Engagement, Incentives, and Impact

Engagement, Incentives, and Impact Wes Alles, Ph.D. Structure of the BeWell incentive Program at Stanford University BeWell engagement from 2008 to 2013 RAND Workplace Wellness Programs Study Answer to mandated questions Data collection Findings Impact of programs Important questions for BeWell Are we preaching to the choir? How do we get impact with just 20% engaged beyond assessment? How do we demonstrate impact? How can we merge health plan data with BeWell data Why do Stanford faculty/staff participate/not participate in BeWell

BeWell Program Design Assess Plan Take Action

Incentives in 2013 Gateway $480 $100 for 5 out of 9

Engagement in BeWell, 2008 to 2013

Number of Participants in 2013* SHALA 9,506 BeWell Berries ** Wellness profile** Biometrics screening Wellness advising session Online wellness plan 6,781 6,781 6,303 Fitness assessment 2,470 Personal training 1,389 Physical activity 3,012 Stress workshop 1,881 Nutrition workshop 1,787 Healthy living class 2,002 Physical exam 2,286 Wellness advocate 1,523 BeWell walkers 1,042 * As of October 31, 2013 **Only those who have taken SHALA are eligible to participate in Wellness profile and BeWell berries.

Participation 2008-2013* 2008 2009 2010 2011 2012 2013 Stanford Health and Lifestyle Assessment 7565 7888 7703 9331 9401 9506 BeWell Numbers/Biometrics N/H N/H 3314 6495 6919 6781 Referral Call/Advising N/H N/H 1017 6326 6895 6781 Personal Wellness Plan N/H 4374 2743 6553 6789 6303 Fitness Assessment 708 1581 1337 2196 2407 2470 Personal Training 621 1289 1069 1356 1501 1389 Stress Workshop N/H N/H 1109 1626 1594 1881 NutriNon Workshop N/H N/H 1273 1943 1793 1787 Healthy Living Class N/H 973 1212 2093 2589 2002 Physical AcNvity 1200 1764 1654 3352 3627 3012 Wellness Advocate N/H N/H 872 1588 1475 1523 Well- Visit N/H 2095 1621 2109 2167 2286 BeWell Walkers N/H N/H N/H N/H 800 1042 Completed 5+ Berry AcNviNes N/H 1368 1392 2208 2323 2129 * Data through October 31, 2013

Change in participation between 2008 and 2013 Assessment: 26% increase in SHALA 105% increase in Biometric Screening Personal Plan: 43% increase in Personal Wellness Plan Take Action: 56% increase in people completing 5+ BeWell berries 70% increase in physical activity class participation 106% increase in healthy living class participation

The Affordable Care Act required data analysis and reporting in order to shape national policy with regard to workplace wellness Early in 2013, RAND Health published a research report titled Workplace Wellness Programs Study This report is intended to satisfy the mandate within ACA. RAND presents research findings and objective analysis. All RAND reports undergo rigorous peer review to ensure high standards for research quality and objectivity. This RAND project represents the most comprehensive analysis of worksite wellness programs to date.

Workplace Wellness Programs Study Final Report Sponsored by the U.S. Department of Labor And the U.S. Department of Health and Human Services

The RAND Report Sought To Answer These Questions 1. WHAT IS THE PREVALENCE AND WHAT ARE THE CHARACTERISTICS OF WORKSITE WELLNESS? 2. WHAT IS THE EVIDENCE FOR PROGRAM IMPACT? 3. WHAT IS THE ROLE OF INCENTIVES? 4. WHAT ARE THE KEY FACILITATORS OF SUCCESSFUL WELLNESS PROGRAMS?

Data Collection RAND used four data collection streams: 1. Health promotion scientific and trade literature 2. A RAND national survey of employer wellness programs 3. Medical claims matched with wellness program data 4. Case studies among a diverse set of employers

Background 1. Unhealthy lifestyle is responsible for increasing the prevalence of chronic disease 2. Chronic Disease leads to: decreased quality of life premature death and disability increased health care costs 3. Chronic disease is occurring at younger age while people are still employed 4. Employers are interested in changing the long term health trajectory of employees to reduce a variety of costs associated with chronic disease 5. Approximately half of the employers with more than 50 employees offer wellness programs

Background 72% of employers with wellness programs characterize them as a combination of screening (HRA and Biometrics), and interventions, including targeted referrals. Among the employees who are targeted for a specific intervention (fitness classes, health education, behavior modification), fewer than 20% chose to participate. The average employee participation rate among those who were referred to specific wellness programs was: Health risk appraisal 46% (assessment) Biometrics 46% (assessment) -------------------------------------- Fitness 21% (engagement) Smoking cessation 7% Weight management 10% Disease management 16%

Evidence for Program Impact 50% of employers with more than 50 employees offer disease management programs Among employers that offer disease management programs, the most popular are:. Diabetes 85% Asthma 60% Coronary artery disease 59% Heart failure 54% Depression 53% Cancer 51%

Evidence for Program Impact Participation in weight control programs was associated with a reduction of BMI and the effect persisted for two additional years. The average weight loss was about one pound for both men and women. Smoking cessation participants quit more often than nonparticipants. Improvements also were shown among participants who became more physically active, who ate more fruits and vegetables, who lowered their fat intake, and who reduced body weight, cholesterol, and blood pressure.

Health Care Cost and Utilization Employers overwhelmingly expressed confidence that wellness programs will reduce medical cost, absenteeism, and health-related productivity, despite the fact that fewer than half of the employers with wellness programs formally evaluated program impacts and just two percent reported estimates of actual savings. None of the five case studies presented by RAND conducted a formal evaluation. Just one of the five requested cost trend data from its health plan. RAND s analyses suggest that participation in a wellness program over five years is associated with a trend toward lower health care costs and decreasing health care use. RAND estimates the annual average difference in health care costs to be $157 when comparing participants vs nonparticipants in a wellness program for five years but the change is not statistically significant.

Role of incentives in workplace wellness programs Laws and regulations at the federal and state level impose requirements and regulate the use of financial incentives. When wellness programs require that individuals satisfy health related standards, both HIPAA and the ACA limit incentives and require accommodation through a reasonable alternative in specific circumstances. Final rules that were published at the same time as the RAND Report allow employers with wellness programs designed to prevent or reduce tobacco use to charge tobacco users up to 50% more in premiums than nonusers. Other federal laws that govern incentives include the Genetic Information and Nondiscrimination Act (GINA), and the Americans with Disabilities Act (ADA). HIPAA and the Affordable Care Act have provisions related to the nondiscriminatory application of incentives when offered through both self-funded and insured group health plan coverage.

Role of incentives in workplace wellness programs 68% of employers with wellness programs use financial incentives to encourage program participation and 10% use results-based incentives. The most common type of incentive is triggered by completion of an HRA, biometric testing, and participation in lifestyle management interventions. Employers that use an incentive for an HRA report participation at 63% compared with 29% that don't provide an incentive. When incentives are offered for biometric testing, the participation rate is 57% vs 38% when incentives are not offered. 84% of employers frame the incentive as a reward rather than as a penalty. RAND analyses conclude that employer incentives for HRA completion and program participation can significantly reduce weight and smoking and increase exercise, HOWEVER, the size of these effects is small and unlikely to be clinically meaningful.

Role of incentives in workplace wellness programs Smoking is the behavior most frequently targeted when incentives are tied to a health standard (i.e., non-smoking). 21% of employers with smoking cessation programs pay for participation 19% of employers tie the incentive to "quit smoking". Smoking also is the only health behavior for which achieving the goal is rewarded with a higher incentive than participation in a program. The average incentive for quitting was $682 (12.6% of the cost of insurance) vs. $203 paid for participation. RAND notes that peer reviewed literature on smoking cessation indicates that incentives increase the initial quit rate but they generally do not achieve longterm behavior change. For other lifestyle behaviors, employers were three to four times more likely to reward participation rather than behavior change.

Summary of Program Impact Evidence of program impact 1. Lifestyle management interventions can reduce risk factors 2. These effects are sustainable over time and are clinically meaningful 3. These findings confirm that workplace wellness programs can help contain the current epidemic of lifestyle related diseases, the primary cause of premature morbidity and mortality's well as health care cost in the United States 4. An important issue for further study is how program design and implementation can improve program impact 5. The RAND estimate of wellness program effects on health care costs are lower than most results reported in the literature, but cautioned the reader that it analyzed the isolated effect of lifestyle management interventions, whereas published studies capture the effect of an employer's overall approach to health and wellness

Summary of Program Impact Evidence of program impact 6. Although RAND did not detect significant decreases in cost for health care, the difference in costs between participants and nonparticipants diverge over time. There is reason to believe that a reduction in medical costs would materialize if employees continued to participate. RAND estimates these programs would be cost-neutral after five program years Role of incentives in workplace wellness programs 7. Employers are using incentives to increase employee engagement 8. An incentive over $50 appears to be effective in getting employees to complete an HRA 9. The use of incentives tied to health standards remains uncommon- 10% of employers with wellness programs

Summary of Program Impact 10. The use of incentives tied to health premiums remains uncommon--7% of employers with wellness programs 11. Maximum incentive amounts average less than 10% of the employee premium for health care coverage 12. Smoking is the only lifestyle risk factor for which results-based incentives are more common than incentives paid for participation Limitations 13. RAND notes that its survey results may be subject to response bias, as there were some differences between respondents and non-respondents. The database represents a convenience sample of employers with long-standing commitment and substantial investment in wellness programs. The results, therefore, may not be generalizable.

Summary of Program Impact 14. RAND notes that as with any non-experimental design, it cannot rule out that unobservable differences (such as motivation) between program participants and non-participants have influenced the results. 15. Limited variation of incentive among employers prevent definite conclusions regarding the effect of incentives on wellness program participation, health behaviors, and outcomes.

This Question Was Raised By Professor Alain Enthoven From Professor Enthoven: What should we think about the report that fewer than 20% of those who need it (wellness) participated? I have been worried about "preaching to the choir.

Response from BeWell From BeWell Our goal from the beginning was to move participants further along a continuum of responsibility. We suggest that if a participant receives a recommendation for an intervention, some portion of the incentive offered by the university should be based on fulfillment of this recommendation. Considering the $157 dollar difference between participants and nonparticipants reported by RAND Health, the more participants who commit to the recommended interventions, the greater the impact will be on employee health and positive impact to the university.

Comparison of SHALA Results: 2008-2013 YTD (10/30/2013) Stanford Employees, 6-year Repeat Participants: (N=2,225; 25% male; Mean age 47.9 years old in 2013)

Healthy Lifestyle Index Score: First timers vs. Non-first timers 2008 2009 2010 2011* 2012** All participants (n=6665) (n=7094) (n=7277) (n=9054) (n=9470) 70.8323 72.5338 73.2511 73.5091 74.2646 First timers (n=6665) (n=2218) (n=1681) (n=2718) (n=3041) 70.8323 70.8499 70.5027 72.8067 72.7186 Non-first timers N/A (n=4876) (n=5596) (n=6336) (n=6429) N/A 73.2998 74.0767 73.8104 74.9959 Note: *One question (that was worth 3 points of Lifestyle Index) was missing in 2011 online SHALA due to a technical problem. To make a year-over-year comparison, the Lifestyle Index score in 2011 was transformed to a 100-point scale. Therefore the Index score in 2011 may not be completely comparable with those of the previous years. **A revision was made to 2012 SHALA, affecting response options of 3 questions included in the Lifestyle Index score calculation. To calculate the Lifestyle Index scores in 2012, these response options have been recoded to match the scoring of old responses as close as possible; however the Index score in 2012 may not be completely comparable with those of the previous years.

Healthy Lifestyle Index Score: First timers vs. Non-first timers Healthy Lifestyle index score 80 78 76 74 72 70 68 66 64 62 60 2008 2009 2010 2011* 2012** First timers Non-first timers Note: *One question (that was worth 3 points of Lifestyle Index) was missing in 2011 online SHALA due to a technical problem. To make a year-over-year comparison, the Lifestyle Index score in 2011 was transformed to a 100-point scale. Therefore the Index score in 2011 may not be completely comparable with those of the previous years. **A revision was made to 2012 SHALA, affecting response options of 3 questions included in the Lifestyle Index score calculation. To calculate the Lifestyle Index scores in 2012, these response options have been recoded to match the scoring of old responses as close as possible; however the Index score in 2012 may not be completely comparable with those of the previous years.

SHALA Health Report Card, 2013 (up to 10/30/2013) (N=8,264; 33% Males; Mean age 42.7 years old)

SHALA Health Report Card, 2013 (up to 10/30/2013) (N=8,264; 33% Males; Mean age 42.7 years old)

Percent with Healthy Biometrics, Stanford University (2012) vs. National data 100 90 80 70 60 50 40 30 Stanford National 20 10 0 Blood pressure Total cholesterol Fasting glucose BMI * National numbers from: US DHHS, CDC, American Diabetes Association; not 100% comparable to Stanford numbers

Changes in % with Healthy Biometric Measures: From 2011 to 2013 YTD 100% 90% * 80% 70% 60% *** 50% 40% 30% 2011 2013 20% 10% 0% Blood pressure Total/HDL cholesterol ratio Fasting glucose BMI Healthy biometric measures are defined as follows: Blood pressure: <120/80mmHg Total/HDL cholesterol ratio: < 5 Fasting glucose: 70 - <100mg/dL BMI: 18.5 - < 25 *p < 0.05; **p < 0.01; ***p < 0.001 Note: This analysis was performed among employees who participated in the Biometric Screening both in 2011 and 2013 (as of 10/30/2013), excluding pregnant women (N=3,540).

Summary of BeWell Findings Participants in BeWell are healthier than national averages, but still have biometric risks. Participation in BeWell among the Stanford community has increased over time. All self-reported lifestyle behaviors have improved from 2008-2013. It took five years to see a reduction in stress and finally, in 2013 we saw improvement. This is a priority that we chose to focus on beginning in 2009. Between 2011 and 2012 we noticed that fasting glucose was getting worse. We began to focus on this in advising and in classes. The 2013 data show that the trend has now reversed and glucose levels are going down.

BeWell Needs to Define Its Impact We will compare the Blue Shield data for diagnoses and costs comparing participants and non-participants for those diseases that are heavily influenced by lifestyle. We have tracked participation in BeWell using SHALA lifestyle scores, motivational assets, biometric data, and participation in programs. We need to merge the Blue Shield data with the BeWell data. Our goal is to quantify the impact of the BeWell program. These data will inform our decisions about incentives, program priorities, and intensity of follow-up on recommended steps among high risk participants.

Are We Preaching to The Choir? While the participants are healthier than national norms, they still have room for improvement, especially in the areas of blood pressure, BMI, and blood glucose. We found that those individuals who started the program in 2012 had healthier biometrics than those in previous years. Something in 2012 attracted them to join. It could have been the larger incentive, or they realized they might benefit from the program as others have done. Those who took SHALA repeatedly scored higher than first timers each of the five years. Until we get the merged data from the health plan, it is impossible to say whether we re missing the least healthy members of our population.

BeWell Plan for Analysis to Demonstrate Positive Impact SHALA questions Lifestyle Readiness for Change BeWell data Biometrics Participation in Wellness Programs Wellness Berries Healthy Living Physical Activity Behavior Change Diagnoses Costs Trends Merge With Data from Blue Shield

Agreed to share SHALA, Biometric Scores, and Wellness Participation with health plan 2012 Total Employees Female 92% Male 93%

BeWell and Health Plan Data Merge

Why do Stanford Faculty and Staff participate/not participate in the BeWell incentive program? Results are from an anonymous survey asking about participation in the 2012 BeWell program

Why did you participate in BeWell? Reason Percentage of respondents I'm interested in my health 91% For the monetary incentive 78% For the free and/or reduced price classes/ assessments Stanford administration encouraged employees to participate 55% 25% My coworkers/supervisor wanted me to 8% My family wanted me to 7% Other (please specify) 5%

Why did you not participate in BeWell? Reason Percentage I did not have time 51% Just didn't get around to it 34% I thought it was too complicated 27% I did not want to reveal my personal information 24% I could not participate while at work 22% Other (please specify) 19% It was too difficult to participate because I don't work on main campus 14% I did not know how to participate 14% I'm already healthy 12% I did not believe it would improve my health 12% I did not think the monetary incentive for completing it was enough 11% I participated in earlier years and didn't enjoy the experience 5% I did not know about it 4%

Why did you not sign up for a fitness class? Reason Percentage I get my exercise in other ways 34% I could not attend them while at work 32% Just didn't get around to it 24% I did not like the scheduled times 20% I do not enjoy group fitness classes 18% Other (please specify) 12% I did not like the locations 10% I already fulfilled 5 other BeWell berry activities 8% I did not know about them 4% I did not like the classes offered 4% I was not ready to make any changes at that time 4% I did not know how to register 3% I did not realize the fee was only $20 after taking SHALA 3% A physical/medical condition limits my physical activity level 3% I did not believe it would improve my health 2% I thought it was too expensive 2%

Why did you not attend at least 60% of your fitness class? Reason Percentage of respondents I could not get out of work 62% My schedule/workload changed since I registered 38% Other (please specify) 28% I felt guilty leaving work 22% Too much trouble to get to the class 18% The location was too inconvenient 17% I developed a medical problem that prevented me from participating 10% I did not like the instructor and/or class 8% I was discouraged by coworkers or supervisors 1%

Summary Employees are motivated to participate because: They are interested in their health They are interested in the monetary incentive They enjoy the free and reduced price wellness programs The university administration supports it Employees cite the following as reasons for not participating: Just didn t get around to it Lack of time Felt guilty leaving work Didn t know about it How can we use these known barriers to reach the late adopters?

Summary Participation in BeWell has increased from 2008 to 2013 Lifestyle scores have improved every year between 2008 and 2013 Percent with healthy biometrics improved from 2011 to 2013. Merger of data between Blue Shield and Stanford is now active Most people signed a waiver to send SHALA, biometric data, and participation data to our Benefits Administrator so we can merge the data with Blue Shield Analysis of this data will be a rich source of information to help us understand the impact of BeWell