Welcome to Scribd, the world's digital library. Read, publish, and share books and documents. See more
Standard view
Full view
of .
Look up keyword
Like this
0 of .
Results for:
No results containing your search query
P. 1
2010 Harvard Wellness Program Meta Study Health Affairs

2010 Harvard Wellness Program Meta Study Health Affairs

Ratings: (0)|Views: 40 |Likes:
Published by jmrazote

More info:

Published by: jmrazote on Jun 26, 2012
Copyright:Attribution Non-commercial


Read on Scribd mobile: iPhone, iPad and Android.
download as PDF, TXT or read online from Scribd
See more
See less





By Katherine Baicker, David Cutler, and Zirui Song
Workplace Wellness Programs CanGenerate Savings
Amid soaring health spending, there is growing interest in workplace disease prevention and wellness programs to improve healthand lower costs. In a critical meta-analysis of the literature on costs andsavings associated with such programs, we found that medical costs fall by about $3.27 for every dollar spent on wellness programs and thatabsenteeism costs fall by about $2.73 for every dollar spent. Althoughfurther exploration of the mechanisms at work and broader applicability of the findings is needed, this return on investment suggests that the wider adoption of such programs could prove beneficial for budgets andproductivity as well as health outcomes.
nanenvironmentofsoaringhealthcarespending, policymakers, insurers, andemployers express growing interest inmethodsofimprovinghealthwhilelow-ering costs. Much discussion has takenplace about investment in disease preventionand health promotion as a way of achieving bet-ter health outcomes at lower costs. PresidentBarack Obama has highlighted prevention as acentralcomponentofhealthreform,ashavema- jor congressional reform proposals.
 Work-place-based wellness programs, which could af-fect prevention, have been showcased in thesereform proposals, the popular press, and con-gressional hearings.
This enthusiasm for workplace programsstemsinpartfromthefactthatmorethan60per-centofAmericansgettheirhealthinsurancecov-erage through an employment-based plan,
aswell as from the recognition that many employ-ees spend the majority of their waking hours inthe workplace
which makes it a natural venuefor investments in health. There are several rea-sons that employers might benefit from invest-ments in employee wellness. First, such pro-grams might lead to reductions in health carecosts and thus health insurance premiums. Sec-ond, healthier workers might be more produc-tive and miss fewer days of work. These benefitsmay accrue at least partially to the employer (such as through improved ability to attractworkers), even if the primary benefits accrueto the employee.These factors may motivate the increasing in-terest in such programs among employers
andespecially large employers. In 2006, 19 percentofcompanieswith500ormoreworkersreportedofferingwellnessprograms,whilea2008survey of large manufacturing employers reported that77 percent offered some kind of formal healthandwellnessprogram.
Consistentwiththeevi-dence presentedbelow,small firms seemsloweto offer such programs, and many of the pro-grams offered are still quite limited in scope.
Several well-publicized case studies have sug-gested a positive return to employers
invest-ment in prevention. For every dollar investedin the program, the employer saves more thanthe dollar spent. The Citibank Health Manage-ment Program reported an estimated savingsof $4.50 in medical expenditures per dollar spent on the program.
Studies from the Cali-fornia Public Employees Retirement System(CalPERS), Bank of America, and Johnson andJohnson have similarly estimated sizable healthcaresavingsfromwellnessprograms.
doi:10.1377/hlthaff.2009.0626HEALTH AFFAIRS 29,NO. 2 (2010):
©2010 Project HOPE
The People-to-People HealthFoundation, Inc.
Katherine Baicker
(Kbaicker@hsph.harvard.edu) is aprofessor of health economicsat the School of PublicHealth, Harvard University, inBoston, Massachusetts.
David Cutler
is a professor ofeconomics at HarvardUniversity.
Zirui Song 
is a doctoralcandidate at Harvard MedicalSchool.FEBRUARY 2010 29:2HEALTH AFFAIRS
this anecdotal evidence of high returns, how-ever, most employers do not engage in wide-scale workplace wellness promotion practices.The 2004 National Worksite Health PromotionSurvey showed that only 7 percent of employersoffered comprehensive programs
of the typespecified in the recommendations of the influ-entialInstituteofMedicinereport
 HealthyPeople 2010
Theseincludehealtheducation,worksitescreenings linked to appropriate medical care,and the integration of the program into corpo-rate or organizational structure.Someempiricalstudiesattempttoestimatethereturnoninvestmentforemployerwellnesspro-gramsmore systematically, but shortcomings inthisliteratureleavethequestionunresolved.
Inparticular, most studies lack an adequate com-parisonorcontrolgroup,andarethusnotabletoaccount for possible unobserved variables or alternative pathways that might be responsiblefor observed differences in costs between well-ness program participants and nonparticipants(rather than those differences
being attributa-ble to the wellness program itself). This leavesopen the possibility of selection bias
for exam-ple, if the healthiest employees were most likely toenrollinvoluntarywellnessprograms,acom-parison of participants and nonparticipantsmight suggest that the programs are improvinghealth more than they really are.Low response rates, inexact case-controlmatching, and potential publication bias (stud-iesfindinghighreturnsmaybemorelikelytobepublished)alsocallintoquestiontheevidenceof high returns. In addition, Sean Nicholson andcolleagues show that common methods used by employers to calculate costs and benefits of health-related investments might not reflectthetrueimpactoftheseprograms.
Theseshort-comings mean that even the limited evidenceavailable might not be robust or generalizable.In this study we conducted a meta-analysis of the literature on costs and savings associatedwith employer-based wellness promotion poli-cies. We began by screening existing studiesfor analytical rigor, and then we compiled stan-dardizedestimatesofreturnoninvestmentfromthose studies. We focused on studies for whichthere was a comparison group of nonpartici-pants, and we examined effects of wellness pro-gram interventions on health care costs andabsenteeism. Wefoundalargepositivereturnoninvestmentacross these rigorous studies, which suggeststhat the wider adoption of such programs couldprove beneficial for budgets as well as health.That they have been implemented so selectively,however, necessitates further research into thelikely effects of broader adoption.
Study Data And Methods
 We conducted a primary literature search fromprior peer-reviewed meta-analyses of employeewellness programs, as well as a computerizedsearch of MEDLINE, Lexis-Nexis, and other health and social science databases. Searchterms included
disease management,
return oninvestment.
This produced an initial sample of more than 100 peer-reviewed studies of employ-ee wellness programs spanning the past threedecades. Among these peer-reviewed studies, we re-stricted our analysis to studies that satisfiedthefollowingcriteria:(1)theyhadawell-definedintervention; (2) they had a well-defined treat-mentandcomparisongroup,evenifthecompar-ison group was not strictly randomly assigned;and (3) they represented analysis of a distinctnewintervention,ratherthanfurtheranalysisof an intervention already examined in one of theother studies.We performed additional analysison the subset of these studies that reported
estimatesofthestudyout-come (comparing the change in the outcomefrom before the program to after the programin the treatment group to the change in the out-come over the same period for the comparisongroup), or the raw data allowing for this calcu-lation.
 Applying these criteria narrowed our sampletothirty-twooriginalpublications.Thesestudiesare listed in Appendix Table 1.
Two of thesestudies reported results of multiple separate in-terventions;wetreatedtheseasseparatestudies.Severalotherstudiesreportedtheresultsofmul-tiple interventions, but because participantswere allowed to self-select into interventionarms, we treated these as a single case each.Thus, the thirty-two original publications gaveus an effective sample of thirty-six studies. Of these,twenty-twolookedatemployeehealthcarecosts,andtwenty-twolookedatemployeeabsen-teeism (eight examined both). We catalogued the characteristics of the firmsthat undertook these employee wellness pro-gramsandthequalitativedimensionsofthepro-grams themselves.We analyzed the health carecost and absenteeism studies separately, but wealso converted the absenteeism results intodollar cost units using a uniform wage rate toconstruct comparable estimates of return oninvestment.
Study Results
Morethan90percentof employeewellnessprogramsinoursamplewereimplemented in large firms (those with more
than1,000workers).One-fourthexaminedwell-ness programs at employers with more than10,000 workers. A number of industries wererepresented: 25 percent of sampled employerswere in financial services; 22 percent in manu-facturing;and16percentinschooldistricts,uni- versities, and municipalities. Other industriesrepresented included utilities, telecommunica-tions, energy, pharmaceuticals, and makers of consumer products. Ten studies took placeacross multiple locations, often the employer headquarters and satellite locations; some wereimplemented across multiple employers.
Wecan characterize the employee wellness pro-grams in the study sample along two dimen-sions: the method of delivery and the focus of intervention(Exhibit 1). The method of delivery characterizes how the intervention was carriedout. By far the most frequently used method of delivery is the health risk assessment
a survey that gathers baseline self-reported health datafrom the employee, which are in turn used by the employer to tailor the subsequent interven-tion.
The health risk assessment is used in80 percent of the studies in our sample; it mostcommonly serves as the initial intervention or requirement for participation in the wellnessprogram.Participation is almost always voluntary among employees at the treatment site, makingselection bias a major concern. Assessments arecommonly used in conjunction with a clinicalscreening of risk factors, including blood pres-sure, cholesterol, and body mass index (BMI).Importantly, the assessment tool provides theemployee with information on risk factors thatmotivate participation. The majority of pro-grams that did not use the assessment methodfeaturedanon-sitegymnasiumorworkout facil-ity, which employees were encouraged to use.The second most common wellness interven-tion mechanism was the provision of self-helpeducation materials, individual counseling withhealthcareprofessionals,oron-sitegroupactiv-ities led by trained personnel. In our sample,about 40 percent of studies included the use of self-helpmaterials;40percentofferedindividualcounseling; and 35 percent featured on-sitegroup activities, classes, or seminars. Most pro-grams offered a combination of these inter- ventions.Theuseofincentivestomotivateparticipationwas seen in 30 percent of programs. Incentiveswere most commonly bonuses and reimburse-ments for program participation, but they alsoincludedthepaybackofdownpaymentspriortoparticipation.Suchcasesmayinvolveanemploy-er 
s withholding a small portion of employeecompensation until program participation oc-curs. Incentives have become more common inrecent interventions.The most common foci of the programs wereobesity and smoking, the two top causes of pre- ventable death in the United States. More than60percentoftheprogramsexplicitlyfocusedonweight loss and fitness. All but three of the re-maining programs focused on either multiplerisks or risks specific to the participant. Half of the programs focused on smoking, often inconjunctionwithobesity.Seventy-fivepercentof programs focused on more than one risk factor,including stress management, back care, nutri-tion, alcohol consumption, blood pressure, andpreventive care, in addition to smoking andobesity.
Twenty-two studies reported on the impact of wellness programs on employee health carecosts (Exhibit 2). The average sample size of intervention groups exceeded 3,000 employees,and the size of comparison groups averaged
Summary Of Characteristics Of Worksite Wellness Programs StudiedMethod of delivery Percent of firms
Health risk assessment 81Self-help education materials 42Individual counseling 39Classes, seminars, group activities 36Added incentives for participation 31
Focus of intervention
Weight loss and fitness 66Smoking cessation 50Multiple risk factors 75
calculations based on 36 studies described in Appendix Table 1, available online at http://content.healthaffairs.org/ cgi/content/full/29/2/hlthaff.2009.0626/DC1FEBRUARY 2010 29:2HEALTH AFFAIRS

You're Reading a Free Preview

/*********** DO NOT ALTER ANYTHING BELOW THIS LINE ! ************/ var s_code=s.t();if(s_code)document.write(s_code)//-->