML] (349,997)
68 "comment on".cm. or review.ti,pt. or randomized controlled trial.pt. [ML] (2,618,576)
69 review.ti. [EM] (219,847)
70 (rat or rats or cow or cows or chicken? or horse or horses or mice or mouse or bovine or animal?).ti. (1,265,081)
71 exp animals/ not humans.sh. [ML] (3,736,637)
72 (animal$ not human$).sh,hw. [EM] (3,642,720)
73 *experimental design/ or *pilot study/ or quasi experimental study/ [EM] (18,154)
74 ("quasi-experiment$" or quasiexperiment$ or "quasi random$" or quasirandom$ or "quasi control$" or quasicontrol$ or ((quasi$ or experimental) adj3 (method$ or study or trial or design$))).ti,ab. [EM] (88,635)
75 ("time series" adj2 interrupt$).ti,ab. [EM] (720)
76 (or/51-67) not (or/68,70-71) [EPOC Methods Filter ML 2.4] (1,851,189)
77 (or/51-58,61-62,65-66,73-75) not (or/69,72) [EPOC Methods Filter EM 1.9-2.4] (1,873,176)
78 4 and 50 [Public Hlth Dentistry & RCT] (79)
79 (and/16,38,50) not 78 [Dentists/Dentistry & Hlth Promo & RCT] (832)
80 (and/16,38,76) not (or/78-79) [Dentists/Dentistry & Hlth Promo & EPOC Filter] (4246)
81 (and/16,47,50) not (or/78-80) [Dentists/Dentistry & Role Change & RCT] (141)
82 (and/16,47,76) not (or/78-81) [Dentists/Dentistry & Role Change & EPOC Filter] (696)
83 (and/4,76) not (or/78-82) [Public Hlth Dentistry & EPOC] (216)
Appendix 2. Grey literature
- Trials Register of Promoting Health (TRoPHI) eppi.ioe.ac.uk/webdatabases/Intro.aspx?ID=5
- Open Grey www.opengrey.eu/
- Grey Literature Report (New York Academy of Medicine) www.nyam.org/library/online-resources/grey-literature-report/
- Agency for Healthcare Research and Quality (AHRQ) www.ahrq.gov/
- Joanna Briggs Institute www.joannabriggs.edu.au/Search.aspx
- National Institute for Health and Care Excellence (NICE) www.nice.org.uk/
- Campbell Library www.campbellcollaboration.org/c2_systematic_reviews/index.php
- CISMEF (Catalog and Index of medical sites in French Language) www.chu-rouen.fr/ssf/dent/dentisterie.html
- Word Dental Federation (WDF) www.fdiworldental.org/
- Center of evidence based dentistry (CEBD) ebd.ada.org/
- American Dental Association (ADA) www.ada.org/
- American Dental Education Association (ADEA) www.adea.org/Pages/default.aspx
- Canadian Dental Association (CDA) www.cda-adc.ca/
Contributions of authors
Pascaline Kengne Talla led the writing of the protocol with contributions from Marie-Pierre Gagnon.
Sophie Dupéré, Aimée Dawson, France Légaré and Christophe Bedos reviewed the draft version, proposed some relevant studies and approved this protocol version.
Declarations of interest
Pascaline Kengne Talla: none known.
Marie-Pierre Gagnon: none known.
Sophie Dupéré: none known.
Christophe Bedos: none known.
France Légaré: none known.
Aimée Brennan Dawson: none known.
Sources of support
- None provided, Not specified.
- None provided, Not specified.