Worksite Wellnes Index with Ron Goetzel
Upcoming SlideShare
Loading in...5
×
 

Worksite Wellnes Index with Ron Goetzel

on

  • 3,667 views

 

Statistics

Views

Total Views
3,667
Views on SlideShare
3,094
Embed Views
573

Actions

Likes
3
Downloads
92
Comments
0

5 Embeds 573

http://healthpromotionlive.com 487
http://hplive.org 83
http://healthpromotionlive.org 1
http://edu.idwellness.org 1
http://webcache.googleusercontent.com 1

Accessibility

Upload Details

Uploaded via as Adobe PDF

Usage Rights

© All Rights Reserved

Report content

Flagged as inappropriate Flag as inappropriate
Flag as inappropriate

Select your reason for flagging this presentation as inappropriate.

Cancel
  • Full Name Full Name Comment goes here.
    Are you sure you want to
    Your message goes here
    Processing…
Post Comment
Edit your comment

Worksite Wellnes Index with Ron Goetzel Worksite Wellnes Index with Ron Goetzel Presentation Transcript

  • REVIEW OF WORKFORCE HEALTH INDICESHOW ORGANIZATIONS CAN MEASURE AND IMPROVE WORKFORCEWELLNESSRON Z. GOETZEL, PH.D.August 12, 2011
  • WHAT IS A WORKFORCE WELLNESS INDEX• Qualitative tool to assess the extent to which a an employer or worksite has adopted ―best practices‖ for population health improvement• Quantitative tool that aligns employees‘ health risk profile with outcomes of interest to the organization (e.g., medical care costs and worker productivity)• Produces a ―single number‖ reflecting the interaction of population health risks and cost that can be compared and contrasted over time 2
  • EXAMPLES OF QUALITATIVE TOOLS• HERO Best Practice Scorecard• National Business Group on Health Wellness Score Card• Checklist of Health Promotion Environments at Worksites (CHEW)• Employers‘ Health and Productivity Management Inventory, Emory• Environmental Assessment Tool (EAT), UGA/Emory• Leading by Example (LBE) – Leadership Support Tool, Emory, UGA• Healthy Employees in Healthy Organizations, ENWHP• Heart Check: Assessing Worksite Support for a Healthy Lifestyle, NYSDH• Heart Check Lite, Fisher & Golaszewski• Well Workplace Checklist, WELCOA 3
  • EXAMPLE OF A WORKSITE HEALTH INDEX  Sample Results Nation ABC Maximu al Category Inc.’s m Averag Score Points e 1. Strategic Planning 7 5 11 2. Leadership 18 16 33 Engagement 3. Program Level 7 11 22 Management 4. Programs 14 28 56 5. Engagement Methods 40 29 67 6. Measurement and 2 5 11 Evaluation TOTAL 88 94 200 Based on ABC Inc.‘s response and database average as of [May 1, 2009]. 4
  • HPM Tool (Screenshot)
  • LEADING BY EXAMPLE (LBE) ASSESSMENT
  • LBE ITEMS 7
  • ENVIRONMENTAL ASSESSMENT TOOL
  • PHYSICAL ACTIVITY POLICIES ANDENVIRONMENTAL SUPPORT– Make available educational information on physical activity (print, web, video, audio) (e.g., brochures in common areas, links from company website, video or audio library)– Lay out walking routes and trails (onsite or offsite in surrounding community)– Post signs at elevators, entrances, by exit signs, etc. that identify stairwell location and encourage use– Make available bikes free of charge for onsite transportation– Install bike racks/bike lockers at common building entrance ways– Offer pedometer programs (distribute free pedometers)– Offer onsite fitness center or fitness room– Encourage use of off-site fitness club subsidies (partial/full reimbursement to employees)– Encourage use of fitness club discounts (discounts arranged with local fitness/athletic center to reduce employee out-of-pocket costs)– Offer time off for physical activity during work hours– Install fitness equipment at the workstation (e.g., cardio equipment, hand weights/dumb bells, stretching mats, exercise balls)– Install sport-specific exercise areas (e.g., basketball, volleyball, racquet ball or tennis courts)– Offer sports team sponsorship or organized physical activities– Provide showers/locker rooms– Develop a newsletter or column for physical activity related information (print or computer-based; providing information on programs, feature articles, high-risk targeted messaging, etc.)– Install posters/bulletin boards designated for physical activity information– Develop policy statement supporting physical activity 9
  • CHANGE AGENT CULTURE OF HEALTH SURVEY 10
  • CDC Worksite Health Index (WHI) ProjectPurpose – Why we need a Worksite Health Index: • The workplace provides many opportunities for promoting health and preventing disease. • There is a need for widely available, recognized tools to assist employers in their assessment of workplace programs, particularly small and medium sized businesses. • Employers are increasingly looking to experts for practical guidance and population-based solutions. 11
  • CDC Initiative – Goals• Develop a tool for use by employers of all sizes and types to assess their organization‘s ―health‖ in terms of: • Worker health/risk factors • Program, policies, environment, culture • Other relevant areas important to the success of workplace health programs• Allow employers to receive immediate feedback and link to additional tools and resources 12
  • Stakeholder Panel• David R. Anderson, PhD, LP – StayWell Health Management• Catherine M. Baase, MD, FAAFP, FACOEM –The Dow Chemical Company• Ken Holtyn, MS – Holtyn & Associates Health Promotion Consultants• Pamela Hymel MD, MPH, FACOEM – Cisco Systems• Laura Linnan, ScD, CHES – University of North Carolina• Dyann Matson-Koffman DrPH, MPH, CHES – CDC, National Center for Chronic Disease Prevention and Health Promotion• Nico Pronk, PhD, FACSM, FAWHP – Health Partners, Center for Health Promotion• Paul Schulte, PhD – CDC, National Institute for Occupational Health and Safety• Andrew Spaulding, MS – Maine CDC/DHHS Cardiovascular Health Program• Cristie Travis, MS – Memphis Business Group on Health• Tonya Vyhlidal, MEd, CHPD, CPT – Lincoln Industries• Ed Watt, MS – Transport Workers Union 13
  • CDC Approach• Phase I – Environmental Scan and Planning • Literature review • Expert consultation• Phase II - Develop Worksite Health Index • Finalize the domains, indicators, and metrics for the index • Build and pilot test a prototype• Phase III - Develop Web Application and Disseminate • Translate prototype into functional application • Promote adoption and utilization of tool 14
  • CDC Environmental Scan• Used four sources to construct an organizing framework of WHI best and promising practices• Three main domains (with 25 subcategories) • Leadership and Corporate Culture • Program Design and Implementation • Program Evaluation 15
  • CDC Worksite Health Index Project Domain 1: Leadership and Corporate Culture CATEGORY/CONCEPT DESCRIPTION/EXAMPLES 1. Leadership and Management Support Demonstrate organizational commitment and leadership support by engaging mid-level management, sharing program ownership with all staff levels, and leading by example. 2. Organizational Culture and Policies A healthy company norm/culture that includes a supportive physical environment and supportive policies (e.g., healthy food, no tobacco, flex time). 3. Alignment of Business and Health Goals Explicit connection of health goals and programs to organization‘s core business objectives and principles. 4. Wellness Champion Identified wellness coordinator/champion, council, or employee-driven advisory board. 5. Sustainability Scalable and accessible programs.8/11/2011 16 16
  • CDC Worksite Health Index Project Domain 2: Program Design &Implementation CATEGORY/CONCEPT DESCRIPTION/EXAMPLES 6. Planning and Program Goals Establish clear, consistent, theory and evidence-based principles and a clearly defined plan of operations with specific program goals and objectives (and with realistic expectations). 7. Diagnostics and Assessment Use/analysis of claims data, health risk data, biometrics, and measures of productivity. 8. Integration, Data Systems Efficient and effective data practices and informatics, integration of and Informatics relevant data systems across multiple organizational functions and departments (e.g., with employee health risk data). 9. Incentives Consider meaningful incentives/rewards and incentives linked to participation (not to changes in biometrics). 10. Adequate Resources Dedicated, adequate resources spent to achieve desired ROI. 11. Multi-Component Multi-component programs (e.g., health education, Interventions and Effective counseling, behavior change/chronic disease risk reduction, Implementation emergency preparedness, safety and the elimination of recognized occupational hazards), integration of program components at the point of implementation. Integrated staff (multi-disciplinary; cross departmental); Integrate/ensure vendor, partners engagement. 12. Tailored Interventions Tailor programs to the specific workplace and provide individualized interventions.8/11/2011 17 17
  • CDC Worksite Health Index ProjectDomain 2: Program Design & Implementation CATEGORY/CONCEPT DESCRIPTION/EXAMPLES13. Screening and Triage Scalable and effective assessment and screening to identify the highest risk individuals, triaging of individuals into programs that produce the biggest payoff/impact, providing public health interventions to keep people at low risk.14. Piloting Start small/simple and scale up using success of pilot results.15. Engagement of Local Community Coordinating with insurance and health care providers (especially primary care providers), public health partners, and community based organizations, using community resources and linkages.16. Accessibility/ Reducing Barriers Accessible/attractive programs and initiatives at the worksite and in the community with services that balance personal, face-to-face interactions with the latest advancements in computers/technology, the promotion of employee participation.17. Confidentiality Relentless focus on safeguarding personal health information, privacy and protecting confidentiality.18. Ecological Interventions Environmental/ecological interventions, the social environment, the built environment in the workplace and community, (e.g., LEED buildings).19. Communications Regular, strategic, multi-channel, effective marketing and communication of results (to management, employees and their dependents).20. Health Benefits Insurance plan design (coverage; payment structure, degree of innovation in plan), vacation and sick leave.8/11/2011 18 18
  • CDC Worksite Health Index Project Domain 3: Program Evaluation CATEGORY/CONCEPT DESCRIPTION/EXAMPLES 21. Measurement and Evaluation Program measurement, analysis and evaluation (e.g. claims data, evaluation data, audit tools) using rigorous methods that stand up to peer review. 22. Effective Tools Find and use effective, valid, and reliable tools. 23. Accountability Build accountability at all levels that is linked to rewards. 24. Learn from Results Learn from experience; adjust the program as needed, explicit connection of results to core values. 25. Economics Return-on-investment (ROI), health care costs, workers‗ compensation, disability.8/11/2011 19 19
  • DERIVATIVE INSTRUMENT – CDC HEALTHSCORE CARD 20
  • OTHER EXCITING DEVELOPMENTS• Development of quantitative health indices – Novartis – PepsiCo – Thomson Reuters• International applications: Discovery Holding (South Africa) 21
  • EVOLUTION OF THE WORKFORCE WELLNESS©2010 Thomson Reuters INDEX 22
  • PRIOR WORK ON INDEXES • The Workforce Wellness Index evolved from prior work carried out in-house at Thomson Reuters – Health indexes for employer clients such as Pepsi Bottling Company and Novartis • Goetzel RZ, Carls GS, Wang S, Kelly E, Mauceri E, Columbus D, Cavuoti A. ―The relationship between modifiable health risk factors and medical expenditures, absenteeism, short-term disability and presenteeism among employees at Novartis. Journal of Occupational and Environmental Medicine. 2009. 51(4): 487-499, April 2009. • Henke RM, Carls GS, Short ME, Pei X, Wang S, Moley S, Sullivan M, Goetzel RZ. The Relationship between Health Risks and Health and Productivity Costs among Employees at Pepsi Bottling Group. Journal of Occupational and Environmental Medicine. 52(5):519-527 May 2010 • Kelly E, Carls GS, Lenhart G, Mauceri E, Columbus D, Cavuoti A, Goetzel RZ. The Novartis Health Index: A method for valuing the economic impact of risk reduction in a©2010 Thomson Reuters workforce. Journal of Occupational and Environmental Medicine. May 2010; 52(5): 528- 535. 23
  • PBC AND NOVARTIS HEALTH INDEXES • Indexes based on relationship between observed health risks and various employer health care and productivity costs – Include medical+Rx, workers‘ compensation, short-term disability, absenteeism and presenteeism • Indexes link employee health risks and cost data to produce a single number, which can be used by management to gauge employee health risks and costs simultaneously©2010 Thomson Reuters 24
  • HEALTH INDEX PHASES • Phase I: Develop Descriptive Statistics: Characteristics of Employees at High vs. Low Health Risks • Phase II: Investigate Relationships Between Health Risks, Medical Expenditures, Productivity, and Other Outcomes • Phase III: Publish Finding • Phase IV: Develop an Excel-Based Model to Forecast the Financial Impact of Interventions Designed to Improve Health and Lower Health Risks©2010 Thomson Reuters 25
  • NOVARTIS – DATA INTEGRATION • Collect data needed to measure employee health risks, productivity, and medical expenditures and merge these data sets into a single analytic file: – Medical claims data for inpatient, outpatient, and ancillary services – Pharmaceutical claims – Health plan enrollment data – Mayo Health risk appraisal (HRA) – Work Limitations Questionnaire (WLQ) (i.e., presenteeism) – Incidental absence data©2010 Thomson Reuters – Short-term disability data 26
  • ©2010 Thomson Reuters 100% 0% 25% 50% 75% General Health 34% Nutrition Risk 81% Emotional Health Risk 70% Safety Risk 63% Weight Risk27 Blood Pressure Risk 53% 49% Exercise Risk NOVARTIS RISK FACTORS 33% Cholesterol Risk 15% Triglycerides Risk 10% Tobacco Risk 8% PROPORTION OF STUDY POPULATION AT HIGH RISK Blood Sugar Risk 7% Alcohol Risk 5%
  • PHASE I EXAMPLE: PRODUCTIVITY AT WORK* General Health Low High Risk Risk p-value1 Sample Size 2,282 1,174 Percent Productivity Lost 1.1% 2.0% 0.00 Workdays Lost (Assuming 2.84 4.93 250-day Work Year) 1P-value for test of difference between low risk and high risk.©2010 Thomson Reuters 28
  • PHASE II: FACTOR ANALYSIS RESULTS Figure 1: Factor Loadings (importance) of each risk to each factor for all employees Factor 1: Biometric Risk Factor 2: Alcohol and Tobacco Factor 3: Emotional Health Risk Tobacco Tobacco Tobacco Alcohol Alcohol Alcohol Weight Weight Weight Emotional Emotional Emotional Exercise Exercise Exercise Triglycerides Triglycerides Triglycerides Cholesterol Cholesterol Cholesterol Blood sugar Blood sugar Blood sugar Blood Blood pressure Blood pressure pressure 0 0.5 1 -0.5 0 0.5 1 -1 -0.5 0 0.5 1 Loading (importance) Loading (importance) Loading (importance) • Males and females: includes • Males and Females: includes • Females: includes those at risk those at risk for blood pressure, those at risk for alcohol or for weight, emotional health or blood sugar, cholesterol, tobacco exercise triglycerides, or weight • Males: includes those at risk for©2010 Thomson Reuters emotional health or cholesterol Females Males 29
  • PHASE II: RISK FACTORS AND MEDICAL EXPENDITURES Outcomes and group of Predicted Predicted Impact on Impact as percent health risks scenario Mean dollars or days difference from scenario (95% CI) without the risk (95% CI) Medical Care Expenditures Annual expenditures Females High Biometric Lab Without risk(s) $3,952 $516 13.1% Values With risk(s) $4,468 ($146, $885) (3.7%, 22.4%) Alcohol - Tobacco Without risk(s) $3,910 $247 6.3% Use With risk(s) $4,157 (-$366, $861) (-9.4%, 22.0%) Emotional Health Without risk(s) $3,925 $500 12.7% With risk(s) $4,425 ($137, $863) (3.5%, 22.0%) Males High Biometric Lab Without risk(s) $2,540 $557 21.9% Values With risk(s) $3,097 ($200, $914) (7.9%, 36.0%) Alcohol - Tobacco Without risk(s) $2,652 $568 21.4% Use With risk(s) $3,220 (-$106, $1,243) (-4.0%, 46.9%) Emotional Health Without risk(s) $2,530 $561 22.2%©2010 Thomson Reuters With risk(s) $3,091 ($166, $956) (6.6%, 37.8%) Indicates a statistically significant difference between those at risk and those without risk. 30
  • PHASE II: RISK FACTORS AND PRESENTEEISM Impact as percent Impact on difference from scenario Outcomes and group of Predicted Predicted dollars or days without the risk health risks scenario Mean (95% CI) (95% CI) Presenteeism Annual unproductive days Females High Biometric Lab Without risk(s) 0.73 0.88 121.6% Values With risk(s) 1.61 (0.77, 1.00) (105.9%, 137.2%) Alcohol - Tobacco Without risk(s) 0.69 1.65 238.1% Use With risk(s) 2.34 (1.34, 1.95) (193.8%, 282.3%) Emotional Health Without risk(s) 0.74 0.86 115.7% With risk(s) 1.60 (0.75, 0.97) (100.7%, 130.7%) Males High Biometric Lab Without risk(s) 0.50 0.73 146.2% Values With risk(s) 1.23 (0.65, 0.81) (129.6%, 162.8%) Alcohol - Tobacco Without risk(s) 0.59 1.33 224.0% Use With risk(s) 1.93 (1.07, 1.59) (180.6%, 267.3%) Emotional Health Without risk(s) 0.54 0.87 159.7%©2010 Thomson Reuters With risk(s) 1.41 (0.76, 0.97) (139.8%, 176.9%) Indicates a statistically significant difference between those at risk and those without risk. 31
  • PHASE III: JOEM PUBLICATION©2010 Thomson Reuters 32
  • PHASE IV: BUSINESS APPLICATION DEVELOPING AN EXCEL-BASED MODEL Model Inputs The Model Consists of Enter the demographics Formulas that Combine characteristics and the the Inputs to Calculate baseline health risk profile Estimated Savings from: for a target population. • Medical Care Choose small, medium, or • Short-term Disability large risk reduction for each of the different health risk • Incidental Absence factors. • Workplace Productivity References tables of • Sales Performance regression equations and©2010 Thomson Reuters factor loadings from Phase II. 33
  • EXCEL MODEL INPUTS • Data points that can be customized for each run of the model include: – Population size – Percent female – Age distribution – Geographic distribution (by region) – Health plan distribution – Percent participation in program – Percent of participants ‗at risk‘ (by 9 risk factors) at baseline©2010 Thomson Reuters – Average daily wage and benefits load (for monetized presenteeism) 34
  • CHANGES IN RISK B C D • After customizing inputs, 2 Specify Hypothetical Changes in Risk Level the predicted impact on 3 Females Hypothetical Reduction Males Hypothetical Reduction risk level of the program 4 in Risk Level in Risk Level considered can be 5 FALSE FALSE modeled. 6 Biometric Risk • For each factor – 7 biometric, alcohol and 8 tobacco, and emotional 9 TRUE TRUE health, the impact on risk10 Alcohol & Tobacco can be selected for males 11 Risk and females. 12 – No change 13 FALSE FALSE – Small decrease 14 Emotional Health –©2010 Thomson Reuters Medium decrease 15 Risk – Large decrease 16 35
  • RESULTS FROM RISK CHANGES Potential Savings Due to Reduction in Health Risk Baseline Risk Reduction in Risk Change Minus Level Level Baseline Medical $15,912,606 $15,788,088 -$124,518 Expenditure Absence $4,218,869 $4,150,426 -$68,443 Payment Presenteeism $8,320,131 $7,870,190 -$449,940©2010 Thomson Reuters 36
  • MODEL OUTPUT: HEALTH INDEX Predicted Average Annual Cost per Employee by Population Health Index $16,000 Annual Cost per Employee Predicted by Model $14,000 Annual Cost per Employee Predicted by Model $12,000 $10,000 $8,000 $6,000 Baseline Health Index = 79 $4,000 Model Predicted PEPY Cost = $6,989 $2,000 Baseline Health Index = 79 Health Index after Reduction = 81 Model Predicted PEPY Cost = $6,989 Model Predicted PEPY Cost = $6,841©2010 Thomson Reuters $0 0 10 20 30 40 50 60 70 80 90 100 High Risk High Risk Population Health index Population Health Index Low Risk Low Risk Med & Rx Med & Rx Absence+STD Absence+STD Presenteeism Presenteeism Total Cost Total Cost Baseline Reduction Reduction Baseline Reduction 37
  • ©2010 Thomson Reuters PEPSICO STUDY38
  • PBC - OVERWEIGHT/OBESE ANALYSIS Adjusted predicted annual costs for employees by BMI *At least one difference significant at the 0.05 level $10,000 Diff = Diff = 25%, 29%, Normal $8,000 $613* $987 Overweight $6,000 Class I Diff = Diff = Class II $4,000 Diff = 26%, 7%, Diff = Class III 58%, $186* $49 10%, $2,000 $111* $28 $0 Medical STD Total Absences Presenteeism WC Difference between combined overweight/ obese categories and 74% of the sample is©2010 Thomson Reuters normal weight is displayed overweight or obese 39
  • MODEL INPUTS TAB Model Inputs (you can click in any of the white boxes and type a new value) Enter distribution (%) for demographics and health risks 41.1% 18-34 27,000 Number of employees Employee Age 28.4% 35-44 $204 Average daily wage & benefit Statistics 22.9% 45-54 0% Percent who will participate (Total = 100%) 7.6% 55-64 Intervention $0 Annual cost per participant Gender 11.7% Female $2,505 Medical expenditure Baseline 20.2% Northeast $664 Workers Compensation Annual Geographic Health and 10.6% North Central $293 STD payment Region Productivity Costs per 42.4% South 2.4% Presenteeism (%) Employee (Total = 129.1%) 26.8% West 2.7 Health-related absence days 29.1% Sales Select a work site from the drop-down list: 7.9% Professional/Non-manager National Total Selecting a work site will populate the model inputs with 16.3% Manager values specific for the site. Sites with 100+ employees are Job Type listed individually, sites with 50-99 employees are grouped 9.1% Technician by geographic region; sites with fewer than 50 employees 5.4% Clerical/Office are grouped nationally. 31.6% Laborer/Production (Total = 100%) 0.7% Unknown 42.8% Overweight 21.8% Obese Class I Health Risks 7.1% Obese Class II 3.1% Obese Class III 17.2% High blood pressure 3.0% High blood glucose 12.1% High total cholesterol 14.3% Physical inactivity 14.9% Poor diet©2010 Thomson Reuters 14.6% Stress 5.0% Depression 23.9% Tobacco use 10.1% Alcohol 0.6% Type I Diabetes 40
  • MODEL RESULTS – PROJECTED CHANGE 27,000 Number of employees Employee $204 Average daily wage & benefit Statistics 30% Percent who will participate Intervention $100 Annual cost per participant Model Results (you can click in any of the white boxes and type a new value for percent risk reduction) Enter percent risk reduction for the health risks; savings will update dynamically. Risk Per Total Health Risk Projected Savings from Risk Reduction Reduction Employee Participants Overweight 10% Baseline $2,505 $20,287,665 Health Index Obese Class I 10% Medical expenditure Risk Reduction $2,426 $19,654,463 (after Risk Reduction) Obese Class II 10% Savings = $78 $633,202 92.5 Obese Class III 10% Baseline $664 $5,381,721 High blood pressure 10% Workers Compensation Risk Reduction $644 $5,215,554 0 => highest possible risk 100 => lowest possible risk High blood glucose 10% Savings = $21 $166,167 High total cholesterol 10% Baseline $293 $2,374,353 Return on Investment Physical inactivity 10% STD payment Risk Reduction $279 $2,259,365 (ROI) Poor diet 10% Savings = $14 $114,988 $1.26 Stress 10% Baseline $323 $2,620,156 Depression 10% Presenteeism cost Risk Reduction $318 $2,578,173 ROI is the net savings for each dollar invested. Tobacco use 10% Savings = $5 $41,983 An ROI of $1.00 indicates break even. Alcohol 10% Baseline $542 $4,388,019©2010 Thomson Reuters Health-related Absence Risk Reduction $533 $4,320,296 Annual Savings/Employee With a 1 Point Increase Savings = $8 $67,723 in the Health Index Total Savings $126 $1,024,064 $124 41
  • HEALTH INDEX SCORE BY WORKSITE 93.9 88.3 92.0 91.4 92.1 91.5 100 Healthier 80 Population Health Index 60 Higher Risk 40 20©2010 Thomson Reuters 0 Loc A Loc B Loc C Loc D Loc E PBC Avg 42
  • THOMSON REUTERS WORKFORCE WELLNESS INDEX • Background – Modifiable health risk factors are associated with increased healthcare and productivity costs • Objectives – Devise a methodology to create a health risk score that can be applied to health risk assessment (HRA) data and correlates with costs associated with health risk factors – Devise a methodology that allows comparison of a population subset to a total health risk score©2010 Thomson Reuters 43
  • DATA SOURCES • MarketScan Medical and Drug Claim Database – Eligibility, Medical and Drug Paid Claims Data for Self-Insured Employers and Health Plans, 2005-2009 – Over 25 Million Covered Lives in 2009 – Eligibility and Medical Claims were used to Derive Employee Demographics and Comorbidities (for Risk Adjustment) – Medical and Drug Claims were used to Estimate Prospective Healthcare Costs • MarketScan Health and Productivity Management Database – Health Risk Assessment Survey, Absenteeism, Workers Compensation and Short Term Disability Data, 2005-2009 – Over 2 Million HRAs in 2009 – Linkable to the MarketScan Medical and Drug Claim Database©2010 Thomson Reuters – HRA Survey Questions were used to Estimate Behavioral Risk Prevalence Rates and to Identify the Presence/Absence of High Risks 44
  • THOMSON REUTERS WORKFORCE WELLNESS INDEX • Two indexes were constructed • MARKETSCAN® INDEX: Prevalence and cost of 8 risk factors based on MarketScan medical and drug claims matched to Health Risk Assessment (HRA) data (privately insured; adjusted to U.S. demographics) - Body Mass Index (BMI) - Tobacco Use - Blood Glucose - Alcohol Use - Blood Pressure - Stress (U.S. rates not available) - Cholesterol - Exercise (U.S. rates not available) • U.S. INDEX: Prevalence of 6 risk factors for U.S. employed, privately insured population age 18-64 with MarketScan cost©2010 Thomson Reuters weights applied 45
  • SAMPLE • Active full-time employees • Ages 18-64 • Enrolled in non-capitated health plans • Continuously enrolled for 365 days before and after the index HRA date • Non-pregnant individuals©2010 Thomson Reuters 46
  • HEALTH RISK INDEX—DATA SOURCES High Risk Definitions Risk Factor High Risk Definition BMI BMI >= 30 Blood Pressure Systolic >=140 or diastolic >=90 Cholesterol Total Cholesterol >= 240 Glucose Total Glucose >= 126 Currently smoke cigarettes or use any form of Tobacco tobacco Alcohol More than 2 drinks per day Sometimes or Often feel stressed and have Stress trouble coping Exercise less than two days per week or less than 20 minutes per day or non-exerciser/light exerciser in the previous©2010 Thomson Reuters Exercise month 47
  • METHODS • Prevalence of health risks was calculated for the sample population – To compensate for possible differences in demographic composition of the MarketScan HRA sample and the national employed workforce, adjustment weights were applied when computing the yearly prevalence rates from the MarketScan HRA sample – Adjustment weights were derived from the Current Population Survey for the years 2005-2009©2010 Thomson Reuters 48
  • METHODS • Regression models were used to estimate the importance weights used to derive the overall index score – Importance weights were computed from risk factor coefficients from the regression model that estimated the cost effect of the risk factors and other covariates. – Each risk factor importance weight can be interpreted as the annual percentage increase in medical and drug costs due to presence of a risk factor, controlling for all other risk factors, comorbid conditions and employee characteristics©2010 Thomson Reuters 49
  • METHODS • The U.S. Index was computed as follows: – The Behavioral Risk Factor Surveillance Survey (BRFSS) and the National Health and Nutrition Examination Survey (NHANES) were used to estimate behavioral risk prevalence rates for the insured, employed population of the U.S. as a whole – Importance weights were derived from the MarketScan claims database – The U.S. Index was then computed in a manner similar to the Workforce Wellness Index©2010 Thomson Reuters 50
  • WORKFORCE WELLNESS INDEXES U.S. AND MARKETSCAN 2005-2009 87.0 86.5 86.0 85.5 85.0 84.5 84.0 83.5 83.0 82.5 2005 2006 2007 2008 2009 U.S. Wellness Index MarketScan Wellness Index©2010 Thomson Reuters Note: Each index is a composite of 6 risk factors: BMI, Blood Glucose, Blood Pressure, Cholesterol, Tobacco Use, Alcohol Use 51
  • RESULTS • Between 2005 and 2009, the U.S. Workforce Wellness Index worsened, declining from 86.4 to 84.4 • The MarketScan sample improved, increasing from 84.1 to 86.2. An index of 100 represents the ideal state where there are no behavioral risk factors present in the employed population and, therefore, no healthcare costs due to these risks©2010 Thomson Reuters 52
  • WORKFORCE WELLNESS INDEX ESTIMATED ANNUAL COST IMPACT1 $450 $400 $350 $300 $250 $200 $150 $100 $50 $- $(50) BMI Blood Pressure Cholesterol Glucose Tobacco Alcohol Implied Cost Impact (based on 2009 Prevalence Rates) 1Based on cost and prevalence rates in MarketScan data sets©2010 Thomson Reuters Note: Cholesterol and Alcohol Use statistically have no medical/drug cost impact 53
  • DISCOVERY VITALITY WELLNESS HEALTHY COMPANY INDEX©2010 Thomson Reuters 54
  • EMPLOYEE HEALTH ASSESSMENT©2010 Thomson Reuters 55
  • REPORTING BMI AND NUTRITION©2010 Thomson Reuters 56
  • ORGANIZATIONAL HEALTH©2010 Thomson Reuters 57
  • ©2010 Thomson Reuters58 LEADERSHIP SUPPORT
  • AND THE WINNERS ARE…©2010 Thomson Reuters 59
  • SUMMARY WORKFORCE WELLNESS INDEXES • There are lots of them out there • Some are qualitative, others quantitative, and yet others are both • They aim to connect organizational health, individual risk factors, and financial metrics • The goals – to come up with one number that reflects a composite health/cost score – like the ―Dow Jones‖ industrial average • Measures the impact of improving behavioral risk©2010 Thomson Reuters factors on healthcare cost in employed populations 60