Powered toothbrushes for oral health
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To compare powered toothbrushes with different actions, in everyday use, by people of any age, in relation to:
(1) removal of plaque;
(2) health of the gingivae;
(3) adverse effects.
Background
Dental plaque is the primary causal factor both in gingivitis (gum inflammation) and periodontitis (loss of bone around the teeth) although the link between the two is complex and not well understood (Löe 1965).
Thus, good oral hygiene (the removal of plaque) by effective toothbrushing has a key role in oral health. Effective toothbrushing depends on a number of factors including motivation, knowledge and manual dexterity.
The relationship between oral cleanliness and caries is not clear‐cut (Richardson 1977; Addy 1986) unless a fluoride toothpaste is used. This is due more to the effect of fluoride than brushing per se (Marinho 2003).
Powered toothbrushes were first introduced commercially in the early 1960s (Chilton 1962; Cross 1962; Hoover 1962; Elliot 1963) and have become established as an alternative to manual methods of toothbrushing. They simulate the manual motion of toothbrushes although both lateral and rotary movements are present in the commercial designs available. A recent review (Heanue 2003) comparing whether powered toothbrushes were better for oral health than manual toothbrushes concluded that only certain powered toothbrushes were more effective at removing plaque and reducing gingivitis. This review did not allow direct comparison of the different powered toothbrush types against each other. Therefore the question remains. Which powered toothbrushes may perform better, at reducing plaque and gingivitis?
Objectives
To compare powered toothbrushes with different actions, in everyday use, by people of any age, in relation to:
(1) removal of plaque;
(2) health of the gingivae;
(3) adverse effects.
Methods
Criteria for considering studies for this review
Types of studies
The review is confined to studies where a comparison of two or more powered brushes with different modes of action has been carried out. Trials comparing powered toothbrushes with the same mode of action will not be included in the review. Only randomised controlled trials (RCTs) that clearly imply randomisation will be included. Where the method of randomisation is unclear, the authors will be contacted to request clarification. Cross‐over trials are eligible. Split‐mouth trials are excluded as these are not considered representative of 'everyday use'.
Types of participants
Included individuals will be of any age with no reported disability that may affect their ability to practice normal self‐toothbrushing. Individuals wearing orthodontic appliances will be included.
Types of interventions
Toothbrushes to be reviewed are all forms of powered brushes with mechanical movement of the brush head (i.e. electric, electronic, sonic, ultrasound etc.) and power delivery to the brush head (i.e. ionic). The brushes will be grouped according to their mode of action. These include.
-
Side to side action, indicates a brush head action that moves laterally side to side.
-
Counter oscillation, indicates a brush action in which adjacent tufts of bristles (usually 6 to 10 in number) rotate in one direction and then the other, independently. Each tuft rotating in the opposite direction to that adjacent to it.
-
Rotation oscillation, indicates a brush action in which the brush head rotates in one direction and then the other.
-
Circular, indicates a brush action in which the brush head rotates in one direction.
-
Ultrasonic, indicates a brush action where the bristles vibrate at ultrasonic frequencies (> 20 kHz).
-
Ionic, indicates a brush which utilises a low electric current to the bristles during toothbrushing. The aim of the brush is to change the charge polarity of the tooth and this will result in attracting dental plaque away from the tooth towards the bristles. No mechanical action is powered by the brush.
-
Unknown, indicates a brush action that the reviewers have been unable to establish based on the trial report or confirm with the manufacturers.
Combined interventions, e.g. brushing combined with the use of mouthrinses or irrigation will be excluded. However trials where subjects are permitted to continue with their usual personal adjuncts to oral hygiene, such as flossing, in both groups, will be included. Professional or supervised brushing will be excluded. Trials of 28 days and over will be eligible, and a subgroup analysis will be carried out on duration of trials for the different outcome measurements.
As with the previous review comparing powered and manual toothbrushes (Heanue 2003) it has been considered that analysis of filament arrangement, orientation, size, shape and flexibility, brush head size and shape along with presence or absence and characteristics of a timer would prove difficult to define across time and brush types.
Types of outcome measures
The outcome measures will be quantified levels of plaque and/or gingivitis. Values recorded on arrival at the assessment will be used (not those after subjects have been instructed to brush their teeth). Where several outcome measures are reported, the data entered for analysis will be selected according to an index hierarchy developed on a previous review (Heanue 2003). This was developed as many different indices of plaque and gingivitis were used across trials and some trials reported multiple indices. A frequencies table was prepared of the indices used and they were ranked based on common usage and simplicity.
Therefore for plaque we will extract, where possible, data reported as the Turesky et al modification of the Quigley‐Hein plaque index of 1962 (Turesky 1970). For gingival inflammation we will extract where possible data reported as the gingival index of Löe and Silness (Löe 1963) or, if unavailable, bleeding on probing (Ainamo 1975). Data for 'Russell's periodontal index' will be excluded because this index fails to distinguish between gingivitis and periodontitis (Russell 1967).
Where available, data will be extracted for whole mouth scores as opposed to part mouth scores. Where only part mouth scores are reported in a study, they will be extracted and a sensitivity analysis carried out to consider their impact on the results of the review. Part mouth scoring will be assumed if plaque and/or gingivitis were not recorded around all erupted teeth, except third molars.
Search methods for identification of studies
Identification of trials will be effected through electronic searches of the following databases:
Cochrane Oral Health Group's Trials register
Cochrane Central Register of Controlled Trials (CENTRAL ‐ The Cochrane Library, current issue)
MEDLINE (1966 to present)
EMBASE (1980 to present)
CINAHL (1982 to present)
Web of Science (1945 to present).
The MEDLINE search (Appendix 1) will combine the subject search with Cochrane sensitive search filters for identifying randomised controlled trials (RCTs) as published in the Cochrane Handbook for Systematic Reviews of Interventions.
Handsearching of key journals, which have not already been searched by the Cochrane Oral Health Group, will also be carried out, including the Journal of Dental Research and Journal of Dentistry.
Additionally all references in the identified trials will be checked and the authors contacted to identify any additional published or unpublished data. All identified manufacturers will be contacted and additional published or unpublished trial reports requested. Publications in all languages will be considered.
Data collection and analysis
Titles and abstracts identified through the searches will be checked by two review authors. The full text of all studies of possible relevance will be obtained for independent assessment by two review authors. The review authors will decide which trials fit the inclusion criteria, and grade their methodological quality independently. Any disagreement will be resolved by discussion between the review authors. Authors will be contacted for clarification where necessary.
The methodological quality of the included trials will be assessed with particular emphasis on allocation concealment, which will be ranked using the Cochrane approach:
(A) Adequate
(B) Unclear
(C) Inadequate.
Where there is uncertainty authors will be contacted for clarification. The agreement on methodology assessment will be reported using Kappa statistics.
The methodological quality of studies will also consider issues such as:
-
power of study/a priori calculation of sample size;
-
blind outcome assessment;
-
comparability of groups at baseline/adjustment for confounding factors;
-
clear inclusion/exclusion criteria;
-
method of assessment (valid assessment criteria);
-
duration of follow up;
-
intention‐to‐treat analysis;
-
withdrawals.
Data extraction will be performed independently by two review authors and the authors of trials contacted to provide missing data where possible. Data will be checked and entered into Review Manager (RevMan) software by one review author.
The data extraction protocol will include:
-
patient characteristics (age, number, gender, special group such as dental students, orthodontic treatment);
-
intervention characteristics (type of brush, duration of use, delivery of instructions etc.);
-
outcome characteristics (plaque, gingivitis indices etc.);
-
sponsors of trial;
-
publication status.
Primary analyses will compare the relative effectiveness of the toothbrush designs. Studies will be grouped according to time of outcome measurements. Short term will be defined as an outcome measured between 4 weeks and 3 months. Long term will be defined as an outcome measured after 3 months. For trials presenting more than one outcome measurement within a time period, the last outcome in the time period will be entered.
Different indices for plaque measure the same concept on different scales, with high correlation between the different indices. The same is true for gingivitis. As it is not possible to combine the results from different indices, for trials presenting continuous outcomes, the effects will be expressed as standardised values, which have no units, before combining. The standardised mean difference (SMD) will therefore be calculated along with the appropriate 95% confidence intervals (CI) and used as the effect measure for meta‐analysis. For dichotomous outcomes, the results for individual trials will be presented as risk ratios (RR) with 95% confidence intervals. Random‐effects models will be used for all pooling of data.
It is planned to combine the data from cross‐over trials with that of similar parallel group trials, using the techniques described by Elbourne et al (Elbourne 2002).
Heterogeneity will be assessed by inspection of a graphical display of the estimated treatment effects from the trials along with their 95% CI and by Cochran's test for homogeneity undertaken before each meta‐analysis. Subgroup analyses of the following will be undertaken for age, presence of orthodontic appliances, professional instruction, and duration of trial, assessments based on full mouth recording versus those based on a partial recording and to examine the effects of concealed allocation, randomisation generation and blind outcome assessment on the overall estimates of effect for important outcomes. Baseline data for plaque will be recorded. If the spread of data is sufficient then a subgroup analyses of high versus low baseline plaque levels will be carried out.
Sensitivity analyses will be performed on the basis of methodological quality. A funnel plot (plots of effect estimates versus the inverse of their standard errors) will be drawn. Asymmetry of the funnel plot may indicate publication bias and other biases related to sample size, though it may also represent a true relationship between trial size and effect size. A formal investigation of the degree of asymmetry will be performed using the method proposed by Egger et al (Egger 1997). A further method proposed by Begg and Mazumdar which tests for publication bias by determining if there is a significant correlation between the effect estimates and their variances will also be carried out (Begg 1994). Both methods will use Stata version 7.0 (Stata Corporation, USA) program Metabias.
We will also examine adverse effects such as hard or soft tissue injury and damage to orthodontic appliances and prostheses. The reporting of these will be dependent on the data recorded within the trials.
