Presentation is loading. Please wait.

Presentation is loading. Please wait.

Bonnie Spring, PhD Northwestern University

Similar presentations


Presentation on theme: "Bonnie Spring, PhD Northwestern University"— Presentation transcript:

1 Optimizing weight loss treatment: Design and implementation of the Opt-IN study
Bonnie Spring, PhD Northwestern University Director, Center for Behavior and Health OBSSR & NIDA: Optimization of Behavioral Interventions, 5/16/16, Bethesda, Maryland

2 Overview Obesity as public health problem
Gold standard obesity treatments: challenges to scalability Are new treatment packages tested by RCT the best way forward? Treatment optimization by factorial experiment: Conceptual model, intervention components, optimization criteria The Opt-IN study One optimization experiment: Tips, tricks, and a workaround

3 Obesity – The Public Health Problem
Prevalence: U.S: 30% obese; 69% overweight/obese (CDC, 2008) World: 1.4 bill overweight/obese (WHO, 2013) U.S. employer annual cost of obesity-related medical expense, absenteeism, and presenteeism for full-time workers $73.1 billion (Finkelstein et al, 2010)

4 The Intervention Challenge
Effective treatment (intensive lifestyle treatment – e.g., DPP, Look AHEAD) produces 7% sustained weight loss and metabolic improvement but burdensome (16-36 sessions) and costly ($1800/patient) Diabetes Prevention Program, Diabetes Care, 2012

5 real-time feedback and decision support on diet and PA
Treatment Package Approach: The ENGAGED Trial Reduce DPP treatment intensity by half; reconfigure components to increase efficiency 8 in-person treatment sessions Telephone coaching Peer support (groups, message board, adherence info) 4. Incentives 5. Technology (app, accelerometer, texts) “Happy Birthday,Philly! May you eat well, but stay within the Fan Meter safe zone…” “Love Shimmer – so reinforcing!” real-time feedback and decision support on diet and PA supportive accountability via coach monitoring and remote counseling supportive accountability via group monitoring and online social networking RC1DK087126

6 What don’t we know at end of RCT?
If Treatment > Control If Treatment < Control Which components are making positive contribution Whether all components are needed Which components’ contribution to effect offsets its cost How to make intervention more effective Whether any components are worth retaining Whether one component had a negative effect that offset a positive effect Explicitly what went wrong and how to do it better next time Courtesy of Kari Kugler and Linda Collins

7 Opt-In Study Optimization of Remotely Delivered Intensive Lifestyle Treatment for Obesity Principal Investigators Bonnie Spring, Ph.D. (Northwestern University) Linda Collins, Ph.D (Pennsylvania State University) Funded by the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK R01 DK097364)

8 Who We Are Shirlene Wang Coach Gwen Ledford Coach
Christine Pellegrini, Ph.D. Project Coordinator Angela Pfammatter, Ph.D. Project Coordinator Elyse Daly Assessment Coordinator Amanda Paluch, PhD Postdoctoral Fellow Sara Hoffman Coach/Coordinator Kara DeWalt Coach Gwen Ledford Coach Shirlene Wang Coach Garseng .Wong Coach

9 Five components Primary Aim 1 Coaching Intensity Text Messaging
Identify which components/component levels, contribute most to average weight loss percent achieving ≥ 7 % weight loss among overweight and obese adults over a 6-month period. Coaching Intensity 12 v. 24 phone sessions Text Messaging No v. Yes Progress Reports for PCP Recommendations to use meal replacements Training participants’ self selected buddies to be supportive Five components

10 Primary Aim 2 Apply these results to build an intervention strategy
Made up of only active components >ES .25 Intervention package < $500 dollars (scalable)

11 Step 1: Opt-IN Conceptual Model
Intra-Individual The two intra-individual mediators are: self-regulation (skill at controlling oneself through self-monitoring, goal-setting, and use of feedback) and self-efficacy (confidence in one’s personal ability to behave in ways that bring about desired. Facilitation entails reliable access to tools (e.g., Opt-IN Smartphone app) and environmental resources (e.g., meal replacements) that make healthy behavior changes easier to perform. Supportive accountability is the manner in which a coaching relationship conveys positive expectations, support, and a sense of holding the participant responsible to attain behavioral goals17,49,61. Environmental

12 Candidate Intervention Components
For example, text messaging for obesity treatment has been bundled with telephone coaching.54,59 Informing the primary care provider (PCP) about the patient’s weight loss progress has been confounded with having the physician provide behavioral weight loss treatment or having allied health professionals deliver treatment in high-cost medical space, neither of which is cost-efficient.71,37 Other components, e.g., meal replacements, have been tested in such expensive formats (e.g., free food provision) as to preclude real world implementation.41,71 Another example involves the use of peers to provide social support who are given so much training (e.g., 36 hours) that they can more aptly be described as semiprofessionals38.

13 Implemented Design

14 Real World Constraints Impeding Implementation of EBPs
Optimization Goals Brevity Economy Effect size Efficiency Reach Time Money Small benefit Workflow Low uptake

15

16 Opt-IN Goals Design maximally effective and efficient weight loss intervention such that all components are active feasible for real world implementation make lowest possible resource demands 2. Components meet 2 criteria: relative cost/burden is known relative impact on weight loss is unknown 3. Build an intervention: made up of active components [ES > .25] yielding largest treatment effect obtainable for <$500

17 Opt-IN Study Protocol 11/11/2018

18 Enrollment Criteria Exclusion Inclusion Unstable medical conditions
History of diabetes w/insulin, CVD, Crohn’s, sleep apnea (CPAP) Previous hospitalization for psychiatric reasons in past 5 yrs. At risk for cardiovascular events with MVPA or use assistive device for mobility Taking weight loss/gain meds Meet criteria for eating disorders or substance abuse Pregnant, trying to get pregnant, or lactating Living with another Opt-IN participant** Plantar fasciitis diagnosed by physician or podiatrist*** 18-60 years (at least half ≥ 45) BMI kg/m2* Weight stable Must have Android or iPhone smartphone No other major medical concerns Willing to use smartphone app *Extended BMI range from to on 8/26/14 **Approved 9/21/15 ***Approved 11/16/15

19 Study Sequence Web Screen Telephone Screen Orientation Session
Baseline Assessment Randomization Study Sequence 3 Month Assessment 6 Month Assessment

20 Core Treatment (All receive)
Behavioral Coaching and Curriculum Online lessons Phone coaching (12 vs. 24) Self-Monitoring Tool Smartphone application Dietary Intervention Calorie/fat goals, feedback Physical Activity Intervention Moderate intensity (brisk walking) physical activity goal, increasing duration over time Pellegrini CA, Hoffman SA, Collins LM, Spring B. Optimization of remotely delivered intensive lifestyle treatment for obesity using the multiphase optimization strategy: Opt-in study protocol. Contemporary clinical trials. 2014;38:

21 Intervention Components
Coaching Calls over 6 Months 12 calls 24 calls Progress Report Mailed to Primary Care Physician Report mailed at 3- and 6-months to participant and Provider Outlines weight loss progress, behavioral recommendations Text Messages Encouraging words, personalized feedback based on self-monitoring Automated messages in real-time

22 Intervention Components (continued)
Meal Replacement Recommendations Provided with one week’s supply at randomization session Recommendations to continue use made by coach throughout intervention Buddy Training One telephone training session Four minute online webinar sessions Buddies receive $5 for each session complete and additional $20 for completing 3 out of 4 webinars 2 Core: Introduction to Healthy Weight Loss for Buddies, How to Be a Supportive Buddy 4 Additional (choose 2): Support During Setbacks, Building Motivation, Problem Solving, Healthy Eating Out

23 Behind the Scenes: Implementation

24 How to Explain this to Participants…
Ensure they know the 5 components Show them the 16 different conditions Highlight: No control group Everyone gets something Study staff & coaches have no idea which conditions/components yield greatest weight loss In some ways easy because coaches have to follow protocol – can’t establish 32 difft treatment biases.

25 How to Implement a Study with 16 Conditions?
Checks, double checks, and triple checks!

26 Sample visual aids in coaching folder

27 12 24

28 And how to course correct..
What can go wrong And how to course correct.. 11/11/2018 Presentation or Section Title

29 The Corrigendum Copy Paste Error – implemented design (proposed in R01) did not match intended fractional factorial Resolution V design Error discovered when enrollment almost halfway completed Solution: expand the design to a complete 25 factorial with 32 experimental conditions Eliminates aliasing of effects among experimental factors and enables us to examine all five components Does not require randomization of additional subjects Res 5: each main effect is aliased with one four-way interaction or the five-way interaction, and each two way interaction is aliased with one three-way or higher-order interaction. Pellegrini CA, Hoffman SA, Collins LM, Spring B. Corrigendum to optimization of remotely delivered intensive lifestyle treatment for obesity using the multiphase optimization strategy: Opt-in study protocol; Contemporary clinical trials. 2015;45:

30 Project Recruitment Timeline
CTAs Projected Actual Average carried forward CTA CTA CTA CTA Randomization Schedule ( ) Total # to Randomize: 560 (current: 425) Time Remaining: 7 Months (June 2016 – Dec 2016) Monthly Randomization Rate: 17 (need 19.3/month) May Randomization = 5 so far, 3 scheduled

31 Current Recruitment Status
Web Screen N = 6175 Common IE Reasons BMI too high = 1214 BMI too low = 440 Weight unstable = 712 Common IE Reasons Unable to contact = 608 Didn’t attend session = 628 Not interested = 256 Phone Screen N = 2949 Eligible N = 1649 Randomization N = 425 Baseline N = 848 560 Gender: Female = 266 / Male = 52 Cond 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 n 18 19 9/13-7/15 Cond 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 n 7 8 6 7/15-12/16

32 Approaching the Finish Line
Current Progress: Randomized 405 participants since October 2013 Goal = 560 participants by December 2016 (March 2017?) Evidence-based intervention design: genuine equipoise Appealing to ppts: no inert control Some implementation challenges, but manageable Error highlights design’s efficiency (no additional ppts needed), flexibility, ability to course correct

33 Thank you!

34


Download ppt "Bonnie Spring, PhD Northwestern University"

Similar presentations


Ads by Google