This web page was created programmatically, to learn the article in its authentic location you possibly can go to the hyperlink bellow:
https://www.jmir.org/2025/1/e71548
and if you wish to take away this text from our web site please contact us
Introduction
Background
Pregnancy represents a vital window of alternative for bettering well being outcomes for each mom and youngster. This transformative interval catalyzes vital modifications in a lady’s conception of self and social roles, essentially altering her capabilities, alternatives, and motivations for well being conduct modification [-]. Throughout this paper, we use the time period “woman’ to refer to individuals assigned female at birth, while acknowledging and respecting all gender identities. As a stage of transition, pregnancy presents an optimal time for encouraging healthy lifestyles, including weight management, with benefits that may persist well beyond pregnancy [,].
Current recognized medical guidelines, such as those by the Institute of Medicine (IOM), emphasize the importance of appropriate gestational weight gain (GWG) depending on prepregnancy BMI (ie, underweight BMI <18.5 kg/m²: 12.5‐18 kg; normal weight BMI 18.5‐24.9 kg/m²: 11.5‐16 kg; overweight BMI 25‐29.9 kg/m²: 7‐11.5 kg; and obese BMI ≥30 kg/m²: 5‐9 kg) for improved maternal and fetal outcomes [,]. Deviation from these recommendations – either excessive or insufficient GWG – has been associated with adverse outcomes. For women, these include increased risk of gestational diabetes, hypertensive disorders of pregnancy, cesarean delivery, and postpartum weight retention. For infants, risks encompass preterm birth, inappropriate birth weight, macrosomia, and increased likelihood of childhood obesity [-].
Despite the well-documented health implications, achieving recommended GWG remains a challenge for many women. Globally, 18%-30% of pregnant women gain insufficient weight, while 37%-51% exceed recommendations [,]. This challenge stems from multiple factors: many women are unaware that healthy weight gain targets differ per individual and depend on their preconception BMI [,], and even if women are acquainted with recommendations, they may not know how to operationalize them []. For example, the majority of pregnant women fail to meet dietary recommendations, such as those for vegetables, fats, and grains [,,], and fall short of achieving the recommended 150 minutes of moderate-intensity aerobic activity per week [-]. In addition, common pregnancy symptoms such as nausea, fatigue, and pain, along with lack of knowledge and limited access to information, form additional barriers to maintaining healthy lifestyle practices [,-].
Therefore, pregnant women require support in following healthy lifestyles and achieving healthy GWG. While health care professionals are ideally positioned to deliver this support through prenatal visits [], they often face significant implementation barriers despite their motivation to address these guidelines [,]. For example, they may lack time, access to quality resources, training, and organizational support and perceive their efforts as being ineffective [-]. Consequently, many pregnant women report receiving either insufficient or inaccurate counseling on these topics, while they may desire it [,,].
This issue is further exacerbated in maternity care deserts, where many women receive minimal or no prenatal care [-]. For example, between 2017 and 2023, in several low- and middle-income countries, less than 60% of pregnant women received at least 4 antenatal care visits while the World Health Organization recommends a minimum of 8 [,]. The problem extends to high-income countries, for example, the United States, where over 2.3 million women live in areas without obstetric services, with an additional 3 million having limited access to maternity care [].
Digital Lifestyle Interventions for Managing Healthy GWG
Digital lifestyle interventions have emerged as a promising solution to the challenges of supporting healthy GWG. Several studies have shown the potential of using digital lifestyle programs in a pregnant population. For example, Redman et al [] reported that their eHealth intervention combining digital and nondigital components positively impacted healthy GWG, while Feng et al [] found reduced GWG in participants using a smartphone-based intervention compared with matched controls. However, some interventions have not achieved their intended outcomes [,], while others have shown differential effects depending on prepregnancy BMI classifications [,]. Despite these mixed outcomes, digital lifestyle interventions offer unique advantages in transcending geographical and social barriers, offering cost-effective, scalable support on top of usual care [,-].
Literature reveals considerable variability in how digital lifestyle interventions are designed and implemented, which contributes to the inconsistent results observed across studies. Whether digital behavior change interventions achieve their intended outcomes depends on multiple factors working in concert, necessitating mapping of intervention characteristics, theoretical foundations, and reported findings to better understand the current landscape of approaches. Even among interventions that have shown promise, uncertainty persists regarding which specific design components and implementation features may contribute to their success.
Some studies suggest there are no clear optimal specifications for implementation features of GWG management interventions, such as duration, contact frequency, intensity of use, delivery format (group vs individual), or dietary approach []. Conversely, other research indicates that factors like the timing of intervention initiation, delivery modes (digital-only vs digital-mixed), and the type and frequency of digital components may influence outcomes [,]. A possible explanation for these mixed findings is how behavior change interventions are described: variation in reporting practices impedes our understanding of intervention mechanisms, evidence synthesis, and the development of more effective interventions [].
Design and Implementation of Digital Lifestyle Interventions
Understanding the specific components that contribute to intervention success requires examination of both design and implementation features. Digital lifestyle interventions are a form of behavior change interventions, defined as programs that aim to change behavior with a clear objective and target group []. When designing interventions to change behavior, researchers draw on established theories to understand which components are most likely to produce change and what contextual factors might strengthen or weaken the intervention’s impact. The design of these interventions incorporates multiple behavior change techniques (BCTs), which represent the smallest active ingredients of an intervention capable of inducing behavior change [-]. BCTs are typically implemented based on theory, and understanding their proposed mechanisms of action may illuminate the processes by which BCTs influence behavior []. For example, the BCT “goal setting (behavior)” – outlined as setting or agreeing on a aim outlined when it comes to the conduct achieved, similar to agreeing on a 3-mile each day strolling aim – works by means of rising behavioral self-regulation.
To come to standardized reporting, researchers created a standardized classification system known as the Behavior Change Technique Taxonomy to systematically determine and classify the methods used to alter conduct and higher perceive what makes conduct change interventions efficient. Systematically figuring out and quantifying BCTs used throughout interventions may also help characterize intervention depth to a point, determine widespread patterns or clusters, and validate theory-based ideas [,]. This method has confirmed worth in earlier evaluations of interventions concentrating on smoking cessation [] and bodily exercise (PA) [].
The extent to which a BCT can obtain its supposed goal could fluctuate relying on the goal inhabitants, the particular conduct being addressed, and the way the intervention is carried out []. As beforehand famous, implementation options similar to timing, length, frequency, and supply modalities are vital as they characterize the sensible elements of how interventions are delivered in real-world settings. It is subsequently vital to additional discover how these options work together with BCTs to affect outcomes.
For wholesome GWG interventions, implementation options are significantly vital given the time-sensitive nature of being pregnant and the necessity to accommodate evolving physiological and psychological states. Research signifies that key implementation options influencing end result embrace the timing of initiation (preconception, early being pregnant, or later levels), intervention length (spanning the complete being pregnant or specializing in particular trimesters), frequency of supply (each day contact vs weekly check-ins), and supply modes (digital-only platforms vs hybrid approaches combining digital and face-to-face parts). The alternative of a digital platform itself introduces extra implementation issues (eg, consumer interface design, accessibility throughout gadgets, integration with present well being care methods, and the extent of personalization provided) that are past the scope of this paper.
The relationship between these design and implementation options and optimistic outcomes for wholesome GWG stays unclear and not using a systematic analysis of present proof. Understanding which BCTs are most used, how they’re sometimes carried out, and whether or not sure implementation approaches are related to higher outcomes requires a complete mapping of the present intervention panorama.
Goal of This Study
To tackle the data hole surrounding the traits and implementation of digital life-style interventions for wholesome GWG, we carried out a scoping evaluate specializing in interventions developed between 2014 and 2024. Our aims have been to systematically map intervention traits and determine design and implementation options that seem like related to optimistic outcomes. We examined theoretical foundations, intervention timing, length, frequency, supply modes, and the BCTs used throughout research.
We addressed 2 major analysis questions. First, what’s the scope and nature of proof on digital life-style interventions for wholesome GWG? Second, what design and implementation options characterize digital interventions that report optimistic outcomes?
By addressing these questions, we goal to synthesize present data about intervention parts and supply insights that will inform future analysis and improvement of digital interventions for supporting wholesome GWG along with traditional care. This paper presents a descriptive and narrative evaluation of our findings.
Methods
Overview
We used a scoping evaluate methodology to map the breadth of present proof and determine patterns within the design and implementation of digital life-style interventions. This methodology is especially well-suited for exploring the extent of present literature, mapping and summarizing proof, figuring out key traits, and informing future analysis instructions []. This method was most acceptable given the heterogeneity in digital intervention designs and reporting practices on this discipline, the place the aim was to map proof patterns fairly than to conduct a quantitative synthesis for scientific decision-making.
Our evaluate was guided by Chapter 10 of the Manual for Evidence Synthesis from the Joanna Briggs Institute, particularly the part on scoping evaluations []. For drafting this paper, we adopted the PRISMA-ScR (Preferred Reporting Items for Systematic Reviews and Meta-Analyses extension tips for scoping evaluations) [] (). A evaluate protocol was developed for inside use however was not registered.
Eligibility Criteria
To outline eligibility standards, we used the Population, Concept, and Context framework [], with inhabitants being “pregnant women” and idea “digital lifestyle interventions for managing gestational weight gain.” With “digital” interventions, we seek advice from these with not less than 1 digital element (eg, app, SMS textual content message, and web site), doubtlessly along with nondigital parts (eg, paper booklets and in-person classes). The context was left unspecified to permit for the inclusion of research carried out throughout various well being care settings and populations.
Both articles on major information (eg, pilot research and randomized managed trials [RCTs]) and on secondary information (eg, evaluations and meta-analyses) have been eligible. To guarantee relevance to present digital well being practices, the search was restricted to articles printed inside the final 10 years. A preliminary scan of literature earlier than 2014 indicated that earlier research both didn’t match our inclusion standards or lacked adequate deal with digital parts. Given the speedy evolution of digital well being applied sciences, we thought of this timeframe acceptable for capturing probably the most related and relevant proof. Articles have been excluded from our choice in the event that they weren’t out there in full in English; weren’t based mostly on human participant information (eg, computer-generated information); have been examine protocols with out outcomes; weren’t associated to being pregnant; didn’t embrace GWG as a major or secondary end result; weren’t peer-reviewed publications; didn’t examine digital life-style interventions; or targeted on high-risk pregnancies, similar to placenta previa and gestational diabetes (to take care of deal with common GWG administration fairly than condition-specific interventions). A whole checklist of inclusion and exclusion standards is supplied in .
Search Strategy
The search technique was developed collaboratively by all 4 authors. To retrieve a complete set of related research, we translated our major analysis query into 3 broad search phrases: “smartphone application,” “mHealth,” and “gestational weight gain.” These phrases have been used to go looking 4 digital databases: PubMed, Embase, Cochrane, and Web of Science. These databases have been chosen for his or her in depth protection of biomedical and life science literature. Our search technique confirmed the appropriateness of this choice: whereas every extra database yielded some distinctive data, few of those met our inclusion standards, suggesting that our chosen method achieved near-saturation of the related literature.
The search strings have been tailored to satisfy the syntax necessities of every database. Refer to for the precise strings used per database. The searches in PubMed, Embase, and Cochrane have been carried out on March 8, 2024, and the search in Web of Science was accomplished on March 25, 2024. To match eligibility standards, the searches have been restricted to articles printed in English between 2014 and 2024 and inside the outlined inhabitants and context scope. Search outcomes from every database have been exported as .csv recordsdata and merged right into a single Microsoft Excel file for screening and deduplication.
Article Selection
First, duplicates have been eliminated utilizing a customized Python script. It recoded every article’s title to an ID by eradicating all punctuation marks, turning all capitals into lowercase, and eradicating all areas. These IDs have been used to determine and delete duplicate entries. The remaining, nonduplicate articles have been imported and saved in a Zotero library (Corporation for Digital Scholarship) for additional screening.
Then, every article was screened independently by completely different pairs of the 4 authors utilizing the eligibility guidelines, inspecting titles, abstracts, and outcomes sections. Ineligible articles have been assigned exclusion criterion codes to make sure traceability. Any disagreements between reviewers have been resolved by means of structured dialogue, with a 3rd writer consulted when consensus couldn’t be reached.
To improve comprehensiveness, we additionally searched the web for follow-up publications of excluded examine protocols and hand-searched the reference lists of all included articles. Any extra articles recognized by means of these strategies underwent the identical rigorous screening and choice course of.
Data Extraction
Selected articles have been learn in full by not less than 2 of the 4 authors in numerous mixtures, and related data was extracted. If an article was deemed ineligible at this stage, it was assigned an exclusion criterion code and faraway from the database. All exclusion selections have been mentioned and agreed upon by not less than 2 authors to make sure consistency and transparency.
Extracted information objects included examine kind, examine aim, inhabitants particulars, pattern dimension, theoretical basis, timing, length, frequency, supply mode, BCTs used, and outcomes. The full checklist of information objects and their definitions, together with the methodology for dealing with assumptions and simplifications for sure information objects to make sure consistency throughout research, is supplied in .
Data Synthesis
Data synthesis was carried out by means of a multistage descriptive and narrative evaluation led by RAO, with assist from the opposite authors. First, for major and secondary information articles, we analyzed extracted examine traits (publication 12 months, nation, examine design, and so forth) to map the scope and distribution of the proof base. These findings have been visualized utilizing frequency tables.
Second, for major information articles, we analyzed intervention traits together with theoretical foundations, timing of initiation, length, frequency, supply mode, and BCTs. Third, for nonpilot major information articles and secondary evaluation of RCT information and meta-analyses, we categorized reported examine outcomes as profitable in attaining supposed outcomes (sure or no) based mostly on the examine authors’ conclusions. We in contrast intervention traits throughout these end result classes to determine patterns and potential distinguishing options. Throughout this information synthesis course of, we used fixed comparability strategies to determine patterns, contradictions, and gaps within the proof base.
Results
Overview
We recognized 349 doubtlessly eligible articles by means of database searches, with a further 15 articles discovered by means of analysis protocols (n=3) and hand-searching reference lists (n=12). Following full-text screening, 44 articles met the eligibility standards: 23 major information articles and 21 secondary information articles (meta-analyses, systematic evaluations, and scoping evaluations that included articles on digital life-style interventions). reveals the article identification and choice circulate. The full information extraction file will be present in .
Descriptives of Primary and Secondary Data Articles
Primary Data Articles
The 23 major information articles included 10 pilot RCTs (43%) [,-], 11 full RCTs (48%) [,,,,,-], 1 real-world consumer information examine (4%) [], and 1 nonrandomized intervention examine (4%) []. These research reported on information gathered between 2011 and 2022, with 6 research (26%) accumulating information (partially) throughout the COVID-19 pandemic [,,,,,]. As illustrated in , pilot RCTs have been carried out in most years besides 2016 and 2018. From 2016 onwards, full RCTs have been additionally carried out yearly, though these weren’t follow-ups of the pilot research. A notable peak in RCTs occurred between 2016 and 2019, doubtlessly reflecting elevated curiosity and funding in digital well being interventions throughout that interval.
Studies have been carried out in 10 completely different nations: United States (n=13, 57%) [,,,,-,], Australia (n=2, 9%) [,], Canada (n=1, 4%) [], China (n=1, 4%) [], Finland (n=1, 4%) [], Singapore (n=1, 4%) [], Spain (n=1, 4%) [], Sweden (n=1, 4%) [], Taiwan (n=1, 4%) [], and the United Kingdom (n=1, 4%) []. Sample sizes ranged from 12 to fifteen,468 individuals (median 68, common 916). One examine utilizing industrial app information from over 15,000 girls drove the excessive imply []. summarizes the 23 major information articles, with full particulars in .
| Study | Study kind | Sample dimension | Intervention | Timing (in gestational age) | Delivery medium | Results |
| Digital-mixed interventions | ||||||
| Soltani et al [] | Pilot RCT | 14 | IG and CG. IG obtained the MOMTech intervention, together with 2 each day SMS textual content messages, 4 appointments with a wholesome life-style midwife, aim setting for weight-reduction plan and PA, and use of self-monitoring diaries. | 14‐16 | SMS textual content message, face-to-face |
|
| Smith et al [] | RCT | 51 | IG and CG. All individuals had entry to an internet site on which CG may view common suggestions on weight-reduction plan and PA. IG had extra entry to PA goal-setting modules, problem-solving instruments, journal, calendar, and group discussion board. | 10‐14 | Website, pen-and-paper |
|
| Redman et al [] | RCT | 54 | CG, IG distant, and IG in-person. IG app included customized dietary consumption prescription, weight self-monitoring, exercise monitoring with a pedometer, receipt of well being data, and steady customized suggestions from counselors. | 10‐13 | App, face-to-face |
|
| Willcox et al [] | Pilot RCT | 91 | IG and CG. Both obtained CAU, together with brochures with recommendation on weight-reduction plan and PA. IG additionally obtained tailor-made SMS textual content messages, entry to an internet site, video messages, chat room interplay, and steering from skilled researcher who educated them on diet, PA, and GWG objectives, and helped them observe weight and set objectives. | 13‐17 | SMS textual content messages, social media, pen-and-paper, face-to-face | There was a big distinction in GWG between teams, with IG individuals gaining a median 7.8 (SD 4.7) kg and CG common 9.7 (SD 3.9) kg (P=.04). |
| Van Horn et al [] | RCT | 280 | IG and CG. CG obtained biweekly newsletters and hyperlinks to publicly out there maternity web sites. IG obtained DASH weight-reduction plan and PA teaching. A commercially out there app was used for self-monitoring of weight-reduction plan and PA, with extra assist supplied by means of phone, SMS textual content message, and electronic mail. | 15 | App, face-to-face, mHealth instruments |
|
| Altazan et al [] | RCT | 54 | CG, IG distant, and IG in-person. IG app included customized dietary consumption prescription, weight self-monitoring, exercise monitoring with a pedometer, receipt of well being data, and steady customized suggestions from counselors. | 10-13 | App, face-to-face |
|
| Darvall et al [] | Pilot RCT | 27 | CG, app IG, and app+coach IG. All individuals wore pedometers. In CG, the pedometer show was obscured and couldn’t be synced. Participants in each IGs had a pedometer synced to their private smartphone. They additionally obtained a 4-session behavioral change program. | 13‐19 | App, face-to-face, phone, mHealth instruments | There was no vital distinction between teams in GWG: app IG: imply −5.46, SD 2.8 kg, P=.07; app-coach IG: imply −0.40, SD 2.8 kg, P=.89, each in contrast with CG. |
| Ferrara et al [] | RCT | 394 | IG and CG. CG obtained CAU. IG obtained a program to enhance GWG, weight-reduction plan, PA, and stress administration, together with in-person and phone classes. | 8‐15 | Pen-and-paper, face-to-face, phone |
|
| Downs et al [] | Pilot RCT | 24 | IG and CG. All individuals obtained CAU and monitored their GWG, PA, and dietary consumption. Participants wore 1 exercise monitor each day and 1 in 2-week cycles to trace PA. Intake was recorded by way of app on 2 weekdays and 1 weekend day. In addition, the IG obtained weekly dietitian conferences, individualized caloric objectives, and academic booklets on weight-reduction plan and PA. Intervention dosage was reviewed and adjusted each 3‐4 weeks as wanted. | 8‐12 | App, electronic mail, pen-and-paper, face-to-face, mHealth instruments |
|
| Thomas et al [] | Pilot RCT | 68 | IG and CG. All individuals obtained CAU. IG additionally set PA objectives working as much as 150 energetic minutes/week, wore accelerometer, and weighed themselves each day. They obtained month-to-month calls to evaluate and reset objectives, entry to an internet site with collected information and GWG sources, and private messages. | 10‐12 | SMS, electronic mail, web site, pen-and-paper, phone, mHealth instruments |
|
| Digital-only interventions | ||||||
| Pollak et al [] | Pilot RCT | 33 | Txt4Baby IG and PregCHAT IG. Txt4Baby IG obtained common pregnancy-related SMS textual content message 3 days/week. PregChat IG obtained customized suggestions by means of SMS textual content messages 3 days/week based mostly on consumption of sweetened drinks, vegatables and fruits, quick meals, each day step depend, and weight. | 16 | SMS textual content messages | Participants in IG gained 6 kilos lower than these in CG, however this was not statistically vital (P=.24, 95% CI −15.9 to 4.0). |
| Herring et al [] | RCT | 66 | IG and CG. All individuals obtained CAU. IG moreover obtained steering by means of customized well being coach calls, texts, and suggestions; pedometer; and DVD; on power consumption, PA, and weight. Also, individuals obtained schooling and shared updates in a Facebook (Meta) group. | 8‐17 | SMS textual content message, social media, phone |
|
| Olson et al [] | RCT | 1689 | Placebo, being pregnant IG and postpartum CG, and being pregnant and postpartum IG. All individuals obtained CAU and behavioral change instruments on an internet site and app platform, however the placebo group didn’t obtain weight achieve tracker, weight-reduction plan and PA aim setting, and self-monitoring software. | 12‐20 | App, electronic mail, web site, mHealth instruments | Authors reported no vital distinction in proportion of girls with extreme GWG in IG versus CG (542/1126, 48% vs 260/563, 46%, RR 1.09, 95% CI 0.98-1.20, P=.12). |
| Li et al [] | Pilot RCT | 26 | IG and CG. All individuals obtained normal dietary steering for being pregnant throughout recruitment. IG additionally obtained 8 weeks of real-time meals teaching by way of app. They may add photographs of meals, drinks, or desserts and obtain suggestions and steering from skilled dietitians. | 18‐20 | App |
|
| Litman et al [] | Real-world consumer information | 15,468 | BabyScripts IG. App is supplied by means of HCP to trace GWG. Participants may enter weight manually or by way of a linked scale. App provides focused, gestational-age-specific academic supplies. | <20 | App |
|
| Sandborg et al [] | RCT | 305 | IG and CG. All individuals obtained CAU. IG obtained a 6-month app program, encouraging nutritious diet and PA. This included push notifications 4 occasions/week for data, assist, methods, steering, encouraging data, and reminders. | 13‐14 | App, mHealth instruments |
|
| Gonzalez-Plaza et al [] | RCT | 120 | IG and CG. All individuals obtained CAU. IG used a smartband-connected app to watch PA and for communication with midwife, who supplied customized well being steering. | 12‐28 | App, SMS textual content messages, mHealth instruments |
|
| Souza et al [] | Nonrandomized intervention examine | 27 | IG solely divided into increased or decrease app utilization group that obtained entry to SmartMoms Canada app, Google Fitbit, and Withings scale. App supplied real-time suggestions on diet, PA, sleep, and GWG. Participants have been inspired to make use of it each day. | 12‐20 | App, mHealth instruments |
|
| Chen et al [] | RCT | 80 | IG and CG. All individuals obtained CAU. IG used app and a wearable exercise tracker. App options included prenatal historical past, aim setting, chart historical past, data, prenatal data, rewards, reminders, and prenatal instruments. | 17 | App, SMS textual content messages, mHealth instruments |
|
| Feng et al [] | RCT | 268 | CG and IG. All individuals obtained CAU. IG monitored weight, weight-reduction plan, and PA with an app. App contained sections for weight administration (by means of weighing, weight-reduction plan, and PA), each day recordings (with suggestions), information traits, reminders, and being pregnant schooling per gestational week. | 6‐7 | App | Overall median GWG in IG was considerably decrease than that in CG (median 8.5, IQR 5.5-11 kg vs median 10.0, IQR 6-14 kg; P=.008; Cohen d=0.42). |
| Koivuniemi et al [] | Pilot RCT | 1038 | Standard-app IG and enhanced-app IG. All individuals recorded life-style habits (PA and weight-reduction plan), monitored modifications by viewing graphs of their recordings, and obtained reminders for recording data by way of app. Enhanced-app IG obtained extra, nonpersonalized, motivating data on health-promoting life-style throughout being pregnant. | 9‐20 | App |
|
| Waring et al [] | Pilot RCT | 12 | IG. Participants used an internet site to trace weight-reduction plan, PA, and GWG, and an open group. Participants have been inspired to verify their feed each day, put up recurrently themselves, observe weight-reduction plan and PA, and weigh themselves weekly. Researchers posted messages each day and interacted asynchronously with individuals. | 14‐18 | Website | The authors reported that 70% (7/10) of individuals had eGWG, 10% (1/10) had insufficient GWG, and 20% (2/10) gained inside the really useful ranges. |
| Mattson and Barger [] | Pilot RCT | 22 | Historical CG, self-weighing IG (WA), and self-weighing+counseling IG (WC). WA and WC teams have been requested to weigh themselves weekly. WC group additionally obtained 6×30 min on-line counseling on nutritious diet. | 10‐25 | Telehealth system, mHealth instruments |
|
aTiming of initiation of intervention.
bFor nonpilot research with a management group, impact sizes for variations in imply gestational weight achieve are reported.
cRCT: randomized managed trial.
dIG: intervention group.
eCG: management group.
fPA: bodily exercise.
gGWG: gestational weight achieve.
hIOM: Institute of Medicine.
ieGWG: extreme gestational weight achieve.
jOR: odds ratio.
okayCAU: care as traditional.
lDASH: dietitian-led dietary approaches to cease hypertension.
mmHealth: cell well being.
nThis paper belongs to a sequence of papers reporting on the identical examine because the one by Redman et al []. Hence, it describes the identical interventions.
oRR: relative threat.
pHCP: well being care skilled.
qMET: metabolic equal of activity.
rMVPA: moderate-to-vigorous bodily exercise.
sIDQ: Index of Diet Quality.
Secondary Data Articles
The 21 secondary information articles included 9 meta-analyses (43%) [,,,-], 5 systematic evaluations (24%) [-], 3 scoping evaluations (14%) [,,], and 4 secondary analyses of RCT information (19%) [,-]. Of the 9 meta-analyses, 4 (44%) investigated life-style interventions normally, whereas the opposite 5 (56%) particularly targeted on digital life-style interventions. All systematic evaluations focused digital life-style interventions solely. Of the three scoping evaluations, 2 (67%) addressed common life-style interventions and 1 (33%) examined digital interventions particularly []. illustrates the temporal pattern in analysis methodologies used within the secondary information articles. Meta-analyses, probably the most prevalent method, peaked in 2017 with 3 publications. Notably, 2 meta-analyses printed in 2017 [,] particularly addressed the effectiveness of digital interventions for managing GWG, regardless of a restricted variety of RCTs on the subject at the moment (). Systematic evaluations maintained constant utilization all through the search interval. Scoping evaluations have been printed much less steadily total however confirmed a rise after 2021, which roughly coincides with the publication of the PRISMA Extension for Scoping Reviews in 2018 []. Secondary analyses of RCT information have been sometimes printed 1-4 years following the unique information assortment.
The secondary information articles synthesized research printed between 1992 and 2022, incorporating 36 publications on common (vary 7‐97, median 21), with a median pattern dimension of 526 individuals per article (vary 53‐2833, median 89). Overlap with our major information articles diverse throughout the secondary information article varieties: meta-analyses shared a median of three articles (vary 2‐6), systematic evaluations 2.2 articles (vary 0‐5), and scoping evaluations 3 articles (vary 1‐13) with our number of 23 major research.
In phrases of inclusion standards, nearly all of the secondary information articles primarily included RCTs or quasi-RCTs inspecting wholesome life in pregnant girls and GWG administration, though systematic evaluations allowed for wider inclusion standards encompassing observational and qualitative research.
The goal populations have been predominantly pregnant girls normally, with exceptions together with 3 articles (14%) focusing particularly on girls with obese or weight problems [,,], 1 article (5%) included each wholesome girls and people with (gestational) diabetes [], and 1 article (5%) concentrating on customers of pregnancy-related cell phone interventions [].
A abstract of all included secondary information articles will be present in , with detailed information extraction out there in .
| Study | Study kind | Target group | Overlap major articles/variety of articles included (%) | Results |
| O’Brien et al [] | Systematic evaluate | All wholesome pregnant girls (with out pregnancy-related situations) | 0/7 (0) |
|
| Graham et al [] | Secondary evaluation RCT information | Normal weight to reasonably overweight pregnant girls | — |
|
| The International Weight Management in Pregnancy (i-WIP) Collaborative Group [] | Meta-analysis | All pregnant girls besides these with GDM | 2/81 (2) |
|
| Lau et al [] | Meta-analysis | Overweight and overweight pregnant girls | 2/17 (12) |
|
| Olson et al [] | Secondary evaluation RCT information | Normal weight to reasonably overweight pregnant girls | — |
|
| Sherifali et al [] | Meta-analysis | All pregnant girls | 4/10 (40) |
|
| Overdijkink et al [] | Systematic evaluate | All pregnant girls | 3/29 (10) |
|
| Walker et al [] | Meta-analysis | All pregnant girls besides these with diabetes or GDM | 4/89 (5) |
|
| Chan and Chen [] | Meta-analysis | All pregnant girls | 3/16 (19) |
|
| Mertens et al [] | Systematic evaluate | All wholesome pregnant girls | 2/11 (18) |
|
| Olson et al [] | Secondary evaluation RCT information | Normal weight to reasonably overweight pregnant girls | — |
|
| Vincze et al [] | Meta-analysis | Pregnant girls with (gestational) diabetes | 2/48 (4) |
|
| Hussain et al [] | Systematic evaluate | All pregnant girls who used pregnancy-related cell phone interventions | 5/28 (18) |
|
| Hutchesson et al [] | Scoping evaluate | All pregnant girls | 4/90 (4) |
|
| Rhodes et al [] | Meta-analysis | All pregnant girls, besides with points that might preclude them from collaborating in diet- or PA-based intervention | 5/11 (50) |
|
| Iyawa et al [] | Systematic evaluate | All pregnant girls | 1/18 (6) |
|
| Leonard et al [] | Meta-analysis | All pregnant girls | 6/21 (29) |
|
| Barroso et al [] | Scoping evaluate | Overweight & overweight pregnant girls | 1/8 (13) |
|
| Henriksson et al [] | Secondary evaluation RCT information | All wholesome pregnant girls | — |
|
| Wu et al [] | Meta-analysis | Overweight and overweight pregnant girls | 2/23 (9) |
|
| Raab et al [] | Scoping evaluate | All pregnant girls | 13/97 (13) |
|
aRCT: randomized managed trial.
bNot relevant.
cCG: management group.
dIG: intervention group.
eGWG: gestational weight achieve.
feGWG: extreme gestational weight achieve.
gGDM: gestational diabetes mellitus.
hIPD: particular person participant information
iPA: bodily exercise.
jRR: relative threat.
okaymHealth: cell well being.
lPPWR: postpartum weight retention.
mPPWL: postpartum weight reduction.
nITT: intention-to-treat.
oPPD: per-protocol information.
pBCT: conduct change approach.
qHCP: well being care skilled.
rCAU: care as traditional.
Intervention Characteristics of Primary Data Articles
The 23 major information articles analyzed 22 distinct interventions on which we focus on this part. In common, research usually supplied inadequate element in intervention descriptions, representing a niche in reporting practices that difficult information extraction and interpretation of profitable design and implementation options. Despite this, we have been capable of extract the next data.
Most interventions focused pregnant girls of their first or second trimester who have been obese or overweight, had singleton pregnancies, and no medical situations or being pregnant problems that might have an effect on metabolism or physique weight. Exceptions included 2 research (9%) that additionally included girls with regular weight [,], 1 (5%) specializing in sedentary girls [], and 1 (5%) concentrating on girls who had entered their weight achieve in a particular app [].
Recruitment was principally carried out by means of prenatal clinics (n=9), adopted by (college) hospital obstetric items (n=6), maternity items (n=3), and social media platforms (n=4). Most interventions used the 2009 IOM GWG tips (n=20, 91%). Regarding life-style, most interventions focused each PA and weight-reduction plan (n=18, 82%), whereas 2 interventions (9%) targeted solely on PA [,], and a couple of (9%) on weight-reduction plan solely [,].
Theoretical Foundation
Interventions have been underpinned by a myriad of behavioral and social theories (), with social cognitive principle being most prevalent (n=11, 50%). Notably, 4 interventions (18%) used greater than 1 principle [,,,], whereas 5 interventions (23%) supplied no identifiable theoretical foundation.
| Theory | Number of research | Study reference |
| Social cognitive principle | 11 | Ferrara et al [], Koivuniemi et al [], Sandborg et al [], Pollak et al [], Willcox et al [], Thomas et al [], Waring et al [], Herring et al [], Smith et al [], Gonzalez-Plaza et al [], Chen et al [] |
| Control principle | 1 | Soltani et al [] |
| Social ecological mannequin | 1 | Herring et al [] |
| Integrative mannequin of conduct prediction | 1 | Olson et al [] |
| Behavior mannequin for persuasive design | 1 | Olson et al [] |
| Theory of deliberate conduct | 1 | Downs et al [] |
| Transtheoretical mannequin of change | 4 | Ferrara et al [], Redman et al [], Thomas et al [], Souza et al [] |
| Self-determination principle | 1 | Darvall et al [] |
| No principle talked about | 5 | Feng et al [], Li et al [], Mattson et al [], Van Horn et al [], Litman et al [] |
Timing
Intervention timing was comparatively constant throughout research, with most beginning between gestational weeks 10 and 16 on common. Specifically, 4 interventions began earlier than 14 weeks of gestation (vary 6‐13, 18%) [,,,]; 13 interventions began earlier than 28 weeks of gestation (vary 8‐28, 59%) [,-,,,-,-]; and 5 interventions began between 14 and 28 weeks of gestation (vary 14‐20, 23%) [,,,,].
Duration
The common intervention length was 21.1 (SD 7.3, vary 6‐29, median 24) weeks. Interventions beginning in trimester 1 (<14 wk gestation) lasted about 7 weeks longer (common 28, SD 1.0 wk) than these beginning in both trimester 1 or 2 mixed (<28 wk gestation; common 21, SD 6.9 wk) and about 11 week longer than these beginning in trimester 2 (≥14 wk and <28 wk gestation; common 16 wk, SD 1.5 wk).
Frequency
Frequency patterns diverse significantly throughout research, starting from on-demand content material entry to structured each day check-ins mixed with in-person classes and a few interventions combining a number of codecs (detailed descriptions in ).
Delivery Mode
Most interventions have been digital-only (n=13, 59%) [-,,,-,-], with 7 incorporating cell well being (mHealth) instruments, similar to sensible watches, sensible scales, or pedometers [,,,-]. The different 9 interventions used digital-mixed codecs (41%) [,,,,-,,], of which 3 additionally used mHealth instruments [-].
BCTs
Across all interventions, a complete of 227 BCTs have been recognized, averaging 9 BCTs per intervention (vary 3‐18). The majority of BCTs originated from the clusters “goals and planning,” “feedback and monitoring,” and “social support.” The 3 commonest particular person BCTs have been “self-monitoring of behavior” (code 2.3), “goal setting (behavior)” (code 1.1), and “self-monitoring of outcome(s) of behavior” (code 2.4). A full taxonomy of BCTs and their frequency is introduced in .
| Cluster and code | BCT | Count, n (%) |
| Goals and planning | ||
| Goal setting (conduct) | 18 (82) | |
| Problem fixing | 11 (50) | |
| Goal setting (end result) | 12 (55) | |
| Action planning | 6 (27) | |
| Review conduct objectives | 8 (36) | |
| Discrepancy between present conduct and aim | 5 (23) | |
| Review end result aim(s) | 6 (27) | |
| Behavioral contract | 0 (0) | |
| Commitment | 0 (0) | |
| Feedback and monitoring | ||
| Monitoring of conduct by others with out suggestions | 0 (0) | |
| Feedback on conduct | 12 (55) | |
| Self-monitoring of conduct | 19 (86) | |
| Self-monitoring of outcomes of conduct | 16 (73) | |
| Monitoring end result(s) of conduct by others with out suggestions | 1 (5) | |
| Biofeedback | 1 (5) | |
| Feedback on outcomes of conduct | 10 (45) | |
| Social assist | ||
| Social assist (unspecified) | 11 (50) | |
| Social assist (sensible) | 4 (18) | |
| Social assist (emotional) | 4 (18) | |
| Shaping data | ||
| Instruction on the best way to carry out a conduct | 15 (68) | |
| Information about antecedents | 3 (14) | |
| Reattribution | 0 (0) | |
| Behavioral experiments | 0 (0) | |
| Natural penalties | ||
| Information about well being penalties | 13 (59) | |
| Salience of penalties | 0 (0) | |
| Information about social and environmental penalties | 0 (0) | |
| Monitoring of emotional penalties | 0 (0) | |
| Anticipated remorse | 0 (0) | |
| Information about emotional penalties | 1 (5) | |
| Comparison of conduct | ||
| Demonstration of the conduct | 4 (18) | |
| Social comparability | 3 (14) | |
| Information about others’ approval | 0 (0) | |
| Associations | ||
| Prompts or cues | 13 (59) | |
| Cue signaling reward | 0 (0) | |
| Reduce prompts or cues | 0 (0) | |
| Remove entry to the reward | 0 (0) | |
| Remove aversive stimulus | 0 (0) | |
| Satiation | 0 (0) | |
| Exposure | 0 (0) | |
| Associative studying | 0 (0) | |
| Repetition and substitution | ||
| Behavioral apply or rehearsal | 2 (9) | |
| Behavior substitution | 0 (0) | |
| Habit formation | 0 (0) | |
| Habit reversal | 0 (0) | |
| Overcorrection | 0 (0) | |
| Generalization of a goal conduct | 0 (0) | |
| Graded duties | 0 (0) | |
| Comparison of outcomes | ||
| Credible supply | 13 (59) | |
| Pros and cons | 0 (0) | |
| Comparative imagining of future outcomes | 0 (0) | |
| Reward and risk | ||
| Material incentive (conduct) | 2 (9) | |
| Material reward (conduct) | 1 (5) | |
| Nonspecific reward | 1 (5) | |
| Social reward | 3 (14) | |
| Social incentive | 0 (0) | |
| Nonspecific incentive | 0 (0) | |
| Self-incentive | 0 (0) | |
| Incentive (end result) | 1 (5) | |
| Self-reward | 2 (9) | |
| Reward (end result) | 1 (5) | |
| Future punishment | 0 (0) | |
| Regulation | ||
| Pharmacological assist | 0 (0) | |
| Reduce adverse feelings | 0 (0) | |
| Conserving psychological sources | 0 (0) | |
| Paradoxical directions | 0 (0) | |
| Antecedents | ||
| Restructuring the bodily setting | 2 (9) | |
| Restructuring the social setting | 0 (0) | |
| Avoidance or lowering publicity to cues for the conduct | 1 (5) | |
| Distraction | 0 (0) | |
| Adding objects to the setting | 0 (0) | |
| Body modifications | 0 (0) | |
| Identity | ||
| Identification of self as function mannequin | 0 (0) | |
| Framing or reframing | 0 (0) | |
| Incompatible beliefs | 0 (0) | |
| Valued self-identity | 0 (0) | |
| Identity related to modified conduct | 0 (0) | |
| Scheduled penalties | ||
| Behavior price | 0 (0) | |
| Punishment | 0 (0) | |
| Remove reward | 0 (0) | |
| Reward approximation | 0 (0) | |
| Rewarding completion | 0 (0) | |
| Situation-specific reward | 0 (0) | |
| Reward incompatible conduct | 0 (0) | |
| Reward different conduct | 0 (0) | |
| Reduce reward frequency | 0 (0) | |
| Remove punishment | 0 (0) | |
| Self-belief | ||
| Verbal persuasion about functionality | 1 (5) | |
| Mental rehearsal of profitable efficiency | 0 (0) | |
| Focus on previous success | 1 (5) | |
| Self-talk | 0 (0) | |
| Covert studying | ||
| Imaginary punishment | 0 (0) | |
| Imaginary reward | 0 (0) | |
| Vicarious penalties | 0 (0) | |
aBCT: conduct change methods.
Outcomes
Digital life-style interventions confirmed diverse outcomes for wholesome GWG throughout each major and secondary research, with sure design and implementation options showing extra steadily in interventions that achieved their supposed outcomes. Among 12 major information articles coded for attaining their supposed outcomes, 7 (58%) reported helpful results of digital life-style interventions on wholesome GWG [,,,,,,]. These interventions resulted in decrease complete imply or median GWG (common Cohen d=0.42; vary 0.33-0.55) [,,,,], diminished weekly price of GWG [,], diminished threat for extreme GWG [], and improved adherence to GWG tips [,,,]. The remaining 5 research (42%) didn’t obtain their supposed end result [,,,,]. Although the ten pilot RCTs have been underpowered for definitive conclusions and thus not coded for attaining their supposed outcomes (seek advice from Methods part), most reported helpful traits for wholesome GWG (n=7, 70%) [-,] and a couple of (20%) described higher adherence to IOM tips [,].
Secondary information articles reinforce these findings. Of the 9 meta-analyses, 7 (78%) discovered that interventions achieved their supposed outcomes [,,-], with pregnant girls in intervention teams gaining a median of 1.4 (SD 0.71) kg lower than management teams. All 4 secondary analyses of RCT information reported optimistic outcomes for digital life-style interventions, together with results for subgroups of individuals with mid-high earnings who constantly used digital interventions (common −2.1, SD 0.26 kg) [,], associations between elevated digital engagement with decrease GWG [], and differential results by BMI standing []. All 5 systematic evaluations discovered digital life-style interventions promising for wholesome GWG [-]. The 3 scoping evaluations equally discovered that roughly half of the digital interventions included improved wholesome GWG outcomes [,,].
Design and Implementation Features Linked to Achieving Outcomes
represents a abstract of design and implementation options of the 12 interventions that we coded for attaining their supposed outcomes, adopted by an outline of the primary findings per characteristic.
Theoretical Foundation
The relationship between theoretical grounding and intervention outcomes was inconsistent throughout research. In the first information articles, social cognitive principle was probably the most steadily used theoretical framework, utilized in interventions with various outcomes. Among the 7 interventions that achieved their said aims, 2 have been based mostly on 2 theories (combining the social cognitive principle and the social ecological mannequin [] [14%] or the transtheoretical mannequin of change [] [14%]). Notably, 3 interventions (43%) that met their aims reported no express theoretical basis [,,]. Of the 5 interventions that didn’t obtain their said objectives, 3 (60%) have been based mostly on social cognitive principle [,,], and 1 (20%) mixed the integrative mannequin of conduct prediction with the conduct mannequin for persuasive design []. Secondary information articles didn’t look at theoretical foundations as determinants of intervention outcomes.
Timing
Primary information articles confirmed that interventions attaining their said aims have been initiated earlier in being pregnant (common 12, SD 4.1 wk, vary 6‐28 wk) in contrast with those who didn’t (common 14, SD 1.9 wk, vary 10‐20 wk, ). This sample was per findings from secondary information sources reporting that digital interventions assembly their outcomes sometimes commenced in early-to-mid being pregnant [,,,,].
Duration
Intervention length diverse between research with completely different outcomes, with interventions that achieved their said aims having longer durations (imply 25.5, SD 3.3 wk, vary 21‐29 wk, median 27 wk) in contrast with those who didn’t (imply 21.4, SD 6.9 wk, vary 9‐28 wk, median 25 wk). Secondary information sources described comparable patterns, with Barroso et al [] noting that digital interventions assembly their outcomes sometimes spanned the remaining being pregnant length. Hutchesson et al [] recognized intervention length and depth as areas requiring additional analysis.
Frequency
As said earlier, the frequency of intervention supply confirmed giant variations between interventions. There have been no clear variations in supply cadence between interventions that met their supposed outcomes and those who didn’t. Secondary information articles additionally reported that frequency didn’t have a constant impression on interventions attaining their aims [,,,].
Delivery Mode
Primary information articles revealed variation in outcomes between supply modes, with 4 digital-mixed interventions (57%) attaining their supposed outcomes [,,,] and three digital-only ones (43%) [,,]. Among the interventions that didn’t meet their aims, 1 was digital-mixed (20%) [] and 4 have been digital-only (80%) [,,,]. Secondary information sources corroborate these patterns, with Barroso et al [] noting that digital interventions attaining their objectives sometimes included frequent contact factors with well-trained life-style coaches by means of in-person classes. Hutchesson et al [] equally recognized supply modes in interventions and involvement of well being care suppliers as areas for additional investigation.
BCTs
Primary information articles indicated that interventions assembly their aims used the identical variety of BCTs (common 10, SD 4.3, vary 3‐16) in contrast with those who didn’t (common 10, SD 2.6, vary 6‐14). maps the BCTs used throughout interventions that achieved their said aims and those who didn’t meet their objectives. BCTs used generally (≥70%) throughout each intervention classes have been aim setting (conduct) and self-monitoring of end result(s) of conduct.
| Code | BCT | Interventions attaining supposed outcomes, n (%) | Interventions not attaining supposed outcomes, n (%) | Total |
| 1.1 | Goal setting (conduct) | 5 (71) | 4 (80) | 9 |
| 1.2 | Problem fixing | 2 (29) | 3 (60) | 5 |
| 1.3 | Goal setting (end result) | 5 (71,) | 2 (40) | 7 |
| 1.4 | Action planning | 1 (14) | 1 (20) | 2 |
| 1.5 | Review conduct aim(s) | 2 (29) | 1 (20) | 3 |
| 1.6 | Discrepancy between present conduct and aim | 3 (43) | 0 (0) | 3 |
| 1.7 | Review end result aim(s) | 1 (14) | 1 (20) | 2 |
| 2.2 | Feedback on conduct | 3 (43) | 2 (40) | 5 |
| 2.3 | Self-monitoring of conduct | 6 (86,) | 3 (60) | 9 |
| 2.4 | Self-monitoring of end result(s) of conduct | 6 (86) | 4 (80) | 10 |
| 2.5 | Monitoring end result(s) of conduct by others with out suggestions | 0 (0) | 1 (20) | 1 |
| 2.6 | Biofeedback | 0 (0) | 1 (20) | 1 |
| 2.7 | Feedback on end result(s) of conduct | 4 (57) | 3 (60) | 7 |
| 3.1 | Social assist (unspecified) | 5 (71,) | 1 (20) | 6 |
| 3.2 | Social assist (sensible) | 1 (14) | 1 (20) | 2 |
| 3.3 | Social assist (emotional) | 1 (14) | 2 (40) | 3 |
| 4.1 | Instruction on the best way to carry out a conduct | 5 (71) | 3 (60) | 8 |
| 4.2 | Information about antecedents | 1 (14) | 1 (20) | 2 |
| 5.1 | Information about well being penalties | 5 (71) | 3 (60) | 8 |
| 6.1 | Demonstration of the conduct | 1 (14) | 1 (20) | 2 |
| 6.2 | Social comparability | 1 (14) | 0 (0) | 1 |
| 7.1 | Prompts or cues | 4 (57) | 4 (80) | 8 |
| 8.1 | Behavioral apply or rehearsal | 0 (0) | 1 (20) | 1 |
| 9.1 | Credible supply | 5 (71,) | 2 (40) | 7 |
| 10.1 | Material incentive (conduct) | 0 (0) | 1 (20) | 1 |
| 10.10 | Reward (end result) | 0 (0) | 1 (20) | 1 |
| 10.2 | Material reward (conduct) | 1 (14) | 0 (0) | 1 |
| 10.3 | Nonspecific reward | 0 (0) | 1 (20) | 1 |
| 10.4 | Social reward | 1 (14) | 1 (20) | 2 |
| 10.9 | Self-reward | 1 (14) | 0 (0) | 1 |
aBCT: conduct change approach.
bBCTs that have been used usually, that’s, in ≥70% of interventions (additionally seek advice from );
cBCTs that have been used extra, that’s, ≥25% in profitable interventions as in contrast with an unsuccessful one or vice versa (additionally seek advice from ).
In complete, 5 BCTs confirmed notable variations in utilization patterns between intervention classes, showing ≥25% extra steadily in interventions that achieved their aims: aim setting (end result; 71% vs 40%), discrepancy between present conduct and aim (43% vs 0%), self-monitoring of conduct (86% vs 60%), social assist (unspecified; 71% vs 20%), and credible supply (71% vs 40%). Conversely, interventions that didn’t meet their aims used drawback fixing (60% vs 29%) and social assist (emotional; 40% vs 14%) ≥25% extra steadily than those who achieved their objectives.
Interestingly, with respect to supply modes, we noticed completely different patterns within the variety of BCTs utilized in interventions assembly their goal: digital-mixed interventions (n=4, 57%) used 13 BCTs, whereas digital-only ones used 7 (n=3, 43%). Most generally used BCTs in profitable digital-mixed interventions have been aim setting (conduct and end result) and self-monitoring of conduct. In distinction, profitable digital-only interventions primarily relied on self-monitoring of end result(s) of conduct, details about well being penalties, and a reputable supply.
From the secondary information articles, solely the meta-analysis by Rhodes et al [] examined BCTs in relation to intervention outcomes, reporting a median of 9 BCTs per intervention and confirming that interventions assembly their aims used twice as many BCTs from the teams “goals and planning” and “feedback and monitoring” classes in contrast with interventions that didn’t. They recognized 7 BCTs generally current throughout interventions that attained their aim: aim setting (conduct), drawback fixing, evaluate of conduct objectives, suggestions on conduct, social assist, details about well being penalties, and details about emotional penalties. Notably, evaluate of conduct aim(s) was noticed solely in interventions attaining their aims, with the authors concluding that interactivity is essential for driving engagement and bettering intervention success. A secondary evaluation of RCT information [] described BMI-specific patterns for aim setting, noting helpful associations for ladies with regular BMI however antagonistic associations for these with increased BMI.
Discussion
Evidence Landscape of Digital Lifestyle Interventions for Healthy GWG
Digital life-style interventions alongside traditional care present potential for supporting wholesome life and GWG administration amongst pregnant girls. This scoping evaluate systematically mapped present literature to determine key design and implementation options, inspecting theoretical frameworks, timing, length, frequency, supply modes, and BCTs. Our mapping of 44 articles (23 major and 21 secondary information) revealed various approaches to digital life-style interventions concentrating on GWG administration and highlighted gaps in present analysis. These findings present insights into figuring out analysis priorities and informing future intervention improvement.
Our preliminary analysis query targeted on the scope and nature of proof relating to digital life-style interventions for GWG. The major information articles revealed various patterns in reported outcomes. Among the 23 major research reviewed [,-,-], 12 have been coded for fulfillment [,,,,,-,-] with 7 attaining their supposed outcomes [,,,,,,]. In addition, 7 of the ten pilot research [,-], though underpowered, indicated optimistic traits [-,]. Notably, a number of research that didn’t report enhancements in GWG did doc advantages similar to elevated PA and improved dietary high quality [,,]. The secondary information articles, significantly meta-analyses, generally reported that digital interventions supported GWG, though the reported impacts have been described as modest. These findings are per broader literature suggesting potential advantages of digital life-style interventions for GWG [].
The substantial physique of secondary literature recognized on this evaluate signifies ongoing curiosity in synthesizing proof on digital life-style interventions and their parts. This can also be mirrored within the temporal evaluation of the systematic evaluations in our choice that exposed an evolving analysis focus and enhancements within the high quality of digital life-style interventions. Early work by O’Brien et al [] in 2014 concluded that digital interventions maintain potential for complementing conventional well being care fashions however emphasised the low high quality and amount of the interventions out there. By 2018, Overdijkink et al [] reported that app interventions had good total usability and efficacy, though excessive dropout charges for a number of apps continued. Mertens et al [] in 2019 confirmed that the standard of interventions improved, however challenges with accessibility and engagement with goal populations remained. Subsequently, Hussain et al [] strengthened the potential of digital interventions for managing wholesome GWG however highlighted the continued heterogeneity of research and their usually small pattern sizes. They additionally known as for cost-effectiveness analysis. In 2021, Iyawa et al [] reported that apps may positively impression self-management, similar to GWG, throughout being pregnant whereas noting gaps in longitudinal research and analysis in low- and middle-income nations.
The continued heterogeneity of interventions described by Hussain et al [] was additionally evident within the major information articles we studied, with variations in examine populations (similar to individuals from completely different prepregnancy BMI classes), intervention length, depth, and supply modes. This variety was explicitly acknowledged in a number of meta-analyses as a problem for proof synthesis [,,,,]. In addition, research used diverse approaches to end result measurement, with some specializing in adherence to IOM tips [,,,], others on the whole common [,,,,] or median GWG [,], weekly weight achieve charges [], patterns of intervention engagement in relation to weight outcomes [], or any mixture of those. This heterogeneity in each intervention design and end result measurement approaches underscores the complexity of this analysis discipline. It factors to areas the place higher methodological consistency may benefit future analysis synthesis efforts.
Patterns in Key Design and Implementation Features
Overview
Our second analysis query explored the important thing parts and traits of digital life-style interventions for wholesome GWG. This exploration concerned mapping intervention parts and inspecting patterns throughout several types of interventions to determine generally reported options. We examined intervention theoretical foundations, timing, length, frequency, supply modes, and BCTs.
A major hole emerged relating to the comprehensiveness of intervention descriptions throughout the reviewed articles. Many lacked adequate element for dependable interpretation of parts and potential analysis replication. This discovering was strengthened by our secondary information articles, which equally recognized gaps in reporting on interventions and their outcomes as a key limitation within the discipline. These documentation limitations align with earlier analysis highlighting challenges in evaluating interventions and synthesizing proof about their traits [,,,,,]. Despite the publication of the Behavior Change Technique Taxonomy in 2013 [], and the truth that included research have been printed afterwards, the bulk haven’t included these standardized reporting tips. Despite these limitations, we recognized tentative patterns between design options and intervention outcomes, mentioned under in relation to present literature.
Theoretical Foundations
Regarding theoretical foundations, we discovered that this characteristic didn’t seem to distinguish between interventions that achieved their supposed outcomes and those who didn’t: each varieties of interventions generally drew upon social cognitive principle. Potential explanations embrace inappropriate choice or poor utility of the idea for a given intervention context []. Notably, 5 interventions in our evaluate lacked a clearly outlined theoretical basis, with 3 of those reporting achievement of their supposed outcomes. This sample is noteworthy given out there literature suggesting that theory-driven interventions are extra doubtless to achieve success [,]. Possible explanations embrace that theoretical frameworks have been utilized however not explicitly reported, or that intervention builders targeted on implementing particular BCTs fairly than grounding their method in a central principle.
Timing and Duration
Earlier intervention initiation and longer length have been extra generally noticed in interventions that achieved their said aims. This aligns with the understanding that weight achieve accelerates within the second half of being pregnant, necessitating early intervention [,]. None of the interventions in our evaluate commenced within the third trimester, which aligns with this rationale.
Frequency
No clear patterns emerged between contact frequency and intervention outcomes. Interventions confirmed substantial heterogeneity in touch factors with suppliers and participant management over interplay with intervention parts. Reasons for these variations doubtless stem from theory-based selections, for instance, providing extra frequent however much less customized parts versus much less frequent however extra tailor-made ones.
Delivery Mode
Digital-mixed interventions have been extra usually related to success than digital-only codecs. As such, in-person parts, particularly these involving well being care supplier endorsement, could improve credibility and conduct change, which is according to present analysis. However, as Rhodes et al [] noticed, methodological heterogeneity between interventions could have restricted the power to detect constant patterns of affiliation with intervention outcomes. Therefore, research through which a comparability is made between delivering the identical intervention by means of digital-mixed and digital-only means, similar to that by Redman et al [] that gives an attention-grabbing venue for future analysis.
BCTs
BCTs associated to the classes objectives and planning, and suggestions and monitoring have been steadily related to profitable interventions. While earlier analysis by Webb et al [] and Rhodes et al [] steered {that a} increased variety of BCTs could improve effectiveness, our evaluate didn’t discover a constant relationship between the variety of BCTs and intervention success. This discrepancy could mirror variations in scope: whereas each Webb et al and Rhodes et al examined digital-only interventions, our evaluate included each digital-only and digital-mixed supply. Despite this, the typical variety of BCTs per intervention in our evaluate (common 9) matched that reported by Rhodes et al [], suggesting some consistency in utilization patterns.
More importantly, our evaluation highlighted that particular BCTs, fairly than the whole quantity, have been extra generally related to profitable outcomes. These included: aim setting (end result), discrepancy between present conduct and aim, self-monitoring of conduct, social assist (unspecified), and a reputable supply. These findings are supported by broader literature on weight-reduction plan and PA interventions, through which interventions combining self-monitoring with not less than one different approach from management principle [] – on this case, aim setting (end result) – have been extra generally related to attaining supposed outcomes [].
When inspecting supply modes, we noticed that digital-mixed interventions used extra BCTs than digital-only ones. This could recommend that digital-only interventions will be efficient with fewer, well-targeted BCTs – an space that warrants additional analysis, significantly given the challenges of adapting sure BCTs to digital codecs [].
In Summary
Summarizing, our evaluate recognized a number of implementation and design options that will contribute to digital life-style interventions for GWG in attaining their supposed objectives. While theoretical basis and frequency of intervention supply weren’t generally related to success, interventions that began earlier in being pregnant, lasted longer, and mixed digital with nondigital parts have been extra prone to obtain helpful outcomes. In addition, the BCTs’ aim setting (end result), discrepancy between present conduct and aim, self-monitoring of conduct, social assist (unspecified), and credible supply have been generally current throughout interventions carrying out their supposed goal.
Limitations and Strengths
This evaluate has a number of limitations. First, we excluded articles involving high-risk pregnancies. Given the higher well being stakes in such populations, it’s attainable that digital life-style interventions could have completely different ranges of impression amongst at-risk girls. Future analysis ought to discover whether or not outcomes differ between normal- versus high-risk pregnancies.
A second limitation is the lack to manage for variations in traditional care throughout completely different localities, which might fluctuate considerably between well being care settings. This could affect girls’s alternatives, motivations, and capabilities to interact in wholesome life-style behaviors. Consequently, the practical hole {that a} digital life-style intervention wants to deal with could differ relying on this native well being care context.
Third, most included research have been carried out in high-income nations, primarily the United States. This limits the generalizability of our findings to low- and middle-income nations, the place digital interventions could also be particularly useful as a result of restricted entry to maternity care.
In addition, our searches included solely publications within the English language and people printed till March 25, 2024. This could have excluded related research in different languages or printed extra not too long ago.
Despite these limitations, our evaluate has a number of strengths. We included each digital-only and digital-mixed interventions, enabling comparative insights. Our structured and replicable literature search technique ensured complete protection of the literature. By specializing in wholesome (normal-risk) pregnant populations, we have been in a position to attract clearer comparisons throughout research.
Finally, we carried out an in depth evaluation of key intervention and design options used throughout the included interventions, together with BCTs. This addressed a niche within the literature, as just a few earlier evaluations have supplied an in depth breakdown of which particular BCTs are generally utilized in digital life-style interventions with each digital-mixed and digital-only supply modes for GWG administration in pregnant girls. Our evaluation, subsequently, supplies a basis for understanding the present panorama of BCT implementation on this discipline and should inform future analysis instructions and intervention improvement issues. Therefore, we imagine that regardless of the constraints of our evaluate, our conclusions about the important thing design and implementation options contributing to interventions attaining their supposed objectives stay legitimate.
Future Research
Based on our findings, we suggest 5 key priorities for future analysis, which we define under.
Healthy Lifestyles Versus GWG
First, digital life-style interventions reported modest variations in GWG between interventions that met their supposed objectives and those who didn’t (common discount of 1.4, SD 0.71 kg). This is according to Dodd et al [], whose meta-analysis of 78 research on nondigital life-style interventions discovered a median lower in GWG of 1.1 (95% CI –1.46 to –0.74) kg in pregnant intervention individuals in contrast with controls. In mixture with earlier analysis in nonpregnant populations suggesting that extra weight reduction is required earlier than well being advantages begin exhibiting [,] and our findings that digital life-style interventions can positively have an effect on life even with out profitable outcomes for wholesome GWG [,,], future research ought to discover whether or not these behavioral modifications, fairly than GWG alone, are the first drivers of improved being pregnant outcomes.
Engagement and Standardization
Second, extra work on methodological elements is required, similar to differentiating between digital and behavioral engagement in longitudinal interventions. Digital engagement refers to customers’ interplay with intervention parts, whereas behavioral engagement describes the adoption of goal well being behaviors. Understanding their temporal relationship is essential: interventions sometimes require preliminary digital engagement till customers set up habits, after which behavioral engagement turns into extra significant for assessing success. Measuring one kind when desiring to measure the opposite may result in deceptive conclusions about what contributes to attaining supposed outcomes. It is vital to notice that long-term engagement—whether or not digital or behavioral—could not all the time be crucial for intervention success []. Rather than pursuing most engagement, future analysis ought to deal with figuring out “effective engagement” to realize wholesome GWG. In addition, broader use of the BCT taxonomy [] and clearer reporting of implementation methods would improve comparability and replication.
Intervention Development
Third, regarding intervention improvement, a greater understanding of particular person BCTs and their implementation is required. Our findings reveal an overlap between probably the most distinguished BCTs in each efficient and ineffective interventions, suggesting that BCT implementation strategies could affect outcomes otherwise than the methods themselves. For occasion, social assist will be delivered by means of boards, group settings, or peer-to-peer pairs, every with doubtlessly completely different traits. Future analysis may discover BCT implementation approaches and their underlying mechanisms of motion.
Another consideration entails current technological traits in intervention improvement. As most being pregnant apps are developed by non-public corporations [], their options mirror present technological capabilities, together with complete information assortment, wearable integration, and algorithm-driven teaching [,]. While these advances could improve accessibility and personalization, they introduce vital moral issues [,]. Particularly, personalization raises considerations about consumer autonomy [] and trustworthiness [], particularly as intervention guidelines change by means of utilization, doubtlessly making one-time knowledgeable consent inadequate. In addition, pregnant girls could place various ranges of belief in these methods regardless of doubtlessly restricted understanding of their capabilities []. Users could find yourself over- or undertrusting the system, each of which come at a value []. Future analysis may discover how moral issues is likely to be built-in into intervention improvement from the beginning, with consideration to significant and moral personalization.
Collaboration Between Academia and Industry
Fourth, we advocate enhanced collaboration between academia and trade within the analysis and improvement of digital life-style interventions for pregnant girls. While our evaluate recognized many pilot RCTs, we noticed a notable shortage of follow-up research scaling these preliminary findings. This hole could stem from the inherent challenges of transitioning from pilot research to full-scale RCTs with technologically mature interventions inside tutorial settings, the place analysis sometimes progresses at a extra measured tempo and fewer sources for technological improvement can be found. In distinction, trade companions function on accelerated improvement cycles with direct entry to rising applied sciences. This distinction in throughput time turns into significantly related provided that whereas most printed analysis originates from tutorial establishments, nearly all of pregnancy-related apps in use are developed by non-public expertise corporations []. Strategic partnerships between these sectors may mix academia’s methodological rigor with trade’s technological agility, accelerating the interpretation of evidence-based interventions into broadly accessible instruments whereas sustaining scientific integrity, to collectively enhance the well being conduct of pregnant girls and their households.
Societal Aspects
Finally, whereas there’s a widespread assumption amongst researchers and policymakers that digital interventions for being pregnant may considerably profit well being care desert communities, this speculation stays largely unexplored in empirical analysis [,]. This hole is especially noteworthy as a result of, as a result of their distinctive circumstances and restricted different sources, people in well being care deserts may very well reveal increased success charges with digital interventions when it comes to elevated functionality and alternative to interact with wholesome behaviors. In our scoping evaluate, solely Iyawa et al [] addressed this significant facet. Furthermore, the well being care desert context emphasizes the significance of conducting rigorous cost-effectiveness analyses, evaluating completely different intervention approaches. Such financial evaluations are important for making knowledgeable selections about useful resource allocation, significantly in areas the place well being care sources are already scarce. This financial perspective turns into particularly related when contemplating the potential scalability and sustainability of digital life-style interventions in underserved communities [].
Conclusion
This scoping evaluate examined the panorama of digital life-style interventions aimed toward supporting wholesome (GWG), with a deal with 6 key design and implementation options: theoretical basis, intervention timing, length, frequency, supply modes, and BCTs. Our findings verify that digital interventions maintain promise for selling wholesome GWG. While theoretical underpinnings and frequency of supply didn’t constantly predict success, interventions that started earlier in being pregnant and lasted longer have been extra prone to obtain helpful outcomes. Digital-mixed supply modes—these combining digital instruments with in-person contact—appeared more practical than digital-only codecs. Importantly, 5 BCTs emerged as extra generally utilized in profitable interventions: aim setting (end result), discrepancy between present conduct and aim, self-monitoring of conduct, social assist (unspecified), and credible supply.
These findings present a basis for designing more practical, evidence-based digital interventions to assist maternal well being. Future analysis ought to proceed to refine these parts, discover their implementation in various populations, and tackle gaps in reporting and standardization.
We are grateful to Joyce Westerink, PhD and Sumit Raurale, PhD, for his or her invaluable evaluations of our manuscript. In addition, we thank Dimple Bhadani, MDes, for designing the figures.
During the preparation of this manuscript, the authors made use of Philips Enterprise AI Chat (model October 2023; to be used by Philips staff solely) and Claude Sonnet 4 (claude-sonnet-4-20250514, supplied by Anthropic). Artificial intelligence session initially targeted on manuscript writing help for readability and consistency and subsequently supplied methodological steering for scoping evaluate conduct and recommendation on aligning analysis aims with analytical method. All synthetic intelligence–generated content material was reviewed, edited, and verified by the authors, who take full accountability for the manuscript’s content material and conclusions.
None declared.
Edited by Naomi Cahill; submitted 30.Jan.2025; peer-reviewed by Danielle Downs, Wendy Bennett; remaining revised model obtained 27.Jun.2025; accepted 30.Jul.2025; printed 14.Nov.2025.
© Renée A Otte, Lucie Duracher, Ozge Demir, Hanne A A Spelt. Originally printed within the Journal of Medical Internet Research ( 14.Nov.2025.
This is an open-access article distributed beneath the phrases of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which allows unrestricted use, distribution, and copy in any medium, supplied the unique work, first printed within the Journal of Medical Internet Research (ISSN 1438-8871), is correctly cited. The full bibliographic data, a hyperlink to the unique publication on in addition to this copyright and license data have to be included.
This web page was created programmatically, to learn the article in its authentic location you possibly can go to the hyperlink bellow:
https://www.jmir.org/2025/1/e71548
and if you wish to take away this text from our web site please contact us
