Swissmilk.ch

Annals of Internal Medicine
Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate
Versus Low-Fat Diet
A Randomized Trial
Gary D. Foster, PhD; Holly R. Wyatt, MD; James O. Hill, PhD; Angela P. Makris, PhD, RD; Diane L. Rosenbaum, BA; Carrie Brill, BS;
Richard I. Stein, PhD; B. Selma Mohammed, MD, PhD; Bernard Miller, MD; Daniel J. Rader, MD; Babette Zemel, PhD;
Thomas A. Wadden, PhD; Thomas Tenhave, PhD; Craig W. Newcomb, MS; and Samuel Klein, MD
Background: Previous studies comparing low-carbohydrate and
toms, bone mineral density, and body composition throughout the low-fat diets have not included a comprehensive behavioral treat- ment, resulting in suboptimal weight loss.
Results: Weight loss was approximately 11 kg (11%) at 1 year and
Objective: To evaluate the effects of 2-year treatment with a
7 kg (7%) at 2 years. There were no differences in weight, body low-carbohydrate or low-fat diet, each of which was combined composition, or bone mineral density between the groups at any with a comprehensive lifestyle modification program.
time point. During the first 6 months, the low-carbohydrate dietgroup had greater reductions in diastolic blood pressure, triglyceride Design: Randomized parallel-group trial. (ClinicalTrials.gov registra-
levels, and very-low-density lipoprotein cholesterol levels, lesser re- ductions in low-density lipoprotein cholesterol levels, and more Setting: 3 academic medical centers.
adverse symptoms than did the low-fat diet group. The low-carbohydrate diet group had greater increases in high-density Patients: 307 participants with a mean age of 45.5 years (SD, 9.7
lipoprotein cholesterol levels at all time points, approximating a years) and mean body mass index of 36.1 kg/m2 (SD, 3.5 kg/m2).
Intervention: A low-carbohydrate diet, which consisted of limited
Limitation: Intensive behavioral treatment was provided, patients
carbohydrate intake (20 g/d for 3 months) in the form of low– with dyslipidemia and diabetes were excluded, and attrition at 2 glycemic index vegetables with unrestricted consumption of fat and protein. After 3 months, participants in the low-carbohydrate dietgroup increased their carbohydrate intake (5 g/d per wk) until a Conclusion: Successful weight loss can be achieved with either a
stable and desired weight was achieved. A low-fat diet consisted of low-fat or low-carbohydrate diet when coupled with behavioral limited energy intake (1200 to 1800 kcal/d; Յ30% calories from treatment. A low-carbohydrate diet is associated with favorable fat). Both diets were combined with comprehensive behavioral changes in cardiovascular disease risk factors at 2 years.
Primary Funding Source: National Institutes of Health.
Measurements: Weight at 2 years was the primary outcome. Sec-
ondary measures included weight at 3, 6, and 12 months and
Ann Intern Med. 2010;153:147-157.
www.annals.org
serum lipid concentrations, blood pressure, urinary ketones, symp- For author affiliations, see end of text.
Data from several randomized trials over the past 6 either a low-carbohydrate or low-fat, calorie-restricted diet
years have demonstrated that low-carbohydrate diets on key clinical end points, namely body weight, cardiovas- produced greater short-term (6 months) weight loss than cular risk factors, bone mineral density, and general symp- low-fat, calorie-restricted diets (1–5). The longer-term (1 toms. The primary outcome was weight loss at 2 years. All to 2 years) results are mixed. Some studies found greater participants received comprehensive behavioral treatment weight loss with low-carbohydrate diets than with low-fat (13, 14) to enhance weight loss associated with both diets.
diets (5, 6), whereas others found no difference (1, 7–9).
We hypothesized that a low-carbohydrate diet would pro- However, weight loss with either diet was usually minimal duce greater weight loss at 2 years than a low-calorie, low- (10 –12), presumably because of the modest dose of behav- ioral treatment provided in these studies (1, 6). The only2-year randomized, controlled trial of a low-carbohydratediet 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 se- Editors’ Notes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 148 lection of the main meal (lunch). Whether the results Summary for Patients. . . . . . . . . . . . . . . . . . . . . . . I-35 would be similar in different settings and cultures is un-known. In addition, few previous studies have evaluated Web-Only
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 2010 American College of Physicians 147
Article Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet sician referral, and self-referral. After a scripted phone screening, eligible patients attended an in-person screen- Previous studies comparing low-carbohydrate with low-fat ing during which the study’s purpose and requirements diets focused on short-term outcomes and did not uni- were fully discussed, eligibility was confirmed, and writ- formly include interventions to change physical activity ten informed consent was obtained. The institutional review boards of each of the 3 participating institutions Contribution
This randomized trial compared outcomes of a behavioral Randomization and Interventions
intervention combined with either a low-carbohydrate or Using a random-number generator, we randomly as- low-fat diet and found that after 2 years, participants in signed participants within each site to treatment with ei- both groups lost about 7% of body weight. Greater im- ther a low-carbohydrate or low-fat, calorie-restricted diet provement in high-density lipoprotein cholesterol levels for 2 years (Figure 1).
was observed with a low-carbohydrate diet, but othermetabolic measures were similar in both groups.
Implication
Table 1. Baseline Participant Characteristics*
Overweight persons can achieve substantial weight loss at2 years if they participate in a behavioral intervention Characteristic
combined with a low-fat or a low-carbohydrate diet.
Diet Group
Carbohydrate
(n ؍ 154)
Diet Group
(n
Race (non-Hispanic or Latino), n (%) Our study was a randomized, controlled trial con- ducted over 2 years with outcome assessments at baseline, Recruitment and data collection were completed at the University of Colorado Denver, Denver, Colorado; Wash- ington University, St. Louis, Missouri; and the University of Pennsylvania, Philadelphia, Pennsylvania.
Participants
Mean systolic blood pressure (SD), mm Hg Mean diastolic blood pressure (SD), mm Hg 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 signifi- cant 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 wereconsistent across sites and included newspaper advertise- BMD ϭ bone mineral density; HDL ϭ high-density lipoprotein; LDL ϭ low- density lipoprotein; VLDL ϭ very-low-density lipoprotein.
ments, flyers in the university or hospital setting, phy- * There were no significant differences between the 2 groups.
148 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Article Figure 1. Study flow diagram.
Completed phone screening (n = 3906)
Excluded by phone
screening (n = 3140)
Passed phone screening (n = 766)
Missed in-person
screening (n = 355)
Assessed for eligibility (n = 411)
Excluded (n = 104)
Did not meet inclusion
criteria: 57
Declined to participate:
Randomly assigned (n = 307)
Other reasons: 15
Low-fat diet (n = 154)
Low-carbohydrate diet (n = 153)
Not assessed at 3 mo (n = 9)
Not assessed at 3 mo (n = 14)
In treatment (n = 3)
In treatment (n = 10)
Discontinued treatment (n = 6)
Discontinued treatment (n = 4)
Time constraints: 1
Time constraints: 1
Life stressors: 1
Dissatisfied with the program: 1
Relocated: 2
Life stressors: 1
Pregnancy: 1
Relocated: 1
No reason: 1
Not assessed at 6 mo (n = 19)
Not assessed at 6 mo (n = 25)
In treatment (n = 9)
In treatment (n = 18)
Discontinued treatment (n = 10)
Discontinued treatment (n = 7)
Time constraints: 4
Time constraints: 2
Dissatisfied with the program: 1
Dissatisfied with the program: 2
Life stressors: 1
Life stressors: 1
Relocated: 2
Relocated: 1
Pregnancy: 1
Pregnancy: 1
No reason: 1
Not assessed at 12 mo (n = 39)
Not assessed at 12 mo (n = 40)
In treatment (n = 23)
In treatment (n = 27)
Discontinued treatment (n = 16)
Discontinued treatment (n = 13)
Time constraints: 6
Time constraints: 3
Dissatisfied with the program: 2
Dissatisfied with the program: 3
Life stressors: 4
Life stressors: 4
Relocated: 2
Relocated: 1
Pregnancy: 1
Pregnancy: 2
No reason: 1
Not assessed at 24 mo (n = 49)
Not assessed at 24 mo (n = 64)
In treatment (n = 13)
In treatment (n = 23)
Discontinued treatment (n = 36)
Discontinued treatment (n = 41)
Time constraints: 7
Time constraints: 6
Dissatisfied with the program: 4
Dissatisfied with the program: 6
Life stressors: 4
Life stressors: 5
Relocated: 3
Pregnancy: 3
Pregnancy: 1
Relocation: 1
No reason: 1
Lost to follow-up: 20
Lost to follow-up: 16
“In treatment” refers to the participants who were still in treatment but did not complete the assessment. “Discontinued treatment” refers to theparticipants who formally withdrew from the study or could not be contacted (that is, lost to follow-up).
www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 149
Article Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Low-Carbohydrate Diet
Serum Lipoproteins
Approximately half of the participants (n ϭ 153) were We obtained blood samples after participants fasted assigned to a low-carbohydrate diet, which limited carbo- overnight (12 hours). Plasma lipid levels were analyzed hydrate intake but allowed unrestricted consumption of fat (16) in a lipid laboratory that participates continuously in and protein. During the first 12 weeks of treatment, par- the Centers for Disease Control and Prevention Lipid ticipants were instructed to limit carbohydrate intake to 20 Standardization Program. We measured plasma high- g/d in the form of low– glycemic index vegetables. After density lipoprotein (HDL) cholesterol and triglyceride lev- the first 12 weeks, participants gradually increased carbo- els enzymatically on a Hitachi autoanalyzer by using Sigma hydrate intake (5 g/d per week) by consuming more vege- reagents (Sigma Chemical Company, St. Louis, Missouri).
tables, a limited amount of fruits, and eventually small Very-low-density lipoprotein (VLDL) cholesterol and low- quantities of whole grains and dairy products, until a stable density lipoprotein (LDL) cholesterol concentrations were and desired weight was achieved. They followed guidelines directly measured by ␤-quantification after ultracentrifuga- described in Dr. Atkins’ New Diet Revolution (15) but were tion at a density of 1.006 g/mL to separate VLDL.
not provided with a copy of the book. Participants wereinstructed to focus on limiting carbohydrate intake and to Blood Pressure
eat foods rich in fat and protein until they were satisfied.
We assessed blood pressure by using automated instru- The primary behavioral target was to limit carbohydrate ments (Dinamap, GE Health Care, Milwaukee, Wiscon- sin) with cuff sizes based on measured arm circumference.
After participants were sitting quietly for 5 minutes, 2readings of blood pressure were obtained, separated by a Low-Fat Diet
1-minute rest period. The average of the 2 readings was The remaining 154 participants were assigned to con- sume a low-fat diet, which consisted of limiting energyintake to 1200 to 1500 kcal/d for women and 1500 to1800 kcal/d for men, with approximately 55% of calories Urine Ketones
from carbohydrate, 30% from fat, and 15% from protein.
Dipsticks (Bayer Ketostix 2880, Elkhart, Indiana) Participants were instructed to limit calorie intake, with a were used to measure fasting urinary ketones and were focus on decreasing fat intake. However, limiting overall characterized as negative (0 mg/dL) or positive (trace, 5 energy intake (kcal/d) was the primary behavioral target.
mg/dL; small, 15 mg/dL; moderate, 40 mg/dL; or large, 80to 160 mg/dL).
Common Instructions
All participants received comprehensive, in-person Symptoms
group behavioral treatment (13, 14) weekly for 20 weeks, We assessed general symptoms with a symptom check- every other week for 20 weeks, and then every other month list used in previous weight-loss studies (17). The checklist for the remainder of the 2-year study period. Each treat- contains 26 symptoms rated as none, mild, moderate, or ment session lasted 75 to 90 minutes. The Appendix
severe. Symptoms were categorized as either absent (none) (available at www.annals.org) provides details of the treat- or present (mild, moderate, or severe) because the symp- ment. Topics included self-monitoring, stimulus control, tom data were not normally distributed (most symptoms 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 Bone Mineral Density and Body Composition
and progressing by week 19 to 4 sessions of 50 minuteseach. Group sessions reviewed participants’ completion of We assessed bone mineral density and body composi- their eating and activity records, as well as other skill build- tion (percentage of body fat) by using dual-energy x-ray ers. Participants in both groups were instructed to take a absorptiometry at baseline and at 6, 12, and 24 months.
daily multivitamin supplement (provided by the study).
All sites used a Hologic (Bedford, Massachusetts) Delphi The lifestyle intervention is described in greater detail in or Discovery model bone densitometer. Whole-body, pos- the Appendix.
teroanterior lumbar spine (L1 to L4), and left proximalfemur scans were acquired according to manufacturer Outcomes and Measurements
guidelines for participant positioning. We cross-calibrated scanners by using the same Hologic anthropomorphic Body weight was measured at each treatment visit on spine and whole-body phantom set before data collection.
calibrated scales while participants wore light clothing and Long-term calibration was monitored at each site with a no shoes. Height was measured by a stadiometer at base- spine phantom scanned daily and a whole-body phantom line. The primary outcome was weight at 2 years.
scanned 3 times a week. Based on these phantoms, the The following measurements were collected at baseline long-term precision was less than 1% for spine bone min- eral density and less than 2% for percentage of body fat. A 150 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Article single technician analyzed all scans centrally by using Ho- ways. The generalized estimating equation– based longitu- logic software, version 11.2, and one investigator indepen- dinal logistic models assume that missing data are unre- dently reviewed for scan and analysis quality. We excluded lated to previously observed outcomes but can be related to poor-quality scans (movement artifacts and improper posi- the treatment because it is a covariate in the model. (that tion) from the analysis (0.7% for spine; 3.9% for hip; and is, covariate-dependent missing completely at random) (18). The likelihood-based mixed-effects models further re-lax the covariate-dependent missing-completely-at-random Follow-up Procedures
assumption by allowing missing data to be dependent on All randomly assigned participants, regardless of previously observed outcomes and treatment (that is, miss- whether they were actively attending treatment, were con- ing at random). To assess departures from the missing-at- tacted by phone, mail, and e-mail to schedule a follow-up random assumption under informative withdrawal—that is, the missing weights are informative for which patients Statistical Analysis
chose to withdraw or continue to participate in the Sample Size
study—we present sensitivity analyses. As such, we assume To detect a 3% (SD, 5%) difference between the that all participants who withdraw would follow first the groups in the primary outcome— body weight at 24 maximum and then minimum patient trajectory of weight months—with 90% power and an ␣ value of 0.05, we needed 85 participants per treatment group. To detect a The ␣ value was set at 5% for weight loss at 24 10% (SD, 20%) difference in LDL cholesterol level and months and 1% for all other outcomes to account for com- other secondary outcomes, 119 participants per group were parisons at 3, 6, 12, and 24 months (or whatever the pair- required. We aimed to enroll 150 participants per group to wise comparisons are). Adding site to the above models account for attrition and to provide power for secondary revealed no site effects for weight loss or attrition at 3, 6, 12, or 24 months, so the entire sample (n ϭ 307) was We used a random-effects linear model that was fitted collapsed and analyzed together. Triglyceride values were to all observed data for each variable on each of the 307 not normally distributed, so analyses were done on the participants for the primary analysis. Each random-effects model consisted of a random intercept and slope to adjustfor individual participant variability due to within-participant correlations among the observed longitudinal Attrition
data. These models also contained the following fixed ef- There were no statistically significant differences be- fects: main effects for each follow-up visit, group assign- tween the 2 groups in attrition, defined as not undergoing ment, interactions between each follow-up visit and group an assessment at a specific time point, independent of the indicator variables, and baseline value as a covariate. We reason. Attrition included participants who withdrew and estimated with maximum likelihood by using the PROC intermittent missingness at each time point. In the low-fat MIXED procedure in SAS, version 9 (SAS Institute, Cary, group, 6%, 12%, 25%, and 32% of participants did not North Carolina). A parallel longitudinal model structure participate in assessments at 3, 6, 12, and 24 months, re- based on main effects for visit, treatment group, and base- spectively. Values for the low-carbohydrate participants line value and visit-treatment interactions was imple- were 9%, 16%, 26%, and 42%, respectively (Figure 1).
mented with logistic regression for binary outcomes. We Under the sensitivity analysis based on imputing missing did estimates by using generalized estimating equations un- outcomes with the highest (13.795) and lowest (Ϫ18.355) der the logistic regression model for correlated longitudinal random-effects slopes (that is, change in weight per binary outcomes implemented in the GENMOD proce- month) under the mixed-effects model for weight, our dure in SAS, version 9. Predicted values for each treatment qualitative findings were not sensitive to either imputation and visit combination at the mean level of the baseline outcome, with corresponding lower and upper confidence Role of the Funding Source
bounds, were produced under each model for the figures.
The National Institutes of Health funded this study.
The previously mentioned longitudinal models pre- The funding source had no role in the design, conduct, or clude the use of less robust approaches, such as fixed- imputation methods (for example, last observation carriedforward or the analysis of participants with complete data[that is, complete case analyses]). These alternative ap- proaches assume that missing data are unrelated to previ- Body Weight
ously observed outcomes or baseline covariates, including Participants in both groups lost approximately 11% treatment (that is, missing completely at random). The of initial weight at 6 and 12 months, with subsequent longitudinal models implemented for this study relax this weight regain to a 7% weight loss at 2 years (Table 2
missing-completely-at-random assumption in different and Figure 2). We found no statistically significant dif-
www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 151
Article Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Table 2. Predicted Mean Changes in Body Weight, Cardiovascular Disease Risk Factors, Bone Mineral Density, and Body
Composition Over 2 Years
Variable
Absolute Change From Baseline (95% CI)
Low-Fat Diet
Low-Carbohydrate Diet
Weight, kg
Triglyceride level, mg/dL†
VLDL cholesterol level, mg/dL†
LDL cholesterol level, mg/dL†
HDL cholesterol level, mg/dL†
Total cholesterol/HDL cholesterol level,
mg/dL†
Systolic blood pressure, mm Hg
Diastolic blood pressure, mm Hg
Hip bone mineral density, g/cm2
152 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Article Table 2—Continued
Variable
Absolute Change From Baseline (95% CI)
Low-Fat Diet
Low-Carbohydrate Diet
Spine bone mineral density, g/cm2
Lean mass, kg
Fat mass, kg
HDL ϭ high-density lipoprotein; LDL ϭ low-density lipoprotein; VLDL ϭ very-low– density lipoprotein.
* P values are for the differences between the 2 groups at each time point.
† To convert values for triglycerides to mmol/L, multiply by 0.01129. To convert values for cholesterol to mmol/L, multiply by 0.02586.
ferences in weight loss at any time point between the blood pressure did not significantly differ between low-carbohydrate and low-fat diet groups, although there groups at any time. However, reductions in diastolic was a strong trend (P ϭ 0.019) for greater weight loss in pressure were significantly greater (2 to 3 mm Hg) in the low-carbohydrate group at 3 months.
the low-carbohydrate than in the low-fat group at 3 and Urinary Ketones
6 months with a strong trend (P ϭ 0.016) at 24 months The percentage of participants who had positive test (Table 2).
results for urinary ketones was greater in the low-carbohydrate than in the low-fat group at 3 months (63% Plasma Lipid Concentrations
vs. 20%; P Ͻ 0.001) and 6 months (28% vs. 9%; P Ͻ The macronutrient content of the 2 diets influenced 0.01). We found no statistically significant differences be- the effect of weight loss on plasma lipid concentrations.
tween groups after 6 months. The decrease from 3 to 24 Most of the differences in plasma lipid concentrations be- months is consistent with liberalization of carbohydrate tween groups were observed during the first 6 months of intake over time, as part of the study protocol.
the diets (Table 2, Figure 3, and Appendix Table, avail-
Blood Pressure
able at www.annals.org). We found a significantly greater Systolic blood pressure decreased with weight loss decrease in LDL cholesterol levels at 3 and 6 months in the in both diet groups relative to baseline, but systolic low-fat group than in the low-carbohydrate group, but thisdifference did not persist at 12 or 24 months. Decreases intriglyceride levels were greater in the low-carbohydrate Figure 2. Predicted absolute mean change in body weight
for participants in the low-fat and low-carbohydrate diet
than in the low-fat group at 3 and 6 months but not at 12 groups, based on a random-effects linear model.
or 24 months. Decreases in VLDL cholesterol levels weresignificantly greater in the low-carbohydrate than in thelow-fat group at 3, 6, and 12 months but not at 24 months. Increases in HDL cholesterol levels were signifi- Low-fat diet group
cantly greater in the low-carbohydrate than in the low-fat Low-carbohydrate diet group
group at 3, 6, 12 and 24 months. The ratio of total- cholesterol to HDL cholesterol levels decreased signifi- cantly in both groups through 24 months but did not significantly differ between groups at any time. There Change in W
was a trend for greater reductions in the low- carbohydrate group at 6 months (P ϭ 0.035) and 12 ϭ 0.016) (Table 2). Therefore, the only
effect on plasma lipid concentrations that persisted at 2 years was the significantly greater increases in HDL cho- lesterol levels among low-carbohydrate participants.
www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 153
Article Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Figure 3. Predicted absolute mean change in serum triglyceride, VLDL cholesterol, LDL cholesterol, and HDL cholesterol
concentrations in the low-fat and low-carbohydrate diet groups, based on a random-effects linear model.
riglyceride Level, –30
Change in T
Change in VLDL Cholesterol Level,
Low-fat diet group
Low-carbohydrate diet group
Change in LDL Cholesterol Level,
Change in HDL Cholesterol Level,
Error bars represent 95% CIs. To convert triglycerides to mmol/L, multiply by 0.0113. To convert HDL, LDL, and VLDL cholesterol to mmol/L,multiply by 0.0259. HDLϭ high-density lipoprotein cholesterol; LDLϭ low-density lipoprotein cholesterol; VLDL ϭ very-low-density lipoprotein cholesterol.
* P Ͻ 0.001.
† P Ͻ 0.01 for between-group differences.
Bone Mineral Density and Body Composition
cardiovascular events (for example, stroke, myocardial in- We found no differences between groups in changes in farction) were reported. The Appendix includes all serious
bone mineral density or body composition over 2 years adverse events (type, time, and attribution to diet).
(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
DISCUSSION
groups. For body composition, both groups experienced Our study has 2 main findings. First, neither dietary similar reductions in lean mass (approximately 5%) and fat fat nor carbohydrate intake influenced weight loss when mass (11% to 20%), and we found no differences between combined with a comprehensive lifestyle intervention. Sec- groups at anytime during the study (Table 2). Finally, the
ond, because both diet groups achieved nearly identical groups did not differ in the percentage of weight lost from weight loss, we were able to determine that a low- carbohydrate diet has greater beneficial long-term effects Symptoms
on HDL cholesterol concentrations than a low-fat diet.
A significantly greater percentage of participants who Our participants had similar and clinically significant consumed the low-carbohydrate than the low-fat diet re- weight losses with either a low-carbohydrate or low-fat diet ported bad breath, hair loss, constipation, and dry mouth at 1 year (11%) and 2 years (7%), demonstrating that (Table 3). Except for constipation, all of these differences
either diet can be used to achieve successful long-term were limited to the first 6 months of treatment. No serious weight loss if coupled with behavioral treatment. The 154 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Article weight losses are similar to those obtained with the best Table 3. Significant Differences in Symptom Reporting
available pharmacotherapy for obesity (19, 20). Data fromthe most previous studies found greater weight loss amonglow-carbohydrate than low-fat dieters (1– 4, 6), presum- Patients (95% CI), %
ably because short-term adherence to a low-carbohydrate Low-Carbohydrate
diet was easier than complying with a low-fat diet. We Diet Group
Diet Group
found a strong trend for greater short-term (3 month) Bad breath
weight loss among the low-carbohydrate participants, but the difference was small (1.3%) and not clinically signifi- cant. Our data suggest that the difference in adherence may be overcome by behavioral treatment, although a 2 ϫ 2 analysis (both diets with and without behavioral treat- Hair loss
ment) would be required to rigorously test this hypothesis.
The similar weight losses observed with low-carbohydrate and low-fat diets demonstrate that the comprehensive life- style intervention produced the same energy deficit in both groups, despite marked differences in their behavioral tar- Constipation
gets (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 Dry mouth
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 haddifferential effects on plasma lipid concentrations. At 3 and * P values are for the difference between the 2 groups for each time point.
6 months, LDL cholesterol concentrations increased in thelow-carbohydrate group but decreased in the low-fatgroup, such that the differences between groups were sta- reduction observed with a low-fat diet at 3, 6, and 12 tistically significant. These differences cannot be explained months. However, at 2 years, plasma triglyceride concen- by differences in weight loss and are probably due to the tration returned toward baseline in the low-carbohydrate increase in total fat intake in participants who consumed group to values that did not differ from those in the low- the carbohydrate-restricted diet. Over the long-term, how- fat group. Similarly, the decline in directly measured ever, plasma LDL cholesterol concentration in the low- VLDL cholesterol concentration was also greater in the carbohydrate diet group was similar to baseline values, and low-carbohydrate than in the low-fat group at 3, 6, and 12 changes in LDL cholesterol concentrations did not statis- months. However, as with triglyceride levels, at 2 years we tically differ between groups. Therefore, the short-term in- found no significant differences between groups. The close creases in plasma LDL cholesterol concentration in the relationship and tracking between fasting plasma triglycer- low-carbohydrate diet group are unlikely to be of clinical ide concentrations (which are primarily contained within importance. Moreover, assessment of LDL cholesterol con- VLDL) and VLDL cholesterol concentrations supports a centration without information on LDL particle size has model in which the low-carbohydrate diet decreased he- limitations as an indicator of coronary heart disease risk patic VLDL secretion, enhanced VLDL clearance, or both because small, dense LDL particles are more atherogenic compared with the low-fat diet during the first year of the than large LDL particles (24). Data from carefully con- trolled studies demonstrated that isocaloric replacement of The low-carbohydrate diet produced a much greater dietary carbohydrate with fat increases plasma LDL choles- increase in plasma HDL cholesterol concentration than did terol concentration but shifts LDL particle size from the low-fat diet at all assessments during the 2-year study.
smaller to larger and less atherogenic LDL (25). Nonethe- Plasma HDL cholesterol concentration increased by ap- less, weight loss with the low-carbohydrate diet was not proximately 20% at 6 months in the low-carbohydrate diet associated with the decrease in LDL cholesterol observed in group, which persisted throughout the study and was more the low-fat diet group and usually observed with weight than twice the increase observed in the low-fat diet group.
The magnitude of the changes observed in the low- The low-carbohydrate diet caused a decrease in plasma carbohydrate group approximates that obtained with the triglyceride concentration that was more than double the maximal doses of nicotinic acid (niacin), the most effective www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 155
Article Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet HDL-raising pharmacologic intervention currently avail- weight loss (11% at 6 months and 7% at 24 months), and able (28). The fact that the HDL cholesterol levels re- persons who received the low-carbohydrate diet had greater mained substantially elevated at 24 months, when the 24-month increases in HDL-cholesterol concentrations plasma triglyceride levels had returned to baseline in the than persons who received the low-fat diet. We found no low-carbohydrate group, argues against the conventional differences between the groups for changes in bone or body explanation that the increase in plasma HDL cholesterol composition. These long-term data suggest that a low- concentration is solely secondary to a reduction in plasma carbohydrate approach is a viable option for obesity treat- triglyceride levels. The increased HDL cholesterol during a low-carbohydrate diet could result, at least in part, fromthe increased intake of dietary fat (29). Although weight From Temple University, University of Pennsylvania School of Medi- loss and increased physical activity undoubtedly contrib- cine, and Children’s Hospital of Philadelphia, Philadelphia, Pennsylva- uted to the elevation of HDL cholesterol in both groups, nia; University of Colorado Denver, Denver, Colorado; and WashingtonUniversity School of Medicine and University of Missouri, St. Louis, the marked difference in HDL cholesterol between the 2 groups, despite similar weight loss, demonstrates that ma-cronutrient composition has independent effects on HDL.
Note: Dr. Foster had full access to all of the data in the study and takes
The mechanism responsible for the robust and sustained in- responsibility for the integrity of the data and the accuracy of the data crease in HDL cholesterol levels among low-carbohydrate par- ticipants is unknown and will require additional mechanisticstudies. The clinical implications of this increase in HDL cho- Acknowledgment: The authors thank Brooke Bailer, Eva Greenberg,
lesterol, which is conventionally believed to be beneficial, are Eileen Ford, Joan Heins, Jennifer Lundgren, Jennifer McCrea, Donna uncertain and will probably depend on the mechanism re- 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.
Weight loss caused a decrease in bone mineral density, which was within the range reported in previous weight- Grant Support: By Washington University (grant UL1 RR024992);
loss studies (30). The changes in bone mineral density did Temple University (grant R01 AT1103); University of Pennsylvania not differ between diet groups, suggesting the hypothetical (grant UL1RR024134); University of Colorado (grant UL1 RR000051); concerns that weight loss induced by a low-carbohydrate and the National Center for Research Resources, a component of the diet causes greater bone loss than weight loss induced by a National Institutes of Health (DK 56341), to Washington University low-fat diet (31) are unfounded. In addition, the decrease in body fat mass and fat-free mass were within the rangereported in previous weight-loss studies, and no differences Potential Conflicts of Interest: Disclosures can be viewed at www.acponline
.org/authors/icmje/ConflictOfInterestForms.do?msNumϭM09-1901.
Our study has several important strengths, including a Reproducible Research Statement: Study protocol: Available from Dr.
long duration, a large sample that contained both men and Foster (e-mail, [email protected]). Statistical code: Available from Dr. Ten- women, and the first long-term assessment of bone and have (e-mail, [email protected]). Data set: Available from Dr. Foster (e- adverse symptoms. Our study also has several limitations.
mail, [email protected]), subject to study group approval and National First, the comprehensive behavioral therapy program used in this study makes it difficult to extrapolate our results togeneral weight management in the community. However, Requests for Single Reprints: Gary D. Foster, PhD, Center for Obesity
the clinically significant weight losses achieved at 24 Research and Education, Temple University, 3223 North Broad Street, months underscore the need for providing patients with Suite 175, Philadelphia, PA 19140; e-mail, [email protected].
long-term behavioral support, whether by registered dieti-tians or other allied health professionals (32, 33). Our pro- Current author addresses and author contributions are available at www.annals.org.
tocol was based on an Atkins version of a low-carbohydrateplan, which prescribes an increase in carbohydrate intakeover time; thus, the effects of longer than 12 weeks of References
severe (20 g/d) carbohydrate restriction could not be as- 1. Foster GD, Wyatt HR, Hill JO, McGuckin BG, Brill C, Mohammed BS,
sessed. Finally, our findings should not be generalized to et al. A randomized trial of a low-carbohydrate diet for obesity. N Engl J Med.
obese persons who have obesity-related diseases that were excluded from our study population, such as diabetes and 2. Samaha FF, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory J, et al.
A low-carbohydrate as compared with a low-fat diet in severe obesity. N Engl J
In conclusion, this 2-year, multicenter study of more 3. Brehm BJ, Seeley RJ, Daniels SR, D’Alessio DA. A randomized trial com-
than 300 participants revealed that neither dietary fat nor paring a very low carbohydrate diet and a calorie-restricted low fat diet on body carbohydrate intake influenced weight loss when combined weight and cardiovascular risk factors in healthy women. J Clin EndocrinolMetab. 2003;88:1617-23. [PMID: 12679447] with a comprehensive lifestyle intervention. Both diet 4. Yancy WS Jr, Olsen MK, Guyton JR, Bakst RP, Westman EC. A low-
groups achieved clinically significant and nearly identical carbohydrate, ketogenic diet versus a low-fat diet to treat obesity and hyperlipid- 156 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
Weight and Metabolic Outcomes After 2 Years on a Low-Carbohydrate Versus Low-Fat Diet Article emia: a randomized, controlled trial. Ann Intern Med. 2004;140:769-77.
STORM Study Group. Sibutramine Trial of Obesity Reduction and Mainte- nance. Lancet. 2000;356:2119-25. [PMID: 11191537] 5. Gardner CD, Kiazand A, Alhassan S, Kim S, Stafford RS, Balise RR, et al.
20. Sjo¨stro¨m L, Rissanen A, Andersen T, Boldrin M, Golay A, Koppeschaar
Comparison of the Atkins, Zone, Ornish, and LEARN diets for change in weight HP, et al. Randomised placebo-controlled trial of orlistat for weight loss and
and related risk factors among overweight premenopausal women: the A TO Z prevention of weight regain in obese patients. European Multicentre Orlistat Weight Loss Study: a randomized trial. JAMA. 2007;297:969-77. [PMID: Study Group. Lancet. 1998;352:167-72. [PMID: 9683204] 21. Yang MU, Van Itallie TB. Composition of weight lost during short-term
6. Shai I, Schwarzfuchs D, Henkin Y, Shahar DR, Witkow S, Greenberg I,
weight reduction. Metabolic responses of obese subjects to starvation and low- et al; Dietary Intervention Randomized Controlled Trial (DIRECT) Group.
calorie ketogenic and nonketogenic diets. J Clin Invest. 1976;58:722-30. [PMID: Weight loss with a low-carbohydrate, Mediterranean, or low-fat diet. N Engl J 22. Golay A, Allaz AF, Morel Y, de Tonnac N, Tankova S, Reaven G. Similar
7. Stern L, Iqbal N, Seshadri P, Chicano KL, Daily DA, McGrory J, et al. The
weight loss with low- or high-carbohydrate diets. Am J Clin Nutr. 1996;63: effects of low-carbohydrate versus conventional weight loss diets in severely obese adults: one-year follow-up of a randomized trial. Ann Intern Med. 2004;140: 23. Boden G, Sargrad K, Homko C, Mozzoli M, Stein TP. Effect of a low-
carbohydrate diet on appetite, blood glucose levels, and insulin resistance in obese 8. Dansinger ML, Gleason JA, Griffith JL, Selker HP, Schaefer EJ. Comparison
patients with type 2 diabetes. Ann Intern Med. 2005;142:403-11. [PMID: of the Atkins, Ornish, Weight Watchers, and Zone diets for weight loss and heart disease risk reduction: a randomized trial. JAMA. 2005;293:43-53. [PMID: 24. Berneis KK, Krauss RM. Metabolic origins and clinical significance of LDL
heterogeneity. J Lipid Res. 2002;43:1363-79. [PMID: 12235168] 9. Yancy WS Jr, Westman EC, McDuffie JR, Grambow SC, Jeffreys AS,
25. Krauss RM, Blanche PJ, Rawlings RS, Fernstrom HS, Williams PT. Sep-
Bolton J, et al. A randomized trial of a low-carbohydrate diet vs orlistat plus a
arate effects of reduced carbohydrate intake and weight loss on atherogenic dys- low-fat diet for weight loss. Arch Intern Med. 2010;170:136-45. [PMID: lipidemia. Am J Clin Nutr. 2006;83:1025-31. [PMID: 16685042] 26. Dattilo DJ, Ige JT, Nwana EJ. Intraoral lipoma of the tongue and subman-
10. Malik VS, Hu FB. Popular weight-loss diets: from evidence to practice. Nat
dibular space: report of a case. J Oral Maxillofac Surg. 1996;54:915-7. [PMID: Clin Pract Cardiovasc Med. 2007;4:34-41. [PMID: 17180148] 11. Makris AP, Foster GD, Astrup A. Diet composition and weight loss. In:
27. Clinical Guidelines on the Identification, Evaluation, and Treatment of Bray GA, Bouchard C, eds. Handbook of Obesity. Clinical Applications. 3rd ed.
Overweight and Obesity in Adults—The Evidence Report. National Institutes of New York: Informa Healthcare USA; 2008:269-90.
Health. Obes Res. 1998;6 Suppl 2:51S-209S. [PMID: 9813653] 12. Nordmann AJ, Nordmann A, Briel M, Keller U, Yancy WS Jr, Brehm BJ,
28. Szapary PO, Rader DJ. Pharmacological management of high triglycerides
et al. Effects of low-carbohydrate vs low-fat diets on weight loss and cardiovas-
and low high-density lipoprotein cholesterol. Curr Opin Pharmacol. 2001;1:113- cular risk factors: a meta-analysis of randomized controlled trials. Arch Intern 29. Kris-Etherton PM, Etherton TD, Yu S. Efficacy of multiple dietary therapies
13. Foster GD, Makris AP, Bailer BA. Behavioral treatment of obesity. Am J
in reducing cardiovascular disease risk factors [Editorial]. Am J Clin Nutr. 1997; Clin Nutr. 2005;82:230S-235S. [PMID: 16002827] 14. Wadden TA, Butryn ML, Wilson C. Lifestyle modification for the manage-
30. Villareal DT, Fontana L, Weiss EP, Racette SB, Steger-May K, Schecht-
ment of obesity. Gastroenterology. 2007;132:2226-38. [PMID: 17498514] man KB, et al. Bone mineral density response to caloric restriction-induced
15. Atkins RC. Dr. Atkins’ New Diet Revolution. New York: Avon Books;
weight loss or exercise-induced weight loss: a randomized controlled trial. Arch Intern Med. 2006;166:2502-10. [PMID: 17159017] 16. Allain CC, Poon LS, Chan CS, Richmond W, Fu PC. Enzymatic determi-
31. Adam-Perrot A, Clifton P, Brouns F. Low-carbohydrate diets: nutritional
nation of total serum cholesterol. Clin Chem. 1974;20:470-5. [PMID: 4818200] and physiological aspects. Obes Rev. 2006;7:49-58. [PMID: 16436102] 17. Wadden TA, Stunkard AJ, Day SC, Gould RA, Rubin CJ. Less food, less
32. Wing RR, Tate DF, Gorin AA, Raynor HA, Fava JL. A self-regulation
hunger: reports of appetite and symptoms in a controlled study of a protein- program for maintenance of weight loss. N Engl J Med. 2006;355:1563-71.
sparing modified fast. Int J Obes. 1987;11:239-49. [PMID: 3667060] 18. Little RJ. Modeling the drop-out mechanism in repeated-measures studies. J
33. Svetkey LP, Stevens VJ, Brantley PJ, Appel LJ, Hollis JF, Loria CM, et al;
Weight Loss Maintenance Collaborative Research Group. Comparison of strat-
19. James WP, Astrup A, Finer N, Hilsted J, Kopelman P, Ro¨ssner S, et al.
egies for sustaining weight loss: the weight loss maintenance randomized con- Effect of sibutramine on weight maintenance after weight loss: a randomised trial.
trolled trial. JAMA. 2008;299:1139-48. [PMID: 18334689] www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 157
Annals of Internal Medicine
Current Author Addresses: Dr. Foster: Center for Obesity Research
There was 1 brief (15 minute) individual session at week 30 that and Education, Temple University, 3223 North Broad Street, Suite 175, focused on assessing progress and goal setting for the future.
During weeks 1 to 20, participants were instructed in tradi- Drs. Wyatt and Hill and Ms. Brill: Center for Human Nutrition, Uni- tional behavioral methods of weight control, such as self- versity of Colorado Denver, University North Pavilion Building, 4455 monitoring, stimulus control, slowed eating, shaping, and rea- East 12th Avenue, 300Z, Denver, CO 80220.
Dr. Makris: 45 Morning Glory Way, Huntingdon Valley, PA 19006.
sonable goal setting. During weeks 21 to 104, there was a focus Ms. Rosenbaum: Department of Psychology, Washington University on skills to maintain weight loss, such as continuing to record School of Medicine, 212 Stadler Hall, St. Louis, MO 63121.
food intake regularly, measuring and recording body weight reg- Drs. Stein, Mohammed, and Miller: Division of Geriatrics and Nutri- ularly, consuming a low-carbohydrate or a high-carbohydrate tional Science, Washington University School of Medicine, 660 South diet, identifying high-risk situations, differentiating lapse from relapse, responding effectively to overeating episodes, and learn- Dr. Rader: Institute for Translational Medicine and Therapeutics, Room ing to reverse small weight gains as they occur. Group sessions 654, Biomedical Research Building II/III, 421 Curie Boulevard, Phila-delphia, PA 19104-6160.
varied between the 2 treatment conditions only in the type of Dr. Zemel: Division of Gastroenterology, Hepatology, and Nutrition, diet plan that was prescribed. Sample group leader protocols The Children’s Hospital of Philadelphia, 3535 Market Street, Room (week 2) for each treatment condition are included under “week 2” of the section “Low Carbohydrate.” Dr. Wadden: Center for Weight and Eating Disorders, 3535 Market Groups were conducted by a registered dietitian or psychol- Street, Suite 3029, Philadelphia, PA 19104.
ogist with experience in weight control. Group leaders attended Dr. Tenhave and Mr. Newcomb: Center for Clinical Epidemiology and an initial, 2-day, in-person training in Philadelphia, and all group Biostatistics, Department of Biostatistics and Epidemiology, 8th Floor, leaders attended biweekly calls throughout the study. The calls Blockley Hall, University of Pennsylvania School of Medicine, 423Guardian Drive, Philadelphia, PA 19104-6021.
were led by a psychologist with extensive experience in behavioral Dr. Klein: Center for Human Nutrition, Washington University School methods of weight control. The calls focused on any clarifications of Medicine, 660 South Euclid Avenue, St. Louis, MO 63110.
of the protocol and the discussion of nonadherent participants.
Low Carbohydrate
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.
A. Begin with reintroduction (names only). If new mem- 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.
bers, include reasons for weight loss as in week 1 but Drafting of the article: G.D. Foster, H.R. Wyatt, J.O. Hill, B. Zemel, T.
keep abbreviated and limit your comments.
B. Ask for volunteers to recall as many names as possible.
Critical revision of the article for important intellectual content: G.D.
C. Address any questions left from last week.
Foster, H.R. Wyatt, J.O. Hill, A.P. Makris, D.L. Rosenbaum, R.I. Stein, D. Briefly review tonight’s agenda. This week we will focus 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, A. Indicate that we want to provide a way for members to Provision of study materials or patients: G.D. Foster, H.R. Wyatt, J.O.
check in briefly at the beginning of each group. For the Hill, B.S. Mohammed, B. Miller, T.A. Wadden.
next few weeks, everyone will check in but over time Statistical expertise: T. Tenhave, C.W. Newcomb.
Obtaining of funding: G.D. Foster, H.R. Wyatt, J.O. Hill, T. Tenhave, (depending on the number of persons in the group, guest lecturers, etc), participants may take turns. Remind Administrative, technical, or logistic support: G.D. Foster, J.O. Hill, about the need to avoid spending too much time on any D.L. Rosenbaum, C. Brill, R.I. Stein, B.S. Mohammed, B. Miller, B.
B. SAFE was chosen to remind us that we want this to be a Collection and assembly of data: G.D. Foster, H.R. Wyatt, J.O. Hill, safe place to discuss eating and exercise habits. (Remind A.P. Makris, D.L. Rosenbaum, C. Brill, R.I. Stein, B.S. Mohammed, B.
about confidentiality). It also reminds us about the key Miller, D.J. Rader, B. Zemel, T.A. Wadden, S. Klein.
things to concentrate on each week.
1. S~self care—Important to view weight loss as self-
care rather than as punitive. It’s something to do for APPENDIX
yourself rather than some punishment that is im- The group treatment sessions were 75 to 90 minutes and posed. Also important to develop non-food alterna- were held weekly from weeks 1 to 20, every other week from tives to nurture self. Each week participants to report weeks 21 to 40, and every 8 weeks from weeks 41 to 104. Groups things they did to take care of themselves that did not included 8 to 12 participants and only contained persons as- include food. Should be things focused on the partic- signed to the same diet condition (low-carbohydrate or low-fat).
ipant rather than her/his family, job, etc. (e.g., mas- Once the group sessions began, no additional members were sage, going to movie that they have been wanting to added, and participants could not attend other group sessions.
see, pedicure, manicure, small “gift”; being inaccessi- W-48 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
ble to others for brief times; going home on time).
E. Indicate that you will collect food record booklets each See “Self-Care” handout for more examples. Part of week and make brief comments about any patterns long-term success is being nice to yourself. Complete you observe. This review should be brief (2 minutes) “Self-Care” handout and pick at least one thing each and include positive comments. Emphasize that these records are for the participants’ benefit not yours. You 2. A~adherence—How were you able to achieve your
are trying to provide a structure to make record keep- goals this week? This includes skill development each week (slow eating, limiting times, etc) as well as in- IV. Goal Setting (15 minutes) (Brownell, pp. 61– 62) dividualized goals (special situations, behaviors from goal worksheet). Review particular successes or diffi- 1. Ask participants to think about how much weight culties. This is a way to get individual attention as they expect to lose over the next 20 weeks. Record well as help the group sharpen its problem-solving them on the board. Ask several participants to de- scribe how they arrived at their numbers. Point out 3. F~food records—Review progress with keeping
that they are probably making assumptions about the records of food and other activities. This is the pri- benefits (e.g., losing 40 pounds will make me feel/ look twice as good as losing 20 pounds) as well as the 4. E~exercise—The physical activity that you per-
costs (e.g., losing the second 20 pounds will be sim- formed this week (type, duration, frequency).
ilar to losing the first 20 pounds). Are these assump- C. Note that W~(weight) is not included in the weekly tions about additional weight loss correct? Review review. Review reasons why weight is a poor short-term faulty assumptions briefly. Avoid getting into a con- measure of success (Brownell, pp. 48 – 49).
test of wills about how much weight people can or should lose. Ultimately, the decision is the 2. water shifts, menstrual cycle, humidity 3. no relation between weight and weekly behavior 2. Compare participants’ goals on the board to what can D. Focus on SAFE and weight loss will follow.
be reasonably expected (1–2 lb per week) (see E. Next week, we will use SAFE to check in.
Brownell, p. 38). Use Brownell diagram (pp. 100 –101) to illustrate that when outcomes (what is achieved) do not match goals (what is expected) there A. Ask participants to describe rationale for self-monitoring are typically negative effects on self-evaluation. How from week 1 (Brownell, pp. 14 –15).
would participants feel if they did not reach their B. Ask about participants’ experiences with recording.
desired weight goals? Probably tend to blame self rather than program or unrealistic goals. Use exam- ples (based on their weight goals) of how same out- comes can be viewed differently based on what was What were participants’ experiences with recording in 3. Actual weight loss will vary due to differences in Was it difficult to record overeating episodes? weight, metabolism and genetics (we will review Did friends or family members comment about record causes of overweight next week). Typical weight loss is 1–2 pounds per week (see Atkins, p. 177). Rather C. It is especially important that participants believe in the than setting a final weight goal now, we recommend utility of keeping records, so be sure to assess this before that participants focus on behavior change and ob- suggesting ways to record better. Focus on any barriers serve what weight loss is accomplished. Weight loss (time, size of record booklets, embarrassment, forget- after week 12 will probably be representative of ting) with specific suggestions. Use group to come up monthly weight loss during the program.
with benefits and suggestions. Emphasize that this is a 4. We recommend an initial goal of a 10% reduction skill that is critical for individualized treatment.
because it is associated with improvements in medical D. Review the weekly food records and how to complete conditions and most persons can achieve it with mod- them (time, amount, type and description of food, add est changes in eating and exercise. When 10% is carbohydrates this week). Stress importance of recording reached, another goal can be set based on costs/ben- ASAP after eating or it will be difficult to recall. Tally efits. Remind participants that body composition will carbohydrate later if necessary. Recommend that they be measured at week 26 so they can make an in- subtotal carbohydrate throughout the day. They can cal- formed decision about further weight loss. It is im- culate carbohydrate using the carbohydrate counter we possible and imprudent to set a long-term weight goal will distribute tonight. Briefly review how the book is now because of the lack of information about costs/ www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 W-49
or off the diet; you’ve had either a good day or a bad 1. Have participants think about one change in their day. There is no middle ground (Brownell, pp. 220 – eating that they would like to make (over the next 4 weeks) that would lead to weight loss. Use several 2. Long-term weight control is based on a regular pat- examples to discuss the following characteristics of tern of eating that avoids extremes and deprivation. It effective goal setting (see handout).
is important to note that the Induction stage of the a. specific— define precisely what is to be accom-
program is only a temporary period designed to ini- plished. Specific goals such as “walk two times this tiate the process of consuming a low carbohydrate week after work on Tuesday and Thursday in the diet. Subsequent stages of the program incorporate a park are more likely to be accomplished that gen- larger variety of foods. Long-term weight control em- eral ones such as, “walk more this week.” Simi- phasizes changes that last. It is based on choosing larly, “eat 20 grams of carbohydrate per day” is foods that you enjoy while making healthy carbohy- more likely to be accomplished than “eat less car- drate choices. The basic theme of any good nutri- tional approach is adaptability. Adding new carbohy- b. reasonable—make small changes. If you’re not
drate containing foods slowly and carefully will help walking at all, do not try to walk every day. The you learn good eating habits. You will be less prone to smaller the difference between your current behav- feeling hungry, irritated, and unhappy. These are feel- ior and your goal behavior the greater the chances you will accomplish it. Small successes lead to bigsuccesses.
Some days will be better than others; it is not realistic to c. active— define your goals in terms of what to do
assume that you should eat the same amount every day. The goal rather than what not to do. For example, “eat ev- is to consume a variety of acceptable foods that you enjoy. The ery four hours” instead of “stop going all day with- goal is not perfection. Eating is not a moral issue. It is inaccurate out eating” or “walk after dinner” rather than an ineffective to make self-evaluations based on eating and exer- “stop lying on the couch after dinner.” d. short-term—assess your goals over short intervals
VI. Induction (15 minutes) (Atkins’ New Diet Revolution, pp. (no more than a week). Sometimes, even shorter intervals are helpful (day by day). Reviewing your progress after short periods will enable you to re- 1. To switch from a high carbohydrate eating plan to a view your accomplishments and troubleshoot any controlled carbohydrate eating plan.
2. To lose weight while eating palatable foods.
e. limited—select no more than two goals per week.
3. To realize that the Induction phase is not going to be Selecting more will decrease your focus and make adherence more difficult. Once your goals have B. In order for Induction to work, it must be followed been accomplished and maintained, you can select precisely; therefore, we suggest that participants follow these guidelines (see Induction Guidelines handout, also f. record—it is helpful to keep a written record of
on pages 122–124 in Atkins’ New Diet Revolution): your goals and progress each week. It will increase 1) Eat 3 regular-size meals a day or 4 –5 smaller meals awareness of your goal and provide an accurate and do not go for more than 4 waking hours without record of your progress. The simplest and easiest records work best. Do what works for you.
2) Eat liberal amounts of fat and protein foods (i.e., 2. Instruct each participant to select one behavioral goal poultry, fish, shellfish, red meat). When you consume for the next week (using the handout as a guide) and fat, use butter, mayonnaise, olive oil, safflower oil, have them record it in the front of their weekly sunflower oil and other vegetable oils rather than record. There will not be time in group to review margarine. See pamphlet for rules regarding egg and each goal. They will discuss this goal under “A” of 3) Eat no more than 20 grams of carbohydrate per day V. Weight Loss: Short-Term Dieting Versus Long-Term Behav- (primarily in the form of salad greens or other per- A. Before establishing a carbohydrate prescription for 4) Do not eat any fruit, bread, pasta, grains, starchy weight loss, let’s review how this approach to long-term vegetables, dairy product (other than cheese, cream, weight control differs from dieting. (Brownell, pp. 6 –7, or butter), or protein/carbohydrate foods (legumes).
5) Only eat acceptable foods (group leaders, see Atkins, 1. Diets are all-or-none. For many people a diet implies pp. 124 –129) listed in the Instruction for Induction short-term dietary change. You’re either on the diet W-50 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
6) Adjust quantities of non-carbohydrate containing are based on serving size, so measuring utensils and foods to suit your appetite (amount that makes you scale (distributed during baseline food intake mea- surement) should be used to accurately determine the 7) Read food labels and check carbohydrate content (see amount consumed. Need to weigh and measure foods Be a Carbohydrate Detective handout).
in the short-term (2 weeks) to become accustomed to 8) Be aware for hidden carbohydrates in gravies, sauces, actual portions. Review guidelines for estimating por- and dressings when eating out. For example, gravy is tion sizes when measuring utensils are not available often made with flour or cornstarch and sugar is (see weekly record). Over time, can perform occa- 9) Drink at least eight 8-oz glasses of water per day (for B. Review two principal benefits of carbohydrate counting.
C. You can lose weight while eating high protein and/or 10) Alcohol is not a source of nutritive carbohydrate fat foods. Fish, shellfish, fowl, meat, and butter are and shouldn’t be consumed in place of food (Atkins for Life, p. 46). Alcoholic beverages should be D. Does not involve self-deprivation or hunger. This eating avoided during Induction for a variety of reasons: plan consists of a variety of foods that are palatable, pleasant, and filling (Atkins, pg. 5, 19, 32).
E. Tips for consuming a reduced carbohydrate diet.
1. Eliminate unnecessary, hidden carbohydrates, which C. Do not try to do a low-fat version of the program as it you do not need such as sugar in sodas and coffee, will disrupt weight loss (Atkins, p. 127).
and choose lower carbohydrate alternatives such as D. In addition to these rules, we ask you to also take a saccharin or sucralose. Limit sweeteners to 3 packets a E. This approach counts carbohydrates rather than calories.
2. Plan ahead. Examine your schedule and prime your Although you will not be counting calories, calories do environment. Stock up on low carbohydrate snacks matter. Gaining weight results from eating more calories and eliminate high carbohydrate temptations. Some than you burn, so eat until satisfied and do not gorge examples of low carbohydrate snacks are turkey and cheese roll, single serving can of tuna, hard boiledeggs, Laughing Cow cheese or string cheese, seeds, VII. Carbohydrate Counting (10 minutes) nuts or nut butter on a celery stick, and sugar free A. Explain the general concept of carbohydrate counting Jello. Carbohydrate controlled foods are generally using a household budget or bank account as a model.
found around the periphery of the grocery store.
Review the basic principles of carbohydrate counting.
3. Avoid deprivation. Eat regularly (every 4 hours) to 1. You receive a 20-gram carbohydrate deposit each day, prevent hunger. Have a small carbohydrate controlled which you can spend according to your own personal snack high in fat or protein if you are hungry between preferences. You decide how to spend your carbohy- drates. This will require you to consider how much 4. Eat primarily unprocessed foods but when you eat pack- you enjoy a particular food versus what it costs. You aged foods (i.e., cheese), read the food labels carefully.
can have 4 cups of salad vegetables per day. However, Generally, “low fat” means “high carbohydrate.” if you would like to add vegetables that containslightly higher carbohydrate contents (limited to 1 NOTE: Although ATKINS Ready to Drink Shakes (up to 1
cup per day), you must reduce your intake of salad per day), ATKINS Shake Mix (up to 2 scoops per day), and vegetables from 4 to 3 cups per day. Emphasize that ATKINS ADVANTAGE BARS (up to 1 per day) can be con- participants cannot simply eat the maximum amount sumed in place of whole foods during Induction, this option of servings from each food group listed in the pam- should only be initiated when it has been determined that the phlet because they will likely go over the 20-gram individual cannot incorporate whole foods into his/her eating limit. The sample menu handout provides some meal plan (like during crunch times). At this point it would be pre- mature to offer this as an option. ATKINS Endulge products 2. Using your carbohydrate counter and food labels, cannot be consumed during Induction.
record the number of carbohydrates that you spend F. Inform participants that they may experience some un- desirable symptoms (i.e., headaches, constipation) after 3. Using the handout, briefly review key aspects of the second day of Induction (see back of Instructions for the food label and review how to calculate net Induction sheet). Call participants after the third day of Induction and ask about their progress and whether they 4. Tally your carbohydrate consumption. The key factor are experiencing any problems. Do not specifically ask is to consume no more than 20 grams of carbohy- about symptoms. Example: “I am calling to see how you drate per day. Point out that grams of carbohydrate are doing on your new eating plan and to find out whether www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 W-51
you have any questions or are experiencing any problems so see, pedicure, manicure, small “gift”; being inaccessi- ble to others for brief times; going home on time).
See “Self-Care” handout for more examples. Part of VIII. Skill Building (Handout) (5 minutes) long-term success is being nice to yourself. Complete “Self-Care” handout and pick at least one thing each 3. Record all food (time, amount, type and description of 2. A~adherence—How were you able to achieve your
goals this week? This includes skill development each 4. Use the carbohydrate counter and food labels to deter- week (slow eating, limiting times, etc) as well as in- mine carbohydrate intake. Key thing is to eat a wide dividualized goals (special situations, behaviors from goal worksheet). Review particular successes or diffi- 5. Record one personal goal for this week in the beginning of the weekly record and assess progress as appropriate.
culties. This is a way to get individual attention aswell as help the group sharpen its problem-solving 3. F~food records—Review progress with keeping
records of food, exercise and other activities. This is the primary tool of weekly assessment.
4. E~exercise—The physical activity that you per-
formed this week (type, duration, frequency).
C. Note that W~(Weight) is not included in the weekly review. Review reasons why weight is a poor short-term measure of success (Brownell, pp. 48 – 49).
2. water shifts, menstrual cycle, humidity 3. no relation between weight and weekly behavior D. Focus on SAFE and weight loss will follow.
E. Next week, we will use SAFE to check in.
A. Begin with reintroduction (names only). If new mem- bers, include reasons for weight loss as in week 1, but A. Ask participants to describe rationale for self-monitoring keep abbreviated and limit your comments.
from week 1 (Brownell, pp. 14 –15).
B. Ask for volunteers to recall as many names as possible.
B. Ask about participants’ experiences with recording.
C. Address any questions left from last week.
D. Briefly review tonight’s agenda. This week we will focus 4. Did they have difficulty estimating portions? A. Indicate that we want to provide a way for members to 5. What were participants’ experiences with recording in check in briefly at the beginning of each group. For the next few weeks, everyone will check in but over time 6. Was it difficult to record overeating episodes? (depending on the number of persons in the group, 7. Did friends or family members comment about guest lecturers, etc), participants may take turns. Remind about the need to avoid spending too much time on any C. It is especially important that participants believe in the utility of keeping records, so be sure to assess this before B. SAFE was chosen to remind us that we want this to be a suggesting ways to record better. Focus on any barriers safe place to discuss eating and exercise habits. (Remind (time, size of record booklets, embarrassment, forget- about confidentiality). It also reminds us about the key ting) with specific suggestions. Use group to come up with benefits and suggestions. Emphasize that this is a 1. S~self care—Important to view weight loss as self-
skill that is critical for individualized treatment.
care rather than as punitive. It’s something to do for D. Review the new food record booklets and how to com- yourself rather than some punishment that is im- plete them (time, amount, type and description, add posed. Also important to develop non-food alterna- calories this week). Stress importance of recording ASAP tives to nurture self. Each week participants to report after eating or it will be difficult to recall. Tally calories things they did to take care of themselves that did not later if necessary. Recommend that they subtotal calories include food. Should be things focused on the partic- throughout the day. They can calculate calories using ipant rather than her/his family, job, etc. (e.g., mas- the calorie counter we will distribute tonight. Briefly sage, going to movie that they have been wanting to W-52 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
E. Indicate that you will collect food record booklets each week and make brief comments about any patterns 1. Have participants think about one change in their you observe. This review should be brief (2 minutes) eating that they would like to make (over the next 4 and include positive comments. Emphasize that these weeks) that would lead to weight loss. Use several records are for the participants’ benefit not yours. You examples to discuss the following characteristics of are trying to provide a structure to make record keep- effective goal setting (see handout).
a. specific— define precisely what is to be accom-
plished. Specific goals such as “walk two times this IV. Goal Setting (Brownell, pp. 61– 62) (15 minutes) week after work on Tuesday and Thursday in the park are more likely to be accomplished that gen- 1. Ask participants to think about how much weight eral ones such as, “walk more this week.” Simi- they expect to lose over the next 20 weeks. Record larly, “eat 1200 –1400 calories per day” is more them on the board. Ask several participants to de- likely to be accomplished than “eat less this week.” scribe how they arrived at their numbers. Point out b. reasonable—make small changes. If you’re not
that they are probably making assumptions about the walking at all, do not try to walk every day. If benefits (e.g., losing 40 pounds will make me feel/ you’re eating 10 candy bars each week, do not look twice as good as losing 20 pounds) as well as the attempt to eat only 2. The smaller the difference costs (e.g., losing the second 20 pounds will be sim- between your current behavior and your goal be- ilar to losing the first 20 pounds). Are these assump- havior the greater the chances you will accomplish tions about additional weight loss correct? Review it. Small successes lead to big successes.
faulty assumptions briefly. Avoid getting into a con- c. active— define your goals in terms of what to do
test of wills about how much weight people can or rather than what not to do. For example, “eat ev- should lose. Ultimately, the decision is the ery four hours” instead of “stop going all day with- out eating” or “walk after dinner” rather than 2. Compare participants’ goals on the board to what can “stop lying on the couch after dinner.” be reasonably expected (1–2 lb per week) (see d. short-term—assess your goals over short intervals
Brownell p. 38). Use Brownell diagram (pp. 100 – (no more than a week). Sometimes, even shorter 101) to illustrate that when outcomes (what is intervals are helpful (day by day). Reviewing your achieved) do not match goals (what is expected) there progress after short periods will enable you to re- are typically negative effects on self-evaluation. How view your accomplishments and troubleshoot any would participants feel if they did not reach their desired weight goals? Probably tend to blame self e. limited—select no more than two goals per week.
rather than program or unrealistic goals. Use exam- Selecting more will decrease your focus and make ples (based on their weight goals) of how same out- adherence more difficult. Once your goals have comes can be viewed differently based on what was been accomplished and maintained, you can select 3. Actual weight loss will vary due to differences in f. record—it is helpful to keep a written record of
weight, metabolism and genetics (we will review your goals and progress each week. It will increase causes of overweight next week). Typical weight loss awareness of your goal and provide an accurate is 1–2 pounds per week (see Brownell p. 38). Rather record of your progress. The simplest and easiest than setting a final weight goal now, we recommend records work best. Do what works for you.
that participants focus on behavior change and ob- 2. Instruct each participant to select one behavioral goal serve what weight loss is accomplished. Weight loss for the next week (using the handout as a guide) and after week 12 will probably be representative of have them record it in the front of their weekly monthly weight loss during the program.
record. There will not be time in group to review 4. We recommend an initial goal of a 10% reduction each goal. They will discuss this goal under “A” of because it is associated with improvements in medical conditions and most persons can achieve it with mod-est changes in eating and exercise. When 10% is V. Weight Loss: Short-Term Dieting Versus Long-Term Behav- reached, another goal can be set based on costs/ben- efits. Remind participants that body composition will A. Before establishing a caloric prescription for weight loss, be measured at week 26 so they can make an in- let’s review how our approach to long-term weight con- formed decision about further weight loss. It is im- trol differs from dieting (Brownell, pp. 6 –7, 12–13).
possible and imprudent to set a long-term weight goal 1. Diets are all-or-none. For many people a diet implies now because of the lack of information about costs/ short-term dietary change. You’re either on the diet or off the diet; you’ve had either a good day or a bad www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 W-53
day. There is no middle ground (Brownell, pp. 220 – record. Point out that calories are based on serving 221). Furthermore, many diets are based on fads, ex- size, so measuring utensils and scale (distributed dur- tremes, and severe deprivation. As such, they are only ing baseline food intake measurement) should be successful in the short-term. Can only make dramatic used to accurately determine the amount consumed.
Need to weigh and measure foods in the short-term 2. Long-term weight control is based on a regular pat- (2 weeks) to become accustomed to actual portions.
tern of eating that avoids extremes and deprivation. It Review guidelines for estimating portion sizes when emphasizes small changes that last. It is based on measuring utensils are not available (see weekly choosing foods that you enjoy while staying within record). Over time, can perform occasional checks or the boundaries of daily caloric allowances. There are weigh novel foods. Briefly review some basic compo- no forbidden foods. Allow yourself to have some spe- nents (e.g. serving size, calories) of the food label us- cial foods or treats and work them into daily/weekly/ ing the “Be a Calorie Detective” handout.
monthly allowances. You will be less prone to feeling 3. Tally your calorie account as expenditures are made. Al- deprived, irritated, unhappy. These are feelings that though you are allotted a certain number or calories each day, you will not spend this amount every day. You can B. There are no absolutes (never, always, must) in successful save calories for special occasions, just as you save weight control. Some days will be better than others; it is money. For example, a person could save 100 calories not realistic to assume that you should eat the same per day, Monday through Friday, and spend the extra amount every day. The goal is to consume a variety of 500 calories over the weekend. You may also spend less foods that you can enjoy and sustain balanced eating to adjust for an unusually large expenditure.
over the long haul. The goal is not perfection. Eating is 4. The key factor is that the calorie ledger must balance not a moral issue. It is inaccurate and ineffective to make (i.e., average 1200 –1500 calories/day or 1500 –1800 self-evaluations based on eating and exercise behavior.
calories/day) in order for your rate of weight loss to VI. Picking a Calorie Range (10 minutes) (Brownell, pp. remain constant. It is best to review your ledger for a weekly balance (8400 –10,500 calories per week).
B. Review two principal benefits of keeping a calorie account.
1. Allows for flexibility and variety.
2. No single overeating episode is paramount, since you 3. Intake ϭ Output ϭ Weight Maintenance can balance your calorie account with adjustments.
B. 3500 calories ϭ 1 pound. To lose one pound/week, C. Tips for reducing caloric intake.
you’ll need to eat 3500 calories less than you burn. Eas- 1. Reduce or eliminate unnecessary, hidden calories ier to decrease intake than to increase output (i.e., easier which you do not need or really enjoy (e.g., eliminate to eat 500 calories less per day than it is to exercise 500 butter, sugar in sodas and coffee, reduce use of cream, calories more per day). Give examples. That’s why exer- choose lower calorie alternatives if similar taste).
cise alone is not the best method for weight loss. Regular 2. Plan ahead. Examine your schedule and prime your physical activity, however, is the best predictor of main- environment. Stock up on low calorie snacks and eliminate high fat and calorie temptations. Be con- C. We are recommending a calorie range (1200 –1500 cal- scious of the caloric cost of food choices. Are the ories/d for women and 1500 –1800 calories/d for men); calories worth it? If they are, fine; if not, skip it or participants will decide how to “spend” those calories choose an alternative. Examples of low calorie snacks using the principles below. As above, need to observe can be found on the handout. Encourage participants VII. A Calorie Account (10 minutes) NOTE: Although meal replacement shakes and bars (e.g.,
A. Explain the general concept of a calorie account using a Slim-Fast) can be consumed in place of whole foods, this option household budget or bank account as a model. Review should only be initiated when it has been determined that the the basic principles of using a calorie account. The sam- individual cannot incorporate whole foods into his/her eating ple menu handouts provide some meal ideas.
plan (like during crunch times). At this point it would be pre- 1. You receive a 1200 –1500 (women) or 1500 –1800 (men) calorie deposit each day, which you can spend 3. Avoid deprivation. It’s a short-term solution to a according to your own personal preferences. You de- long-term problem. Do not totally eliminate foods cide how to spend your calories. This will require you that you really enjoy. Find a way to work them in.
to consider how much you enjoy a particular food Make changes that you can live with.
4. Eat regularly (every 4 hours) to prevent hunger. This 2. Using your fat and calorie counter, record the num- will be addressed in greater detail next week when we ber of calories that you spend each day in your weekly discuss developing an eating schedule.
W-54 3 August 2010 Annals of Internal Medicine Volume 153 • Number 3
www.annals.org
D. Remind participants to take a multivitamin each day.
3. Use the calorie counter and food labels to determine E. Inform participants that they may experience some un- calorie intake. Key thing is to eat a wide variety of desirable symptoms. Call participants within the first three days to ask about their progress and whether they 4. Record one personal goal for this week in beginning of are experiencing any problems. Do not specifically ask the food record and assess progress as appropriate.
about symptoms. Example: “I am calling to see how you are doing on your new eating plan and to find out whether you have any questions or are experiencing any problems so VIII. Skill Building (Handout) (5 minutes) 4. Tip the Calorie Balance Handout5. Be a Calorie Detective Handout 1. Eat a diet that is consistent with your calorie goal (1200 – 6. Enjoy the Variety-Healthy Food Choices 1500 calories per day for women, 1500 –1800 calories 2. Record all food (time, amount, type and description, Appendix Table. Serious Adverse Events Among 307 Participants Over 2 Years
Study Group
Possibly, but not likely; weight loss was www.annals.org
3 August 2010 Annals of Internal Medicine Volume 153 • Number 3 W-55

Source: http://www.swissmilk.ch/de/services/ernaehrungs-fachleute/beratungsmaterial/low-carb-plus/-dl-/fileadmin/filemount/foster-10-weight-and-metabolic-outcomes-after-2-years-an-a-low-carbohydrate-versus-low-fat-diet.pdf