Mobile health (m‐health) smartphone interventions for overweight or obese adolescents and adults

Objectives

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

To assess the effects of smartphone‐based m‐health interventions for overweight or obese adolescents and adults.

Background

Description of the condition

Obesity is considered to be a risk factor for various diseases, leading to considerable health effects and costs. Worldwide obesity has almost tripled since 1975 (WHO 2018). In 2016, according to global estimates, 39% of adults were overweight or obese (of these, 26% were overweight and 13% were obese), and 18% of children and adolescents aged 5 to 19 were overweight or obese. The main factors influencing the increasing prevalence of obesity are, amongst others, the oversupply of energy‐dense foods and a decrease in physical activity. This applies not only to high‐income nations, but increasingly to low‐ and middle‐income countries, where, in recent decades, traditional eating habits, and working and living conditions have changed into a more Westernised lifestyle, especially in the urban setting (WHO 2018).

Overweight and obesity are usually measured by using the body mass index (BMI), which is a person´s weight in kilograms divided by the square of their height in meters (kg/m²). BMI is considered the most effective population‐level measure, as it applies to all sexes and ages in adults. However, the amount and distribution of body fat is sometimes considered a more precise marker of individual health risks (Gallagher 2000). For example, increased abdominal fat has been identified as a special cardiometabolic risk factor, and is measured more accurately by waist‐to‐height ratio (Ashwell 2012; Lee 2008). On the individual level, BMI is often considered a proxy measure, as it neither permits conclusions about a person´s fat‐to‐muscle ratio nor provides information on the person's distribution of body fat.

According to the World Health Organization (WHO), the current international BMI cutoff points for adults are as follows:

• 18.5 to 24.9 kg/m² (normal range)

• 25 to 29.9 kg/m² (overweight)

• 30 to 34.9 kg/m² (obesity class I)

• 35 to 39.9 kg/m² (obesity class II)

• 40 kg/m² and above (obesity class III).

For some ethnic populations, e.g. Asians, lower BMI thresholds have been discussed (Kanazawa 2005; Low 2009; WHO 2004).

For children aged between 5 and 19 years, overweight and obesity are usually defined as follows:

• overweight is BMI‐for‐age greater than one standard deviation; and

• obesity is more than two standard deviations above the WHO growth reference median (WHO 2007).

Obesity is considered a risk factor associated with several diseases, predominantly coronary heart disease, ischaemic stroke, and type 2 diabetes, but also osteoarthritis and some cancers (Ashwell 2012; WHO 2018). Evidence of associated risk is available for cancers of digestive organs, endometrial and breast cancer in women, kidney cancer, and multiple myeloma (Kyrgiou 2017; Lauby‐Secretan 2016).

In the past 15 years, different conclusions have been drawn regarding the risk of mortality associated with obesity. Berrington de Gonzalez 2010 pooled 19 prospective cohort studies, which included 1.46 million non‐Hispanic white adults, and found a J‐shaped association between BMI and all‐cause mortality, concluding that both excess weight and underweight increased the risk of death. Similar results were obtained in an individual participant data meta‐analysis by the Global BMI Mortality Collaboration (Global BMI Mortality Collaboration 2016). It concluded that all‐cause mortality was minimal at a BMI between 20 kg/m² to 25 kg/m² and increased in people with BMIs both below 20 kg/m² and throughout the overweight range. For a BMI above 25 kg/m², the increase in overall mortality was approximately log‐linear.

Other analyses report conflicting results. McGee 2005 pooled 26 observational studies with 389,000 people from many different national and ethnic groups. They found excess mortality among people with obesity, but not among overweight people. Flegal 2013 pooled 97 prospective studies with 2.88 million adults. They concluded that obesity class II and class III were both associated with higher all‐cause mortality, but class I obesity was not. Overweight was associated with a lower all‐cause mortality than normal weight. The most extensive analysis on the topic to date, although in a specific population, is a South Korean prospective cohort study with 12.8 million Korean adults (Yi 2015). It concluded that mortality risk associated with BMI depended on sex and age. For men, the optimal BMI was 23 kg/m² to 26 kg/m² at 18 years to 34 years; 24 kg/m² to 28 kg/m² at 45 years to 54 years; and 25 kg/m² to 29 kg/m² at 65 years to 74 years. Among women, it was 15.5 kg/m² to 25 kg/m² at 18 years to 34 years; 21 kg/m² to 27 kg/m² at 45 years to 54 years; and 24 kg/m² to 29 kg/m² at 65 years to 74 years. Again, these results suggest that being overweight tends to be a protective factor rather than a risk factor, especially with increasing age.

The fundamental problem of these observational studies is that they can only describe associations but are not able to establish causality. Confounders, such as socioeconomic status, must be taken into account; none of the above mentioned analyses adjusted adequately for them (Broady 2015; Corsi 2019; Hoebel 2019; Hwang 2019; McLaren 2007). Finally, some researchers emphasise that cardiorespiratory fitness, at every body size, is a more important health factor than body weight (Barry 2014; Ekelund 2019; Gaesser 2015; Lavie 2016, Ross 2015).

In addition to being at risk for adverse health outcomes, obese people suffer societal discrimination and stigma across multiple domains, such as in employment, education, health care, interpersonal relationships, and in the mass media (Puhl 2009). Obesity should not be oversimplified as the result of an individual´s behaviour and attributed lack of control. An individual's psychosocial, financial, and structural environments play key roles in creating or sustaining obesogenic behaviours (Kirk 2010; Lakerveld 2017; Marks 2015). Change involves the affected individuals, but on the interpersonal level, also involves their healthcare providers, employers, and social networks, who can support them by providing a more respectful and less blaming environment (Marks 2015). Environmental factors that support or hinder healthy behaviours, such as taxation of unhealthy foods, food advertisement control, investment in active transportation infrastructure, and regulation of the nutritional environment at school and work, must also be addressed if obesity is to be tackled in a holistic way (Roberto 2015; Schwartz 2017; Vallgarda 2015; Wolfenden 2018; Wolfenden 2020).

Description of the intervention

Interventions that target overweight or obesity can aim at environmental or individual factors. As described in the previous section, environmental approaches seem as important as individual approaches, since they can support the prevention of conditions and circumstances leading to overweight or obesity and associated stigmata, which are beyond an individual´s control. This review focusses on an individual approach, by examining the effectiveness of a smartphone‐delivered behaviour change intervention.

Behaviour change is a challenging process. People trying to increase physical activity, modify dietary habits, and achieve weight loss are constantly managing psychological tensions (Greaves 2017). Health behaviour is strongly influenced by the social environment throughout one's entire lifespan (Marks 2015). On the individual level, behaviour change interventions are key components in the treatment of obesity. To support overweight and obese people, a wide variety of multicomponent interventions have been devised and implemented (LeBlanc 2018; Olson 2017). Their goal is to change various determinants of behaviour, and ideally, they are based on theories that conceptualise behaviour change, i.e. explain how behaviours are developed, maintained, and influenced.

A multitude of behaviour change theories are available. They differ according to constructs, relationship between constructs, measurability and testability of constructs, incorporation of explanation, description of causality, generalisability, and evidence‐base (Davis 2015). The most prominent theories used in obesity intervention trials are the Transtheoretical model/Stages of change (Prochaska 1997; Prochaska 2015), Social cognitive theory (Bandura 1986; Bandura 2004), and the Theory of planned behaviour (Ajzen 1991).

Behaviour change interventions aimed at increasing physical activity, reducing energy intake, and achieving weight loss are usually multicomponent programmes, delivered face to face by specialised healthcare personnel. They include dietary counselling, physical activity, and behavioural strategies to increase adherence to goals, and have been shown to achieve 5% to 10% weight loss after 6 to 12 months (LeBlanc 2018; Lv 2017; Mastellos 2014; Olson 2017). The components of these interventions (education, self‐monitoring, feedback, problem‐solving of barriers to adherence, goal setting, social support) ideally contain specific, evidence‐based behaviour change techniques, which help to alter the regulation of behaviour (Michie 2013; Tate 2016).

Nowadays, smartphones and other mobile devices, usually worn close to the body (e.g. wearables), are believed to support healthy behaviours on the individual level, in a low‐threshold and cost‐effective manner (Free 2013; Vandelanotte 2016). Mobile health (m‐health) is defined as 'medical and public health practice supported by mobile devices, such as mobile phones, patient monitoring devices, personal digital assistants (PDAs), and other wireless devices' (WHO 2011). M‐health interventions are purported to promote self‐efficacy and adherence (e.g. by continuously recording health‐related information, keeping digital diaries, sending reminders). They are also thought to influence abilities and behaviour (e.g. through therapeutic suggestions, by providing actionable feedback or information on disease self‐management (Bradway 2017; Vandelanotte 2016; WHO 2011)). Thus, m‐health interventions can potentially encourage overweight and obese people to become and stay active, promote a change in dietary patterns, and lose weight. Proponents also propose that m‐health interventions are more cost‐effective than traditional in‐person weight loss programmes, and offer the possibility of low‐threshold, population‐wide access, including for hard to reach populations, such as people with a low socioeconomic background (Bradway 2017; Olson 2017; Vandelanotte 2016).

This review will include m‐health interventions developed for overweight or obese adolescents and adults, regardless of the intervention´s main aim (e.g. increasing physical activity, reducing sedentary behaviour, promoting cardiorespiratory fitness, inducing weight loss, and transitioning to a healthier diet) delivered via an application (app) installed on a smartphone. As current m‐health interventions are often multimodal, and can also include in‐person components, we will consider those interventions that use the smartphone as the main intervention component. We define this to be the case if the smartphone is used from the beginning to the end of the intervention, and is the main access point for participant interaction. Concomitant in‐person components are possible, but should not be the dominating mode of intervention delivery.

Adverse effects of the intervention

Behaviour change interventions delivered via an application on a smartphone could potentially induce feelings of disappointment, failure, and lack of self‐efficacy when participants are unable to achieve their goals or sustain a newly acquired health behaviour. With respect to weight loss, in addition to potentially being ineffective in the long term, dieting has been associated with weight cycling (Gaesser 2015; Montani 2015). This phenomenon of episodic variation in body weight over a period of time has been discussed as a risk factor that influences morbidity and mortality, but current evidence supporting this association is sparse (Mehta 2014).

Potential risks attributed specifically to m‐health interventions can be classified as technical (data security and data privacy‐related, malfunction, usability problems), clinical (decrease in quality of care), or inadequate content (lack of updates, lack of evidence‐based functions or information), all potentially resulting in ineffective or harmful treatment (Albrecht 2016; Bradway 2017). Furthermore, following the NICE Evidence Standards for Digital Health Technologies, m‐health interventions can imply a higher risk if:

• they are aimed at vulnerable populations, such as children or at‐risk adults;

• consequences to the consumer are serious should the intervention fail;

• they are intended to be used by consumers on their own (without support from qualified health care professionals);

• they include machine learning algorithms or artificial intelligence;

• they imply high financial or organisational risks (such as complex changes in working practice or care pathways (NICE 2019)).

Why it is important to do this review

The rationale for focusing on m‐health as the main intervention component for overweight and obesity management is to be able to evaluate the effectiveness of this specific mode of intervention delivery. To date, it is unclear whether m‐health interventions are theory‐based, and which components, namely behaviour change techniques, these interventions actually comprise (Free 2013; Hutchesson 2015; Vandelanotte 2016).

Evidence from Cochrane Reviews related to the topic of this review can be summarised as follows: eight years ago, de Jongh and colleagues evaluated the effectiveness of mobile phone messaging for facilitating self‐management of long‐term illnesses. They found a very limited number of studies, and concluded that further research was needed (de Jongh 2012). In the same year, Wieland and colleagues investigated interactive computer‐based interventions for weight loss and weight maintenance. They found them to be effective, but less so than in‐person interventions, and pointed out that the clinical significance of this difference was unclear (Wieland 2012). Two more recent reviews evaluating m‐health interventions, albeit for chronic obstructive pulmonary disease (McCabe 2017), and smoking cessation (Whittaker 2019), found a certain benefit of m‐health interventions compared to standard interventions, but noted that the evidence base was of poor quality and of very low‐certainty, respectively. A recent qualitative evidence synthesis on perceptions and experiences of targeted digital communication accessible via mobile devices, provided a summary of participants' views on acceptability and preferences, technical access, content preferences, confidentiality, and programme impact (Ames 2019).

The aim of this review is to assess the effectiveness of delivering a behaviour change intervention for overweight or obese people mainly via a smartphone‐based application. This is important as other systematic reviews on the topic to date, produced outside of Cochrane, are characterised by a striking lack of detail in their data extraction. This could also be due to a lack of reporting in original m‐health studies, which has been addressed in the CONSORT‐EHEALTH framework (Eysenbach 2011). As a result, most of the systematic reviews to date have not been able to either adequately establish a fair comparison of interventions, or to assess the effectiveness of specific intervention components, modalities, and intensity, resulting in syntheses of questionable validity. A notable exception with regard to data extraction and pooling of comparable interventions is the review by Hutchesson and colleagues, which had a broader perspective and evaluated e‐health interventions for overweight or obesity, i.e. interventions which could be delivered via the internet, computers, tablets, or smartphones (Hutchesson 2015). Their analyses indicated that e‐health interventions achieved a modest weight loss compared to no or minimal treatment (mean difference in weight loss of 1.4 kg and 2.7 kg), but that the clinical significance was inferior to traditional behavioural weight loss interventions, where participants lose an average of 10.7 kg. They highlighted the diversity of technologies used, the lack of adequate description of intervention components, and the challenge in determining their effectiveness.

Due to the high degree of uncertainty in this area, as well as the potential for use on a population‐wide scale, an up‐to‐date, state of the art Cochrane Review is needed. It will evaluate m‐health interventions based on a careful and detailed data extraction, as well as make use of a contemporary framework for assessing m‐health interventions (NICE Evidence Standards Framework for Digital Health Technologies (NICE 2019)).

Objectives

To assess the effects of smartphone‐based m‐health interventions for overweight or obese adolescents and adults.

Methods

Criteria for considering studies for this review

Types of studies

We will include randomised controlled trials (RCTs).

Types of participants

We will include adolescents (aged 13 years to 17 years) and adults (≥ 18 years) who are overweight or obese, regardless of comorbidities (e.g. diabetes, hypertension, or dyslipidaemia). If children younger than 13 years are included in the study, we will only include the study if the mean age of participants is over 13.

Diagnostic criteria for overweight and obese adults

We will define adults with a body mass index (BMI) between 25 kg/m² and 29.9 kg/m² as overweight and those with a BMI 30 kg/m² or greater as obese.

Diagnostic criteria for overweight and obese adolescents

We will use the World Health Organization (WHO) growth reference, which provides BMI‐for‐age (5 to 19 years) tables, with percentiles and z‐scores (WHO 2007).

Types of interventions

We plan to investigate the following comparisons of intervention versus control or comparator.

Intervention

Smartphone application (m‐health intervention)

a) as a sole mode of delivery;

b) as part of a multimodal behavioural intervention, where the smartphone is used from the beginning to the end of the intervention, and is the main access point for patient interaction (concomitant in‐person components are possible, but should not be the dominating mode of intervention delivery).

Comparisons

• Usual care

• A different smartphone application

• Any other active intervention (e.g. face‐to‐face counselling or multimodal behavioral interventions)

• The same smartphone application with a different intensity (dosage) or partially different components

• No intervention

Concomitant interventions will have to be the same in both the intervention and comparator groups to establish fair comparisons. If a study includes multiple arms, we will include any arm that meets the review inclusion criteria.

Minimum duration of intervention

We will include interventions of any duration.

Minimum duration of follow‐up

Minimal duration of follow‐up will be three months, as outcomes should be assessed after a meaningful period of time to be able to establish efficacy.

We will define any follow‐up period going beyond the original time frame for the primary outcome measure as specified in the power calculation of the trial's protocol as an extended follow‐up period (usually called 'open‐label extension study' (Buch 2011; Megan 2012)).

Summary of specific exclusion criteria

We will exclude:

• studies including children aged 12 or less;

• studies including pregnant women;

• studies where the primary aim is the management of concomitant medical conditions (e.g. diabetes, hypertension, or dyslipidaemia); and

• studies where m‐health interventions are:

  • based solely on text messages delivered to a mobile phone;

  • based on devices that do not connect to an application on a smartphone (such as wearables alone);

  • embedded into multimodal interventions, in which the smartphone is not the main participant access point for an interaction.

Types of outcome measures

We will not exclude a study if it fails to report one or several of our primary or secondary outcome measures. If none of our primary or secondary outcomes are reported in the trial, we will not include the trial but provide some basic information in the 'Characteristics of studies awaiting classification' table.

We will investigate the following outcomes using the methods and time points specified below.

Primary outcomes

• Change in physical activity

• Change in anthropometric measures

• Adverse events

Secondary outcomes

• Health‐related quality of life

• Self‐efficacy

• Well‐being

• Participant satisfaction

• Adherence or engagement

• Change in dietary behaviour

• Evidence for effectiveness

Method of outcome measurement

• Change in physical activity, as measured in included studies, e.g. number of days per month of physical activity, movement or step counts, cardiorespiratory fitness, heart rate, or via instruments, such as the International Physical Activity Questionnaire (IPAQ).

• Change in anthropometric measures, such as BMI, BMI percentiles, body weight, percentage of body fat, waist or hip circumference, waist‐to‐hip ratio.

• Adverse events, such as feelings of disappointment, failure.

• Health‐related quality of life, evaluated by validated instruments, such as the Paediatric Quality of Life Inventory, 36‐, 12‐, or 6‐item Short Form Health Survey (SF‐36; SF‐12, SF‐6D), EQ‐5D, WHO‐Quality of life‐BREF (WHOQOL‐BREF).

• Self‐efficacy, evaluated by validated instruments, such as the General Self‐Efficacy Scale (GSES), Self Efficacy Survey (SES), Strenghts Self‐Efficacy Scale (SSES), Self‐Efficacy Scale for Exercise (SEE).

• Well‐being, evaluated by validated instruments, such as the Quality of Well‐being Scale, Satisfaction with Life Scale, Positive and Negative Affect Scale.

• Participant satisfaction (includes similar outcomes, such as 'participant acceptability' and 'patient experience'), as measured and defined in included trials, e.g. measured via surveys or interviews.

• Adherence or engagement, such as frequency and duration of interactions, number of features accessed.

• Change in dietary behaviour, such as higher intake of vegetables and fruits, decreased intake of high caloric foods and sweetened beverages.

• Evidence for effectiveness, based on the effectiveness standards extracted from NICE 2019.

Timing of outcome measurement

• Adverse events: at any point after start of intervention or baseline assessment (if the start of the intervention cannot be determined).

• All other outcomes: short‐term (three to six months), medium‐term (6 to 12 months), long‐term (over 12 months). We will not extract data for time points before three months.

Search methods for identification of studies

Electronic searches

To identify relevant studies, we will conduct a search from 2007 onwards, as this was the year the first smartphone performing computing functions with internet access and operating system capable of running third‐party applications was introduced. We will search the following sources, and will place no restrictions on the language of publication:

• Cochrane Central Register of Controlled Trials (CENTRAL);

• MEDLINE Ovid;

• PsycINFO Ovid;

• CINAHL (Cumulative Index to Nursing and Allied Health Literature);

• LILACS (Latin American and Caribbean Health Sciences Literature);

• ClinicalTrials.gov (www.clinicaltrials.gov);

• World Health Organization International Clinical Trials Registry Platform (ICTRP; www.who.int/trialsearch/).

We will not include Embase in our search, as RCTs indexed in Embase are now prospectively added to CENTRAL via a highly sensitive screening process (Cochrane 2020). For detailed search strategies, see Appendix 1.

Searching other resources

We will try to identify other potentially eligible studies or ancillary publications by searching the reference lists of included studies, systematic reviews, meta‐analyses, and health technology assessment reports identified by our search. In addition, we will contact the authors of included studies to obtain additional information.

We will not use abstracts or conference proceedings for data extraction because this information source does not fulfil the CONSORT requirements, which consist of 'an evidence‐based, minimum set of recommendations for reporting randomized trials' (CONSORT 2020; Scherer 2018). We will present information on studies reported solely as abstracts or conference proceedings in the 'Characteristics of studies awaiting classification' table.

We define grey literature as records from ClinicalTrials.gov or WHO ICTRP, as well as dissertations identified via the electronic searches detailed above (e.g. via PsycINFO or CINAHL). We will not search additional sources for other types of grey literature.

Data collection and analysis

Selection of studies

Two review authors (MIM, LSW) will independently screen the abstract, title, or both, of every record retrieved by the literature searches, to determine which studies we should assess further. We will obtain the full‐text of all potentially relevant records, and independently screen them to select studies for inclusion. We will resolve disagreements through consensus or by recourse to a third review author (BR). If we cannot resolve a disagreement, we will categorise the study as a 'Study awaiting classification' and will contact the study authors for clarification. We will present an adapted PRISMA flow diagram to show the process of study selection (Liberati 2009). We will list all articles excluded after full‐text assessment in a 'Characteristics of excluded studies' table, and will provide the reasons for exclusion.

Data extraction and management

For studies that fulfil our inclusion criteria, two review authors (MIM, LSW) will independently extract key participant and intervention characteristics. We will describe interventions according to the 'template for intervention description and replication' (TIDieR) checklist (Hoffmann 2014; Hoffmann 2017). In addition, we will use a matrix detailing the effectiveness standards extracted from NICE 2019 (Appendix 2).

We will report data on efficacy outcomes and adverse events using standardised data extraction sheets from the Cochrane Metabolic and Endocrine Disorders (CMED) Group. We will resolve disagreements by discussion, or, if required, by consultation with a third review author (BR).

We will provide information, including the study identifier, for potentially relevant ongoing trials in the 'Characteristics of ongoing trials' table and in a joint appendix 'Matrix of study endpoint (publications and trial documents)'. We will try to find the protocol for each included study, and in a joint appendix, will report primary, secondary, and other outcomes in comparison with data in publications.

We will email all authors of included studies to enquire whether they would be willing to answer questions regarding their studies. We will present the results of this survey in an appendix. We will thereafter seek relevant missing information on the study from the primary study author(s), if required.

Dealing with duplicate and companion publications

In the event of duplicate publications, companion documents, or multiple reports of a primary study, we will maximise the information yield by collating all available data, and we will use the most complete data set aggregated across all known publications. We will list duplicate publications, companion documents, multiple reports of a primary study, and trial documents of included trials (such as trial registry information) as secondary references under the study ID of the included study. Furthermore, we will list duplicate publications, companion documents, multiple reports of a study, and trial documents of excluded trials (such as trial registry information) as secondary references under the study ID of the excluded study.

Data from clinical trials registers

If data from included trials are available as study results in clinical trials registers, such as ClinicalTrials.gov or similar sources, we will make full use of this information and extract the data. If a full publication of the study is also available, we will collate and critically appraise all available data. If an included study is marked as a completed study in a clinical trials register but no additional information (study results, publication, or both) is available, we will add this study to the 'Characteristics of studies awaiting classification' table.

Assessment of risk of bias in included studies

Two review authors (MIM, LSW) will independently assess the risk of bias for each included study. We will resolve disagreements by consensus, or by consulting a third review author (BR). If adequate information is unavailable from the publications, trials protocols, or other sources, we will contact the study authors for more detail, and to request missing data on 'Risk of bias' items.

We will use the Cochrane 'Risk of bias 2' (RoB 2) tool (Higgins 2019a; https://archie.cochrane.org/sections/documents/RoB 2 tool).

We will focus on the assessment of the effect of assignment to the interventions at baseline. The https://archie.cochrane.org/sections/documents/RoB 2 tool evaluates the following domains:

• bias arising from the randomisation process;

• bias due to deviations from the intended interventions;

• bias due to missing outcome data;

• bias in measurement of the outcome; and

• bias in selection of the reported results.

Within each domain, signalling questions provide information about features of the study that are relevant to risk of bias. Possible answers to the signalling questions are 'yes', 'probably yes', 'probably no', 'no' and 'no information'. After answering the signalling questions, we will make a judgement of the risk of bias, assigning one of three levels ('low risk of bias', 'some concerns', 'high risk of bias') to each domain.

For each specific outcome, we will establish an overall 'Risk of bias' judgement using the following criteria.

• Low risk of bias: the study is judged to be at low risk of bias for all domains for this result.

• Some concerns: the study is judged to raise some concern in at least one domain for this result, but not to be at high risk of bias for any domain.

• High risk of bias: the study is either judged to be at high risk of bias in at least one domain for this result, or the study is judged to have some concerns for multiple domains in a way that substantially lowers confidence in the result.

We will distinguish between participant‐reported outcomes, observer‐reported outcomes not involving judgement, observer‐reported outcomes involving some judgement, outcomes reflecting decisions made by intervention providers, and composite outcomes.

• Participant‐reported outcomes: change in physical activity (if self‐reported), change in anthropometric measures (if self‐reported), change in dietary behaviour (if self‐reported), adherence or engagement (if self‐reported), health‐related quality of life, well‐being, participant satisfaction, adverse events.

• Observer‐reported outcomes not involving judgement: change in physical activity (if measured by application); change in anthropometric measures (if measured by investigators); adherence or engagement (if measured by application).

• Observer‐reported outcomes involving some judgement: none.

• Outcomes reflecting decisions made by intervention providers: none.

• Composite outcomes: evidence for effectiveness (as assessed by applying the NICE 2019 framework to effectiveness standards.

Measures of treatment effect

We will express dichotomous data as a risk ratio (RR) or an odds ratio (OR) with 95% confidence intervals (CIs); we will express continuous data as the mean difference (MD) or standardised mean difference (SMD) with 95% CIs. When combining data across studies for continuous outcomes measured on the same scale (e.g. weight loss in kg), we will estimate the intervention effect using the MD with 95% CIs. When combining data across studies for outcomes that measure the same underlying concept (e.g. health‐related quality of life) but use different measurement scales, we will calculate the SMD with 95% CI. We will express time‐to‐event data as a hazard ratio (HR) with 95% CIs.

Unit of analysis issues

We will take into account the level at which randomisation occurred, such as cross‐over studies, cluster‐randomised trials, and multiple observations for the same outcome. If more than one comparison from the same study is eligible for inclusion in the same meta‐analysis, we will either combine groups to create a single pair‐wise comparison, or we will appropriately reduce the sample size so that the same participants do not contribute data to the meta‐analysis more than once (splitting the 'shared' group into two or more groups). Although the latter approach offers some solution for adjusting the precision of the comparison, it does not account for correlation arising from inclusion of the same set of participants in multiple comparisons (Higgins 2019b).

If we identify eligible cross‐over studies, we will only consider the first period of the study, as carry‐over effects are likely. However, we will only include the study if outcomes were measured at three months before participants were switched to the subsequent intervention (Higgins 2019c).

We will attempt to re‐analyse cluster‐RCTs that have not appropriately adjusted for potential clustering of participants within clusters in their analyses. Variance of the intervention effects will be inflated by a design effect. Calculation of a design effect involves estimation of an intracluster correlation coefficient (ICC). We will obtain estimates of ICCs by contacting study authors, or by imputing ICC values using either estimates from other included studies that report ICCs or external estimates from empirical research (e.g. Bell 2013). We plan to examine the impact of clustering by performing sensitivity analyses.

Dealing with missing data

If possible, we will obtain missing data from the authors of included studies. We will carefully evaluate important numerical data, such as screened, and randomly assigned participants, as well as intention‐to‐treat and as‐treated and per‐protocol populations. We will investigate attrition rates (e.g. dropouts, losses to follow‐up, withdrawals), and we will critically appraise issues concerning missing data and use of imputation methods (e.g. last observation carried forward).

For studies in which the standard deviation (SD) of the outcome is not available at follow‐up or we cannot re‐create it, we will standardise by the mean of the pooled baseline SD from studies that reported this information.

When included studies do not report means and SDs for outcomes, and we do not receive requested information from study authors, we will impute these values by estimating the mean and the variance from the median, the range, and the size of the sample (Hozo 2005).

We will investigate the impact of imputation on meta‐analyses by performing sensitivity analyses, and for every outcome, we will report which studies had imputed SDs.

Assessment of heterogeneity

In the event of substantial clinical or methodological heterogeneity, we will not report study results as the pooled effect estimate in a meta‐analysis.

We will identify heterogeneity (inconsistency) by visually inspecting the forest plots, and by using a standard Chi² test with a significance level of α = 0.1 (Deeks 2019). In view of the low power of this test, we will also consider the I² statistic, which quantifies inconsistency across studies, to assess the impact of heterogeneity on the meta‐analysis (Higgins 2002; Higgins 2003).

When we find heterogeneity, we will attempt to determine possible reasons for this by examining individual study and subgroup characteristics.

Assessment of reporting biases

If we include 10 or more studies that investigate a particular outcome, we will use funnel plots to assess small‐study effects. Several explanations may account for funnel plot asymmetry, including true heterogeneity of effect with respect to study size, poor methodological design (and hence bias of small studies), and publication bias (Page 2019). Therefore, we will interpret the results carefully (Sterne 2011).

Data synthesis

We plan to undertake (or display) a meta‐analysis only if we judge participants, interventions, comparisons, and outcomes to be sufficiently similar to ensure an answer that is clinically meaningful. Unless good evidence shows homogeneous effects across studies of different methodological quality, we will primarily summarise low risk of bias data using a random‐effects model (Wood 2008). We will interpret random‐effects meta‐analyses with due consideration for the whole distribution of effects, and will present a prediction interval (Borenstein 2017a; Borenstein 2017b; Higgins 2009). A prediction interval requires at least three studies to be calculated, and specifies a predicted range for the true treatment effect in an individual study (Riley 2011). For rare events, such as event rates below 1%, we will use the Peto odds ratio method, provided there is no substantial imbalance between intervention and comparator group sizes, and intervention effects are not exceptionally large. In addition, we will perform statistical analyses according to the statistical guidelines presented in the Cochrane Handbook for Systematic Reviews of Interventions (Deeks 2019).

Subgroup analysis and investigation of heterogeneity

We expect the following characteristics to introduce clinical heterogeneity, and we plan to carry out the following subgroup analyses including investigation of interactions (Altman 2003).

• Gender

• Overweight, obesity class I, II, and III

• Ethnicity

• Age groups (adolescents, adults)

• Educational level (low, medium, high)

• Sole m‐health versus combined m‐health and in‐person interventions

• Dosage, according to data

• Adherence or engagement, according to data

• Region (low‐, middle‐ or high‐income countries)

If results permit, equity‐specific intervention effects will be examined by gender, educational level, and region (Czwikla 2019).

Sensitivity analysis

We plan to perform sensitivity analyses to explore the influence of the following factors (when applicable) on effect sizes, by restricting analysis to the following.

• Published studies

• Effect of risk of bias, as specified in the Assessment of risk of bias in included studies section

• Very long or large studies, to establish the extent to which they dominate the results

• Cluster‐randomized studies

We will use of the following filters, if applicable: diagnostic criteria, imputation used, language of publication (English versus other languages), source of funding (industry versus other), or country (depending on data).

We will also test the robustness of results by repeating analyses using different measures of effect size (i.e. RR, OR, etc.) and different statistical models (fixed‐effect and random‐effects models).

Summary of findings and assessment of the certainty of the evidence

We will present the overall certainty of the evidence for each outcome specified below, according to the GRADE approach, which takes into account issues related not only to internal validity (risk of bias, inconsistency, imprecision, publication bias), but also to external validity, such as directness of results (Schünemann 2013). Two review authors (MIM, LSW) will independently rate the certainty of the evidence for each outcome. We will resolve differences in assessment by discussion or by consultation with a third review author (BR).

We will include an appendix entitled 'Checklist to aid consistency and reproducibility of GRADE assessments', to help with standardisation of the 'Summary of findings' tables (Meader 2014). Alternatively, we will use GRADEpro GDT software and will present evidence profile tables as an appendix (GRADEpro GDT). We will present results for outcomes as described in the Types of outcome measures section. If meta‐analysis is not possible, we will present the results in a narrative format in the 'Summary of findings' table. We will justify all decisions to downgrade the certainty of the evidence by using footnotes, and we will make comments to aid the reader's understanding of the Cochrane Review when necessary.

'Summary of findings' table

We will present a summary of the evidence in a 'Summary of findings' table. This will provide key information about the best estimate of the magnitude of effect, in relative terms and as absolute differences, for each relevant comparison of alternative management strategies, numbers of participants and studies that address each important outcome, and a rating of overall confidence in the effect estimates for each outcome. We will create the 'Summary of findings' table using the methods described in the Cochrane Handbook for Systematic Reviews of Interventions (Schünemann 2019), along with Review Manager 5 software (Review Manager 2014).

In the 'Summary of findings' table we will report on the interventions versus comparators as detailed in Types of interventions.

We will report the following outcomes, listed according to priority.

• Change in physical activity

• Change in anthropometric measures

• Health‐related quality of life

• Self‐efficacy

• Well‐being

• Evidence for effectiveness

• Adverse events.

History

Protocol first published: Issue 4, 2020

Notes

We based parts of the methods of this Cochrane Protocol on a standard template established by the CMED Group.

Appendices

Appendix 1. Search strategies

MEDLINE Ovid

1. Cell Phone/ 2. Smartphone/ 3. Mobile Applications/ 4. (app or apps).tw. 5. ((smartphone or mobile or phone) adj6 application*).tw. 6. ((cell or mobile or smart) adj2 phone*).tw. 7. (cellphone* or mobile or mobiles or smartphone*).tw. 8. (mobile adj2 (device* or technolog* or app*)).tw. 9. (mhealth or m‐health or mobile health).tw. 10. or/1‐9 11. Obesity/ 12. Overweight/ 13. Pediatric Obesity/ 14. Weight Loss/ 15. Weight Reduction Programs/ 16. Health Promotion/ 17. (overweight or obes*).tw. 18. (fitness or physical activity).tw. 19. (diet* or nutrition*).tw. 20. weight.tw. 21. or/11‐20 22. 10 and 21 [23‐33: Cochrane Handbook 2019 RCT filter ‐ sensitivity max. version (Lefebvre 2019)] 23. randomized controlled trial.pt. 24. controlled clinical trial.pt. 25. randomi?ed.ab. 26. placebo.ab. 27. drug therapy.fs. 28. randomly.ab. 29. trial.ab. 30. groups.ab. 31. or/23‐30 32. exp animals/ not humans/ 33. 31 not 32 34. 22 and 33 35. ("2007*" or "2008*" or "2009*" or "201*").dt. 36. 34 and 35 = 2075 [Test if strategy retrieves 21 relevant PubMed records known to the authors] 37. ("30779333" or "30576840" or "31486407" or "30381313" or "30361140" or "29233169" or "27818304" or "29074513" or "27039178" or "27098449" or "27573315" or "26995281" or "27335237" or "26530929" or "26076688" or "26048581" or "25919921" or "26229119" or "26033349" or "25157000" or "25402403").ui. = 21 38. 36 and 37 = 21

Appendix 2. NICE evidence standards framework for digital health technologies: evidence for effectiveness standards

Tier	Evidence category	Standard	Content
1	Credibility with UK health and social care professionals	Minimum	Plausible mode of action viewed as useful and relevant by professional experts, either health care professionals have been involved in design, development or testing of DHT or been involved in signing‐off and giving informed approval of DHT
		Best practice	Published or publicly available evidence documenting role of healthcare professionals in design, development, testing or sign‐off of DHT
	Relevance to current care pathways in the UK heath and social care system	Minimum	Evidence of successful piloting of DHT in the healthcare system showing its relevance to current care pathways and service provision, also evidence of performance in intended function to needed scale (server capacity)
		Best practice	Evidence of successful implementation of DHT in healthcare system
	Acceptability with users	Minimum	Show that representatives of intended user groups were involved in design, development or testing of DHT, provide data on user satisfaction with DHT
		Best practice	Published or publicly available evidence showing that representatives of intended user groups were involved in design, development or testing of DHT and show user satisfaction with DHT
	Equalities considerations	Minimum	Evidence that DHT contributes to challenging healthcare inequalities in healthcare system or improving access to care among hard‐to‐reach populations or contributes to equality and eliminating unlawful discrimination
		Best practice	Evidence of DHT being used in hard‐to‐reach populations
	Accurate and reliable measurements (if relevant)	Minimum	Data showing that DHT‐generated or recorded data are accurate, reproducible, relevant to expected range in target population, and data showing that DHT is able to detect clinically relevant changes or responses
		Best practice	Same as for minimum standard
	Accurate and reliable transmission of data (if relevant)	Minimum	Technical data showing that numerical, text, audio, image, graphic, or video information is not changed during transmission process nor biased by data 'value' expected from target user population
		Best practice	Same as for minimum standard, but with quantitative data
2	Reliable information content	Minimum	Show that health information provided is valid, accurate, up to date, reviewed by experts at defined intervals, comprehensive
		Best practice	Evidence of endorsement, accreditation or recommendation by healthcare system organisations, relevant professional bodies or patient organisations; alternatively validation or accreditation through independent certification (such as 'The information standard' or 'HONcode')
	Ongoing data collection to show use of the DHT	Minimum	Commitment of data collection showing use of DHT in target population and commitment to share with decision‐makers in clear and useful format
		Best practice	Evidence that data on DHT use is being collected and can be made available to decision makers
	Ongoing data collection to show value of the DHT	Minimum	Commitment of data collection to show user outcomes or user satisfaction and commitment to share with decision makers
		Best practice	Evidence that date on outcomes or user satisfaction is being collected and can be made available to decision makers
	Quality and safeguarding	Minimum	Show appropriate safeguarding measures around peer support and other communication functions, describe who, and provide reasons why they have access to platform, describe measures to ensure safety in peer to peer communication (e.g. user agreements or moderation)
		Best practice	Same as for minimum standard
3a	Demonstrating effectiveness	Minimum	High quality observational or quasi‐experimental studies demonstrating relevant outcomes (comparative data on relevant outcomes in control group, use of historical controls, routinely collected data); relevant outcomes (behavioural or condition‐related user outcomes, evidence of positive behaviour change, user satisfaction)
		Best practice	High quality intervention studies (quasi‐experimental or experimental) that incorporate a comparison group, showing improvement in relevant outcomes (PROs, other clinical measures, healthy behaviours, physiological measures, user satisfaction and engagement, healthcare use (such as admissions or appointments)); comparator should be reflective of standard of care
	Use of appropriate behaviour change techniques (if relevant)	Minimum	Show that techniques used are consistent with behaviour change theory, appropriate for target population
		Best practice	Published qualitative or quantitative evidence showing that techniques used are based on recognised behaviour change techniques, aligned with recommended practice, appropriate for target population
3b	Demonstrating effectiveness	Minimum	High quality intervention study (experimental or quasi‐experimental) showing improvement in relevant outcomes, such as diagnostic accuracy, PROs, clinical measures, healthy behaviours, physiological measures, user satisfaction and engagement; comparator should reflect standard care
		Best practice	High quality RCT or studies in relevant healthcare settings, comparing DHT with relevant comparator and demonstrating consistent benefit, including in clinical outcomes in target population, using validated, condition‐specific outcome measures; alternatively, well‐conducted meta‐analysis of RCTs on DHT

Notes: evidence standards of higher tiers must also meet the standards of the lower tiers.

DHT: digital health technology; PRO: participant‐reported outcome; RCT: randomised controlled trial.

Contributions of authors

All protocol authors contributed to, read and approved the final protocol draft.

Declarations of interest

Maria‐Inti Metzendorf (MIM): none known.

Lisa Susan Wieland (LSW): none known.

Bernd Richter (BR): none known.

New