Comparação bem estar com doenças crônicas

CME AVAILABLE FOR THIS ARTICLE AT ACOEMORG

Comparing the Contributions of Well-Being and Disease Statusto Employee Productivity

William M Gandy EdD Carter Coberley PhD James E Pope MD Aaron Wells PhDand Elizabeth Y Rula PhD

Objective To compare employee overall well-being to chronic disease statuswhich has a long-established relationship to productivity as relative contrib-utors to on-the-job productivity Methods Data from two annual surveys ofthree companies were used in longitudinal analyses of well-being as a pre-dictor of productivity level and productivity change among 2629 employeeswith diabetes or without any chronic conditions Results Well-being wasthe most significant predictor of productivity cross-sectionally in a modelthat included disease status and demographic characteristics Longitudinallychanges in well-being contributed to changes in productivity above and be-yond what could be explained by the presence of chronic disease or otherfixed characteristics Conclusions These findings support the use of well-being as the broader framework for understanding explaining and improvingemployee productivity in both the healthy and those with disease

ldquoT he level of productivity is the single most important deter-minant of a countryrsquos standard of living rdquo1 Poor employee

health not only contributes to reduced profitability for companiesthrough increased operational costs it also results in indirect costfrom decreased productivity23 Moreover these issues threaten USwell-being and competitiveness in the global environment4

Although the direct costs of poor employee health havebeen well established5ndash10 recent literature has begun to focus onthe indirect costs of poor employee health in the form of lostproductivity11ndash14 By focusing on improving employee health em-ployers can reduce not only the direct costs (ie medical and phar-macy) of poor health but also indirect costs due to absenteeismand presenteeism13ndash21 The costs of lost productivity are substantialMultiple studies have found that the indirect costs of poor employeehealth can exceed those due to medical and pharmacy131422 On thebasis of his research Loeppke et al1314 conclude that a new moreintegrated and comprehensive approach to health care managementthat not only takes account of traditional medical and drug costs butalso addresses the costs of absenteeism and presenteeism is needed

Presenteeism has traditionally been defined as productivityloss as a consequence of attending work while ill20 Neverthelessproductivity loss from suboptimal performance on the job may arisefrom various nonndashhealth-related causes such as stress family dif-ficulties and financial concerns23 Because of advancements in thetheory and measurement of productivity research has begun to rede-fine presenteeism as an outcome that is not solely a consequence ofsuboptimal health and several instruments have been developed tomeasure presenteeism stemming from both health- and nonndashhealth-

From the Center for Health Research Healthways Inc Franklin TennThis study was funded by Healthways Inc and all authors are employees and

shareholders of this companyAuthors Gandy Coberley Pope Wells and Rula have no relationshipsconditions

circumstances that present potential conflict of interestThe JOEM editorial board and planners have no financial interest related to this

researchAddress correspondence to Elizabeth Y Rula PhD Center for Health

Research Healthways Inc 701 Cool Springs Blvd Franklin TN 37067(elizabethrulahealthwayscom)

Copyright Ccopy 2014 by American College of Occupational and EnvironmentalMedicine

DOI 101097JOM0000000000000109

Learning Objectives Become familiar with recent reports suggesting that pro-

ductivity is affected by nonndashhealth-related as well as health-related factors

Summarize the new findings on well-being as a predictor ofproductivity in employees with and without diabetes

Discuss the construct of well-being as part of emerging morecomprehensive strategies to optimize productivity

related sources (eg Health and Work Performance Questionnaire[HPQ] global presenteeism measure Well-Being Assessment forProductivity [WBA-P])1524

Similarly researchers have expanded the view of factors thatmay influence productivity There is a considerable body of evidencethat demonstrates that exposure to stress can adversely affect workperformance and prolonged exposure can produce negative healtheffects such as obesity heart disease depression etc1625ndash30 Morerecent research has explored aspects of work-related factors em-ployee engagement financial stress and depressionanxiety as ad-ditional factors that influence employee productivity Each of thesewere found as significant contributors to self-rated job performancein addition to physical heath factors3132 Merrill and colleaguesconclude from their work that ldquoemployers can maximize theiremployeesrsquo job performance by using a multipronged integratedapproach to well-being improvementrdquo32

Despite the evidence that efforts to improve work performanceshould focus on areas beyond physical health to date most re-search documenting the effects of lost productivity has focused morenarrowly on physical causes as the contributing factor12ndash141733ndash38

Collins et al39 evaluated the relative influence of employee demo-graphics health risk factors chronic conditions health care claimsand employment information and found that chronic conditions werethe most important determinant of work impairment Neverthelesseven among data available in health care claims conditions beyondthe ldquocore chronicsrdquo including depression anxiety and fatigue haveemerged as significant drivers of productivity loss14

With the recognition that factors beyond chronic disease mustbe considered to make the largest impact on productivity the selec-tive approach to examining these factors in existing studies repre-sents a limited view There is still the gap of a more comprehen-sive evaluation of the elements that influence performance In factLoeppke et al14 argue that health risk assessments (HRAs) representan advancement over claims data for use by employers focused ona broader set of ldquobusiness-relevant outcomesrdquo yet HRAs still col-lect information that is primarily focused on physical health and thebehaviors that influence physical health Any performance deficitsamong employees without existing disease or health risks or amongemployees who are unaware of their health risks cannot be identi-fied for intervention by using the information collected by an HRATherefore an important limitation of prior research given advancesin the field showing that a range of factors influence productivityis the absence of a more comprehensive measure of the factors thatinfluence onersquos ability to perform optimally

Copyright copy 2014 Lippincott Williams amp Wilkins Unauthorized reproduction of this article is prohibited

252 JOEM Volume 56 Number 3 March 2014

JOEM Volume 56 Number 3 March 2014 Well-Being and Employee Productivity

In their groundbreaking article ldquoBeyond Money Toward anEconomy of Well-Beingrdquo Diener and Seligman40 contend that well-being should be the ultimate goal around which economic healthand social policies are developed not simply because well-being isan important indicator in itself but also because well-being is manytimes a cause of other valued outcomes such as worker productivityand rewarding relationships They conclude that organizations andnations would benefit from regular assessment of well-being andfinding ways to improve it An individualrsquos well-being is sensitivenot only to physical health factors that contribute to absenteeism andpresenteeism but also to nonndashhealth-related factors (eg psychoso-cial environmental financial work etc) that may serve as barriersto optimal productivity

In addition to the macrolevel argument put forth by Dienerand Seligman40 that well-being is an important indicator of societalhealth this construct is gaining acceptance as an integral part ofhealth care measurement continues to gain adherents Both the Na-tional Institutes of Health and the Centers for Disease Control andPrevention have recognized well-being as an important indicator ofphysical and mental health and therefore an essential component inthe measurement of public health outcomes4142

The purpose of this study was to evaluate well-being as mea-sured by the Well-Being Assessment (WBA) as such a comprehen-sive measure with the aim of maximizing opportunity for produc-tivity improvement across entire populations regardless of healthstatus Individual well-being scores (IWBSs) as measured by theWBA have demonstrated strong relationships to a number of stan-dard health care outcomes including expenditures inpatient admis-sions and emergency department visits4344 as well as productiv-ity outcomes in the form of absenteeism and presenteeism

2431324546

lending support to this well-being measure as the broader frameworkfor understanding what influences business outcomes including pro-ductivity Nevertheless research to date has not directly comparedwell-being with the traditional factors known to affect productivityHere IWBS is compared with the prevalent chronic condition di-abetes to determine whether well-being contributes to productivitylevels and changes over and above what can be explained by physicalhealth alone

METHODSStudy Design

The study used a longitudinal retrospective panel study de-sign using survey data collected in 2 consecutive years (T1 and T2)The data for the panel were composed of employees from three UScompanies One company was a large health care vendor anotherwas a nationally based insurance company and a third was an inter-nationally based manufacturing company Study eligibility requiredcompletion of the WBA at T1 and T2 with valid well-being andproductivity scores from both time points and a documented agebetween 18 and 64 years Two study groups were identified fromthe data panel (1) nondiseasedmdashindividuals who self-reported nochronic conditions at T1 and T2 (n = 1858) and (2) diabetesmdashindividuals who self-reported a diagnosis of diabetes at T1 and T2(n = 771) Individuals in the diabetes group were not restrictedon the basis of the presence or absence of other comorbid condi-tions Because of the negligible risk retrospective design and use ofde-identified data this study was exempt from institutional reviewboard approval based on exclusion criteria outlined in the US Codeof Federal Regulations (45 CFR sect46101)

Measures

Well-Being WBA and IWBSThe WBA was developed as an extension of the Gallup-

Healthways Well-Being Index (WBI)4748 a community survey ofwell-being for use with employer populations and other organiza-

tions such as health plans and health systems interested in totalpopulation health and well-being The instrument includes healthrisk and productivity measures in addition to the WBI item set Indi-vidual well-being score was developed from WBI items and domainsto allow calculation of well-being at the individual level49 The IWBSis calculated using 40 questions from the following six domains ofwell-being that are included in the WBI and WBA physical healthemotional health healthy behaviors work environment basic ac-cess and life evaluation Each domain is weighted equally in thecalculation of the IWBS because they are in the WBI and scoresrange from 0 to 100 for each respondent

Productivity WBA-P Overall Score and WBAP_Alt ScoreThe primary measure of productivity in the study was the

WBA-P that is administered as part of the WBA and provides an in-formative evaluation of on-the-job productivity loss (presenteeism)due to well-beingndashrelated barriers Criterion-related validity of theWBA-P has been established through multivariate analysis to a num-ber of health and well-being measures24

The WBA-P score is taken from 11 items on the WBA withthe shared question stem ldquoDuring the past 4 weeks (28 days) howoften have you had trouble at work concentrating or doing your bestbecause of rdquo and then lists 11 possible reasons or barriers Scor-ing of this measure ranges from 0 (not at all) to 100 (a lot for all11 reasons)24 For the purposes of the study the WBA-P overallscore was converted from a measure of productivity loss to one ofproductivity functioning This was accomplished by taking the com-plement of the WBA-P that is WBAP_Alt = 100 minus (WBA-P) Thisconverted score is designed to reduce or eliminate zero scores for sta-tistical analyses Higher scores reflect higher levels of productivity

Productivity (Secondary Measure) HPQ Self-RatedPerformance Scale

The HPQ contains a global self-rating of job performancemeasured on a 0 to 10-point scale that is considered an absolutemeasure of presenteeism15 this question was included in the WBAThe item reads ldquoOn a scale from 0 to 10 where 0 is the worst jobperformance anyone could have at your job and 10 is the performanceof a top worker how would you rate the usual performance of mostworkers in a job similar to yoursrdquo Responses to this global item wereused in sensitivity analyses to further strengthen conclusions drawnwith the primary productivity measure of the study the WBA-P

Statistical MethodsFirst cross-sectional linear models were used to examine

the relative contribution of well-being and other selected covari-ates (Table 1) including disease status on productivity among

TABLE 1 Study Model Variables

Variable Type CategoriesRange

Productivity score(WBA-P_Alt)

Continuous 0ndash100

Disease status Categorical Nondiseased or diabetes

Well-being (IWBS) Continuous 0ndash100

Age yrs Continuous 18ndash64

Gender Categorical Men or women

Comorbidities count Continuous 1ndash21

Employer Categorical A B or C

Time Categorical T1 = 2010 T2 = 2011

Dependent variableIWBS individual well-being score WBAP_Alt Well-Being Assessment for

Productivity complement of score


Ccopy 2014 American College of Occupational and Environmental Medicine 253

Gandy et al JOEM Volume 56 Number 3 March 2014

nondiseased and those with a chronic condition (ie nondiseased vsdiabetes)

To assess how changes in well-being relate to changes in pro-ductivity fixed-effects regression (ie first differencing) was used tofurther test the robustness of well-being in explaining changes in pro-ductivity Fixed-effects techniques have often been referred to as theldquocriterion standardrdquo for observational studies50 Among fixed-effectstechniques fixed-effects regression was chosen over fixed-effects es-timation found in mixed models because it provides a more straight-forward way to assess the impact of changes in well-being (IWBS)on changes in productivity (WBAP_Alt) The appeal of fixed-effectstechniques is their ability to control for bias from all time-invariantsources whether measured or unmeasured thus addressing thepotential limitation of omitted variable bias51ndash54 Also fixed-effectsregression removes hierarchical sources of bias due to nesting (egemployees nested within companies) These attributes of the tech-nique are achieved through use of the within-person variation (eachcase in essence serves as its own control)

Although fixed-effects techniques cannot provide reliable es-timates for the effects of time-invariant sources this limitation wasof minor consequence to this study given that a detailed evaluation ofcovariate contribution was not an objective Aligned with our studygoals first differencing provides a powerful test of the robustness ofobserved time-varying variables50 Additional sensitivity analyseswere also conducted to assess the stability of results All statisticalanalyses were conducted using SAS 92 (SAS Institute Inc CaryNC)

RESULTSOf the 2629 employees who qualified for the data panel 771

(293) self-reported they had diabetes at T1 and T2 with 1858(707) self-reporting the absence of the surveyed disease conditionsat T1 and T2 Participant demographics are presented in Table 2 bydisease status The study group was largely composed of men thoughgender is more balanced in the diabetes group As the table revealsthe nondiseased members have higher productivity and well-beingscores than their diabetes counterparts and are also younger onaverage Of the three employer groups employer B had the lowestpercentage of members with diabetes

Contribution of Well-Being on Productivity in theContext of Disease Status

Cross-Sectional Examination at T1 and T2General linear models were used in a cross-sectional analysis

of well-being and other variables as contributors to productivity

TABLE 2 Participant Demographics (N = 2629)

Disease Status Group

Diabetes NondiseasedVariable (n = 771) (n = 1858)

Productivity score (WBAP_Alt mean) 806 878

Well-being (IWBS mean) 693 816

Age (mean) yrs 505 379

Women 488 415

Number of comorbid conditions (mean) 53 0

Employer A ( by group n = 881) 342 658

Employer B ( by group n = 1472) 252 748

Employer C ( by group n = 276) 359 641

IWBS individual well-being score WBAP_Alt Well-Being Assessment forProductivity complement of score

(WBAP_Alt) at each time point (Table 3) A positive coefficient forthe continuous variables of age and well-being indicates a positiverelationship with productivity A negative coefficient for categoricalvariables (ie gender disease status and customer) reflects lowerproductivity compared with the reference group In both years well-being is the most influential predictor and is fairly consistent in effectNevertheless there seems to be a general increase in the influenceof the other covariates going from T1 to T2 suggesting an effect fortime

Fixed-Effects Regression (First Differencing)The results of fixed-effects regression evaluating the influence

of well-being change on changes in productivity (WBAP_Alt) arepresented in Table 4 Because this approach controls for all time-invariant factors whether measured or unmeasured fixed covariatesare not included in the structural model Results revealed a largestatistically significant coefficient for the relationship between well-being and productivity indicating that changes in well-being explainchanges in productivity beyond what can be explained by diseasestatus or other fixed characteristics

Sensitivity TestingBecause of the disproportionate number of nondiseased to

diabetes group members sensitivity tests were conducted on thegroups separately to confirm that the results were not driven by aparticularly strong relationship between well-being and productivityin one group or the other Specifically by evaluating the stability ofthe fixed-effects regression results between the two groups it is pos-sible to evaluate whether well-being is similarly predictive regardlessof disease status Results presented in Tables 5 and 6 reveal thatchange in productivity (WBAP_Alt) is significantly related to changein well-being in both the nondiseased and the diabetes groups

As an additional sensitivity test to ensure that conclusionsare robust to different measures of productivity changes in well-being (IWBS) were modeled in predicting changes in the globalwork performance measure from the HPQ (Table 7) Well-being wasfound to make a statistically significant contribution to performanceself-rating corroborating findings with the WBA-P

DISCUSSIONIn the new globally competitive marketplace human capi-

tal has become the competitive advantage that employers can nolonger afford to take for granted Research to date linking pro-ductivity to individual factors has focused on health problems asthe source of worker productivity loss with particular focus onchronic conditions12ndash141733ndash395556 Nevertheless a growing bodyof literature is demonstrating that a wider array of personal factorsnamely individual well-being together represent powerful predictorsof outcomes that affect business performance and competitivenessmdashworker health care costs hospital utilization performance produc-tivity and retention43ndash4657ndash60 Nevertheless the research has yet todirectly compare worker well-being to the traditional view of poorhealth (ie chronic disease) as variables predicting performance

The analyses conducted in this study address this gap andshow that well-being provides explanatory power for productivityabove and beyond what can be attributed to disease status usingthe prevalent condition diabetes as the focus disease As others haverecognized a focus solely on physical health as the basis to addressproductivity has important limitations61 Nevertheless research ex-amining the relative influence of physical health factors includingchronic conditions health risk factors and health care claims to-gether with demographic factors and work factors on productivitystill found that the presence of a chronic condition was the mostimportant predictor39 The present findings offer IWBS as a highlysignificant predictor of on-the-job productivity and more predictivecompared with other factors including disease status Specifically


254 Ccopy 2014 American College of Occupational and Environmental Medicine


TABLE 3 Cross-Sectional Examination of Well-Being and Employee Characteristics as Predictors of Productivity in 2Consecutive Years (N = 2629)

T1 T2

Parameter Estimate t P Estimate t P

Intercept 4227 2235 lt00001 4366 2419 lt00001

Disease status (nondiseased relative to diabetes) minus 007 minus 011 09117 164 267 00077

Well-being (IWBS) 055 2962 lt00001 052 2963 lt00001

Gender (women relative to men) minus 058 minus 113 02581 minus 091 minus 187 00609

Age 001 047 06403 004 171 00878

Employer A (relative to C) 026 033 0743 minus 198 minus 261 00091

Employer B (relative to C) minus 089 minus 111 02689 126 165 00989

IWBS individual well-being score T1 initial evaluation year T2 second evaluation year

cross-sectional analysis in 2 consecutive years found IWBS to bethe most significant variable associated with productivity with othercovariates (including disease status) having inconsistent or not sig-nificant associations beyond what was explained through IWBSBolstering these findings a longitudinal analysis of change in thesemeasures found that controlling for all time-invariant factors changein IWBS was a robust predictor of change in productivity

The WBA-P the productivity score calculated from the WBAwas chosen as the productivity measure for this analysis because of itsmultidimensionality which has the advantages of measuring morevariance in a population and also providing useful information toemployers about the prevalent sources of productivity loss24 Theanalysis was replicated using the overall job performance item fromthe HPQ and found that results were consistent with those using theWBA-P thus lending generalizability of the results across measuresof presenteeism

Diener and Seligman40 make the compelling case that tradi-tional economic indicators like gross domestic product are insuf-ficient to represent societal health and quality of life In line withthis broader notion the present findings demonstrate that physicalhealth is not sufficient to represent the vicissitudes of productiv-ity in the modern workplace but that the more global measure ofindividual well-being has a more important role in explaining pro-ductivity variance among workers Does this mean that we dismissthe role that health plays in productivity To the contrary the rolethat health plays as a foundation for productivity must be acknowl-edged Yet just as income is insufficient to fully reflect quality oflife in postindustrial societies physical health alone is insufficientto explain changes in productivity now that individual well-being isrecognized to have a stronger relationship with this outcome Alsojust as Diener and Seligman40 argued that money is not an end itselfbut a means to an end well-being is not an end itself for employersbut a means to greater productivity and profitability Therefore it isin employersrsquo interest to support and improve employee well-beingas a competitive strategy62ndash64

TABLE 4 Combined Model of Well-Being Change (T1 toT2) Contribution to Productivity Level Change AmongDiabetes and Nondiseased Groups Collectively (N = 2629)

Parameter Estimate Standard Error t P

Intercept 182 02346 776 lt00001

Change in well-being(IWBS)

032 00206 1529 lt00001

IWBS individual well-being score

Well-being may have more explanatory power with respectto productivity because it takes account of the important role thatphysical health plays and also addresses an array of other psychoso-cial lifestyle and environmental factors that influence productivityThese other well-being factors may also represent the ldquoroot causesrdquoor issues that one must address as a means to optimize health andproductivity outcomes The broad construct of well-being would log-ically relate to higher levels of productive functioning like creativityinnovation employee engagement adapting to a changing marketand environment that manifests as added economic value areas thatshould be explored in future research6566

Traditionally companies have relied on disease managementand wellness programs as a means of reducing or avoiding produc-tivity loss due to poor health To enroll employees in these programsrisks are identified using either claims data which apply only to em-ployees already using the health care system or traditional HRAswhich identify only existing health risk factors and requires aware-ness by the employee of their risks A strategy of well-being im-provement on the contrary applies to the total population not justthose with known physical health risks and conditions By evaluat-ing the diabetes and nondiseased groups separately this study showsthat well-being explains productivity within the groups at each endof this spectrum Therefore well-beingndashimprovement initiatives canbenefit all employees and have the opportunity to improve produc-tivity for an entire workforce not just those with existing healthproblems to allow employers the opportunity to shift the focus fromproductivity loss to productivity gain

The concept of well-being improvement is aligned with theparadigm shift in which employees are considered ldquohuman capitalrdquoand are thus of strategic importance to success in the marketplacerather than an expense to be managed6768 Innovative companies likeIBM realize that looking out for the physical health of employees isnot enough On their Web site dedicated to well-being IBM statesldquoAdvancing the health safety and well-being of our global workforceis an absolute priority itrsquos a commitment that encompasses the en-vironments in which employees work and the communities in which

TABLE 5 Nondiseased Model of Well-Being Change (T1 toT2) Contribution to Productivity Level Change (n = 1858)


Intercept 224 02637 850 lt00001


028 00249 1127 lt00001





TABLE 6 Diabetes Model of Well-Being Change (T1 to T2)Contribution to Productivity Level Change (n = 771)


Intercept 081 04836 168 0093


037 00370 1005 lt00001


TABLE 7 Contribution of Well-Being Change to HPQSelf-Rated Performance Change (N = 2629)


Intercept 151 02255 672 lt00001


033 00199 1638 lt00001

IWBS individual well-being score HPQ Health and Work PerformanceQuestionnaire

they liverdquo It is further stated that ldquothese [global] challenges require alsquototal health managementrsquo framework that transcends the implemen-tation of traditional employee well-being programs by recognizingthe importance of promoting physical and psychological healthrdquo69

Other organizations are adopting this view A large international sur-vey by the World Economic Forum found that organizations viewedas actively promoting health and well-being were at least 25 timesmore likely to be rated a best performer and to encourage creativityand innovation and 4 times less likely to lose talent66

Limitations to consider when evaluating the results of thisstudy include the potential for self-report bias and the fact that non-random samples across three employer groups may restrict gener-alizability Diabetes was chosen as the disease for analysis becauseof its prevalence documented impact on productivity and commonoccurrence as a comorbid condition Another potential limitation isthe under-diagnosis of diabetes in the United States giving rise tothe possible contamination of nondiseased group with undiagnoseddiabetics Nevertheless any such contamination would likely haveserved to attenuate the measured effect found between productivityand well-being

Although the diabetes group did have a high prevalence ofcomorbidity (53 conditions on average) future research should ex-pand this focus to understand whether the results are consistent acrossother specific diagnoses Although first differencing was used to con-trol for bias due to all time-invariant sources potential model mis-specification in the form of omission of time-varying variables andthe issue of simultaneity must be acknowledged Despite the afore-mentioned limitations the promise of well-being as a frameworkfor understanding and impacting worker productivity is supportedby the results Future research should address a broader range ofproductivity issues not available through current models and shoulddirectly test the impact of well-being improvement initiatives onpresenteeism and other forms of productivity loss

CONCLUSIONSThe results presented here show that employee well-being is a

significant predictor of productivity and suggest strategies for well-being improvement as an important means to optimize productivityabove and beyond what might be achieved through approaches aimedat preventing or managing chronic conditions The construct of well-

being provides the opportunity for a more comprehensive approachto addressing the issue of employee productivity in an entire popula-tion because workers who do not have health conditions often stillhave opportunities for well-being improvement

REFERENCES1 Economic Policy Institute The link between productivity growth and

living standards Available at httpwwwepiorgpublicationwebfeaturessnapshots archive 03222000 Published 2000 Accessed July 24 2013

2 Hymel PA Loeppke RR Baase CM et al Workplace health protection andpromotion a new pathway for a healthiermdashand safermdashworkforce J OccupEnviron Med 201153695ndash702

3 Loeppke R Christian J Gochfeld M et al Healthy workforcehealthy econ-omy the role of health productivity and disability management in addressingthe nationrsquos health care crisis why an emphasis on the health of the workforceis vital to the health of the economy J Occup Environ Med 200951114ndash119

4 Schulte PA Pandalai S Wulsin V Chun H Interaction of occupational andpersonal risk factors in workforce health and safety Am J Public Health2012102434ndash448

5 Burton WN Chen CY Schultz AB Conti DJ Pransky G Edington DWWorker productivity loss associated with arthritis Dis Manag 20069131ndash143

6 Ng YC Jacobs P Johnson JA Productivity losses associated with diabetes inthe US Diabetes Care 200124257ndash261

7 Oliva-Moreno J Loss of labour productivity caused by disease and healthproblems what is the magnitude of its effect on Spainrsquos economy Eur JHealth Econ 201213605ndash614

8 Kahn ME Health and labor market performance the case of diabetes J LaborEcon 199816878ndash899

9 Nurmagambetov T Atherly A Williams S Holguin F Mannino DM ReddSC What is the cost to employers of direct medical care for chronic obstructivepulmonary disease COPD 20063203ndash209

10 Colombi AM Wood GC Obesity in the workplace impact on cardiovasculardisease costs and utilization of care Am Health Drug Benefits 20114271ndash278

11 Birnbaum HG Berger WE Greenberg PE et al Direct and indirect costs ofasthma to an employer J Allergy Clin Immunol 2002109264ndash270

12 Goetzel RZ Long SR Ozminkowski RJ Hawkins K Wang S Lynch WHealth absence disability and presenteeism cost estimates of certain physicaland mental health conditions affecting US employers J Occup Environ Med200446398ndash412

13 Loeppke R Taitel M Richling D et al Health and productivity as a businessstrategy J Occup Environ Med 200749712ndash721

14 Loeppke R Taitel M Haufle V Parry T Kessler RC Jinnett K Health andproductivity as a business strategy a multiemployer study J Occup EnvironMed 200951411ndash428

15 Kessler RC Ames M Hymel PA et al Using the World Health OrganizationHealth and Work Performance Questionnaire (HPQ) to evaluate the indirectworkplace costs of illness J Occup Environ Med 200446(suppl)S23ndashS37

16 Colligan TW Higgins EM Workplace stress etiology and consequences JWorkplace Behav Health 20052189ndash97

17 Burton WN Pransky G Conti DJ Chen CY Edington DW The asso-ciation of medical conditions and presenteeism J Occup Environ Med200446(suppl)S38ndashS45

18 Burton WN Chen CY Conti DJ Schultz AB Pransky G Edington DWThe association of health risks with on-the-job productivity J Occup EnvironMed 200547769ndash777

19 Burton WN Chen CY Conti DJ Schultz AB Edington DW The associationbetween health risk change and presenteeism change J Occup Environ Med200648252ndash263

20 Johns G Presenteeism in the workplace a review and research agenda JOrgan Behav 201031519ndash542

21 Goetzel RZ Carls GS Wang S et al The relationship between modifiablehealth risk factors and medical expenditures absenteeism short-term disabil-ity and presenteeism among employees at Novartis J Occup Environ Med200951487ndash499

22 Hemp P Presenteeism at workmdashbut out of it Harv Bus Rev 20048249ndash58

23 Willingham JG Managing presenteeism and disability to improve productiv-ity Benefits Compens Dig 20084513

24 Prochaska JO Evers KE Johnson JL et al The Well-Being Assessment forProductivity a well-being approach to presenteeism J Occup Environ Med201153735ndash742




25 Dallman MF Pecoraro NC la Fleur SE Chronic stress and comfort foodsself-medication and abdominal obesity Brain Behav Immun 200519275ndash280

26 Glaser R Kiecolt-Glaser JK Stress-induced immune dysfunction implica-tions for health Nat Rev Immunol 20055243ndash251

27 Kiecolt-Glaser JK Glaser R Depression and immune function central path-ways to morbidity and mortality J Psychosom Res 200253873ndash876

28 Padgett DA Glaser R How stress influences the immune response TrendsImmunol 200324444ndash448

29 Schnall PL Belkic KL Landsbergis PA Baker DB The workplace and car-diovascular disease Occup Med 20001597ndash122

30 Tennant C Work-related stress and depressive disorders J Psychosom Res200151697ndash704

31 Merrill RM Aldana SG Pope JE Anderson DR Coberley CR WhitmerRW Presenteeism according to healthy behaviors physical health and workenvironment Popul Health Manag 201215293ndash301

32 Merrill RM Aldana SG Pope JE et al Self-rated job performance and ab-senteeism according to employee engagement health behaviors and physicalhealth J Occup Environ Med 20135517

33 Wang PS Beck A Berglund P et al Chronic medical conditions and workperformance in the Health and Work Performance Questionnaire calibrationsurveys J Occup Environ Med 2003451303ndash1311

34 Stewart WF Ricci JA Chee E Morganstein D Lipton R Lost productivetime and cost due to common pain conditions in the US workforce JAMA20032902443ndash2454

35 Stewart WF Ricci JA Chee E Morganstein D Lost productive work timecosts from health conditions in the United States results from the AmericanProductivity Audit J Occup Environ Med 2003451234ndash1246

36 Finkelstein EA DiBonaventura M Burgess SM Hale BC The costs of obesityin the workplace J Occup Environ Med 201052971ndash976

37 Tunceli K Bradley CJ Nerenz D Williams LK Pladevall M Elston Lafata JThe impact of diabetes on employment and work productivity Diabetes Care2005282662ndash2667

38 Schultz AB Edington DW Employee health and presenteeism a systematicreview J Occup Rehabil 200717547ndash579

39 Collins JJ Baase CM Sharda CE et al The assessment of chronic healthconditions on work performance absence and total economic impact foremployers J Occup Environ Med 200547547ndash557

40 Diener E Seligman MEP Beyond money toward an economy of well-beingPsychol Sci Public Interest 200451ndash31

41 National Institutes of Health NIH challenge grants in health and scienceresearch (RC1) 05-AG-102 prevention and risk factor reduction strategiesfor disabilities Available at httpgrantsnihgovgrantsfundingchallengeawardHigh Priority Topicspdftopic 05 Accessed October 2 2012

42 Centers for Disease Control and Prevention Well-being concepts Availableat httpwwwcdcgovhrqolwellbeinghtm Published 2013 Accessed July25 2013

43 Gandy WM Coberley C Pope JE Rula EY Well-being and employeehealthmdashhow employeesrsquo well-being scores interact with demographic fac-tors to influence risk of hospitalization or an emergency room visit PopulHealth Manag 2013 April 5 [epub ahead of print]

44 Harrison PL Pope JE Coberley CR Rula EY Evaluation of the relationshipbetween individual well-being and future health care utilization and costPopul Health Manag 201215325ndash330

45 Shi Y Sears LE Coberley CR Pope JE Classification of individual well-beingscores for the determination of adverse health and productivity outcomes inemployee populations Popul Health Manag 20121690ndash98

46 Shi Y Sears LE Coberley CR Pope JE The association between modifiablewell-being risks and productivity a longitudinal study in pooled employersample J Occup Environ Med 201355353ndash364

47 Harter JK Gurley V Measuring well-being in the United States Assoc Psy-chol Sci 20082123ndash26

48 Gallup Gallup-Healthways Well-Being Index methodology report forindexes Available at httpwell-beingindexcomfilesGallup-Healthways

20Index20Methodology20Report20FINAL203ndash25ndash08pdf Pub-lished 2009 Accessed August 13 2013

49 Evers KE Prochaska JO Castle PH Johnson JL Prochaska JM Harrison PLDevelopment of an individual well-being scores assessment Psychol Well-Being 201221

50 Schurer S Yong J Personality well-being and the marginal utility ofincome what can we learn from random coefficient models Availableat httpwwwmelbourneinstitutecomdownloadshildaBibliographyWorking+Discussion+Research Papers2012Schurer etal PersonalityWellbeing and the Marginal Utility of Incomepdf Health Economics andData Group Working Paper University of York Published 2012 AccessedAugust 15 2013

51 Greene WH Econometric Analysis 7th ed Upper Saddle River NJ PrenticeHall 2012

52 Allison P Fixed Effects Regression Methods for Longitudinal Data UsingSAS Cary NC SAS Institute 2005

53 Woodridge JM Introductory Econometrics A Modern Approach 5th edMason OH Cengage Learning 2012

54 Bell A Jones K Explaining Fixed Effects Random Effects Mod-elling of Time-Series Cross-Sectional and Panel Data Center for Mul-tilevel Modelling Bristol UK University of Bristol 2013 Avail-able at httppolmethwustledumediaPaperFixedversusRandom_1_2pdfAccessed January 10 2014

55 Vogenberg FR The economic burden of COPD in the workplace Am HealthDrug Benefits 20092198ndash200

56 Rodbard HW Fox KM Grandy S Impact of obesity on work productivityand role disability in individuals with and at risk for diabetes mellitus Am JHealth Promot 200923353ndash360

57 Mills PR The development of a new corporate specific health risk measure-ment instrument and its use in investigating the relationship between healthand well-being and employee productivity Environ Health 200541

58 Sears LE Shi Y Coberley CR Pope JE Overall well-being as a predictor ofhealth care productivity and retention outcomes in a large employer PopulHealth Manag 201316397ndash405

59 Torrisi B Academic productivity correlated with well-being at work Scien-tometrics 201394801ndash815

60 Cooper CL Mental health and well being at work the disposal workforce IntJ Public Health 200853225ndash226

61 Taylor B Wellness viewed as physical health misses the mark EmployerBenefits Advis 2013 Available at httpebabenefitnewscomnewswellness-viewed-physical-health-misses-the-mark-2730520-1html Accessed January10 2014

62 Johnson S Cooper C Cartwright S Donald I Taylor P Millet C The ex-perience of work-related stress across occupations J Managerial Psychol200520178ndash187

63 Wright TA Cropanzano R The happyproductive worker thesis revisited InMartocchio JJ ed Research in Personnel and Human Resources ManagementVol 26 Bingley Yorkshire UK Emerald Group 2007269ndash307

64 Harter JK Schmidt FL Hayes TL Business-unit-level relationship betweenemployee satisfaction employee engagement and business outcomes a meta-analysis J Appl Psychol 200287268ndash279

65 Holbeche L Aligning Human Resources and Business Strategy 2 ed Ams-terdam the Netherlands ElsevierButterworth-Heinemann 2009

66 Dornan A The Wellness Imperative Creating More Effective Organi-zations Geneva Switzerland The World Economic Forum and RightManagement 2010 Available at httpwww3weforumorgdocsWEF HEWellnessImperativeCreatingMoreEffectiveOrganizations Report 2010pdf Accessed August 14 2013

67 James KC The HR paradigm shift and the federal human capital opportunityPublic Manager 20013013ndash16

68 Ip P Developing a concept of workplace well-being for Greater China SocIndic Res 20099159ndash77

69 IBM Web site Employee well-being Available at httpwwwibmcomibmresponsibilityemployee well beingshtml Accessed August 14 2013




In their groundbreaking article ldquoBeyond Money Toward anEconomy of Well-Beingrdquo Diener and Seligman40 contend that well-being should be the ultimate goal around which economic healthand social policies are developed not simply because well-being isan important indicator in itself but also because well-being is manytimes a cause of other valued outcomes such as worker productivityand rewarding relationships They conclude that organizations andnations would benefit from regular assessment of well-being andfinding ways to improve it An individualrsquos well-being is sensitivenot only to physical health factors that contribute to absenteeism andpresenteeism but also to nonndashhealth-related factors (eg psychoso-cial environmental financial work etc) that may serve as barriersto optimal productivity

In addition to the macrolevel argument put forth by Dienerand Seligman40 that well-being is an important indicator of societalhealth this construct is gaining acceptance as an integral part ofhealth care measurement continues to gain adherents Both the Na-tional Institutes of Health and the Centers for Disease Control andPrevention have recognized well-being as an important indicator ofphysical and mental health and therefore an essential component inthe measurement of public health outcomes4142

The purpose of this study was to evaluate well-being as mea-sured by the Well-Being Assessment (WBA) as such a comprehen-sive measure with the aim of maximizing opportunity for produc-tivity improvement across entire populations regardless of healthstatus Individual well-being scores (IWBSs) as measured by theWBA have demonstrated strong relationships to a number of stan-dard health care outcomes including expenditures inpatient admis-sions and emergency department visits4344 as well as productiv-ity outcomes in the form of absenteeism and presenteeism

2431324546

lending support to this well-being measure as the broader frameworkfor understanding what influences business outcomes including pro-ductivity Nevertheless research to date has not directly comparedwell-being with the traditional factors known to affect productivityHere IWBS is compared with the prevalent chronic condition di-abetes to determine whether well-being contributes to productivitylevels and changes over and above what can be explained by physicalhealth alone

METHODSStudy Design

The study used a longitudinal retrospective panel study de-sign using survey data collected in 2 consecutive years (T1 and T2)The data for the panel were composed of employees from three UScompanies One company was a large health care vendor anotherwas a nationally based insurance company and a third was an inter-nationally based manufacturing company Study eligibility requiredcompletion of the WBA at T1 and T2 with valid well-being andproductivity scores from both time points and a documented agebetween 18 and 64 years Two study groups were identified fromthe data panel (1) nondiseasedmdashindividuals who self-reported nochronic conditions at T1 and T2 (n = 1858) and (2) diabetesmdashindividuals who self-reported a diagnosis of diabetes at T1 and T2(n = 771) Individuals in the diabetes group were not restrictedon the basis of the presence or absence of other comorbid condi-tions Because of the negligible risk retrospective design and use ofde-identified data this study was exempt from institutional reviewboard approval based on exclusion criteria outlined in the US Codeof Federal Regulations (45 CFR sect46101)

Measures

Well-Being WBA and IWBSThe WBA was developed as an extension of the Gallup-

Healthways Well-Being Index (WBI)4748 a community survey ofwell-being for use with employer populations and other organiza-

tions such as health plans and health systems interested in totalpopulation health and well-being The instrument includes healthrisk and productivity measures in addition to the WBI item set Indi-vidual well-being score was developed from WBI items and domainsto allow calculation of well-being at the individual level49 The IWBSis calculated using 40 questions from the following six domains ofwell-being that are included in the WBI and WBA physical healthemotional health healthy behaviors work environment basic ac-cess and life evaluation Each domain is weighted equally in thecalculation of the IWBS because they are in the WBI and scoresrange from 0 to 100 for each respondent

Productivity WBA-P Overall Score and WBAP_Alt ScoreThe primary measure of productivity in the study was the

WBA-P that is administered as part of the WBA and provides an in-formative evaluation of on-the-job productivity loss (presenteeism)due to well-beingndashrelated barriers Criterion-related validity of theWBA-P has been established through multivariate analysis to a num-ber of health and well-being measures24

The WBA-P score is taken from 11 items on the WBA withthe shared question stem ldquoDuring the past 4 weeks (28 days) howoften have you had trouble at work concentrating or doing your bestbecause of rdquo and then lists 11 possible reasons or barriers Scor-ing of this measure ranges from 0 (not at all) to 100 (a lot for all11 reasons)24 For the purposes of the study the WBA-P overallscore was converted from a measure of productivity loss to one ofproductivity functioning This was accomplished by taking the com-plement of the WBA-P that is WBAP_Alt = 100 minus (WBA-P) Thisconverted score is designed to reduce or eliminate zero scores for sta-tistical analyses Higher scores reflect higher levels of productivity

Productivity (Secondary Measure) HPQ Self-RatedPerformance Scale

The HPQ contains a global self-rating of job performancemeasured on a 0 to 10-point scale that is considered an absolutemeasure of presenteeism15 this question was included in the WBAThe item reads ldquoOn a scale from 0 to 10 where 0 is the worst jobperformance anyone could have at your job and 10 is the performanceof a top worker how would you rate the usual performance of mostworkers in a job similar to yoursrdquo Responses to this global item wereused in sensitivity analyses to further strengthen conclusions drawnwith the primary productivity measure of the study the WBA-P

Statistical MethodsFirst cross-sectional linear models were used to examine

the relative contribution of well-being and other selected covari-ates (Table 1) including disease status on productivity among

TABLE 1 Study Model Variables

Variable Type CategoriesRange

Productivity score(WBA-P_Alt)

Continuous 0ndash100

Disease status Categorical Nondiseased or diabetes

Well-being (IWBS) Continuous 0ndash100

Age yrs Continuous 18ndash64

Gender Categorical Men or women

Comorbidities count Continuous 1ndash21

Employer Categorical A B or C

Time Categorical T1 = 2010 T2 = 2011

Dependent variableIWBS individual well-being score WBAP_Alt Well-Being Assessment for

Productivity complement of score


















Women 488 415



















T1 T2


Intercept 4227 2235 lt00001 4366 2419 lt00001




Age 001 047 06403 004 171 00878









Intercept 182 02346 776 lt00001


032 00206 1529 lt00001







Intercept 224 02637 850 lt00001


028 00249 1127 lt00001







Intercept 081 04836 168 0093


037 00370 1005 lt00001




Intercept 151 02255 672 lt00001


033 00199 1638 lt00001




































































































Women 488 415



















T1 T2


Intercept 4227 2235 lt00001 4366 2419 lt00001




Age 001 047 06403 004 171 00878









Intercept 182 02346 776 lt00001


032 00206 1529 lt00001







Intercept 224 02637 850 lt00001


028 00249 1127 lt00001







Intercept 081 04836 168 0093


037 00370 1005 lt00001




Intercept 151 02255 672 lt00001


033 00199 1638 lt00001























































































T1 T2


Intercept 4227 2235 lt00001 4366 2419 lt00001




Age 001 047 06403 004 171 00878









Intercept 182 02346 776 lt00001


032 00206 1529 lt00001







Intercept 224 02637 850 lt00001


028 00249 1127 lt00001







Intercept 081 04836 168 0093


037 00370 1005 lt00001




Intercept 151 02255 672 lt00001


033 00199 1638 lt00001
























































































Intercept 081 04836 168 0093


037 00370 1005 lt00001




Intercept 151 02255 672 lt00001


033 00199 1638 lt00001





















































































Comparação bem estar com doenças crônicas

Documents

Transcript of Comparação bem estar com doenças crônicas