Innovative Program

Successful Internet-Based Lifestyle

Change Program on Body Weight
and Markers of Metabolic Health
Holly R. Wyatt, M.D.,1 Lorraine G. Ogden, Ph.D.,1 Kristen S. Cassic, M.A.,1
Emily A. Hoagland, M.S., R.D.,1 Toni McKinnon, R.N.,2 Natalie Eich, B.S.,2
Vasiliy Chernyshev, B.S.,2 Tim Wood, Ph.D.,2 John Cuomo, Ph.D.,2
and James O. Hill, Ph.D.1
Center for Human Nutrition, University of Colorado Denver, Denver, CO
1

USANA Health Sciences, Inc., Salt Lake City, UT

2

Abstract Summary:
This study dem-
Objective: The objective of this study was to evaluate onstrates than an
the effectiveness of an Internet-based 12-Week Healthy online lifestyle
for Life Program in supporting weight loss and improve- change program
ments in metabolic and cardiovascular health among sub- that prescribes a
jects with metabolic syndrome. low-glycemic diet,
Research methods and procedures: Sixty subjects nutritional supple-
with metabolic syndrome were studied before, during, and ments, and moder- Holly R. Wyatt, James O. Hill,
after a 12-week online lifestyle intervention program that ate exercise can M.D. Ph.D.
prescribed a low-glycemic diet, nutritional supplementa- successfully pro-
tion, and moderate exercise. duce meaningful
Results: The intervention produced an average weight weight loss, significant improvements in glycemic control,
loss of 5.5 kg (5.4%). Measures of glycemic control and significant reductions in risk factors for heart disease in
improved significantly during the study. Fasting insulin individuals with metabolic syndrome.
was reduced by 32.3% and 120-minute insulin during an
oral glucose tolerance test was reduced by 43.6%. Insulin
sensitivity was increased as evidenced by a reduction in the
homeostatic model assessment (HOMA) index (by 31.6%)
Introduction

H
and an increase in the insulin sensitivity index. There were igh rates of overweight and obesity in the United
also significant improvements in triglycerides, total choles- States1 suggest that many Americans are at increased
terol, and blood pressure. At the end of the study, 58.5% risk for several chronic diseases.2 Most notably,
of the study completers met criteria for the metabolic syn- overweight and obesity are often associated with a cluster of
drome compared to 84.9% at baseline (p = 0.002). risk factors for diabetes and cardiovascular disease. These

DOI: 10.1089/obe.2009.0406 © Mary Ann Liebert, Inc. August 2009 Obesity and Weight Management 167
factors include a large waist circumference, elevated blood at least two other risk factors described below) were
pressure, elevated triglycerides and fasting glucose, low recruited from the Denver metropolitan area. All subject
high-density lipoprotein (HDL) cholesterol, and poor insulin recruitment and selection was performed by University of
sensitivity. Individuals possessing several of these symptoms Colorado Denver, independently of USANA Health Sci-
are now often diagnosed as having metabolic syndrome,3 ences and The Healthy for Life Program. Eligible subjects
a prediabetic state that recent research indicates may be were 20 to 60 years of age with a body mass index ≤ 42
reversible, in large measure through lifestyle change. Given kg/m2 and a waist circumference > 40 inches (males) or >
the rising rates of type 2 diabetes, there is an urgent need to 35 inches (females). In addition, subjects had to have at
develop lifestyle intervention programs for people with met- least two of the following measurements at screening and/
abolic syndrome to prevent the progression of their disease. or baseline: elevated blood pressure (systolic > 130 mm
Weight loss is an indicated treatment for both obesity Hg and/or diastolic > 85 mm Hg); elevated triglycerides
and metabolic syndrome. Modest weight loss (5%–10% (> 150 mg/dL); elevated fasting glucose (> 100 mg/dL);
of initial weight) can improve cardiometabolic risk fac- or low HDL cholesterol (< 40 mg/dL for males, < 50 mg/
tors and reduce the risk of developing type 2 diabetes.4 dL for females). Participants also had to have access to
The challenge lies in designing and providing programs e-mail and be willing to make changes to their diet and
that can effectively help the large numbers of people with increase their activity level. Participants were excluded
metabolic syndrome to achieve modest weight loss. Sever- if they were pregnant or lactating, had been diagnosed
al approaches are available. In research settings, behavioral with type 2 diabetes or were taking medication for blood
group treatment,5 individual treatment by counselors,5 meal glucose control. Subjects on lipid-lowering medications,
replacement programs,5 and pharmaceutical interventions6 with allergies, or significant intolerance to soy foods or
have shown some success. But given the large number with acute or chronic illnesses that prevented participa-
of people who are overweight or obese, and/or who have tion in the study were also excluded.
metabolic syndrome, scale-up remains an issue. All participants provided informed written consent. The
The Internet provides one means to easily and inexpen- study protocol was approved by the Western Institutional
sively deliver weight loss interventions to large numbers of Review Board (WIRB). Recruitment was via newspaper
people. That said, success to date in using this tool has been and email advertising; a telephone number was provided
modest.7 The intent of this trial was to determine whether a for subjects to obtain more information and participate in
12-week, Internet-based lifestyle modification program pre- a prescreening interview. Research staff prescreened 262
scribing a low-glycemic diet (including low-glycemic func- subjects via telephone. Of these, 80 subjects were screened
tional foods), vitamin and mineral supplements, and modest in person, at which time data were obtained on weight,
exercise could reduce body weight and improve symptoms waist circumference, blood pressure, fasting blood lipids,
related to the metabolic syndrome and cardiovascular risk. and fasting blood glucose. There were 20 screen failures,
and 60 subjects were enrolled (24 males, 36 females).
Methods and Procedures Study Design
Participants Enrolled subjects were asked to participate in a 12-week
Male and female subjects with metabolic syndrome Healthy for Life Internet program that prescribed a low-
(defined as having abdominal adiposity combined with glycemic diet and modest exercise. They were provided

Table 1. Study Design for Nutritional Products and Coaching

WEEKS 1–4 Weeks 5–12
Lifestyle Coaching Subjects received daily/weekly emails and kept an online log of food intake and physical activity
Supplements (A.M.) 2 USANA Mega Antioxidant pills & 2 USANA Chelated Mineral pills
Breakfast 1 USANA Nutrimeal™ Shake 1 USANA Nutrimeal™ Shake
Snack 1 USANA Nutrition Bar 1 USANA Nutrition Bar
Lunch 1 USANA Nutrimeal™ Shake low-glycemic meal
Snack 1 USANA Nutrition Bar healthy snack
Dinner low-glycemic meal low-glycemic meal
Supplements (P.M.) 2 USANA Mega Antioxidant pills & 2 USANA Chelated Mineral pills

168 Obesity and Weight Management August 2009

with vitamin and mineral supplements, low-glycemic rises in blood glucose, and instead to eat low-glycemic
meal replacement shakes, and nutrition bars as part of foods that cause modest but sustained increases in blood
the prescribed diet (Table 1). Moreover, they were asked glucose. During the first 4 weeks, subjects were instruct-
to keep an online food and physical activity diary for the ed to consume one low-glycemic meal replacement
duration of the study. shake for breakfast, another for lunch, two low-glycemic
The following measurements were obtained at baseline, nutrition bars as snacks, and a low-glycemic dinner pre-
week 6, and week 12: weight, waist circumference, blood pared according to instructions provided by the website.
pressure, and fasting blood measures of glucose, insulin,
hemoglobin A1c (HbA1c) , serum triglycerides, total cho-
lesterol, low-density lipoprotein (LDL) cholesterol, HDL
cholesterol, high-sensitivity C-reactive protein, vitamin There is an urgent need
E (α and γ), plasma-induced isoprostanes, and urinary
isoprostanes. In addition, oral glucose tolerance tests
to develop lifestyle
were administered at each visit; pre- and post-glucose intervention programs
and insulin were obtained (at 0 minutes and 120 minutes,
respectively), and a finger-stick glucose was obtained for people with metabolic
at 90 minutes. Indexes of homeostatic model assess-
ment (HOMA) and insulin sensitivity were calculated
syndrome to prevent the
from these results.8,9 Subjects also completed the Food progression of their disease.
Craving Inventory10 at each visit. All data collection was
preformed independent of USANA Health Sciences by
University of Colorado Denver staff.
During the final 8 weeks of the study, subjects were
instructed to consume one meal replacement shake for
Internet-based lifestyle change either breakfast or lunch, and two low-glycemic meals
again prepared according to instructions. They were
program also told to consume two low-glycemic snacks per day
The 12-week program was developed by a family prac- during this period; one nutrition bar and one snack of
tice physician, and it was administered over the Internet. their choice. Throughout the study, subjects were told to
The Internet-based behavior modification program served take the prescribed vitamin/mineral supplements daily.
two purposes: (1) to educate participants about the prin- The commercially available supplements, shakes and
ciples and practices of healthy lifestyle change and (2) bars were provided free of charge to the participants
to help hold participants accountable for adhering to the by USANA Health Sciences, the study sponsor. The
recommended interventions. products are outlined in Table 1. Study participants also
Participants received daily motivational and instruc- received $50 per study visit for a total of $200 compen-
tional e-mails that provided guidance in making healthy sation for participation in the study.
food choices and increasing physical activity. Partici-
pants were also asked to make daily entries in their own Statistical methods
online lifestyle journals, recording what they ate, how Demographic and baseline characteristics are reported
they exercised, and whether they took their nutritional as mean ± standard deviation (SD) for continuous vari-
supplements. Lifestyle journals were reviewed weekly by ables or number and percentage of participants for
a lifestyle coach who was then able to offer feedback and categorical variables. To examine changes in outcome
personalized guidance concerning program adherence. measures over time (baseline, 6 weeks, and 12 weeks)
The coach was also available via e-mail to answer partici- primary analyses utilized repeated measures mixed mod-
pants’ questions regarding the program. In addition, each els performed using the SAS PROC MIXED procedure.
participant received a copy of the book Healthy for Life, The repeated measures mixed model includes all available
which described the lifestyle program and the principles data for each participant and accounts for missing data in
of a low-glycemic diet in greater detail.11 the model. An unstructured covariance (type = UN) was
specified for the covariance structure and restricted maxi-
Low-glycemic diet mum likelihood (REML) estimation was used. Several
During the 12-week intervention period, subjects were outcome measures were log-transformed prior to analy-
instructed to avoid high-glycemic foods that cause rapid sis to better approximate a Gaussian distribution and to

August 2009 Obesity and Weight Management 169

reduce the impact of outliers on the analysis. Results for At baseline 84.9% enrolled subjects met criteria for the
these outcome measures were back-transformed and are metabolic syndrome, 80% had an elevated blood pressure
thus interpreted on the multiplicative scale (i.e., percent (> 130/85 mm Hg), 65% had triglycerides >150 mg/dL,
change). All analyses were performed using SAS statisti- 33% had a blood glucose > 100 mg/dL, and 68% had low
cal software (SAS Institute Inc., Cary, NC). HDL based on gender.

Attrition
Results Of the 60 enrolled subjects, 53 (23 males, 30 females)
Baseline characteristics completed the study (88%) and 7 discontinued from
Two hundred sixty-two subjects were phone screened the study early. Of the 7 early terminations, 5 subjects
for this study and 80 were consented. There were 20 con- withdrew consent and 2 subjects withdrew because of
sented screen failures; 60 subjects (24 males, 36 females) illness.
were enrolled in the study.
Demographic and screening characteristics for enrolled Weight and waist circumference change
participants are presented in Table 2 for the overall sam- Over 12 weeks, study participants lost an average of
ple and by gender. Data are presented as mean ± SD for 12.1 pounds (95% confidence interval [CI]: −14.2 to
continuous variables or number and percentage of partici- −9.9 pounds p < 0.001). This equates to an average 5.4%
pants for categorical variables. - weight loss (95% CI: −6.4% to −4.4%). Table 3 presents

Table 2. Demographic and Screening

Characteristics of Enrolled Participants (n = 60)
MALES FEMALES TOTAL The Internet provides
n = 24 n = 36 n = 60
Age (yrs) 51.1 ± 8.3 53.2 ± 6.0 52.4 ± 7.0 one means to easily and
Race/ethnicity
White 19 (79.2%) 26 (72.2%) 45 (75.0%)
inexpensively deliver
Latino/Hispanic
Other
5 (20.8%)
0 (0.0%)
7 (19.4%)
3 (8.3%)
12 (20.0%)
3 (5.0%)
weight-loss interventions
Weight (pounds) 245.6 ± 37.5 207.0 ± 31.4 222.5 ± 38.7 to large numbers of people.
BMI (kg/m2) 34.4 ± 4.6 34.6 ± 3.9 34.5 ± 4.2
Waist circumference (in) 46.1 ± 3.8 43.4 ± 4.2 44.5 ± 4.2
SBP (mm Hg) 138.0 ± 10.6 131.2 ± 16.2 133.9 ± 14.5
changes in body weight during the 12-week program. As
DBP (mm Hg) 92.9 ± 6.9 92.8 ± 9.2 92.8 ± 8.3
shown, the majority of the weight loss (−9.1 pounds, 95%
Fasting glucose (mg/dL) 101.2 ± 16.2 101.0 ± 13.8 101.1 ± 14.7
CI: −10.5 to −7.7 pounds, p < 0.001) occurred during
Triglycerides (mg/dL) 232.1 ± 128.6 210.5 ± 117.7 219.2 ± 121.5
the first 6 weeks of the program. On average, there was
LDL cholesterol (mg/dL) 141.1 ± 24.8 127.7 ± 32.1 132.9 ± 30.0 an additional 3.0 pound weight loss during the second 6
HDL cholesterol (mg/dL) 39.7 ± 10.8 44.0 ± 7.4 42.3 ± 9.1 weeks (95% CI: −4.1 to −1.9 lbs, p < 0.001). There were
Data presented as mean ± standard deviation (SD) or number of also significant reductions in BMI (−1.9 kg/m2, 95% CI
participants (% of participants). −2.2 to −1.6 kg/m2, p < 0.001) and waist circumference
SBP, systolic blood pressure; DBP, diastolic blood pressure;
LDL, low-density lipoprotein; HDL, high-density lipoprotein. (−2.0 inches, 95% CI −2.2 to −1.6 inches, p < 0.001) dur-
ing the 12-week program.

Table 3. Changes in Weight and Waist Circumference, All Available Data (n = 60)
VARIABLE BASELINE 6 WEEKS 12 WEEKS 6-WEEK CHANGE 12-WEEK CHANGE
Weight (lbs) 222.53 213.46 210.47 -9.07 a
-12.06a
Percent Weight Loss (%) -4.1% -5.4%
Waist Circumference (in) 44.61 43.27 42.58 -1.34 a
-2.02a
Mixed model estimates (95% confidence interval [CI]).
a
p < 0.001.

170 Obesity and Weight Management August 2009

Changes in glycemic control Changes in cardiovascular health
Measures of glycemic control improved significantly and inflammation
during the 12-week intervention. Table 4 summarizes
these changes. At the end of 12 weeks, fasting insulin Table 5 presents the changes in measures of car-
was reduced on average by 32.3% (95% CI: −41.1% diovascular health and markers of inflammation that
to −22.2%, p < 0.001) and 120-minute oral glucose occurred during the 12-week program. There were sig-
tolerance test (OGTT) insulin by an average of 43.6% nificant improvements in all measures at the end of 6
(95% CI: −54.8% to −29.8%, p < 0.001). There were weeks, but by the end of 12 weeks, significant changes
no significant changes in fasting glucose (p = 0.345), were only observed for triglycerides, total cholesterol,
90-minute OGTT glucose (p = 0.067) or 120-minute and systolic and diastolic blood pressure.
OGTT glucose (p = 0.119). HOMA index was reduced Changes in antioxidants and inflammatory mark-
by an average of 31.6% (95% CI: −41.0% to −20.8%, ers were also measured. C-reactive protein levels
p < 0.001) and the insulin sensitivity index (ISI) declined by an average of 26.5% during the 12-week
increased an average of 0.03 points (95% CI: 0.02 to intervention (95% CI: −38.2% to −12.7%, p < 0.001).
0.04, p < 0.001). At the end of the 12-week interven- Vitamin E (α- tocopherol) levels increased by an aver-
tion, HbA1c levels were 1.8% lower than baseline (95% age of 30.4% (95% CI: 21.2% to 40.4%, p < 0.001).
CI: −3.4% to −0.2%, p < 0.001). This was the result Plasma antioxidant reserve 12 was increased 20.4% as
of a large reduction in HbA1c during the first 6 weeks measured by a 20.4% reduction in induced isopros-
of the program (−2.8%, 95% CI: −4.3% to −1.3%, p =
0.001) followed by a 1.1% increase (95% CI: 0.1% to
1.9%, p = 0.029; data not shown) during the second Table 5. Changes in Measures of
6-week period. Cardiovascular Health, Antioxidants and
Inflammatory Markers All Available Data
(n = 60): Mixed Model Estimates (95% CI)
VARIABLE 6-WEEK CHANGE 12-WEEK CHANGE
Table 4. Changes in Glycemic Control,
Triglycerides, mg/dLa -17.19% -15.67%
All Available Data (n = 60): Mixed Model (-25.25%, -8.37%) (-23.73%, -6.76%)
Estimates (95% CI) p < 0.001 p = 0.001
VARIABLE 6-WEEK CHANGE 12-WEEK CHANGE Total Cholesterol, mg/dL -17.99 -10.35
Fasting glucose, 0.10% 1.29% (-25.53, -10.46) (-17.17, -3.53)
mg/dL (-2.78%, 3.07%) (-1.40%, 4.06%) p < 0.001 p = 0.004
p = 0.947 p = 0.345 LDL Cholesterol, mg/dL -9.81 -5.25
Fasting insulin, -20.92% -32.33% (-16.41, -3.22) (-11.85, 1.34)
μIU/mL (-29.27%, -11.59%) (-41.11%, -22.23%) p = 0.004 p = 0.116
p < 0.001 p < 0.001 HDL Cholesterol, mg/dL -1.77 0.25
HOMA Index -20.97 % -31.62 % (-3.47, -0.06) (-1.20, 1.71)
(-30.46%, -10.19%) (-40.95%, -20.82%) p = 0.043 p = 0.729
p < 0.001 p < 0.001 Systolic Blood Pressure, -6.51 -7.88
OGTT glucose, mg/dL 90 -5.31% -4.94% mm Hg (-10.27, -2.75) (-11.82, -3.95)
minutes (-10.35%, 0.00%) (-9.96%, 0.38%) p = 0.001 p < 0.001
p = 0.050 p = 0.067 Diastolic Blood Pressure, -5.69 -6.76
OGTT glucose, mg/dL 120 -13.92% -7.74% mm Hg (-8.14, -3.23) (-9.09, -4.43)
minutes (-22.40%, -4.51%) (-16.68%, 2.16%) p < 0.001 p < 0.001
p = 0.005 p = 0.119 C-reactive protein, mg/dLa -21.26% -26.53%
OGTT insulin, μIU/mL 120 -39.49% -43.64% (-33.08%, -7.34%) (-38.19%, -12.67%)
minutes (-49.76%, -27.13%) (-54.77%, -29.77%) p = 0.005 p < 0.001
p < 0.001 p < 0.001 Induced Isoprostanes, -18.65% -20.44%
HbA1c, % -2.82% -1.84% pg/mLa (-24.88%, -11.90%) (-26.70%, -13.65%)
(-4.34%, -1.27%) (-3.41%, -0.24%) p < 0.001 p < 0.001
p < 0.001 p = 0.025 Urinary Isoprostanes, -17.15% -28.96%
Mixed model estimates of % change in outcome measures (all variables ng/mg cra (-28.51%, -4.00%) (-39.16%, -17.04%)
were log transformed prior to analysis and parameter estimates were back p = 0.013 p < 0.001
transformed after analysis to be interpreted as % change). a
Log-transformed variable.
CI, confidence interval; HOMA, homeostatic model assessment; OGTT, Mixed model estimates of absolute change in outcome measures
oral glucose tolerance test; HbA1c, hemoglogin A1c. (backtransformed to % change for log-transformed variables).

August 2009 Obesity and Weight Management 171

tanes (95% CI: −26.7% to −13.7%, p < 0.001) and Discussion
urinary isoprostanes fell by 29.0% (95% CI: −39.2%
to −17.0% p < 0.001). These results demonstrate that an Internet-based life-
style change program can successfully produce mean-
Changes in food cravings ingful weight loss in obese individuals with metabolic
Table 6 presents changes in scores on the Food Crav- syndrome. The average weight loss was 5.4% and was
ings Inventory (FCI) during the 12-week study. Both associated with clear improvements in glycemic control
FCI total scores as well as scores on each of the four and reductions in cardiometabolic risk factors.
subscales were reduced significantly during the 12-week
intervention.

Percentage of completers with metabolic syndrome Enrolled subjects were

measurements
Figure 1 illustrates the changes over the course of
asked to participate in
the study in percentages of completers for each of a 12-week Healthy for
the metabolic syndrome criteria and prevalence of
the metabolic syndrome. At baseline, 84.9% of study Life Internet program that
completers had three or more criteria defining the
metabolic syndrome and at 12 weeks, 58.5% had three
prescribed a low-glycemic
or more criteria. The percentage of completers with diet and modest exercise.
elevated triglycerides at baseline was 60.4%. This
percentage decreased significantly to 45.3% at week
12. In addition, the percentage of completers meeting
the criteria for elevated blood pressure (>130/85 mm The 12-week Healthy for Life program was delivered
Hg) decreased from 79.3% at baseline to 43.4% at 12 via the Internet but involved using standard weight loss
weeks. Significant changes in the percentage of com- tools such as meal replacements, self-monitoring, behav-
pleters were not noted in the criteria for elevated waist ioral change strategies, and low-glycemic diets. The results
circumference, fasting glucose, or low HDL. are significant in that they show the feasibility of using
Identical analyses were also performed for the 53 standard tools delivered via an Internet format to achieve
complete cases (participants who attended all three study weight loss in obese individuals with metabolic syndrome.
visits: baseline, 6 weeks, and 12 weeks) and the results Given the need for strategies to help large numbers of
were similar to those reported using all available data. obese individuals achieve weight loss, this is significant.

Table 6. Changes in Food Craving Inventory (FCI) Scores, All Available Data (n = 60): Mixed Model
Estimates (95% CI)
VARIABLE BASELINE 6 WEEKS 12 WEEKS 6-WEEK CHANGE 12-WEEK CHANGE
FCI total score (average) 2.47 1.81 1.71 -0.66 -0.76
(2.35, 2.59) (1.67, 1.96) (1.60, 1.82) (-0.81 , -0.51 ) (-0.89, -0.63 )
p < 0.001 p < 0.001
FCI Subscales:
High fat 2.15 1.77 1.69 -0.38 -0.46
(1.99, 2.31 ) (1.62, 1.92 ) (1.56, 1.82 ) (-0.53 , -0.23 ) (-0.61 , -0.31 )
p < 0.001 p < 0.001
Sweets 2.64 1.66 1.66 -0.99 -0.99
(2.44, 2.85 ) (1.48, 1.84 ) (1.49, 1.83 ) (-1.19, -0.78 ) (-1.18, -0.79 )
p < 0.001 p < 0.001
Carbohydrate/starches 2.53 1.93 1.70 -0.60 -0.83
(2.37, 2.70 ) (1.76, 2.11 ) (1.57, 1.83 ) (-0.80, -0.41 ) (-1.01, -0.66 )
p < 0.001 p < 0.001
Fast food fats 2.71 2.08 1.98 -0.63 -0.74
(2.54, 2.89 ) (1.91, 2.26 ) (1.83, 2.12 ) (-0.81, -0.45 ) (-0.91, -0.57 )
p < 0.001 p < 0.001

172 Obesity and Weight Management August 2009

 100
P = 0.002
P = 0.082 Acknowledgments
■ Baseline
P < 0.001
■ 6 weeks P = 0.132 This research was supported by an unrestricted gift to
80 ■ 12 weeks
the Center of Human Nutrition at the University of Colo-
P = 0.030 rado Denver from USANA for translational research on
Percentage

60 overweight and obesity.

P = 0.779
40
Author Disclosure Statement
20 Toni McKinnon, Natalie Eich, Vasiliy Chernyshev, Tim
Wood, and John Cuomo are employees of USANA and
receive salary support from USANA. ■
0 Metabolic syndrome Waist BP >130/85 TG >150 Fasting glucose HDL
(3 or more >40 in male; mm Hg mg/dL >100 mg/dL <50 mg/dL male;
components) >35 in female <40 mg/dL female

References
Figure 1. Percentage of completers with metabolic syndrome and metabolic
1. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Fle-
syndrome criteria.
gal KM. Prevalence of overweight and obesity in the United States,
1999–2004. JAMA. 2006;295:1549–1555.
2. Pi-Sunyer FX. Health implications of obesity. Am J Clin Nutr.
Weight loss was not the only end point measured in this 1991;53(6 Suppl):1595S–1603S.
trial. On average, the 53 subjects who completed the study 3. Grundy SM, Brewer B, Cleeman JI, Smith SC, Lenfant C. Defini-
experienced improvements in many of the metabolic and tion of metabolic syndrome: Report of the National Heart, Lung, and
cardiovascular risk factors that were studied. Moreover, Blood Institute/American Heart Association Conference on Scientific
by the end of the study, the number of subjects who met Issues Related to Definition. Circulation. 2004;109:433–438.
the criteria for metabolic syndrome had dropped by one 4. Orchard TJ, Temprosa M, Goldberg R, Haffner S, Ratner R, Marc-
ovina S, Fowler S; Diabetes Prevention Program Research Group.
third, adding more support to the clinical significance of The effect of metformin and intensive lifestyle intervention on the
lifestyle modification in promoting overall health. metabolic syndrome: The Diabetes Prevention Program Random-
A weakness of the present study was the absence of ized Trial. Ann Intern Med. 2005;142:611–619.
a no-treatment control group. However, given the natu- 5. Wadden TA, Butryn ML, Wilson C. Lifestyle modification for the
ral course of development of chronic disease, we think management of obesity. Gastroenterology. 2007;132:2226–2238.
it is unlikely that a no-treatment control group would 6. Kushner RF. Anti-obesity drugs. Expert Opin Pharmacother.
have shown similar improvements. It will be important 2008;9:1339–1350.
to continue to evaluate this online program and to com- 7. Tate DF, Jackvony EH, Wing RR. A randomized trial comparing
pare it to a control group in the future. It will also be human e-mail counseling, computer-automated tailored counseling,
and no counseling in an Internet weight loss program. Arch Intern
important to evaluate the long-term sustainability of the
Med. 2006;166:1620–1625.
intervention. However, given the urgent need to develop
8. Matthews DR, Hosker JP, Rudenski AS, Naylor BA, Treacher DF,
effective interventions that can easily be delivered to Turner RC. Homeostasis model assessment: Insulin resistance and
large numbers of people with metabolic syndrome, these beta-cell function from fasting plasma glucose and insulin concen-
results are promising. trations in man. Diabetologia. 1985;28:412–419.
9. Stumvoll M, Mitrakou A, Pimenta W, Jenssen T, Yki-Jarvinen H,
Availability of programs and tools utilized in this Van Haeften T, Renn W, Gerich J. Use of the oral glucose toler-
research study ance test to assess insulin release and insulin sensitivity. Diabetes
Care. 2000;23:295–301.
The nutritional supplements (the USANA Essentials™)
meal replacement shakes and snack bars used in this study 10. White MA, Whisenhunt BL, Williamson DA, Greenway FL, Nete-
meyer RG. Development and validation of the food craving inven-
were provided by USANA Health Sciences, Inc. (Table tory. Obes Res. 2002;10:107–114.
1). To learn more about USANA's products, visit www.
11. Strand RD. Healthy For Life. Rapid City, SD: Real Life Press
usana.com. Products can be ordered online or by calling Printing, 2005.
a customer service representative at 888-950-9595. The 12. Rabovsky A; Cuomo J, Eich N. Measurement of plasma antioxidant
12-week Healthy for Life program used in this clinical reserve after supplementation with various antioxidants in healthy
trial is available at www.releasingfat.com/. subjects. Clin Chim Acta. 2006;371:55–60.

August 2009 Obesity and Weight Management 173