Untitled

Editorials represent the opinions of the authors and JAMA and not those of the American Medical Association.
A Diet by Any Other Name Is Still About EnergyLinda Van Horn, PhD, RD The obesity epidemic has fostered increasing interest among
sons. Further considerations regarding the inclusion of exer- many people to seek effective treatment strategies. Relatively cise and behavioral support are important.
few research studies have compared the effectiveness of The authors are careful to point out that while the strengths various diets with different of these analyses include rigorous application of criteria used to determine the eligibility of these studies and ranking of the Related article page 923 and even fewer studies have overall quality in the review, the limitations remain impor- compared named or branded tant. There are more low-carbohydrate diet (Atkins, Zone, and (trade-marked) weight loss programs using comparable, well- South Beach) studies reported than any other study type in- controlled assessment methods and outcome measures. There cluding low fat. The numerous moderate macronutrient stud- is significant interest among the public as well as health care ies, which are based on diet composition, are mixed-branded professionals regarding the efficacy, safety, and long-term fea- diets that involve use of food replacement products, differ- sibility of adhering to these branded diets for the purposes of ent types of behavioral approaches, and encompass a variety guiding decisions regarding the best choices.
of other differences that make it difficult to differentiate and In this issue of JAMA, Johnston and colleagues1 report find- compare with other diets. Also, no sensitivity analyses were ings from a network meta-analysis designed to assess the com- provided to help readers compare the level of adherence to parative effectiveness of branded competing diets that have these different diets and whether the diet composition or the the availability of published data from randomized clinical trials level of adherence (representing successful caloric restric- (RCTs). Network meta-analysis offers a method of estimating tion) were the underlying causes of less weight lost.
the relative effectiveness of available diets in the absence of As well done as these comparative analyses are and as rea- direct comparisons by using indirect comparisons of 2 diets vs sonable as the conclusions by Johnston et al1 appear to be (eg, common comparators. This is helpful for standardizing these that both low-carbohydrate and low-fat diets appear to achieve analyses in a manner that is unlikely to be conducted by the similar weight loss results), several questions and additional commercial diet vendors and also offers results that can help details would facilitate overall interpretation of these find- manage expectations when attempting to choose one dietary ings. For example, even though the results are based solely on approach over another.
weight loss, it would be helpful to know more about the dif- This carefully conducted meta-analysis describes eligibil- ferences in nutrient quality, long-term levels of dietary adher- ity criteria for inclusion, duration and intervention details, and ence, and energy intake associated with these diets involving specific outcomes regarding weight loss at 6 and 12 months of very different diet compositions. Protein intakes of 30% of follow-up.1 The search strategy, study selection process, and kilocalories, or double what the other diets provide, raise ques- data extraction system further lend credibility to this review tions about possible long-term influences on kidney func- and analysis. The choice of comparators including wait-listed tion, calcium losses, and other questions that should be ex- controls and no previously assigned diet or competing di- plored. The adverse effects reported are certainly benign, etary program is sound. The use of the Lifestyle, Exercise, At- ranging from headache to halitosis during the experimental titudes, Relationships, and Nutrition (LEARN) dietary pro- dieting period, but what, if any, further changes occur with ad- gram first described by Brownell2 reflects a reasonable standard ditional months or years of this high protein intake? Are meal for comparison. The approaches to data analyses and de- replacement diets affordable and tolerable long term and are tailed discussion of the confidence estimates and assessment there economic, social isolation, and other limitations that put of publication bias further attest to the strong study design.
these diets out of reach for medium to low income sub- A total of 48 RCTs were identified, met eligibility criteria, groups? Are there compromises in the intake of sodium, solid and were evaluated (N = 7286 individuals). Of these studies, fat, or sugar, or are there other adverse factors that may en- 43 reported weight loss at 6 months and 25 trials reported hance adherence to these meal replacements but limit nutri- weight loss at 12 months. Ultimately, Johnston et al1 con- cluded that both the low-carbohydrate and low-fat diets were The Preventing Overweight Using Novel Dietary Strate- associated with average weight losses of approximately 8 kg gies (POUNDS LOST) trial3,4 was an RCT that tested differ- at 6 months and 6 to 7 kg at 12 months compared with ences in weight loss achieved by 4 different approaches to no diet.1 The LEARN and moderate macronutrient distribu- diet composition: (1) a low-fat, average protein diet (20% fat, tion diets (including Biggest Loser, Jenny Craig, Nutrisystem, 15% protein, and 65% carbohydrate), (2) a low-fat, high- Volumetrics, and Weight Watchers) were associated with loss protein diet (20% fat, 25% protein, and 55% carbohydrate), of body weight of about 2 kg less overall in these compari- (3) a high-fat, average protein diet (40% fat, 15% protein, and JAMA September 3, 2014 Volume 312, Number 9
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Northwestern University User on 09/03/2014
Editorial Opinion 45% carbohydrate), and (4) a high-fat, high protein diet (40% continue to accumulate regarding the benefits of consuming fat, 25% protein, and 35% carbohydrate).4 In addition, satu- high-quality diets associated with reduced risk of cardiovas- rated fat was reduced to 8% or less of kilocalories per day, cular and all-cause mortality, the preferred dietary approach dietary fiber was 20 g/d or greater, and cholesterol of 150 mg to weight loss and maintenance may become increasingly evi- or less per 1000 kcal. Low glycemic index foods were recom- dent as well.6-8 The POUNDS LOST results,3-5 although not spe- mended and a 750 kcal/d deficit from estimated require- cifically representing branded diets, provide relevant find- ments was tailored to each individual. Not only were weight ings based on the generic differences in the macronutrient losses similar across all 4 diets at 2 years, the specific compo- composition that defines them and similarly concluded that nents of weight loss including body fat, abdominal fat, and weight loss is achieved by adherence to any diet that success- hepatic fat were likewise similar with no differences in loss fully reduces calorie intake. Even the well-documented re- of lean body mass.3 ductions in resting energy expenditure9 that are known to ac- In an additional analysis from the POUNDS LOST study,5 company weight loss did not differ by dietary composition at no differences were reported among the 4 diet groups when 2 years, rather apparently it is weight loss that influences this examining other longer-term outcomes (such as food crav- ings or mood changes). Despite some short-term differences Overall, the findings from the study by Johnston et al,1 by the 2-year end point, regardless of their macronutrient com- along with other recent data, underscore the importance of ef- position, weight loss was associated with significant reduc- fective diet and lifestyle interventions that promote behav- tions in cravings for fats, sweets, and starches while cravings ioral changes to support adherence to a calorie-restricted, nu- for fruits and vegetables increased. This suggests that over time trient-dense diet that ultimately accomplishes weight loss.
people not only adapted but actually preferred the taste of the Choosing the best diet suited to an individual's food prefer- nutrient-dense foods, such as fruits and vegetables, reflect- ences may help foster adherence, but beyond weight loss, diet ing enhanced adherence to recommended intake of foods that quality including micronutrient composition may further ben- are typically underconsumed by the population at large. As data efit longevity.
ARTICLE INFORMATION 3. de Souza RJ, Bray GA, Carey VJ, et al. Effects of 4
Ramipril Global End Point Trial (ONTARGET)/ Author Affiliation: Department of Preventive
weight-loss diets differing in fat, protein, and Telmisartan Randomized Assessment Study in ACEI Medicine, Northwestern University Feinberg School carbohydrate on fat mass, lean mass, visceral Intolerant Subjects With Cardiovascular Disease of Medicine, Chicago, Illinois.
adipose tissue, and hepatic fat: results from the (TRANSCEND) Trial Investigators. Relationship POUNDS LOST trial. Am J Clin Nutr. 2012;95(3):614- between healthy diet and risk of cardiovascular Corresponding Author: Linda Van Horn, PhD, RD,
disease among patients on drug therapies for Northwestern University Feinberg School of secondary prevention: a prospective cohort study Medicine, 680 N Lake Shore Dr, Ste 1400, Chicago, 4. Sacks FM, Bray GA, Carey VJ, et al. Comparison
of 31 546 high-risk individuals from 40 countries.
IL 60611 ([email protected]).
of weight-loss diets with different compositions offat, protein, and carbohydrates. N Engl J Med.
Conflict of Interest Disclosures: The author has
8. Oude Griep LM, Wang H, Chan Q.
completed and submitted the ICMJE Form for Empirically-derived dietary patterns, diet quality Disclosure of Potential Conflicts of Interest and 5. Anton SD, Gallagher J, Carey VJ, et al. Diet type
scores, and markers of inflammation and none were reported.
and changes in food cravings following weight loss:findings from the POUNDS LOST Trial. Eat Weight endothelial dysfunction. Curr Nutr Rep. 2013;2(2): 6. Atkins JL, Whincup PH, Morris RW, Lennon LT,
9. de Jonge L, Bray GA, Smith SR, et al. Effect of
1. Johnston BC, Kanters S, Bandayrel K, et al.
Papacosta O, Wannamethee SG. High diet quality is diet composition and weight loss on resting energy Comparison of weight loss among named diet associated with a lower risk of cardiovascular expenditure in the POUNDS LOST study. Obesity programs in overweight and obese adults: disease and all-cause mortality in older men. J Nutr.
a meta-analysis. JAMA. doi:10.1001/jama.2014.10397.
2. Brownell K. The Learn Program for Weight
7. Dehghan M, Mente A, Teo KK, et al; Ongoing
Management. Euless, TX: American Health Telmisartan Alone and in Combination With Publishing; 2004.
JAMA September 3, 2014 Volume 312, Number 9
Copyright 2014 American Medical Association. All rights reserved.
Downloaded From: http://jama.jamanetwork.com/ by a Northwestern University User on 09/03/2014

Source: https://www.womenshealth.northwestern.edu/sites/womenshealth/files/ckfinder/userfiles/files/vanhornjed140074(1).pdf