A systematic review and meta-analysis of the use of TB in paediatric patients with a chronic illness, to teach the skills required for self-management and treatment adherence compared with a controlled group.
We have developed this protocol in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocol (PRISMA-P) (31) statement and have registered it with Prospero [CRD42021229025].
Type of participants
This review will include only studies that involved children (less than 18 years of age) in any healthcare setting (community, primary,secondary or tertiary care). Included study participants were those with any chronic disease including (but not exclusively); asthma, allergy, diabetes and epilepsy.
Types of intervention
Eligible studies are those which reported on the use of the Teach-Back method to educate patients on any aspect of the self-management of their disease. The intervention could be delivered by any healthcare worker. The comparator was any educational method that did not involve Teach-Back.
Types of studies
This review will consider quantatative studies including randomised control trials (RCT’s), quasi-experimental studies and observational studies that evaluated the effect of Teach-Back. All publications up until February 2021 will be included. Non-english plublications will be excluded.
Types of outcomes
Selected outcomes will be disease specific self-management skill acquisiton, treatment adherence and disease specific knowledge.
Information sources and search strategy
Under the guidance of the Queen’s University Belfast medical librarian, a search strategy will be developed and performed electronically. We will include a range of terms and keywords related to the research question, such as: “teach-back”, “teach-to-goal”, “closing the loop”, “closing the cycle”, “repeated instruction”, “children” and “paediatrics.” To maximise the potential of all potential literature being captured by the search, the following digital databases will be used: Medline, Embase, Web of Science, Scopus, CINAHL, and the Cochrane Library, along with eligible publications sourced from the citations within papers elicited from the search. If a full-text publication, which meets the eligibility criteria is not available then contact will be made with the study author to request access.
Data collection and analysis
EndnoteX9TM will be the reference management system used to save the results from each online search. Separate files for each online database will be kept and then all results will be imported into one library. At this point, all duplicates will be removed. All results will then be exported onto a Microsoft ExcelTM file. During the screening and selection process, documentation will be recorded on Microsoft ExcelTM.
We will screen each of the studies by title and abstract to ensure they fit the eligibility criteria. Titles and or abstracts which clearly fit the criteria will be brought forward for full text analysis. Studies which do not fit the eligibility criteria will be reviewed by a second independent reviewer before excluding them from further analysis. Those studies which have been deemed to fit the eligibility criteria as per title and abstract will be brought forward for full text analysis. In the event that a full text article is not available or that there is insufficient data within the publication we will contact the author to try to obtain the data.
Once all full text articles are available, the papers will be reviewed by each member of the research team to ensure they meet eligibility criteria. During this review process, any articles which are deemed relevant from the references of included studies will be screened in the same process as described above and considered for inclusion.
A draft data extraction table will be created and then piloted by two members of the research team independently. This will allow for iterative changes to ensure we have captured all of the relevant information. Any disagreements will be resolved by consensus between all four members of the research team. Amendments made will be noted and documented before a final data extraction table is decided upon.
Information will be extracted from each included trial regarding 1. The participants’ chronic disease type 2. The skill/knowledge under investigation and 3. A measure of the effect size that the TB educational intervention had on this skill/knowledge. These have been prioritised in order to best answer the research question which is whether the use of TB methodology is effective in teaching children and/or their carers to self-manage their chronic disease.
Risk of bias in individual studies
The development and registration of this protocol in line with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses Protocol (PRISMA-P) (31) statement will help protect from reporting bias. Using two independent reviewers to screen the articles will aid with selection bias along with two further independent reviewers to aid with ambiguous cases. Ideally, the articles would be masked by author and institution. However a consensus was achieved, and this was deemed unnecessary. By accepting only peer-reviewed published studies, we accept a potential for evidence selection bias. This bias is difficult to overcome, as to answer the research question, we need to review studies which have reported outcomes with a statistical measure of effect size. We will report the methods used to assess the risk of bias in each included study and we will use the Cochrane Collaboration’s tool for assessing risk of bias (32). The quality of evidence will be evaluated by all four reviewers using the ‘Grading of Recommendations Assessment, Development and Evaluation’ (GRADE) approach (33). There are no competing interests from the research team to report in performing this systematic review.
Data synthesis and meta-analysis
To test for heterogeneity in the quantitative data, we will use the I2 test, taking an I2 of > 75% as being high heterogeneity
Using the main outcomes of effect size of change in desired skill/knowledge, if appropriate, a meta-analysis will be performed using a random-effects model, with the difference between means (for continuous data) and the odds ratio (for categorical data) calculated as the principal summary measure. These findings will be illustrated using a forest plot based on 95% confidence intervals. StatsDirect v 3.34, (StatsDirect Limited, www.statsdirect.com) will be used to analyse the data.
Where statistical pooling is not possible, the findings will be presented in a narrative form.
Sub-group analysis looking at asthma patients will be carried out to explore the effectiveness of Teach-Back as an educational intervention for asthma self-management in children (in particular, adherence to inhaled medications, inhaler technique and understanding of personalised asthma action plans).