Scolaris Content Display Scolaris Content Display

Base de datos Cochrane de Revisiones Sistemáticas Protocolo - Intervención

Oral hygiene programmes for people with intellectual disabilities

Declaraciones de intereses de los autores

Versión publicada: 06 abril 2017 Historial de versiones

https://doi.org/10.1002/14651858.CD012628

Esta versión no es la más reciente

ver la versión actual 31 mayo 2019
Contraer todo Desplegar todo

Abstract

This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:

To assess the effects (benefits and harms) of oral hygiene interventions for people with intellectual disabilities (ID).

Background

Description of the condition

Dental plaque has been shown to have a major role in the development of periodontal disease and dental caries, the two most common oral diseases. Oral hygiene is the process of cleaning the hard and soft tissues of the oral cavity (teeth, gums, and tongue) and fixed dental prostheses, oral appliances, and dentures. Failure to perform regular and effective oral hygiene will result in increased levels and varieties of the dental bacteria that make up dental plaque.

Poor oral hygiene can affect a person’s quality of life because of pain and discomfort when eating, poor self esteem, or sleep disturbance as a result of dental caries (Bonetti 2015). Evidence also exists linking levels of oral hygiene and periodontal status to diabetes, aspiration pneumonia, and rheumatoid arthritis, and evidence is emerging of a link with other systemic diseases and conditions such as cardiovascular disease, liver disease, and obesity (Borgnakke 2015; Horwitz 2000).

People with an intellectual disability (ID) have a higher prevalence and greater severity of periodontal disease than the general population (Anders 2010; Campanaro 2014). The prevalence of dental caries in children with ID is similar to those of the general population (Whelton 2009). However, their oral health deteriorates at a faster rate as they move into adulthood. There are more untreated dental caries, more missing teeth, and fewer restorations found in adults with ID than in the general population (Anders 2010; Catteau 2011; Crowley 2005; De Jongh 2008; Hennequin 2008; Morgan 2012; Oliveira 2013).

In the general population, brushing regularly with fluoridated toothpaste has been shown to play an important role in preventing dental caries and periodontal disease (Axelsson 2004; Broadbent 2011; Chestnutt 1998; Löe 1965; Löe 2000; Van der Weijden 2011; Zimmermann 2015). Whether this also applies to people with an ID is less clear, as many research studies on oral health exclude people with disabilities (Glassman 2009).

The removal of dental plaque can reduce the prevalence of periodontal disease and dental caries (Axelsson 1978; Axelsson 2004; Löe 1965; Löe 2000). However, the plaque must be removed thoroughly and regularly to achieve this effect. This health‐related behaviour requires an individual to have the relevant capabilities, opportunities, and motivation to perform the necessary routines (Broadbent 2011; Stewart 1996; Watt 2005). For people with an ID, understanding the importance of removing dental plaque and developing the skills to do so requires more effort, and achieving the standard of oral hygiene necessary for good oral health may require the support of others (Crowley 2005).

When considering the issues possibly influencing oral hygiene practices for the population with ID, it is important first to have a clear definition of ID; second to consider the specific barriers to regular oral hygiene practices that might exist for people with ID; and third to consider facilitators to promoting regular oral hygiene practices that might exist for people with ID.

Definition of intellectual disability

The current definition of disability comes from the World Health Organization's (WHO) International Classification of Functioning, Disability and Health (ICF), which incorporates the complex interactions between health conditions, environmental factors, and personal factors. Regarding a person with an ID, this definition would consider how their personal factors, health condition, and environment affect their lives (WHO 2001). Three elements are common for people with ID: a significant impairment of intelligence; a resultant significant reduction in adaptive behaviour/social functioning; and the development of the condition before the age of 18, which persists throughout life (Schalock 2010).

The WHO has also developed an operational definition of ID, which focuses mostly on the functional elements of ID, in the International Statistical Classification of Diseases and Related Health Problems, 10th version (ICD‐10) (WHO 2011). The ICF definition of disability is designed to complement the ICD‐10 definition.

The ICD‐10 describes the four levels of ID as follows:

Mild intellectual disability

(ICD‐10 F70): IQ range 50 to 69

“Ability to use speech in everyday situations; usually full independence in self care; Difficulties in identification of this population arise, as those with borderline disability may not be in contact with service providers.”

Moderate intellectual disability

(ICD‐10 F71): IQ between 35 and 49

“Slow in comprehension; supervision of self care, retarded motor skills.”

Severe intellectual disability

(ICD‐10 F72): IQ between 20 and 34

“Marked impairment of motor skills; clinically significant damage to Central Nervous System (CNS).”

Profound intellectual disability

(ICD‐10 F73): IQ less than 20

“Severely limited understanding; Immobility or restricted mobility; Incontinence; requires constant supervision; usually organic aetiology.”

We acknowledge that other terms and definitions have been used to describe people with ID, both historically and currently. Whilst we acknowledge that the ICF is currently accepted as the most holistic definition of disability, it is unlikely to be used in the studies that are potentially eligible for this review. We therefore plan to use the ICD‐10 definition as the reference to determine the inclusion or exclusion of studies, as it provides a detailed nomenclature of diseases (Kurbasic 2008).

Barriers to oral hygiene practice for people with ID

Capability, opportunity, and motivation are required to achieve adequate oral hygiene. Oral hygiene tasks are more difficult for people with ID; they may lack the manual dexterity or cognitive skills to carry out or understand the need for effective oral hygiene (Anders 2010; Löe 2000). There may be other commonly associated conditions such as sensory impairment, anxiety, or chronic medical or behavioural conditions that militate against carrying out oral hygiene effectively (Gordon 1998; Minihan 2014; Prangnell 2008). In a 2003 Irish study, carers of people with ID identified some of the physical barriers caused by the disability such as an inability to rinse, difficulty getting a toothbrush into someone's mouth, an oversensitive mouth, and a tendency to gag (Taylor‐Dillon 2003). Nearly 10% of people with ID have dysphagia, which is associated with increased tendency to choke or gag during toothbrushing (Chadwick 2009).

People with ID have been shown to have chronic, poor oral hygiene and to require more support with their oral hygiene. Carers for people with ID are often poorly trained (Crowley 2005). Oral hygiene is not seen as a priority for overstretched carers, particularly if it is not a personal priority (Kaye 2005; Rawlinson 2001). Lack of co‐operation and challenging behaviour on the part of the person with ID may add to this burden.

Facilitators for oral hygiene practice for people with ID

Behaviour management techniques may improve the co‐operation of individuals with ID when performing oral hygiene. Education may change carers’ attitudes and improve their skills (Faulks 2000; Mac Giolla Phadraig 2013). Management support and guidance, stakeholder engagement, policy documents, and increased resources may improve levels of oral hygiene provision in community and institutionalised settings (Glassman 2003; Glassman 2006; Glassman 2009). Specially designed or adapted oral hygiene aids may improve the oral hygiene skills of people with ID or their carers (Dougall 2008).

Description of the intervention

The interventions to be evaluated in this Cochrane Review will focus on two distinct groups: people with an ID, and the carers supporting, assisting, or providing oral hygiene care for them. The range of carers involved is likely to be diverse, that is formal and non‐formal personal carers with and without relevant training, with and without dental qualifications, and with and without psychology or related qualifications.

We will define oral hygiene in this review as the mechanical removal of plaque with either manual or electric toothbrushes, interdental aids, or other mechanical aids. Interventions designed to impact on levels of oral hygiene tend to group outcomes into clinical outcomes such as reductions in plaque levels, periodontal diseases, or dental caries, and psychological outcomes such as changes in comprehension, self efficacy, or motivation (Cooper 2013; Cooper 2014). This review will include interventions that measure clinical and behavioural/psychological outcomes, or both. We will consider any intervention that may impact on the duration, frequency, and/or effectiveness of oral hygiene practices.

How the intervention might work

Behavioural change models have been shown to relate to oral hygiene behaviour and to be effective when used in interventions for people with ID (Renz 2007; Willems 2017). In a systematic review, Newton 2015 looked at using the constructs of changing behaviour rather than the models themselves and found this to be a more reliable way of predicting behavioural change. Michie 2013 has identified a taxonomy of 93 behavioural change techniques (BCTs) that provides a standardisation of the terminology used. In order to make the taxonomy of BCTs more accessible for clinicians, the BCTs can be grouped into 16 labeled clusters (Cane 2015; Michie 2011; Michie 2013). These labels provide a means to identifying the elements in an intervention that are likely to improve adherence and make routines easier to establish in relation to oral hygiene, even if they are not formally identified as BCTs. To further assist the analysis of interventions, Michie 2011 developed the Behavioural Change Wheel, which characterises the elements at three levels: a “behaviour system” at the hub, encircled by intervention functions, and then by policy categories.

We will consider interventions under the “behaviour system” elements of Capability (which includes knowledge and skills), Opportunity, and Motivation, as outlined by Michie as the COM‐B framework for understanding behaviour and behavioural change (Michie 2011).

Capability‐based interventions, defined as the psychological and physical capacity to engage in the activity concerned, might be a knowledge‐based intervention in relation to the comprehension and reasoning around why oral hygiene is required or a skills‐based intervention that trains a person with ID in the physical skills of toothbrushing or that trains carers in how to brush the teeth of a person with ID.

Opportunity‐based interventions, which include both physical and social opportunities, should consider the elements beyond the individual, that is contexts such as the variety of settings in which the intervention is delivered or undertaken, for example home, school, dental clinic, day care, or residential care setting, or the influences of the families or carers' attitudes regarding oral hygiene practices.

Motivation‐based interventions, which include both automatic and reflective processes, might result in conscious decision‐making where a goal or reward is identified to instigate or perpetuate a daily oral hygiene routine.

An intervention may change one or more of these elements resulting in the desired behaviour of performing regular oral hygiene.

Why it is important to do this review

The difficulties of providing dental care for those with disabilities have been acknowledged (Al Kindi 2016). Access to care is known to be more difficult; resources are stretched for parents, carers, and healthcare professionals (Dougall 2008; Kaye 2005; Prabhu 2010; Sagheri 2013). The consequences of oral diseases for people with ID are more profound. Interventions designed to suit people with disabilities must show benefit (WHO 2011). Whilst an increasing number of good‐quality systematic reviews of oral health interventions for the general population have been produced (Agnihotry 2016), Cochrane has not specifically reviewed the effectiveness of oral health interventions for people with disabilities (Cochrane 2016).

A recent scoping review of interventions aimed at the wider disability population has identified those with ID as being the most common target subpopulation (Waldron 2016). Oral hygiene status is one of the most common targets of oral health intervention for those with disabilities. Our systematic review of oral hygiene interventions for people with ID will allow us to draw conclusions about the effectiveness of these types of interventions. Furthermore, the review may increase dental professionals' understanding of behavioural change principles, inform the training of the dental team, and have an impact on the design and implementation of future oral health interventions for this population.

Cochrane Oral Health undertook an extensive prioritisation exercise in 2014 to identify a core portfolio of titles that were the most clinically important ones to maintain on the Cochrane Library (Worthington 2015). This exercise identified prevention and maintenance as priorities, which included oral hygiene adherence and caries prevention. Whilst not specifically identifying people with disabilities, the exercise also prioritised oral health promotion, specifically in institutionalised settings (Worthington 2015).

Objectives

To assess the effects (benefits and harms) of oral hygiene interventions for people with intellectual disabilities (ID).

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs), including cluster‐RCTs, evaluating interventions targeted at groups such as parents, carers, and other healthcare professionals caring for people with ID.

As the number of RCTs is likely to be low, we will include some types of non‐randomised studies (NRS); however, we will report RCTs and NRS separately. The types of NRS eligible for this review will include: non‐randomised controlled trials, controlled before‐after studies, interrupted time series studies, and repeated measures studies.

We will apply no language restrictions.

Types of participants

People with an ID, living either at home, in day care, in hospital, or in a residential care setting. We will consider all intellectual disabilities as defined by the ICD‐10.

Types of interventions

We will consider all oral hygiene interventions, which may include but are not limited to the following components.

  1. Oral hygiene‐related knowledge‐based interventions for participants or their carers, e.g. education in the care of the teeth and gums, diseases of the teeth and gums, and prevention of diseases of the teeth and gums.

  2. Oral hygiene‐related behavioural interventions for participants or their carers that result in a behavioural change that can be observed and is replicable, e.g. goal setting or positive reinforcement.

  3. Oral hygiene‐related skills training for participants, e.g. interventions related to toothbrushing skills or other oral hygiene routines.

  4. Oral hygiene‐related skills training for carers, e.g. interventions related to assessing toothbrushing skills or assisting participants with toothbrushing or other oral hygiene routines, or providing oral hygiene care to participants.

To be eligible, controlled studies must include a comparison with at least one of the following: usual care, no care, or a similar alternative intervention. Examples of studies comparing similar alternative interventions might be a study comparing education provided in different formats, for example, face to face versus an information leaflet, or a study comparing oral hygiene care provided after a training session versus oral hygiene care provided without training.

Types of outcome measures

Outcome measures will include both behavioural and clinical measures.

We will base the behavioural outcomes measured on the Capabilities, Opportunities, and Motivation elements of COM‐B method and include indicators such as improved toothbrushing skills, changes in knowledge of participants and carers with regard to oral health, and increased episodes and duration of toothbrushing. We will accept any reasonable instrument used in the included studies such as questionnaires, interviews, observational outcomes, video recordings, diaries, self reported outcomes, or observational measures such as measuring toothpaste weight to check adherence with toothbrushing.

The clinical outcomes measured will include indicators such as the prevalence/incidence of dental plaque, prevalence/incidence of periodontitis (gum disease), and prevalence/incidence of dental caries. We will use only published or validated assessments of the clinical outcomes or those using a measure that can be compared to one of these measurements. For example, dental plaque may be measured using the Plaque Index, in Silness 1964, or Quigley and Hein's Plaque Index as modified by Turesky and colleagues (Turesky 1970). Gingivitis may be measured using an index such as the Gingival Index, in Löe 1963 and Löe 1967, or the Modified Gingival Index (Lobene 1986). Dental caries may be measured using the Decayed, Missing or Filled Teeth/Surface Index (dmft/s; DMFT/s) (Petersen 2013).

We will consider all time frames for the delivery and follow‐up of the intervention. The follow‐up time frames may be grouped for analysis into short, medium, and long term, based on the outcomes being measured. A systematic review of the effects of oral hygiene on chronic periodontitis in the general population excluded studies lasting less than nine months, in order to differentiate between gingivitis and periodontitis. Of the three RCTs included in that review, all were followed up for over three years (Hujoel 2005). Mombelli 1998 reflected on the limitations of longitudinal studies of periodontal disease when considering the issues of measurement error in periodontal probing and the impact of concepts such as bursts of disease activity versus a continuous disease process in relation to the progression of periodontal disease. Mombelli 1998 concluded that studies of 20 years or more in length would be required to achieve sufficient confidence in the results (Mombelli 1998). A recent systematic review on the effects of toothbrushing frequency on the prevention of dental caries showed that the eligible interventions were all followed up for between 11 months and 15 years (Kumar 2016). The following division of time frames for this current review was guided by a review of the literature on the effects of oral hygiene practices on oral health.

  • Short‐term follow‐up (< 6 weeks) may allow for some clinically visible signs of early changes in gingival health to become apparent, permitting the impact of possible differences in the microbial load, immune responses, and environmental factors in this population group to manifest in clinically observable signs (Garmyn 1998). It may also be useful to observe the clinical changes resulting from an intervention without confounding the result with other variables, e.g. the effect of oral hygiene following skills training without the variable of adherence to oral hygiene practices over time (Egelberg 1994). The early, clinically visible signs would include changes in plaque prevalence/incidence and changes in gingival bleeding or inflammation (Löe 1965).

  • Medium‐term follow‐up (6 weeks to 12 months) may allow for some additional signs of changes in the gingival health to be recorded compared to the baseline status, depending on the success of the intervention, such as gingival pocket probing depths, as well as consistent changes in knowledge and behaviour.

  • Long‐term follow‐up (> 12 months) would allow some for assessment of those measures that require a longer time frame to show change, such as dental caries prevalence/incidence or clinical attachment loss or bone loss in relation to gingival disease, as well as continued follow‐up on all other measurements. Long‐term studies would also allow for follow‐up on any short‐ or medium‐term success regarding changes in oral hygiene behaviour that is crucial to maintaining oral health (Schou 1998). It must be accepted that any findings from these long‐term interventions will be confounded by factors outside the study setting.

Primary outcomes

  1. Gingival health of people with an ID.

Secondary outcomes

  1. Oral hygiene skills acquisition by people with an ID, or the acquisition of skills by their carers in providing or supporting oral hygiene care.

  2. Oral hygiene knowledge acquisition by people with an ID or the acquisition of knowledge by their carers with regard to oral hygiene.

  3. Adherence to oral hygiene routines by people with an ID or their carers.

  4. Quality of life changes or long‐term, patient‐centred outcomes for people with an ID (e.g. tooth retention).

  5. Dental caries levels of people with an ID.

  6. Any unintended (positive or negative) effects of the interventions.

Outcome measures relating to clinical outcome must be published or validated or comparable to measurements such as dental plaque levels (i.e. Plaque Score), gingival health (i.e. Gingival Index), or dental caries prevalence/incidence (i.e. DMFT/S). Behavioural outcome assessments with regard to changes in capability, opportunity, or motivation of the patient or carer for oral hygiene practices will include any reasonable instrument used in the included studies (i.e. questionnaires or interviews).

Search methods for identification of studies

Cochrane Oral Health’s Information Specialist will conduct systematic searches for randomised controlled trials and controlled clinical trials. Because of the Cochrane Embase Project to identify all clinical trials on the database and add them to the Cochrane Central Register of Controlled Trials (CENTRAL), we will search only recent months of the Embase database. Please see the searching page on the Cochrane Oral Health website for more information. We will place no restrictions on the language or date of publication when searching the electronic databases.

Electronic searches

We will search the following electronic databases:

  • Cochrane Oral Health’s Trials Register;

  • Cochrane Central Register of Controlled Trials (CENTRAL) in the Cochrane Library;

  • MEDLINE Ovid (from 1946 onwards);

  • Embase Ovid (previous six months to date);

  • PsycINFO Ovid (from 1806 onwards).

We will model the subject strategies for databases on the search strategy designed for MEDLINE Ovid, shown in Appendix 1.

Searching other resources

We will search the following trials registries:

We will also handsearch the specialist conference abstracts from the International Association of Disability and Oral Health (2000 to 2016).

We will check the bibliographies of included studies and any relevant systematic reviews identified for further references to relevant trials (Horsley 2011).

We will consider any unpublished studies discovered using the search methods described above if they meet the inclusion criteria for this review.

We will consider adverse effects described in included studies only.

Data collection and analysis

Selection of studies

Five review authors will independently examine studies identified by the searches and ensure that each study is examined by at least two review authors, firstly by the titles and abstracts and then by obtaining and reading the full text if the study is considered potentially relevant. We will use specific inclusion criteria in our decision‐making including the participants and interventions as outlined below. A sixth review author will be the arbiter.

Inclusion criteria:

  • Participants must have an intellectual disability as defined by the ICD‐10, or be a carer of a person with an intellectual disability.

  • The intervention must relate to oral hygiene.

  • Complex interventions with an oral hygiene element will be considered if the outcome from the oral hygiene component can be appraised separately.

Details such as keywords, authors, publication, and institutions will be available to the review authors. Any disagreements will be discussed by the review authors and an arbiter. We will record justifications for all decisions. If we identify multiple reports of the same study, we will link these together in so far that this is possible. We will contact study authors to clarify details regarding eligibility where necessary. We will attempt to identify all relevant studies irrespective of language. Relevant non‐English language papers will be translated. Any studies identified by the searches, involving any of the review authors, will be examined by other review authors.

Data extraction and management

Two review authors will independently extract data using a standardised data collection sheet and enter the data into Review Manager 5 software (RevMan 2014). We will extract data under the following headings: profile of the participants, number of participants, settings, details of interventions including details of the carers and/or personnel delivering or supporting the intervention and any behavioural change techniques identified within the interventions, outcomes measured, duration of intervention and follow‐up, and study design.

Assessment of risk of bias in included studies

Two review authors will independently assess the risk of bias for each study, discussing any disagreements with the arbiter. We will record justifications for decisions. To avoid any conflict of interest, we will exclude team members who are authors of included studies from the 'Risk of bias' assessment of those studies, and alternative review authors will undertake this assessment.

We will consider bias for RCTs for the following domains: selection bias, performance bias, detection bias, attrition bias, reporting bias, and other bias, as per the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011a). We will consider bias for cluster‐RCTs under the two different units of measurement, the cluster and the participant, if provided and under for the same domains as the RCTs (Campbell 2012).

We will consider bias for non‐randomised controlled trials using the Risk Of Bias In Non‐randomised Studies ‐ of Interventions (ROBINS‐I) assessment tool, which has similar but slightly broader domains and ratings than the 'Risk of bias' tool for RCTs or the Effective Practice and Organisation of Care (EPOC) criteria for assessing risk of bias (EPOC 2016; Sterne 2016). We will rate each study as either high risk, low risk, or unclear risk of bias for the separate domains. We will report objective and subjective outcomes separately. We will complete a 'Risk of bias' table for each included study and will also present results graphically.

Measures of treatment effect

For continuous data, we will use the mean difference (MD) when the same scale is used or standardised mean difference (SMD) if more than one scale is used to measure the same outcome, along with the 95% confidence interval (CI). An example of likely continuous data is plaque levels as measured by plaque indices between the baseline and all other measured time frames, and between the control and intervention groups if a control was used. Another example is a self efficacy scale for carers measured before and after an intervention. We will use the risk ratio (RR) and its 95% CI for dichotomous data, for example toothbrushing undertaken or not, before or after the intervention.

Unit of analysis issues

We will consider the unit of analysis with regard to whether the intervention was targeted at an individual or at a group level, if there was more than one element to the intervention, or if there were different time points measured in the intervention. Where possible, we will adjust for clustering if participants were allocated to the same intervention in groups (e.g. family members or care homes) for example in cluster‐RCTs. To account for the effect of clustering, we will estimate inflated standard errors (SEs)(Deeks 2011).

With regard to RCTs and other comparative studies, we will consider adjustments to avoid multiple use of participants in multi‐arm studies. If multiple arms are to be included in the meta‐analysis in such a way that one intervention arm is to be included in more than one treatment comparison, we will divide the number of participants (in that arm) and the number of events by the total number of treatment comparisons. This approach retains original information but reduces the precision of the pooled estimates.

Dealing with missing data

We will make all reasonable efforts to contact authors for details and reasons for any missing outcome data (Young 2011). If data are missing and no information is available in the paper or is not obtainable by contacting the authors, we will assume for dichotomous data that drop out was due to the intervention being unsuccessful. We will perform sensitivity analysis to assess the effects of these assumptions, for example missing data for intervention studies can be dealt with by using intention‐to‐treat numbers. For continuous data, we may use this information if we are able to calculate the data from other data provided, for example, calculating SD from SE and P value (Higgins 2011b). We will reflect on the overall effect of missing data in the final discussion.

Assessment of heterogeneity

We will assess levels of statistical heterogeneity by observing the confidence interval within studies and to a lesser extent, statistically by use of the Chi2 test to determine whether observed differences in results are compatible with chance alone. Interpretation of the Chi2 will take into consideration the sample size and number of studies included in the meta‐analysis.

We will quantify heterogeneity with the I2 statistic, using the following guide. Some level of heterogeneity is expected (Higgins 2011a).

  • 0% to 40%: might not be important

  • 30% to 60%: may represent moderate heterogeneity

  • 50% to 90%: may represent substantial heterogeneity

  • 75% to 100%: very substantial ("considerable") heterogeneity.

Assessment of reporting biases

Thorough searching of multiple databases should limit general reporting biases. We will keep reporting biases due to such things as time lag bias, Hopewell 2007, selective outcome reporting, Dwan 2014, and publication bias, Hopewell 2009, to a minimum by inclusion of prospective trial registers; if we identify at least 10 studies, we will use funnel plots to explore publication bias and other biases related to study size. We will contact authors for results of outcomes included in their study objectives but not reported.

Data synthesis

We will only undertake meta‐analyses when studies are sufficiently homogeneous in their participants, interventions, and outcomes. We will combine RRs for dichotomous data, and MDs or SMDs for continuous data, using random‐effects models if at least four studies are included in a meta‐analysis. We will use fixed‐effect models if there are fewer than four studies and if heterogeneity is reasonably low.

Where we do not undertake meta‐analysis, we will present a narrative synthesis.

Subgroup analysis and investigation of heterogeneity

We may consider subgroup analysis for topics such as the following if sufficient data are available, but will do so cautiously if the amount of data is small (Bender 2008).

  • Level of intellectual disability

  • With versus without support of dental professionals

  • Non‐formal carers (i.e. parents or siblings) versus formal carers

  • Setting of the intervention (i.e. home‐based versus residential/institutional/school‐based interventions)

  • Behavioural change technique used based on the COM‐B method

  • Short‐term interventions versus long‐term interventions

Sensitivity analysis

Providing there are sufficient included studies, we will undertake sensitivity analysis based on risk of bias.

Summarising findings and assessing the quality of the evidence

We will create one or more 'Summary of findings' tables to show the effects of different interventions on different outcome measures, using GRADEpro software (Schünemann 2013). We will supplement this with information on participant and study characteristics where appropriate. We will assess the quality of the body of evidence with regard to the overall risk of bias of the included studies, directness of the evidence, consistency of the results, precision of the estimates, and risk of reporting bias. We will categorise the quality of the body of evidence for each of the main outcomes for each comparison as high, moderate, low, or very low.