PROSPERO International prospective register of systematic reviews
Text messaging and smartphone app interventions to improve adherence in adolescents and young adults: a systematic review and meta-analysis
Sherif Badawy, Mohamed Sino, Saara Kaviany, Linda O’Dwyer, Robert Liem, Alexis Thompson, Lisa Kuhns
Sherif Badawy, Mohamed Sino, Saara Kaviany, Linda O’Dwyer, Robert Liem, Alexis Thompson, Lisa Kuhns. Text messaging and smartphone app interventions to improve adherence in adolescents and young adults: a systematic review and meta-analysis.
Available from http://www.crd.york.ac.uk/PROSPERO/display_record.asp?ID=CRD42015025907
1. What are the feasibility, acceptability and efficacy of text messaging interventions in improving medication adherence in adolescents and young adults with chronic conditions?
2. What are the feasibility, acceptability and efficacy of smartphone apps interventions in improving medication adherence in adolescents and young adults with chronic conditions?
3. What are the feasibility, acceptability and efficacy of text messaging and smartphone apps interventions in improving adherence to preventive, wellness or health promoting and acute care measures in adolescents and young adults?
We will search the following electronic databases:
• Cochrane Library (including Cochrane Central Register of Controlled Trials (CENTRAL), Cochrane Database of Systematic Reviews, Cochrane Methodology Register)
• Web Of Science
• IEEEExplore (Institute of Electrical and Electronics Engineers Explore)
• CRD (Centre for Review and Dissemination)
• NHS (National Health Services) Health Technology Assessment Database
• Others (engineering and computer science databases)
Additional studies will be sought by hand searching the reference lists of retrieved primary trials, systematic reviews, meta-analyses, health-technology assessment reports, relevant studies, and other personal collections of articles.
No study will be excluded on the basis of Language.
Grey literature will be searched through the following sources:
• Google Scholar;
• Conference proceedings;
• ProQuest Dissertations;
• Mobile Active: a user-created directory of mobile health solutions.
• Databases of completed and ongoing trials will be searched in the following resources (2008-present):
2. Current Controlled Trials (http://www.controlled-trials.com/)
3. WHO International Clinical Trials Registry Platform (ICTRP) (http://www.who.int/ictrp/en/)
Unlike studies identified from electronic databases, non-English articles in the grey literature will be included only if there is an English abstract.
We will contact study authors where additional contextual information is required to clarify detail and where it is thought they may be able to identify other, new or ongoing studies relevant to the review.
• Quasi-experimental studies (including interrupted time series studies);
• Pilot or feasibility studies
• Non-experimental observational (cross-sectional, case-series, case studies)
Condition or domain being studied
Adolescents and young adults with chronic conditions and those who require adherence to preventive, wellness or health promoting and acute care measures
Adolescents and young adults (age 12 – 24 years old)
• Technology-based interventions that include text messaging and smartphone apps
• Interventions to improve medication adherence in patients with chronic conditions
• Interventions to improve adherence to preventive, wellness or health promoting and acute care measures
• Technology-based interventions other than text messaging and smartphone apps
• Parent targeted text messaging and smartphone apps
The comparison is the usual standard of care, other intervention variant, or in the case of a randomized control trial, the comparison is control condition. Studies with no comparison will still be eligible for inclusion.
Medical and non-medical (community-based) contexts will be considered. There are no exclusion criteria for context.
• Adherence to the recommended treatment or intervention
• Feasibility, usability, accessibility, satisfaction, and utilization or uptake of text messaging or smartphone app interventions
• Clinical or disease related outcomes, such as complications, hospitalizations and emergency room visits
• Disease related biomarkers or anthropometric measures
• Self-report quality of life (QoL) or patient reported outcomes (PROs) measures
• Self-efficacy or self-management measures
• Other knowledge or behavior change measures
• Barriers and facilitators to adoption for both patients and providers
• Any cost-effectiveness assessment
We will include outcomes measured at completion of an intervention and any subsequent time points (follow-up). We define short-term follow-up as outcomes monitoring that completes within 30 days of the completion of the intervention, long-term follow-up as monitoring that continues 6 months or more after the completion of the intervention and medium-term follow-up as intermediate of the two.
Data extraction, (selection and coding)
A medical librarian will be responsible for creating a search strategy and conduct literature search. Reference management software EndNote will be used to combine search results from electronic databases, and remove duplicated records. Two other researchers will independently screen the articles to determine if they meet the criteria for the systematic review, meta-analysis or both. The two researchers will then view the abstracts of the remaining articles independently to determine eligibility based on the specified inclusion/exclusion criteria. Full-text articles will be referred to where required to clarify eligibility. Agreement on inclusion or inter-rater reliability will be calculated using the kappa statistics to ensure consistency. Any disagreements or differences in the articles selected by the two researchers will be discussed to reach a decision regarding inclusion, resolved by a third reviewer or both. If resolving disagreement is not possible, the author will be contacted for clarification.
Reference lists of all included articles will also be searched for additional articles. Once the subset of publications meeting inclusion criteria is finalized, each publication will be reviewed and its characteristics documented using a standardized pre-tested data extraction form. The study flow and reasons for exclusion of full text papers will be documented in an adapted PRISMA (preferred reporting items for systematic reviews and meta-analyses) study flowchart.
Data to be extracted by core team members would include: first author's name, publication year, country, study design, condition or disease of the study, theoretical framework, duration of intervention, follow-up, components/type of technology-intervention (text messaging or smartphone apps), control group, characteristics of participants and adolescents/young adults (age, gender), setting, recommended treatment or condition-specific intervention, adherence measures, adherence rates and outcome measures such as disease-related outcomes “morbidity and mortality”, self-report PROs or Qol, self-efficacy or self-management, cost-effectiveness etc. A random sample of 10% of records will be crosschecked for accuracy. If discrepancies are encountered, a second 10% sample will be checked to assess if the problem is widespread. The reviewers will attempt to contact the authors of studies that are missing key data. The reviewers will attempt to translate included studies written in other languages, or use trusted online or commercial translation services.
Risk of bias (quality) assessment
To evaluate the methodological quality of included RCT studies, the Cochrane Collaboration’s Risk of Bias Tool will be used. This method uses the following bias criteria:
• Random sequence generation (selection bias);
• Allocation concealment (selection bias);
• Blinding of participants and personnel (performance bias and detection bias);
• Blinding of outcome assessment (performance bias and detection bias);
• Incomplete outcome data (attrition bias);
• Selective outcome reporting (reporting bias);
• Other bias.
A modified version of the risk of bias tool or checklist will be used to assess the risk of bias and individual bias items using the criteria as stipulated by the International Cochrane Collaboration. Risk of bias criteria will be judged as 'low risk', 'high risk' or 'unclear risk', and then will be summarized as 'yes' (indicating a low risk of bias in all domains), 'no' (indicating a high risk of bias in at least one of the domains) or 'unclear' (indicating an uncertain risk of bias in at least one of the domains). Other observational studies will be assessed using the NOS score (Newcastle-Ottawa Quality Assessment Scale). The NOS score rates quality based on high risk (1-3 stars), medium risk (4-5 stars), or low risk (6-9 stars) NOS score.
Independent assessment will take place as quality assessment will be undertaken by one reviewer and checked by a second reviewer. Assessment of internal validity (bias, confounding, etc) and assessment would include (a) RCTs: adequacy of randomization, intervention blinding, blinded outcome assessment, intention to treat analysis, covariates, power, baseline characteristics, dropout, extent of missing data, summary results, and other relevant data; and (b) observational trials: comparability of groups, control for confounding factors, extent of missing data, intention to treat analysis.
Any disagreements between the two reviewers regarding the quality of a trial will be discussed and resolved by a consensus, resolved by a third reviewer or both. Where data are missing, reviewers will attempt to contact the author/s at their last known address.
To evaluate the possibility of publication bias, we will use the Peters test and a color-enhanced funnel plot that can be done using STATA software.
For the quality assessment of qualitative studies, the quality criteria defined by Critical Appraisal Skills Programme (CASP) will be used. This tool contains ten items relating to rigorousness, credibility and relevance. Three categories will be established for each item: clearly met, clearly not met, and unclear.
Strategy for data synthesis
Quantitative evidence synthesis:
Where appropriate, meta-analysis will be performed according to the statistical guidelines referenced in the newest version of the Cochrane Handbook for Systematic Reviews of Interventions. However, if meta-analysis is considered unsuitable due to the heterogeneity and/or small number of studies, a narrative overview of the findings of included studies will be presented with tabular summaries of extracted data. Primarily, the narrative synthesis will be structured according to the intended purpose of the intervention (text messaging or smartphone apps), and then, studies will be organized by the objectives addressed in each of them. Typically, study, intervention, population, and outcome characteristics will be summarized in text, evidence, and summary tables. Results for each outcome of interest will be presented separately and stratified according to the purpose and the type of intervention, and compared qualitatively and, where possible, quantitatively in text and summary tables. We will summarize the effects associated with quantitative outcomes in a separate table with summary of the findings. We will calculate effect sizes (average difference, mean differences, risk ratios, percentage difference) for intervention, and use I-squared test for heterogeneity to generate forest plots. Software will be used to conduct the meta-analyses.
Evidence will not be formally graded, as the purpose of this review is descriptive rather than prescriptive.
Qualitative thematic synthesis:
A qualitative thematic synthesis will be performed for all identified relevant qualitative studies, which utilize qualitative methods of data collection and analysis only if was a discrete part of a mixed-method study, but not a stand-alone study. Studies assessing the same outcome will be grouped and coded, and the free text of included studies will be extracted and coded using available software. Data extracted from each of the studies will align with data required to answer the research questions and will be used in aggregate form and we will structure the narrative synthesis so that, when appropriate, information is segregated. Information will be stratified according to the purpose and type of intervention, common evaluation tools/approaches, implementation experiences for diverse mHealth technology and other relevant data, and we will also summarize included studies in a separate table of characteristics of included studies.
Analysis of subgroups or subsets
The following a priori subgroup analyses, mainly for quantitative outcomes, are planned:
• Age groups (12 – 17, 18 – 24)
• Patient population (underlying condition) and their characteristics (gender, race, ethnicity, etc.)
• Purpose of the intervention (adherence to chronic medication or adherence to preventive, wellness or health promoting and acute care measures)
• Type of study design (pre-experimental, experimental, quasi-experimental) and other relevant study characteristics
• Type of the intervention (text messaging, smartphone app)
• Duration of the intervention
• Type of outcomes (quantitative ones)
• Quality rating
Presentation at Academy Health Annual Research Meeting 2016 and Peer-reviewed publication (Systematic Review and Meta-analysis).
Contact details for further information
225 E Chicago Ave., Box #30, Chicago, IL 60611
Organisational affiliation of the review
Ann & Robert H. Lurie Children's Hospital of Chicago - Feinberg School of Medicine at Northwestern University
Dr Sherif Badawy, Department of Pediatrics, Division of Hematology, Oncology and Stem Cell Transplant, Ann & Robert H. Lurie Children's Hospital of Chicago - Feinberg School of Medicine at Northwestern University Dr Mohamed Sino, Department of Pediatrics, University of Kansas Medical Center Dr Saara Kaviany, Department of Pediatrics, Advocate Children's Hospital - Oak Lawn Ms Linda O’Dwyer, Galter Health Sciences Library, Feinberg School of Medicine at Northwestern University Dr Robert Liem, Department of Pediatrics, Division of Hematology, Oncology and Stem Cell Transplant, Ann & Robert H. Lurie Children's Hospital of Chicago - Feinberg School of Medicine at Northwestern University Professor Alexis Thompson, Department of Pediatrics, Division of Hematology, Oncology and Stem Cell Transplant, Ann & Robert H. Lurie Children's Hospital of Chicago - Feinberg School of Medicine at Northwestern University Dr Lisa Kuhns, Department of Pediatrics, Division of Adolescent Medicine, Ann & Robert H. Lurie Children's Hospital of Chicago - Feinberg School of Medicine at Northwestern University
Anticipated or actual start date
05 July 2015
Anticipated completion date
01 March 2016
Conflicts of interest
United States of America
Subject index terms status
Subject indexing assigned by CRD
Subject index terms
Adolescent; Adult; Cell Phones; Humans; Telephone; Text Messaging; Young Adult
Stage of review
Date of registration in PROSPERO
07 September 2015
Date of publication of this revision
07 September 2015
Stage of review at time of this submission
Piloting of the study selection process
Formal screening of search results against eligibility criteria
Risk of bias (quality) assessment
PROSPERO This information has been provided by the named contact for this review. CRD has accepted this information in good faith and registered the review in PROSPERO. CRD bears no responsibility or liability for the content of this registration record, any associated files or external websites.