Combined diet and physical activity effects on health-related outcomes in people with overweight or obesity: an overview of systematic reviews

Abstract

Background:

This overview synthesized evidence from systematic reviews (SRs) of combined physical activity (PA) and diet interventions versus diet-only or PA-only approaches on obesity-related anthropometric, cardiometabolic, and physical fitness outcomes, and identified program features linked to better effects.

Methods:

We conducted an Overview of SRs following Cochrane guidance and PRIOR standards. Five databases were searched until June 13, 2024, and Google Scholar was updated on January 25, 2025. Eligible SRs synthesized randomized controlled trials of combined PA and diet interventions in people with overweight or obesity. We extracted key intervention characteristics, outcomes, and meta-analytic estimates where available. Methodological quality was assessed with AMSTAR 2, and overlap was quantified using the corrected covered area (CCA).

Results:

Thirty-two SRs (19 meta-analyses) met inclusion criteria. In adults, combined PA and diet interventions generally outperformed single-component comparators for reducing weight, BMI, waist circumference, and fat mass, and improving cardiorespiratory fitness. More favorable and sustained effects were most often reported in programs lasting ≥6–12 months with frequent, structured contact and supervised, resistance-inclusive exercise. In children and adolescents, face-to-face and family-engaged programs produced the most consistent improvements in BMI/BMI-z, but effects often attenuated 6–12 months after program end without maintenance. In older adults, combined programs reduce fat mass while preserving lean mass, strength, and function. Across diverse settings (including type 2 diabetes, workplaces, and digital delivery), combined interventions improved glucose control, insulin resistance, lipids, and inflammation, and some benefits occurred even without additional weight loss. Long-term lifestyle programs reduced incident type 2 diabetes by about half. Review quality varied (31% high; 69% low/critically low) and overlap between SRs was minimal (CCA 0.0–0.6%).

Conclusion:

Combined PA and diet interventions tended to outperform diet-only and PA-only approaches. The most consistent and durable benefits were linked to longer duration, higher contact, supervision, resistance training, and family engagement in youth, underscoring maintenance planning. These findings support prioritizing structured lifestyle programs, including scalable hybrid/digital delivery models, while interpreting effects cautiously given the predominance of low-quality SRs. Future research should improve rigor, standardize outcomes, report intervention dose and behavior-change methods, include longer follow-up, and improve representation of underserved groups.

1 Introduction

Obesity has emerged as a major global public health crisis. By 2022, its burden encompassed more than one billion individuals, about one in eight worldwide, and its prevalence has more than doubled among adults and quadrupled among young people since 1990 (1). However, temporal increases have not been uniform across high-income countries with stable food availability. For example, in the United States, measured adult obesity increased from 13.4% in 1960–1962 to 40.3% in August 2021–August 2023, whereas Japan has remained among the leanest OECD countries, underscoring the importance of differences in food environments, lifestyles, and policy contexts when interpreting global trends (2). Beyond its health consequences, the economic implications are equally profound: systematic reviews estimate that obesity already imposes multi-trillion-dollar annual costs on health-care systems and productivity worldwide (3). These figures underscore the urgency of scalable, evidence-based interventions that address positive energy balance while also acknowledging that obesity is shaped by interacting biological, behavioral, and environmental determinants, including appetite regulation, food quality and processing, physical activity, and the broader food environment. Even in settings with stable food availability, substantial cross-country variation suggests that obesogenic exposures and dietary patterns differ meaningfully across populations.

Obesity has been addressed through multiple strategies, most commonly dietary modification, increased physical activity (PA), and, often most effective, their combination. The World Health Organization Guidelines (2020) on PA and Sedentary Behavior recommend that adults accumulate 150–300 min week⁻¹ of moderate-intensity activity (or the vigorous equivalent), emphasizing that “any activity is better than none” (4). Complementary WHO dietary guidance emphasizes not only limiting free sugars, sodium, saturated fat, and trans-fat, but also improving overall dietary quality by prioritizing whole grains, vegetables, fruits, and pulses, with adults encouraged to consume at least 400 g/day of fruits and vegetables and 25 g/day of naturally occurring dietary fiber (5, 6). Emerging evidence further suggests that obesity risk is shaped not only by excess energy intake in a general sense but also by low consumption of fiber-rich minimally processed foods and high exposure to ultra-processed dietary patterns, which may influence satiety, metabolizable energy, and downstream metabolic health (7–9). At the same time, cross-population doubly labeled water studies indicate that reduced daily energy expenditure alone is unlikely to fully explain rising obesity prevalence in industrialized settings, even though physical activity remains critically important for cardiometabolic health, body composition, and weight-loss maintenance (10). However, despite these clear prescriptions, population adherence remains low, prompting clinical and public-health guidelines, including the WHO Acceleration Plan to Stop Obesity (2023–2030), to endorse multicomponent lifestyle programs that integrate structured PA counseling with tailored dietary modification.

A growing evidence base underscores the added value of addressing both diet and PA. Consistent with this, a meta-analysis of randomized trials found that combined PA and diet intervention achieved greater weight loss than diet-only or exercise-only approaches at 12 months (11). A recent meta-analysis supports those findings; Rotunda, Rains (12) pooled short-term trials and reported greater weight loss versus usual care, with most interventions yielding clinically meaningful reductions for a sizeable proportion of participants. Evidence in younger populations is convergent; a meta-analysis showed significant reductions in body weight and metabolic-syndrome markers when combined PA and diet intervention were reinforced with behavior-change techniques (13). The available evidence further suggests that combined PA and diet intervention are associated with clinically meaningful weight loss and more favorable improvements in insulin sensitivity, lipid profiles, inflammation, and visceral fat compared with single-component approaches, with effects on visceral fat appearing more pronounced under dietary energy restriction (14). Evidence further suggests that specific pairings may matter. Resistance training combined with higher-protein, moderate-energy-deficit diets has been associated with better preservation of fat-free mass, whereas aerobic training is more consistently linked to greater fat loss (15). In adults with obesity and type 2 diabetes, combined PA and diet intervention are also associated with more favorable cardiometabolic profiles, even when weight loss is modest (15).

Nonetheless, important uncertainties remain that limit translation into practice and policy. Although numerous systematic reviews have examined combined PA and diet intervention, heterogeneity in intervention intensity, supervision, theoretical grounding, and delivery mode complicates comparison across reviews and challenges the consistency of guideline development. Outcomes are reported inconsistently, ranging from weight alone to waist circumference, glycemic indices, quality of life, and healthcare use, which further limits the ability of decision-makers to prioritize outcomes most relevant to specific populations and settings. Evidence on equity and scalability is limited, with few analyses by sex, age, or socioeconomic status. The therapeutic landscape has also changed rapidly with the emergence of highly effective incretin-based anti-obesity pharmacotherapies. In large contemporary trials, semaglutide 2.4 mg and tirzepatide have produced mean weight reductions substantially greater than those usually reported in older lifestyle-only trials (16, 17). At the same time, such therapies are used alongside, rather than instead of, reduced-calorie diets and increased physical activity. This broader context makes it important to interpret the present overview not only in terms of relative superiority over single-component lifestyle approaches, but also in relation to the absolute magnitude and clinical role of lifestyle-related benefits. Importantly, pharmacotherapy and lifestyle treatment should not be viewed as competing options; current obesity care increasingly combines anti-obesity medications with structured dietary and physical-activity support, while metabolic bariatric surgery remains the most effective durable treatment for severe obesity.

Given the rapid expansion of systematic reviews in this field, an overview of reviews is particularly valuable to synthesize evidence across the life course, identify intervention characteristics consistently associated with more favorable outcomes, and appraise the robustness of conclusions in light of review overlap and methodological quality. Such synthesis is especially relevant for clinicians designing lifestyle-based obesity services, public-health and workplace health programs, developers of digital and hybrid interventions, and pediatric and family-based care pathways, as well as for policymakers tasked with guideline development and resource allocation.

Accordingly, the present Overview of systematic reviews (OoSRs) aims to synthesize and critically appraise evidence on the effectiveness of combined PA and diet intervention on anthropometric, metabolic, physical-fitness, and patient-centered outcomes across the life course in populations with overweight or obesity.

2 Methods

The present OoSRs was conducted in accordance with the Cochrane Guide for Overviews of Reviews (18) and reported following the Preferred Reporting Items for Overviews of Systematic Reviews (PRIOR) Checklists (19).

2.1 Eligibility criteria

We prespecified eligibility using the PICOS framework (Participants, Interventions, Comparators, Outcomes, Study design). We included systematic reviews, with or without meta-analysis, which synthesized randomized controlled trials of combined PA and diet intervention in individuals living with overweight or obesity. To be considered a systematic review, publications were required to report, at minimum, a reproducible search strategy, explicit eligibility criteria, a systematic study selection process, and a structured synthesis of findings. Participants were children (≥ 5 years), adolescents, adults, or older adults classified as having overweight or obesity. Interventions were combined PA and diet intervention. Comparators were passive controls (usual care, wait-list), single-domain controls (diet-only or exercise-only), or attention controls. Outcomes had to include at least one post-intervention or follow-up measure related to anthropometry, metabolic health, patient-centered outcomes, or implementation and health-economic endpoints. We excluded reviews focused primarily on pharmacotherapy or bariatric surgery unless lifestyle-only arms were analyzed separately, reviews restricted to weight-neutral conditions (for example, sarcopenia), and narrative, scoping, or rapid reviews lacking systematic methodology. Only reviews published in peer-reviewed journals were included; non-English articles were translated prior to the study selection and data extraction phases. Cochrane reviews were eligible for inclusion. When multiple versions of the same systematic review were identified, the most recent and comprehensive version was retained.

2.2 Search strategy, study selection, and data extraction

The systematic literature search was conducted using five online databases (PubMed, Web of Science, CINAHL, ProQuest, and PsycInfo) from database inception to June 13th, 2024. The full research strategy and keywords were presented in Supplementary Table S1. References for all included papers were manually screened for additional relevant reviews. Google Scholar was searched on January 25th, 2025, for potentially published reviews.

The selection process and data extraction were conducted by two authors independently. After each step, the authors double-checked the spreadsheets’ accuracy, and any disagreements were solved by discussion between the two authors. Given the exploratory nature of the overview and the use of consensus resolution, formal inter-rater reliability statistics were not calculated. Duplicated articles were removed using the Endnote software (version 20). All articles were screened by the title, abstract, and full text using the PICOS inclusion criteria.

The following data were extracted and presented in Table 1: (i) reference, (ii) design, (iii) number of the included studies (N), (iv) participants characteristics, (v) intervention characteristics (PA. Diet, Delivery Mode, Duration, Comparator), (vi) measured outcomes (Outcome Domain, Specific Outcome, Follow-up), (vii) results, and (viii) conclusion.

2.3 Methodological quality assessment

The methodological quality of the SRs was assessed using the AMSTAR2 (A MeaSurement Tool to Assess Systematic Reviews) (20) with 16-items. Seven items are considered critical, and three items concern meta-analytical methods and are not applicable for SRs without accompanying meta-analysis (Supplementary Table S4). The AMSTAR2 rates SRs as critically low (more than one critical weakness, with or without noncritical weaknesses), low (one critical weakness, with or without noncritical weaknesses), moderate (no critical weakness, with more than one non-critical weakness), or high quality (neither critical nor non-critical weaknesses) (20). AMSTAR2 was conducted independently by two authors, with any disagreements solved by discussion between the two authors. AMSTAR 2 was used to assess methodological quality at the review level; no separate tool for risk of bias across reviews [e.g., ROBIS (21)] was applied.

AMSTAR 2 was selected because it is designed to appraise the methodological quality of systematic reviews at the review level. By contrast, GRADE is primarily intended to rate the certainty of evidence for specific outcomes, and its application in overviews of reviews remains methodologically challenging and lacks consistent guidance, particularly when included reviews vary in scope, quality, outcome definitions, and overlap of primary studies (22–24). Therefore, formal GRADE assessments were not performed in the present overview. To mitigate the risk of overinterpretation, conclusions were instead contextualized by explicitly considering methodological quality (AMSTAR 2), consistency of findings across reviews, and the degree of overlap between primary studies.

2.4 Overlap of included studies

The degree of overlap was quantified using the corrected covered area (CCA) method (25). The CCA is calculated by dividing the frequency of repeated occurrences of index studies (first occurrence of a primary study) in other reviews (of the same domain) by the product of the number of index studies and the number of reviews, minus the number of reviews. The CCA is represented as a percentage between 0 and 100%. A CCA of 0–5% is considered a slight overlap, 6–10% moderate overlap, 11–15% high overlap, and more than 15% very high overlap (25). To visualize pairs of overlapping SRs in each domain, a heatmap was generated using the “ccaR” package in the R programming language¹ (26, 27). In addition to quantifying overlapping, explicit decision rules were applied when interpreting findings from overlapping reviews. Priority was given to systematic reviews with higher AMSTAR 2 ratings, more recent search dates, and broader primary study coverage, to minimize double counting and undue influence of methodologically weaker reviews.

2.5 Data synthesis

Data synthesis followed best-practice recommendations for OoSRs. Where included reviews conducted meta-analyses, the pooled number of participants, pooled effect estimates (e.g., mean difference or standardized mean difference with 95% confidence intervals), heterogeneity statistics (I²), and model specifications were extracted as reported by the review authors. No re-analysis of primary study data was undertaken.

For reviews without meta-analysis, findings were synthesized narratively using a structured approach based on: (i) the direction of effects reported by the systematic reviews, (ii) consistency of findings across primary studies and reviews, (iii) clinical relevance of the reported effects, and (iv) the methodological quality of each review as assessed by AMSTAR 2. Narrative-synthesis judgments were made by two authors independently and resolved by consensus using these predefined criteria, although formal inter-rater reliability statistics were not calculated. When overlapping reviews addressed similar populations and outcomes, greater interpretive weight was given to reviews with higher methodological quality, more recent search coverage, and broader primary-study inclusion. Vote counting based solely on statistical significance was avoided, as this approach may bias interpretation by disregarding study size, precision, and between-study heterogeneity (28, 29).

When multiple systematic reviews addressed overlapping populations and outcomes, interpretation was guided by a structured hierarchy prioritizing higher methodological quality, more recent search coverage, and greater comprehensiveness of included trials.

3 Results

3.1 Study selection

A total of 7,019 articles were identified via online databases. A total of 2,534 duplicates and 212 citations were removed. After screening 4,273 articles based on title and abstract, 4,197 were excluded. Full texts were sought via institutional subscriptions and interlibrary loan; corresponding authors were also contacted when needed. Despite these efforts, two records could not be obtained and were excluded from full-text assessment (30, 31). After careful reviews of full-text articles (n = 74), 42 reviews were excluded. The list of excluded studies is presented in Supplementary Table S2. A total of 32 systematic reviews were included in this OoSRs (Figure 1). Reviewer agreement was high during title/abstract screening (percent agreement = 98%), full-text screening (100%), and the Google Scholar update search (100%). Given consensus resolution procedures, Cohen’s κ was not calculated.

Figure 1

3.2 Studies characteristics

The characteristics of the 32 studies included in this OoSRs are presented in Table 1. Systematic reviews were published between 2008 and 2024, of which 19 (60%) performed meta-analyses. Only six reviews reported an overall total sample size aggregated across included trials. Across all reviews, the median number of primary studies per review was 22 (IQR 13–32; range 4–118).

Adult-focused reviews enrolled non-athlete adults with overweight or obesity, including subgroups with metabolic syndrome or severe obesity and worksite cohorts (32–35). Reviews of children and adolescents covered preschoolers through teenagers in school, community, and outpatient/primary-care settings, often with explicit parent or family engagement (36–40). One review focused on older adults, including geriatric-obesity phenotypes, delivered in community (41). Special populations included infertile patients undergoing lifestyle programs (42) and children and adults enrolled in culturally tailored programs (43).

Programs delivered combined PA and diet intervention. PA components were typically aerobic, resistance, mixed, or High-Intensity Interval Training (HIIT), usually moderate-to-vigorous intensity, 2–5 days·week⁻¹. Dietary components included hypocaloric balanced menus; Mediterranean/Dietary Approaches to Stop Hypertension (DASH); low-fat/low-carbohydrate; or meal-replacement/very-low-energy phases. Interventions were delivered by dietitians, exercise professionals, or multidisciplinary teams. Delivery ranged from fully face-to-face to hybrid or fully remote formats using web/apps, tele-monitoring, and step-count goals; several reviews contrasted supervised center-based sessions with home-based programming. Family-based delivery was common in pediatric combined PA and diet intervention and was often associated with higher adherence and more favorable anthropometric responses. Program length ranged from 2 weeks to 61 months, with most trials clustered in the 3–12-month window; higher contact frequency and ≥12-month duration was commonly associated with larger or more durable effects. In older adults, combined PA and diet intervention mitigated lean-mass and bone losses during weight reduction across 6–18 months. Control conditions included usual care/minimal advice, diet-only, exercise-only, or alternative active comparators; pediatric prevention studies typically used usual-curriculum comparators at the school or class level.

Reviews commonly reported anthropometric and body composition outcomes (weight, body mass index (BMI), BMI/BMI-for-age z-score (BMI/BMI-z), waist circumference, body-fat%), cardiometabolic markers (blood pressure, lipids, glycaemia/insulin/Homeostatic Model Assessment of Insulin Resistance (HOMA-IR)), inflammation/adipokines (C-reactive protein (CRP), Interleukin-6 (IL-6), Tumor Necrosis Factor-alpha (TNF-α); adiponectin/leptin), physical fitness and strength, behavioral outcomes (Moderate to Vigorous PA (MVPA), diet quality, sedentary time), quality of life/psychosocial endpoints, and, where reported, autonomic function (e.g., heart rate variability (HRV)).

Most syntheses used meta-analysis for anthropometric and cardiometabolic outcomes (e.g., long-term adult weight change and workplace BMI/waist effects), while network meta-analysis was employed to contrast multi-component nodes (e.g., face-to-face combined PA and diet intervention vs. single-component arms) in pediatric cohorts.

Across population subgroups, intervention domains showed distinct patterns. Among adults, combined PA and diet intervention reduced body weight and central adiposity and improved cardiometabolic risk, with remote/digital delivery demonstrating short-term weight loss comparable to in-person formats (12, 32, 42). In children and adolescents, family-engaged, school-centered, multi-component combined PA and diet intervention produced the most consistent improvements in BMI/BMI-z, whereas single-component PA alone rarely shifted BMI (36–38). Among older adults, resistance-inclusive combined PA and diet intervention preserved lean mass and improved strength more than diet-only or aerobic-only comparators over 8–16 weeks (41).

3.3 Narrative synthesis of results

Across populations and settings, SRs generally suggested that combined PA and diet intervention consistently tended to outperform single-component comparators for obesity management and prevention (11). The strength and consistency of evidence varied by outcome domain, population, and review quality, with effects appearing more consistently in longer-term interventions that featured structured or supervised delivery, higher contact frequency, and embedded behavior-change techniques such as goal setting, self-monitoring, feedback, and motivational interviewing (11, 12, 34, 35, 44).

In adults with overweight or obesity, combined PA and diet intervention was generally associated with greater and more durable reductions in body weight, BMI, and waist circumference than diet-only or PA-only strategies, with effects commonly larger in longer, higher-contact, and more structured or supervised programs (11, 35, 44). Relative to diet-only approaches, combined PA and diet intervention enhances fat-mass loss while helping preserve lean mass, which improves the quality of weight loss and simultaneously raises cardiorespiratory fitness, muscular strength, and physical function, even when incremental weight loss over diet alone is moderate (11, 32, 45, 46).

Compared to diet alone, combined PA and diet intervention showed better glycemic control, healthier lipid profiles, lower blood pressure, and favorable inflammatory and adipokine changes, including small but significant improvements in fasting glucose and insulin resistance even when additional weight loss is minimal (15, 47–49). Among adults with metabolic syndrome, combined PA and diet intervention was commonly associated with reductions in waist circumference (often ~5–15 cm) and weight loss (~3–12 kg), although estimates varied across reviews and trial designs; outcomes appeared to depend more on overall energy deficit and exercise structure than on macronutrient ratios (35).

In class II to III obesity, combined PA and diet intervention was associated with moderate but clinically relevant improvements in weight, BMI, blood pressure, triglycerides, and total or low-density lipoprotein (LDL) cholesterol, with larger effects more often reported in longer and higher-contact programs (33, 44). Over longitudinal studies (up to 15 years), multi-component lifestyle interventions approximately halve incident type 2 diabetes risk versus usual care, with greater weight loss and higher baseline fasting insulin predicting larger risk reductions (50). In adults with established type 2 diabetes, combined PA and diet intervention was associated with improvements in Glycated hemoglobin (HbA1c), fasting glucose, insulin resistance, and lipid outcomes, reinforcing therapeutic relevance beyond weight control (15).

In children and adolescents, face-to-face, family-engaged, multi-component combined PA and diet intervention appears to outperform usual care and either component alone, improving BMI, BMI-z, waist circumference, and body fat percentage at about 6 months, with higher effects when programs run at least 6–12 months and actively engage parents or the school environment (36, 37, 40, 51–54). Pre-school prevention programs rarely affect BMI, whereas management programs with active parent involvement in clinical or community settings more consistently reduce BMI or BMI-z (36, 37). In older adults, combining PA and diet intervention generates the most favorable balance of outcomes, reducing fat mass while attenuating losses of lean mass and bone typically observed with diet alone, alongside improvements in function and quality of life, while exercise-only arms improve function without meaningful weight loss (41). Workplace programs yield small to moderate improvements in BMI, waist circumference, lipids, and blood pressure, with more improvements when coaching and exercise opportunities are included rather than information-only approaches (34).

Digitally supported and telehealth modalities are associated with short-term weight loss across 8–26 weeks, particularly when delivered entirely online, although durable effects beyond the initial period depend on behavioral support, high-frequency contact, and planned follow-up (12). Evidence syntheses of culturally tailored programs indicate small weight losses in South-Asian adults and no clear anthropometric effects for school-aged children, again pointing to intensity, duration, and family engagement as effect modifiers in real-world implementation (43).

Interpretation of these patterns should be cautious given the substantial variability in methodological quality across included reviews, and the predominance of low or critically low ratings.

3.4 Meta-analytical results

All quantitative estimates reported below are pooled effects extracted directly from the included MAs, without re-analysis of primary trial data.

3.4.1 Body weight, BMI, and weight-loss magnitude

In adults, MAs generally indicate that behavioral programs lasting at least 1 year that deliver combined PA and diet intervention tend to achieve more favorable weight outcomes than either component alone. Across 22 trials, combined PA and diet intervention decreased body weight by 1.38 kg beyond diet alone and 4.13 kg beyond exercise alone (32). Head-to-head contrasts reinforce this added value: meta-analysis of eight adult RCTs showed that, compared with PA alone, combined PA and diet intervention achieved 5.33 kg greater loss at 3 to 6 months and 6.29 kg greater loss at 12–18 months, while versus diet alone the advantage was 0.62 kg at 3 to 6 months and 1.72 kg at 12 to 18 months (11). Shorter adult programs also show benefits: across 14 trials of 8 to 26 weeks, combined PA and diet intervention reduced body weight by 2.59 kg versus usual care, with larger effects in fully online delivery and male-only trials (12). In professionally guided adult programs, 1-year absolute weight loss was only marginally smaller in adults with overweight than in those with class I or II obesity (≈1.1 kg less than class I; ≈1.5 kg less than class II), while percentage weight loss did not differ by BMI class, supporting applicability across BMI 25 to 40 kg·m⁻² (55). In adults with class II to III obesity, pooled data across 56 studies indicated mean weight loss of ≈8.9 kg and a BMI reduction of ≈2.8 kg·m⁻², with larger weight loss when programs lasted longer than 12 months (44), and independent reviews of severe obesity reported 1.0 to 11.5 kg more loss than controls over 3 to 24 months and BMI reductions of 0.3 to 4.0 kg·m⁻² more than controls (33). In workplace trials in adults, pooled estimates across 13 RCTs showed BMI decreases of 0.86 kg·m⁻² (34). In women of reproductive age with overweight or obesity, combined PA and diet intervention was associated with a mean BMI reduction of 1.42 kg·m⁻² versus minimal care (56), and in obesity-related infertility pooled weight loss was 3.98 kg (with the best individual trials around 10 kg) (42). In South-Asian adults, culturally tailored combined PA and diet intervention reported pooled weight loss of −1.82 kg versus control, while pooled effects for BMI were not statistically significant (43). In pediatric and adolescent populations, a 6-month network meta-analysis of 118 RCTs found that face-to-face combined PA and diet intervention reduced BMI by 0.98 kg·m⁻² and BMI-z by 0.10 SD versus control (52), while broader syntheses report a standardized mean difference of −1.09 for BMI or BMI-z at ~6 months with attenuation after contact ends (57). Pooled prevention-setting analyses reported BMI-z mean difference of −0.12 and body weight change of −1.59 kg (57), and earlier work suggested BMI reductions were approximately doubled when parents were actively engaged (36). In South-Asian school-age children, network meta-analysis did not identify a clear between-group effect for BMI or BMI-z (43).

3.4.2 Waist circumference and central adiposity

In adults, combined PA and diet intervention reduced waist circumference by a further 1.68 cm versus diet and 3.00 cm versus exercise (32). In adults with class II to III obesity, waist circumference decreased but fasting glucose and HDL/LDL responses were inconsistent (44). In workplace adult trials, waist circumference decreased by 2.06 cm (34). In pediatric/adolescent RCTs, face-to-face combined PA and diet intervention reduced waist circumference by 1.49 cm versus control and outperformed face-to-face PA or diet alone for BMI and waist outcomes (37, 52). In pooled school/community prevention settings, waist outcomes improved modestly overall, though heterogeneity was high (57). In South-Asian adults, culturally tailored combined PA and diet intervention did not show significant pooled effects on waist circumference, and in South-Asian school-age children pooled waist effects were also non-significant (43).

3.4.3 Fat mass, lean mass, and “quality” of weight loss

In adults, combined PA and diet intervention removed an additional 1.65 kg of fat versus diet and 3.60 kg versus exercise (32). Across adult syntheses, when structured aerobic or resistance training is delivered within combined PA and diet intervention, additional fat mass loss is ~17 to 18% from baseline, and lean-mass loss is attenuated (~2.2 to 2.5% vs. ~ 3.6% with diet alone) (46). In older adults (community-dwelling ≥60 years with obesity), diet-only groups reduced fat mass but also fat-free mass and showed small yet significant bone-density loss; combined PA and diet intervention better preserved lean tissue and attenuated bone loss (41).

3.4.4 Blood pressure

In adults, diastolic blood pressure decreased by ~1.2 mm Hg versus diet and ~2.1 mm Hg versus exercise, while systolic blood pressure differences were minimal versus diet and about −2.8 mm Hg versus exercise (32). In adults with obesity and type 2 diabetes, combined PA and diet intervention was associated with modest decreases in systolic and diastolic blood pressure (15). In class II to III obesity, diastolic blood pressure fell by ~4 to 6 mm Hg and systolic pressure also declined (44), and severe-obesity reviews similarly noted modest blood pressure improvements (33). In workplace adult trials, systolic blood pressure decreased by 3.39 mm Hg and diastolic blood pressure by 2.89 mm Hg (34). In pooled school/community prevention settings, blood pressure improved in roughly one third of studies, particularly in programs ≥12 months with structured exercise plus nutrition plans (57).

3.4.5 Cardiorespiratory fitness, strength, and physical function

In adults, cardiorespiratory fitness increased by ~3.6 mL·kg⁻¹·min⁻¹ over diet and ~2.1 mL·kg⁻¹·min⁻¹ over exercise (32), and broader adult syntheses report VO₂peak rises of ~2 to 5 mL·kg⁻¹·min⁻¹ with combined PA and diet intervention, with routine strength gains when resistance or combined training is included (46). In older adults (≥60 years with obesity), exercise-only interventions improved performance without corresponding weight loss (e.g., VO₂peak + ~ 1.4 mL·kg⁻¹·min⁻¹ and upper-body strength from 174 to 190 lb. one-rep-max), whereas combined PA and diet intervention produced larger VO₂peak gains (~ + 3.1 mL·kg⁻¹·min⁻¹) and consistent improvements in chair-rise and 6-min walk tests (41). In comparative adult work, portion-controlled diets plus ≥175 min/week combined endurance + strength training were associated with ~5% extra body-weight loss over 6 to 24 months versus usual care and outperformed diet alone by 2 to 4 kg at 6 to 12 months in programs using supervised aerobic sessions (45).

3.4.6 Lipids

In adults, lipid responses favored combined PA and diet intervention over exercise alone (≈10–11 mg·dL⁻¹ lower total and LDL cholesterol), while differences versus diet alone were small; HDL increased slightly versus diet (≈1.6 mg·dL⁻¹) and triglycerides were modestly lower versus diet (≈10 mg·dL⁻¹) (32). In the same meta-analysis of adults without diabetes, triacylglycerides decreased by −0.258 ± 0.037 mmol·L⁻¹, whereas HDL showed no significant pooled change (58). In adults with obesity and type 2 diabetes, combined PA and diet intervention improved lipid profile by reducing total cholesterol and triglycerides, while HDL-C and LDL-C showed no clear pooled effect (15). In class II to III obesity, total cholesterol, LDL-C, and triglycerides decreased, with inconsistent HDL-C responses (44). In workplace adult trials, total cholesterol decreased by 6.83 mg·dL⁻¹, HDL-C increased by 0.83 mg·dL⁻¹, LDL-C decreased by 6.20 mg·dL⁻¹, and triglycerides fell by 12 mg·dL⁻¹ (34). In pediatric/adolescent RCTs, pooled estimates indicated total cholesterol SMD − 0.45 and triglycerides SMD − 0.42, with non-significant pooled effects on HDL-C and LDL-C (15), and in pooled school/community prevention settings lipids improved in roughly one third of studies (57).

3.4.7 Glycemic control and insulin sensitivity

In adults, adding structured exercise within combined PA and diet intervention was associated with reductions in fasting plasma glucose of ~1.46 mg·dL⁻¹ and improvements in insulin sensitivity of ~0.33 SD, despite negligible additional weight loss relative to diet alone (48). In adults without diabetes, a meta-analysis of controlled trials similarly showed that diet-plus-exercise education (≈1 year) reduced fasting glucose by −0.18 ± 0.04 mmol·L⁻¹ and fasting insulin by −2.56 ± 0.58 mU·L⁻¹ versus controls (58). In adults with obesity and type 2 diabetes, combined PA and diet intervention was associated with reductions in HbA1c, fasting glucose, and fasting insulin (15). In class II to III obesity, fasting glucose responses were inconsistent (44). In workplace adult trials, fasting glucose decreased by 1.23 mg·dL⁻¹ (34). In pediatric/adolescent RCTs, pooled effects indicated improved glycemic control (HbA1c SMD − 0.52; fasting insulin SMD − 0.32), with a large but imprecise reduction in fasting glucose (15), and pooled school/community prevention settings also reported improvements in a subset of studies (57).

3.4.8 Inflammation and adipokines

In adults, pooled meta-analytic estimates suggested small inflammatory effects when combined PA and diet intervention was contrasted with diet-only approaches (CRP SMD − 0.16, 95% CI − 0.29 to −0.03), while IL-6 and TNF-α pooled effects were near null (49). In adults with obesity and type 2 diabetes, pooled analyses suggested rises in adiponectin of ~1.4 μg·mL⁻¹ and reductions in leptin of ~7.3 ng·mL⁻¹ (15), while broader syntheses similarly report small CRP reductions with pooled null effects on IL-6 and TNF-α (15, 47, 49). In pediatric/adolescent RCTs, inflammation markers showed generally favorable pooled changes in limited trials (e.g., CRP SMD − 0.46) with no clear pooled change in TNF-α (15), and longitudinal cohorts reported endocrine adaptations tracking weight trajectories (e.g., leptin falling with loss and rising with regain; TSH and FT3 predicting later regain) (53).

3.4.9 Physical activity and dietary behavior change (process outcomes)

In adults, PA indices were not consistently higher at 12 months in combined PA and diet intervention arms, yet weight and fitness outcomes still favored the dual-component approach (11). In pooled school/community prevention settings, PA increased by at least 10% in about half of studies assessing it, and dietary intake improved by ~41%, with stronger effects in programs of at least 12 months featuring structured exercise plus nutrition plans (57). In school-centered pediatric programs, classroom and after-school initiatives increased moderate-to-vigorous activity (e.g., ~26 extra minutes/day in APPLE; ~47 min of active time in after-school free play) and improved dietary choices (e.g., reduced fizzy-drink intake; higher fruit selection), although BMI changes were not universal (54).

3.4.10 Quality of life, functional outcomes, and adverse events

In older adults (≥60 years with obesity), quality of life (when reported) improved more in combined PA and diet intervention arms, with SF-36 physical composite rising by ~8 to 15% compared with ~3% in diet-only and no change in control; adverse events were generally minor (typically transient musculoskeletal pain) (41). In pediatric contexts, family-based outpatient training improved health-related quality of life at 12 months (53).

3.4.11 Long-term disease incidence

Over a median 13-year follow-up in adults, combined PA and diet intervention was associated with an approximately 50% lower incidence of type 2 diabetes versus usual care (odds ratio ~0.44), with larger risk reductions accompanying greater weight loss and higher baseline fasting insulin (50).

3.5 The methodological quality assessment

The AMSTAR 2 appraisal of 32 systematic reviews showed substantial variation in methodological rigor, with important implications for confidence in the synthesized findings (Supplementary Table S3). Using the decision rules, overall confidence ratings were High 10/32 (31.2%), Moderate 2/32 (6.2%), Low 10/32 (31.2%), and critically low 10/32 (31.2%), with approximately 69% of reviews rated as low to critically low (22/32, 68.8%). For the critical domains, preregistration or a priori protocol (Item 2): Y 10/32 (31.2%); N 22/32 (68.8%); PY 0/32 (0.0%). Comprehensive multi-database searching (Item 4): Y 23/32 (71.9%); N 1/32 (3.1%); PY 8/32 (25.0%). Transparency around study exclusion remained a consistent weakness, few reviews listed excluding studies with reasons (Item 7: Y 8/32, 25.0%; N 24/32, 75.0%; PY 0/32, 0.0%). Study-level risk of bias was assessed in over half of reviews (Item 9: Y 18/32, 56.2%; N 13/32, 40.6%; PY 1/32, 3.1%). Among reviews that conducted a meta-analysis (denominator = Y + N + PY = 23), statistical methods were almost always appropriate (Item 11: Y 22/23, 95.7%; N 1/23, 4.3%; PY 0/23, 0.0%), yet incorporation of risk of bias into interpretation was inconsistent (Item 13: Y 11/32, 34.4%; N 13/32, 40.6%; PY 8/32, 25.0%). Publication-bias diagnostics were common but not universal in meta-analyses (Item 15: Y 17/23, 73.9%; N 5/23, 21.7%; PY 1/23, 4.3%). Overall, critical weaknesses most frequently related to the absence of preregistered protocols, limited transparency around excluded studies, and inconsistent integration of risk of bias into result interpretation.

Regarding non-critical domains, most reviews framed a clear PICOS question (Item 1: Y 31/32, 96.9%; N 0/32, 0.0%; PY 1/32, 3.1%) and justified eligible study designs (Item 3: Y 29/32, 90.6%; N 1/32, 3.1%; PY 2/32, 6.2%). Duplicate processes were less consistently reported, particularly for study selection (Item 5: Y 23/32, 71.9%; N 5/32, 15.6%; PY 4/32, 12.5%) and data extraction (Item 6: Y 21/32, 65.6%; N 6/32, 18.8%; PY 5/32, 15.6%). Descriptions of included studies were typically detailed (Item 8: Y 31/32, 96.9%; N 0/32, 0.0%; PY 1/32, 3.1%), while reporting of primary-study funding was rare (Item 10: Y 1/32, 3.1%; N 31/32, 96.9%; PY 0/32, 0.0%). The extent to which the impact of risk of bias on results was examined varied (Item 12: Y 11/32, 34.4%; N 12/32, 37.5%; PY 9/32, 28.1%). Statistical heterogeneity was addressed where applicable (Item 14: Y 22/32, 68.8%; N 2/32, 6.2%; PY 3/32, 9.4%). Finally, review-level funding and conflicts of interest were usually disclosed (Item 16: Y 29/32, 90.6%; N 2/32, 6.2%; PY 1/32, 3.1%).

3.6 Overlap of the included studies

Overlap results is presented in Supplementary Table S4 and illustrated heatmap for each outcome in Supplementary Figures S1–S10 Across outcome domains, CCA values ranged from 0.0 to 0.6%, indicating slight overlap. The BMI or body weight domain (28 reviews, 715 total study instances, 647 unique studies) showed a CCA of 0.4%, reflecting minimal duplication of primary trials. Similarly small overlaps were observed for waist circumference (0.2%), body composition (0.1%), metabolic markers (0.6%), and blood pressure (0.3%). Physical fitness and function showed a CCA of 0.2%, still within the slight category. Several domains exhibited zero overlap, including quality of life, dietary behavior, and inflammatory markers. However, in domains with few reviews or limited primary study pools, a CCA of 0.0% likely reflects limited opportunity for overlap rather than comprehensive coverage of the evidence base. Overall, the overview is therefore at low risk of double counting across reviews, supporting cautious synthesis of findings across reviews.

Given the generally low methodological quality of many included SRs, these findings should be interpreted conservatively, emphasizing patterns that are consistent across higher-quality reviews and outcomes that are supported by multiple independent syntheses.

3.7 Summary of evidence

A synthesis of 32 systematic reviews shows that multicomponent programs combining structured PA with an energy-restricted or quality-improving diet tend to achieve more favorable outcomes than single-component approaches across the life course.

In adults, reviews consistently report larger reductions in weight, BMI, waist circumference, and fat mass in interventions lasting at least 12 months compared with diet-only or exercise-only strategies, with exercise contributing to improved weight-loss quality through better preservation of lean mass and fitness (32, 46). Cardiometabolic risk markers generally improve modestly yet consistently, including fasting glucose, insulin resistance, CRP, and adipokines, and long-term programs are associated with a substantially lower incidence of type 2 diabetes (35, 47, 48, 50).

In youth, face-to-face, family-engaged diet-plus-activity interventions are most often associated with more durable improvements BMI and BMI-z changes, whereas brief or single-component formats commonly show attenuation of effects in the absence of maintenance (15, 37, 49, 52).

In older adults, hypocaloric diets combined with supervised, resistance-inclusive exercise are associated with favorable trade-offs between weight reduction, preservation of lean mass, limiting sarcopenia, and improvements in function and quality of life (41). Evidence from specific settings, including infertility care, workplaces, and digitally supported or hybrid programs, suggests modest but clinically meaningful short-term benefits, typically on the order of 2–3 kg of weight loss (12, 34, 42, 56).

Across populations, more favorable outcomes are reported more often in combined diet and PA programs of longer duration (≥6–12 months), characterized by higher contact frequency, supervised or structured delivery, inclusion of resistance training, family engagement in youth, and maintenance of an overall energy deficit (40).

These findings remain clinically useful, but they should be interpreted as reflecting an evidence base dominated by traditional lifestyle-intervention models; contemporary obesity science increasingly suggests that treatment response may also be shaped by gut-derived biology, including satiety signaling, incretin pathways, and microbiome-related differences in host energy handling (59, 60).

4 Discussion

This overview indicates that combined PA and diet intervention tends to outperform single-component approaches across weight, adiposity, body-composition quality, fitness, and cardiometabolic risk markers. In adults, combined PA and diet intervention lasting at least 12 months produces larger reductions in weight, BMI, waist, and fat mass than diet-only or exercise-only, and adding exercise preserves lean mass and improves fitness, especially with resistance training. Cardiometabolic markers improve modestly but consistently, driven more by sustained energy deficit and structured exercise than by macronutrient composition, and longer combined PA and diet intervention is associated with lower type 2 diabetes incidence. In youth, face-to-face, family-engaged combined PA and diet intervention yields the largest and most durable BMI gains, while brief or single-component efforts often fade. Benefits are also seen in older adults and in workplace and digital or hybrid settings, typically 2 to 3 kg short-term, and are strongest when contact is high, sessions are structured or supervised, resistance training is integrated, families are engaged, and an energy deficit is maintained.

These findings should also be interpreted relative to the current hierarchy of obesity treatments. Although combined diet and physical activity interventions generally outperform single-component lifestyle approaches, the absolute magnitude of average weight loss in this overview is typically modest compared with modern incretin-based pharmacotherapy and metabolic bariatric surgery. In contemporary randomized trials, semaglutide 2.4 mg and tirzepatide have produced substantially greater mean weight reductions than those typically seen in the lifestyle reviews included here, while current bariatric-surgery guidance continues to regard metabolic bariatric surgery as the most effective durable treatment for severe obesity (16, 17). Accordingly, pharmacological and surgical approaches generally achieve larger average weight-loss magnitudes, while combined PA and diet interventions remain essential as foundational or complementary strategies that support cardiometabolic health, functional capacity, and long-term weight management. Dietary research also increasingly suggests that food processing and dietary composition may influence appetite regulation and weight change through gut-derived pathways, including endogenous GLP-1 signaling, although this evidence remains evolving rather than definitive (6, 7).

Taken together, these patterns suggest that, within the lifestyle reviews included in this overview, intervention features such as duration, contact frequency, supervision, resistance training, and maintenance support were more consistently associated with favorable outcomes than differences in reported macronutrient composition alone (32, 35, 46). This does not imply that dietary composition or gut-derived biology is unimportant; rather, the available review-level evidence was better able to distinguish the contribution of program structure than the contribution of specific mechanistic pathways, including satiety signaling, incretin responses, or microbiome-related differences in host energy handling (6, 59–62). This apparent discrepancy reflects differences in explanatory level. Indeed, while energy imbalance remains a necessary condition for weight change, intervention-level outcomes appear to be more strongly influenced by behavioral and structural factors that determine adherence, implementation, and long-term sustainability.

Accordingly, the present overview supports the importance of sustained, well-designed multicomponent lifestyle programs. At the same time, these findings should be interpreted within a broader and evolving biological framework, in which treatment response is understood to arise from interactions between behavior, diet quality, and underlying physiological processes. In this context, combined lifestyle interventions may operate primarily through behavioral and energy-balance pathways, but their effectiveness may also be modulated by individual biological responsiveness, including gut–brain signaling and microbiome-related mechanisms, although these pathways were not directly assessed in the included reviews and are supported primarily by emerging experimental and translational evidence (6, 59–62).

Where mechanistic explanations are discussed below, these should be interpreted as plausible interpretations consistent with observed patterns across the included systematic reviews, rather than as direct causal inferences from primary experimental evidence.

Relative to diet alone, combined PA and diet intervention produced small to moderate additional reductions in body weight and waist circumference, with larger advantages over PA alone, and parallel improvements in blood pressure and cardiorespiratory fitness. These patterns are consistent with head-to-head meta-analytical contrasts reported in the Results section, indicating modest but clinically relevant additional weight loss relative to single-component interventions, together with small reductions in diastolic blood pressure and meaningful gains in VO₂max (11, 32). Although the kilogram differences are modest, the clinical gain is larger because exercise appears to improve the quality of weight loss, preserves lean mass, and stabilizes resting energy expenditure, which supports function and may reduce weight-regain risk; the stronger advantage over activity-only arms underscores that diet supplies the primary energy deficit while exercise shapes composition and fitness (11, 32, 46). This pattern is consistent with the interpretation that the added benefit of combined PA and diet intervention may reflect the contribution of supervised, structured exercise alongside a sustained dietary energy deficit, rather than changes in self-reported PA alone.

These findings should also be interpreted in the context of the rapidly evolving obesity-treatment landscape. Most reviews included in this overview synthesized trials developed before the current era of highly effective incretin-based anti-obesity pharmacotherapies. In contemporary trials, semaglutide 2.4 mg produced a mean body-weight reduction of 14.9% at 68 weeks, while tirzepatide produced mean reductions approaching 20% at 72 weeks, depending on dose (16, 17). Accordingly, our overview should not be interpreted as suggesting that combined diet-plus-physical-activity interventions achieve weight-loss magnitudes comparable to newer pharmacological therapies. Rather, the present evidence supports lifestyle intervention as a foundational treatment strategy that improves body composition, fitness, and cardiometabolic risk, and that remains relevant whether used alone or alongside pharmacotherapy, which is itself indicated as an adjunct to a reduced-calorie diet and increased physical activity (16, 17).

Importantly, short-term digital and hybrid combined PA and diet intervention also revealed clinically meaningful weight loss over 8 to 26 weeks, and the superiority of combined PA and diet intervention was similar across BMI strata, supporting generalizability from BMI 25 to 40 kg·m⁻² in professionally guided settings (12, 55). Digital delivery appears effective when contact is frequent and structured, but without planned maintenance weight commonly drifts back as coaching intensity falls, mirroring attenuation after face-to-face combined PA and diet intervention once support ends (11, 12).

In severe obesity, combined PA and diet intervention produced clinically significant weight loss, approximately 9 kg on average, and improved cardiometabolic profiles, with effect sizes increasing when contact was more frequent and program duration exceeded 12 months (33, 44). For people with higher BMI, longer combined PA and diet intervention with more frequent support helps the body adapt and improve movement. Supervised sessions that include strength training build function and can also make future treatments more effective if needed (33, 44, 46).

Otherwise, workplace implementations demonstrated moderate but consistent benefits on BMI, waist circumference, and lipids, underscoring feasibility of combined PA and diet intervention in real-world delivery models that include coaching and structured activity rather than information alone (34). These effects are modest per person but meaningful at population scale, suggesting that even relatively small individual-level benefits may translate into relevant public-health gains when implemented broadly, provided that sustained engagement, maintenance support, and enabling system-level conditions are in place. Such effects also depend on behavioral scaffolding such as coaching, self-monitoring, and structured activity, since information-only approaches rarely move BMI or lipids (34).

Over longer horizons, combined PA and diet intervention halved incident type 2 diabetes risk compared with usual care, indicating durable disease-prevention potential even as between-group weight differences attenuate with time (50). This durability points to weight-independent pathways, improved insulin sensitivity, and higher cardiorespiratory fitness as mediators beyond early weight loss (15, 50).

Within combined PA and diet intervention, structured exercise alongside a hypocaloric diet improved fasting glucose by roughly 1.46 mg·dL⁻¹ and insulin resistance by about one third of a standard deviation despite negligible extra weight loss, and it shifted adipokines toward a more cardioprotective profile (lower leptin and higher adiponectin) (48). Pooled inflammation signals showed a small reduction in C-reactive protein, with near-null average effects for IL-6 and TNF-α. In adults with obesity and type 2 diabetes, combined PA and diet intervention reduced HbA1c, fasting glucose, fasting insulin, triglycerides, and blood pressure, with standardized effects in the small to moderate range across outcomes (15, 47). Clinically, these findings support consideration of progressive, structured exercise as part of combined PA and diet intervention, even when extra weight loss is expected to be small, since glycemic control and blood pressure can still improve through exercise-mediated insulin action and vascular effects (15, 48).

Program outcomes tended to improve more consistently with longer duration and higher contact frequency, as combined PA and diet intervention lasting 6–12 months with regular structured sessions produced larger and more lasting effects than shorter or low-contact formats (35, 40, 55). Supervision and structure, particularly the deliberate inclusion of resistance training, were repeatedly linked to superior body-composition and functional outcomes, including better preservation of fat-free mass and larger gains in strength and cardiorespiratory fitness compared with diet alone or aerobic-only exercise (41, 46). In applied settings, workplace and digital combined PA and diet intervention worked best when coaching, self-monitoring, and feedback were built in; fully online delivery achieved short-term losses comparable to in-person models when contact was frequent, but maintenance strategies were essential to limit post-program regain (12, 34). A practical sequence is to first establish an appropriate energy-deficit diet, then add structured exercise including resistance training two to three times per week, and finally build behavior-change support through goal setting, self-monitoring, feedback, and a planned maintenance phase after 3–6 months (11, 35, 46).

Importantly, however, confidence in these conclusions varies across outcome domains. Where higher-quality systematic reviews (AMSTAR 2 high or moderate confidence) were available, particularly for adult weight, adiposity, and fitness outcomes, conclusions are comparatively more robust. In contrast, findings related to some metabolic, inflammatory, and culturally tailored combined PA and diet intervention are derived predominantly from low or critically low-quality reviews and should therefore be interpreted cautiously as provisional rather than definitive.

From an implementation perspective, the evidence synthesized in this overview suggests that effective combined PA and diet intervention is most commonly characterized by dietary energy restriction or quality-improving dietary counseling alongside structured PA; a duration of at least 6–12 months; regular professional contact with supervised or structured sessions; inclusion of resistance-based exercise alongside aerobic activity; and a planned maintenance phase incorporating self-monitoring and booster contacts.

The pediatric pattern echoed adult results but was more sensitive to delivery details. Face-to-face, family-engaged combined PA and diet intervention outperformed usual care and single-component comparators for BMI, BMI-z, waist circumference, and body fat percentage at approximately 6 months, whereas school-based prevention alone often produced small and heterogeneous anthropometric effects. Sustained benefits typically required at least 6 to 12 months of contact with explicit parent involvement, and weight regain was common once support ceased, highlighting the need for planned maintenance and environmental scaffolding in homes and schools (36, 37, 52, 53, 57). Family engagement likely enhances adherence through home meal structure, healthier food availability, and logistics for organized activity, while school-only prevention tends to underdose intensity and contact, which blunts BMI effects; programs should therefore combine parent components with school or community supports for at least one academic year (36, 37, 52).

Culturally tailored combined PA and diet intervention for South-Asian adults yielded small average weight losses, suggesting that tailoring improves acceptability but does not substitute for adequate intensity and duration; in South-Asian school-age children, controlled evidence did not show clear anthropometric benefits, again pointing to dose and family engagement as key effect modifiers (43). Tailoring should be layered on top of sufficient duration and contact rather than seen as a replacement for program dose (43).

Two additional clinical subgroups warrant emphasis. In older adults with obesity, combined PA and diet intervention was consistently superior to diet alone or aerobic-only training for preserving lean mass and bone while improving strength, function, and quality of life; exercise alone improved function but did not reduce weight, highlighting the complementary roles of both components in this population (41). For older adults, resistance training is pivotal to protect muscle and bone during energy restriction; reported adverse events were minor, supporting scalability in community settings (41).

Among individuals with obesity-related infertility and women of reproductive age, combined PA and diet intervention improved body weight and BMI, with some trials reporting larger individual losses and ancillary reproductive benefits in the broader narrative, which strengthens the case for integrating combined PA and diet intervention into fertility care pathways (42, 56). Given low risk and broader cardiometabolic gains, combined PA and diet intervention can run in parallel with fertility care and may improve responsiveness through weight and insulin-sensitivity pathways (42).

The apparent superiority of combined diet and physical activity interventions should not be reduced to a simplistic “calories in versus calories out” model. While sustained dietary energy deficit and structured exercise remain central to the effects summarized in this overview, emerging evidence suggests that obesity-related responses are also shaped by gut-derived mechanisms, including enteroendocrine signaling, endogenous GLP-1 responses, bile-acid pathways, and gut-microbiome-related influences on satiety and metabolizable energy (6, 7, 62, 63). However, these mechanisms should be interpreted as plausible complementary explanations rather than definitive drivers of the observed effects. Controlled human data indicate that diet–microbiome interactions can alter host metabolizable energy, and recent reviews further suggest that modern dietary and pharmacological obesity treatments interact with gut–brain and gut–microbiome pathways (6, 7, 62, 63). Similarly, the metabolic benefits of Roux-en-Y gastric bypass are increasingly understood to involve more than restriction or malabsorption alone; rapid nutrient delivery to the distal intestine is associated with marked postprandial increases in GLP-1 and related gut peptides, alongside other neurohormonal and metabolic adaptations (62, 63). Nevertheless, these pathways were not directly assessed in the primary studies included in this overview, and their contribution to the observed effects cannot be determined. Accordingly, interpretation is limited to comparative effectiveness patterns rather than mechanistic attribution. Future trials should also examine whether combining exercise with higher-quality dietary patterns influences immunometabolic outcomes through pathways beyond weight loss alone, including inflammation, appetite regulation, gut-derived signaling, and microbiome-related metabolites.

Implementation implications follow directly from these patterns. Across adult strata, including metabolic syndrome and higher BMI classes, outcomes tracked more closely with total energy deficit, exercise structure and supervision, and behavior-change components than with macronutrient composition per se within combined PA and diet intervention frameworks. This supports prioritizing adherence-friendly, culturally acceptable dietary patterns paired with progressive, resistance-inclusive PA rather than seeking marginal gains from macronutrient tweaking alone (27, 33, 44, 64). Health systems and employers can scale access through hybrid and online delivery, but should retain coaching, self-monitoring, and feedback loops that drive adherence and outcomes; workplace evidence already demonstrates that such scaffolding is associated with better anthropometric and lipid changes than information-only approaches (12, 34). Health systems can scale access with hybrid models if they preserve coaching and monitoring, and employers can integrate step goals, supervised sessions, and nutrition support to move beyond information-only approaches (34).

In pediatrics, services should privilege combined PA and diet intervention with family engagement, sufficient duration and intensity, and school-environment changes, and avoid reliance on PA-only strategies for BMI change (36, 37, 40, 52). For fertility care, integrating combined PA and diet intervention offers a pragmatic, low-risk adjunct that improves BMI and may enhance reproductive endpoints in broader care pathways (42).

Meta-analytic evidence clarifies both durability and maintenance needs: in adults, resistance-inclusive combined PA and diet intervention outperforms single-component approaches for 12–18 months, and combined PA and diet intervention confers a ~ 56% relative reduction in incident type 2 diabetes on long-term follow-up, whereas in children and adolescents, BMI/BMI-z improvements seen at ~6 months often attenuate within 6–12 months after contact ends (11, 32, 37, 50, 57). Accordingly, maintenance should be planned from the outset, using low-burden booster contacts, continued self-monitoring, environmental supports, and periodic supervised blocks, to sustain fitness and limit regain; especially in youth where withdrawal of support often precedes relapse (11, 37).

Building on this evidence, clinicians may consider prioritizing combining PA and diet intervention with supervised resistance training, frequent contact, and relapse-prevention plans for adults with overweight or obesity, including those with type 2 diabetes or metabolic syndrome (15, 35). For older adults, resistance training is even more critical to preserve musculoskeletal health during weight reduction (41). Health systems and employers should leverage hybrid or online delivery to scale access, while retaining coaching, self-monitoring, and feedback loops that drive adherence and outcomes (12, 34). In pediatrics, policies and services should privilege combined PA and diet intervention with family engagement, sufficient duration and intensity, and school-environment changes while avoiding reliance on PA-only approaches for BMI change (36, 37, 52). Finally, for fertility care in people with obesity, integrating combined PA and diet intervention improves BMI (42).

These cardiometabolic benefits should also be interpreted within the rapidly evolving obesity-treatment landscape. Although combined physical activity and diet interventions improve glycaemic control, lipid profiles, blood pressure, and selected inflammatory markers, the absolute magnitude of average weight loss is generally modest compared with contemporary incretin-based pharmacotherapy and metabolic bariatric surgery (16, 17, 63). Emerging dietary evidence further suggests that replacing ultra-processed foods with minimally processed dietary patterns can improve weight and cardiometabolic health, although the mechanistic contribution of endogenous GLP-1 and related gut-derived pathways remains an active area of investigation rather than a settled explanation (6, 7, 62). Accordingly, the scalability of the present findings should be understood in the context of an evidence base composed largely of traditional lifestyle trials that were not designed to compare directly with modern GLP-1 receptor agonists, bariatric procedures, or microbiome-focused dietary strategies. At the same time, this gap highlights an important research priority, namely the direct evaluation of combined exercise with minimally processed, fiber-rich, polyphenol-rich dietary patterns using integrated immunometabolic, gut-hormone, and microbiome outcomes (62, 65, 66).

4.1 Methodological quality considerations

The AMSTAR 2 appraisal indicates that while a substantial evidence base exists, confidence in specific effect estimates varies considerably across outcome domains. Primary conclusions of this overview are therefore driven mainly by systematic reviews rated as moderate to high quality, particularly in adult populations, and should be interpreted cautiously where evidence is derived predominantly from low or critically low-quality reviews. The AMSTAR 2 appraisal further indicates marked variability in review conduct and reporting across the 32 systematic reviews, with overall confidence skewed toward the lower end of the scale. Ratings were High 10/32 (31.2%), Moderate 2/32 (6.2%), Low 10/32 (31.2%), and critically low 10/32 (31.2%). This distribution warrants caution when interpreting pooled or narrative conclusions derived from the corpus, as multiple critical weaknesses, rather than minor reporting lapses, drove most downgrades.

Three items emerge from the critical domains. First, protocol preregistration (Item 2) was uncommon (10/32 “Yes” [31.2%]; 22/32 “No” [68.8%]), which heightens the risk of post hoc decision making (e.g., selective eligibility criteria and flexible outcome handling) and undermines reproducibility. Routine registration (e.g., PROSPERO or OSF), with timestamped, prespecified methods, should be considered foundational. Second, transparency of study exclusion (Item 7) was poor (8/32 “Yes” [25.0%]; 24/32 “No” [75.0%]), limiting auditability of the search-to-inclusion pipeline and raising concerns about selection bias. Publishing a PRISMA-compliant exclusions table with reasons should be standard practice. Third, while study-level risk of bias assessment (Item 9) was performed in over half of reviews (18/32 “Yes” [56.2%]; 13/32 “No” [40.6%]; 1/32 “Partial Yes” [3.1%]), its integration into interpretation (Item 13) was inconsistent (11/32 “Yes” [34.4%]; 13/32 “No” [40.6%]; 8/32 “Partial Yes” [25.0%]). This disconnect likely weakens inferential strength even when risk of bias is assessed, because credibility judgments are not propagated into the analytic or narrative synthesis (e.g., via sensitivity analyses, stratification, or explicit downgrading of confidence).

Where meta-analyses were conducted (23/32 reviews; 71.9%), statistical methods were almost always appropriate (Item 11: 22/23 “Yes” [95.7%]), and publication-bias diagnostics were attempted in most cases (Item 15: 17/23 “Yes” [73.9%]). These are notable strengths. However, small-study bias tests are underpowered when k is small, and their absence in roughly a quarter of meta-analyses, combined with weak protocol use and exclusion-list transparency, means residual reporting and selection biases cannot be ruled out. Moreover, even technically sound meta-analysis cannot compensate for upstream shortcomings in protocolization, selection transparency, or the failure to carry risk-of-bias judgments into conclusions.

The non-critical domains were stronger. Almost all reviews framed clear PICOS questions (Item 1: 31/32 “Yes” [96.9%]; 1/32 “Partial Yes” [3.1%]) and described included studies in detail (Item 8: 31/32 “Yes” [96.9%]; 1/32 “Partial Yes” [3.1%]). Most justified eligible designs (Item 3: 29/32 “Yes” [90.6%]), and heterogeneity was commonly addressed where applicable (Item 14: 22/32 “Yes” [68.8%]; 2/32 “No” [6.2%]; 3/32 “Partial Yes” [9.4%]). Nevertheless, duplicate processes were not universal: study selection was performed in duplicate in 23/32 (Item 5: 71.9%) and duplicate data extraction in 21/32 (Item 6: 65.6%), leaving room for avoidable screening and transcription errors. Most strikingly, only 1/32 reviews reported funding sources of the primary studies (Item 10: 3.1%), a pervasive gap that limits assessment of sponsor influence and constrains bias-related sensitivity analyses.

Collectively, these patterns suggest that the credibility of effect estimates in this literature is constrained less by analytic technique and more by design transparency, selection accountability, and evidence-appraisal practice. For decision makers; and for any umbrella review or overview drawing on this corpus; two safeguards are advisable: (i) weight conclusions toward reviews with moderate to high AMSTAR 2 confidence and better compliance with critical items, and (ii) conduct sensitivity/best-evidence syntheses that down-weight or explicitly flag findings dominated by critically low reviews, especially where conclusions hinge on small, heterogeneous, or industry-funded primary evidence.

4.2 Limitations of the evidence base and of this OoSRs

This overview has several limitations. Our findings rely on the quality of 32 source reviews, which varied substantially: approximately 31% were high quality and roughly 69% were low or critically low; 69% were not preregistered, 75% did not list excluded studies, and only 34% formally incorporated risk-of-bias; any of which may have influenced pooled or narrative conclusions. We did not apply GRADE at the overview level because methods are unsettled, so we cannot formally rate certainty. Primary studies differed substantially in intervention length, contact frequency, staffing, exercise type and dose (aerobic, resistance, HIIT), diet approach (balanced hypocaloric, Mediterranean/DASH, lower-fat/carbohydrate, meal replacements), and delivery (in-person vs. digital), making it hard to credit specific components and pushing some reviews toward narrative summaries. Outcomes were defined and measured on different schedules (e.g., BMI vs. BMI-z in youth; mixed fitness tests), which complicates harmonization and can inflate variability. Durability is a concern: several reviews reported that benefits fade 6–12 months after programs end, highlighting the need for maintenance or booster plans and longer follow-up. Although overall overlap across reviews was small (corrected covered area 0.0–0.6%), very low overlap in thinly covered topics likely reflects limited coverage rather than true independence, so some duplication or gaps may remain. Selective reporting and publication bias cannot be ruled out; bias tests were not universal, funding of primary studies was rarely reported, and adverse events and cost outcomes were sparse and inconsistent. Implementation factors (adherence, fidelity) were inconsistently measured; often by self-report; and were seldom linked to outcomes; cost reporting was limited. Generalizability is uneven: while there is evidence for severe obesity, older adults, and infertility, data are thin in low-resource settings and among adolescents in transition. Finally, we used the effect estimates as reported by each systematic review; without reanalyzing primary data, any analytic choices or errors in those reviews carry into this overview. An additional limitation is that most included reviews synthesized traditional lifestyle trials conducted before the current GLP-1RA era and outside direct comparison with metabolic bariatric surgery or minimally processed-food interventions; therefore, this overview is better suited to evaluating the relative effectiveness of combined versus single-component lifestyle strategies than to ranking lifestyle treatment against contemporary pharmacological, surgical, or gut-microbiota-oriented dietary approaches.

5 Conclusion

Across the included systematic reviews, combined physical activity and diet interventions generally demonstrated more favorable outcomes than single-component strategies, with the most consistent benefits observed in programs that were longer, structured, supervised, and supported by behavior-change strategies, and, in youth, included active family engagement. Although the average magnitude of weight loss was usually modest, combined interventions contributed importantly to improved body-composition quality, fitness, and cardiometabolic health, with some benefits extending beyond additional weight loss alone. Their value may therefore lie less in producing large absolute weight reductions than in improving a multicomponent behavioral and clinical foundation for long-term obesity care. At the population level, even modest individual effects may still be relevant when delivered at scale, although their public-health impact is likely to depend on sustained engagement, maintenance support, and enabling system-level conditions.

In contemporary obesity management, lifestyle intervention and pharmacotherapy should not be viewed as competing options; rather, structured physical activity and dietary support remain foundational components of obesity care, and are increasingly used alongside anti-obesity medications to enhance weight reduction, improve cardiometabolic health, and support long-term weight maintenance. At the same time, contemporary pharmacological and surgical treatments generally achieve substantially larger average weight loss than the lifestyle-only interventions synthesized here, which reinforces the importance of interpreting lifestyle treatment as a foundational and complementary component of obesity care rather than a direct substitute for higher-efficacy therapies.

Future work should prespecify and transparently report behavior-change components and contact dose to enable dose–response analyses; standardize core outcome sets (e.g., BMI and BMI-z, waist circumference, body-fat percentage, VO₂peak, HbA1c, HOMA-IR) and include objective device-based activity and sedentary metrics; use dismantling and factorial designs to isolate the incremental value of resistance training, supervision, and digital features; extend follow-up with explicit maintenance and relapse-management strategies; report cost-effectiveness to inform scaling decisions; and improve representation of older adults, adolescents, and underserved groups. Future studies should also examine whether combining exercise with minimally processed, fiber-rich, higher-quality dietary patterns yields additional immunometabolic benefits, including effects on appetite-related, gut-derived, and microbiota-related pathways, as the evidence base becomes more mature.

References

• 1.

  PhelpsNHSingletonRKZhouBHeapRAMishraABennettJEet al. Worldwide trends in underweight and obesity from 1990 to 2022: a pooled analysis of 3663 population-representative studies with 222 million children, adolescents, and adults. Lancet. (2024) 403:1027–50. doi: 10.1016/S0140-6736(23)02750-2,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 2.

  FryarCDAffulJSaifNTPrevalence of Overweight, Obesity, and Severe Obesity among Adults Age 20 and Older: United States, 1960–1962 through August 2021–August 2023. NCHS Health E Stats (2026):1–7. doi: 10.15620/cdc/174643

  • CrossRef
  • Google Scholar

• 3.

  TremmelMGerdthamU-GNilssonPMSahaS. Economic burden of obesity: a systematic literature review. Int J Environ Res Public Health. (2017) 14:435. doi: 10.3390/ijerph14040435,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4.

  BullFCAl-AnsariSSBiddleSBorodulinKBumanMPCardonGet al. World Health Organization 2020 guidelines on physical activity and sedentary behaviour. Br J Sports Med. (2020) 54:1451–62. doi: 10.1136/bjsports-2020-102955,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 5.

  TouvierMda Costa LouzaMLMozaffarianDBakerPJuulFSrourB. Ultra-processed foods and cardiometabolic health: public health policies to reduce consumption cannot wait. BMJ. (2023) 383:e075294. doi: 10.1136/bmj-2023-075294

  • CrossRef
  • Google Scholar

• 6.

  HuberHSchierenAHolstJJSimonM-C. Dietary impact on fasting and stimulated Glp-1 secretion in different metabolic conditions–a narrative review. Am J Clin Nutr. (2024) 119:599–627. doi: 10.1016/j.ajcnut.2024.01.007,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7.

  DickenSJJassilFCBrownAKalisMStanleyCRansonCet al. Ultraprocessed or minimally processed diets following healthy dietary guidelines on weight and cardiometabolic health: a randomized, crossover trial. Nat Med. (2025) 31:3297–308. doi: 10.1038/s41591-025-03842-0,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 8.

  DaiSWellensJYangNLiDWangJWangLet al. Ultra-processed foods and human health: an umbrella review and updated meta-analyses of observational evidence. Clin Nutr. (2024) 43:1386–94. doi: 10.1016/j.clnu.2024.04.016,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9.

  PontzerHRaichlenDAWoodBMMabullaAZRacetteSBMarloweFW. Hunter-gatherer energetics and human obesity. PLoS One. (2012) 7:e40503. doi: 10.1371/journal.pone.0040503,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10.

  McGroskyALukeAArabLBedu-AddoKBonomiAGBovetPet al. Energy expenditure and obesity across the economic Spectrum. Proc Natl Acad Sci. (2025) 122:e2420902122. doi: 10.1073/pnas.2420902122,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 11.

  JohnsDJHartmann-BoyceJJebbSA, Group BWMR. Diet or exercise interventions vs combined behavioral weight management programs: a systematic review and meta-analysis of direct comparisons. J Acad Nutr Diet (2014) 114:1557–1568. doi: 10.1016/j.jand.2014.07.005

  • CrossRef
  • Google Scholar

• 12.

  RotundaWRainsCJacobsSRNgVLeeRRutledgeSet al. Weight loss in short-term interventions for physical activity and nutrition among adults with overweight or obesity: a systematic review and meta-analysis. Prev Chronic Dis. (2024) 21:E21. doi: 10.5888/pcd21.230347

  • CrossRef
  • Google Scholar

• 13.

  Jiménez-PeláezCCFernández-AparicioÁMontero-AlonsoMAGonzález-JiménezE. Effect of dietary and physical activity interventions combined with psychological and behavioral strategies on preventing metabolic syndrome in adolescents with obesity: a Meta-analysis of clinical trials. Nutrients. (2025) 17:2051. doi: 10.3390/nu17132051,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14.

  VerheggenRMaessenMGreenDJHermusAHopmanMThijssenD. A systematic review and meta-analysis on the effects of exercise training versus hypocaloric diet: distinct effects on body weight and visceral adipose tissue. Obes Rev. (2016) 17:664–90. doi: 10.1111/obr.12406,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15.

  Al-MhannaSBRocha-RodriguescSMohamedMBatrakoulisAAldhahiMIAfolabiHAet al. Effects of combined aerobic exercise and diet on cardiometabolic health in patients with obesity and type 2 diabetes: a systematic review and meta-analysis. BMC Sports Sci Med Rehabil. (2023) 15:165. doi: 10.1186/s13102-023-00766-5,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16.

  WildingJPHBatterhamRLCalannaSDaviesMvan GaalLFLingvayIet al. Once-weekly semaglutide in adults with overweight or obesity. N Engl J Med. (2021) 384:989–1002. doi: 10.1056/NEJMoa2032183

  • CrossRef
  • Google Scholar

• 17.

  JastreboffAMle RouxCWStefanskiAAronneLJHalpernBWhartonSet al. Tirzepatide for obesity treatment and diabetes prevention. N Engl J Med. (2025) 392:958–71. doi: 10.1056/NEJMoa2410819

  • CrossRef
  • Google Scholar

• 18.

  PollockMFernandesRMBeckerLAPieperDHartlingL. "Chapter V: overviews of reviews". In: Cochrane Handbook for systematic Reviews of Interventions Version, London: Cochrane. vol. 6 (2020).

  • Google Scholar

• 19.

  GatesMGatesAPieperDFernandesRMTriccoACMoherDet al. Reporting guideline for overviews of reviews of healthcare interventions: development of the prior statement. BMJ. (2022) 378:e070849. doi: 10.1136/bmj-2022-070849,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 20.

  SheaBJReevesBCWellsGThukuMHamelCMoranJet al. Amstar 2: a critical appraisal tool for systematic reviews that include randomised or non-randomised studies of healthcare interventions, or both. BMJ. (2017) 358:j4008. doi: 10.1136/bmj.j4008

  • CrossRef
  • Google Scholar

• 21.

  WhitingPSavovićJHigginsJPCaldwellDMReevesBCSheaBet al. Robis: a new tool to assess risk of bias in systematic reviews was developed. J Clin Epidemiol. (2016) 69:225–34. doi: 10.1016/j.jclinepi.2015.06.005,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 22.

  SantaguidaPLKeshavarzHCarlessoLCLomotanMGrossAMacDermidJCet al. A description of the methodology used in an overview of reviews to evaluate evidence on the treatment, harms, diagnosis/classification, prognosis and outcomes used in the management of neck pain. Open Orthop J. (2013) 7:461. doi: 10.2174/1874325001307010461,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23.

  HartlingLChisholmAThomsonDDrydenDM. A descriptive analysis of overviews of reviews published between 2000 and 2011. PLoS One. (2012) 7:e49667. doi: 10.1371/journal.pone.0049667,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24.

  BallardMMontgomeryP. Risk of Bias in overviews of reviews: a scoping review of methodological guidance and four-item checklist. Res Synth Methods. (2017) 8:92–108. doi: 10.1002/jrsm.1229,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25.

  PieperDAntoineS-LMathesTNeugebauerEAEikermannM. Systematic review finds overlapping reviews were not mentioned in every other overview. J Clin Epidemiol. (2014) 67:368–75. doi: 10.1016/j.jclinepi.2013.11.007,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 26.

  BougioukasKIDiakonidisTMavromanoliACHaidichAB. Ccar: a package for assessing primary study overlap across systematic reviews in overviews. Res Synth Methods. (2023) 14:443–54. doi: 10.1002/jrsm.1610,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27.

  Pérez-BracchiglioneJMezaNBangdiwalaSIde GuzmánENUrrútiaGBonfillXet al. Graphical representation of overlap for overviews: groove tool. Res Synth Methods. (2022) 13:381–8. doi: 10.1002/jrsm.1557

  • CrossRef
  • Google Scholar

• 28.

  McKenzieJEBrennanSE. "Synthesizing and presenting findings using other methods". In: Cochrane Handbook for systematic Reviews of Interventions. 2nd ed. Chichester, UK: Wiley, (2019) 321–47.

  • Google Scholar

• 29.

  KorichevaJGurevitchJMengersenK. Handbook of Meta-Analysis in Ecology and Evolution. Princeton, NJ: Princeton University Press (2013).

  • Google Scholar

• 30.

  LaraMAmigoH. What kind of intervention has the Best results to reduce the weight in Overweighted or obese adults?Archivos Latinoamericanos Nutr. (2011) 61:45–54. doi: 10.37527/2011.61.1.006,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 31.

  HoMJensenMEBurrowsTNeveMGarnettSPBaurLet al. Best practice dietetic management of overweight and obese children and adolescents: a 2010 update of a systematic review. JBI Database System Rev Implement Rep. (2013) 11:190–293. doi: 10.11124/jbisrir-2013-890

  • CrossRef
  • Google Scholar

• 32.

  SchwingshacklLDiasSHoffmannG. Impact of Long-term lifestyle Programmes on weight loss and cardiovascular risk factors in overweight/obese participants: a systematic review and network Meta-analysis. Syst Rev. (2014) 3:130. doi: 10.1186/2046-4053-3-130,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 33.

  HassanYHeadVJacobDBachmannMDiuSFordJ. Lifestyle interventions for weight loss in adults with severe obesity: a systematic review. Clinical obesity. (2016) 6:395–403. doi: 10.1111/cob.12161,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 34.

  Gea CabreraACaballeroPWanden-BergheCSanz-LorenteMLópez-PintorE. Effectiveness of workplace-based diet and lifestyle interventions on risk factors in workers with metabolic syndrome: a systematic review, Meta-analysis and Meta-regression. Nutrients. (2021) 13:4560. doi: 10.3390/nu13124560,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35.

  Albert PerezEPoveda GonzálezMMartínez-EspinosaRMMolina VilaMDReig Garcia-GalbisM. Practical guidance for interventions in adults with metabolic syndrome: diet and exercise vs. changes in body composition. Int J Environ Res Public Health. (2019) 16:3481. doi: 10.3390/ijerph16183481,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 36.

  McGovernLJohnsonJNPauloRHettingerASinghalVKamathCet al. Treatment of pediatric obesity: a systematic review and meta-analysis of randomized trials. J Clin Endocrinol Metab. (2008) 93:4600–5. doi: 10.1210/jc.2006-2409,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 37.

  PeirsonLFitzpatrick-LewisDMorrisonKWarrenRAliMURainaP. Treatment of overweight and obesity in children and youth: a systematic review and meta-analysis. CMAJ Open. (2015) 3:E35–46. doi: 10.9778/cmajo.20140047,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 38.

  LingJRobbinsLBWenF. Interventions to prevent and manage overweight or obesity in preschool children: a systematic review. Int J Nurs Stud. (2016) 53:270–89. doi: 10.1016/j.ijnurstu.2015.10.017,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39.

  AngawiKGaissiA. Systematic review of setting-based interventions for preventing childhood obesity. Biomed Res Int. (2021) 2021:4477534. doi: 10.1155/2021/4477534,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 40.

  Albornoz-GuerreroJGarciaSSevillaGGPCigarroaIZapata-LamanaR. Characteristics of multicomponent interventions to treat childhood overweight and obesity in extremely cold climates: a systematic review of a randomized controlled trial. Int J Environ Res Public Health. (2021) 18:3098. doi: 10.3390/ijerph18063098,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 41.

  BatsisJAGillLEMasutaniRKAdachi-MejiaAMBluntHBBagleyPJet al. Weight loss interventions in older adults with obesity: a systematic review of randomized controlled trials since 2005. J Am Geriatr Soc. (2017) 65:257–68. doi: 10.1111/jgs.14514,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 42.

  BestDAvenellABhattacharyaS. How effective are weight-loss interventions for improving fertility in women and men who are overweight or obese? A systematic review and Meta-analysis of the evidence. Hum Reprod Update. (2017) 23:681–705. doi: 10.1093/humupd/dmx027,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43.

  BrownTSmithSBhopalRKasimASummerbellC. Diet and physical activity interventions to prevent or treat obesity in south Asian children and adults: a systematic review and Meta-analysis. Int J Environ Res Public Health. (2015) 12:566–94. doi: 10.3390/ijerph120100566,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 44.

  BaillotARomainAJBoisvert-VigneaultKAudetMBaillargeonJPDionneIJet al. Effects of lifestyle interventions that include a physical activity component in class ii and iii obese individuals: a systematic review and Meta-analysis. PLoS One. (2015) 10:e0119017. doi: 10.1371/journal.pone.0119017,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45.

  OlatejuIVOpaleye-EnakhimionTUdeoguJEAsuquoJOlaleyeKTOsaEet al. A systematic review on the effectiveness of diet and exercise in the management of obesity. Diabetes Metab Syndr Clin Res Rev. (2023) 17:102759. doi: 10.1016/j.dsx.2023.102759,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 46.

  MillerCTFraserSFLevingerIStraznickyNEDixonJBReynoldsJet al. The effects of exercise training in addition to energy restriction on functional capacities and body composition in obese adults during weight loss: a systematic review. PLoS One. (2013) 8:e81692. doi: 10.1371/journal.pone.0081692,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 47.

  KhalafiMSakhaeiMHKheradmandSSymondsMERosenkranzSK. The impact of exercise and dietary interventions on circulating leptin and adiponectin in individuals who are overweight and those with obesity: a systematic review and Meta-analysis. Adv Nutr. (2023) 14:128–46. doi: 10.1016/j.advnut.2022.10.001,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 48.

  KhalafiMAzali AlamdariKSymondsMERohaniHSakhaeiMH. A comparison of the impact of exercise training with dietary intervention versus dietary intervention alone on insulin resistance and glucose regulation in individual with overweight or obesity: a systemic review and meta-analysis. Crit Rev Food Sci Nutr. (2023) 63:9349–63. doi: 10.1080/10408398.2022.2064424,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 49.

  LiuYHongFLebakaVRMohammedAJiLZhangYet al. Calorie restriction with exercise intervention improves inflammatory response in overweight and obese adults: a systematic review and Meta-analysis. Front Physiol. (2021) 12:754731. doi: 10.3389/fphys.2021.754731,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 50.

  MerlottiCMorabitoACerianiVPontiroliAE. Prevention of type 2 diabetes in obese at-risk subjects: a systematic review and meta-analysis. Acta Diabetol. (2014) 51:853–63. doi: 10.1007/s00592-014-0624-9,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 51.

  SelvendranSSPenneyNCAggarwalNDarziAWPurkayasthaS. Treatment of obesity in young people—a systematic review and meta-analysis. Obes Surg. (2018) 28:2537–49. doi: 10.1007/s11695-018-3285-x,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 52.

  LiangJHZhaoYChenYCJiangNZhangSXHuangSet al. Face-to-face physical activity incorporated into dietary intervention for overweight/obesity in children and adolescents: a Bayesian network Meta-analysis. BMC Med. (2022) 20:325. doi: 10.1186/s12916-022-02462-6,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53.

  Aguilar CorderoMJOrtegón PiñeroABaena GarcíaLNoack SegoviaJPLevet HernándezMCSánchez LópezAM. Efecto Rebote De Los Programas De Intervención Para Reducir El Sobrepeso Y La Obesidad De Niños Y Adolescentes: Revisión Sistemática. Nutr Hosp. (2015) 32:2508–17. doi: 10.3305/nh.2015.32.6.10071

  • CrossRef
  • Google Scholar

• 54.

  AlMarzooqiMANagyMC. Childhood obesity intervention programs: a systematic review. Life Sci. (2011) 8:45–60.

  • Google Scholar

• 55.

  BarteJCVeldwijkJTeixeiraPJSacksFMBemelmansWJ. Differences in weight loss across different BMI classes: a meta-analysis of the effects of interventions with diet and exercise. Int J Behav Med. (2014) 21:784–93. doi: 10.1007/s12529-013-9355-5

  • CrossRef
  • Google Scholar

• 56.

  Ruiz-GonzálezDCavero-RedondoIHernández-MartínezABaena-RayaAMartínez-ForteSAltmäeSet al. Comparative efficacy of exercise, diet and/or pharmacological interventions on BMI, ovulation, and hormonal profile in reproductive-aged women with overweight or obesity: a systematic review and network meta-analysis. Hum Reprod Update. (2024) 30:472–87. doi: 10.1093/humupd/dmae008,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 57.

  SalamRAPadhaniZADasJKShaikhAYHoodbhoyZJeelaniSMet al. Effects of lifestyle modification interventions to prevent and manage child and adolescent obesity: a systematic review and meta-analysis. Nutrients. (2020) 12:2208. doi: 10.3390/nu12082208,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 58.

  AppuhamyJRNKebreabESimonMYadaRMilliganLPFranceJ. Effects of diet and exercise interventions on diabetes risk factors in adults without diabetes: Meta-analyses of controlled trials. Diabetol Metab Syndr. (2014) 6:127. doi: 10.1186/1758-5996-6-127,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 59.

  CorbinKDCarneroEADirksBIgudesmanDYiFMarcusAet al. Host-diet-gut microbiome interactions influence human energy balance: a randomized clinical trial. Nat Commun. (2023) 14:3161. doi: 10.1038/s41467-023-38778-x,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 60.

  MasiDWatanabeMClémentK. Gut microbiome and obesity care: bridging dietary, surgical, and pharmacological interventions. Cell Rep Med. (2026) 7:102573. doi: 10.1016/j.xcrm.2025.102573,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 61.

  CarmodyRNVaradyKTurnbaughPJ. Digesting the complex metabolic effects of diet on the host and microbiome. Cell. (2024) 187:3857–76. doi: 10.1016/j.cell.2024.06.032,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 62.

  ChaoJColemanRAKeatingDJMartinAM. Gut microbiome regulation of gut hormone secretion. Endocrinology. (2025) 166:bqaf004. doi: 10.1210/endocr/bqaf004,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 63.

  EisenbergDShikoraSAAartsEAminianAAngrisaniLCohenRVet al. American Society of Metabolic and Bariatric Surgery (ASMBs) and International Federation for the Surgery of Obesity and Metabolic Disorders (IFSO) Indications for Metabolic and Bariatric Surgery. WLS (2023) 33:3–14. doi: 10.1007/s11695-022-06332-1

  • CrossRef
  • Google Scholar

• 64.

  Bondyra-WiśniewskaBMyszkowska-RyciakJHartonA. Impact of lifestyle intervention programs for children and adolescents with overweight or obesity on body weight and selected Cardiometabolic factors—a systematic review. Int J Environ Res Public Health. (2021) 18:2061. doi: 10.3390/ijerph18042061,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 65.

  Pérez-PrietoIPlaza-FloridoAUbago-GuisadoEOrtegaFBAltmäeS. Physical activity, sedentary behavior and microbiome: a systematic review and meta-analysis. J Sci Med Sport. (2024) 27:793–804. doi: 10.1016/j.jsams.2024.07.003,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 66.

  AslamHTrakmanGDissanayakeTToddEHarrisonPAlbyCet al. Dietary interventions and the gut microbiota: a systematic literature review of 80 controlled clinical trials. J Transl Med. (2026) 24:39. doi: 10.1186/s12967-025-07428-9,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 67.

  MattosSDa CunhaMRSilvaMIBSerfatyFTarvainenMPKleinMRSTet al. Effects of weight loss through lifestyle changes on heart rate variability in overweight and obese patients: a systematic review. Clin Nutr. (2022) 41:2577–86. doi: 10.1016/j.clnu.2022.09.009

  • CrossRef
  • Google Scholar