Comparison of weight loss induced by daily caloric restriction versus intermittent fasting (DRIFT) in individuals with obesity: study protocol for a 52-week randomized clinical trial

Study setting

Healthy men and women with overweight or obesity will be recruited from the region around Denver, Colorado. All participants will be recruited and studied at a single site, the University of Colorado Anschutz Medical Campus (CU-AMC).

Eligibility criteria

Participants will be included in this trial if they are age 18–60 years, have a body mass index (BMI) of 27–46 kg/m2, self-report <150 min/week of voluntary exercise at moderate intensity or greater, and <60 min/day of total habitual PA (i.e., work and/or transportation related) at moderate intensity or greater over the past 13 weeks, and if they live or work within 30 min of CU-AMC. Participants will be excluded if they have (1) plans to relocate within the next 52 weeks; (2) plans for extended travel (>2-week continuous travel) within the next 52 weeks; (3) elevated diastolic (>100 mm/Hg) or systolic (>160 mm/Hg) blood pressure; (4) elevated resting heart rate (>100 beats/minute); (5) history of diabetes or fasting glucose ≥126 mg/dL or glycated hemoglobin A1C ≥6.5%); (6) elevated triglycerides (>400 mg/dL); (7) elevated low-density lipoprotein (LDL) cholesterol (>200 mg/dL); (8) untreated thyroid disorder or thyroid stimulating hormone (TSH) level out of the normal reference range; (9) hematocrit, white blood cell count, or platelet count significantly outside the normal reference range; (10) cardiovascular disease, peripheral vascular disease, cerebrovascular disease, significant cardiac arrhythmias, or cardiac valve disease; (11) cancer (within the last 5 years, except skin cancer or other cancers considered cured with excellent prognosis); (12) HIV infection; (13) significant renal, musculoskeletal, neurologic, or hematologic disease; (14) significant gastrointestinal disorders (chronic malabsorptive conditions, peptic ulcer disease, Crohn’s disease, ulcerative colitis, chronic diarrhea, or active gallbladder disease); (15) significant pulmonary disorders (chronic obstructive pulmonary disease , interstitial lung disease, cystic fibrosis, or uncontrolled asthma); (16) regular use of an anti-obesity medication within the last 26 weeks; (17) regular use of prescription or over-the-counter medications known to significantly impact appetite, weight, or energy metabolism within the last 26 weeks (e.g., appetite suppressants, lithium, stimulants, anti-psychotics, tricyclic antidepressants); (18) regular use of systemic steroids (other than oral contraceptive pills) within the last 26 weeks; (19) previous obesity treatment with surgery or weight loss device, except (a) liposuction and/or abdominoplasty if performed >1 year before screening, (b) laparoscopic banding if the band has been removed >1 year before screening, (c) intragastric balloon if the balloon has been removed >1 year before screening, (d) duodenal-jejunal bypass sleeve, if the sleeve has been removed >1 year before screening, or (e) AspireAssist or other endoscopically placed weight loss device if the device has been removed >1 year before screening; (20) current alcohol or substance abuse; (21) current or past (within last 26 weeks) nicotine use; (22) history of clinically diagnosed eating disorders (including anorexia nervosa, bulimia, or binge eating disorder); (23) current severe depression or history of severe depression within the previous year based on DSM-IV-TR criteria for major depressive episode; (24) history of other significant psychiatric illness (e.g., psychosis, schizophrenia, mania, bipolar disorder) which would interfere with the ability to adhere to dietary or exercise interventions; (25) current participation in or plans to participate in any other formal weight loss or PA programs or clinical trials; (26) currently following an intermittent fasting weight loss diet plan; (27) self-reported weight loss >5 kg in past 13 weeks for any reason except post-partum weight loss; (28) self-reported weight loss of >50 lbs. in past 3 years for any reason except post-partum weight loss; (29) urinary incontinence or retention to a degree that would interfere with planned doubly labeled water (DLW) measures; or (30) prescription medication(s) that must be taken with food that would preclude the ability to adhere to the IMF intervention. Females will also be excluded if they are currently pregnant or lactating, were pregnant within the past 26 weeks, are planning to become pregnant in the next 52 weeks, or are sexually active and not using contraception.

Who will take informed consent?

Volunteers will be initially assessed for eligibility via a preliminary online screening questionnaire or telephone interview. Potential participants will be asked general eligibility questions regarding age, height, weight, and medical history and receive a brief explanation of the purpose of the study and study expectations, including participant responsibilities and potential risks related to study participation. Interested individuals will be scheduled for a consent and screening visit to determine eligibility. Study coordinators will review the consent form with participants in a private setting. After providing written informed consent, potential participants will undergo a detailed health history and physical examination with the study physician. Body weight and height will be measured to confirm the self-reported BMI. Resting blood pressure and heart rate will be measured, and a resting 12-lead electrocardiogram will be obtained. A fasting venous blood sample will be drawn for measurement of a complete blood count, comprehensive metabolic panel, lipid profile, hemoglobin A1c, and TSH. A urine pregnancy test will be performed in women. Laboratory values available within the past 52 weeks may be substituted for these screening labs at the discretion of the study physician (except for the urine pregnancy test). Participants will also complete the Physical Activity Readiness Questionnaire [29] to assist in screening for exclusionary medical conditions. In addition, participants will complete the Beck Depression Inventory [30] to screen for depression (scores >18 will require further assessment by the study MD to determine if it is appropriate for the subject to participate in the study), the Eating Attitudes Test [31] (scores >20 will require further assessment by the Study MD to determine if it is appropriate for the subject to participate in the study), and the Questionnaire on Eating and Weight Patterns (QEWP-5) [32, 33] to screen for eating disorders (scores that indicate possible diagnosis of a binge eating disorder and/or bulimia nervosa will require further assessment by the Study MD to determine if it is appropriate for the subject to participate in the study). Prior to randomization, all eligible participants will undergo a test fast day to ensure they understand the requirements of the IMF intervention. Participants will be asked to limit EI to 500 kcal/day (women) or 600 kcal/day (men) throughout the test fast day until awakening the following day and rate the difficulty of completing the fast. Study staff will discuss the test fast experience with participants and confirm their willingness to be randomized to the study.

Additional consent provisions for collection and use of participant data and biological specimens

During informed consent, participants will be provided an opportunity to consent for additional, optional ancillary studies and storage of data and biological specimens (including blood and stool) to be used in future research, including genetic research.

InterventionsExplanation for the choice of comparators

We chose to compare IMF to DCR as DCR is the current standard-of-care dietary weight loss intervention. The targeted weekly energy deficit and diet macronutrient content (55% carbohydrates, 15% protein, 30% fat) will be equivalent in the IMF and DCR groups, permitting us to compare the specific effect of the dietary intake pattern on outcomes of interest. Both groups will receive a guideline-based PA prescription [34] and a comprehensive, group-based behavioral weight loss intervention consistent with current guidelines for obesity treatment [17].

Intervention description Intermittent fasting (IMF) group

Participants in this group will be instructed to limit EI to 20% of estimated individual daily weight maintenance energy requirements on three non-consecutive days per week. Weight maintenance energy requirements will be calculated as resting energy expenditure (REE) measured by indirect calorimetry multiplied by an activity factor of 1.5 [35]. On fed days, IMF participants will eat ad libitum, but will be encouraged to make healthy food and portion choices. This 80% energy restriction on three fast days per week translates to a targeted weekly energy deficit of ~34%. Participants will be guided to target diet macronutrient content of 55% carbohydrates, 15% protein, and 30% fat. Sample fast day menus and individualized fast day calorie goals will be provided to assist in achieving EI targets. On fast days, participants will be encouraged to consume their calories in their dinner meal. However, they will be allowed to consume their fast day calories at other times of the day if they have difficulty with the recommendation to consume all calories at dinner. A previous study using a similar IMF protocol [36] found that altering times of caloric intake during the modified fast day (i.e., consuming calories at lunch, dinner, or throughout the day) did not impact weight loss or compliance. Participants in this group will be instructed in calorie counting and food logging but will be asked to count calories and log food intake only on fast days.

We chose a modified IMF paradigm rather than a zero-calorie IMF paradigm (i.e., zero-calorie intake on fast days) for several reasons. First, our review of the scientific and lay literature revealed that the strategy largely endorsed is a modified IMF protocol (rather than a zero-calorie IMF protocol); thus, we felt a modified IMF protocol was the most clinically relevant paradigm to study. Second, in a prior study of a zero-calorie IMF intervention, the level of self-reported hunger remained high throughout the study [37]. In contrast, in studies using a modified IMF approach, hunger decreased [38, 39] or remained unchanged [40, 41] compared to pre-intervention levels, suggesting that long-term compliance may be better with a modified IMF protocol. Third, a cross-over study among n = 10 adults with overweight or obesity examined food intake 2 days after varying degrees of energy restriction: (1) isoenergetic intake, (2) partial 75% energy restriction (i.e., modified fast day), and (3) 100% energy restriction (i.e., zero-calorie fast day) [42]. Results suggested that a modified fast day produces a similar energy deficit as a zero-calorie fast day over a 3-day period because there is less compensatory increase in EI on the post-fast day with the modified fast [42]. Lastly, we administered a brief online survey to 82 adults with overweight or obesity, who had recently completed a behavioral weight loss interventional trial with similar inclusion and exclusion criteria at our CU-AMC Anschutz Health and Wellness Center; 78% of respondents indicated that they would be more likely to adhere to a modified IMF protocol as compared to a zero-calorie IMF protocol in the context of a 52-week behavioral weight loss program.

Daily caloric restriction (DCR) group

Participants in this group will be given a calorie goal designed to produce a 34% energy deficit from baseline estimated individual weight maintenance energy requirements for the duration of the 52-week intervention. As in the IMF group, weight maintenance energy requirements will be determined as (measured REE × activity factor of 1.5) [35]. Participants will be guided to target diet macronutrient content of 55% carbohydrates, 15% protein, and 30% fat.

PA prescription

Participants in both groups will also receive a recommendation to gradually increase moderate intensity PA to 300 min/week over the initial 26 weeks and to maintain this level of PA for the duration of the study (see Table 1 for ramp-up). This target is consistent with current PA guidelines for weight management [34, 43]. Participants will be instructed in how to use a relative intensity scale [34] to achieve the target of moderate intensity.

Table 1 Description of timeline for PA ramp-upStrategies to improve adherence to interventions Group-based behavioral support

To improve adherence to the interventions, both groups will receive a 52-week group-based behavioral weight loss program with equivalent contact and support (see Table 2 for a list of weekly behavioral support topics). The programs will fulfill all recommendations for behavioral interventions for treatment of obesity outlined in current guidelines [17]. Randomized groups will meet separately. Curriculum for DCR will be based on the Colorado Weigh behavioral weight loss program which was developed at CU-AMC in 2000 and uses a skills-based approach and cognitive behavioral strategies for lifestyle modification with a dietary focus on DCR [44, 45]. Curriculum for IMF will also be based on the Colorado Weigh behavioral weight loss program and features similar weekly curriculum themes as used in DCR but adapted by the study PI (VC) with input from a registered dietitian (KB) and a behavioral psychologist (AC) to focus on behavioral support specific to IMF. Group meetings will be taught by a registered dietitian with experience in leading group-based behavioral weight loss interventions. Groups will meet weekly during weeks 0–13 and every 2 weeks during weeks 14–52. Weight will be obtained at each group meeting. Group sessions will last ~60 min and the curriculum will be delivered using a mix of large group discussion, small breakout discussions, visual demonstrations, and written exercises. Participants in the IMF group will be instructed in specific strategies to support IMF including strategies to deal with hunger on fast days, distraction techniques, and choosing a balanced diet/appropriate portions on fed days. Participants in the DCR group will be instructed in specific strategies to support DCR with a focus on daily calorie counting and food logging. Topics covered in both groups during the first 26 weeks include realistic weight loss goal setting, self-monitoring strategies, mindful eating, stress management, cognitive restructuring, improving personal food environments and social networks, and strategies to overcome barriers to healthy eating and increasing PA. Later in the curriculum (weeks 27–52), topics will focus on strategies for weight loss maintenance and include impact of weight loss on EE and propensity for weight regain, relapse prevention, and identifying motivation for long-term dietary and PA changes. Participants will also be provided two 10–15 min 1:1 phone calls with their group leader during the initial 26 weeks of the intervention for individualized dietary goal setting using a standardized protocol.

Table 2 Weekly behavioral support topics covered in group-based sessions by randomized group

Behavioral support for PA will be provided within the weight loss program and will include discussion of health and weight benefits of PA, weekly PA goals to gradually achieve the target of 300 min/week moderate intensity PA, strategies to overcome barriers to PA, and strategies to improve exercise self-efficacy including exercise goal setting. Participants will also be provided a 20–30-min 1:1 in-person PA support session and a 10–15-min 1:1 follow-up telephone call with an exercise specialist during the initial 26 weeks of the intervention for individualized PA goal setting using a standardized protocol. IMF participants will be guided in strategies to adjust exercise duration on fast days (if needed) and still meet weekly PA targets. Participants will be provided access to the CU-AMC Anschutz Health and Wellness Center Fitness Center for the duration of the intervention.

Procedures for monitoring adherence

Adherence to the prescribed dietary interventions for both randomized groups (DCR, IMF) will be monitored via (1) 7-day diet diaries, (2) monthly dietary adherence surveys, and (3) the DLW intake-balance method.

Criteria for discontinuing or modifying allocated interventions

Participants will not be withdrawn for non-adherence in this intent-to-treat study. Participants in both groups will be encouraged to adhere to the dietary prescriptions without modification for the initial 2 weeks as a prior study suggests individuals with obesity become habituated to IMF after ~2 weeks [38]. After the initial 2 weeks, a standardized dietary modification will be offered if a participant (1) expresses a desire to withdraw due to intolerance of study diet and/or (2) experiences adverse effects related to the study diet (i.e., insomnia, impaired concentration, headaches, irritability) that impair ability to function. DCR participants will be allowed to raise their calorie goal to target an energy deficit of 20% from weight maintenance requirements. IMF participants will be allowed to reduce fasting to 2 days per week (i.e., reduce targeted weekly energy deficit to ~20%). Participants will be allowed to continue these strategies for 2 weeks and will then be asked to re-try the original dietary prescription. If they are still unable to tolerate the original dietary prescription after a second attempt, they will be allowed to continue at the modified levels for the study duration. Percentage of participants in each group requiring sustained intervention modification will be recorded. The allocated interventions will not be modified other than as described above. However, participants may discontinue the intervention and/or choose to withdraw from the study at any time.

Relevant concomitant care permitted or prohibited during the trial

Participants will not be permitted to (1) engage in any other formal weight loss or exercise programs or clinical trials during the intervention, (2) use prescription or over-the-counter anti-obesity medications or supplements, and/or (3) undergo obesity treatment with bariatric surgery or endoscopic weight loss device placement.

Provisions for post-trial care

Participants will not be provided any compensation if they experience harm or injury from participating in the study. Participants will continue to have access to the printed behavioral weight loss program materials that will be provided during the intervention after completion of study participation.

Outcomes

Participants will be asked to complete assessment visits at baseline and weeks 13, 26, 39, and 52 during the intervention as well as a follow-up assessment 26 weeks after completion of the 52-week study intervention (i.e., at week 78). See Table 3 for assessments by study timepoint. Body weight, measured in the clinic, is the primary study outcome. Among participants allocated to IMF, outcomes will be measured after a fed day.

Table 3 Schedule of enrollment, interventions, and assessments for the DRIFT study participantsAnthropometric measures

Body weight will be measured using a digital scale in the CU-AMC Anschutz Health and Wellness Center clinic accurate to ±0.1 kg at baseline and at weeks 13, 26, 39, 52, and 78 in the AM, after an overnight fast. Participants will also be given a smart scale (©BodyTrace smart scales (Palo Alto, CA)) for daily home weighing use during the 52-week study intervention. These daily weights will be transmitted via wireless cellular network to a secure website. Height will be measured to the nearest 1 mm with a stadiometer at baseline. Blood pressure will be measured with a manual sphygmomanometer (average of 2 seated values taken after 5 min rest) at baseline and weeks 13, 26, 52, and 78. Waist circumference (cm) will be measured at baseline and weeks 13, 26, 52, and 78 with a tape measure parallel to the floor, just superior to the iliac crest. Fat mass and lean mass will be measured using dual-energy x-ray absorptiometry (DXA) at baseline and weeks 26, 52, and 78.

Resting energy expenditure (REE)

At baseline and weeks 26 and 52, REE will be measured using indirect calorimetry (Parvo Medics TrueOne 2400, Salt Lake City, UT) in the AM in a thermoneutral (68–74 °F) quiet room after a 12-h overnight fast and 24-h abstention from exercise. After the participant has quietly rested for 30 min, a transparent plastic hood connected to the cart will be placed over the participant’s head. For the duration of the test, the participant will be asked to remain motionless and awake. Prior to each measurement, the pneumotach flowmeter will be calibrated using a 3-L calibration syringe (Hans Rudolph Inc., Shawnee, KS, USA), and gas analyzers will be calibrated using standardized gas mixtures. REE will be measured for 20 min and the average of the last 15 min of the measurement will be used to calculate REE using the Weir equation [46].

Free-living total daily energy expenditure (TDEE) and energy intake (EI)

TDEE will be measured using the DLW method at baseline and at weeks 26 and 52. Participants will arrive in the AM after a 12-h fast. A baseline urine sample will be collected prior to administration of the oral dose of DLW containing H218O (10% atom percent excess, APE) and 2H2O (99.8% APE) mixed in a ratio of 15:1. Participants will consume a dose based on body weight and an estimation of total body water. Delivered doses will range from 0.85 to 1.10 g/kg of body weight for females and 0.95 to 1.15 g/kg of body weight for males. The dosing cup will be rinsed twice with 30 mL of water and the rinsing dose consumed. The exact time of dosing will be recorded. Urine will be collected after a 4-h and 5-h post-dose equilibrium period and then again on day 8 at the same time points. Sample aliquots (4 mL) will be frozen at −80°C until analysis. Frozen urine samples will be thawed and prepared by centrifugation and analyzed for 18O and 2H2 enrichment by Off-Axis Integrated Cavity Output Spectroscopy (OA-ICOS, Los Gatos Research Inc, Mountain View CA), as previously described [47]. TDEE will be determined using the two-point method according to Speakman et al. [48]; PAEE will be calculated as TDEE (0.9) − REE and PA level (PAL) as TDEE/REE.

EI will be calculated using the intake-balance method. Change in body composition (determined by DXA scan) will be used to calculate change in energy stores (ΔES) using 9.3 and 1.1 kcal/g as the energy coefficients of fat mass and fat free mass respectively [49]. EI will be calculated as TDEE + ΔES/Δtime [50]. However, because DLW cannot quantify macronutrient intake or provide information on patterns of EI on fed and fast days, 7-day diet records will also be collected at baseline and weeks 13, 26, 52, and 78. Diet records will be analyzed by Colorado Clinical and Translational Sciences Institute Nutrition Core personnel blinded to study group assignment using Nutrition Data System for Research software (University of Minnesota).

Patterns of PA and sedentary behavior

The activPALTM micro (activPAL4, PALTechnologies, Glasgow, Scotland) will be used to estimate time spent in moderate-to-vigorous PA (MVPA), light activity, and sedentary behavior over a 7-day period at baseline and weeks 13, 26, 52, and 78. The activPAL4 micro is a small (23.5 × 43 × 5 mm) and light (9.5 g) device that uses accelerometer-derived information about thigh position to estimate time spent in different body positions (i.e., sitting/lying, standing, and stepping). The device is attached to the anterior thigh and is waterproofed by wrapping in a nitrile sleeve. Thus, it can be worn during bathing and overnight, allowing for 24-h measurement of wake and sleep behavior. Raw activPAL.datx files will be processed using PALBatch software using the CREA – 24-h wear protocol (allows for 4 hours of non-wear per day) and auto-correcting for inverted wear (PAL Technologies Ltd., 2019). The algorithm will be used to determine time spent engaging in sedentary behavior, light activity, and MVPA. Many studies have validated the activPAL for use in adults and report very high levels of accuracy (96.2%) for estimating time in activity intensity categories [51,52,53,54,55]. Participants will also be asked to record all exercise and sleep times (i.e., bedtime and waketime) during the 7-day wear period.

Self-reported dietary adherence, effort, and self-efficacy

Self-reported dietary adherence and ratings of effort to adhere to the prescribed study diets will be obtained on a monthly basis during weeks 1–52 using methodology described by Dansinger et al. [2]. Participants will be asked to rate on a 1–10 Likert scale: (1) how adherent they were to the prescribed study diet over the past week, (2) how hard was it to adhere to the prescribed study diet over the past week, and (3) how likely they feel they can adhere to the prescribed diet for the next month. Participants in the IMF group will be asked to report the number of fast days over the past week. During weeks 53–78, monthly dietary follow-up questionnaires will be administered to evaluate participant efforts to continue weight loss or weight maintenance after the 52-week intervention has ended.

Metabolic/hormonal evaluations

A 12-h fasting blood sample will be obtained for assessment of glucose, insulin, hemoglobin A1c, lipid panel, leptin, ghrelin, peptide YY, and highly sensitive C-reactive protein (hs-CRP) at baseline and weeks 26, 52, and 78. Circulating nutrients and hormonal regulators of fuel metabolism (glucose, insulin, triglycerides, free fatty acids, beta-hydroxybutyrate, and cortisol) will be measured at baseline after both a fed (i.e., 12-h fast) and fast day (i.e., 36-h test fast day with 500 kcal/day (women) or 600 kcal/day (men)) to assess if baseline metabolic response to fasting can serve as a predictor of weight loss. Labs will be analyzed by the UCHealth Clinical Laboratory (beta-hydroxybutyrate) or the Clinical Translational Research Institute (CTRC, all other labs). Insulin sensitivity (homeostasis model assessment of insulin resistance, HOMA-IR) will be calculated as ([insulin] × [fasting glucose × 0.055]/22.5).

Psychosocial, behavioral, and environmental predictors of response

To measure psychosocial, behavioral, and environmental constructs which may predict weight loss response, several validated questionnaires and computer-based tasks will be administered to participants. Selection of measures was informed by the recommendations of the NIH ADOPT Core Measures Project [24, 27, 28]. Questionnaires will be administered at baseline and at weeks 13, 26, 52, and 78 (see Table 4). Each month, participants will also be asked to report any changes in their home or work address, to capture changes in their built, social, and community food environments.

Table 4 Psychosocial, behavioral, and environmental constructs by study week

Executive control will be assessed using computerized task-based measures of executive function (i.e., attention, inhibition, working memory, and set-shifting) and impulsivity (monetary 5-trial adjusting delay discounting task [95]) performed at baseline and at weeks 13, 26, 52, and 78. The battery of computerized tasks to assess key domains of executive function will be deployed locally on a secure study computer and iPad coded by participant ID. The Urgency Premeditation Perseverance Sensation Seeking + Personality Pathway (UPPS+P) questionnaire [84] will be administered once at baseline as a measure of impulsivity.

Participant timeline

See Table 3 for the schedule of enrollment, randomization, interventions, and assessments for study participants.

Sample size

An a priori power calculation was performed for the primary outcome (weight) using nQuery 7.0 (Statistical Solutions Ltd, Cork, Ireland). In our power analyses, we made the following assumptions: (1) power ≥ 0.90 and above for the primary outcome is ideal; (2) the significance level under the null hypothesis will be set at α = 0.05; (3) null hypotheses will be tested against two-directional (that is, two-tailed) alternative hypotheses; (4) a 3-kg between-group difference in weight loss was deemed the minimum difference of clinical importance; and (5) the primary analysis will use the intent-to-treat principle in which all randomized participants are analyzed. We will enroll 150 participants (75 participants per randomized group) to provide 90% power at 5% significance to detect a between-group difference of 3 kg in body weight at 52 weeks. With an attrition rate of as high as 20%, we retain 82% power in a sensitivity (completer’s) analysis. Standard deviation (SD) of weight loss used in this calculation was 5.6 kg and was based on a 52-week interventional study using the Colorado Weigh program conducted at our research center [96]. There are no previous data for us to estimate the effect size for any of the outcomes at 52 weeks. If superiority of IMF to DCR cannot be established, we will perform a non-inferiority test using 2.26-kg weight loss as a non-inferiority margin. In this case, we would have 79% power to establish non-inferiority of IMF as compared to DCR assuming no difference in effect of weight loss between the two groups.

Recruitment

Potential participants will be recruited from the Denver Metro area. We will recruit using CU-AMC campus wide e-mails, flyers, and research website (which maintains an updated list of available studies performed on campus which is also accessible to the general population). We will also provide flyers to the University of Colorado Primary Care practices as well as other local primary care practices including Denver Health Medical Center which serves low-income and minority populations. The trial is registered at ClinicalTrials.gov. We will also employ social media advertising strategies including the use of BUMP Digital Marketing. BUMP uses social media networks (Facebook, Instagram, etc.) to target a population of interest using an appealing advertisement for the study. The advertisement will include a link to a study landing page providing more information about the study, including a brief description of the study requirements and eligibility criteria. Interested individuals will be provided a link to a screening questionnaire.

Assignment of interventions: allocationSequence generation

The randomization assignment will be generated by the study biostatistician (ZP) using a computer-generated block randomization performed within strata defined by sex. Stratification by BMI, age, and race/ethnicity is not needed; given the large sample size, randomization alone should produce balance in these factors.

Concealment mechanism

Allocation of treatment will take place after all eligible participants for each cohort have completed baseline measures, based on the a priori randomization list.

Implementation

The study biostatistician will generate the allocation sequence. After all participants have completed screening and baseline assessments, participants will be notified of the randomization assignment on the first day of group class for the intervention. This process will take place separately within each cohort.

Assignment of interventions: blindingWho will be blinded

Due to the nature of the intervention, it is not plausible to blind participants. However, participants will be blinded to the study hypothesis. Due to the nature of the group-based behavioral intervention, it is also not plausible to blind study staff who enroll participants and deliver or monitor intervention delivery. However, the following personnel will be blinded to study group assignment: (1) staff performing laboratory analyses, (2) DLW assessment core staff performing analysis of the DLW samples, and (3) Colorado Clinical and Translational Sciences Institute Nutrition Core personnel who will be using Nutrition Data System for Research software to analyze the 7-day diet diary data. Lastly, all data analyses and reporting will be completed by study investigators who are blinded to study arm assignments.

Procedure for unblinding if needed

Study personnel who are responsible for analyzing data and reporting results will remain blinded throughout the study period.

Data collection and managementPlans for assessment and collection of outcomes

Outcome measures will be assessed during in-person clinic visits with trained, research staff. We will use detailed data collection forms to promote data quality and reduce assessor error. Weight, blood pressure, and waist circumference will be measured twice and the average of the two measures will be used in analyses. Questionnaires will be administered online using REDCap survey or Qualtrics. Scores from questionnaire responses will be calculated directly within REDCap using hidden, calculated fields. See Table 4 for a list of each study questionnaire and the questionnaire references.

Plans to promote participant retention and complete follow-up

The study coordinators will be responsible for establishing and maintaining contact with participants throughout the study. At enrollment, contact information will be obtained including alternate phone numbers and a phone number of a relative/friend who will always know the participant’s whereabouts. If study appointments are missed, the study coordinator will follow-up with a telephone call to reschedule. If participants miss two group-based support classes in a row without prior communication, study coordinators will contact participants to discuss attendance. In our experience, the most important characteristics of high retention is frequent and high-quality interaction with key study staff. We will ensure continuity of key staff during the entire study so that participants build a relationship with study personnel, to provide inspiration and increase accountability. Participants will also be compensated for completing study outcome measures.

Data management

Field and range checks will be programmed to minimize data entry errors. Data distribution will be checked periodically, and outliers verified; missing data will be tracked and checked. All data will be entered into the REDCap database and will be sight verified by a second study staff member.

Confidentiality

A participant identifier code for the data will be used so that data will not have the participant’s name associated with it. The key linking participant name and participant identifier code is kept in a secured location, with only key personnel having direct access to the list. Participant identifier codes will be used for all data entry and data analyses. The investigative team will be trained in accordance to both the Colorado Multiple Institutional Review Board (COMIRB) and the Health Insurance Portability and Accountability Act (HIPAA) compliance issues and will act to maintain confidentiality and protect health information. Electronic data will be stored on secure servers with a high level of security, controlled access, daily back-up, and long-term retention of back-up files. Hard copies of study data will be kept in in participants’ charts in a locked, secured file cabinet in the principal investigator’s or study coordinator’s office. Access to all study data will be restricted to the principal investigator and research personnel.

Plans for collection, laboratory evaluation, and storage of biological specimens for genetic or molecular analysis in this trial/future use

Blood samples will be collected at baseline and weeks 13, 26, 52, and 78 and will be stored securely until funds are available to support future ancillary studies, including genetic/epigenetic or molecular analyses. Stool samples will be self-collected by study participants at baseline and weeks 13, 26, 52, and 78 using the Alpco EasySampler Stool Collection Kit. Participants will sterilely transfer ~1–2 g of stool to the provided collection tubes, store specimens at −20 °C (home freezers), and transport them to the clinic on icepacks within a week of collection. Blood and stool specimens will then be stored at −80°C for future analyses.

Statistical methodsStatistical methods for primary and secondary outcomes

Baseline characteristics will be summarized by treatment groups using descriptive statistics. Normality assumptions will be evaluated, and transformations used (e.g., square root and log) as appropriate. Body weight is the primary outcome variable and others are secondary or exploratory outcomes. Consequently, no adjustment for multiple outcomes will be applied. P-values < 0.05 will be considered significant. Any imbalance in baseline characteristics between groups that could potentially be a confounding factor for the outcome will be adjusted for in the statistical model.

The primary analysis will use the “intent-to-treat” principle and all randomized participants will be analyzed. Each continuous outcome will be analyzed using a linear mixed effects model. The fixed effects of the linear mixed effects model will consist of treatment group (IMF or DCR) and time of measurement (i.e., baseline and follow-up time points) and their interaction. Time of measurement will be treated as a continuous or categorical variable, as appropriate. Test of the interaction assesses the difference in post-intervention change from baseline between two groups and is used to quantify the effectiveness of the intervention. The 90% confidence interval for the interaction will be calculated. Incidence of adverse events, protocol modifications, and attrition will be compared between groups using chi square or Fisher’s exact test as appropriate.

To explore the impact of selected biologic, psychosocial, behavioral, and/or environmental variables on the intervention effect, we will initially apply models similar to those being tested in aim 1, with the addition of a potential effect moderator as a covariate and its interaction term (i.e., intervention × moderator × time interaction) in addition to related two-way interaction terms. Among factors with a significant three-level interaction, we will further use post-intervention weight loss as the outcome and multiple linear regression to examine their association with the outcome and its interaction with intervention. Similar multiple logistic regression analysis will also be conducted with the outcome dichotomized as responder and non-responder using clinical criteria.

Interim analyses

No interim analyses are planned for this trial.

Methods for additional analyses (e.g., subgroup analyses)

Although not formally powered to examine sex differences, results will be reported separately for men and women to enhance transparency and inform data interpretation.

Methods in analysis to handle protocol non-adherence and any statistical methods to handle missing data

The primary analysis will use the “intent-to-treat” principle to test our study hypotheses under practical and realistic conditions. All randomized participants will be analyzed. We expect that missingness condition is either missing completely at random (MCAR) or missing at random (MAR). A linear mixed effects model will serve as the primary method to handle missing values. The primary and secondary outcomes will be further examined using various sensitivity analyses including analysis of completers using the above statistical models; imputing missing values using baseline-observation-carried-forward, and then assessing efficacy by analyzing the change score from baseline using two sample t-tests or linear regression model while adjusting for other covariates; and finally imputing missing values using Markov chain Monte Carlo multiple imputation to create 20 imputed data sets and then analyzing the data using the linear mixed effects model.

Plans to give access to the full protocol, participant-level data, and statistical code

This paper provides the full protocol. Interested individuals should contact the study principal investigator (VC) if interested in other data or documentation of the study.

Oversight and monitoringComposition of the coordinating center and trial steering committee

This single-site randomized trial does not have a coordinating center or trial steering committee. The trial will be supervised by the PI (VC) with input from study co-investigators (co-authors EM, DB, PM, and ZP). Additional overs

留言 (0)

沒有登入
gif