Initial weight loss on an 800-kcal diet as a predictor of weight loss success after 8 weeks: the Diogenes study


T Handjieva-Darlenska, S Handjiev, TM Larsen, MA van Baak, S Jebb, A Papadaki, AFH Pfeiffer, JA Martinez, M Kunesova, C Holst, WHM Saris and A Astrup


European Journal of Clinical Nutrition (2010) 64, 994-999

Patient group:

This is a substudy of the Diogenes study, a multicenter study with 8 European centers located in Bulgaria, Denmark, the Netherlands, the UK, Greece, Germany, Spain and the Czech Republic. It included 932 overweight or obese male and female participants, who completed a period of treatment with a Low Calorie Diet (Modifast LCD diet replacement). 


It has previously been shown that future optimal treatment of overweight and obesity must be based on individualized treatment, with specific markers for what candidates are most likely to succeed with diet, physical activity and behavior modification.

Several studies have previously been carried out to identify predictors of weight loss among overweight and obese patients in treatment programs, but the literature has not identified any specific cutoff criteria to predict weight loss. This has been difficult since obese patients differ with respect to the rate of weight loss during a weight loss program. Heavier patients usually lose more weight initially than those with a lower starting weight, because the former experience a higher total calorie deficit. 

Some earlier studies have shown that greater weight loss during the first five weeks of a treatment program is associated with better long-term results.

Study and method:

The aim of the study was to determine how properties and characteristics prior to treatment and early weight loss during the first few weeks on an LCD can predict total weight loss after an 8-week low-calorie diet (LCD).

Participants completed an 8-week LCD period (soups, beverages and puddings), 800 to 880 kcal per day. Four meals were provided daily to replace breakfast, lunch, dinner and a snack. In addition, participants could eat 200 grams of raw vegetables and drink water, coffee, tea, and calorie-free beverages, as well as chew sugar-free chewing gum or consume sugar-free lozenges. Participants attended group sessions with a dietician on six occasions over the eight-week study period. Anthropometric measurements (e.g., waist and hip circumference, bioelectrical impedance) were measured before and after the eight-week period.


Of the 932 participants, 801 completed the eight-week LCD treatment (86%). Total dropout rate was 14% for all centers (drop-out rate in Denmark was 3% - comment from Impolin). Average weight loss after 8 weeks was 10.8 (±0.13) kg, compared with a 2.8 (±0.05) kg average weight loss after week 1 and 5.4 (±0.07) kg after week 3. Men lost more weight than women, with a loss of 12.67 (±0.24) kg, while women lost 9.87 (±0.13) kg, p=0.0001.

The results of the study showed a positive correlation between total weight loss after week 8 and initial weight, height, BMI, hip measurement, sagittal measurement, fat mass, lean body weight and gender.

In the multivariate regression model of predictors in the study, three variables showed significant correlation: initial weight, weight loss after week 1 and weight loss after week 3 in the LCD period. These three factors predicted 68% of the variation of the final weight loss. None of the other reported markers accounted for such a large percentage of the variation in weight loss results. The study shows that early weight loss after week 1 (≥ 2.6 kg) is an optimal predictor that can foretell a weight loss of 10 kg over an 8-week period of LCD treatment.


Identifying the best treatment for overweight and obese individuals and finding early predictors of weight loss success or failure is a challenge. Although researchers have identified predictors of weight loss for over 30 years, these have only been able to explain a small portion. Several studies have had too few participants, for which reason it has been impossible to substantiate statistical significance.

Standardized markers that can easily predict both short-term and long-term weight loss, using objective predictors with narrow confidence intervals, are difficult to identify. Consequently, this study analyzed anthropometric predictors of weight loss in order to identify factors that are important in initial treatment and weight control, as well as the importance of early weight loss during the LCD.

This study has succeeded in identifying some key factors and their function to predict weight loss during an 8-week LCD. A strong and significant predictor that was identified was higher initial weight. Patients with higher initial weight lose weight faster than those with lower initial weight. Given the same energy intake of 800 kcal, it is obvious that a higher initial weight is associated with a larger calorie deficit, which therefore results in faster weight loss. This study showed that higher initial weight correlates positively with both early weight loss and total weight loss after 8 weeks on the LCD.

Other interesting significant predictors included early weight loss after week 1 and weight loss after week 3. These findings are consistent with earlier studies that have shown that large early weight loss is a strong predictor of success for both short-term and long-term weight loss.

In summary, these data demonstrate the importance of considering initial weight, weight loss during the first week of the diet and gender. In addition, weight loss after week 3 is highly significant and can account for much of the variation in total weight loss. Certain individuals may lose weight more slowly than others in the early stages of the LCD, but if they complete treatment, their total weight loss may be the same as for those who lost more weight during the first week.


The study shows that significant predictors of weight loss during an 8-week LCD include: initial weight, weight loss after week 1 and weight loss after week 3. These factors can be used as predictors of how patients respond to the LCD.