This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
The prevalence of physical chronic or long-term conditions in adolescents aged 10-24 years is rising. Mobile phone and tablet mobile technologies featuring software program apps are widely used by these adolescents and their healthy peers for social networking or gaming. Apps are also used in health care to support personal condition management and they have considerable potential in this context. There is a growing body of literature on app use in health contexts, thereby making a systematic review of their effectiveness very timely.
To systematically review the literature on the effectiveness of mobile apps designed to support adolescents’ management of their physical chronic or long-term conditions.
We conducted a review of the English-language literature published since 2003 in five relevant bibliographical databases using key search terms. Two independent reviewers screened titles and abstracts using data extraction and quality assessment tools.
The search returned 1120 hits. Of the 19 eligible full-text papers, four met our review criteria, reporting one pilot randomized controlled trial and three pretest/post-test studies. Samples ranged from 4 to 18 participants, with a combined sample of 46 participants. The apps reported were targeted at type 1 diabetes, asthma, and cancer. Two papers provided data for calculating effect size. Heterogeneity in terms of study design, reported outcomes, follow-up times, participants’ ages, and health conditions prevented meta-analyses. There was variation in whether adolescents received guidance in using the app or were solely responsible for navigating the app. Three studies reported some level of patient involvement in app design, development, and/or evaluation. Health professional involvement in the modelling stages of apps was reported in all studies, although it was not always clear whether specific clinical (as opposed to academic) expertise in working with adolescents was represented. The dearth of studies and the small overall sample size emphasizes the need for future studies of the development, evaluation, use, and effectiveness of mobile apps to support adolescents’ personal management of their conditions.
A key finding of the review is the paucity of evidence-based apps that exist, in contrast to the thousands of apps available on the app market that are not evidence-based or user or professional informed. Although we aimed to assess the effectiveness of apps, the dearth of studies meeting our criteria meant that we were unable to be conclusive in this regard. Based on the available evidence, apps may be considered feasible health interventions, but more studies involving larger sample sizes, and with patient and health professional input at all stages, are needed to determine apps’ acceptability and effectiveness. This review provides valuable findings and paves the way for future rigorous development and evaluation of health apps for adolescents with chronic or long-term conditions.
Globally, the pattern of illness in young people or adolescents aged 10-24 years (hereafter referred to as adolescents) has shifted from acute to long-term or chronic conditions (hereafter referred to as chronic). A chronic condition in this age group is a health condition that at the time of diagnosis is predicted to last longer than 3 months [
Effective support from the health sector is therefore paramount, especially during the transition from pediatric to adult health services, and particularly if adult services are not young-person friendly [
Contemporaneous reports indicate that utilizing modern mobile electronic technologies in health interventions for young people [
Significant declines in treatment adherence have been observed during adolescence and the transition from pediatric to adult-centered health services [
In a recent commentary, Wu and Hommel [
Personal management of chronic physical conditions involves five core skills: problem solving, decision making, resource utilization, forming patient-health professional relationships, and taking action [
A review of the effectiveness of mobile health technology-based health behavior change or disease management interventions for adults found that only six of the 49 interventions used apps and none of these involved adolescents [
Mobile apps are widely accepted by adolescents living in today’s technology-rich environment. In the United Kingdom for instance, children and adolescents aged 5-15 years are frequent users of mobile technologies. Indeed, 62% of 12-15-year-olds own a mobile phone, and the use of tablet computers by 5-15-year-olds tripled between 2012 and 2013 with 42% using tablets in 2013 [
In 2013, the UK National Health Service (NHS) Commissioning Board unveiled a library of NHS-reviewed health apps [
Yet despite increased use and the significant potential of these technologies for adolescents with chronic conditions, to our knowledge a synthesis of studies of their effectiveness in this population has not been undertaken. This systematic review of the evidence is, therefore, timely as it aimed to assess the effectiveness of mobile phone and tablet apps for adolescents’ personal management of chronic conditions. In this review, young people are defined as those aged 10-24 years (as defined by the World Health Organization [WHO] [
This systematic review aimed to synthesize the evidence on mobile phone and tablet apps. The methodology adhered to that described in the Cochrane Handbook for Systematic Reviews of Intervention [
Eligible studies were identified through a comprehensive literature search of the following five bibliographical databases: MEDLINE, the Cumulative Index to Nursing and Allied Health Literature (CINAHL), Embase, PsycINFO, and the Web of Science. The search strategy, which was developed in consultation with an information scientist, used standardized indexed search terms and free-text terms that relate to the following three key concepts: (1) adolescents, (2) physical chronic conditions, and (3) mobile technology. The search included British and North American terms and spellings. The search strategy was initially devised in MEDLINE and then adapted to the other databases. The Web of Science did not employ any indexed search terms and the other databases did not employ them in a standardized fashion. Free-text terms were used consistently throughout. In addition to testing search sensitivity, journals associated with the most retrieved citations were hand searched from 2009 to 2014, thus decreasing the likelihood of missing relevant studies. The identification of any studies additional to those we had identified from hand searching allowed us to comment on the rigor of the search strategy and the quality of indexing in the bibliographic databases mentioned above. This is a particularly useful strategy in this relatively new domain of mobile technology. Also, due to the emerging nature of mobile technology, the search included conference abstracts published in peer-reviewed journals, and authors were contacted requesting additional related published or unpublished work.
Initially, all papers were independently scrutinized by two reviewers (MR, AH) using a screening tool that outlined the study inclusion criteria (see
Criteria for included studies are shown in
Inclusion criteria
Population: Adolescents aged 10-24 years (WHO definition from 2001 [
Intervention: Any app for a mobile phone or tablet that could be considered a management intervention (or a component of an intervention) in terms of content and/or delivery. This judgment was based on the five core management skills for chronic physical health conditions, as outlined by Lorig [
Comparisons: Intervention versus usual care
Outcomes: Any physiological, attitudinal, behavioral, or knowledge outcomes.
Study design: Randomized controlled trial (RCT)
The Cochrane Collaboration excludes nonrandomized controlled trials due to their greater bias, but because this is a relatively new area, we included studies of various designs to systematically collect an overview of the current evidence.
While international literature was included, non-English-language publications and studies that focused on adolescents with mental health problems, learning disabilities, and/or cognitive impairment were excluded due to resource limitations. Interventions using mobile phone technology only in the context of delivering/receiving text messages or phone calls were also excluded. Given the review focus, the technology context was considered key so we applied a publication start date of 2003. This is the year when 3G networks, which provided the bandwidth required to support advanced mobile Internet apps, were launched in the United Kingdom [
For every included study, two reviewers extracted relevant data independently. A tool based on the data extraction template for Cochrane reviews [
The evidence and quality of the papers included in the systematic review were assessed using the Effective Public Health Practice Project (EPHPP) quality assessment tool [
Except where indicated, extracted data from the papers were analyzed using IBM SPSS Statistics version 22.0 (IBM Corp). For each paper, inter-rater reliability was estimated for each of the domains, the total score, and the final grade of the EPHPP. Following Armijo-Olivo et al [
The combined electronic searches identified 1120 records. Of these, 338 records were removed after accounting for duplicates, leaving 782 records for further consideration. Out of the 782 titles and abstracts that were then screened to test eligibility using the PICOS screening tool (population or participant, intervention or indicator, comparator or control, outcome, and study design), 19 full-text papers were potentially eligible for inclusion. Many of the excluded papers reported observational, noncontrolled studies that did not focus on the population or intervention of interest.
Seven of the 19 studies included from the title and abstract screen were abstracts from conference presentations; a search was undertaken to find the full paper of each abstract, failing which the corresponding author was contacted. For two of the seven conference papers, we found subsequent publications [
PRISMA flow diagram of the review.
The four included studies described four different apps addressing the needs of adolescents with different chronic conditions: type 1 diabetes [
Aldiss et al evaluated the Advanced Symptom Management System (ASyMS) which utilizes mobile phone technology to monitor chemotherapy-related symptoms and promote self-care [
Characteristics of the included studies.
Study | Country | SSa, n | Measurements | IDb, weeks | Fc, % | Age (years), |
Study design | Cond.d |
Aldiss |
UK | 4 | Pediatric Quality of Life Inventory |
2 | 0 | N/Ae
|
Pilot RCTf | Cancer |
Burbank |
USA | 18 | Self-efficacy questionnaire |
8 | N/A | 13.5 |
Feasibility |
Asthma |
Cafazzo |
Canada | 12 | Self-care inventory |
12 | 67 | 14.9 |
Pretest/ |
Type 1 diabetes |
Frøisland |
Norway | 12 | HbA1ch
|
12 | 54 | 16.2 |
Pretest/ |
Type 1 diabetes |
aSS: sample size.
bID: intervention duration.
cF: female.
dCond.: condition.
eN/A: not applicable (information not reported or stated).
fRCT: randomized controlled trial.
gQOL: quality of life.
hHbA1c: hemoglobin A1c (glycated hemoglobin).
In light of the national asthma education program recommendation for a written asthma action plan (AAP) for all patients with asthma, and in recognition that few studies demonstrating acceptance of phone-based self-monitoring have taken place in rural and medically underserved US regions, Burbank et al tested a mobile AAP mobile phone app for adolescents with persistent asthma in Arkansas [
Cafazzo et al [
Where possible, Cohen’s
Cafazzo et al recruited adolescents aged 12-16 years who were diagnosed with type 1 diabetes for more than 1 year, were receiving care at one clinic for at least 6 months, had a hemoglobin A1c (HbA1c) level between 8 and 10%, and were able to communicate in English [
All four apps focused on management of a chronic condition. Cafazzo et al’s [
There was variation in whether apps were delivered as a stand-alone resource or whether they were used alongside other components of a medical intervention. Burbank et al [
There was also variation in whether adolescents received guidance in using the app or whether they were solely responsible for navigating the app. Aldiss et al [
Study quality assessment for each study.
Quality assessment items | Study and quality assessment scorea | |||
|
Aldiss [ |
Burbank [ |
Cafazzo [ |
Frøisland [ |
Selection bias | Weak | Weak | Weak | Weak |
Study design | Moderate | Weak | Moderate | Weak |
Confounders | Weak | Weak | Weak | Weak |
Blinding | Moderate | Weak | Weak | Weak |
Data collection method | Moderate | Strong | Strong | Weak |
Withdrawals and dropouts | Moderate | Strong | Moderate | Strong |
aItems were scored using the Effective Public Health Practice Project (EPHPP) quality assessment tool.
Across the four papers, there was moderate-to-good agreement between the raters on the domains of the EPHPP quality assessment tool, with agreement on 17 of the 24 domains (71% agreement; unweighted κ = .60, 95% CI .33-.87; weighted κ = .71, 95% CI .49-.93). There was excellent agreement on the total domain scores for the four papers (ICC = .83, 95% CI -.17 to .99). The wide confidence intervals reflected the small number of papers assessed.
Only two papers provided data for calculating effect sizes and both featured mobile phone apps to help adolescents with type 1 diabetes. Both papers reported small sample sizes; Cafazzo et al [
Since mobile phones and tablets as the new generation of mobile electronic devices are a consumer product with custom software programs called apps, it is perhaps especially important to take into account adolescents’ attitudes toward using apps developed to help manage their chronic conditions. While we know mobile apps in general are widely accepted by adolescents living in today’s technology-rich environment, we were interested to know whether adolescents had been involved in the development of apps included in this review that were aimed at them and their peers. Three of the four studies we reviewed reported some level of patient involvement in the design, development, and/or evaluation of the app (see
Involvement of adolescents and their families at different stages of mobile app design, development, and evaluation.
Study | Involvement of adolescents and families at each stage | ||
|
Design | Development | Evaluation |
Aldiss et al [ |
Identified the symptoms to be assessed and addressed self-care advice | Tested the symptom report system | Reported perceptions of intervention and reviewed the self-care advice |
Burbank et al [ |
Not reported | Not reported | Not reported |
Cafazzo et al [ |
Requirements were obtained through qualitative interviews and focus group sessions | Themes were derived from focus group sessions, which were incorporated into the prototype app | Not reported |
Frøisland et al [ |
Adolescents suggested improvements for a future app | Not reported | Technical problems were reported, along with improvements of the existing app |
Adolescents were important contributors in the development of ASyMS, an advanced symptom management system utilizing mobile phone technology for adolescents with cancer [
Similarly, in Cafazzo et al’s pilot study [
Frøisland et al [
Appropriate clinical expertise in the specific condition was evident in all four of the research teams, although it was not always clear whether specific clinical (as opposed to academic) expertise in working with adolescents was represented. Health professional involvement in the modelling stages was reported in three of the four studies [
This review clearly demonstrates that despite the large number of health care apps in existence, the evidence base for their benefits for adolescents in personal management of their chronic physical health conditions is limited. The additional contextual challenge is that manufacturers are readily developing apps which are not based on empirical evidence [
As increasing numbers of adolescents with chronic conditions have transferred to adult-centered care, significant declines in treatment adherence have been observed during adolescence and the transition period [
Although these studies [
Parents and carers also play a significant part in promoting the development of adolescents’ personal management skills in chronic conditions [
None of the studies reviewed specifically considered the impact of the technology on parents and their role in the development of adolescents’ self-management skills. Similarly, although the mutual benefits of participatory design for both end users and developers were highlighted in the papers, there was little discussion about the implications of app usage on care delivery for health professionals. In one study, there was a mandatory consultation with a health professional to discuss the app [
Due to resource limitations, this review excluded mental health conditions and learning disabilities and was only able to consider the three common chronic physical conditions reported in the included studies—asthma, cancer, and diabetes. There is, however, emerging data in the area of mental health care, albeit not specifically targeting adolescents, reporting that the majority of existing apps for mental health care lack scientific evidence about their efficacy [
While we used a recognized tool to assess the quality of the four studies, it did not consider factors such as user and health professional involvement in intervention, yet these factors are recognized as being increasingly important in the development and evaluation of complex interventions.
The small evidence base identified by our review emphasizes the need for future high-quality, sophisticated trials in the area of app development for adolescents with chronic conditions, and the total sample size of 46 participants limits generalizability of the findings. The dearth of existing evidence prevented us from commenting on the effectiveness of mobile apps designed to support adolescents’ management of their physical conditions, as had been the objective at the outset. This in itself was an important finding and generated stimulating discussion around what the next step should be, from a multi-professional expert audience at an international conference where preliminary findings from this review were reported [
Our rationale for using the WHO definition of young people aged 10-24 years was based on the specific developmental implications of this age group, which differ considerably from those of younger children and older adults; this meant that papers which were otherwise potentially relevant needed to be excluded. For example, of the 19 full-text papers considered for eligibility, the age range of reported populations for some studies fell outside our predetermined definition. Where we were unable to extract specific data that related to 10-24-year-old participants, either from the papers themselves or by communication with the original authors, we excluded those studies from our analysis. While it is outside the scope of this review to return to the search stage and revise the inclusion and exclusion criteria, it is a consideration for a future review.
Information on adolescents who decline to take part in studies is of major importance in research with this age group, particularly as research nonrespondents have been reported to have poorer health outcomes than those who do respond. In addition, identified barriers to the use of mobile technologies by adolescents, including the disparity of access to mobile devices and the potential for habituation, suggest that the use of IT to address health issues may be limited or even harmful to adolescents [
Recommendations from our systematic review emphasize the value of a multidisciplinary team enabling expert clinical and patient involvement in the app design, development, and evaluation stages, as well as the involvement of technological and research personnel. It is also recommended that future trials are based on sound theory and are tested across age groups (where appropriate, and while recognizing the different developmental stages of adolescence), gender, and ethnicity. Future work should also carefully consider which primary and secondary outcomes are important to assess, and the best medium- to long-term follow-up times in context of the longevity and persistence of any behavior change observed. Furthermore, it is anticipated that initially developing a robust, adolescent-friendly app in one condition may have the additional advantage of subsequent adaptability and/or transferability across other conditions.
In conclusion, a key finding of the review is the paucity of evidence-based apps that exist in contrast to the thousands of apps available on the app market that are not evidence-based or user and professional informed. Although we aimed to assess the effectiveness of apps, the dearth of studies meeting our criteria meant that we were unable to be conclusive in this regard. This review provides valuable findings and paves the way for future rigorous development and evaluation of health apps for adolescents with chronic conditions. There remains a need for a phased approach to well-designed trials of mobile phone and tablet apps which resonate with the lives of adolescents, that can be feasibly transferred into real-life settings and which involve adolescents, parents, and health professionals in their design, development, and evaluation. Based on the available evidence, apps may be considered feasible health interventions, but more studies involving larger sample sizes, and with patient and health professional input at all stages, are needed to determine apps’ acceptability and effectiveness.
Excluded studies and reasons for exclusion.
asthma action plan
Advanced Symptom Management System
Advanced Symptom Management System for young people
Cumulative Index to Nursing and Allied Health Literature
condition
Effective Public Health Practice Project
female
hemoglobin A1c (glycated hemoglobin)
intraclass correlation coefficient
intervention duration
Institute of Medical Science
Institute of Medical Science Institute of Healthcare Informatics
information technology
Life Situation Scale for Adolescents
not applicable
National Health Service
National Institute for Health Research
population or participant, intervention or indicator, comparator or control, outcome, and study design
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
international prospective register of systematic reviews
quality of life
randomized controlled trial
short message service
sample size
State-Trait Anxiety Inventory
University of Manchester Research Institute
wideband code-division multiple access
World Health Organization
This review includes independent research supported by (or funded by) the National Institute for Health Research Biomedical Research Unit Funding Scheme. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the National Institute for Health Research, or the Department of Health. The authors of this review would also like to thank the University of Manchester Research Institute (UMRI) Pump Priming Programme for providing funding to support the project and we gratefully acknowledge the contributions of Professor Helen Foster and Mrs Sharon Douglas to the project.
None declared.