AWM: Meal Replacements (2006)
Heymsfield SB, van Mierlo CAJ, van der Knaap HCM, Heo M, Frier HI. Weight management using a meal replacement strategy: meta and pooling analysis from six studies. Int J Obes Relat Metab Disord 2003; 27(5): 537-549.PubMed ID: 12704397
Searches in Medline, Embase and the Cochrane Clinical Trials Register from 1960 to January 2001 and from reference lists identified 30 studies for analysis.
Meta-analysis of 6 randomized controlled trials.
Blinding used (if applicable)
Blinding of the patients not possible due to the nature of the interventions. Blinding of the outcome assessor not described in any of the protocols or publications. Description of randomization apparent in all protocols.
Intervention (if applicable)
A partial meal replacement plan was defined as a program that prescribes a low calorie (800-1600 kcal/day) diet whereby 1 or 2 meals are replaced by commercially available, energy-reduced products that are vitamin & mineral fortified, and includes at least 1 meal of regular foods. Overweight and obese subjects were randomized to partial meal replacements or a conventional reduced calorie diet, both containing the same number of calories.
Primary data from the six studies were used for both meta- and pooling analysis. Descriptive statistics of variables used in the present study were computed for each group. The significance of between-group differences was evaluated by Student's t and chi-square tests. To estimate the effect of partial meal replacement on weight loss, the 3-month and 1-year weight losses were calculated as the difference in weight at these 2 time points from baseline values. The effectiveness of the partial meal replacement intervention was measured by the weight loss difference between partial meal replacement and control groups. Similarly, biomarker effects were defined as the changes observed at 3 months and 1 year. Study estimates of weight loss in both groups and the group effect adjusting for gender, baseline age and baseline BMI was obtained with a general linear regression model, the variables adjusted for were included as covariates. The estimates from the models for the individual studies were then meta-analyzed with the studies treated as fixed and random effects. Heterogeneity of estimates was followed using a chi-square test. Assessment of possible publication bias was carried out by calculating "fail-safe N" and Kendall's tau statistic. Missing imputation was not applied for this meta-analysis. To obtain single estimates for each effect of interest, the primary data from the 6 studies were pooled into a single data set and the estimates obtained after adjusting for gender, baseline age, baseline BMI, study effects, and interaction between study and group effects, where the study effects were treated as random. Estimate heterogeneity was assessed by means of testing the significance of interaction between study and group effects. This pooling analysis was also applied to data with missing imputations such as the last-observation-carried forward method and a multiple imputation method. A general linear model was applied to the pooled data for each risk factor to evaluate if the risk factor improvements were associated with their respective baseline values, weight loss, and treatment assignment. The models included random study effects as covariates along with baseline risk factor level, weight loss, and treatment effect as independent variables. Statistical significance was set at the 0.05 level.
Timing of Measurements
Weight and biomarker effects at 3 months and 1 year.
- Body weight
- Biomarkers measured: fasting plasma glucose, insulin, cholesterol, HDL cholesterol, LDL cholesterol, triacylglycerols, and systolic and diastolic blood pressure
- Dietary intervention: partial meal replacements or reduced calorie diets
Initial N: 6 studies met criteria - all studies used the same meal replacement product.
Attrition (final N): 249 partial meal replacement subjects and 238 controls. Female subjects comprised 75% of the population.
Age: Mean age 46.1 years
Ethnicity: Not mentioned.
Other relevant demographics: Mean BMI at baseline: 31.0 kg/m2. 119 subjects had diabetes and 368 were non-diabetic.
Anthropometrics Baseline glucose was the only risk factor that differed significantly between the partial meal replacement and reduced calorie diet groups.
Location: 5 conducted in US and 1 in Germany.
|Study||PMR - RCD (3 months)||P-value||PMR - RCD (1 year)||P-value|
|Individual||ULM (n=100)||5.81 +/- 0.60||<0.001|
|UCLA (n=65)||1.89 +/- 1.11||0.094|
|MAYO (n=30)||1.92 +/- 1.67||0.261||0.98 +/- 1.86||0.603|
|NEV (n=50)||2.75 +/- 0.96||0.006||4.08 +/- 1.92||0.039|
|SDA (n=91)||2.37 +/- 0.71||0.001||5.35 +/- 0.98||<0.001|
|TP (n=67)||0.49 +/- 0.71||0.496||-1.37 +/- 1.77||0.452|
|Meta||Fixed (n=403)||3.01 +/- 0.33||<0.001||3.39 +/- 0.72||<0.001|
|Random (n=403)||2.60 +/- 0.96||0.006||2.43 +/- 1.65||0.142|
|Pooled||Completer (n=403)||2.54 +/- 0.37||<0.001||2.63 +/- 0.88||0.003|
|LOCF (n=403)||2.39 +/- 0.35||
2.86 +/- 0.46
|MI (n=403)||2.23 +/- 0.35||
1.62 +/- 0.98
The dropout rate was not significantly different between the partial meal replacement and reduced calorie diet groups at 3 months: 16% and 19% respectively, P = 0.407. None who dropped out reported program or product-related adverse effects. The dropout rate was significantly less in the partial meal replacement group at 1 year.
Subjects prescribed either the partial meal replacement plan or reduced calorie diet plans lost significant amounts of weight at both the 3-month and 1-year evaluation time points.
All methods of analysis indicated a significantly greater weight loss in subjects receiving the partial meal replacement plan compared to the reduced calorie diet plan.
Depending on the analysis and follow-up duration, the partial meal replacement group lost 7-8% body weight and the reduced calorie diet group lost 3-7% body weight.
A random effects meta-analysis estimate indicated a 2.54 kg (P < 0.01) and 2.43 kg (P = 0.14) greater weight loss in the partial meal replacement group for the 3-month and 1 year periods, respectively.
A pooling analysis of completers showed a greater weight loss in the partial meal replacement group of 2.54 kg (P < 0.01) and 2.63 (P < 0.01) during the same time period.
At 3 months of treatment, 34% of reduced calorie diet subjects and 72% of partial meal replacement subjects lost >5% of body weight (P < 0.001). A similar distribution of weight loss and between-group difference (33% vs 74%, P < 0.001) was observed at the 1-year evaluation point.
Risk factors of disease associated with excess weight improved with weight loss in both groups at the 2 time points. The degree of improvement was also dependent on baseline risk factor levels.
Although now used by consumers worldwide for weight control and incorporated as a therapeutic agent in many clinical trials, the effectiveness of partial meal replacements within experimental treatment programs has never been the subject of a systematic review. Moreover, the lack of available information on the topic necessitated our developing a working definition of partial meal replacement program and associated controlled trials. Despite only a few trials with published and original data that met these criteria, our results show a clear pattern with equivalent or greater weight loss efficacy of a partial meal replacement plan compared to reduced calorie diet treatment in 6 individual studies. The collective analysis of 6 trials showed greater partial meal replacement efficacy with biomarker improvement a function of baseline level and weight loss magnitude. Our findings demonstrate the important potential of well-developed partial meal replacement products and plans as a means of weight control. This investigation also provides guidance for and can serve as a basis for use of partial meal replacements in future weight management programs.
|University/Hospital:||Columbia University College,|
Quality Criteria Checklist: Review Articles
|1.||Will the answer if true, have a direct bearing on the health of patients?||Yes|
|2.||Is the outcome or topic something that patients/clients/population groups would care about?||Yes|
|3.||Is the problem addressed in the review one that is relevant to dietetics practice?||Yes|
|4.||Will the information, if true, require a change in practice?||Yes|
|1.||Was the question for the review clearly focused and appropriate?||Yes|
|2.||Was the search strategy used to locate relevant studies comprehensive? Were the databases searched and the search termsused described?||Yes|
|3.||Were explicit methods used to select studies to include in the review? Were inclusion/exclusion criteria specified andappropriate? Wereselectionmethods unbiased?||Yes|
|4.||Was there an appraisal of the quality and validity of studies included in the review? Were appraisal methodsspecified,appropriate, andreproducible?||Yes|
|5.||Were specific treatments/interventions/exposures described? Were treatments similar enough to be combined?||Yes|
|6.||Was the outcome of interest clearly indicated? Were other potential harms and benefits considered?||Yes|
|7.||Were processes for data abstraction, synthesis, and analysis described? Were they applied consistently acrossstudies and groups? Was thereappropriate use of qualitative and/or quantitative synthesis? Was variation in findings among studies analyzed? Were heterogeneity issued considered? If data from studies were aggregated for meta-analysis, was the procedure described?||Yes|
|8.||Are the results clearly presented in narrative and/or quantitative terms? If summary statistics are used, are levels ofsignificance and/or confidence intervals included?||Yes|
|9.||Are conclusions supported by results with biases and limitations taken into consideration? Are limitations ofthe review identified anddiscussed?||Yes|
|10.||Was bias due to the review's funding or sponsorship unlikely?||???|