The effectiveness of self-directed meal replacement-assisted intermittent fasting in adults

Abstract

Objective

This study examined whether the addition of meal replacement (MR)-assisted intermittent fasting increases the effectiveness of a lifestyle intervention.

Methods

In a nonrandomized intervention study, overweight and obese participants aged 18 to 60 years were recruited for weight management with MRs or lifestyle intervention alone (LIA) for 8 weeks. The outcomes were the percent change in body weight from baseline to weeks 4 and 8 and the associated changes in body composition (using a bioimpedance analyzer). Generalized estimating equation (GEE) models were used to compare outcomes between groups.

Results

A total of 126 participants were recruited, and 74 participants in the MR group and 46 participants in the LIA group finished the intervention protocol. The mean age was 35.4 ± 9.7 years, and 75.4% were female. The baseline BMI was 26.6 ± 3.7 kg/m². At 4 weeks, the percent weight loss in the MR group was 6.3%, whereas it was 4.0% in the LIA group (P < 0.001). At 8 weeks, the percent weight loss in the MR group was 8.2%, whereas it was 5.8% in the LIA group (P = 0.004). The GEE models revealed no further differences in weight loss or related measurements between the groups, whereas the time effect was significant.

Conclusions

Our study demonstrated that lifestyle interventions can result in weight loss regardless of the use of MR-assisted intermittent fasting. This finding suggests that nutritional support needs to be provided during short-term weight loss interventions.

Trial registration

Chinese Clinical Trial Registry number: ChiCTR2500099520, Retrospectively registered 25 March 2025.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12906-025-05016-4.

Introduction

The global obesity epidemic has reached alarming proportions, affecting over 2.5 billion adults and 390 million children and adolescents worldwide [1]. China has the highest number of people living with overweight and obesity in the world. The latest national prevalence rates were 34.3% for overweight and 16.4% for obesity among adults [2]. Obesity is a major risk factor for numerous chronic conditions, including type 2 diabetes, cardiovascular diseases, fatty liver, and cancer [3]. A market research survey of 22,008 people from 30 countries revealed that 45–60% of individuals across 30 countries are currently trying to lose weight [4]. Moreover, there has been a significant increase in prescriptions of weight loss medications among active duty service members [5]. Despite widespread efforts to combat obesity, sustained success remains elusive. Even modest weight reductions of 5–10% can markedly improve obesity-related complications and overall health, yet achieving these goals often requires more than lifestyle modifications alone [6].

Lifestyle interventions, encompassing dietary adjustments, physical activity, and behavioral changes, constitute the cornerstone of obesity management [7] and are endorsed as first-line treatments by guidelines in China, Europe and the United States. However, lifestyle interventions alone typically fail to achieve or sustain treatment targets without complementary therapies [8].

Dietary strategies, such as caloric restriction, are central to these efforts. Intermittent fasting (IF), in which individuals alternate between fasting and feeding periods, has emerged as a promising approach that results in moderate weight loss (3–8% of initial body weight) within 8–12 weeks while improving metabolic health [9, 10].

Meal replacements (MRs), defined as prepackaged, calorie-controlled substitutes for one or more daily meals, offer a practical tool to increase dietary adherence [11]. Systematic reviews have shown that the use of meal replacements not only facilitates greater weight loss but also simplifies compliance, especially for individuals with limited cooking skills, owing to the standardized portion sizes and convenience of MRs [12]. When combined with lifestyle interventions, MRs have been shown to amplify reductions in body weight and insulin levels [13]. Recent studies suggest that MRs, particularly high-protein formulations, may further optimize weight management, irrespective of fasting protocols [14].

Despite these advances, critical gaps persist. First, most studies have been focused on body weight or BMI reductions, whereas body fat, particularly visceral fat mass, is recognized as the primary causative factor in obesity [15, 16] and remains insufficiently investigated. Second, the synergistic effects of MR-assisted intermittent fasting within lifestyle interventions are poorly characterized. Further investigation is needed to elucidate the effects of weight-management meal replacements on adiposity and body composition [17, 18]. These gaps are addressed in this study by evaluating the effectiveness of MR-assisted intermittent fasting combined with lifestyle interventions versus lifestyle interventions alone (LIA) on weight loss, body composition, and metabolic parameters improvement in adults.

Methods

Design and participants

A nonrandomized intervention study was conducted to assess the effectiveness of self-directed meal replacement-assisted intermittent fasting in adults. All participants self-selected to receive lifestyle interventions alone (LIA) or lifestyle interventions plus meal replacement-assisted intermittent fasting (MR) for eight weeks. We recruited participants aged 18–60 years with a stable weight change (± 4 kg) over the last three months who met at least one of the following criteria: overweight, with a BMI ≥ 24 kg/m^{2 [19]}; a fat mass percentage ≥ 30% for women and ≥ 25% for men ( [20]; or a waist circumference ≥ 85 cm for women and ≥ 90 cm for men ( [21].

The exclusion criteria were tumors; liver or kidney dysfunction; acute cardiovascular or cerebrovascular disease; acute infectious disease; psychiatric disorder; bulimia nervosa; anorexia; gastrointestinal disease; surgery in the past three months; pregnancy or lactation; oral contraceptive use; hormone replacement therapy; secondary obesity due to an endocrine or monogenetic disorder; and artificial pacemakers. Participants were excluded if they had already been dieting to lose weight. Participants were also excluded if they were taking the following medications 14 days before the initiation of this study: hypoglycemic medications, psychotropic medications, weight-loss medications, and other medications known to affect body weight or energy metabolism.

In 2021, a total of 147 participants were recruited from the Taizhou area via posters, emails, WeChat, social media, and website advertisements. Among them, 16 did not meet the inclusion criteria, and 5 refused to participate; 126 eligible participants agreed to participate in the interventions. After excluding 6 individuals who were lost to follow-up at 8 weeks, 120 participants finished the full management period. First, all the participants provided written informed consent after fully understanding the purpose and nature of the trial. The participants were then divided into two groups according to whether they volunteered to use the meal replacements: the lifestyle intervention plus meal replacement (MR)-assisted intermittent fasting group (n = 74) and the lifestyle intervention alone (LIA) group (n = 46). The workflow for sample selection is depicted in Fig. 1. This trial was conducted at Taizhou Hospital of Zhejiang Province (China), the protocol was approved by the hospital’s ethics committee (K20210607), and the trial was conducted according to the principles of the 2013 Declaration of Helsinki [22]. This study was retrospectively registered in the Chinese Clinical Trial Registry (ChiCTR) with the code number of ChiCTR2500099520 in 25 March 2025.

Fig. 1.

Flow diagram of participant recruitment and study design

Lifestyle and dietary intervention

At the beginning of the trial, all participants were required to complete a sociodemographic information and medical history questionnaire while accompanied by a trained staff member. Throughout the management period, all participants were instructed to avoid taking supplements other than those provided by a dietician and medications that might affect their weight. The participants were encouraged to walk 10,000 steps or perform equivalent physical activity each day without supervision.

All participants were instructed to adhere to prescribed diets. The participants in the LIA group consumed a calorie-restricted diet (1200–1400 kcal/d for men and 1000–1200 kcal/d for women) at home each day [23]. In accordance with the dietary guidelines for Chinese residents [24], the participants were instructed by a dietician to eat a balanced diet consisting of cereals, potatoes, fruits, vegetables, fish, eggs, milk and beans on a daily basis. Total caloric intake was restricted to 1200–1400 kcal/d in males and 1000–1200 kcal/d in females. All participants were required to have an average daily intake of 120–200 g of fish, poultry, eggs and lean meat; 150–200 g of cereal; no less than 300 g of fresh vegetables; and 1.5–1.8 L of water per day. The use of a balanced meal tray was recommended. Before each of the three meals a day were eaten, the individual food was photographed and uploaded to the management WeChat group. The nutritionist assessed the diet structure and total calories according to the type and volume of food and adjusted improper dietary intake in a timely manner.

The participants in the MR group received a dietary intervention that included a 5:2 intermittent fasting regimen. This diet regimen included two nonconsecutive fasting days (800 kcal/d, Tuesdays and Fridays) and five feeding days (the remaining 5 days). During the 8-week weight management period, participants in the MR group took a meal replacement substituting three meals/day (~ 800 kcal) during the fasting days and consumed a calorie-restricted diet (1200–1400 kcal/d for men and 1000–1200 kcal/d for women) at home on each feeding day. Each MR serving contains 47 g of nutritional powder (205 kcal, 12.3 g protein, 26.1 g carbohydrates, and 5.2 g fat). The meal replacement powder (manufactured by Xi’an Lipang Clinical Nutrition Co. Ltd., Xi’an, Shaanxi, China) was prepared by a dietitian, and the detailed nutritional composition is shown in Supplementary Table 1.

The participants were permitted to take probiotic and dietary fiber supplements provided by a dietitian and to consume 150–200 g of nonstarchy and sugar-free vegetables, such as cucumber and tomato, between meals. Participants were encouraged to drink plenty of water on a fasting day. The feeding day management protocol of the MR group was the same as that of the LIA group.

Anthropometric measurements

Body weight was measured to the nearest 0.1 kg on a balance beam scale, with participants in light clothing and barefoot, at the first visit and subsequently at 2, 4, and 8 weeks. Height was measured to the nearest 0.1 cm with a wall-mounted stadiometer. BMI was calculated as the weight in kilograms divided by the square of the height in meters. Body composition, including percent body fat, fat mass, fat-free mass, visceral fat area, and resting metabolic rate, was evaluated with a bioimpedance analyzer (BIA; InBody770, Seoul, Korea) [25] with participants standing on metal foot plates while holding the handles. Waist circumference (WC) was measured at the level of the umbilicus [26]. Blood pressure was measured on the right upper arm with an electronic sphygmomanometer (Omron, HBP-9021, Omron (Dalian) Co. Ltd., Liaoning, China) and the cuff maintained at heart level after 5 min of rest in a seated position.

Biochemical assays

Three-milliliter fasting venous blood samples were collected from each participant after a 12-h overnight fast, both before and after intervention, to assess the serum blood biochemical indices. The serum was isolated from blood samples after centrifugation. Fasting blood glucose (FBG) levels were measured via the hexokinase method. Serum total cholesterol (TC), low-density lipoprotein cholesterol (LDL-C), high-density lipoprotein cholesterol (HDL-C), and triglyceride (TG) levels were measured via enzyme-coupled colorimetry. The serum levels of liver enzymes, including alanine aminotransferase (ALT), aspartate aminotransferase (AST), and gamma-glutamyl transferase (GGT), were measured via the kinetic rate method. Serum levels of creatinine were measured via the creatine oxidase method, and urea levels were measured via the urease–GLDH coupling rate method. Serum uric acid levels were measured via the uricase‒peroxidase coupling method. The estimated glomerular filtration rate (eGFR) was calculated on the basis of the serum creatinine level. All reagents were purchased, and blood biochemical indices were determined with an AU5800 automatic biochemical analyzer (Beckman Coulter). All the laboratory equipment was calibrated. All the measurements were performed by a technician who was blinded to the assigned intervention.

Statistical analysis

The primary outcome of the study was body weight in kg after 4 and 8 weeks of intervention. The secondary outcomes were significant changes in body weight, BMI, fat mass, fat-free mass, WC, and biochemical indices. Kolmogorov‒Smirnov and Levene’s tests were used to check the data distribution normality and equal variances, respectively. Continuous variables are presented as the means ± standard deviations (SDs). Serum TG levels that were nonnormally distributed are presented as the geometric mean (interquartile range) and were natural log transformed prior to the t test. A paired t test was used to compare postintervention measurements and baseline values. Baseline characteristics of the MR and LIA groups were compared with independent sample t tests for continuous variables or the chi-square test for categorical variables. We used analysis of covariance (ANCOVA) to compare the changes in body weight, body composition, and metabolic indices between the MR group and the LIA group after adjusting for age, sex, and baseline levels.

A generalized estimating equation (GEE) with a heterogeneous autoregressive covariance structure was used to analyze and compare the repeated anthropometric measurements (weight, BMI, waist circumference) and body composition measurements (percent body fat, fat mass, fat-free mass, visceral fat area) between the MR group and the LIA group after 4 weeks and after 8 weeks of intervention, adjusting for sex and age. A first-order interaction intervention*time was included in the model. All the statistical analyses were performed with the Statistics Package for the Social Sciences (SPSS) software, version 26.0 (IBM Corp., Armonk, New York). A two-tailed P value less than 0.05 was considered statistically significant.

Results

Clinical characteristics of the participants

A total of 126 participants were included in the analysis. The mean age was 35.4 ± 9.7 years, and 75.4% were female. As shown in the study flow diagram (Fig. 1), 74 participants in the MR group and 46 participants in the LIA group completed the intervention protocol. The baseline BMI was 26.6 ± 3.7 kg/m². The baseline characteristics of the participants in the two groups are presented in Table 1. There were no significant differences between the groups in terms of age (34.6 ± 10.7 years vs. 36.6 ± 7.8 years, P = 0.232) or sex (men: 27.5% vs. 19.6%, P = 0.442). The resting metabolic rate and fat-free mass were also similar between the groups. However, the anthropometric and clinical data, including body weight, BMI, WC, fat mass, visceral fat area, blood pressure, and triglyceride, aminotransferase and uric acid levels at baseline, were significantly greater in the MR group than in the LIA group.

Table 1.

Baseline characteristics of the study participants

Variables	Meal Replacement (N = 80)	Lifestyle Intervention Alone (N = 46)	P
Age (year)	34.6 ± 10.7	36.6 ± 7.8	0.232
Sex (M/F)	22/58	9/37	0.442
Height (cm)	162.3 ± 7.9	161.8 ± 8.7	0.754
Body weight (kg)	81.1 ± 16.2	69.3 ± 12.4	< 0.001
Body mass index (kg/m2)	30.6 ± 4.0	26.3 ± 3.1	< 0.001
Waist circumference (cm)	100.0 ± 11.4	91.6 ± 7.4	< 0.001
Body fat percent (%)	39.7 ± 5.0	34.7 ± 5.8	< 0.001
Fat mass (kg)	32.3 ± 8.0	23.9 ± 5.3	< 0.001
Fat free mass (kg)	48.8 ± 10.3	45.4 ± 10.3	0.075
Visceral fat area (cm2)	156.5 ± 37.1	114.4 ± 28.9	< 0.001
Resting metabolic rate (kcal/d)	1424.9 ± 222.3	1349.9 ± 222.6	0.075
Systolic blood pressure (mm Hg)	127 ± 20	115 ± 14	< 0.001
Diastolic blood pressure (mm Hg)	77 ± 17	68 ± 10	0.001
Glucose (mmol/L)	5.47 ± 1.56	5.16 ± 0.69	0.218
Total cholesterol (mmol/L)	5.03 ± 1.1	4.73 ± 0.75	0.076
Low density lipoprotein-cholesterol (mmol/L)	2.70 ± 0.85	2.66 ± 0.59	0.785
High density lipoprotein-cholesterol (mmol/L)	1.29 ± 0.27	1.33 ± 0.28	0.462
Triglycerides (mmol/L)*	1.73 (1.07–2.76)	1.25 (0.85–1.84)	0.006
γ-glutamyl transpeptidase (GGT, U/L)*	28 (17–42)	19 (13–27)	0.001
Alanine aminotransferase (ALT, U/L)*	27 (14–46)	17 (11–26)	0.001
Aspartate aminotransferase (AST, U/L)*	24 (17–32)	18 (15–22)	0.001
Creatinine (µmol/L)	60 ± 11	61 ± 13	0.572
Urea (mmol/L)	4.50 ± 1.11	4.53 ± 1.10	0.861
Uric acid (µmol/L)	372 ± 108	333 ± 75	0.032
Estimated glomerular filtration rate (ml/min/1.73m2)	116 ± 12	112 ± 12	0.082

*Non-normally distributed data are presented as geometric mean (interquartile range) and are transformed by the natural logarithm before the t-test

P values were obtained from Student’s t test

Body weight changes over time

The participants’ average body weights were 81.1 ± 16.2 kg, 76.0 ± 15.3 kg, and 74.4 ± 14.8 kg at baseline, week 4, and week 8, respectively, in the MR group and 69.3 ± 12.4, 66.4 ± 11.5 and 64.9 ± 11.3 kg, respectively, in the LIA group (Table 2). The weight loss in both groups was statistically significant (P < 0.001). Furthermore, the magnitude of weight loss from baseline to 4 weeks was significantly greater in the MR group (− 5.1 kg, 95% confidence interval [CI], − 5.6 to − 4.6) than in the LIA group (− 2.8 kg, 95% CI, − 3.4 to − 2.2), even when adjusted for age, sex and baseline weight (P < 0.001) (Table 2; Fig. 2A). Weight reduction was also more pronounced in the MR group (mean: − 6.6, 95% CI: − 7.3 to − 6.0 kg) than in the LIA group (mean: − 4.1, 95% CI: − 4.8 to − 3.4 kg) after 8 weeks of intervention (P = 0.004) (Table 2; Fig. 2A). The individual changes in body weight from baseline to 8 weeks ranged from − 16.1 kg to 0.5 kg in the MR group and from − 12.1 kg to -0.7 kg in the LIA group (Fig. 2C).

Table 2.

Effects of weight management on weight loss, body composition and cardiovascular risk factors after 4-week or 8-week intervention [change from baseline (95% CI)]

Variables	Group	Baseline	After 4 weeks	After 8 weeks	Changes from baseline after 4 weeks					Changes from baseline after 8 weeks
		Mean ± SD	Mean ± SD	Mean ± SD	Differences (95%CI)	% Change	P a	P b (MR vs. LIA)		Differences (95%CI)	% Change	P a	P b (MR vs. LIA)
Anthropometric measures
Weight (kg)	MR	81.1 ± 16.2 ***	76.0 ± 15.3	74.4 ± 14.8	-5.1 (-5.6, -4.6)	−6.3	< 0.001	< 0.001		-6.6 (-7.3, -6.0)	−8.2	< 0.001	0.004
	LIA	69.3 ± 12.4	66.4 ± 11.5	64.9 ± 11.3	-2.8 (-3.4, -2.2)	−4.0	< 0.001			-4.1 (-4.8, -3.4)	−5.8	< 0.001
Body mass index (kg/m2)	MR	30.6 ± 4.0 ***	28.6 ± 3.7	28.1 ± 3.6	-1.9 (-2.1, -1.8)	−6.3	< 0.001	< 0.001		-2.5 (-2.7, -2.2)	−8.1	< 0.001	0.012
	LIA	26.3 ± 3.1	25.3 ± 3.0	24.8 ± 2.9	-1.0 (-1.2, -0.8)	−4.0	< 0.001			-1.5 (-1.8, -1.3)	−5.8	< 0.001
Waist (cm)	MR	100.0 ± 11.4 ***	92.6 ± 11.9	89.5 ± 11.4	-7.4 (-8.3, -6.5)	−7.5	< 0.001	0.115		-9.2 (-10.2, -8.2)	−9.2	< 0.001	0.663
	LIA	91.6 ± 7.4	85.4 ± 7.6	82.8 ± 7.7	-6.2 (-6.9, -5.6)	−6.8	< 0.001			-8.9 (-9.8, -8.1)	−9.8	< 0.001
Systolic blood pressure (mm Hg)	MR	127 ± 20 ***	118 ± 15	118 ± 15	-10 (-13, -6)	−6.6	< 0.001	0.981		-9 (-12, -5)	−5.8	< 0.001	0.591
	LIA	115 ± 14	111 ± 15	113 ± 14	-4 (-7, 0)	−3.1	0.024			-2 (-6, 1)	−1.4	0.215
Diastolic blood pressure (mm Hg)	MR	77 ± 17 **	71 ± 13	70 ± 12	-6 (-9, -4)	−6.7	< 0.001	0.635		-7 (-10, -4)	−7.6	< 0.001	0.378
	LIA	68 ± 10	65 ± 10	66 ± 10	-3 (-5, -1)	−4.5	0.001			-2 (-5, 0)	−2.7	0.064
Body composition
Percent body fat	MR	39.7 ± 5.0 ***	37.1 ± 5.4	35.9 ± 5.3	-2.7 (-3.0, -2.3)	−6.9	< 0.001	0.154		-3.9 (-4.3, -3.4)	−9.9	< 0.001	0.003
	LIA	34.7 ± 5.8	32.4 ± 6.1	31.4 ± 6.3	-2.4 (-2.8, -1.9)	−7.1	< 0.001			-3.5 (-4.1, -2.9)	−10.4	< 0.001
Fat mass (kg)	MR	32.3 ± 8.0 ***	28.3 ± 7.6	26.8 ± 7.2	-4.10 (-4.49, -3.71)	−12.8	< 0.001	0.001		-5.51 (-6.02, -5.00)	−17.2	< 0.001	< 0.001
	LIA	23.9 ± 5.3	21.4 ± 5.4	20.4 ± 5.4	-2.46 (-2.87, -2.06)	−10.8	< 0.001			-3.59 (-4.12, -3.05)	−15.5	< 0.001
Fat free mass (kg)	MR	48.8 ± 10.3	47.9 ± 10.2	47.6 ± 10.0	-1.08 (-1.35, -0.81)	−2.2	< 0.001	0.015		-1.17 (-1.48, -0.87)	−2.4	< 0.001	0.019
	LIA	45.4 ± 10.3	45.0 ± 9.6	44.9 ± 9.7	-0.35 (-0.72, 0.02)	−0.5	0.061			-0.50 (-0.91, -0.10)	−0.9	0.016
Visceral fat area (cm2)	MR	156.5 ± 37.1 ***	134.2 ± 37.7	120.3 ± 37.1	-22.6 (-24.8, -20.5)	−15.0	< 0.001	0.002		-30.9 (-33.9, -28.0)	−20.5	< 0.001	< 0.001
	LIA	114.4 ± 28.9	98.6 ± 30.6	93.6 ± 30.2	-15.8 (-18.3, -13.3)	−14.6	< 0.001			-21.3 (-24.5, -18.1)	−19.4	< 0.001
Resting metabolic rate (kcal/d)	MR	1424.9 ± 222.3	1403.6 ± 220.3	1399.5 ± 216.9	-23.3 (-29.2, -17.4)	−1.6	< 0.001	0.016		-25.4 (-32.0, -18.7)	−1.7	< 0.001	0.019
	LIA	1349.9 ± 222.6	1342.3 ± 206.2	1331.8 ± 204.5	-7.6 (-15.7, 0.5)	−0.4	0.064			-10.8 (-19.5, -2.0)	−0.6	0.017
Laboratory tests
Glucose (mmol/L)	MR	5.47 ± 1.56	4.69 ± 0.74	4.83 ± 0.50	-0.78 (-1.11, -0.44)	−11.8	< 0.001	0.01		-0.64 (-0.93, -0.34)	−8.9	< 0.001	< 0.001
	LIA	5.16 ± 0.69	4.94 ± 0.44	4.95 ± 0.48	-0.22 (-0.44, -0.01)	−3.2	0.041			-0.21 (-0.44, 0.01)	−2.9	0.065
Total cholesterol (mmol/L)	MR	5.03 ± 1.10	4.42 ± 1.02	4.61 ± 0.92	-0.61 (-0.80, -0.41)	−11.1	< 0.001	0.038		-0.42 (-0.59, -0.25)	−7.1	< 0.001	0.001
	LIA	4.73 ± 0.74	4.48 ± 0.83	4.58 ± 0.87	-0.25 (-0.41, -0.08)	−5.1	0.004			-0.15 (-0.32, 0.02)	−3.1	0.083
LDL-C (mmol/L)	MR	2.70 ± 0.85	2.53 ± 0.79	2.64 ± 0.75	-0.17 (-0.32, -0.02)	−2.3	0.030	0.835		-0.06 (-0.19, 0.07)	2.1	0.370	0.011
	LIA	2.66 ± 0.59	2.52 ± 0.62	2.60 ± 0.72	-0.14 (-0.26, -0.02)	−4.7	0.028			-0.06 (-0.19, 0.07)	−2.3	0.374
HDL-C (mmol/L)	MR	1.29 ± 0.27	1.22 ± 0.26	1.30 ± 0.22	-0.07 (-0.13, -0.02)	−3.8	0.014	0.009		0.01 (-0.04, 0.04)	2.1	0.788	< 0.001
	LIA	1.33 ± 0.28	1.35 ± 0.29	1.40 ± 0.26	0.02 (-0.03, 0.07)	2.3	0.392			0.07 (0.03, 0.12)	6.8	0.002
Triglycerides (mmol/L)	MR	2.34 ± 2.68 **	1.21 ± 0.69	1.20 ± 0.62	-1.13 (-1.70, -0.55)	−30.8	< 0.001	0.571		-1.14 (-1.69, -0.59)	−31.3	< 0.001	0.159
	LIA	1.43 ± 0.86	1.04 ± 0.45	1.03 ± 0.44	-0.39 (-0.65, -0.12)	−13.2	0.005			-0.40 (-0.62, -0.17)	−15.4	0.001
γ-glutamyl transpeptidase (U/l)	MR	37 ± 33 **	21 ± 14	21 ± 17	-16 (-21, -11)	−34.2	< 0.001	0.241		-16 (-21, -11)	−33.4	< 0.001	0.026
	LIA	23 ± 15	17 ± 9	17 ± 8	-6 (-8, -3)	−18.1	< 0.001			-6 (-9, -3)	−17.0	< 0.001
ALT (U/l)	MR	38 ± 35 **	28 ± 25	20 ± 14	-10 (-15, -6)	−16.0	< 0.001	0.101		-18 (-24, -12)	−29.6	< 0.001	0.025
	LIA	21 ± 17	17 ± 12	15 ± 8	-4 (-7, -1)	−9.6	0.021			-6 (-10, -2)	−13.7	0.002
AST (U/l)	MR	27 ± 17 **	24 ± 12	19 ± 6	-4 (-6, -1)	−3.7	0.008	0.542		-8 (-11, -5)	−16.7	< 0.001	0.001
	LIA	19 ± 7	20 ± 7	19 ± 6	1 (-0.3, 2)	7.8	0.133			-0.5 (-2, 1)	2.2	0.536
Creatinine (µmol/L)	MR	60 ± 11	63 ± 14	64 ± 12	3 (1, 6)	6.1	0.002	0.542		4 (2, 5)	6.4	< 0.001	0.03
	LIA	61 ± 13	64 ± 15	62 ± 13	2 (0, 4)	4.0	0.012			0.6 (-0.9, 2)	1.4	0.441
Urea (mmol/L)	MR	4.50 ± 1.11	4.21 ± 1.03	4.36 ± 1.00	-0.28 (-0.56, -0.01)	−2.7	0.043	0.019		-0.13 (-0.41, 0.14)	0.7	0.334	< 0.001
	LIA	4.53 ± 1.10	4.62 ± 1.13	4.63 ± 1.12	0.09 (-0.02, 0.38)	4	0.528			0.10 (-0.18, 0.39)	4.6	0.476
Uric acid (µmol/L)	MR	372 ± 108 *	366 ± 105	350 ± 90	-6 (-22, 9)	−0.3	0.405	0.524		-22 (-38, -5)	−3.9	0.01	0.043
	LIA	333 ± 75	328 ± 79	311 ± 76	-5 (-24, 14)	−0.1	0.614			-21 (-34, -9)	−0.6	0.001
Estimated GFR (ml/(min*1.73m2))	MR	116 ± 12	111 ± 14	112 ± 14	-5 (-6, -3)	−4.1	< 0.001	0.036		-4 (-6, -2)	−3.4	< 0.001	0.016
	LIA	112 ± 12	110 ± 14	111 ± 12	-2 (-4, -0.2)	−1.8	0.034			-0.5 (-2, 1)	−0.3	0.516

The values are reported as the means ± sds. ^a. P value was obtained from a paired t test. ^b. P value was obtained from ANCOVA adjusted for age, sex and baseline levels

* P < 0.05, ** P < 0.01, *** P < 0.001, differences between meal replacement (MR) and lifestyle intervention alone (LIA) using two independent samples t tests

Abbreviations: SD, standard deviation; CI, confidence interval; MR, meal replacement; LIA, lifestyle intervention alone; LDL-C, low-density lipoprotein cholesterol; HDL-C, high-density lipoprotein cholesterol; GGT, γ-glutamyl transpeptidase; ALT, alanine transaminase; AST, aspartate transaminase; GFR, glomerular filtration rate.

Fig. 2.

Effects of Meal Replacement and Lifestyle Intervention Alone on Body Weight

Panels A and B show the changes from baseline in body weight for participants who received meal replacement (MR) or lifestyle intervention alone (LIA). The bars indicate the means ± SEs

Panels C and D show the observed data on weight loss after 8 weeks of meal replacement (MR) or lifestyle intervention alone (LIA)

To account for baseline differences, weight loss was expressed as percentage change from initial values. The MR group exhibited significantly greater reductions than the LIA group at both time points (6.3% vs. 4.0% at week 4; 8.2% vs. 5.8% at week 8), with corresponding between-group differences evident in the relative body weight trajectories (Fig. 2B).

Similarly, the reduction in BMI was greater in the MR group than in the LIA group, even when adjusted for age, sex and baseline measurements (P < 0.001). In addition, participants in the two groups had similar reductions in waist circumference (Table 2).

Percent weight loss

As displayed in Fig. 2D, the percentages of participants with weight loss greater than 3%, 5%, and 10% at 8 weeks were significantly greater in the MR group than in the LIA group. Among the participants in the MR group, 8 (10.8%) achieved less than 5% weight loss, 47 (63.5%) achieved 5–10% weight loss, and 19 (25.7%) achieved greater than 10% weight loss. Among the participants in the LIA group, 20 (43.5%) achieved less than 5% weight loss, 20 (43.5%) achieved 5–10% weight loss, and 6 (13.0%) achieved greater than 10% weight loss. In addition, 3 (4.1%) and 8 (17.4%) participants in the MR group and LIA group lost less than 3% of their body weight, respectively.

Body composition

Although the body fat percentage was greater in the MR group than in the LIA group at baseline, at 4 weeks, both groups achieved similar reductions. At 8 weeks, the body fat percentage continued to decrease by 3.9 (95% CI, − 4.3 to − 3.4) from baseline in the MR group and by 3.5 (95% CI, − 4.1 to − 2.9) in the LIA group, with significant differences between the groups after adjusting for sex, age, and baseline body fat percentage (P = 0.003). Both the MR and LIA interventions led to a loss of fat mass and visceral fat area; at 8 weeks, participants in the MR group had greater reductions than did those in the LIA group (P < 0.05) (Table 2).

Participant changes in body composition parameters from baseline during the 8 weeks of intervention are shown in Fig. 3. Specifically, body fat mass decreased by 12.8% vs. 10.8% at week 4 (P < 0.001) and 17.2% vs. 15.5% at week 8 (P < 0.001) in the MR and LIA groups, respectively (Fig. 3A). Corresponding reductions in fat free mass were 2.2% vs. 0.5% (P = 0.002) at week 4 and 2.4% vs. 0.9% (P = 0.003) at week 8 (Fig. 3B). After adjusting for sex, age and baseline levels, the reductions in fat mass and fat-free mass were significantly greater in the MR group than in the LIA group (at week 4: P = 0.001 and P = 0.015, respectively; at week 8: P < 0.001 and P = 0.019, respectively) (Table 2). These data indicate that the MR intervention induced greater reductions in both fat mass and fat-free mass compared with LIA (all between-group P < 0.05). Notably, weight loss in both groups was primarily attributed to fat mass reduction (Fig. 3A), while fat-free mass remained preserved throughout the intervention period (Fig. 3B).

Fig. 3.

Absolute changes in body fat mass (A) and fat-free mass (B) from baseline in patients receiving lifestyle interventions alone (LIA) or lifestyle interventions plus meal replacement (MR)-assisted intermittent fasting for 8 weeks. The bars indicate the means ± SEs

Blood pressure, glucose, lipids, and cardiometabolic risk factors

Both MR and LIA were associated with reduced systolic and diastolic blood pressure over 4 weeks, with no substantial between-group differences (Table 2). However, the reduction from baseline at 8 weeks was significant only in the MR group. The changes in fasting glucose levels, lipid profiles, liver enzymes and renal function indices at 8 weeks from baseline were greater in the MR group than in the LIA group after adjusting for sex, age and baseline levels (Table 2).

Impact of lifestyle intervention with or without the MR intervention on weight loss and body composition changes over time

The results of GEE modeling to estimate the effects of lifestyle intervention with or without the MR intervention on weight loss and changes in body composition over time with adjustments for sex and age are shown in Table 3. GEE modeling revealed a statistically significant effect of time (P < 0.001), indicating significant changes in body weight over time (Table 3). For the parameter estimates, the participants’ body weight decreased 4.0% (P < 0.001) at week 4 and 5.7% (P < 0.001) at week 8 from the baseline body weight (Table 3). However, there was no significant effect of the intervention (β = −0.001, P = 0.428, 95% CI: −0.002 to 0.001). In terms of the parameter estimates of the intervention by time interaction, weight loss in the MR group was 2.1% (P < 0.001) and 3.7% (P < 0.001) greater than that in the LIA group from baseline to week 4 and from baseline to week 8, respectively (Table 3).

Table 3.

Generalized Estimation equation (GEE) analysis of the impact of weight loss and composition changes between meal replacement (MR) and lifestyle intervention alone (LIA) after 4-week and 8-week interventions (n = 120)

Dependent variable	Variable	β	SE	95%CI		Wald χ2	P value
				Lower	Upper
Percentage of weight loss	Intervention
	LIA	Ref
	MR	-0.001	0.0007	-0.002	0.001	0.627	0.428
	Time
	Baseline	Ref
	Week 4	0.040	0.0033	0.034	0.047	146.461	< 0.001
	Week 8	0.057	0.0044	0.048	0.065	168.437	< 0.001
	Intervention × Baseline	Ref
	MR × Week 4	0.021	0.0041	0.013	0.029	26.808	< 0.001
	MR × Week 8	0.037	0.0058	0.026	0.048	40.094	< 0.001
Percentage reduction in body mass index	Intervention
	LIA	Ref
	MR	-0.001	0.0007	-0.002	0.001	0.710	0.399
	Time
	Baseline	Ref
	Week 4	0.040	0.0033	0.034	0.047	146.201	< 0.001
	Week 8	0.056	0.0044	0.047	0.065	160.096	< 0.001
	Intervention × Baseline	Ref
	MR × Week 4	0.021	0.0041	0.013	0.029	25.360	< 0.001
	MR × Week 8	0.037	0.0059	0.025	0.048	38.366	< 0.001
Percentage reduction in waist circumference	Intervention
	LIA	Ref
	MR	0.004	0.0037	-0.003	0.012	1.428	0.232
	Time
	Baseline	Ref
	Week 4	0.071	0.0029	0.065	0.076	584.098	< 0.001
	Week 8	0.101	0.0037	0.093	0.108	736.012	< 0.001
Percentage reduction in percent body fat	Intervention
	LIA	Ref
	MR	0.006	0.0063	-0.007	0.018	0.767	0.381
	Time
	Baseline	Ref
	Week 4	0.070	0.0041	0.062	0.078	294.123	< 0.001
	Week 8	0.113	0.0058	0.102	0.124	376.043	< 0.001
Percentage reduction in fat mass	Intervention
	LIA	Ref
	MR	-0.001	0.0012	-0.004	0.001	1.262	0.261
	Time
	Baseline	Ref
	Week 4	0.110	0.009	0.092	0.127	147.249	< 0.001
	Week 8	0.151	0.0123	0.126	0.175	149.537	< 0.001
	Intervention × Baseline	Ref
	MR × Week 4	0.016	0.0105	-0.004	0.037	2.379	0.123
	MR × Week 8	0.055	0.0162	0.023	0.087	11.525	0.001
Percentage reduction in fat free mass	Intervention
	LIA	Ref
	MR	0	0.0004	-0.001	0.001	0.256	0.613
	Time
	Baseline	Ref
	Week 4	0.005	0.0036	-0.002	0.012	2.213	0.137
	Week 8	0.008	0.0041	0	0.016	4.123	0.042
	Intervention × Baseline	Ref
	MR × Week 4	0.016	0.0046	0.007	0.025	12.186	< 0.001
	MR × Week 8	0.025	0.0092	0.007	0.043	7.334	0.007
Percentage reduction in visceral fat area	Intervention
	LIA	Ref
	MR	0	0.0016	-0.004	0.003	0.092	0.761
	Time
	Baseline	Ref
	Week 4	0.149	0.0124	0.125	0.173	144.994	< 0.001
	Week 8	0.189	0.0153	0.159	0.219	152.501	< 0.001
	Intervention × Baseline	Ref
	MR × Week 4	0	0.0145	-0.029	0.028	0.001	0.975
	MR × Week 8	0.056	0.0214	0.014	0.098	6.835	0.009

Adjusted age and sex.

MR, meal replacement; LIA, lifestyle intervention alone.

Consistent with the weight loss results, reductions in BMI, waist circumference, and body composition, including percent body fat, fat mass, fat-free mass, and visceral fat area, were not significantly different between the MR group and the LIA group (all P > 0.05, Table 3). Fat-free mass lost did not differ significantly from baseline until week 8. In addition, the interaction terms involving time and intervention did not significantly affect waist circumference or body fat percentage.

Discussion

Key findings and clinical implication

This 8-week trial demonstrated that both MR-assisted intermittent fasting and LIA effectively induced significant weight loss, aligning with established dietary approaches for obesity management. Notably, MR-assisted intermittent fasting exhibited superior clinical performance: it induced more rapid reductions in body weight, BMI, and waist circumference, alongside more pronounced improvements in body composition (fat and muscle mass) and cardiometabolic parameters (blood glucose and lipid profiles). Specifically, 89.2% of participants receiving MR-assisted intermittent fasting achieved clinically meaningful weight loss (> 5% baseline body weight), markedly exceeding the 56.5% response rate observed with LIA (absolute difference: 32.7%). These findings position MR-assisted intermittent fasting as a preferred initial strategy for patients requiring accelerated metabolic improvement or high-magnitude weight loss.

However, GEE analysis revealed that longitudinal trajectories of weight loss and secondary outcomes did not differ significantly between groups, indicating that intervention duration, rather than MR supplementation, was the primary outcome driver. Consequently, while MR-assisted intermittent fasting should be prioritized for short-term rapid response, both interventions remain viable for sustained weight management. Treatment selection should therefore be individualized based on patient-specific factors, including metabolic urgency, cost-benefit considerations, and therapeutic preferences.

Meal replacements as effective obesity management tools

The evidence strongly supports the use of MRs in the treatment of obesity. In lifestyle interventions, long-term MR use effectively reduces hemoglobin A1c (HbA1c) levels by 0.7% and initial body weight by 8.6–9.0% in patients with obesity and type 2 diabetes [27, 28]. Meal replacements have been shown to effectively limit caloric intake and promote weight loss and maintenance among overweight and obese individuals. On the one hand, meal replacements facilitate portion control and increase weight loss during lifestyle interventions [12]; on the other hand, the use of meal replacements is well tolerated, with no obvious adverse effects reported [29, 30]. A 90-day trial highlighted the ability of MRs to reduce body fat percentage while preserving metabolic health, with a more favorable trend as the MR intervention increased [31]. Meta-analyses confirmed the superiority of both partial and total meal replacements in weight loss [32]. Our study further revealed the role of MRs in metabolic health improvement, including improvements in glycemic control and lipid profiles, as observed in a 16-week Chinese trial in which 5:2 intermittent fasting was combined with MR use [33].

Furthermore, MRs are fortified with essential micronutrients (e.g., folic acid, vitamin D [34, 35]), addressing deficiencies common in obesity [36] and restrictive diets [37]. MR products are scientifically designed to deliver balanced amounts of essential vitamins, minerals, and dietary fiber while controlling caloric intake. Their standardized formulations mitigate the challenges of self-directed dietary management, offering convenience and precise calorie control [38, 39]. The high-fiber, low-calorie MRs used in our study facilitated rapid weight loss during the initial phase of the intervention, potentially increasing participants’ confidence and motivation [29]. Therefore, MRs serve as an effective tool for obesity management.

Behavioral economics and financial incentives in dietary adherence

Traditional dietary management imposes significant cognitive and temporal demands, requiring expertise in nutrition planning, rigorous self-monitoring, and sustained willpower, thereby creating barriers that frequently undermine adherence among time-constrained populations. MRs circumvent these challenges through choice architecture principles, a core tenet of behavioral economics. By providing preportioned, calorie-restricted formulations with standardized micronutrient profiles, MRs simplify decision-making processes, reduce cognitive load, and minimize opportunities for impulsive dietary deviations. This structural scaffolding aligns with nudge theory, wherein environmental modifications steer individuals toward healthier choices without restricting autonomy [40].

The observed superior weight reduction in MR-assisted intermittent fasting combined with lifestyle interventions may align with the principles of motivated behavior in behavioral economics. This framework posits that health-related decisions are contextually driven rather than purely rational [40]. In this study, self-funded procurement of MRs introduced a financial commitment that likely amplified participants’ psychological investment in weight loss—a phenomenon consistent with the sunk cost effect, wherein individuals persist in behaviors to justify prior expenditures. Additionally, the cost-effectiveness and structured nature of meal replacements may have reinforced adherence to prescribed dietary regimens by reducing decision fatigue and simplifying daily calorie management [41]. These findings underscore the synergistic role of financial nudges and behavioral scaffolding in promoting engagement with health interventions, positioning MRs as both practical and psychologically leveraged tools for obesity management.

Challenge of muscle loss during weight reduction

Weight loss interventions inevitably reduce both fat mass (FM) and fat-free mass (FFM), with FFM accounting for 20–30% of the total weight reduction [42]. This lean tissue depletion correlates positively with the magnitude and velocity of weight loss [43, 44] and is associated with adverse outcomes, including a diminished resting metabolic rate, neuromuscular dysfunction, and heightened risks of fatigue, injury, and weight regain [45]. Consequently, mitigating FFM loss represents a critical challenge in obesity management.

Emerging evidence identifies synergistic strategies for muscle preservation during weight loss, integrating nutrient-optimized meal replacements, exercise‒diet synergy, and dietary timing optimization. Essential amino acid-enriched formulations preserve skeletal muscle mass during caloric restriction, particularly in older adults with obesity [46, 47], whereas whey protein-based meal replacements combined with preoperative exercise increase muscle strength and functional capacity in bariatric candidates [48]. Resistance training paired with high-protein diets (≥ 1.6 g/kg/day) maximizes FM reduction while maintaining FFM, as evidenced by meta-analyses of randomized trials [49]. Intermittent fasting regimens (e.g., the 5:2 method) predominantly reduce FM, whereas concurrent high-protein intake (1.2–1.6 g/kg/day) attenuates FFM loss during rapid weight loss phases [50, 51]. Elevated daily protein intake (≥ 25% of total calories) suppresses muscle catabolism through proteolytic pathway modulation, thereby increasing fat oxidation [52]. These findings collectively support a physiologically grounded framework that integrates protein-fortified meal replacement, resistance exercise, and timed nutrient delivery to achieve the dual objectives of FM reduction and FFM preservation.

Limitations and future directions

Our study has several limitations warranting acknowledgment. First, the nonrandomized assignment to intervention groups is a primary limitation. Specifically, participants in the MR group were required to self-fund the MR product. This likely introduced selection bias, as individuals opting for and paying for the MR intervention were probably more motivated to lose weight compared to those in the LIA group. Although ANCOVA was used to adjust for baseline differences, the potential for selection bias and baseline imbalances stemming from the nonrandomized design remains a concern. Second, despite adjusting for baseline levels of measured variables, residual confounding persists owing to unmeasured factors, such as resting metabolic rates, levels of intrinsic motivation, social support, and appetite regulatory mechanisms, which may affect the trajectories of weight loss. Third, the 8-week intervention duration precludes assessment of long-term effectiveness, safety, and weight maintenance outcomes, necessitating extended follow-up studies. Fourth, the dietary intervention protocol requires refinement, notably through age- and weight-adjusted caloric prescriptions. Moreover, the self-funded MR approach currently restricts accessibility to economically advantaged subgroups. Fifth, the recommendation for daily 10,000-step walking without monitoring introduces ambiguity in distinguishing between dietary and physical activity contributions to outcomes. Sixth, while adverse events were not systematically documented, mild symptoms (including constipation, fatigue, dizziness, hypoglycemia, and flatulence) occurred sporadically in isolated cases during the intervention. Critically, these limited events did not affect adherence within the MR or LIA groups. Finally, despite favorable tolerability and recommendation of MRs, their self-funded nature imposes a high cost barrier, limiting real-world accessibility, scalability, and equitable implementation.

Conclusions

Findings from our self-selection intervention study indicated that both LIA and lifestyle interventions with MRs may lead to weight loss, with faster weight loss in those who participated in lifestyle interventions with MR-assisted intermittent fasting. Notably, the decrease in fat content was accompanied by a decrease in fat-free mass. These findings provide real-world evidence for the effects of both LIA and MR-assisted intermittent fasting on weight loss and body composition in overweight and obese adults. This finding has implications for clinical practice and suggests that nutritional support needs to be provided when patients with overweight and obesity participate in a short-term weight loss intervention. However, randomized controlled trials are needed to better define the role of nutritional supplementation in the prevention of fat-free mass loss.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Author contributions

MZ, YC and YY contributed to the study conception and design. SW, XH, YC and YY collected the samples and clinical data. MZ, GS and THT conducted data analysis and interpreted the results. MZ, GS and YY wrote the first draft of manuscript and interpreted the relevant literature. All authors read and approved the final manuscript.

Funding

The study was supported by the Natural Science Foundation of Zhejiang Province (LGF20H260013), the Medical Science and Technology Project of Zhejiang Province (2025KY448), and the Taizhou Municipal Science and Technology Bureau of Zhejiang Province (24ywb21) to MZ and Taizhou Enze Medical Centre to YC and YY (23EZC02).

Data availability

Data is provided within the manuscript or supplementary information files. Raw data can be shared by contacting the authors by email (13958588986@163.com) when the article is published.

Declarations

Ethics approval and consent to participate

This study involving human participants was reviewed and approved by the Ethics Committee of Taizhou Hospital of Zhejiang Province in China. All procedures were performed in accordance with the guidelines of the institutional ethics committee of the authors and adhered to the tenets of the Declaration of Helsinki. All the data, including demographic, biochemical, and genetic information, were anonymized. No personal information was involved in this study, and informed consent was obtained from all study participants.

Competing interests

The authors declare no competing interests.