|Home | About | Journals | Submit | Contact Us | Français|
Drs. Wyatt and Hill and Ms. Brill: Center for Human Nutrition, University of Colorado Denver, University North Pavilion Building, 4455 East 12th Avenue, 300Z, Denver, CO 80220.
Dr. Makris: 45 Morning Glory Way, Huntingdon Valley, PA 19006. Ms. Rosenbaum: Department of Psychology, Washington University School of Medicine, 212 Stadler Hall, St. Louis, MO 63121.
Drs. Stein, Mohammed, and Miller: Division of Geriatrics and Nutritional Science, Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110.
Dr. Rader: Institute for Translational Medicine and Therapeutics, Room 654, Biomedical Research Building II/III, 421 Curie Boulevard, Philadelphia, PA 19104-6160.
Dr. Zemel: Division of Gastroenterology, Hepatology, and Nutrition, The Children’s Hospital of Philadelphia, 3535 Market Street, Room 1560, Philadelphia, PA 19104-4399.
Dr. Wadden: Center for Weight and Eating Disorders, 3535 Market Street, Suite 3029, Philadelphia, PA 19104.
Dr. Tenhave and Mr. Newcomb: Center for Clinical Epidemiology and Biostatistics, Department of Biostatistics and Epidemiology, 8th Floor, Blockley Hall, University of Pennsylvania School of Medicine, 423 Guardian Drive, Philadelphia, PA 19104-6021.
Dr. Klein: Center for Human Nutrition, Washington University School of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110.
Previous studies comparing low-carbohydrate and low-fat diets have not included a comprehensive behavioral treatment, resulting in suboptimal weight loss.
To evaluate the effects of 2-year treatment with a low-carbohydrate or low-fat diet, each of which was combined with a comprehensive lifestyle modification program.
Randomized parallel-group trial. (ClinicalTrials.gov registration number: NCT00143936)
3 academic medical centers.
307 participants with a mean age of 45.5 years (SD, 9.7 years) and mean body mass index of 36.1 kg/m2 (SD, 3.5 kg/m2).
A low-carbohydrate diet, which consisted of limited carbohydrate intake (20 g/d for 3 months) in the form of low–glycemic index vegetables with unrestricted consumption of fat and protein. After 3 months, participants in the low-carbohydrate diet group increased their carbohydrate intake (5 g/d per wk) until a stable and desired weight was achieved. A low-fat diet consisted of limited energy intake (1200 to 1800 kcal/d; ≤30% calories from fat). Both diets were combined with comprehensive behavioral treatment.
Weight at 2 years was the primary outcome. Secondary measures included weight at 3, 6, and 12 months and serum lipid concentrations, blood pressure, urinary ketones, symptoms, bone mineral density, and body composition throughout the study.
Weight loss was approximately 11 kg (11%) at 1 year and 7 kg (7%) at 2 years. There were no differences in weight, body composition, or bone mineral density between the groups at any time point. During the first 6 months, the low-carbohydrate diet group had greater reductions in diastolic blood pressure, triglyceride levels, and very-low-density lipoprotein cholesterol levels, lesser reductions in low-density lipoprotein cholesterol levels, and more adverse symptoms than did the low-fat diet group. The low-carbohydrate diet group had greater increases in high-density lipoprotein cholesterol levels at all time points, approximating a 23% increase at 2 years.
Intensive behavioral treatment was provided, patients with dyslipidemia and diabetes were excluded, and attrition at 2 years was high.
Successful weight loss can be achieved with either a low-fat or low-carbohydrate diet when coupled with behavioral treatment. A low-carbohydrate diet is associated with favorable changes in cardiovascular disease risk factors at 2 years.
National Institutes of Health.
Data from several randomized trials over the past 6 years have demonstrated that low-carbohydrate diets produced greater short-term (6 months) weight loss than low-fat, calorie-restricted diets (1-5). The longer-term (1 to 2 years) results are mixed. Some studies found greater weight loss with low-carbohydrate diets than with low-fat diets (5, 6), whereas others found no difference (1, 7-9). However, weight loss with either diet was usually minimal (10-12), presumably because of the modest dose of behavioral treatment provided in these studies (1, 6). The only 2-year randomized, controlled trial of a low-carbohydrate diet to date found greater 2-year weight loss with a low-carbohydrate than a low-fat diet (6). The Israel-based study used visual prompts in a cafeteria setting to guide the selection of the main meal (lunch). Whether the results would be similar in different settings and cultures is unknown. In addition, few previous studies have evaluated the effect of low-carbohydrate diets on symptoms or bone, and the assessments have been limited to 6 months (3, 4).
The purpose of our randomized, 3-center trial was to evaluate the effects of long-term (2-year) treatment with either a low-carbohydrate or low-fat, calorie-restricted diet on key clinical end points, namely body weight, cardiovascular risk factors, bone mineral density, and general symptoms. The primary outcome was weight loss at 2 years. All participants received comprehensive behavioral treatment (13, 14) to enhance weight loss associated with both diets. We hypothesized that a low-carbohydrate diet would produce greater weight loss at 2 years than a low-calorie, low-fat diet.
Previous studies comparing low-carbohydrate with low-fat diets focused on short-term outcomes and did not uniformly include interventions to change physical activity and other aspects of lifestyle.
This randomized trial compared outcomes of a behavioral intervention combined with either a low-carbohydrate or low-fat diet and found that after 2 years, participants in both groups lost about 7% of body weight. Greater improvement in high-density lipoprotein cholesterol levels was observed with a low-carbohydrate diet, but other metabolic measures were similar in both groups.
Overweight persons can achieve substantial weight loss at 2 years if they participate in a behavioral intervention combined with a low-fat or a low-carbohydrate diet.
Our study was a randomized, controlled trial conducted over 2 years with outcome assessments at baseline, 3, 6, 12, and 24 months.
Recruitment and data collection were completed at the University of Colorado Denver, Denver, Colorado; Washington University, St. Louis, Missouri; and the University of Pennsylvania, Philadelphia, Pennsylvania.
The primary inclusion criteria were age 18 to 65 years, body mass index of 30 to 40 kg/m2, and body weight less than 136 kg. A total of 307 adults (208 women and 99 men) with a mean age of 45.5 years (SD, 9.7 years) and a mean body mass index of 36.1 kg/m2 (SD, 3.5 kg/m2) participated in this study. Most (74.9%) participants were white; 22.1% were African American, and 3% were of other race or ethnicity. There were no statistically significant differences between the 2 diet groups in any baseline variables (Table 1).
All participants completed a comprehensive medical examination and routine blood tests. We excluded study applicants if they had serious medical illnesses, such as type 2 diabetes; took lipid-lowering medications; were pregnant or lactating; or took medications that affect body weight, including antiobesity agents. Participants with blood pressures of 140/90 mm Hg or more were excluded regardless of whether they were treated. We recruited, enrolled, and followed participants from March 2003 to June 2007. Recruitment methods were consistent across sites and included newspaper advertisements, flyers in the university or hospital setting, physician referral, and self-referral. After a scripted phone screening, eligible patients attended an in-person screening during which the study’s purpose and requirements were fully discussed, eligibility was confirmed, and written informed consent was obtained. The institutional review boards of each of the 3 participating institutions approved the study.
Using a random-number generator, we randomly assigned participants within each site to treatment with either a low-carbohydrate or low-fat, calorie-restricted diet for 2 years (Figure 1).
Approximately half of the participants (n = 153) were assigned to a low-carbohydrate diet, which limited carbohydrate intake but allowed unrestricted consumption of fat and protein. During the first 12 weeks of treatment, participants were instructed to limit carbohydrate intake to 20 g/d in the form of low–glycemic index vegetables. After the first 12 weeks, participants gradually increased carbohydrate intake (5 g/d per week) by consuming more vegetables, a limited amount of fruits, and eventually small quantities of whole grains and dairy products, until a stable and desired weight was achieved. They followed guidelines described in Dr. Atkins’ New Diet Revolution (15) but were not provided with a copy of the book. Participants were instructed to focus on limiting carbohydrate intake and to eat foods rich in fat and protein until they were satisfied. The primary behavioral target was to limit carbohydrate intake.
The remaining 154 participants were assigned to consume a low-fat diet, which consisted of limiting energy intake to 1200 to 1500 kcal/d for women and 1500 to 1800 kcal/d for men, with approximately 55% of calories from carbohydrate, 30% from fat, and 15% from protein. Participants were instructed to limit calorie intake, with a focus on decreasing fat intake. However, limiting overall energy intake (kcal/d) was the primary behavioral target.
All participants received comprehensive, in-person group behavioral treatment (13, 14) weekly for 20 weeks, every other week for 20 weeks, and then every other month for the remainder of the 2-year study period. Each treatment session lasted 75 to 90 minutes. The Appendix (available at www.annals.org) provides details of the treatment. Topics included self-monitoring, stimulus control, and relapse management. All participants were prescribed the same level of physical activity (principally walking), beginning at week 4, with 4 sessions of 20 minutes each and progressing by week 19 to 4 sessions of 50 minutes each. Group sessions reviewed participants’ completion of their eating and activity records, as well as other skill builders. Participants in both groups were instructed to take a daily multivitamin supplement (provided by the study). The lifestyle intervention is described in greater detail in the Appendix.
Body weight was measured at each treatment visit on calibrated scales while participants wore light clothing and no shoes. Height was measured by a stadiometer at baseline. The primary outcome was weight at 2 years.
The following measurements were collected at baseline and at 3, 6, 12 and 24 months.
We obtained blood samples after participants fasted overnight (12 hours). Plasma lipid levels were analyzed (16) in a lipid laboratory that participates continuously in the Centers for Disease Control and Prevention Lipid Standardization Program. We measured plasma high-density lipoprotein (HDL) cholesterol and triglyceride levels enzymatically on a Hitachi autoanalyzer by using Sigma reagents (Sigma Chemical Company, St. Louis, Missouri). Very-low-density lipoprotein (VLDL) cholesterol and low-density lipoprotein (LDL) cholesterol concentrations were directly measured by β-quantification after ultracentrifugation at a density of 1.006 g/mL to separate VLDL.
We assessed blood pressure by using automated instruments (Dinamap, GE Health Care, Milwaukee, Wisconsin) with cuff sizes based on measured arm circumference. After participants were sitting quietly for 5 minutes, 2 readings of blood pressure were obtained, separated by a 1-minute rest period. The average of the 2 readings was used to determine blood pressure.
Dipsticks (Bayer Ketostix 2880, Elkhart, Indiana) were used to measure fasting urinary ketones and were characterized as negative (0 mg/dL) or positive (trace, 5 mg/dL; small, 15 mg/dL; moderate, 40 mg/dL; or large, 80 to 160 mg/dL).
We assessed general symptoms with a symptom checklist used in previous weight-loss studies (17). The checklist contains 26 symptoms rated as none, mild, moderate, or severe. Symptoms were categorized as either absent (none) or present (mild, moderate, or severe) because the symptom data were not normally distributed (most symptoms were listed as none or mild).
We assessed bone mineral density and body composition (percentage of body fat) by using dual-energy x-ray absorptiometry at baseline and at 6, 12, and 24 months. All sites used a Hologic (Bedford, Massachusetts) Delphi or Discovery model bone densitometer. Whole-body, posteroanterior lumbar spine (L1 to L4), and left proximal femur scans were acquired according to manufacturer guidelines for participant positioning. We cross-calibrated scanners by using the same Hologic anthropomorphic spine and whole-body phantom set before data collection. Long-term calibration was monitored at each site with a spine phantom scanned daily and a whole-body phantom scanned 3 times a week. Based on these phantoms, the long-term precision was less than 1% for spine bone mineral density and less than 2% for percentage of body fat. A single technician analyzed all scans centrally by using Hologic software, version 11.2, and one investigator independently reviewed for scan and analysis quality. We excluded poor-quality scans (movement artifacts and improper position) from the analysis (0.7% for spine; 3.9% for hip; and 3.1% for whole body).
All randomly assigned participants, regardless of whether they were actively attending treatment, were contacted by phone, mail, and e-mail to schedule a follow-up assessment.
To detect a 3% (SD, 5%) difference between the groups in the primary outcome—body weight at 24 months—with 90% power and an α value of 0.05, we needed 85 participants per treatment group. To detect a 10% (SD, 20%) difference in LDL cholesterol level and other secondary outcomes, 119 participants per group were required. We aimed to enroll 150 participants per group to account for attrition and to provide power for secondary outcomes.
We used a random-effects linear model that was fitted to all observed data for each variable on each of the 307 participants for the primary analysis. Each random-effects model consisted of a random intercept and slope to adjust for individual participant variability due to within-participant correlations among the observed longitudinal data. These models also contained the following fixed effects: main effects for each follow-up visit, group assignment, interactions between each follow-up visit and group indicator variables, and baseline value as a covariate. We estimated with maximum likelihood by using the PROC MIXED procedure in SAS, version 9 (SAS Institute, Cary, North Carolina). A parallel longitudinal model structure based on main effects for visit, treatment group, and baseline value and visit-treatment interactions was implemented with logistic regression for binary outcomes. We did estimates by using generalized estimating equations under the logistic regression model for correlated longitudinal binary outcomes implemented in the GENMOD procedure in SAS, version 9. Predicted values for each treatment and visit combination at the mean level of the baseline outcome, with corresponding lower and upper confidence bounds, were produced under each model for the figures.
The previously mentioned longitudinal models preclude the use of less robust approaches, such as fixed-imputation methods (for example, last observation carried forward or the analysis of participants with complete data [that is, complete case analyses]). These alternative approaches assume that missing data are unrelated to previously observed outcomes or baseline covariates, including treatment (that is, missing completely at random). The longitudinal models implemented for this study relax this missing-completely-at-random assumption in different ways. The generalized estimating equation–based longitudinal logistic models assume that missing data are unrelated to previously observed outcomes but can be related to the treatment because it is a covariate in the model. (that is, covariate-dependent missing completely at random) (18). The likelihood-based mixed-effects models further relax the covariate-dependent missing-completely-at-random assumption by allowing missing data to be dependent on previously observed outcomes and treatment (that is, missing at random). To assess departures from the missing-at-random assumption under informative withdrawal—that is, the missing weights are informative for which patients chose to withdraw or continue to participate in the study—we present sensitivity analyses. As such, we assume that all participants who withdraw would follow first the maximum and then minimum patient trajectory of weight under the random intercept model.
The α value was set at 5% for weight loss at 24 months and 1% for all other outcomes to account for comparisons at 3, 6, 12, and 24 months (or whatever the pair-wise comparisons are). Adding site to the above models revealed no site effects for weight loss or attrition at 3, 6, 12, or 24 months, so the entire sample (n = 307) was collapsed and analyzed together. Triglyceride values were not normally distributed, so analyses were done on the log-transformed values.
There were no statistically significant differences between the 2 groups in attrition, defined as not undergoing an assessment at a specific time point, independent of the reason. Attrition included participants who withdrew and intermittent missingness at each time point. In the low-fat group, 6%, 12%, 25%, and 32% of participants did not participate in assessments at 3, 6, 12, and 24 months, respectively. Values for the low-carbohydrate participants were 9%, 16%, 26%, and 42%, respectively (Figure 1). Under the sensitivity analysis based on imputing missing outcomes with the highest (13.795) and lowest (−18.355) random-effects slopes (that is, change in weight per month) under the mixed-effects model for weight, our qualitative findings were not sensitive to either imputation approach.
The National Institutes of Health funded this study. The funding source had no role in the design, conduct, or reporting of the study.
Participants in both groups lost approximately 11% of initial weight at 6 and 12 months, with subsequent weight regain to a 7% weight loss at 2 years (Table 2 and Figure 2). We found no statistically significant differences in weight loss at any time point between the low-carbohydrate and low-fat diet groups, although there was a strong trend (P = 0.019) for greater weight loss in the low-carbohydrate group at 3 months.
The percentage of participants who had positive test results for urinary ketones was greater in the low-carbohydrate than in the low-fat group at 3 months (63% vs. 20%; P < 0.001) and 6 months (28% vs. 9%; P < 0.01). We found no statistically significant differences between groups after 6 months. The decrease from 3 to 24 months is consistent with liberalization of carbohydrate intake over time, as part of the study protocol.
Systolic blood pressure decreased with weight loss in both diet groups relative to baseline, but systolic blood pressure did not significantly differ between groups at any time. However, reductions in diastolic pressure were significantly greater (2 to 3 mm Hg) in the low-carbohydrate than in the low-fat group at 3 and 6 months with a strong trend (P = 0.016) at 24 months (Table 2).
The macronutrient content of the 2 diets influenced the effect of weight loss on plasma lipid concentrations. Most of the differences in plasma lipid concentrations between groups were observed during the first 6 months of the diets (Table 2, Figure 3, and Appendix Table, available at www.annals.org). We found a significantly greater decrease in LDL cholesterol levels at 3 and 6 months in the low-fat group than in the low-carbohydrate group, but this difference did not persist at 12 or 24 months. Decreases in triglyceride levels were greater in the low-carbohydrate than in the low-fat group at 3 and 6 months but not at 12 or 24 months. Decreases in VLDL cholesterol levels were significantly greater in the low-carbohydrate than in the low-fat group at 3, 6, and 12 months but not at 24 months. Increases in HDL cholesterol levels were significantly greater in the low-carbohydrate than in the low-fat group at 3, 6, 12 and 24 months. The ratio of total-cholesterol to HDL cholesterol levels decreased significantly in both groups through 24 months but did not significantly differ between groups at any time. There was a trend for greater reductions in the low-carbohydrate group at 6 months (P = 0.035) and 12 months (P = 0.016) (Table 2). Therefore, the only effect on plasma lipid concentrations that persisted at 2 years was the significantly greater increases in HDL cholesterol levels among low-carbohydrate participants.
We found no differences between groups in changes in bone mineral density or body composition over 2 years (Table 2). For both hip and spine bone mineral density, the change from baseline was 1.5% or less at 6, 12, and 24 months, and we found no significant differences between groups. For body composition, both groups experienced similar reductions in lean mass (approximately 5%) and fat mass (11% to 20%), and we found no differences between groups at anytime during the study (Table 2). Finally, the groups did not differ in the percentage of weight lost from fat or lean mass.
A significantly greater percentage of participants who consumed the low-carbohydrate than the low-fat diet reported bad breath, hair loss, constipation, and dry mouth (Table 3). Except for constipation, all of these differences were limited to the first 6 months of treatment. No serious cardiovascular events (for example, stroke, myocardial infarction) were reported. The Appendix includes all serious adverse events (type, time, and attribution to diet).
Our study has 2 main findings. First, neither dietary fat nor carbohydrate intake influenced weight loss when combined with a comprehensive lifestyle intervention. Second, because both diet groups achieved nearly identical weight loss, we were able to determine that a low-carbohydrate diet has greater beneficial long-term effects on HDL cholesterol concentrations than a low-fat diet.
Our participants had similar and clinically significant weight losses with either a low-carbohydrate or low-fat diet at 1 year (11%) and 2 years (7%), demonstrating that either diet can be used to achieve successful long-term weight loss if coupled with behavioral treatment. The weight losses are similar to those obtained with the best available pharmacotherapy for obesity (19, 20). Data from the most previous studies found greater weight loss among low-carbohydrate than low-fat dieters (1-4, 6), presumably because short-term adherence to a low-carbohydrate diet was easier than complying with a low-fat diet. We found a strong trend for greater short-term (3 month) weight loss among the low-carbohydrate participants, but the difference was small (1.3%) and not clinically significant. Our data suggest that the difference in adherence may be overcome by behavioral treatment, although a 2 × 2 analysis (both diets with and without behavioral treatment) would be required to rigorously test this hypothesis. The similar weight losses observed with low-carbohydrate and low-fat diets demonstrate that the comprehensive lifestyle intervention produced the same energy deficit in both groups, despite marked differences in their behavioral targets (carbohydrates vs. calories and fat). This long-term finding in an outpatient setting is consistent with data from short-term metabolic ward studies showing that macronutrient composition did not influence weight loss when energy content was fixed (21-23).
The nearly identical weight loss in the 2 diet groups during our study provided a unique opportunity to assess the relative effects of the macronutrient content of the 2 diets on cardiovascular disease risk factors. The results demonstrate that dietary macronutrient composition had differential effects on plasma lipid concentrations. At 3 and 6 months, LDL cholesterol concentrations increased in the low-carbohydrate group but decreased in the low-fat group, such that the differences between groups were statistically significant. These differences cannot be explained by differences in weight loss and are probably due to the increase in total fat intake in participants who consumed the carbohydrate-restricted diet. Over the long-term, however, plasma LDL cholesterol concentration in the low-carbohydrate diet group was similar to baseline values, and changes in LDL cholesterol concentrations did not statistically differ between groups. Therefore, the short-term increases in plasma LDL cholesterol concentration in the low-carbohydrate diet group are unlikely to be of clinical importance. Moreover, assessment of LDL cholesterol concentration without information on LDL particle size has limitations as an indicator of coronary heart disease risk because small, dense LDL particles are more atherogenic than large LDL particles (24). Data from carefully controlled studies demonstrated that isocaloric replacement of dietary carbohydrate with fat increases plasma LDL cholesterol concentration but shifts LDL particle size from smaller to larger and less atherogenic LDL (25). Nonetheless, weight loss with the low-carbohydrate diet was not associated with the decrease in LDL cholesterol observed in the low-fat diet group and usually observed with weight reduction (26, 27).
The low-carbohydrate diet caused a decrease in plasma triglyceride concentration that was more than double the reduction observed with a low-fat diet at 3, 6, and 12 months. However, at 2 years, plasma triglyceride concentration returned toward baseline in the low-carbohydrate group to values that did not differ from those in the low-fat group. Similarly, the decline in directly measured VLDL cholesterol concentration was also greater in the low-carbohydrate than in the low-fat group at 3, 6, and 12 months. However, as with triglyceride levels, at 2 years we found no significant differences between groups. The close relationship and tracking between fasting plasma triglyceride concentrations (which are primarily contained within VLDL) and VLDL cholesterol concentrations supports a model in which the low-carbohydrate diet decreased hepatic VLDL secretion, enhanced VLDL clearance, or both compared with the low-fat diet during the first year of the study.
The low-carbohydrate diet produced a much greater increase in plasma HDL cholesterol concentration than did the low-fat diet at all assessments during the 2-year study. Plasma HDL cholesterol concentration increased by approximately 20% at 6 months in the low-carbohydrate diet group, which persisted throughout the study and was more than twice the increase observed in the low-fat diet group. The magnitude of the changes observed in the low-carbohydrate group approximates that obtained with the maximal doses of nicotinic acid (niacin), the most effective HDL-raising pharmacologic intervention currently available (28). The fact that the HDL cholesterol levels remained substantially elevated at 24 months, when the plasma triglyceride levels had returned to baseline in the low-carbohydrate group, argues against the conventional explanation that the increase in plasma HDL cholesterol concentration is solely secondary to a reduction in plasma triglyceride levels. The increased HDL cholesterol during a low-carbohydrate diet could result, at least in part, from the increased intake of dietary fat (29). Although weight loss and increased physical activity undoubtedly contributed to the elevation of HDL cholesterol in both groups, the marked difference in HDL cholesterol between the 2 groups, despite similar weight loss, demonstrates that macronutrient composition has independent effects on HDL. The mechanism responsible for the robust and sustained increase in HDL cholesterol levels among low-carbohydrate participants is unknown and will require additional mechanistic studies. The clinical implications of this increase in HDL cholesterol, which is conventionally believed to be beneficial, are uncertain and will probably depend on the mechanism responsible for this effect.
Weight loss caused a decrease in bone mineral density, which was within the range reported in previous weight-loss studies (30). The changes in bone mineral density did not differ between diet groups, suggesting the hypothetical concerns that weight loss induced by a low-carbohydrate diet causes greater bone loss than weight loss induced by a low-fat diet (31) are unfounded. In addition, the decrease in body fat mass and fat-free mass were within the range reported in previous weight-loss studies, and no differences were found between diet groups.
Our study has several important strengths, including a long duration, a large sample that contained both men and women, and the first long-term assessment of bone and adverse symptoms. Our study also has several limitations. First, the comprehensive behavioral therapy program used in this study makes it difficult to extrapolate our results to general weight management in the community. However, the clinically significant weight losses achieved at 24 months underscore the need for providing patients with long-term behavioral support, whether by registered dietitians or other allied health professionals (32, 33). Our protocol was based on an Atkins version of a low-carbohydrate plan, which prescribes an increase in carbohydrate intake over time; thus, the effects of longer than 12 weeks of severe (20 g/d) carbohydrate restriction could not be assessed. Finally, our findings should not be generalized to obese persons who have obesity-related diseases that were excluded from our study population, such as diabetes and hypercholesterolemia.
In conclusion, this 2-year, multicenter study of more than 300 participants revealed that neither dietary fat nor carbohydrate intake influenced weight loss when combined with a comprehensive lifestyle intervention. Both diet groups achieved clinically significant and nearly identical weight loss (11% at 6 months and 7% at 24 months), and persons who received the low-carbohydrate diet had greater 24-month increases in HDL-cholesterol concentrations than persons who received the low-fat diet. We found no differences between the groups for changes in bone or body composition. These long-term data suggest that a low-carbohydrate approach is a viable option for obesity treatment for obese adults.
The authors thank Brooke Bailer, Eva Greenberg, Eileen Ford, Joan Heins, Jennifer Lundgren, Jennifer McCrea, Donna Paulhamus, Gary Skolnick, Emily Smith, Philippe Szapary, Adam Tsai, and Leslie Womble and for their assistance in conducting this study and the study participants for their participation.
Grant Support: By the National Institutes of Health (NIH) grant R01 AT1103 to Temple University; NIH grant UL1RR024134 to University of Pennsylvania; NIH grant UL1 RR000051 to University of Colorado; and NIH grant UL1 RR024992 and DK 56341 to Washington University.
The group treatment sessions were 75 to 90 minutes and were held weekly from weeks 1 to 20, every other week from weeks 21 to 40, and every 8 weeks from weeks 41 to 104. Groups included 8 to 12 participants and only contained persons assigned to the same diet condition (low-carbohydrate or low-fat). Once the group sessions began, no additional members were added, and participants could not attend other group sessions. There was 1 brief (15 minute) individual session at week 30 that focused on assessing progress and goal setting for the future.
During weeks 1 to 20, participants were instructed in traditional behavioral methods of weight control, such as self-monitoring, stimulus control, slowed eating, shaping, and reasonable goal setting. During weeks 21 to 104, there was a focus on skills to maintain weight loss, such as continuing to record food intake regularly, measuring and recording body weight regularly, consuming a low-carbohydrate or a high-carbohydrate diet, identifying high-risk situations, differentiating lapse from relapse, responding effectively to overeating episodes, and learning to reverse small weight gains as they occur. Group sessions varied between the 2 treatment conditions only in the type of diet plan that was prescribed. Sample group leader protocols (week 2) for each treatment condition are included under “week 2” of the section “Low Carbohydrate.”
Groups were conducted by a registered dietitian or psychologist with experience in weight control. Group leaders attended an initial, 2-day, in-person training in Philadelphia, and all group leaders attended biweekly calls throughout the study. The calls were led by a psychologist with extensive experience in behavioral methods of weight control. The calls focused on any clarifications of the protocol and the discussion of nonadherent participants.
Some days will be better than others; it is not realistic to assume that you should eat the same amount every day. The goal is to consume a variety of acceptable foods that you enjoy. The goal is not perfection. Eating is not a moral issue. It is inaccurate an ineffective to make self-evaluations based on eating and exercise behavior.
NOTE: Although ATKINS Ready to Drink Shakes (up to 1 per day), ATKINS Shake Mix (up to 2 scoops per day), and ATKINS ADVANTAGE BARS (up to 1 per day) can be consumed in place of whole foods during Induction, this option should only be initiated when it has been determined that the individual cannot incorporate whole foods into his/her eating plan (like during crunch times). At this point it would be premature to offer this as an option. ATKINS Endulge products cannot be consumed during Induction.
NOTE: Although meal replacement shakes and bars (e.g., Slim-Fast) can be consumed in place of whole foods, this option should only be initiated when it has been determined that the individual cannot incorporate whole foods into his/her eating plan (like during crunch times). At this point it would be premature to offer this as an option.
|48||Low-fat diet||Right and left knee replacement||No|
|43||Low-fat diet||Severe allergic reaction to trimethoprim–sulfamethoxazole||No|
|80||Low-fat diet||Cellulitis from dog bite||No|
|7||Low-carbohydrate diet||Ovarian mass||No|
|20||Low-carbohydrate diet||Renal stones or diverticulitis||Possibly, but not likely; weight loss was 1.09 kg at 20 wk|
|39||Low-fat diet||Umbilical hernia repaired||No|
|56||Low-carbohydrate diet||Torn left meniscus||Possibly due to prescribed exercise program|
Author Contributions: Conception and design: G.D. Foster, H.R. Wyatt, J.O. Hill, A.P. Makris, C. Brill, D.J. Rader, T.A. Wadden, S. Klein.Analysis and interpretation of the data: G.D. Foster, H.R. Wyatt, J.O. Hill, D.J. Rader, B. Zemel, T. Tenhave, C.W. Newcomb, S. Klein.
Drafting of the article: G.D. Foster, H.R. Wyatt, J.O. Hill, B. Zemel, T. Tenhave, S. Klein.
Critical revision of the article for important intellectual content: G.D. Foster, H.R. Wyatt, J.O. Hill, A.P. Makris, D.L. Rosenbaum, R.I. Stein, B.S. Mohammed, B. Miller, D.J. Rader, T.A. Wadden, S. Klein.
Final approval of the article: G.D. Foster, H.R. Wyatt, J.O. Hill, C. Brill, R.I. Stein, B.S. Mohammed, B. Miller, D.J. Rader, T.A. Wadden, T. Tenhave, C.W. Newcomb, S. Klein.
Provision of study materials or patients: G.D. Foster, H.R. Wyatt, J.O. Hill, B.S. Mohammed, B. Miller, T.A. Wadden.
Statistical expertise: T. Tenhave, C.W. Newcomb.
Obtaining of funding: G.D. Foster, H.R. Wyatt, J.O. Hill, T. Tenhave, S. Klein.
Administrative, technical, or logistic support: G.D. Foster, J.O. Hill, D.L. Rosenbaum, C. Brill, R.I. Stein, B.S. Mohammed, B. Miller, B. Zemel, T. Tenhave, S. Klein.
Collection and assembly of data: G.D. Foster, H.R. Wyatt, J.O. Hill, A.P. Makris, D.L. Rosenbaum, C. Brill, R.I. Stein, B.S. Mohammed, B. Miller, D.J. Rader, B. Zemel, T.A. Wadden, S. Klein.
Note: Dr. Foster had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Potential Conflicts of Interest: Disclosures can be viewed at www.acponline.org/authors/icmje/ConflictOfInterestForms.do?msNum=M09-1901.
Reproducible Research Statement: Study protocol: Available from Dr. Foster (ude.elpmet@retsofg). Statistical code: Available from Dr. Tenhave (ude.nnepu@evahnett). Data set: Available from Dr. Foster (ude.elpmet@retsofg), subject to study group approval and National Institutes of Health policy.