Scientific Opinion on the substantiation of health claims related to meal
replacements for weight control (as defined in Directive 96/8/EC on energy
restricted diets for weight loss) and reduction in body weight (ID 1417),
and maintenance of body weight after weight loss (ID 1418) pursuant to
Article 13(1) of Regulation (EC) No 1924/2006[sup]1[/sup]
EFSA Panel on Dietetic Products, Nutrition and Allergies (NDA)2, 3
European Food Safety Authority (EFSA), Parma, Italy
Słowa kluczowe:
Meal replacement
body weight
health claims
weight loss
weight maintenance
3. Naukowe uzasadnienia wpływu na zdrowie człowieka -
3.1. Redukcja masy ciała (ID 1417)
A total of 25 references were submitted for the substantiation of the claimed effect. These include two meta-analyses of randomised controlled trials (RCT) in humans (Heymsfield et al., 2003; Anderson et al., 2004) covering most of the pertinent studies cited in relation to this claim and therefore the assessment will be based on these.
The meta-analysis (and pooled analysis) by Heymsfield et al. (2003) is based on six randomised controlled trials where subjects were instructed to replace one (usually for the weight maintenance phase) or two meals (usually for the weight loss phase) with meal replacement products as part of a conventional energy restricted diet (>800 and <1600 kcal per day) or to follow a conventional energy restricted diet providing the same calorie content with durations ranging between three and 51 months. All the studies included in the meta-analysis used commercial products for the replacement of meals, which usually contain up to 250 kcal/serving and generally comply with the characterisation described in section 1 except for a lower protein content expressed on energy basis (generally 18-25 % energy as protein). A total of almost 500 subjects with initial BMI ranging between 25 and 40 kg/m2 (average 31 kg/m2 ) of which 249 were randomised to receive meal replacements (including a small portion of subjects with type 2 diabetes) where evaluated for their weight loss after 3 and 12 months respectively. The drop out rate was equivalent between groups at 3 months and significantly lower in the meal replacement group at one year. Weight loss was significant at both time points in all studies (one study including 75 subjects reported only 3-month data) for both interventions, with significantly greater weight loss in subjects receiving the meal replacement plan (about 7-8 % of initial body weight) compared to the conventional energy-restricted diet (about 3-7 % of initial body weight). When subjects with type 2 diabetes were omitted from the analyses, the weight loss difference between groups was slightly more pronounced. A random effects meta-analysis estimate indicated a significantly greater weight loss in the meal replacement group than in the conventional diet group at three months and one year (weight loss difference between groups=2.54 kg and 2.43 kg, respectively). The Panel notes that the majority of studies evaluated data on completers only (per protocol analyses), and not on the intention-to-treat population, and that the most plausible explanation for the observed effects is a better compliance with the weight reduction program in the meal replacement intervention group.
The meta-analysis by Anderson et al. (2004) is based on 47 intervention studies conducted in adult obese but otherwise healthy subjects (BMI at least 30 kg/m2 at baseline) assessing the effects of meal replacements (at least two meal replacements per day, four studies), energy restricted diets (providing >1500 kcal per day, six studies), low-energy diets (providing 800-1500 kcal per day, 10 studies), very low-calorie diets (providing up to 800 kcal per day, 19 studies), and soy diets (providing up to 800 kcal per day, eight studies) and reporting weight loss data after 24 weeks of treatment. Participants in the four studies on meal replacements were overweight and obese subjects who received at least two meal replacement servings daily as part of an energy restricted diet. All the studies included in the meta-analysis used commercial products for the replacement of meals, which usually contain up to 250 kcal/serving and generally comply with the characterisation described in section 1 except for a lower protein content expressed on energy basis (generally 18-25 % energy as protein). They involved more than 600 subjects of both sexes (470 women and 133 men) with average
BMI ranging from 28 to 34 kg/m2. Two of the studies had been included in the meta-analysis by Heymsfield et al. (2003) described above (Ditschuneit et al., 1999; Ashley et al., 2001), whereas one was published after the end date for the literature search (Bowerman et al., 2001). In these studies, women lost an average of 9.3 % of their initial weight while men lost 8.6 %. These effects were comparable to intensive lifestyle interventions including conventional energy-restricted diets with an intensity factor (integrating number of visits with a doctor, clinic visits, and class hours) three times lower. Body weight loss with meal replacements was only about 50 % of that found in populations with slightly higher BMI receiving VLCD treatments with an intensity factor four times higher. The Panel notes that the majority of studies evaluated present data on completers only (per protocol analyses), and not on the intention-to-treat population.
Reference was made in the consolidated list to the recommendations of the American Dietetic Association (ADA) on meal replacements for weight loss and weight maintenance (http://www.adaevidencelibrary.com/template.cfm?template=guide_summary&key=625&auth=1). The ADA reports that the evidence for that recommendation is strong and based on eight RCTs, three nonrandomised clinical trials, and good-quality meta-analysis reporting equivalent or greater weight loss in subjects receiving a diet containing 1–3 daily meal replacements (Ahrens et al., 2003; Allison et al, 2003; Ashley et al., 2001; Ditschuneit et al., 1999; Ditschuneit et al., 2001; Flechtner-Mors et al., 2000; Hannum et al, 2004; Heber et al, 1994; Heymsfield et al., 2003; Mattes, 2002; Noakes et al, 2004; Rothacker, 2000; Rothacker et al., 2001; Winick et al, 2002).
Different mechanisms have been proposed by which replacing one or two meals per day by meal replacements could promote weight loss as compared to traditional energy-restricted diets. Meal replacements appear to increase compliance with energy-restricted programs for weight loss (Anderson et al., 2004). This finding could be explained in part because they offer an easy and “ready-to-eat” way of restricting energy intake using calorie-controlled meals, in part because their (protein-rich, low-fat) macronutrient composition may induce sustained satiety to a greater extent. On the other hand, the (protein-rich, low-fat) macronutrient composition of the meal replacements may induce energy inefficiency during negative energy balance by several mechanisms (increasing energy expenditure, sparing lean body mass), which may explain in part their effects on weight loss and maintenance (Krieger et al., 2006; Westerterp-Plantenga et al., 2009).
In weighing the evidence, the Panel took into account that in the meta-analysis by Heymsfield et al. (2003), weight loss achieved with meal replacement products was significantly greater (almost double) than with conventional energy restricted diets prescribed with the same energy content, and that in the meta-analysis by Anderson et al. (2004), both weight loss strategies were equally effective, albeit the intensity of the intervention was three times higher for the conventional energy restricted diets than for meal replacements. The Panel also took into account that all the meal replacement studies included in the meta-analyses used commercial products for the replacement of meals that usually contain up to 250 kcal/serving and generally comply with the characterisation described in section 1, except for a lower protein content expressed on energy basis (generally 18-25 % energy as protein), and that biologically plausible mechanisms have been proposed by which meal replacements could exert the claimed effect, mostly in relation to their controlled energy content and relatively high protein, low fat content.
The Panel concludes that a cause and effect relationship has been established between the consumption of meal replacements in substitution of regular meals in the context of energy restricted diets and reduction in body weight.
3.2. Utrzymanie masy ciała po odchudzaniu (ID 1418)
A total of 14 references were submitted for substantiation of the claimed effect. A number of these references were not considered as pertinent to the claim because of their short follow-up (Anderson et al., 2004; Heber et al., 1994; Winick et al., 2002), were based on pharmacologically treated type 2
diabetic subjects, which are not representative of the target population (Redmon et al., 2003; Yip et al., 2001), did not report on original data (e.g., duplicates, letters and comments), or were limited to abstracts with no sufficient information available for a full scientific evaluation.
The meta-analyses by Heymsfield et al. (2003) described in section 3.1. included most of the single studies cited in relation to the claim on the effects of meal replacements on weight loss. As the greatest weight loss with this strategy is achieved during the first three months of the intervention, during which at least two meals are substituted with meal replacements, the Panel considers that data on further weight loss achieved after this period (from three months to one year in the meta-analysis by Heymsfield et al., 2003), during which usually one meal and one snack are substituted with meal replacements, can be taken into consideration for the substantiation of a claim on body weight maintenance after weight loss.
Three additional references reporting longer follow-ups of some of the studies included in the meta- analysis by Heymsfield et al., (2003) were also considered pertinent to the claimed effect (Ditschuneit et al., 1999; Flechtner-Mors et al., 2000; Rothacker, 2000). These references reported sustained weight loss after a total of 27 months (Ditschuneit et al., 1999) that was maintained after 4 years (-8.4 % of initial body weight) (Flechtner-Mors et al., 2000) with one meal and one snack replacements. Rothacker (2000) also reported that 84 overweight women and 50 overweight males who received a self-managed meal replacement weight control program for 5 years lost 4.2 and 5.8 kg (respectively) in that period, compared to matched controls who gained 6.5 and 6.7 kg, respectively, during the same period of time.
Reference was made in the consolidated list to the recommendations of the American Dietetic Association (ADA) on meal replacements for weight loss and weight maintenance (for link see item 3.1.). The ADA reports that the evidence for that recommendation is strong and based on eight RCTs, three nonrandomised clinical trials, and good-quality meta-analysis reporting equivalent or greater weight loss in subjects receiving a diet containing 1–3 daily meal replacements (Ahrens et al., 2003; Allison et al, 2003; Ashley et al, 2001; Ditschuneit et al, 1999; Ditschuneit et al, 2001; Flechtner- Mors et al, 2000; Hannum et al, 2004; Heber et al, 1994; Heymsfield et al, 2003; Mattes, 2002; Noakes et al, 2004; Rothacker, 2000; Rothacker et al., 2001; Winick et al, 2002).
Different mechanisms have been proposed by which replacing one or two meals per day by meal replacements could promote the maintenance of body weight after weight loss as compared to traditional energy-restricted diets. Meal replacements appear to increase compliance with energy- restricted programs for weight maintenance after weight loss (Anderson et al., 2004). This finding could be explained in part because they offer an easy and “ready-to-eat” way of restricting energy intake using energy-controlled meals, in part because their (protein-rich, low-fat) macronutrient composition may induce sustained satiety to a greater extent. On the other hand, the (protein-rich, low-fat) macronutrient composition of meal replacements may induce energy inefficiency during negative energy balance by several mechanisms (increasing energy expenditure, sparing lean body mass), which may explain in part their effects on weight maintenance after weight loss (Krieger et al., 2006; Westerterp-Plantenga et al., 2009).
In weighing the evidence, the Panel took into account that most of the studies presented reported significantly better weight maintenance after weight loss when meal replacements are used in the maintenance phase, that all the meal replacement studies presented used commercial products for the replacement of meals that usually contain up to 250 kcal/serving and generally comply with the characterisation described in section 1, except for a lower protein content expressed on energy basis (generally 18-25 % energy as protein), and that biologically plausible mechanisms have been proposed by which meal replacements could exert the claimed effect, mostly in relation to their controlled energy content and relatively high protein, low fat content.
The Panel concludes that a cause and effect relationship has been established between the consumption of meal replacements in substitution of regular meals and maintenance of body weight after weight loss.