Mobile health support and its effects on mothers and newborns with pregnancy diabetes

A comprehensive electronic literature search was conducted across PubMed, Web of Science, Scopus, Cochrane Library, and EMBASE from the inception of each database up to July 23, 2025. The search strategy incorporated both MeSH and free-text terms. The search strategy for PubMed is shown in. It was adapted for the other databases. Search syntax was adapted according to the specific indexing systems and query rules of each database. A manual search of reference lists from relevant articles was also performed to identify additional eligible studies. 1

Literature retrieval was limited to published English-language articles. Eligible studies were required to meet the following criteria:

All retrieved records from the databases were imported into EndNote X9 software for deduplication and literature management. Two independent reviewers (QFY and YLY) initially screened titles and abstracts based on predefined inclusion criteria. For studies that preliminarily met the criteria, full texts were retrieved and further screened to determine final inclusion. Any disagreements between the two reviewers during the screening process were resolved through discussion; when necessary, a third reviewer (JYL) was consulted to provide arbitration.

This study strictly adhered to the PRISMA statement for data extraction to ensure methodological systematicity. Two reviewers (QFY and YLY) independently extracted data using a pre-tested data extraction form, and a third author (JYL) cross-verified the accuracy of the results. The extracted data included publication details (authors, year of study, country), sample size, intervention methods for both the intervention and control groups, duration of intervention, and primary outcomes.

Two authors (QFY and YLY) independently assessed the risk of bias, methodological quality, and certainty of evidence for the included studies. Any assessment discrepancies were resolved through discussion with a third reviewer (JYL) until consensus was reached. The risk of bias was evaluated using the revised Cochrane Risk of Bias tool (RoB 2), focusing on the following five domains: the randomization process, deviations from intended interventions, missing outcome data, outcome measurement, and selection of the reported result (37).

Given the anticipated heterogeneity among the included studies, a random-effects model was employed to synthesize the pooled estimates of post-intervention effects. Standardized mean differences (SMDs) with 95% confidence intervals (CIs) were calculated using the inverse variance method for continuous variables. Following Cohen (38), effect sizes were interpreted as small (0.2), moderate (0.5), or large (0.8). The degree of heterogeneity was assessed using the I² statistic, with the following thresholds: I² < 25% indicating low heterogeneity, 25% ≤ I² < 50% moderate heterogeneity, 50% ≤ I² < 75% high heterogeneity, and I² ≥ 75% indicating very high heterogeneity (39). Publication bias was evaluated through visual inspection of funnel plot symmetry and by calculating Begg’s and Egger’s test values (40, 41). Sensitivity analysis was conducted using the leave-one-out method to assess the robustness of the pooled results (42). All statistical analyses were performed using STATA 17.0.

There were no other inconsistencies with the pre-registration protocol.

The study selection process and reasons for exclusion are illustrated in Figure 1. A total of 6,279 records were retrieved from five databases. After removing duplicates, 3,441 potentially eligible studies were identified. Based on title and abstract screening, 106 studies were initially included. After full-text retrieval and evaluation, 18 RCTs met the inclusion criteria. Among the 88 studies, 44 were excluded due to irrelevant research topics, 16 due to mismatched intervention methods, 17 due to the absence of relevant outcome measures, and 7 due to incomplete data. Additionally, 10 studies were identified through tracking and supplementary searches of relevant citations. After applying the inclusion and exclusion criteria, one more study was included. Ultimately, 19 studies were included in the meta-analysis (29–35, 43–54).

This meta-analysis included 19 randomized controlled trials conducted between 2007 and 2023, which investigated the effects of mHealth interventions on maternal and neonatal outcomes in pregnant women with GDM. The included studies were conducted across multiple countries and regions, including Asia (China), North America (the United States, Canada), and Europe (Spain, the United Kingdom). 1821 pregnant women with GDM were enrolled across these studies, with sample sizes ranging from 21 to 240 per study. All 19 included studies utilized usual care as the control, while the experimental groups received remote interventions based on mHealth tools such as smartphones and applications. The intervention groups involved diverse types of applications, including mobile phones, the Internet, smartphones (e.g., WeChat), web-based telemedicine systems (e.g., DiabeTIC website), and telehomecare (THCa) systems. Regarding intervention duration, 15 studies continued the intervention until delivery, one study lasted 6 months, one for 3.5 months (14 weeks), and two did not report the duration. The primary outcome measures in this analysis included cesarean section, composite neonatal complications, 2-hour oral glucose tolerance test, and gestational weight gain, among others. More detailed information regarding the main results of each study is presented in Table 1.

As shown in Figure 2, among the 19 included studies, 13 were rated as having an overall low risk of bias, 5 studies were assessed to have some concerns regarding the risk of bias (34, 43, 45, 48, 51), and 1 study was rated as having a high risk of bias (35). Regarding the randomization process, except for one study that was evaluated as having some concerns (45), the remaining 18 studies reported adequate randomization procedures. They were considered to have a low risk of bias. Regarding deviations from intended interventions, 16 studies were rated as having a low risk of bias, while 3 studies were assessed to have some concerns (34, 48, 51). Regarding missing outcome data, 17 studies were judged as having a low risk of bias, one study raised some concerns (43), and another was rated as having a high risk of bias (35). In the domains of outcome measurement and selection of reported results, all 19 studies were evaluated as having a low risk of bias.

A systematic review and meta-analysis of 19 eligible randomized controlled trials was conducted to evaluate the impact of mHealth interventions on maternal and infant outcomes in pregnant women with GDM. The results demonstrated that mHealth interventions significantly reduced the incidence of cesarean section and emergency cesarean delivery among women with GDM, as well as significantly lowering the risk of composite neonatal complications. Additionally, mHealth interventions markedly improved the 2-hour postprandial blood glucose levels in GDM patients. Furthermore, a significant trend toward reduced gestational weight gain was observed in the mHealth intervention group. Detailed statistical results are presented as follows:

The sensitivity analysis results are presented in. The pooled effect estimates remained robust after the sequential exclusion of each study. 1

Both Begg’s test and Egger’s test (P > 0.05) indicated no significant publication bias in the results of this meta-analysis (). 1

In relation to caesarean section-related circumstances, mobile healthcare has played a multifaceted role. First, through wearable devices and mobile applications, mHealth enables real-time and continuous monitoring of maternal physiological indicators such as blood glucose, providing continuous glucose profiles to capture fluctuation patterns (13). If abnormal blood glucose levels are detected, the system automatically triggers alerts, facilitating immediate intervention (57). Healthcare providers can adjust treatment plans based on real-time data (58), effectively preventing macrosomia due to uncontrolled blood glucose and significantly reducing the associated risk of cesarean section (59). Second, mHealth applications can provide personalized dietary and exercise recommendations tailored to the individual condition of pregnant women (23). These behavior intervention plans, tailored to different risk levels, can enhance patient compliance and self-management capabilities (23). Additionally, pregnant women can use mHealth platforms to communicate with healthcare providers in real time and report data (60), thereby enhancing their confidence in self-management and enabling prompt resolution of clinical issues (60), thus avoiding emergency cesarean sections due to disease progression (61).

Regarding the improvement of neonatal outcomes, the real-time monitoring mechanism enables continuous tracking of maternal blood glucose levels, allowing for the timely detection and correction of abnormalities (13), thereby reducing fetal exposure to hyperglycemia (57) and preventing fetal metabolic disorders and related complications. Personalized support, which provides appropriate dietary and exercise plans based on the specific conditions of pregnant women (23), optimizes nutrient intake and energy expenditure (23), lowers the risk of intrauterine hyperglycemia (59), and reduces the occurrence rate of macrosomia, neonatal hypoglycemia, and respiratory distress syndrome (62). The continuous monitoring provided by mHealth offers data support, enabling more precise interventions and safeguarding neonatal health (57).

In terms of glycemic control, real-time monitoring enables pregnant women and healthcare providers to dynamically track blood glucose fluctuations via wearable devices (13). If abnormal postprandial glucose levels are detected, the system provides immediate alerts, prompting adjustments to dietary or physical activity regimens (63). Personalized support involves the development of individualized management plans based on the patient’s glucose profile and risk factors (23, 60). For instance, individuals with suboptimal glycemic control are advised to adopt high-fiber diets and engage in low-intensity exercise, which has been shown to improve the rate of achieving target glucose levels (59). Through mobile platforms, physicians and patients engage in interactive consultations, facilitating the timely resolution of patient concerns (60) and enhancing treatment adherence (23). Such professional guidance reinforces glycemic management and helps maintain glucose levels within a safe range (60).

Regarding gestational weight gain, mHealth interventions help pregnant women achieve appropriate weight control through personalized guidance on dietary and physical activity. These individualized plans are designed based on maternal physiological parameters to optimize energy balance (23) and are dynamically adjusted using real-time feedback on behavioral data (23). For example, when excessive weight gain is observed, the system or healthcare providers prompt the patient to increase physical activity or modify dietary intake, thereby mitigating risks associated with excessive gestational weight gain (12). Successful weight management has been associated with a reduced incidence of macrosomia and cesarean delivery (12).

In summary, mHealth interventions positively influence maternal and infant outcomes, including cesarean section rates, neonatal outcomes, glycemic control, and gestational weight gain among women with GDM through multidimensional mechanisms such as real-time monitoring, personalized support, and enhanced patient–provider interaction.

Additionally, one study has evaluated the association between digital health interventions and maternal and neonatal outcomes (64). We compared the present study with the aforementioned research: Regarding literature search periods and databases, the study by Wang et al. (64) covered publications up to August 2024 and searched four databases (PubMed, Embase, Cochrane Library, and Web of Science). The current study, however, extended the search up to July 2025 and included five databases: PubMed, Web of Science, Scopus, Cochrane Library, and EMBASE. In terms of the number of included studies and study populations, Wang et al. (64) incorporated 42 relevant RCTs involving 148,866 pregnant women, whereas the present study included 19 RCTs involving 1,821 pregnant women with GDM. Regarding outcome measures, the study by Wang et al. (64) evaluated outcomes such as gestational weight gain, gestational hypertension or preeclampsia, gestational age at delivery, miscarriage, shoulder dystocia, preterm birth, infant birth weight, macrosomia, and neonatal hypoglycemia. In contrast, our study focused on five specific outcomes: caesarean section, emergency caesarean, composite neonatal complications, 2-hour postprandial blood glucose, and gestational weight gain. Furthermore, this study explored the potential mechanisms through which mHealth interventions may improve maternal and infant outcomes in GDM, thereby enriching both the research content and theoretical foundation. Finally, the conclusions drawn in this study were based on sensitivity analysis and publication bias detection, rendering them more robust and persuasive.

This study demonstrates that mHealth interventions can effectively optimize glycemic control and reduce the risks of emergency cesarean delivery and composite neonatal complications. These findings suggest that such interventions may compensate for the limitations of traditional healthcare models in follow-up management and real-time intervention, particularly offering a feasible solution for resource-limited settings. Healthcare providers should fully recognize the potential of mHealth in managing GDM. Through remote monitoring, intelligent reminders, and behavioral interventions, mHealth enhances patient compliance and self-management capabilities, thereby reducing hospitalization needs and long-term healthcare burdens. Furthermore, this study provides evidence-based support for updating clinical guidelines and informing policy-making, which may facilitate the integration of mHealth into standard GDM care and promote the advancement of personalized medicine and interdisciplinary collaboration. Future research should further explore AI-driven precision interventions and the long-term effects of mHealth on offspring health, thereby establishing a more solid theoretical and practical foundation for applying digital healthcare in the perinatal period.

The original contributions presented in the study are included in the article/. Further inquiries can be directed to the corresponding author. 1