Effects of lifestyle education interventions on quality of life and fbg in women with gestational diabetes: a systematic review and meta-analysis Authors: Mina Shojaee, Mahsa Sharifnejad, Nastaran Safavi Ardabili, Fatemeh Abdi, Mohammadamin Jandaghian-Bidgoli, Mahboubeh Firouz, Solmaz Norouzi Published: 2026 ABSTRACT Introduction: Gestational diabetes mellitus (GDM) is a prevalent condition with significant maternal and fetal health implications. Lifestyle education is a promising non-pharmacological intervention to improve quality of life (QoL) and fasting blood glucose (FBG) in affected women. For the purposes of this review, lifestyle education interventions are defined as structured programs including dietary counseling, physical activity guidance, and/or self-management education (e.g., blood glucose monitoring), delivered through planned sessions with defined frequency, duration, and content. This systematic review and meta-analysis evaluates its efficacy.MethodsA comprehensive search of PubMed, Embase, Cochrane CENTRAL, Scopus, and Web of Science was conducted up to October 2025, following PRISMA guidelines. Studies were included if they involved women with GDM, evaluated lifestyle education (structured, curriculum-based programs with multiple sessions, specified learning objectives, and follow-up components), and reported QoL or FBG outcomes. Randomized controlled trials and quasi-experimental studies with comparators were eligible. Two reviewers independently screened studies, extracted data, and assessed risk of bias using the Cochrane tool. Meta-analyses used a random effects model, with heterogeneity assessed via I² statistics. Subgroup analyses explored intervention types, and publication bias was evaluated using Egger’s and Begg’s tests.ResultsFrom 304 records identified, 20 studies (19 randomized controlled trials and 1 quasi-experimental study) involving 4,777 women were included. Overall meta-analysis demonstrated a statistically significant effect of lifestyle interventions on quality of life and fasting blood glucose (FBG) outcomes (Hedges’ g = − 0.31, 95% CI: −0.57 to − 0.05, p = 0.02), although substantial heterogeneity was observed (I² = 91.94%). For FBG outcomes, the overall pooled effect estimate was − 4.01 mg/dL (95% CI: −8.13 to 0.10), with high heterogeneity (I² = 97.42%). Subgroup analyses showed that combined individual and group-based interventions significantly reduced FBG levels (pooled effect estimate = − 2.60 mg/dL, 95% CI: −4.90 to − 0.31; I² = 0%), whereas individual-only and group-only interventions were not statistically significant. For quality of life outcomes, the overall pooled effect was not statistically significant (pooled effect estimate = 3.42, 95% CI: −0.55 to 7.39; I² = 91.28%). Exploratory subgroup analyses indicated that combined interventions significantly improved quality of life (pooled effect estimate = 5.20, 95% CI: 3.76 to 6.64; I² = 0%), while standalone individual and group interventions showed no statistically significant effects. Egger’s (p = 0.12) and Begg’s (p = 0.19) tests suggested no significant publication bias. Sensitivity analyses confirmed the robustness of the findings.ConclusionLifestyle education interventions effectively reduce FBG in women with GDM, with group and combined interventions showing most consistent benefits. However, the impact on QoL remains uncertain; while combined individual and group approaches showed some promise in exploratory subgroup analyses, the overall pooled effect on QoL was non-significant with substantial heterogeneity. These subgroup findings should be considered hypothesis-generating rather than definitive. Tailored, scalable interventions integrating individual and group components may optimize GDM management, but further high-quality research is needed to confirm efficacy for QoL improvement.Supplementary InformationThe online version contains supplementary material available at 10.1186/s13098-026-02189-x. Introduction Gestational diabetes mellitus (GDM) is a condition characterized by glucose intolerance that is first diagnosed during pregnancy, typically in the second or third trimester, and generally resolves after delivery [1]. It is distinguished from pre-existing diabetes by its onset and is marked by elevated blood glucose levels that fall below the threshold for type 1 or type 2 diabetes but still pose significant health risks to both mother and child. GDM represents a critical public health concern due to its rising prevalence and multifaceted impact on maternal and fetal health [2]. The global epidemiology of GDM underscores its growing significance. Estimates suggest that GDM affects 7–14% of pregnancies worldwide, though prevalence varies widely based on diagnostic criteria, geographic region, and population characteristics [3]. For example, higher rates are observed in South Asian, Hispanic, and Indigenous populations, with prevalence reaching up to 20–30% in high-risk groups. The increasing incidence of GDM is closely linked to rising rates of obesity, sedentary lifestyles, and advanced maternal age, all of which exacerbate insulin resistance during pregnancy. In low- and middle-income countries, where access to screening and management may be limited, the burden of GDM is particularly pronounced, contributing to disparities in maternal and neonatal health outcomes [4]. The consequences of GDM impose a significant burden on women, their families, and healthcare systems. During pregnancy, women with GDM face an elevated risk of complications such as preeclampsia, gestational hypertension, and cesarean delivery. Their newborns are at increased risk of macrosomia, neonatal hypoglycemia, respiratory distress syndrome, and birth trauma [5]. Beyond the immediate perinatal period, GDM has long-term implications. Women with a history of GDM have a 7- to 10-fold higher risk of developing type 2 diabetes within 5–10 years postpartum, and their offspring are predisposed to obesity, insulin resistance, and metabolic syndrome in later life [6]. These health challenges are compounded by psychosocial impacts, including heightened anxiety, stress, and reduced quality of life (QoL) stemming from the demands of managing a high-risk pregnancy. The cumulative effect of these complications underscores the urgent need for effective interventions to mitigate the burden of GDM. In addition, Quality of life and fasting blood glucose (FBG) are pivotal outcomes in the management of GDM, reflecting both patient-centered and clinical priorities [7]. QoL encompasses physical, emotional, and social dimensions of well-being, which can be significantly disrupted by GDM. The need for constant blood glucose monitoring, adherence to strict dietary regimens, and concerns about fetal health can erode emotional resilience and social functioning, leading to diminished QoL. FBG, a key marker of glycemic control, is directly linked to the risk of adverse maternal and fetal outcomes [8]. Elevated FBG levels are associated with increased rates of macrosomia, cesarean delivery, and neonatal complications, making its regulation a cornerstone of GDM care. Improving QoL and achieving optimal FBG levels are thus interconnected goals that not only enhance maternal health during pregnancy but also reduce the long-term risk of chronic diseases [9]. Lifestyle education has emerged as a cornerstone intervention for managing GDM, offering a non-pharmacological approach to improve glycemic control and overall well-being. This intervention typically includes structured guidance on nutrition, physical activity, and self-management strategies tailored to the unique needs of pregnant women. Nutritional counseling focuses on balanced carbohydrate intake and portion control to stabilize blood glucose levels, while exercise recommendations emphasize moderate activities, such as walking or prenatal yoga, to enhance insulin sensitivity. Additionally, lifestyle education equips women with the knowledge and skills to monitor their blood glucose, recognize symptoms of hyperglycemia, and make informed decisions about their care. By fostering sustainable behavioral changes, lifestyle education empowers women to take an active role in managing their condition, potentially improving both clinical outcomes and QoL [10]. For the purposes of this review, lifestyle education is defined as a structured, curriculum-based program that goes beyond simple advice or information provision. It encompasses organized instructional activities designed to promote sustainable behavioral changes in diet, physical activity, and self-management, delivered through planned sessions (individual or group) with defined frequency, duration, and content [11, 12]. This distinguishes eligible interventions from routine prenatal counseling or brief lifestyle recommendations, which typically involve one-time advice without structured follow-up. Eligible programs included multiple sessions with specified learning objectives, behavior change techniques (e.g., goal setting, self-monitoring, problem-solving), and systematic follow-up components [13]. This operational definition was applied to ensure conceptual coherence across the heterogeneous studies included in the meta-analysis. Despite the promise of lifestyle education, significant gaps remain in the literature regarding its specific effects on QoL and FBG in women with GDM. While numerous studies have explored the impact of lifestyle interventions on obstetric outcomes, such as birth weight or cesarean delivery rates [14, 15], fewer have prioritized patient-centered outcomes like QoL. Moreover, the effectiveness of lifestyle education programs varies widely due to differences in their design, delivery, and intensity. Some programs are delivered in-person by healthcare professionals, while others leverage digital platforms or group-based formats, leading to inconsistent outcomes. Additionally, there is limited consensus on the optimal components of lifestyle education, such as the frequency of counseling sessions or the type of physical activity most beneficial for women with GDM. These inconsistencies highlight the need for a systematic evaluation of existing evidence to determine the efficacy of lifestyle education and identify best practices. This systematic review and meta-analysis addresses these research gaps by synthesizing evidence on the effects of lifestyle education on QoL and FBG in women with GDM. By focusing on these critical outcomes, the study aims to provide a comprehensive understanding of how lifestyle interventions can improve maternal health and well-being. Ultimately, this research seeks to empower women with GDM to lead healthier pregnancies and reduce the long-term burden of this condition on individuals and healthcare systems alike. Therefore, the purpose of this study is to evaluate the effects of lifestyle education on QoL and FBG in women with GDM. Methods Design This study was a systematic review and meta-analysis conducted to evaluate the effects of lifestyle education on QoL and FBG in women with GDM. The review adhered to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines to ensure transparency and rigor in reporting [16]. A meta-analytic approach was employed to quantitatively synthesize data from randomized controlled trials (RCTs) and other relevant study designs, where appropriate, to provide robust evidence on the efficacy of lifestyle education interventions. Additionally, the study has been registered with the International Prospective Register of Systematic Reviews (PROSPERO) under registration number CRD420251026191. Search strategy A comprehensive literature search was conducted to identify relevant studies published up to October 2025. The electronic databases PubMed, Embase, Cochrane Central Register of Controlled Trials (CENTRAL), Scopus, and Web of Science were searched. The search strategy combined Medical Subject Headings (MeSH) and free-text terms related to GDM, lifestyle education, QoL, and FBG, including terms such as “gestational diabetes mellitus,” “GDM,” “lifestyle education,” “lifestyle intervention,” “dietary counseling,” “physical activity,” “quality of life,” “QoL,” “fasting blood glucose,” “FBG,” and “glycemic control.” Boolean operators (AND, OR, NOT) were used to refine the search, and filters were applied to limit results to human studies and English-language publications. Additional sources, including grey literature via Google Scholar and clinical trial registries such as ClinicalTrials.gov, were explored. The reference lists of included studies and relevant reviews were manually screened to identify additional eligible studies. The search strategy was developed in consultation with a medical librarian to optimize sensitivity and specificity. All search results were imported into EndNote to remove duplicates and facilitate screening. The full search strategy for each database is attached as supplementary file. Inclusion and exclusion criteria The inclusion and exclusion criteria were defined using the PICO framework (Population, Intervention, Comparator, Outcome) to ensure the selection of relevant studies. Population: Studies involving women diagnosed with GDM, regardless of gestational age at diagnosis, were included. Studies focusing on women with pre-existing type 1 or type 2 diabetes, or those without a confirmed GDM diagnosis, were excluded.Intervention: Studies evaluating lifestyle education interventions, defined as structured programs including dietary counseling, physical activity guidance, and/or self-management education (e.g., blood glucose monitoring), were included. Interventions could be delivered in-person, via digital platforms, or through hybrid formats. Studies focusing solely on pharmacological interventions or non-lifestyle interventions (e.g., surgical or medical treatments) were excluded.Comparator: Studies with any comparator group, including usual care, routine prenatal care, or alternative interventions (e.g., minimal lifestyle advice), were included. Studies without a comparator group (e.g., single-arm studies) were excluded from the meta-analysis but considered for narrative synthesis.Outcome: Studies reporting at least one of the primary outcomes QoL (measured using validated tools, e.g., SF-36, DQOL, or GDMQ-36) or FBG (measured in mmol/L or mg/dL) were included. Secondary outcomes, such as gestational weight gain, insulin use, or neonatal outcomes, were extracted if reported but were not required for inclusion. Studies that did not report QoL or FBG were excluded. Population: Studies involving women diagnosed with GDM, regardless of gestational age at diagnosis, were included. Studies focusing on women with pre-existing type 1 or type 2 diabetes, or those without a confirmed GDM diagnosis, were excluded. Population: Studies involving women diagnosed with GDM, regardless of gestational age at diagnosis, were included. Studies focusing on women with pre-existing type 1 or type 2 diabetes, or those without a confirmed GDM diagnosis, were excluded. Intervention: Studies evaluating lifestyle education interventions, defined as structured programs including dietary counseling, physical activity guidance, and/or self-management education (e.g., blood glucose monitoring), were included. Interventions could be delivered in-person, via digital platforms, or through hybrid formats. Studies focusing solely on pharmacological interventions or non-lifestyle interventions (e.g., surgical or medical treatments) were excluded. Intervention: Studies evaluating lifestyle education interventions, defined as structured programs including dietary counseling, physical activity guidance, and/or self-management education (e.g., blood glucose monitoring), were included. Interventions could be delivered in-person, via digital platforms, or through hybrid formats. Studies focusing solely on pharmacological interventions or non-lifestyle interventions (e.g., surgical or medical treatments) were excluded. Comparator: Studies with any comparator group, including usual care, routine prenatal care, or alternative interventions (e.g., minimal lifestyle advice), were included. Studies without a comparator group (e.g., single-arm studies) were excluded from the meta-analysis but considered for narrative synthesis. Comparator: Studies with any comparator group, including usual care, routine prenatal care, or alternative interventions (e.g., minimal lifestyle advice), were included. Studies without a comparator group (e.g., single-arm studies) were excluded from the meta-analysis but considered for narrative synthesis. Outcome: Studies reporting at least one of the primary outcomes QoL (measured using validated tools, e.g., SF-36, DQOL, or GDMQ-36) or FBG (measured in mmol/L or mg/dL) were included. Secondary outcomes, such as gestational weight gain, insulin use, or neonatal outcomes, were extracted if reported but were not required for inclusion. Studies that did not report QoL or FBG were excluded. Outcome: Studies reporting at least one of the primary outcomes QoL (measured using validated tools, e.g., SF-36, DQOL, or GDMQ-36) or FBG (measured in mmol/L or mg/dL) were included. Secondary outcomes, such as gestational weight gain, insulin use, or neonatal outcomes, were extracted if reported but were not required for inclusion. Studies that did not report QoL or FBG were excluded. RCTs, quasi-experimental studies, and cohort studies with a control group were included, but case studies, case series, reviews, and editorials were excluded. Peer-reviewed articles and grey literature were included if they provided sufficient data, while non-English studies or those with inaccessible full texts were excluded unless translation or access was feasible. Study selection and data extraction Two independent reviewers screened titles and abstracts against the inclusion and exclusion criteria using a standardized screening tool. Potentially eligible studies proceeded to full-text review, where the same reviewers assessed eligibility based on the PICO criteria and study design. Discrepancies between reviewers were resolved through discussion or consultation with a third reviewer to achieve consensus. A PRISMA flow diagram was generated to document the selection process, including reasons for exclusion at the full-text stage. Data extraction was performed independently by two reviewers using a pre-designed form aligned with the headings in Table 1. For studies with mixed populations, only outcome data from women with prior GDM were extracted when available. If separate GDM-subgroup data were not provided, the study was excluded from the relevant meta-analyses. The first author, publication year, and reference number were extracted to identify each study. The country where the study was conducted was recorded to provide geographical context. The study design, such as RCT or quasi-experimental, was documented to assess methodological rigor. The population, including sample size and gestational age at enrollment, was extracted to describe participant characteristics. Intervention components, including type, delivery mode, frequency, and content of lifestyle education, were detailed to understand the intervention’s structure. The control or comparator group, such as usual care or minimal intervention, was described to contextualize the comparison. The duration of the intervention, including specific time points like gestational weeks or postpartum periods, was recorded to assess intervention length. Primary outcomes, focusing on QoL and FBG, including measurement tools and time points, were extracted, along with secondary outcomes if reported. Key findings related to QoL, FBG, and other relevant outcomes, including statistical significance and effect sizes, were summarized to evaluate intervention efficacy. Extracted data were cross-checked for accuracy, discrepancies were resolved through discussion or consultation with a third reviewer, and corresponding authors were contacted for additional information when needed. Publication bias In this meta-analysis, the possibility of publication bias was examined using Egger’s and Begg’s statistical tests. Egger’s test in quality of life and Begg’s test in FBG studies yielded a p-value greater than 0.05, suggesting that there is no strong evidence to reject the null hypothesis. This implies that the presence of small-study effects and, thus, publication bias appears to be unlikely, although it cannot be entirely ruled out. Risk of bias The ROB-2 tool was used for the randomized controlled trials [17]. Since one of the included studies had a quasi‑experimental design, its risk of bias was assessed using the ROBINS‑I tool for non‑randomized studies. Statistical analysis All statistical analyses were conducted using StataCorp Stata version 17. For outcomes reported on the same measurement scale, the mean difference (MD) and its corresponding 95% confidence interval (CI) were computed. When studies used different scales or reported conceptually similar outcomes with heterogeneous units, the standardized mean difference (SMD) was calculated using Hedges’ g, defined as the mean difference between groups divided by the pooled standard deviation, with a small-sample correction applied. This approach follows the standard meta‑analytic recommendations for effect size estimation. Between‑study heterogeneity was assessed using the I² statistic, with values above 50% considered indicative of substantial heterogeneity. To account for expected variations across study populations and methodologies, a random‑effects model based on restricted maximum likelihood (REML) estimation was applied. Publication bias was evaluated using Egger’s regression test and Begg’s test. Additionally, sensitivity analyses were performed to assess the robustness of the pooled estimates, and subgroup analyses were conducted when sufficient data were available. A two‑sided P‑value < 0.05 was considered statistically significant. Results The process started by gathering potentially relevant records through database searches and other sources, yielding 265 records from databases and 39 from additional channels. After removing 68 duplicates, 236 records were left for screening. These records were assessed using predefined inclusion and exclusion criteria, with their titles and abstracts reviewed to determine eligibility. This initial review resulted in 31 records selected for a thorough full-text evaluation. During this stage, each record was closely scrutinized to confirm adherence to the inclusion criteria. As a result, 6 records were excluded due to an inappropriate study population, 3 were discarded for not addressing the intended outcome, and 2 were removed for lacking peer review. In the end, 20 studies were included in the final synthesis (see Fig. 1). Fig. 1Flowchart of the database search, selection, and review process of articles Flowchart of the database search, selection, and review process of articles Characteristics of included studies A total of 20 studies were included in this systematic review and meta-analysis, comprising 19 randomized controlled trials (RCTs) and 1 quasi-experimental study. The overall sample size across the included studies was 4,777 women, with individual study sample sizes ranging from 21 [18] to 1,601 [19] participants. Of note, one enrolled adults with metabolic syndrome; however, data from women with a history of GDM within that study were extracted and included where available [20]. The mean age of participants, where reported, ranged from approximately 29.0 [21] to 31.0 [22] years, with an overall weighted mean age of approximately 30.2 years. Gestational age at enrollment varied, ranging from early pregnancy (≤ 8 weeks gestation) [23] to the third trimester (up to 34 weeks) [24], with most studies enrolling women between 12 and 32 weeks gestation [22, 24–32]. Some studies also included postpartum women with a recent history of GDM, typically within 6 weeks to 18 months post-delivery [19, 21, 33–35]. The studies were conducted across 14 countries, reflecting geographic diversity. The distribution of studies by country was as follows: Iran (4 studies), China (3 studies), Sri Lanka (2 studies), Switzerland (1 study), Poland (1 study), Belgium (1 study), Hong Kong (1 study), United Arab Emirates (1 study), India (1 study, multi-country with Sri Lanka and Bangladesh), Singapore (1 study), Turkey (1 study), South Korea (1 study), Finland (1 study), Norway (1 study), and Brazil (1 study). The higher number of studies from Iran and China suggests significant research focus in these regions, potentially driven by high GDM prevalence or healthcare priorities. More details are included in Table 1. Intervention characteristics The lifestyle education interventions varied widely in their design, content, delivery mode, frequency, and duration, reflecting diverse approaches to supporting women with GDM. Most interventions incorporated a combination of dietary counseling, physical activity guidance, and self-management education, such as blood glucose monitoring. Dietary components typically emphasized balanced carbohydrate intake, portion control, and culturally appropriate food choices [35]. Physical activity recommendations often included moderate exercises like walking, prenatal yoga, or resistance training, with some studies specifying targets such as 150 min per week of moderate-intensity activity or ≥ 5,000 steps per day [25, 36]. Self-management education frequently focused on empowering women to monitor blood glucose, recognize hyperglycemia symptoms, and adhere to lifestyle recommendations, as seen in interventions incorporating blood glucose monitoring training [21, 22, 28, 32]. Delivery modes were diverse, encompassing face-to-face sessions (individual or group), digital platforms, and hybrid approaches. Face-to-face interventions ranged from weekly small-group sessions, such as 4 weekly lectures [22], to one-time group workshops supplemented by follow-ups, as implemented with four in-pregnancy lifestyle visits [21]. Other face-to-face formats included three 45-minute sessions over three days [24] and weekly group sessions for 12 weeks [20]. Digital interventions leveraged smartphone apps, telephonic coaching, or SMS reminders, exemplified by a mobile app with glucometer integration and tailored coaching [18], a smartphone app with monthly telephone coaching [33], and SMS-based follow-ups [34]. Hybrid models integrated in-person and digital components, such as in-person dietary and exercise sessions with [37] or face-to-face sessions with biweekly phone follow-ups [35]. The frequency of intervention components varied from intensive schedules, such as 50–60-minute exercise sessions three times per week [27], to less frequent interactions, such as five dietitian visits and monthly phone/SMS follow-ups over 12 months [34]. Other examples include biweekly counseling sessions [35], monthly visits [21], and two dietary phone consultations over several months [31]. Intervention durations ranged from short-term programs of 4 weeks [22] to extended interventions lasting up to 12 months, often spanning pregnancy and postpartum periods [19, 21, 35]. Several studies incorporated behavioral or theoretical frameworks to enhance intervention efficacy. For instance, the Health Belief Model was used to structure nutritional education [26], while the 5 A model guided self-management education [28]. Motivational interviewing and SMART goal-setting were employed to foster sustainable behavioral changes [36], and peer support workshops were integrated into the intervention [21]. The details regarding the included studies are presented in Table 1. Table 1Overall Characteristics of included studiesFirst author, Year, ReferenceCountryDesignPopulation (Sample Size, Gestational Age)Intervention ComponentsControl/ComparatorDuration of InterventionPrimary OutcomesKey FindingsQuansah et al. (2024) (21)SwitzerlandRandomized Controlled Trial211 pregnant women, ≥ 18 years, diagnosed with gestational diabetes mellitus at 24–32 weeks gestation, enrolled at mean 29.0 weeks gestationIntervention (N = 105): Added to usual care, four lifestyle visits in pregnancy (dietician, physiotherapist), four postpartum interdisciplinary visits, peer support workshops, and bimonthly lifestylUsual care (N = 106): Active guideline-based care (American Diabetes Association, Endocrine Society, NICE) with lifestyle advice, dietician visit, blood glucose monitoring, and postpartum follow-upFrom enrollment (mean 29.0 weeks gestation) to 1 year postpartum (September 2016–October 2021)Decrease in maternal weight and depression symptom scores (Edinburgh Postnatal Depression Scale) between baseline and 1 year postpartum- No significant weight change difference: -0.38 kg (-2.08 to 1.30, P = 0.653).- No significant depression score change: -0.67 (-1.84 to 0.49, P = 0.254).- Lower gestational weight gain since GDM visit: 1.89 kg vs. 3.07 kg, -1.20 kg (-2.14 to -0.26, P = 0.012).- Reduced weekly weight gain: 0.17 kg vs. 0.31 kg, -0.14 kg (-0.25 to -0.03, P = 0.008).- Higher proportion without weight retention at 1 year: 34% (28/82) vs. 21% (20/96, P = 0.034).- Lower insulin need: 32% (34/105) vs. 43% (45/106, P = 0.040).- No differences in glycaemic parameters, anxiety, wellbeing, or fitness (P ≥ 0.21).- Increased fat-free mass: 0.02 kg (0.01 to 0.03, P = 0.041).Adamczak et al. (2024) (25)PolandRandomized Controlled Trial124 pregnant women, singleton pregnancy, > 18 years, diagnosed with gestational diabetes < 20 weeks, obese (BMI ≥ 30 kg/m² or ≥ 25 kg/m² with complications), enrolled at median 16 weeks gestationPedometer group (N = 62): Advised to take ≥ 5000 steps/day, monitored daily step count using pedometers; both groups received diabetic diet education, caloric content recommendations, and weight gain guidanceControl group (N = 62): Received same diet and physical activity education but no pedometer or step count monitoringFrom enrollment (median 16 weeks) to 37–39 weeks gestationGestational weight gain (GWG), changes in maternal anthropometric parameters, HbA1c between baseline and 37–39 weeks- Significantly fewer newborns with macrosomia in pedometer group: 0% (0/53) vs. 10% (6/62) in control group (p = 0.03).- No significant difference in gestational weight gain between groups: 1.5 kg (-2.5–6.5) in pedometer group vs. 4.6 kg (0.5–8.0) in control group (p = 0.13).- Patients achieving > 5000 steps/day had lower weight gain: -2.7 kg (-7.4–0.75) vs. 6.0 kg (3.0–12.0) in < 5000 steps subgroup (p < 0.001).- ROC curve identified 4210 steps/day as cut-off for avoiding excessive weight gain (> 7 kg): AUC 0.92, sensitivity 0.92, specificity 0.88 (p = 0.00001).- No significant differences in HbA1c, triglycerides, or C-reactive protein levels between pedometer and control groups (p > 0.05).- No significant differences in preterm births, LGA, SGA, or Apgar scores between groups (p > 0.05).Minschart et al. (2024) (33)BelgiumRandomized Controlled Trial240 women with glucose intolerance post-GDM (Intervention: 121, Control: 119), 6–16 weeks postpartumBlended-care lifestyle program: 1 face-to-face session (1.5 h), smartphone app with diet/physical activity coaching, monthly telephone coachingUsual care: referral to primary care with annual diabetes screening recommendation1yearProportion achieving weight goal (≥ 5% weight loss if prepregnancy BMI ≥ 25 kg/m² or return to prepregnancy weight if BMI < 25 kg/m²)- No difference in weight goal achievement: 46.3% (56/121) in IG vs. 43.3% (52/119) in CG (OR 1.13, 95% CI 0.63–2.03, P = 0.680).- Reduced metabolic syndrome in IG: 7.3% (6/82) vs. 16.5% (14/85) in CG (OR 0.40, 95% CI 0.22–0.72, P = 0.002).- IG had less sedentary time (282.7 ± 144.3 vs. 398.4 ± 166.2 min/day, P = 0.032) and higher autonomous/controlled motivation (P = 0.001).- No difference in glucose intolerance, diabetes, or quality of life.Tsoi et al. (2024) (34)Hong KongRandomized Controlled Trial103 Chinese women aged 18–45 with recent GDM, 6–12 weeks postpartum (79 analyzed)12-month lifestyle program: 5 dietitian visits, 1 exercise session, 10 monthly phone/SMS follow-ups for diet, activity, weight lossStandard postnatal care, health booklet, educational videos12 monthsProportion achieving body weight goal- Intervention: 57.5% vs. control: 46.2% achieved weight goal (P = 0.314).- Greater weight loss (-0.93 kg vs. -0.01 kg, P = 0.36).- Reduced energy intake (-497.6 kcal vs. -222.0 kcal, P < 0.01).- Lower IFG/IGT progression (22.5% vs. 30.7%).Sundarapperuma et al. (2024) (35)Sri LankaRandomized Controlled Trial100 postpartum women with prior GDM, post-delivery of healthy singleton infant, from Colombo, Gampaha, and Galle districts (79 analyzed at 12 months)Lifestyle Intervention (N = 50): Culturally tailored diet (carbohydrate restriction, 1800–2425 kcal based on BMI) and exercise (household activities as energy expenditure), 30-min counseling, biweekly phone follow-ups, monthly visitsStandard Care (N = 50): Routine postpartum care, diet/activity diaries, no specific lifestyle guidance12 months post-delivery (2015–2016)Glycemic profile (FPG, 2-h post-OGTT, HbA1c), insulin resistance (HOMA-IR)- Intervention: Reduced FPG (87.3 vs. 123.2 mg/dl), 2-h OGTT (106.5 vs. 156.1 mg/dl), HbA1c (5.3 vs. 6.8%), HOMA-IR (0.9 vs. 2.3) at 12 months (all P < 0.01).- Control: Increased FPG, 2-h OGTT, HbA1c (P < 0.01), no HOMA-IR change.- T2DM incidence: 2% (intervention) vs. 34% (control).- Control group: 33-fold higher insulin resistance risk (adj. OR 32.9, P = 0.01).- Intervention: Lower BMI, WHR (P < 0.05).Sharifat et al. (2024) (26)IranRandomized Controlled Trial84 pregnant women at high risk for GDM (Intervention: 42, Control: 42), gestational age 12–16 weeks, mean age 30.12 ± 4.35 years12 weeks of nutritional education on glycemic index and load, including lectures and group discussions on healthy diet and lifestyle, based on Health Belief Model (HBM)Routine public health education12 weeks (from 12–16 weeks gestation to ~ 24–28 weeks)Health-related quality of life (SF-12), metabolic indicators (FBS, insulin, hs-CRP, lipids), HBM constructs- Intervention group showed increased perceived susceptibility (mean change 1.45 ± 4.21, P = 0.03) and cue to action (mean change 1.22 ± 3.38, P = 0.02) within group, but no significant between-group differences in HBM constructs (P > 0.05). - General Health subscale of SF-12 improved significantly in intervention group vs. control (mean change 13.69 ± 29.83 vs. 0.00 ± 29.58, P = 0.04, adjusted). - Significant between-group differences in hs-CRP (mean change − 0.877 ± 3.47 vs. -0.067 ± 3.40, P = 0.01, adjusted) and triglycerides (mean change − 55.24 ± 111.21 vs. 40.92 ± 142.01, P = 0.001, adjusted). - Total cholesterol increased less in intervention vs. control (mean change 22.75 ± 66.17 vs. 30.12 ± 61.33, P = 0.01, adjusted). - No significant differences in glycemic indices (FBS, insulin, HOMA-IR) between groups (P > 0.05).Xie et al. (2022) (27)ChinaRandomized Controlled Trial100 women with GDM, 24–31 weeks gestation (86 analyzed)Resistance exercise: 50–60 min, 3x/week, upper/lower limb training; Aerobic exercise: 50–60 min, 3x/week, step walking/stretchingAerobic exercise group6 weeksBlood glucose levels- Resistance group: lower 2-hour postprandial glucose (6.03 vs. 6.19 mmol/L, P = 0.019).- No difference in fasting glucose (P = 0.614) or insulin use (P = 1.000).- Higher compliance in resistance group (97.96% vs. 84.31%, P = 0.031).- No difference in adverse pregnancy outcomes (P > 0.05).Sadiya et al. (2022) (36)UAERandomized Controlled Trial63 pregnant women, aged 18–45 years, ≤ 12 weeks gestation, singleton pregnancy, with ≥ 2 risk factors for GDMLifestyle Intervention (LI, N = 30): 12-week moderate-intensity program with two face-to-face dietary consultations, two telephonic counseling sessions by a dietitian, targeting diet (50–55% carbs, 25–30% fat, 20% protein), 150 min/week moderate physical activity, and behavior change via motivational interviewing and SMART goalsUsual Care (UC, N = 33): Standard antenatal care with general lifestyle advice12 weeks, from ≤ 12 weeks gestation to 24–28 weeks gestationIncidence of GDM per IADPSG criteria at 24–28 weeks gestation- GDM incidence lower in LI: 33.3% (10/30) vs. 57.5% (19/33) in UC, RR = 0.59 (0.32–1.04, P = 0.05).- Unadjusted OR for GDM: 0.37 (0.13–1.02, P = 0.05).- Adjusted OR (age, BMI, family history): 0.26 (0.07–0.92, P = 0.04).- Reduced daily calorie intake in LI: -120 kcal (P = 0.004), carbohydrates: -19 g (P = 0.009), fat: -5 g (P = 0.03).- Increased physical activity in LI: +52 min/week (P = 0.05).- No significant differences in gestational weight gain, cesarean section, birth weight, or gestational age at delivery (P ≥ 0.15).- Fewer LI women exceeded IOM weight gain recommendations: 6.6% (2/30) vs. 21.2% (7/33, P = 0.05).Tandon et al. (2022) (19)India, Sri Lanka, BangladeshRandomized Controlled Trial1601 women with recent GDM, no type 2 diabetes at 3–18 months postpartum, OGTT at 24–34 weeks gestation (1308 analyzed)Lifestyle Intervention (N = 800): 12-month program with 4 group sessions, 2 individual sessions for weight gain, 84 voice/text messages, monthly calls, focus on diet and physical activityUsual Care (N = 801): No specific intervention, standard postpartum care12 monthsWorsening glycemic status (normal to prediabetes/type 2 diabetes, or prediabetes to type 2 diabetes)- No reduction in glycemic deterioration (25.5% intervention vs. 27.1% control, HR 0.92, P = 0.42).- No difference in type 2 diabetes incidence (9.3% vs. 10.0%, P = 0.48) or body weight change (+ 0.4 kg both groups).- High early postpartum dysglycemia (8.8% type 2 diabetes, 34.1% prediabetes at baseline).Rokni et al. (2022) (28)IranQuasi-Experimental Study54 women with GDM (Intervention: 27, Control: 27), Gestational age 28.40 ± 3.28 weeks (Intervention), 30.22 ± 3.20 weeks (Control)5 A model-based self-management education: 3 sessions (1.5 h each) on healthy eating, physical activity, stress and blood glucose management, with pamphlets and 2-month follow-up via phone callsRoutine education on healthy eating and physical activity per Ministry of Health recommendations2 months (3 sessions, follow-up via phone calls)Quality of life (DQOL), Blood glucose levels- Improved DQOL total score in IG (37.62 ± 10.89 to 48.29 ± 6.49, P < 0.001) vs. decline in CG (37.29 ± 4.05 to 36.62 ± 5.49, P = 0.02).- Reduced blood glucose in IG (101.96 ± 6.62 to 88.14 ± 6.28 mg/dL, P < 0.001) vs. no change in CG (103.37 ± 12.08 to 104.11 ± 16.01 mg/dL, P = 0.54).- Lower blood glucose in IG vs. CG at study end (P < 0.001).Deng et al. (2021) (37)ChinaRandomized Controlled Trial94 pregnant women with GDM risk factors (Intervention: 47, Control: 47), < 13 weeks gestation, mean age 29.64 ± 3.86 years (intervention), 29.64 ± 3.44 years (control)Diet and exercise program (7 sessions, every 2 weeks) with personalized nutrition (high-fiber, low-fat diet, 50–60% carbohydrates) and moderate exercise (20–30 min, 5 days/week, e.g., walking, yoga), supported by diaries, WeChat reminders, and outpatient/WeChat adjustments, plus routine careRoutine prenatal care only~ 14 weeks (from 14 weeks to 24–28 weeks gestation)Incidence of GDM, gestational weight gain (GWG)- GDM incidence significantly lower in intervention vs. control group (23.9% vs. 51.1%; χ²=7.301, P = 0.007). - FPG significantly lower in intervention vs. control (median: 4.53 vs. 4.74 mmol/L; P = 0.045). - 2-hour PG post-OGTT significantly lower in intervention vs. control (median: 6.42 vs. 7.26 mmol/L; P = 0.002). - No significant difference in 1-hour PG post-OGTT (median: 7.45 vs. 8.18 mmol/L; P = 0.161). - No significant difference in GWG from enrollment to 24–28 weeks (mean: 5.65 vs. 6.13 kg; P = 0.399) or to delivery (mean: 6.49 vs. 5.44 kg; P = 0.089). - Cesarean section rate higher in intervention vs. control (58.5% vs. 27.9%; P = 0.005), but GDM-related cesarean indications similar (3 vs. 3 cases). - No significant differences in gestational hypertension (4.9% vs. 9.3%; P = 0.716), premature infants (2.4% vs. 9.3%; P = 0.386), or macrosomia (9.8% vs. 14.0%; P = 0.797).Yew et al. (2021) (29)SingaporeRandomized Controlled Trial340 women with GDM, 12–30 weeks gestation340 women with GDM, 12–30 weeks gestationUsual care (face-to-face education, manual glucose logging)From recruitment (mean 27 weeks) to delivery (mean 38 weeks)Proportion with excessive gestational weight gain (EGWG)- No difference in EGWG (OR 1.55, P = 0.152) or absolute GWG (P = 0.207).- Lower average glucose (-0.15 mmol/L, P = 0.011) and glucose above targets (premeal: 17.9% vs. 23.3%, P = 0.003; postmeal: 19.9% vs. 50%, P < 0.001).- Lower composite neonatal complications (38.1% vs. 53.7%, P = 0.006).Ural et al. (2021) (24)TurkeyRandomized Controlled Trial88 women with GDM (Intervention: 46, Control: 42), Gestational age 24–34 weeksHealth-Promoting Lifestyle Education Program (HPLEP): 3 sessions on self-monitoring blood glucose, nutrition, physical activity, stress management, sleep, smoking cessation, self-care, breastfeeding; with booklet, diary card, follow-up callsUsual care (routine diabetes follow-up: blood glucose monitoring, dietician/diabetes nurse referral)Three 45-minute sessions over 3 days, follow-up calls at 2 weeks post-intervention and 2nd week postpartumMaternal lifestyle (HPLP-II), quality of life (SF-36), depression (CES-D), neonatal health- Improved lifestyle (HPLP-II total: 141.20 ± 17.82 vs. 128.45 ± 20.72, P = 0.003) and quality of life (e.g., SF-36 physical role: 76.09 ± 32.89 vs. 39.88 ± 37.06, P < 0.001) in intervention group.- No significant depression change (CES-D: 14.93 ± 9.39 vs. 15.74 ± 8.54, P = 0.405).- Lower macrosomia rates (6.5% vs. 28.6%, P = 0.016) and higher breastfeeding (76.1% vs. 50%, P = 0.011) in intervention neonates.− 100% vs. 7.1% postpartum diabetes assessment (P < 0.001).Sung et al. (2019) (18)South KoreaRandomized Controlled Trial21 pregnant women with GDM, diagnosed at 24–28 weeks gestation (19 analyzed)Mobile Management (N = 11): Mobile app with glucometer, accelerometer, tailored coaching (diet/exercise messages twice weekly), standard antenatal careConventional Management (N = 10): Standard antenatal care (biweekly visits, blood glucose/diet logs), optional app access without coachingFrom 24–28 weeks gestation to deliveryPostpartum metabolic outcomes (BMI, HOMA-IR, body fat), obstetric outcomes- MM group: Lower postpartum BMI (20.22 vs. 23.72 kg/m², P = 0.021), weight (54.31 vs. 62.58 kg, P = 0.037), body fat (29.20% vs. 38.12%, P = 0.005), HOMA-IR (1.46 vs. 1.97, P = 0.011).< br>- No difference in HbA1c, fasting glucose, or obstetric outcomes (cesarean, large for gestational age, P > 0.05).< br>- Improved dietary habits, self-efficacy in MM group (P < 0.05).Sahrakorpi et al. (2019) (30)FinlandRandomized Controlled Trial378 pregnant women at high risk for GDM (Intervention: 192, Control: 186), mean gestational age at baseline ~ 13 weeksStructured, individually tailored dietary and physical activity counselling by study nurses at 3 pregnancy visits; one 2-hour group nutrition session by dietitian; based on Nordic Nutrition Recommendations, targeting 150 min/week moderate physical activityUsual care with information leaflets on healthy diet and exercise, plus routine antenatal clinic visitsFrom early pregnancy (~ 13 weeks) to 12 months postpartumHealth-related quality of life (HRQoL) measured by 15D instrument- No significant differences in 15D HRQoL scores between intervention and control groups during pregnancy or up to 12 months postpartum (adjusted for age, pre-pregnancy BMI). - HRQoL decreased similarly in both groups toward third trimester (P-value not reported). - No correlation between GDM diagnosis and HRQoL at any stage. - Cumulative GDM incidence similar (Intervention: 45.7%, Control: 44.5%, P = 0.82).Saboya et al. (2017) (20)BrazilRandomized Controlled TrialWomen with GDM; only GDM-subgroup data extracted for this review: n = 28 with GDM in combined analysis), aged 30–59 yearsGI: Weekly 1h45m group sessions (10–12 participants) with health education, diet/exercise strategies, led by psychologist, nurse, physiotherapist, nutritionist. II: Weekly individual nutrition/psychology sessions, 36 treadmill sessions (60 min, 3x/week, 75–85% max HR)SI: Two individual consultations (baseline, 3 months) with standard diet/exercise/self-care guidance12 weeksChange in metabolic parameters (BMI, WC, BP, glucose, lipids)- Among women with GDM (data extracted for this review): Significant reductions at 3 months in BMI (GI: 33.3 ± 0.3, II: 32.2 ± 0.2 vs. SI: 33.7 ± 0.3, P < 0.001), WC (GI: 106.5 ± 1.0, II: 105.4 ± 1.0 vs. SI: 110.2 ± 1.2, P = 0.009). QoL improved in physical functioning domain (P = 0.024). No significant changes in FG, TGL, HDL-C, DBP. Effects not sustained at 9 months.Zandinava et al. (2017) (22)IranRandomized Controlled Trial92 pregnant women with GDM (Intervention: 46, Control: 46), gestational age 24–28 weeks, mean age 31.0 ± 5.0 years4 weekly sessions of self-care education on diabetes (lecture, Q&A, booklet) focusing on physical activity, nutrition, and complication prevention, plus routine prenatal careRoutine prenatal care only4 weeks (from 24–28 weeks gestation to ~ 28–32 weeks)Self-care behavior score, quality of life (QOL-GRAV), FBS, GTT (1-hour and 2-hour post-load glucose)- Self-care behavior score significantly higher in intervention vs. control group post-intervention (adjusted mean difference: 19.5; 95% CI: 14.4 to 24.6; P < 0.001). - No significant difference in quality of life between groups (adjusted mean difference: -0.82; 95% CI: -2.2 to 0.6; P = 0.264). - No significant difference in FBS between groups (adjusted mean difference: -1.09; 95% CI: -3.8 to 1.7; P = 0.443). - Significant reduction in 1-hour GTT glucose in intervention vs. control (adjusted mean difference: -21.6; 95% CI: -32.1 to -11.1; P < 0.001). - Significant reduction in 2-hour GTT glucose in intervention vs. control (adjusted mean difference: -17.3; 95% CI: -23.0 to -11.6; P < 0.001).Sagedal et al. (2017) (31)NorwayRandomized Controlled Trial606 nulliparous women, aged ≥ 18 years, pre-pregnancy BMI ≥ 19 kg/m², singleton pregnancy, ≤ 20 weeks gestation (557 analyzed)Lifestyle Intervention (N = 281): Two dietary phone consultations (healthy food choices, no calorie restriction), access to twice-weekly exercise classes, encouraged 30 min moderate activity 3 days/week, plus educational materialsRoutine Prenatal Care (N = 276): Standard care with 8 prenatal visits, one ultrasound, and general lifestyle adviceFrom ≤ 20 weeks to 30 weeks gestationGestational weight gain (GWG), maternal glucose levels at 30 weeks- Reduced GWG in intervention: -1.2 kg (P = 0.021).- Lower insulin: -0.91 mU/L (P = 0.045) and leptin: -207.8 pmol/L (P = 0.021).- No effect on fasting or 2-h glucose levels (P ≥ 0.776).- Higher glucose exceeding IADPSG thresholds in intervention, especially overweight/obese (33.7% vs. 13.9%, adj. OR 3.89, P = 0.004).< br>- Normal-weight women: trend to lower insulin/HOMA-IR (P = 0.056).- No difference in GDM incidence (WHO criteria) or neonatal outcomes (P > 0.05).Mirfeizi et al. (2017) (32)IranRandomized Controlled Trial149 pregnant women with GDM (Nutrition+Education: 32, Insulin+Education: 47, Nutrition Control: 35, Insulin Control: 35), mean gestational age 22.78 ± 7.95 weeksEducational intervention: weekly 45-min face-to-face sessions (4 patients/session) for 2 months, covering GDM nature, complications, diet, exercise, SMBG, delivery, postpartum care; booklet provided; bi-monthly telecommunicationStandard prenatal care without additional education2 monthsQoL score measured by Iranian Diabetes Quality of Life Brief Clinical Inventory (IDQOL-BCI)- QoL improved in all groups post-intervention (P < 0.001). - Greatest QoL improvement in Insulin+Education group (mean difference: 16.43, P < 0.001), followed by Nutrition+Education (11.64, P < 0.001), Insulin Control (9.01, P < 0.001), Nutrition Control (7.27, P < 0.001). - Significant reduction in 1-h blood glucose post-intervention (P < 0.001), no significant changes in FBG, 2-h BG, or HbA1c. - No significant between-group differences in QoL after adjusting for base line (P = 0.814).Wang et al. (2015) (23)ChinaCluster-Randomized Controlled Trial272 pregnant women with ≥ 1 GDM risk factor (Intervention: 134, Control: 138), < 8 weeks gestation, mean age 31.01 ± 3.8 years (intervention), 30.27 ± 3.64 years (control)Two-step lifestyle intervention: (1) 3 group courses (40–60 min each, < 6 participants) at 6–8 weeks on balanced diet, physical activity (e.g., 30 min walking daily), and weight gain goals; (2) nutritional counseling and enforcement at 12–13 weeks based on anthropometrics, plus routine prenatal careRoutine prenatal care only~ 6–7 weeks (from 6–8 weeks to 12–13 weeks gestation)Incidence of GDM, gestational weight gain (GWG)- GDM incidence lower in intervention vs. control group (17.16% [23/134] vs. 23.91% [33/138]; P = 0.168), not statistically significant. - GWG in first trimester similar between intervention and control (1.38 ± 2.34 kg vs. 1.41 ± 2.58 kg; P = 0.905). - GWG in second trimester similar between intervention and control (5.51 ± 2.18 kg vs. 5.66 ± 2.25 kg; P = 0.567). - For women with FPG ≥ 5.1 mmol/L in early pregnancy, GDM incidence lower in intervention vs. control (30.55% [11/36] vs. 45.95% [17/37]; P = 0.076), not statistically significant. Table 1Overall Characteristics of included studiesFirst author, Year, ReferenceCountryDesignPopulation (Sample Size, Gestational Age)Intervention ComponentsControl/ComparatorDuration of InterventionPrimary OutcomesKey FindingsQuansah et al. (2024) (21)SwitzerlandRandomized Controlled Trial211 pregnant women, ≥ 18 years, diagnosed with gestational diabetes mellitus at 24–32 weeks gestation, enrolled at mean 29.0 weeks gestationIntervention (N = 105): Added to usual care, four lifestyle visits in pregnancy (dietician, physiotherapist), four postpartum interdisciplinary visits, peer support workshops, and bimonthly lifestylUsual care (N = 106): Active guideline-based care (American Diabetes Association, Endocrine Society, NICE) with lifestyle advice, dietician visit, blood glucose monitoring, and postpartum follow-upFrom enrollment (mean 29.0 weeks gestation) to 1 year postpartum (September 2016–October 2021)Decrease in maternal weight and depression symptom scores (Edinburgh Postnatal Depression Scale) between baseline and 1 year postpartum- No significant weight change difference: -0.38 kg (-2.08 to 1.30, P = 0.653).- No significant depression score change: -0.67 (-1.84 to 0.49, P = 0.254).- Lower gestational weight gain since GDM visit: 1.89 kg vs. 3.07 kg, -1.20 kg (-2.14 to -0.26, P = 0.012).- Reduced weekly weight gain: 0.17 kg vs. 0.31 kg, -0.14 kg (-0.25 to -0.03, P = 0.008).- Higher proportion without weight retention at 1 year: 34% (28/82) vs. 21% (20/96, P = 0.034).- Lower insulin need: 32% (34/105) vs. 43% (45/106, P = 0.040).- No differences in glycaemic parameters, anxiety, wellbeing, or fitness (P ≥ 0.21).- Increased fat-free mass: 0.02 kg (0.01 to 0.03, P = 0.041).Adamczak et al. (2024) (25)PolandRandomized Controlled Trial124 pregnant women, singleton pregnancy, > 18 years, diagnosed with gestational diabetes < 20 weeks, obese (BMI ≥ 30 kg/m² or ≥ 25 kg/m² with complications), enrolled at median 16 weeks gestationPedometer group (N = 62): Advised to take ≥ 5000 steps/day, monitored daily step count using pedometers; both groups received diabetic diet education, caloric content recommendations, and weight gain guidanceControl group (N = 62): Received same diet and physical activity education but no pedometer or step count monitoringFrom enrollment (median 16 weeks) to 37–39 weeks gestationGestational weight gain (GWG), changes in maternal anthropometric parameters, HbA1c between baseline and 37–39 weeks- Significantly fewer newborns with macrosomia in pedometer group: 0% (0/53) vs. 10% (6/62) in control group (p = 0.03).- No significant difference in gestational weight gain between groups: 1.5 kg (-2.5–6.5) in pedometer group vs. 4.6 kg (0.5–8.0) in control group (p = 0.13).- Patients achieving > 5000 steps/day had lower weight gain: -2.7 kg (-7.4–0.75) vs. 6.0 kg (3.0–12.0) in < 5000 steps subgroup (p < 0.001).- ROC curve identified 4210 steps/day as cut-off for avoiding excessive weight gain (> 7 kg): AUC 0.92, sensitivity 0.92, specificity 0.88 (p = 0.00001).- No significant differences in HbA1c, triglycerides, or C-reactive protein levels between pedometer and control groups (p > 0.05).- No significant differences in preterm births, LGA, SGA, or Apgar scores between groups (p > 0.05).Minschart et al. (2024) (33)BelgiumRandomized Controlled Trial240 women with glucose intolerance post-GDM (Intervention: 121, Control: 119), 6–16 weeks postpartumBlended-care lifestyle program: 1 face-to-face session (1.5 h), smartphone app with diet/physical activity coaching, monthly telephone coachingUsual care: referral to primary care with annual diabetes screening recommendation1yearProportion achieving weight goal (≥ 5% weight loss if prepregnancy BMI ≥ 25 kg/m² or return to prepregnancy weight if BMI < 25 kg/m²)- No difference in weight goal achievement: 46.3% (56/121) in IG vs. 43.3% (52/119) in CG (OR 1.13, 95% CI 0.63–2.03, P = 0.680).- Reduced metabolic syndrome in IG: 7.3% (6/82) vs. 16.5% (14/85) in CG (OR 0.40, 95% CI 0.22–0.72, P = 0.002).- IG had less sedentary time (282.7 ± 144.3 vs. 398.4 ± 166.2 min/day, P = 0.032) and higher autonomous/controlled motivation (P = 0.001).- No difference in glucose intolerance, diabetes, or quality of life.Tsoi et al. (2024) (34)Hong KongRandomized Controlled Trial103 Chinese women aged 18–45 with recent GDM, 6–12 weeks postpartum (79 analyzed)12-month lifestyle program: 5 dietitian visits, 1 exercise session, 10 monthly phone/SMS follow-ups for diet, activity, weight lossStandard postnatal care, health booklet, educational videos12 monthsProportion achieving body weight goal- Intervention: 57.5% vs. control: 46.2% achieved weight goal (P = 0.314).- Greater weight loss (-0.93 kg vs. -0.01 kg, P = 0.36).- Reduced energy intake (-497.6 kcal vs. -222.0 kcal, P < 0.01).- Lower IFG/IGT progression (22.5% vs. 30.7%).Sundarapperuma et al. (2024) (35)Sri LankaRandomized Controlled Trial100 postpartum women with prior GDM, post-delivery of healthy singleton infant, from Colombo, Gampaha, and Galle districts (79 analyzed at 12 months)Lifestyle Intervention (N = 50): Culturally tailored diet (carbohydrate restriction, 1800–2425 kcal based on BMI) and exercise (household activities as energy expenditure), 30-min counseling, biweekly phone follow-ups, monthly visitsStandard Care (N = 50): Routine postpartum care, diet/activity diaries, no specific lifestyle guidance12 months post-delivery (2015–2016)Glycemic profile (FPG, 2-h post-OGTT, HbA1c), insulin resistance (HOMA-IR)- Intervention: Reduced FPG (87.3 vs. 123.2 mg/dl), 2-h OGTT (106.5 vs. 156.1 mg/dl), HbA1c (5.3 vs. 6.8%), HOMA-IR (0.9 vs. 2.3) at 12 months (all P < 0.01).- Control: Increased FPG, 2-h OGTT, HbA1c (P < 0.01), no HOMA-IR change.- T2DM incidence: 2% (intervention) vs. 34% (control).- Control group: 33-fold higher insulin resistance risk (adj. OR 32.9, P = 0.01).- Intervention: Lower BMI, WHR (P < 0.05).Sharifat et al. (2024) (26)IranRandomized Controlled Trial84 pregnant women at high risk for GDM (Intervention: 42, Control: 42), gestational age 12–16 weeks, mean age 30.12 ± 4.35 years12 weeks of nutritional education on glycemic index and load, including lectures and group discussions on healthy diet and lifestyle, based on Health Belief Model (HBM)Routine public health education12 weeks (from 12–16 weeks gestation to ~ 24–28 weeks)Health-related quality of life (SF-12), metabolic indicators (FBS, insulin, hs-CRP, lipids), HBM constructs- Intervention group showed increased perceived susceptibility (mean change 1.45 ± 4.21, P = 0.03) and cue to action (mean change 1.22 ± 3.38, P = 0.02) within group, but no significant between-group differences in HBM constructs (P > 0.05). - General Health subscale of SF-12 improved significantly in intervention group vs. control (mean change 13.69 ± 29.83 vs. 0.00 ± 29.58, P = 0.04, adjusted). - Significant between-group differences in hs-CRP (mean change − 0.877 ± 3.47 vs. -0.067 ± 3.40, P = 0.01, adjusted) and triglycerides (mean change − 55.24 ± 111.21 vs. 40.92 ± 142.01, P = 0.001, adjusted). - Total cholesterol increased less in intervention vs. control (mean change 22.75 ± 66.17 vs. 30.12 ± 61.33, P = 0.01, adjusted). - No significant differences in glycemic indices (FBS, insulin, HOMA-IR) between groups (P > 0.05).Xie et al. (2022) (27)ChinaRandomized Controlled Trial100 women with GDM, 24–31 weeks gestation (86 analyzed)Resistance exercise: 50–60 min, 3x/week, upper/lower limb training; Aerobic exercise: 50–60 min, 3x/week, step walking/stretchingAerobic exercise group6 weeksBlood glucose levels- Resistance group: lower 2-hour postprandial glucose (6.03 vs. 6.19 mmol/L, P = 0.019).- No difference in fasting glucose (P = 0.614) or insulin use (P = 1.000).- Higher compliance in resistance group (97.96% vs. 84.31%, P = 0.031).- No difference in adverse pregnancy outcomes (P > 0.05).Sadiya et al. (2022) (36)UAERandomized Controlled Trial63 pregnant women, aged 18–45 years, ≤ 12 weeks gestation, singleton pregnancy, with ≥ 2 risk factors for GDMLifestyle Intervention (LI, N = 30): 12-week moderate-intensity program with two face-to-face dietary consultations, two telephonic counseling sessions by a dietitian, targeting diet (50–55% carbs, 25–30% fat, 20% protein), 150 min/week moderate physical activity, and behavior change via motivational interviewing and SMART goalsUsual Care (UC, N = 33): Standard antenatal care with general lifestyle advice12 weeks, from ≤ 12 weeks gestation to 24–28 weeks gestationIncidence of GDM per IADPSG criteria at 24–28 weeks gestation- GDM incidence lower in LI: 33.3% (10/30) vs. 57.5% (19/33) in UC, RR = 0.59 (0.32–1.04, P = 0.05).- Unadjusted OR for GDM: 0.37 (0.13–1.02, P = 0.05).- Adjusted OR (age, BMI, family history): 0.26 (0.07–0.92, P = 0.04).- Reduced daily calorie intake in LI: -120 kcal (P = 0.004), carbohydrates: -19 g (P = 0.009), fat: -5 g (P = 0.03).- Increased physical activity in LI: +52 min/week (P = 0.05).- No significant differences in gestational weight gain, cesarean section, birth weight, or gestational age at delivery (P ≥ 0.15).- Fewer LI women exceeded IOM weight gain recommendations: 6.6% (2/30) vs. 21.2% (7/33, P = 0.05).Tandon et al. (2022) (19)India, Sri Lanka, BangladeshRandomized Controlled Trial1601 women with recent GDM, no type 2 diabetes at 3–18 months postpartum, OGTT at 24–34 weeks gestation (1308 analyzed)Lifestyle Intervention (N = 800): 12-month program with 4 group sessions, 2 individual sessions for weight gain, 84 voice/text messages, monthly calls, focus on diet and physical activityUsual Care (N = 801): No specific intervention, standard postpartum care12 monthsWorsening glycemic status (normal to prediabetes/type 2 diabetes, or prediabetes to type 2 diabetes)- No reduction in glycemic deterioration (25.5% intervention vs. 27.1% control, HR 0.92, P = 0.42).- No difference in type 2 diabetes incidence (9.3% vs. 10.0%, P = 0.48) or body weight change (+ 0.4 kg both groups).- High early postpartum dysglycemia (8.8% type 2 diabetes, 34.1% prediabetes at baseline).Rokni et al. (2022) (28)IranQuasi-Experimental Study54 women with GDM (Intervention: 27, Control: 27), Gestational age 28.40 ± 3.28 weeks (Intervention), 30.22 ± 3.20 weeks (Control)5 A model-based self-management education: 3 sessions (1.5 h each) on healthy eating, physical activity, stress and blood glucose management, with pamphlets and 2-month follow-up via phone callsRoutine education on healthy eating and physical activity per Ministry of Health recommendations2 months (3 sessions, follow-up via phone calls)Quality of life (DQOL), Blood glucose levels- Improved DQOL total score in IG (37.62 ± 10.89 to 48.29 ± 6.49, P < 0.001) vs. decline in CG (37.29 ± 4.05 to 36.62 ± 5.49, P = 0.02).- Reduced blood glucose in IG (101.96 ± 6.62 to 88.14 ± 6.28 mg/dL, P < 0.001) vs. no change in CG (103.37 ± 12.08 to 104.11 ± 16.01 mg/dL, P = 0.54).- Lower blood glucose in IG vs. CG at study end (P < 0.001).Deng et al. (2021) (37)ChinaRandomized Controlled Trial94 pregnant women with GDM risk factors (Intervention: 47, Control: 47), < 13 weeks gestation, mean age 29.64 ± 3.86 years (intervention), 29.64 ± 3.44 years (control)Diet and exercise program (7 sessions, every 2 weeks) with personalized nutrition (high-fiber, low-fat diet, 50–60% carbohydrates) and moderate exercise (20–30 min, 5 days/week, e.g., walking, yoga), supported by diaries, WeChat reminders, and outpatient/WeChat adjustments, plus routine careRoutine prenatal care only~ 14 weeks (from 14 weeks to 24–28 weeks gestation)Incidence of GDM, gestational weight gain (GWG)- GDM incidence significantly lower in intervention vs. control group (23.9% vs. 51.1%; χ²=7.301, P = 0.007). - FPG significantly lower in intervention vs. control (median: 4.53 vs. 4.74 mmol/L; P = 0.045). - 2-hour PG post-OGTT significantly lower in intervention vs. control (median: 6.42 vs. 7.26 mmol/L; P = 0.002). - No significant difference in 1-hour PG post-OGTT (median: 7.45 vs. 8.18 mmol/L; P = 0.161). - No significant difference in GWG from enrollment to 24–28 weeks (mean: 5.65 vs. 6.13 kg; P = 0.399) or to delivery (mean: 6.49 vs. 5.44 kg; P = 0.089). - Cesarean section rate higher in intervention vs. control (58.5% vs. 27.9%; P = 0.005), but GDM-related cesarean indications similar (3 vs. 3 cases). - No significant differences in gestational hypertension (4.9% vs. 9.3%; P = 0.716), premature infants (2.4% vs. 9.3%; P = 0.386), or macrosomia (9.8% vs. 14.0%; P = 0.797).Yew et al. (2021) (29)SingaporeRandomized Controlled Trial340 women with GDM, 12–30 weeks gestation340 women with GDM, 12–30 weeks gestationUsual care (face-to-face education, manual glucose logging)From recruitment (mean 27 weeks) to delivery (mean 38 weeks)Proportion with excessive gestational weight gain (EGWG)- No difference in EGWG (OR 1.55, P = 0.152) or absolute GWG (P = 0.207).- Lower average glucose (-0.15 mmol/L, P = 0.011) and glucose above targets (premeal: 17.9% vs. 23.3%, P = 0.003; postmeal: 19.9% vs. 50%, P < 0.001).- Lower composite neonatal complications (38.1% vs. 53.7%, P = 0.006).Ural et al. (2021) (24)TurkeyRandomized Controlled Trial88 women with GDM (Intervention: 46, Control: 42), Gestational age 24–34 weeksHealth-Promoting Lifestyle Education Program (HPLEP): 3 sessions on self-monitoring blood glucose, nutrition, physical activity, stress management, sleep, smoking cessation, self-care, breastfeeding; with booklet, diary card, follow-up callsUsual care (routine diabetes follow-up: blood glucose monitoring, dietician/diabetes nurse referral)Three 45-minute sessions over 3 days, follow-up calls at 2 weeks post-intervention and 2nd week postpartumMaternal lifestyle (HPLP-II), quality of life (SF-36), depression (CES-D), neonatal health- Improved lifestyle (HPLP-II total: 141.20 ± 17.82 vs. 128.45 ± 20.72, P = 0.003) and quality of life (e.g., SF-36 physical role: 76.09 ± 32.89 vs. 39.88 ± 37.06, P < 0.001) in intervention group.- No significant depression change (CES-D: 14.93 ± 9.39 vs. 15.74 ± 8.54, P = 0.405).- Lower macrosomia rates (6.5% vs. 28.6%, P = 0.016) and higher breastfeeding (76.1% vs. 50%, P = 0.011) in intervention neonates.− 100% vs. 7.1% postpartum diabetes assessment (P < 0.001).Sung et al. (2019) (18)South KoreaRandomized Controlled Trial21 pregnant women with GDM, diagnosed at 24–28 weeks gestation (19 analyzed)Mobile Management (N = 11): Mobile app with glucometer, accelerometer, tailored coaching (diet/exercise messages twice weekly), standard antenatal careConventional Management (N = 10): Standard antenatal care (biweekly visits, blood glucose/diet logs), optional app access without coachingFrom 24–28 weeks gestation to deliveryPostpartum metabolic outcomes (BMI, HOMA-IR, body fat), obstetric outcomes- MM group: Lower postpartum BMI (20.22 vs. 23.72 kg/m², P = 0.021), weight (54.31 vs. 62.58 kg, P = 0.037), body fat (29.20% vs. 38.12%, P = 0.005), HOMA-IR (1.46 vs. 1.97, P = 0.011).< br>- No difference in HbA1c, fasting glucose, or obstetric outcomes (cesarean, large for gestational age, P > 0.05).< br>- Improved dietary habits, self-efficacy in MM group (P < 0.05).Sahrakorpi et al. (2019) (30)FinlandRandomized Controlled Trial378 pregnant women at high risk for GDM (Intervention: 192, Control: 186), mean gestational age at baseline ~ 13 weeksStructured, individually tailored dietary and physical activity counselling by study nurses at 3 pregnancy visits; one 2-hour group nutrition session by dietitian; based on Nordic Nutrition Recommendations, targeting 150 min/week moderate physical activityUsual care with information leaflets on healthy diet and exercise, plus routine antenatal clinic visitsFrom early pregnancy (~ 13 weeks) to 12 months postpartumHealth-related quality of life (HRQoL) measured by 15D instrument- No significant differences in 15D HRQoL scores between intervention and control groups during pregnancy or up to 12 months postpartum (adjusted for age, pre-pregnancy BMI). - HRQoL decreased similarly in both groups toward third trimester (P-value not reported). - No correlation between GDM diagnosis and HRQoL at any stage. - Cumulative GDM incidence similar (Intervention: 45.7%, Control: 44.5%, P = 0.82).Saboya et al. (2017) (20)BrazilRandomized Controlled TrialWomen with GDM; only GDM-subgroup data extracted for this review: n = 28 with GDM in combined analysis), aged 30–59 yearsGI: Weekly 1h45m group sessions (10–12 participants) with health education, diet/exercise strategies, led by psychologist, nurse, physiotherapist, nutritionist. II: Weekly individual nutrition/psychology sessions, 36 treadmill sessions (60 min, 3x/week, 75–85% max HR)SI: Two individual consultations (baseline, 3 months) with standard diet/exercise/self-care guidance12 weeksChange in metabolic parameters (BMI, WC, BP, glucose, lipids)- Among women with GDM (data extracted for this review): Significant reductions at 3 months in BMI (GI: 33.3 ± 0.3, II: 32.2 ± 0.2 vs. SI: 33.7 ± 0.3, P < 0.001), WC (GI: 106.5 ± 1.0, II: 105.4 ± 1.0 vs. SI: 110.2 ± 1.2, P = 0.009). QoL improved in physical functioning domain (P = 0.024). No significant changes in FG, TGL, HDL-C, DBP. Effects not sustained at 9 months.Zandinava et al. (2017) (22)IranRandomized Controlled Trial92 pregnant women with GDM (Intervention: 46, Control: 46), gestational age 24–28 weeks, mean age 31.0 ± 5.0 years4 weekly sessions of self-care education on diabetes (lecture, Q&A, booklet) focusing on physical activity, nutrition, and complication prevention, plus routine prenatal careRoutine prenatal care only4 weeks (from 24–28 weeks gestation to ~ 28–32 weeks)Self-care behavior score, quality of life (QOL-GRAV), FBS, GTT (1-hour and 2-hour post-load glucose)- Self-care behavior score significantly higher in intervention vs. control group post-intervention (adjusted mean difference: 19.5; 95% CI: 14.4 to 24.6; P < 0.001). - No significant difference in quality of life between groups (adjusted mean difference: -0.82; 95% CI: -2.2 to 0.6; P = 0.264). - No significant difference in FBS between groups (adjusted mean difference: -1.09; 95% CI: -3.8 to 1.7; P = 0.443). - Significant reduction in 1-hour GTT glucose in intervention vs. control (adjusted mean difference: -21.6; 95% CI: -32.1 to -11.1; P < 0.001). - Significant reduction in 2-hour GTT glucose in intervention vs. control (adjusted mean difference: -17.3; 95% CI: -23.0 to -11.6; P < 0.001).Sagedal et al. (2017) (31)NorwayRandomized Controlled Trial606 nulliparous women, aged ≥ 18 years, pre-pregnancy BMI ≥ 19 kg/m², singleton pregnancy, ≤ 20 weeks gestation (557 analyzed)Lifestyle Intervention (N = 281): Two dietary phone consultations (healthy food choices, no calorie restriction), access to twice-weekly exercise classes, encouraged 30 min moderate activity 3 days/week, plus educational materialsRoutine Prenatal Care (N = 276): Standard care with 8 prenatal visits, one ultrasound, and general lifestyle adviceFrom ≤ 20 weeks to 30 weeks gestationGestational weight gain (GWG), maternal glucose levels at 30 weeks- Reduced GWG in intervention: -1.2 kg (P = 0.021).- Lower insulin: -0.91 mU/L (P = 0.045) and leptin: -207.8 pmol/L (P = 0.021).- No effect on fasting or 2-h glucose levels (P ≥ 0.776).- Higher glucose exceeding IADPSG thresholds in intervention, especially overweight/obese (33.7% vs. 13.9%, adj. OR 3.89, P = 0.004).< br>- Normal-weight women: trend to lower insulin/HOMA-IR (P = 0.056).- No difference in GDM incidence (WHO criteria) or neonatal outcomes (P > 0.05).Mirfeizi et al. (2017) (32)IranRandomized Controlled Trial149 pregnant women with GDM (Nutrition+Education: 32, Insulin+Education: 47, Nutrition Control: 35, Insulin Control: 35), mean gestational age 22.78 ± 7.95 weeksEducational intervention: weekly 45-min face-to-face sessions (4 patients/session) for 2 months, covering GDM nature, complications, diet, exercise, SMBG, delivery, postpartum care; booklet provided; bi-monthly telecommunicationStandard prenatal care without additional education2 monthsQoL score measured by Iranian Diabetes Quality of Life Brief Clinical Inventory (IDQOL-BCI)- QoL improved in all groups post-intervention (P < 0.001). - Greatest QoL improvement in Insulin+Education group (mean difference: 16.43, P < 0.001), followed by Nutrition+Education (11.64, P < 0.001), Insulin Control (9.01, P < 0.001), Nutrition Control (7.27, P < 0.001). - Significant reduction in 1-h blood glucose post-intervention (P < 0.001), no significant changes in FBG, 2-h BG, or HbA1c. - No significant between-group differences in QoL after adjusting for base line (P = 0.814).Wang et al. (2015) (23)ChinaCluster-Randomized Controlled Trial272 pregnant women with ≥ 1 GDM risk factor (Intervention: 134, Control: 138), < 8 weeks gestation, mean age 31.01 ± 3.8 years (intervention), 30.27 ± 3.64 years (control)Two-step lifestyle intervention: (1) 3 group courses (40–60 min each, < 6 participants) at 6–8 weeks on balanced diet, physical activity (e.g., 30 min walking daily), and weight gain goals; (2) nutritional counseling and enforcement at 12–13 weeks based on anthropometrics, plus routine prenatal careRoutine prenatal care only~ 6–7 weeks (from 6–8 weeks to 12–13 weeks gestation)Incidence of GDM, gestational weight gain (GWG)- GDM incidence lower in intervention vs. control group (17.16% [23/134] vs. 23.91% [33/138]; P = 0.168), not statistically significant. - GWG in first trimester similar between intervention and control (1.38 ± 2.34 kg vs. 1.41 ± 2.58 kg; P = 0.905). - GWG in second trimester similar between intervention and control (5.51 ± 2.18 kg vs. 5.66 ± 2.25 kg; P = 0.567). - For women with FPG ≥ 5.1 mmol/L in early pregnancy, GDM incidence lower in intervention vs. control (30.55% [11/36] vs. 45.95% [17/37]; P = 0.076), not statistically significant. Overall Characteristics of included studies - No significant weight change difference: -0.38 kg (-2.08 to 1.30, P = 0.653). - No significant depression score change: -0.67 (-1.84 to 0.49, P = 0.254). - Lower gestational weight gain since GDM visit: 1.89 kg vs. 3.07 kg, -1.20 kg (-2.14 to -0.26, P = 0.012). - Reduced weekly weight gain: 0.17 kg vs. 0.31 kg, -0.14 kg (-0.25 to -0.03, P = 0.008). - Higher proportion without weight retention at 1 year: 34% (28/82) vs. 21% (20/96, P = 0.034). - Lower insulin need: 32% (34/105) vs. 43% (45/106, P = 0.040). - No differences in glycaemic parameters, anxiety, wellbeing, or fitness (P ≥ 0.21). - Increased fat-free mass: 0.02 kg (0.01 to 0.03, P = 0.041). - Significantly fewer newborns with macrosomia in pedometer group: 0% (0/53) vs. 10% (6/62) in control group (p = 0.03). - No significant difference in gestational weight gain between groups: 1.5 kg (-2.5–6.5) in pedometer group vs. 4.6 kg (0.5–8.0) in control group (p = 0.13). - Patients achieving > 5000 steps/day had lower weight gain: -2.7 kg (-7.4–0.75) vs. 6.0 kg (3.0–12.0) in < 5000 steps subgroup (p < 0.001). - ROC curve identified 4210 steps/day as cut-off for avoiding excessive weight gain (> 7 kg): AUC 0.92, sensitivity 0.92, specificity 0.88 (p = 0.00001). - No significant differences in HbA1c, triglycerides, or C-reactive protein levels between pedometer and control groups (p > 0.05). - No significant differences in preterm births, LGA, SGA, or Apgar scores between groups (p > 0.05). - No difference in weight goal achievement: 46.3% (56/121) in IG vs. 43.3% (52/119) in CG (OR 1.13, 95% CI 0.63–2.03, P = 0.680). - Reduced metabolic syndrome in IG: 7.3% (6/82) vs. 16.5% (14/85) in CG (OR 0.40, 95% CI 0.22–0.72, P = 0.002). - IG had less sedentary time (282.7 ± 144.3 vs. 398.4 ± 166.2 min/day, P = 0.032) and higher autonomous/controlled motivation (P = 0.001). - No difference in glucose intolerance, diabetes, or quality of life. - Intervention: 57.5% vs. control: 46.2% achieved weight goal (P = 0.314). - Greater weight loss (-0.93 kg vs. -0.01 kg, P = 0.36). - Reduced energy intake (-497.6 kcal vs. -222.0 kcal, P < 0.01). - Lower IFG/IGT progression (22.5% vs. 30.7%). - Intervention: Reduced FPG (87.3 vs. 123.2 mg/dl), 2-h OGTT (106.5 vs. 156.1 mg/dl), HbA1c (5.3 vs. 6.8%), HOMA-IR (0.9 vs. 2.3) at 12 months (all P < 0.01). - Control: Increased FPG, 2-h OGTT, HbA1c (P < 0.01), no HOMA-IR change. - T2DM incidence: 2% (intervention) vs. 34% (control). - Control group: 33-fold higher insulin resistance risk (adj. OR 32.9, P = 0.01). - Intervention: Lower BMI, WHR (P < 0.05). - Intervention group showed increased perceived susceptibility (mean change 1.45 ± 4.21, P = 0.03) and cue to action (mean change 1.22 ± 3.38, P = 0.02) within group, but no significant between-group differences in HBM constructs (P > 0.05). - General Health subscale of SF-12 improved significantly in intervention group vs. control (mean change 13.69 ± 29.83 vs. 0.00 ± 29.58, P = 0.04, adjusted). - Significant between-group differences in hs-CRP (mean change − 0.877 ± 3.47 vs. -0.067 ± 3.40, P = 0.01, adjusted) and triglycerides (mean change − 55.24 ± 111.21 vs. 40.92 ± 142.01, P = 0.001, adjusted). - Total cholesterol increased less in intervention vs. control (mean change 22.75 ± 66.17 vs. 30.12 ± 61.33, P = 0.01, adjusted). - No significant differences in glycemic indices (FBS, insulin, HOMA-IR) between groups (P > 0.05). - Resistance group: lower 2-hour postprandial glucose (6.03 vs. 6.19 mmol/L, P = 0.019). - No difference in fasting glucose (P = 0.614) or insulin use (P = 1.000). - Higher compliance in resistance group (97.96% vs. 84.31%, P = 0.031). - No difference in adverse pregnancy outcomes (P > 0.05). - GDM incidence lower in LI: 33.3% (10/30) vs. 57.5% (19/33) in UC, RR = 0.59 (0.32–1.04, P = 0.05). - Unadjusted OR for GDM: 0.37 (0.13–1.02, P = 0.05). - Adjusted OR (age, BMI, family history): 0.26 (0.07–0.92, P = 0.04). - Reduced daily calorie intake in LI: -120 kcal (P = 0.004), carbohydrates: -19 g (P = 0.009), fat: -5 g (P = 0.03). - Increased physical activity in LI: +52 min/week (P = 0.05). - No significant differences in gestational weight gain, cesarean section, birth weight, or gestational age at delivery (P ≥ 0.15). - Fewer LI women exceeded IOM weight gain recommendations: 6.6% (2/30) vs. 21.2% (7/33, P = 0.05). - No reduction in glycemic deterioration (25.5% intervention vs. 27.1% control, HR 0.92, P = 0.42). - No difference in type 2 diabetes incidence (9.3% vs. 10.0%, P = 0.48) or body weight change (+ 0.4 kg both groups). - High early postpartum dysglycemia (8.8% type 2 diabetes, 34.1% prediabetes at baseline). - Improved DQOL total score in IG (37.62 ± 10.89 to 48.29 ± 6.49, P < 0.001) vs. decline in CG (37.29 ± 4.05 to 36.62 ± 5.49, P = 0.02). - Reduced blood glucose in IG (101.96 ± 6.62 to 88.14 ± 6.28 mg/dL, P < 0.001) vs. no change in CG (103.37 ± 12.08 to 104.11 ± 16.01 mg/dL, P = 0.54). - Lower blood glucose in IG vs. CG at study end (P < 0.001). - GDM incidence significantly lower in intervention vs. control group (23.9% vs. 51.1%; χ²=7.301, P = 0.007). - FPG significantly lower in intervention vs. control (median: 4.53 vs. 4.74 mmol/L; P = 0.045). - 2-hour PG post-OGTT significantly lower in intervention vs. control (median: 6.42 vs. 7.26 mmol/L; P = 0.002). - No significant difference in 1-hour PG post-OGTT (median: 7.45 vs. 8.18 mmol/L; P = 0.161). - No significant difference in GWG from enrollment to 24–28 weeks (mean: 5.65 vs. 6.13 kg; P = 0.399) or to delivery (mean: 6.49 vs. 5.44 kg; P = 0.089). - Cesarean section rate higher in intervention vs. control (58.5% vs. 27.9%; P = 0.005), but GDM-related cesarean indications similar (3 vs. 3 cases). - No significant differences in gestational hypertension (4.9% vs. 9.3%; P = 0.716), premature infants (2.4% vs. 9.3%; P = 0.386), or macrosomia (9.8% vs. 14.0%; P = 0.797). - No difference in EGWG (OR 1.55, P = 0.152) or absolute GWG (P = 0.207). - Lower average glucose (-0.15 mmol/L, P = 0.011) and glucose above targets (premeal: 17.9% vs. 23.3%, P = 0.003; postmeal: 19.9% vs. 50%, P < 0.001). - Lower composite neonatal complications (38.1% vs. 53.7%, P = 0.006). - Improved lifestyle (HPLP-II total: 141.20 ± 17.82 vs. 128.45 ± 20.72, P = 0.003) and quality of life (e.g., SF-36 physical role: 76.09 ± 32.89 vs. 39.88 ± 37.06, P < 0.001) in intervention group. - No significant depression change (CES-D: 14.93 ± 9.39 vs. 15.74 ± 8.54, P = 0.405). - Lower macrosomia rates (6.5% vs. 28.6%, P = 0.016) and higher breastfeeding (76.1% vs. 50%, P = 0.011) in intervention neonates. − 100% vs. 7.1% postpartum diabetes assessment (P < 0.001). - No significant differences in 15D HRQoL scores between intervention and control groups during pregnancy or up to 12 months postpartum (adjusted for age, pre-pregnancy BMI). - HRQoL decreased similarly in both groups toward third trimester (P-value not reported). - No correlation between GDM diagnosis and HRQoL at any stage. - Cumulative GDM incidence similar (Intervention: 45.7%, Control: 44.5%, P = 0.82). - Self-care behavior score significantly higher in intervention vs. control group post-intervention (adjusted mean difference: 19.5; 95% CI: 14.4 to 24.6; P < 0.001). - No significant difference in quality of life between groups (adjusted mean difference: -0.82; 95% CI: -2.2 to 0.6; P = 0.264). - No significant difference in FBS between groups (adjusted mean difference: -1.09; 95% CI: -3.8 to 1.7; P = 0.443). - Significant reduction in 1-hour GTT glucose in intervention vs. control (adjusted mean difference: -21.6; 95% CI: -32.1 to -11.1; P < 0.001). - Significant reduction in 2-hour GTT glucose in intervention vs. control (adjusted mean difference: -17.3; 95% CI: -23.0 to -11.6; P < 0.001). - Reduced GWG in intervention: -1.2 kg (P = 0.021). - Lower insulin: -0.91 mU/L (P = 0.045) and leptin: -207.8 pmol/L (P = 0.021). - No effect on fasting or 2-h glucose levels (P ≥ 0.776). - Higher glucose exceeding IADPSG thresholds in intervention, especially overweight/obese (33.7% vs. 13.9%, adj. OR 3.89, P = 0.004).< br>- Normal-weight women: trend to lower insulin/HOMA-IR (P = 0.056). - No difference in GDM incidence (WHO criteria) or neonatal outcomes (P > 0.05). - QoL improved in all groups post-intervention (P < 0.001). - Greatest QoL improvement in Insulin+Education group (mean difference: 16.43, P < 0.001), followed by Nutrition+Education (11.64, P < 0.001), Insulin Control (9.01, P < 0.001), Nutrition Control (7.27, P < 0.001). - Significant reduction in 1-h blood glucose post-intervention (P < 0.001), no significant changes in FBG, 2-h BG, or HbA1c. - No significant between-group differences in QoL after adjusting for base line (P = 0.814). - GDM incidence lower in intervention vs. control group (17.16% [23/134] vs. 23.91% [33/138]; P = 0.168), not statistically significant. - GWG in first trimester similar between intervention and control (1.38 ± 2.34 kg vs. 1.41 ± 2.58 kg; P = 0.905). - GWG in second trimester similar between intervention and control (5.51 ± 2.18 kg vs. 5.66 ± 2.25 kg; P = 0.567). - For women with FPG ≥ 5.1 mmol/L in early pregnancy, GDM incidence lower in intervention vs. control (30.55% [11/36] vs. 45.95% [17/37]; P = 0.076), not statistically significant. Risk of bias The RoB-2 assessment of the included randomized controlled trials showed overall low risk in most domains for both Quality of Life and FBG outcomes, although some concerns were identified in outcome measurement and deviations from intended interventions (Fig. 2). The quasi-experimental study (Rokni et al., 2022) was assessed using the ROBINS-I tool and demonstrated an overall moderate risk of bias. Moderate concerns were mainly related to confounding, deviations from intended interventions, and measurement of outcomes; therefore, the findings should be interpreted with caution (Table 2). Fig. 2RoB-2 assessment of the included randomized controlled trials (A: Quality of Life, B: FBG) RoB-2 assessment of the included randomized controlled trials (A: Quality of Life, B: FBG) Table 2ROBINS-I assessment for quasi-experimental study (Rokni et al., 2022(DomainRisk LevelBias due to confoundingModerateSelection of participantsLowClassification of interventionsLowDeviations from intended interventionModerateMissing dataLowMeasurement of outcomesModerateSelection of reported resultsLowOverall Risk of BiasModerate Table 2ROBINS-I assessment for quasi-experimental study (Rokni et al., 2022(DomainRisk LevelBias due to confoundingModerateSelection of participantsLowClassification of interventionsLowDeviations from intended interventionModerateMissing dataLowMeasurement of outcomesModerateSelection of reported resultsLowOverall Risk of BiasModerate ROBINS-I assessment for quasi-experimental study (Rokni et al., 2022( Sensitivity analyses Leave-one-out sensitivity analysis indicated that exclusion of any single study did not materially affect the pooled effect size. The Hedges’s g values ranged from − 0.23 to − 0.34, with all 95% confidence intervals crossing zero, confirming the robustness of the meta-analysis results (Fig. 3). Fig. 3Leave-one-out sensitivity analysis Leave-one-out sensitivity analysis Overall Forest Plot The overall meta-analysis indicates that lifestyle interventions have a statistically significant effect on improving quality of life and reducing FBG levels among women with gestational diabetes (Hedges’ g = − 0.31, 95% CI:−0.57 to − 0.05, p = 0.02. However, substantial heterogeneity was observed across the included studies (I²=91.94%), suggesting considerable variability in study results; therefore, the findings should be interpreted with caution (Fig. 4). Fig. 4Overall meta-analysis of lifestyle interventions on quality of life and FBG Overall meta-analysis of lifestyle interventions on quality of life and FBG Subgroup analysis Subgroup analysis by study population type The forest plot (Fig. 5) presents results of subgroup analyses based on study population type, including antenatal (pregnant women with GDM receiving treatment), postpartum follow-up populations, and prevention studies among high-risk women. A random-effects model with the REML estimator was applied to each subgroup and for the overall effect size. In the antenatal subgroup (6 studies; 293 interventions vs. 293 control participants), the pooled effect size was Hedges’ g = − 0.19 (95% CI: −0.49 to 0.10), which was not statistically significant (p = 0.20) and showed substantial heterogeneity (I² = 84.3%, p < 0.001). The postpartum follow-up subgroup (3 studies; 174 interventions vs. 184 control participants) yielded a pooled effect size of Hedges’ g = − 0.45 (95% CI: −1.33 to 0.43; p = 0.32), with very high heterogeneity (I² = 93.7%, p < 0.001). For prevention studies in high-risk women (6 studies; 929 interventions vs. 928 control participants), Hedges’ g was − 0.41 (95% CI: −0.96 to 0.14; p = 0.15; I² = 92.1%, p < 0.001). The pooled effect size across all 15 studies was Hedges’ g = − 0.31 (95% CI: −0.57 to − 0.05; p = 0.02), indicating a statistically significant overall effect despite high heterogeneity (I² = 91.9%, p < 0.001). Importantly, the test for subgroup differences was not statistically significant (Q [2] = 0.65, p = 0.72), suggesting no evidence that study population type moderates the effect of lifestyle interventions. Subgroup analysis by type of lifestyle intervention (Quality of Life Outcome) A total of seven studies evaluating quality of life outcomes were included across three subgroups: individual interventions (4 studies), group interventions (1 study), and combined individual-plus-group interventions (2 studies). Individual interventions showed a pooled mean difference of 2.20 (95% CI: −4.18 to 8.58), which was not statistically significant, and were associated with very high heterogeneity (I² = 94.03%, p < 0.001). The group intervention subgroup consisted of a single study and demonstrated a mean difference of 8.86 (95% CI: −2.78 to 20.50), also statistically nonsignificant. For combined interventions, the pooled mean difference was 5.20 (95% CI: 3.76 to 6.64), indicating a statistically significant improvement in quality of life, with no heterogeneity (I² = 0%). Across all seven studies, the overall pooled effect was 3.42 (95% CI: −0.55 to 7.39), which did not reach statistical significance. Heterogeneity was extremely high (I² = 91.28%, p < 0.001). Importantly, the test for subgroup differences was not statistically significant (Q [2] = 1.21, p = 0.55), indicating no evidence that intervention type moderated the effect on quality of life (Fig. 6). Subgroup analysis by type of lifestyle intervention (FBG) A total of 15 studies assessed FBG outcomes across three subgroups: individual interventions (6 studies), group interventions (3 studies), and combined interventions (4 studies). The individual intervention subgroup demonstrated a pooled mean difference of − 8.43 mg/dL (95% CI: −19.38 to 2.52), which was not statistically significant (p > 0.05) and exhibited very high heterogeneity (I² = 96.90%, p < 0.001). The group intervention subgroup yielded a pooled mean difference of 0.17 mg/dL (95% CI: −0.77 to 1.12), also not statistically significant and with no heterogeneity (I² = 0%). The combined interventions subgroup showed a pooled mean difference of − 2.60 mg/dL (95% CI: −4.90 to − 0.31), indicating a statistically significant reduction in FBG, with negligible heterogeneity (I² = 0%). Across all 15 included studies, the overall pooled mean difference was − 4.01 mg/dL (95% CI: −8.13 to 0.10), suggesting a borderline statistically significant reduction in fasting glucose under lifestyle interventions. However, heterogeneity remained high (I² = 97.42%, p < 0.001). Importantly, the test for subgroup differences was statistically significant (Q [2] = 6.94, p = 0.03), indicating that intervention type may moderately influence the magnitude of fasting glucose reduction (Fig. 7). Overall, given the clinical differences among antenatal GDM treatment, GDM prevention in high-risk women, and postpartum follow-up populations, the overall pooled estimates should be interpreted with caution. In addition, the subgroup analyses were considered exploratory and should not be interpreted as definitive findings. Fig. 5Forest plot of subgroup analyses Based on Study Population Type Forest plot of subgroup analyses Based on Study Population Type Fig. 6Subgroup forest plot of lifestyle interventions type on quality of life in women with gestational diabetes Subgroup forest plot of lifestyle interventions type on quality of life in women with gestational diabetes Fig. 7Subgroup forest plot of lifestyle interventions type on FBG levels in women with gestational diabetes Subgroup forest plot of lifestyle interventions type on FBG levels in women with gestational diabetes Discussion This systematic review and meta-analysis synthesized data from 20 studies involving 4,777 women with GDM to evaluate the effects of lifestyle education on FBG and QoL. The findings revealed that lifestyle education interventions led to a statistically significant reduction in FBG levels. However, the effect on QoL was mixed and only statistically significant in subgroup analyses involving combined (individual + group) intervention formats. The consistent reduction in FBG across multiple studies confirms the utility of lifestyle education in glycemic management for women with GDM. Lifestyle interventions, particularly those incorporating dietary modifications and physical activity, are grounded in well-established physiological mechanisms. Improved insulin sensitivity through moderate physical activity, regulation of carbohydrate intake, and enhanced metabolic awareness via self-monitoring are all essential pathways through which such interventions act to control blood glucose levels. This aligns with other another study, which advocate for structured lifestyle programs as a first-line strategy in the management of GDM [38]. The robustness of FBG findings, supported by sensitivity analyses and relatively consistent subgroup effects, provides strong evidence for recommending lifestyle education interventions for glycemic control. Several included studies demonstrated significant improvements in FBG, even in high-risk or postpartum populations, suggesting that education-based interventions are not only effective during pregnancy but may also prevent long-term complications such as type 2 diabetes [39, 40]. Nonetheless, the impact on QoL was less uniform. Although some studies reported statistically significant improvements in domains such as physical functioning, general health, or emotional well-being [41], others found no significant change post-intervention [42]. Importantly, the overall pooled effect on QoL was non-significant, and the significant improvement observed in the combined intervention subgroup should be interpreted with caution given the high overall heterogeneity (I² = 76%). This inconsistency may be attributed to several factors. First, QoL is a multidimensional construct influenced not only by physical health but also by psychological, social, and cultural factors. Lifestyle education, if not designed holistically, may address physical symptoms without substantially influencing psychosocial aspects of pregnancy and chronic disease management. The discrepancy between FBG and QoL outcomes suggests that glycemic improvement does not automatically translate to enhanced well-being, highlighting the need for interventions that explicitly target psychosocial outcomes alongside metabolic control. Moreover, studies with positive QoL outcomes often included components that extended beyond standard education, such as peer support, psychosocial counseling, postpartum follow-up, or family involvement. These features likely contributed to enhanced emotional coping, reduced stress, and improved social functioning [43]. In contrast, interventions with a narrow focus on diet and exercise without attention to emotional or cultural contexts may fall short in improving QoL [44]. Heterogeneity in intervention design further complicates the interpretation of QoL outcomes. Variations were noted in frequency and duration of education sessions, the expertise of facilitators, the setting (hospital-based vs. community), and the use of digital versus face-to-face delivery modes. For example, digital interventions, while convenient, may lack the interactivity and emotional resonance of group-based sessions, limiting their effect on QoL despite offering glycemic benefits. Furthermore, the inconsistency in QoL measurement tools (e.g., SF-36, GDMQ-36, DQOL) hinders comparability, as each instrument captures different dimensions and may vary in sensitivity to change [45]. Subgroup analyses revealed that combined individual and group interventions produced the most favorable outcomes for QoL, likely due to the synergistic effects of personalized care and social reinforcement. In contrast, individually delivered or group-only formats showed limited efficacy, possibly due to a lack of peer interaction or insufficient tailoring to individual needs. These findings emphasize the value of hybrid educational models that blend personal support with community-based motivation and shared learning. The observed differences in intervention effectiveness across geographical regions also warrant consideration. Cultural beliefs, health literacy levels, and healthcare system infrastructure can substantially influence how interventions are received and acted upon. For example, in low- and middle-income countries where GDM-related stigma and limited healthcare access may pose barriers, culturally adapted interventions and community health worker involvement may be critical for success [46]. Finally, while improvements in FBG are encouraging, they represent only one component of comprehensive GDM care. The psychological burden of GDM, which includes heightened anxiety, fear of fetal harm, and loss of control over dietary freedom, can severely impair QoL. Therefore, interventions that ignore the emotional context of pregnancy may achieve physiological improvements without meaningfully enhancing overall well-being. In summary, lifestyle education effectively improves glycemic control in women with GDM, but its impact on QoL depends on the comprehensiveness and delivery mode of the intervention. Interventions that integrate psychosocial support, use behavioral frameworks, and are culturally tailored appear most promising. These findings underscore the importance of multidimensional intervention design and reinforce the need for standardized outcome measures in future research. Limitations This study has some limitations that should be considered. First of all, the inclusion of only English-language studies and peer-reviewed articles may have introduced selection bias, potentially excluding relevant non-English or grey literature. Second, the reliance on validated QoL tools, while ensuring rigor, may have overlooked nuanced psychosocial outcomes not captured by standardized measures. The inclusion of one quasi‑experimental study alongside randomized controlled trials may slightly affect the overall certainty of the findings; however, sensitivity analysis showed that excluding this study did not result in any significant change in the results. Physiological differences among pregnant women, postpartum women, and high-risk individuals may contribute to heterogeneity across the included studies. Although our subgroup analysis did not identify a statistically significant difference between these populations, such inherent variation should still be considered an important limitation of this meta-analysis. Conclusion This systematic review and meta-analysis provide strong evidence that lifestyle education interventions significantly reduce FBG levels in women with GDM, with group and combined interventions showing consistent benefits. However, the impact on QoL is uncertain. Although combined individual and group approaches showed a statistically significant improvement in exploratory subgroup analyses, the overall pooled effect on QoL was non-significant, and substantial heterogeneity was observed across studies. These subgroup findings should be considered exploratory and hypothesis-generating rather than conclusive. The discrepancy between robust FBG reduction and uncertain QoL improvement underscores the importance of intervention design that addresses both physiological and patient-centered outcomes. Current evidence does not permit firm conclusions regarding the efficacy of lifestyle education for improving QoL in women with GDM. For clinical practice, the strong evidence for FBG reduction supports the implementation of structured lifestyle education programs as a first-line strategy for glycemic control in GDM. However, recommendations regarding QoL improvement cannot be made with confidence given the uncertain evidence base. To optimize the effectiveness of lifestyle education, future interventions should prioritize culturally tailored programs that integrate individual counseling with group-based support and explicitly include psychosocial components. High-quality RCTs with standardized QoL measures and adequately powered samples are needed to determine whether lifestyle education can reliably improve QoL in this population. Policymakers and healthcare providers should invest in scalable delivery modes, such as digital platforms, to improve access, particularly in low-resource settings. By addressing these gaps, healthcare systems can better empower women with GDM to achieve healthier pregnancies and reduce the long-term risks associated with this condition. Supplementary Information Supplementary material 1. Supplementary material 1. Supplementary material 2. Supplementary material 2.