Weight Gain
Conditions
Brief summary
The Study of Novel Approaches for Prevention (SNAP) is randomized trial designed to test whether behavioral interventions based on self-regulation can prevent weight gain in young adults (18-35 years; body mass index (BMI) 21-30 kg/m2). Two different self-regulation interventions for weight gain prevention will be compared in this trial; one intervention will focus on making small, consistent, changes in eating and exercise behavior to prevent weight gain or reverse weight gain if it occurs, whereas the other will emphasize larger changes in eating and exercise that occur periodically, with a goal of producing weight loss and thereby providing a buffer against anticipated weight gains. The primary aim of the trial is to test whether the magnitude of weight gain from baseline across an average three-year follow-up differs across the three groups, with the hypotheses that weight gain will be greater in the Control group than in either intervention and greater in the Small Changes than Large Changes group. SNAP-E (Extension) will determine whether the effects of the intervention can be maintained over an additional 3 years (i.e. through a total of 6 years).
Detailed description
Young adults, aged 20-35 years, experience the greatest rate of weight gain, averaging 1-2 lbs/yr. Over time, this weight gain is associated with a worsening in cardiovascular disease risk factors and an increase in the prevalence of metabolic syndrome. Given the difficulties in producing sustained weight loss later in life, preventing weight gain from occurring during this critical period is key to curbing the obesity epidemic. The Study of Novel Approaches for Prevention (SNAP) is a 2-center randomized trial designed to test whether behavioral interventions based on self-regulation can prevent weight gain in young adults (18-35 years; body mass index (BMI) 21-30 kg/m2). Approximately 600 participants will be recruited over two years and randomly assigned to a control condition (N=200), self-regulation with small changes (N=200) or self-regulation with large changes (N=200). The Small Changes group will be taught to make small, consistent, changes in eating and exercise behavior to prevent weight gain or reverse weight gain if it occurs whereas the Large Changes group will emphasize periodic, larger changes in eating and exercise, with a goal of producing weight loss and thereby providing a buffer against anticipated weight gain. The primary aim of the trial is to test whether the magnitude of weight gain from baseline across an average planned follow-up of three years differs across the three groups, with a priori hypotheses that weight gain will be greater in the Control group than in either intervention and greater in the Small Changes than Large Changes group. Secondary aims are to compare the three groups on a) the proportion of participants in the three groups who gain less than 1 pound over the planned follow-up, b) the mean difference in weight gain from baseline to 24-month follow-up, c) the changes in behaviors and psychosocial measures (diet, physical activity, dietary restraint, frequency of self-weighing,depression, and occurrence of abnormal eating behaviors), and d) the changes in cardiovascular disease risk factors (blood pressure, lipids, insulin sensitivity, and waist circumference). The trial will also examine the association among changes in behaviors, weight, and cardiovascular disease risk factors and examine variables that may moderate the effects of the intervention (including gender, ethnicity, initial BMI, age) and potential mediators of the effects of the intervention (including changes in diet, activity, and self-regulatory behaviors). SNAP is member of the Early Adult Reduction in Weight Through Lifestyle Interventions (EARLY) consortium of clinical trials funded by the National Heart, Lung, and Blood Institute.
Interventions
BEHAVIORALSmall Behavior Changes
The Self-Regulation Plus Small Behavior Changes Intervention will focus on making small changes in diet and physical activity on a daily basis to prevent weight gain. Diet: The dietary approach used in this group is to identify small changes in what and how much participants eat each day. The general concept is that these are small, manageable changes that will produce small reductions in overall intake and can easily be made on a daily basis and maintained over time. Exercise: At the start of the program, participants will be given a pedometer and asked to record their current or baseline number of steps. They will then be given the goal of increasing their daily steps by 2000 steps per day over this baseline level.
BEHAVIORALLarge Behavior Changes
The focus of this intervention group will be on periodically making large changes in diet and physical activity, with the goal of losing 5-10 pounds to buffer against the weight gain that often occurs during young adulthood. Diet: Individuals with a BMI of 21-24.9 kg/m2 will be encouraged to lose 5 pounds; those with a BMI of 25-30 kg/m2 will be encouraged to lose 10 pounds. Exercise: The Large Changes group will be instructed to gradually increase their minutes of physical activity until achieving 250 minutes per week (5 days/week with 50 minutes per day) using activities similar in intensity to brisk walking.
Sponsors
National Heart, Lung, and Blood Institute (NHLBI)
Wake Forest University
Study design
Allocation
RANDOMIZED
Intervention model
PARALLEL
Primary purpose
PREVENTION
Masking
DOUBLE (Investigator, Outcomes Assessor)
Eligibility
Sex/Gender
ALL
Age
18 Years to 35 Years
Healthy volunteers
Yes
Inclusion criteria
1\. BMI of 21 - 30 kg/m2
Exclusion criteria
1. Untreated hypertension, hyperlipidemia, or type 2 diabetes, unless permission is provided by their health care provider. 2. Heart disease, heart problems or report being prescribed drugs for blood pressure or a major heart condition, unless permission is provided by their health care provider. 3. Type 1 diabetes or treatment of type 2 diabetes with insulin or oral medication that may cause hypoglycemia. 4. Health problems which may influence the ability to walk for physical activity (e.g. lower limb amputation) or other reasons why a person should not do physical activity, unless permission is provided from their health care provider. 5. Health problems that may be associated with unintentional weight change or affect the safety of a weight loss program, including report of a heart attack or stroke, chest pain during periods of activity or rest, loss of consciousness, active tuberculosis, HIV, chronic hepatitis B or C, inflammatory bowel disease requiring treatment within the past year, thyroid disease, renal disease, liver disease, hospitalization for asthma in the past year, or cancer within the past 5 years (except for non-melanoma skin cancers or early stage cervical cancer) or chronic use of steroid medication. 6. Report of a past diagnosis of or treatment for a Diagnostic and Statistical Manual of Mental Disorders-Fourth Edition-Text Revision (DSM-IV-TR) eating disorder (anorexia nervosa or bulimia nervosa) or meet criteria for anorexia or bulimia nervosa during screening for this trial 7. Report of a past diagnosis of or current symptoms of alcohol or substance dependence. 8. Currently pregnant, pregnant within the past 6 months, or planning to become pregnant within the next 6 months. 9. History of schizophrenia, manic depression, or bipolar disorder. 10. Hospitalization for depression or other psychiatric disorder within the past 12 months. 11. Having lost and maintained a weight loss of 10 pounds or more within the past 6 months or are currently participating in a weight loss program, trying to gain weight, using steroids for muscle mass or weight gain, taking weight loss medication, or have had surgery for weight loss. 12. Participation in another weight loss or physical activity study that would interfere with this study. 13. Another member of the household (or roommate) is a participant or staff member on this trial. 14. Reason to suspect that the participant would not adhere to the study intervention or assessment schedule (i.e., can't come to group on a regular basis; will be away for more than two weeks during initial intervention phase or planning to move from the area within next year). 15. Not able to speak and understand English. 16. Residence or place of work further than 30 miles from the intervention site. 17. Perceived inability to attend the 2 year data collection visit. 18. Does not have Internet access on a regular basis.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Weight Changes From Baseline Over Follow-up. | 3 years | Mean weight change from baseline across an average planned follow-up of three years. These mean changes will be compared among the three arms of the trial. |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Mean Changes From Baseline to 2 Years in Total Cholesterol | 2 years | Mean changes from baseline to 2 years in total cholesterol among participants with Year 2 measurements (mg/dl) |
| Weight Gain 1 Pound or More at Any Time Over Follow-up | 3 years | Average over time (average follow-up of 3 years) of the percent of participants within each arm of the trial who gain 1 pound or more at each visit. These percentages will be compared among the three arms generalized estimating equations. Note that weight changes in units of pounds were used to define this outcome so that it may be more clear to participants. Elsewhere in the protocol, weight is reported in kilograms. Percentages at each visit are the percent who gained 1 pound or more from baseline among all who were weighed at that visit. Participants were assigned values of 0 or 1 at each visit depending on their weight gain status. |
| Mean Weight Changes | 2 years | Mean differences in weight changes among intervention groups at 24 months post-randomization |
| Mean Changes in Systolic Blood Pressure | Measured at 2 Years | Compare changes in systolic blood pressure across the three intervention groups |
| Mean Changes From Baseline in Diastolic Blood Pressure | 2 years | Change from baseline to 2 years in diastolic blood pressure |
| Obesity | 3 years | Percentage of those participants whose body mass index at baseline was less than 30 kg/m2 who subsequently transitioned to a body mass index of 30 kg/m2 or more (i.e. met criteria for obesity) sometime during 3 years of follow-up (i.e. at least one visit). Percentages will be compared among the three arms of the trial and summarized with odds ratios Participants were assigned values of 0 or 1 at each exam depending on their obesity level. Inference is based on generalized estimating equations. |
| Dietary Restraint: Mean Change From Baseline to 2 Years | 2 years | The Eating Inventory (Stunkard, 1988) is a 51-item self-report instrument, was used to assess the subscale of dietary restraint (e.g., degree of conscious control exerted over eating behaviors; range from 0-21 with higher scores reflecting greater levels of restraint). Reference: Stunkard, A. J. M., S. (1988). Eating Inventory Manual. New York: Psychological Corporation. |
| Disinhibition | Changes at 2 years | The Eating Inventory (TFEQ(Stunkard, 1988), a 51-item self-report instrument, was used to assess the subscale of disinhibition (e.g., susceptibility to loss of control over eating; range 0-16, with higher scores reflecting greater levels of disinhibition). REF: Stunkard, A. J. M., S. (1988). Eating Inventory Manual. New York: Psychological Corporation. |
| Flexible Dietary Control | Changes from baseline to 2 years | Flexible control is characterized by a balanced approach to eating (e.g., taking smaller portions to control weight, engaging in healthy compensation) and is associated with better weight management outcomes (Westenhoefer, Stunkard, & Pudel, 1999). Scores range from 0 to 12 with higher scores reflecting greater levels of flexible control. REF: Westenhoefer, J., Stunkard, A. J., & Pudel, V. (1999). Validation of the flexible and rigid control dimensions of dietary restraint. Int J Eat Disord, 26(1), 53-64. |
| Rigid Dietary Control | Changes from baseline to 2 years | Rigid control is characterized by an all-or-nothing inflexibility around dietary rules (e.g., strict calorie counting, with guilt following if calorie-dense foods are consumed) that is associated with poor weight outcomes and more binge eating (Westenhoefer, Stunkard, & Pudel, 1999). Scores range from 0 to 16 with higher scores reflecting greater rigid control. REF: Stunkard, A. J. M., S. (1988). Eating Inventory Manual. New York: Psychological Corporation. |
| Mean Changes in High Density Lipoprotein Cholesterol (HDL-C) | 2 years | Mean changes in HDL-C from baseline to year 2 in (mg/dl) for compared among the 3 arms using analysis of variance |
| Mean Changes in Low Density Lipoprotein Cholesterol (LDL-C) | 2 years | Mean changes between baseline and 2 years in low density lipoprotein cholesterol: LDL-c (mg/dl) |
| Mean Change in Fasting Glucose From Baseline to 2 Years | 2 years | Mean change in fasting glucose from baseline to 2 years in mg/dl for all participants with year 2 measures |
| Mean Change in Fasting Insulin From Baseline to 2 Years | 2 years | Mean change in fasting insulin (uU/ml) from baseline to 2 years |
| Depression Symptomatology | 2 years | Mean changes in the Center for Epidemiologic Studies Depression (CES-C) Scale. Reference: Turvey, C. L., Wallace, R. B., & Herzog, R. (1999). A revised CES-D measure of depressive symptoms and a DSM-based measure of major depressive episodes in the elderly. Int Psychogeriatr, 11(2), 139-148. 20 item questionnaire with a possible range of scores is zero to 60, and higher scores indicating the presence of more symptomatology. |
| Insulin Resistance | Change from baseline to 2 years | We calculated homeostatic model assessment insulin resistance (HOMA-IR): fasting glucose in (mg/dl) \* fasting insulin in (uU/mL). |
| Total Energy Dietary Intake Per Day (Kcals) | Changes from baseline to 2 years in kilocalories | Dietary intake was assessed using the 2005 Block Food Frequency Questionnaire (Block FFQ) at baseline and 2 years. This validated, quantitative 110-food item questionnaire is designed to assess relative intake of energy. REF: Block G, Woods M, Potosky A, Clifford C. Validation of a self-administered diet history questionnaire using multiple diet records. J Clin Epidemiol 1990; 43:1327-1335. |
| Change in Waist Circumference (cm) | Change from baseline to 2 years | Waist circumference will be measured using a Gulik tape measure and following a standardized protocol. Two measures of waist circumference will be taken; if the difference exceeds 1.0 cm, a third measure will be taken. Changes are measured from baseline to year 2. |
| Self-weighing | 2 years | Number of days per week the participant reports weighing themselves. This is divided into two groups: 1) more than once per week and 2) no more than once per week |
| General Health Index | Changes from baseline to 2 years | The General Health Index, a one-item question from the CDC's Health-Related Quality of Life measure (Measuring Healthy Days, 2000) required participants to report whether in general their health is excellent (1), very good (2), good (3), fair (4), or poor (5). Lower scores denotes better outcomes. Ref: Measuring Healthy Days. (2000). Atlanta, Georgia: Centers for Disease Control and Prevention |
Other
| Measure | Time frame | Description |
|---|---|---|
| 6 Year Weight Changes | 6 years | Changes from baseline to year 6 in body weight |
Countries
United States
Participant flow
Recruitment details
SNAP targeted an enrollment of 600 participants, aged 18-35 years, with a BMI of 21.0 to 30.9. Participants were recruited from 2 clinical sites (Providence, RI and Chapel Hill, NC). The final sample of 599 participants (27% minority, 22% male) was recruited over a 19-month period (Aug 2010-Feb 2012).
Pre-assignment details
Participants were pre-screened online and by telephone before attending an orientation, followed by two screening visits. 609 participants were randomized; although 10 participants never attended the randomization visit and did not learn of their group assignment.
Participants by arm
| Arm | Count |
|---|---|
| Control Group Self-Guided Behavior Changes. This group will be used to determine average rate of weight gain over 3 years with little intervention. Participants randomized to this group will receive one face-to-face session that will provide general education of self-weighing and information about both the small and large changes approach. Participants will also be provided with quarterly newsletters describing study events and very limited information on health eating. | 202 |
| Self-regulation With Small Behavior Changes Participants randomized to this group will be taught to self-weigh daily and report weights regularly. They will be reinforced for maintaining weight below randomization weight and taught that if weight exceeds randomization weight, they should reinstate small changes (e.g., modify diet by approximately 100 kcal, decrease portion sizes or change types of food, increase activity by 2000 steps/day).
Self-regulation theory: Participants randomized to either of these two groups will be taught to self-weigh daily and report weights regularly. They will be reinforced for maintaining weight below randomization weight and taught that if weight exceeds randomization weight, they should reinstate either large or small changes. | 200 |
| Self-regulation With Large Behavior Changes Participants randomized to this group will be taught to self-weigh daily and report weights regularly. They will be reinforced for maintaining weight below randomization weight and taught that if weight exceeds randomization weight, they should reinstate large changes (e.g., modify diet to 1200-1500 or 1500-1800 kcal/day with \< 30% fat, increase exercise to 250 minutes/week of moderate intensity activity).
Self-regulation theory: Participants randomized to either of these two groups will be taught to self-weigh daily and report weights regularly. They will be reinforced for maintaining weight below randomization weight and taught that if weight exceeds randomization weight, they should reinstate either large or small changes. | 197 |
| Total | 599 |
Baseline characteristics
| Characteristic | Total | Self-regulation With Large Behavior Changes | Self-regulation With Small Behavior Changes | Control Group |
|---|---|---|---|---|
| Age, Customized Age Age : 18-24 years | 169 Participants | 56 Participants | 60 Participants | 53 Participants |
| Age, Customized Age Age : 25-35 years | 430 Participants | 141 Participants | 140 Participants | 149 Participants |
| BMI <25.0 kg/m2 | 277 Participants | 96 Participants | 84 Participants | 97 Participants |
| BMI >=25.0 kg/m2 | 322 Participants | 101 Participants | 116 Participants | 105 Participants |
| Education College graduate | 479 Participants | 162 Participants | 154 Participants | 163 Participants |
| Education Not college graduate | 120 Participants | 35 Participants | 46 Participants | 39 Participants |
| Employment Status Employed full time | 377 Participants | 125 Participants | 125 Participants | 127 Participants |
| Employment Status Other | 57 Participants | 17 Participants | 14 Participants | 26 Participants |
| Employment Status Student full time | 165 Participants | 55 Participants | 61 Participants | 49 Participants |
| Race/Ethnicity, Customized Race/Ethnicity African American | 66 Participants | 22 Participants | 25 Participants | 19 Participants |
| Race/Ethnicity, Customized Race/Ethnicity Asian/Pacific Islander | 25 Participants | 7 Participants | 11 Participants | 7 Participants |
| Race/Ethnicity, Customized Race/Ethnicity Hispanic | 46 Participants | 21 Participants | 10 Participants | 15 Participants |
| Race/Ethnicity, Customized Race/Ethnicity Multiple/Refused | 24 Participants | 3 Participants | 8 Participants | 13 Participants |
| Race/Ethnicity, Customized Race/Ethnicity Non-Hispanic white | 438 Participants | 144 Participants | 146 Participants | 148 Participants |
| Sex: Female, Male Female | 469 Participants | 154 Participants | 157 Participants | 158 Participants |
| Sex: Female, Male Male | 130 Participants | 43 Participants | 43 Participants | 44 Participants |
| Weight | 71.4 kilograms STANDARD_DEVIATION 10.7 | 70.8 kilograms STANDARD_DEVIATION 11 | 71.9 kilograms STANDARD_DEVIATION 11 | 71.4 kilograms STANDARD_DEVIATION 10.2 |
Adverse events
| Event type | EG000 affected / at risk | EG001 affected / at risk | EG002 affected / at risk |
|---|---|---|---|
| deaths Total, all-cause mortality | — / — | — / — | — / — |
| other Total, other adverse events | 119 / 202 | 118 / 200 | 109 / 197 |
| serious Total, serious adverse events | 4 / 202 | 11 / 200 | 9 / 197 |
Outcome results
Primary
Weight Changes From Baseline Over Follow-up.
Mean weight change from baseline across an average planned follow-up of three years. These mean changes will be compared among the three arms of the trial.
Time frame: 3 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Weight Changes From Baseline Over Follow-up. | 0.26 kilograms | Standard Error 0.22 |
| Self-regulation With Small Behavior Changes | Weight Changes From Baseline Over Follow-up. | -0.56 kilograms | Standard Error 0.22 |
| Self-regulation With Large Behavior Changes | Weight Changes From Baseline Over Follow-up. | -2.37 kilograms | Standard Error 0.22 |
Comparison: Mean changes from random effects model applied to repeated measures to compute the average differences among groups over time.p-value: <0.0595% CI: [0.23, 1.41]Mixed Models Analysis
Comparison: Mean changes from a random effects model applied to repeated measures to compute the average differences between groups over time. This entry is for the comparison between groups 1 and 3.p-value: <0.0595% CI: [2.05, 3.22]Mixed Models Analysis
Secondary
Change in Waist Circumference (cm)
Waist circumference will be measured using a Gulik tape measure and following a standardized protocol. Two measures of waist circumference will be taken; if the difference exceeds 1.0 cm, a third measure will be taken. Changes are measured from baseline to year 2.
Time frame: Change from baseline to 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Change in Waist Circumference (cm) | 0.21 centimeters | Standard Error 0.46 |
| Self-regulation With Small Behavior Changes | Change in Waist Circumference (cm) | -1.06 centimeters | Standard Error 0.47 |
| Self-regulation With Large Behavior Changes | Change in Waist Circumference (cm) | -2.21 centimeters | Standard Error 0.46 |
Comparison: 2 degree of freedom F-test from ANOVA to compare differences among 3 armsp-value: 0.00195% CI: [-2.55, 0.02]ANOVA
Comparison: Mean change in waist girth from baseline to year 2p-value: 0.00195% CI: [-3.69, -1.14]ANOVA
Secondary
Depression Symptomatology
Mean changes in the Center for Epidemiologic Studies Depression (CES-C) Scale. Reference: Turvey, C. L., Wallace, R. B., & Herzog, R. (1999). A revised CES-D measure of depressive symptoms and a DSM-based measure of major depressive episodes in the elderly. Int Psychogeriatr, 11(2), 139-148. 20 item questionnaire with a possible range of scores is zero to 60, and higher scores indicating the presence of more symptomatology.
Time frame: 2 years
Population: All Participants with 2 year data
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Depression Symptomatology | 1.68 units on a scale | Standard Error 0.5 |
| Self-regulation With Small Behavior Changes | Depression Symptomatology | 0.60 units on a scale | Standard Error 0.51 |
| Self-regulation With Large Behavior Changes | Depression Symptomatology | 1.73 units on a scale | Standard Error 0.51 |
Comparison: Mean differences from baseline to year 2 among the 3 arms were assessed using analysis of variance.p-value: 0.295% CI: [-0.84, 3]ANOVA
Comparison: Analysis of variance was used to compare differences among the 3 groups.p-value: 0.295% CI: [-1.45, 1.35]ANOVA
Secondary
Dietary Restraint: Mean Change From Baseline to 2 Years
The Eating Inventory (Stunkard, 1988) is a 51-item self-report instrument, was used to assess the subscale of dietary restraint (e.g., degree of conscious control exerted over eating behaviors; range from 0-21 with higher scores reflecting greater levels of restraint). Reference: Stunkard, A. J. M., S. (1988). Eating Inventory Manual. New York: Psychological Corporation.
Time frame: 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Dietary Restraint: Mean Change From Baseline to 2 Years | 0.39 Units on a scale | Standard Error 0.18 |
| Self-regulation With Small Behavior Changes | Dietary Restraint: Mean Change From Baseline to 2 Years | 0.59 Units on a scale | Standard Error 0.19 |
| Self-regulation With Large Behavior Changes | Dietary Restraint: Mean Change From Baseline to 2 Years | 1.27 Units on a scale | Standard Error 0.19 |
Comparison: Mean changes from baseline to 2 years among the 3 groups were assessed using a 2 degree of freedom F-testp-value: 0.00295% CI: [-0.61, 0.45]ANOVA
Comparison: Differences among the 3 groups were based on analyses of variance.p-value: 0.00295% CI: [0.02, 1.08]ANOVA
Secondary
Disinhibition
The Eating Inventory (TFEQ(Stunkard, 1988), a 51-item self-report instrument, was used to assess the subscale of disinhibition (e.g., susceptibility to loss of control over eating; range 0-16, with higher scores reflecting greater levels of disinhibition). REF: Stunkard, A. J. M., S. (1988). Eating Inventory Manual. New York: Psychological Corporation.
Time frame: Changes at 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Disinhibition | 0.07 Units on a scale | Standard Error 0.17 |
| Self-regulation With Small Behavior Changes | Disinhibition | -0.78 Units on a scale | Standard Error 0.17 |
| Self-regulation With Large Behavior Changes | Disinhibition | -0.03 Units on a scale | Standard Error 0.17 |
Comparison: Changes in units of the scale from baseline to 2 yearsp-value: <0.00195% CI: [-1.32, -0.38]ANOVA
Comparison: Analysis of variance was used to compare mean changes from baseline to 2 years among the 3 groups.p-value: <0.00195% CI: [-0.57, 0.37]ANOVA
Secondary
Flexible Dietary Control
Flexible control is characterized by a balanced approach to eating (e.g., taking smaller portions to control weight, engaging in healthy compensation) and is associated with better weight management outcomes (Westenhoefer, Stunkard, & Pudel, 1999). Scores range from 0 to 12 with higher scores reflecting greater levels of flexible control. REF: Westenhoefer, J., Stunkard, A. J., & Pudel, V. (1999). Validation of the flexible and rigid control dimensions of dietary restraint. Int J Eat Disord, 26(1), 53-64.
Time frame: Changes from baseline to 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Flexible Dietary Control | 0.39 units on a scale | Standard Error 0.18 |
| Self-regulation With Small Behavior Changes | Flexible Dietary Control | 0.59 units on a scale | Standard Error 0.19 |
| Self-regulation With Large Behavior Changes | Flexible Dietary Control | 1.27 units on a scale | Standard Error 0.19 |
Comparison: 2 degree of freedom F-test from analysis of variance to compare 2 year differences from baseline among the 3 arms of the studyp-value: <0.00195% CI: [-0.31, 0.71]ANOVA
Comparison: Analysis of variance was used to compare mean changes from baseline to year 2 among the 3 groups.p-value: 0.00295% CI: [0.37, 1.39]ANOVA
Secondary
General Health Index
The General Health Index, a one-item question from the CDC's Health-Related Quality of Life measure (Measuring Healthy Days, 2000) required participants to report whether in general their health is excellent (1), very good (2), good (3), fair (4), or poor (5). Lower scores denotes better outcomes. Ref: Measuring Healthy Days. (2000). Atlanta, Georgia: Centers for Disease Control and Prevention
Time frame: Changes from baseline to 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | General Health Index | 0.04 2 year changes in units on a scale | Standard Error 0.06 |
| Self-regulation With Small Behavior Changes | General Health Index | -0.04 2 year changes in units on a scale | Standard Error 0.06 |
| Self-regulation With Large Behavior Changes | General Health Index | -0.11 2 year changes in units on a scale | Standard Error 0.06 |
Comparison: 2 degree of freedom F-test from analysis of variance to compare 2 year differences in general health index values among the 3 armsp-value: 0.2495% CI: [-0.24, 0.09]ANOVA
Comparison: Analysis of variance was used to compare differences in changes from baseline to 2 years among the three arms.p-value: 0.2495% CI: [-0.32, 0.02]ANOVA
Secondary
Insulin Resistance
We calculated homeostatic model assessment insulin resistance (HOMA-IR): fasting glucose in (mg/dl) \* fasting insulin in (uU/mL).
Time frame: Change from baseline to 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Insulin Resistance | -0.03 HOMA-IR | Standard Error 0.08 |
| Self-regulation With Small Behavior Changes | Insulin Resistance | -0.15 HOMA-IR | Standard Error 0.08 |
| Self-regulation With Large Behavior Changes | Insulin Resistance | -0.33 HOMA-IR | Standard Error 0.08 |
Comparison: 2 degree of freedom F-test to compare mean 2 year differences among 3 armsp-value: 0.0295% CI: [-0.34, 0.1]ANOVA
Comparison: Analysis of variance was used to compare mean differences in changes from baseline to 2 years among the 3 groups.p-value: 0.0295% CI: [-0.52, -0.08]ANOVA
Secondary
Mean Change in Fasting Glucose From Baseline to 2 Years
Mean change in fasting glucose from baseline to 2 years in mg/dl for all participants with year 2 measures
Time frame: 2 years
Population: All participants with measurements at 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Change in Fasting Glucose From Baseline to 2 Years | 1.48 Mean change from baseline in mg/dl | Standard Error 0.49 |
| Self-regulation With Small Behavior Changes | Mean Change in Fasting Glucose From Baseline to 2 Years | 0.40 Mean change from baseline in mg/dl | Standard Error 0.49 |
| Self-regulation With Large Behavior Changes | Mean Change in Fasting Glucose From Baseline to 2 Years | -0.18 Mean change from baseline in mg/dl | Standard Error 0.48 |
Comparison: Mean changes from baseline to year 2 among the 3 arms were compared using analysis of variance.p-value: 0.0595% CI: [-2.44, 0.28]ANOVA
Comparison: Analysis of variance was used to compare mean differences in changes from baseline to 2 years among the 3 groups.p-value: 0.0595% CI: [-3, -0.32]ANOVA
Secondary
Mean Change in Fasting Insulin From Baseline to 2 Years
Mean change in fasting insulin (uU/ml) from baseline to 2 years
Time frame: 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Change in Fasting Insulin From Baseline to 2 Years | -0.27 Mean changes at 2 years in uU/ml | Standard Error 0.33 |
| Self-regulation With Small Behavior Changes | Mean Change in Fasting Insulin From Baseline to 2 Years | -0.73 Mean changes at 2 years in uU/ml | Standard Error 0.33 |
| Self-regulation With Large Behavior Changes | Mean Change in Fasting Insulin From Baseline to 2 Years | -1.48 Mean changes at 2 years in uU/ml | Standard Error 0.32 |
Comparison: Mean changes between baseline and year 2 were compared among the 3 arms using analysis of variance.p-value: 0.0395% CI: [-1.37, 0.45]ANOVA
Comparison: Analysis of variance was used to compare mean differences from baseline to 2 years among the 3 groups.p-value: 0.0395% CI: [-2.11, -0.3]ANOVA
Secondary
Mean Changes From Baseline in Diastolic Blood Pressure
Change from baseline to 2 years in diastolic blood pressure
Time frame: 2 years
Population: All participants providing data at 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Changes From Baseline in Diastolic Blood Pressure | -0.41 mmHg | Standard Error 0.51 |
| Self-regulation With Small Behavior Changes | Mean Changes From Baseline in Diastolic Blood Pressure | -2.14 mmHg | Standard Error 0.52 |
| Self-regulation With Large Behavior Changes | Mean Changes From Baseline in Diastolic Blood Pressure | -1.33 mmHg | Standard Error 0.51 |
Comparison: Mean differences from baseline to 2 years were compared among the 3 arms using analysis of variance.p-value: 0.0695% CI: [0.32, 3.14]ANOVA
Comparison: Analysis of variance was used to compare mean changes from baseline to 2 years in diastolic blood pressure.p-value: 0.0695% CI: [-0.49, 2.33]ANOVA
Secondary
Mean Changes From Baseline to 2 Years in Total Cholesterol
Mean changes from baseline to 2 years in total cholesterol among participants with Year 2 measurements (mg/dl)
Time frame: 2 years
Population: All participants with Year 2 measures
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Changes From Baseline to 2 Years in Total Cholesterol | -0.56 mg/dl | Standard Error 1.75 |
| Self-regulation With Small Behavior Changes | Mean Changes From Baseline to 2 Years in Total Cholesterol | 0.74 mg/dl | Standard Error 1.73 |
| Self-regulation With Large Behavior Changes | Mean Changes From Baseline to 2 Years in Total Cholesterol | 1.33 mg/dl | Standard Error 1.75 |
Comparison: Mean changes from baseline to 2 years were compared between Groups 1 and 2.p-value: 0.7395% CI: [-6.12, 3.52]ANOVA
Comparison: Analysis of variance was used to compare mean differences in total cholesterol changes from baseline among groups.p-value: 0.7395% CI: [-4.36, 0.58]ANOVA
Secondary
Mean Changes in High Density Lipoprotein Cholesterol (HDL-C)
Mean changes in HDL-C from baseline to year 2 in (mg/dl) for compared among the 3 arms using analysis of variance
Time frame: 2 years
Population: Mean changes among all participants with measurements at Year 2
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Changes in High Density Lipoprotein Cholesterol (HDL-C) | -0.09 Mean Change from Baseline in mg/dl | Standard Error 0.83 |
| Self-regulation With Small Behavior Changes | Mean Changes in High Density Lipoprotein Cholesterol (HDL-C) | 1.43 Mean Change from Baseline in mg/dl | Standard Error 0.82 |
| Self-regulation With Large Behavior Changes | Mean Changes in High Density Lipoprotein Cholesterol (HDL-C) | 2.12 Mean Change from Baseline in mg/dl | Standard Error 0.83 |
Comparison: Mean differences between baseline and year 2 were compared among the 3 arms using analysis of variance.p-value: 0.1695% CI: [-0.77, 3.81]ANOVA
Comparison: Analysis of variance was used to compared mean differences among the 3 groups.p-value: 0.1695% CI: [-0.09, 4.51]ANOVA
Secondary
Mean Changes in Low Density Lipoprotein Cholesterol (LDL-C)
Mean changes between baseline and 2 years in low density lipoprotein cholesterol: LDL-c (mg/dl)
Time frame: 2 years
Population: Mean changes from baseline among all participants with year 2 measures
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Changes in Low Density Lipoprotein Cholesterol (LDL-C) | -0.01 Mean change in LDL-C in mg/dl | Standard Error 1.46 |
| Self-regulation With Small Behavior Changes | Mean Changes in Low Density Lipoprotein Cholesterol (LDL-C) | 0.16 Mean change in LDL-C in mg/dl | Standard Error 1.44 |
| Self-regulation With Large Behavior Changes | Mean Changes in Low Density Lipoprotein Cholesterol (LDL-C) | 1.43 Mean change in LDL-C in mg/dl | Standard Error 1.46 |
Comparison: Mean changes from baseline to year 2 among the 3 arms were compared using analysis of variance.p-value: 0.7595% CI: [-3.89, 4.23]ANOVA
Comparison: Analysis of variance was used to compare mean changes among the 3 groups.p-value: 0.7595% CI: [-2.61, 5.49]ANOVA
Secondary
Mean Changes in Systolic Blood Pressure
Compare changes in systolic blood pressure across the three intervention groups
Time frame: Measured at 2 Years
Population: All participants with measurements at Year 2
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Changes in Systolic Blood Pressure | 1.73 mmHg | Standard Error 0.69 |
| Self-regulation With Small Behavior Changes | Mean Changes in Systolic Blood Pressure | -3.72 mmHg | Standard Error 0.7 |
| Self-regulation With Large Behavior Changes | Mean Changes in Systolic Blood Pressure | -2.66 mmHg | Standard Error 0.69 |
Comparison: Analysis of variance to assess mean differences in changes from baseline in systolic blood pressure. This analysis compares groups 1 and 2.p-value: 0.1395% CI: [0.06, 3.92]ANOVA
Comparison: Mean differences in systolic blood pressure between groups 1 and 3 over time.p-value: 0.1395% CI: [-0.98, 2.84]ANOVA
Secondary
Mean Weight Changes
Mean differences in weight changes among intervention groups at 24 months post-randomization
Time frame: 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Mean Weight Changes | 0.54 kilograms | Standard Error 0.33 |
| Self-regulation With Small Behavior Changes | Mean Weight Changes | -0.77 kilograms | Standard Error 0.33 |
| Self-regulation With Large Behavior Changes | Mean Weight Changes | -1.50 kilograms | Standard Error 0.34 |
Comparison: Mean differences at 2 years are calculated from a linear contrast within a mixed effects model. Note that this comparison is between Groups 1 and 2.p-value: <0.0595% CI: [0.39, 2.24]Mixed Models Analysis
Comparison: A linear contrast from a mixed effects model was used to compare mean differences at 24 months between groups 1 and 3.p-value: <0.0595% CI: [1.11, 2.98]Mixed Models Analysis
Secondary
Obesity
Percentage of those participants whose body mass index at baseline was less than 30 kg/m2 who subsequently transitioned to a body mass index of 30 kg/m2 or more (i.e. met criteria for obesity) sometime during 3 years of follow-up (i.e. at least one visit). Percentages will be compared among the three arms of the trial and summarized with odds ratios Participants were assigned values of 0 or 1 at each exam depending on their obesity level. Inference is based on generalized estimating equations.
Time frame: 3 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Obesity | 16.9 percentage of participants | Standard Error 2.7 |
| Self-regulation With Small Behavior Changes | Obesity | 7.9 percentage of participants | Standard Error 2 |
| Self-regulation With Large Behavior Changes | Obesity | 8.6 percentage of participants | Standard Error 2 |
Comparison: The average percentage of participants who were obese across follow-up were compared among the 3 arms using a generalized estimating equations approach for repeated binary measures. Participants were assigned values of 0 or 1 depending on their obesity status at each exam. Differences were summarized with odds ratios.p-value: <0.00195% CI: [1.23, 4.52]generalized estimating equations
Comparison: Generalized estimating equations were used. Differences were summarized with odds ratios.p-value: 0.00895% CI: [1.12, 4.1]generalized estimating equations
Secondary
Rigid Dietary Control
Rigid control is characterized by an all-or-nothing inflexibility around dietary rules (e.g., strict calorie counting, with guilt following if calorie-dense foods are consumed) that is associated with poor weight outcomes and more binge eating (Westenhoefer, Stunkard, & Pudel, 1999). Scores range from 0 to 16 with higher scores reflecting greater rigid control. REF: Stunkard, A. J. M., S. (1988). Eating Inventory Manual. New York: Psychological Corporation.
Time frame: Changes from baseline to 2 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Rigid Dietary Control | 0.35 2 year changes in units on a scale | Standard Error 0.21 |
| Self-regulation With Small Behavior Changes | Rigid Dietary Control | 0.22 2 year changes in units on a scale | Standard Error 0.21 |
| Self-regulation With Large Behavior Changes | Rigid Dietary Control | 1.75 2 year changes in units on a scale | Standard Error 0.21 |
Comparison: 2 degree F-test from analysis of variance applied to 2 year changes in scores from baselinep-value: <0.00195% CI: [-0.71, 0.45]ANOVA
Comparison: Analysis of variance was used to compare mean changes from baseline to 2 years among the 3 groups.p-value: <0.00195% CI: [0.82, 1.98]ANOVA
Secondary
Self-weighing
Number of days per week the participant reports weighing themselves. This is divided into two groups: 1) more than once per week and 2) no more than once per week
Time frame: 2 years
| Arm | Measure | Value (COUNT_OF_PARTICIPANTS) |
|---|---|---|
| Control Group | Self-weighing | 30 Participants |
| Self-regulation With Small Behavior Changes | Self-weighing | 44 Participants |
| Self-regulation With Large Behavior Changes | Self-weighing | 56 Participants |
Comparison: Logistic regression to compare differences among the 3 groupsp-value: 0.00495% CI: [1.04, 3.02]Regression, Logistic
Comparison: Logistic regression analysis was used to compare the rates of daily self-weighing at 2 years among the 3 groups.p-value: 0.00495% CI: [1.4, 3.94]Regression, Logistic
Secondary
Total Energy Dietary Intake Per Day (Kcals)
Dietary intake was assessed using the 2005 Block Food Frequency Questionnaire (Block FFQ) at baseline and 2 years. This validated, quantitative 110-food item questionnaire is designed to assess relative intake of energy. REF: Block G, Woods M, Potosky A, Clifford C. Validation of a self-administered diet history questionnaire using multiple diet records. J Clin Epidemiol 1990; 43:1327-1335.
Time frame: Changes from baseline to 2 years in kilocalories
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Total Energy Dietary Intake Per Day (Kcals) | -134 kilocalories | Standard Error 40 |
| Self-regulation With Small Behavior Changes | Total Energy Dietary Intake Per Day (Kcals) | -186 kilocalories | Standard Error 41 |
| Self-regulation With Large Behavior Changes | Total Energy Dietary Intake Per Day (Kcals) | -185 kilocalories | Standard Error 41 |
Comparison: 2 degree of freedom F-test to compare mean differences among 3 arms in changes in kilocalories from baseline to year 2p-value: 0.5895% CI: [-164, 60]ANOVA
Comparison: Analysis of variance was used to assess mean differences in 2 year changes in kilocalories intake among the 3 groups.p-value: 0.5895% CI: [-162, 61]ANOVA
Secondary
Weight Gain 1 Pound or More at Any Time Over Follow-up
Average over time (average follow-up of 3 years) of the percent of participants within each arm of the trial who gain 1 pound or more at each visit. These percentages will be compared among the three arms generalized estimating equations. Note that weight changes in units of pounds were used to define this outcome so that it may be more clear to participants. Elsewhere in the protocol, weight is reported in kilograms. Percentages at each visit are the percent who gained 1 pound or more from baseline among all who were weighed at that visit. Participants were assigned values of 0 or 1 at each visit depending on their weight gain status.
Time frame: 3 years
| Arm | Measure | Value (MEAN) | Dispersion |
|---|---|---|---|
| Control Group | Weight Gain 1 Pound or More at Any Time Over Follow-up | 40.8 percentage gaining 1 pound or more among | Standard Error 4.4 |
| Self-regulation With Small Behavior Changes | Weight Gain 1 Pound or More at Any Time Over Follow-up | 32.5 percentage gaining 1 pound or more among | Standard Error 3.8 |
| Self-regulation With Large Behavior Changes | Weight Gain 1 Pound or More at Any Time Over Follow-up | 23.6 percentage gaining 1 pound or more among | Standard Error 2.8 |
Comparison: Generalized estimating equations were used to compare the average percent of weight gainers over time among the three groups. The null hypothesis was that there was no difference in these average percentages. Participants were assigned values of 0 or 1 at each visit depending on their weight gain status. The percentages were summarized with odds ratios for weight gain over time.p-value: <0.0595% CI: [1.02, 1.9]Generalized Estimating Equations
Comparison: This is a parallel analysis, comparing groups 1 and 3, using generalized estimating equations to summarize the odds ratio for weight gain in group 1 versus group 3,p-value: <0.0595% CI: [1.64, 3.19]generalized estimating equations
Other Pre-specified
6 Year Weight Changes
Changes from baseline to year 6 in body weight
Time frame: 6 years