Portland State University Portland State University

PDXScholar PDXScholar

Psychology Faculty Publications and Presentations Psychology

2013

Designing Work, Family & Health Organizational Designing Work, Family & Health Organizational

Change Initiatives Change Initiatives

Ellen Ernst Kossek Purdue University

Leslie B. Hammer Portland State University

Erin L. Kelly University of Minnesota

Phyllis Moen University of Minnesota

Follow this and additional works at: https://pdxscholar.library.pdx.edu/psy_fac

Part of the Industrial and Organizational Psychology Commons, and the Social Psychology Commons

Let us know how access to this document benefits you.

Citation Details Citation Details Kossek, E. E., Hammer, L. B., Kelly, E. L., & Moen, P. (2014). Designing Work, Family & Health Organizational Change Initiatives. Organizational dynamics, 43(1), 53-63.

This Article is brought to you for free and open access. It has been accepted for inclusion in Psychology Faculty Publications and Presentations by an authorized administrator of PDXScholar. Please contact us if we can make this document more accessible: pdxscholar@pdx.edu.

Designing Work, Family & HealthOrganizational Change Initiatives§,§§

Ellen Ernst Kossek, Leslie B. Hammer, Erin L. Kelly, Phyllis Moen

EACH PERSON HAS THE SUPPORT THEY NEEDTO HAVE CONTROL OVER THEIR WORK ANDLIFE, AS LONG AS THE WORK GETS DONE

What if all workplaces were designed to change organiza-tional cultures and the structure of work to truly supportemployees’ work and family needs and reduce conflicts? How

can employers and researchers create initiatives to improveemployment settings to prevent work—family conflict andburnout? Despite a burgeoning literature and the prolifera-tion of work—life consultants and policies, work—familyresearch has had relatively limited impact on how work ismanaged in many companies today. Yet work—family andpersonal life conflicts and stress are growing managementand public health concerns that impact employees, employ-ers, and families across the globe.

Work—family conflict (from work to families and fromfamilies to work) is an increasingly critical issue in today’sworkplace. It has been consistently linked to adverse mental,behavioral, and physical health outcomes, including cardio-vascular disease risk, sleep quality, depressive symptoms,burnout, workplace safety, obesity, and addictive behaviors(i.e., smoking and alcohol use). Work—family conflicts arealso related to employee productivity, turnover, absentee-ism, well-being, and engagement.

Despite the importance of work—family conflict for healthand productivity, researcher-organizational partnershipshave not fostered major change in practice. Poor qualitystudies have weakened the business case. For example, manystudies simply compare workers with and without work—family conflict, overlooking evaluation how the design ofworkplaces may be fostering conflict. Or policies are intro-duced with poor implementation such as weak linkage towork procedures, career systems, or management practice.These gaps have resulted in limited employer evidence forprioritizing systemic reduction in work—family conflict in theway work is organized. It has also slowed the diffusion ofevidence-based practice.

Employers need to use best practice approaches, such asrandomized control trials (use of control and experimentalgroups) of interventions aimed at preventing or reducingwork—family conflict in order to foster healthy workplaces.Top management needs to take an active role in preventingwork—family stress in how work is managed and organized.

Organizational Dynamics (2014) 43, 53—63

§ This is an open-access article distributed under the terms of theCreative Commons Attribution-NonCommercial-No Derivative WorksLicense, which permits non-commercial use, distribution, and repro-duction in any medium, provided the original author and source arecredited.§§ Author Note: The authors would like to thank their colleagues fortheir valued assistance and input into the writing of this paper: RyanOlson, Brad Wipfli & Kent Anger, Oregon Health Sciences; KristaBrockwood, Portland State University; Cassandra Okechukwu, Har-vard University, Georgia Karuntzos, RTI International, Rosalind King,National Institutes of Health, and Lisa Burke, Purdue, University. Thisresearch was conducted as part of the Work, Family, and HealthNetwork, which is funded by a cooperative agreement through theNational Institutes of Health and the Centers for Disease Control andPrevention: The Eunice Kennedy Shriver National Institute of ChildHealth and Human Development (Grant # U01HD051217,U01HD051218, U01HD051256, U01HD051276), National Institute onAging (Grant # U01AG027669), Office of Behavioral and SocialSciences Research, and National Institute for Occupational Safetyand Health (Grant # U010H008788). The contents of this publicationare solely the responsibility of the authors and do not necessarilyrepresent the official views of these institutes and offices. Specialacknowledgment goes to Extramural Staff Science Collaborator,Rosalind Berkowitz King, Ph.D. (NICHD) and Lynne Casper, Ph.D.(now of the University of Southern California) for design of theoriginal Workplace, Family, Health and Well-Being Network Initia-tive. Persons interested in learning more about the Network shouldgo to: http://www.kpchr.org/workfamilyhealthnetwork/public/researchers.aspx.

Available online at www.sciencedirect.com

ScienceDirect

jo u rn al h om ep ag e: ww w.els evier .c o m/lo c ate /o rg d yn

0090-2616/$ — see front matter # 2013 The Authors. Published by Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.orgdyn.2013.10.007

Reviews published by the authors of this article in theAcademy of Management Annals and Personnel Psychologyunderscore the need for organizational interventions speci-fically focusing on job stress and improving relationshipsbetween work and employees’ family and personal lives.While rigorous change partnerships are clearly needed,how do leaders and scholars go about designing and imple-menting them?

OBJECTIVES

In this paper, we describe the development of the mostcomprehensive work—family organizational change initia-tive to date in the United States. Our goal is to share an in-depth case study with examples and critical lessonsthat emerged. We draw on our years of experience workingwith major employers from two industries representativeof today’s workforce (health care and IT professionals).Employers and applied researchers can draw on thisstudy and lessons to create, customize, and deliver evi-dence-based interventions to improve work, family andhealth.

THE WORK, FAMILY AND HEALTH NETWORKINTERVENTION

The Work, Family and Health Intervention is a comprehensivemulti-faceted organizational intervention that is designed tofoster a healthy psychosocial work environment by prevent-ing stressors in the organization of the workplace that canlead to work—family conflict.

A national interdisciplinary team of researchers devel-oped the intervention. The Work Family and HealthNetwork (WFHN) is a collaboration of scholars with back-grounds in public health, medicine, family studies, orga-nizational psychology, occupational health psychology,sociology, economics and many other fields. The interven-tion benefited from having multiple disciplinary scientificperspectives on contemporary work—family conflictchallenges. It also was informed by employee andemployer advisory groups providing practical stakeholderinput.

Below we describe a series of pilot studies conducted toevaluate the effectiveness of intervention components. Tocreate adaptive design, we also assessed the contextualinfluences on work—family conflict across the health careand IT (information technology) industries. We describe thekey intervention features and design stages, followed by theseven principles that emerged (see Table 1 for a summarywith examples), as a template for work—life interventionresearch and practice.

This intervention is innovative, as it is designed to proac-tively change work conditions to reduce work—family con-flict. Traditionally, most work—life policies and practices arereactive, ad hoc, or stigmatize employees with work—lifestresses. Typically they are viewed as an individual accom-modation, not mainstream work practice. They do not pre-emptively eliminate the stress caused by work—familyconflict in the general work environment of all workers acrossan entire organization.

KEY INTERVENTION COMPONENTS FROMPILOT STUDIES

Early pilot studies were useful for identifying whether keyfactors identified as important in the work—family literaturecould be delivered in different occupations. The first is toincrease employees’ control over their work schedules and afocus on results, not time. The second is to increase work—family specific social support through supervisor behaviortraining.

Schedule control and results orientation. One set ofstudies led by sociologists Erin Kelly and Phyllis Moen atthe University of Minnesota focused on a natural experi-ment. They examined a corporate-led initiative called‘‘ROWE’’ (Results Oriented Work Environment) targetingprofessionals at Best Buy’s headquarters in Minneapolis.ROWE aimed at increasing employees’ control over theirwork time and fostering team-level job redesign keyingin on results, not time spent in meetings or at the office.This is considered a ‘‘natural’’ experiment because ROWEwould have occurred whether or not the researchers stu-died it.

The researchers chose to assess the effects of ROWEbecause it aligned with concepts developed by seminal jobstress researchers Robert Karasek and Tores Theorell on theimportance of employees’ job control, for health. Theresearchers extended this concept to control over time.The pilot studies showed that work teams following ROWEpractices had higher schedule control, lower work—familyconflict, lower turnover intentions, and improved healthbehaviors, than other teams.

Work—family specific social support through supervisorbehavior training. The other main intervention pilot studywas led by Leslie Hammer of Portland State University andEllen Ernst Kossek of Michigan State University (now atPurdue University). The researchers partnered with SpartanStores in Michigan and Ohio to develop, validate, and eval-uate the Family Supportive Supervisor Behavior (FSSB) train-ing and self-monitoring intervention.

The self-paced, computer-based and behavioral self-mon-itoring intervention was designed to increase supervisors’level of family supportive supervisor behaviors. Seminaltheorists Sheldon Cohen and Thomas Wills suggest increasedsocial support perceptions have positive psychological, well-being and performance effects. The researchers operationa-lized behaviors indicative of manager social support forfamily and non-work roles.

Behavioral science researchers W. Kent Anger and RyanOlson at Oregon Health Sciences collaborated on the devel-opment of the FSSB training. The content was based onratings of employee experience with four supervisor beha-viors that was validated in another study led by Hammer andKossek. They are:

� Instrumental — behaviors helping workers manage sche-dules and working with employees to solve schedule con-flicts. For example, helping an employee find areplacement, if absent.� Emotional — behaviors demonstrating a worker is beingcared for, and their feelings are being considered.For example, increasing face-to-face contact with

54 E.E. Kossek et al.

Table 1 Seven Design Principles for Organizational Work, Family and Health Interventions.

Intervention Design Principles Description Intervention Examples

1. Take a primary preventionsystemic approach to organizationalchange management

� Leverages individual and organizationalapproaches to change the leaders andorganizational structure to preventwork—family conflict before it occurs inwork organization� Focus on changing work site andmanagement approaches to theorganization of work that may creatework—family conflict in work structureand culture� Examine formal and informal aspects ofthe organization of work (e.g., formalpolicies and informal cultural) for holisticalignment to enhance resources acrosswork—family systems for a ‘‘dual agenda’’

� Proactively teaches individual supervisorsimproved leadership behaviors toincrease social support for family andperformance (Leader)� Structural job redesign to increasecontrol over work time, empowerworker’s autonomy, and to reduce lowvalue work (Organizational)� Align formal supervisor performancegoals with supportive organizationalcultural redesign practices jointlysupporting work and family

2. Identify theoretically-based keyintervention ingredients thattarget work—family conflictreduction

� Increase socially supportive behaviors forpersonal/family and performance roles� Increase employees’ perceived controlover work and work time� Improve the design of the organization offormal and informal work processes andcultural norms to become resultsoriented, reducing low value work andnonproductive face time.

� Family Supportive Supervisor Behaviors(FSSB) manager/supervisor training� Employees given more say over when andhow work is done and encouraged to helpeach other� Group training to de-stigmatizecomments and judgments about howface time relates to productivity

3. Include formative research andidentify comparative evaluationgroups leading to evidence-basedresearch findings

� Evaluate intervention components inpilot studies� Validate measurement of interventionchange targets� Pilot in several contextsIdentify control and experimental groups

� Pilot studies conducted in lower wage andhigher wage industries� FSSB validatedSchedule control and low value workevaluated� Randomized field method

4. Use Interdisciplinarydesign teams

� Integrate knowledge and variousperspectives from multiple disciplines� Can combine a positive and negativeapproach to workplace stress andassumptions about change processesrelated to the reduction of work—familyconflict as an occupational healthpathway� Theoretically comprehensive approachesenhance resource-enhancing strength ofintervention

� FFSB Training focused on increasingindividual leader demonstration ofwork—family and performance specificsupportive behaviors and self-monitoring(a psychological view)� Group empowerment social changeactivities, focusing on empoweringworkers to have greater control over howand when work is done (sociological &organizational behavior focused)

5. Bottom up participation withstrong management supportfor the intervention

� Combination of high employeeparticipation with top downmanagement buy-in and support

� Highly engaging, participative employeesessions� Management support needed for trainingdelivery during work time

6.Interventions should becustomizable and adaptivein design

� Adapting intervention content anddelivery across socio-economic work,family and organizational contexts

� Training delivery (e.g., number ofsessions) and examples of control andsupport at the two contrasting industriesdiffered.

7. Plan and devote resources topromote intervention transferto the work environment

� Plan for transfer of interventionexperiences during and post facilitation

� ‘‘Homework’’ between sessions� Do Something Scary/Different self-monitoring� Moving Forward session at Leef fourweeks after sessions ended

Work, family and health change initiatives 55

employees, asking how employees are doing, or commu-nicating genuine concern about employees’ work/lifechallenges.� Role modeling — behaviors that show how a supervisor istaking care of her or his own work/life challenges. Forexample, discussing taking time out to attend a child’sschool activities and talking about one’s own family. Orleaving work at reasonable hours, and showing that man-agers value involvement in non-work life.� Creative work—family management — behaviors aimedat redesigning work to support the conflicting employeework—life demands in a manner that is a win—win for bothemployees and employers. For example, promoting cross-training and the ability for shift trades to jointly enableemployee scheduling flexibility needs and work coverage.

The training combined theory on knowledge disseminationwith behavioral role-modeling principles delivered in pro-grammed instruction. The program of behavioral role-model-ing training draws on five cumulative steps identified bypsychologist P.J. Taylor and colleagues in a review. The firstis to have a clear description of the behaviors or skills to belearned. The second step is to develop models of effectiveuse. Third, create opportunity to practice the behaviors.Fourth, provide feedback and social reinforcement. Fifth,use motivators to foster on-the-job transfer.

The training intervention was delivered by Hammer andKossek and their team in a randomized control trial fieldstudy at 12 grocery and drug stores. The use of a randomized

approach for work—family training was pioneering, as it wasone of the first work—family studies to do so. Most work—family change efforts are only studied in more innovativefirms or do not use control and intervention treatmentgroups. This makes it harder to know if change is due tothe initiative or the overall better context. Replication is alsomore difficult as it can be difficult to adapt initiatives fromcontexts that are highly open to work family change to lesspositive settings.

INTERVENTION DESIGN, TRAINING CONTENT,AND STAGES

Hammer, Kossek, Kelly and Moen formed the Network’sintervention team. Together they integrated key conceptsfrom the pilot studies (shown in Figure 1) to develop theWFRN intervention and supporting training content(Figure 2). Examples include improving the work environ-ment to increase employees’ control over their schedules, afocus on results-oriented time use and the four FSSB beha-viors. New content was added, such supervisor performancesupport behaviors in the managerial training to better linkFSSB training to the results orientation of ROWE. For exam-ple, besides FSSB behaviors, managers also set goals andtracked how often they provided job direction and helpedprioritize work, or gave feedback and coaching on good andbad work behaviors.

The integrated WFRN intervention was called STAR (Sup-port. Transform. Achieve. Results.). It focused on whole

B. Goals: Change s inEmplo yee Psychos ocialOrgani za�onalEnvi ronment

A. Interven�on Deliv eryStrategies

D. Mo dera� ng Fa cto rs

C. Work-Fa mily In terface

D3. Fa mily Health a ndWell -Being

D2. Emplo yee Healthand Well -Being

D1. Wo rkplaceOut comes

Group par�cipatory,facil itated sessionsA�e r-session work-improvement redesignac�vi�esLeader computer-basedtrainingBehavioral se lf-monitoring by leader sand co-workers

Incre ased SOCIALSUPPORT FOR WOR KAND FAMILY INTERFA CE(supervisors & co-workers)Incre ased em ployeeCONTROL OVER WORKAND WOR K TIMEImprove design of workcondi�o ns an dprocesses and increasevalues/norms to beRESULTS-ORIENTE D ,REDU CE LOW VALU EWORK

Work-to-Famil y Con flictFamil y-to-Work Con flictWork-to-Famil y Posi�veSpillo verPer ceived Time Adequacy

Demogra phicsFamily Ch aracteris�csJob Ch aracter is�csSupe rvisor Cha racteris�csSocial Suppo rtHealth StatusOrganiza�on (Tomo, Le ef)

Marital Sa�sfac�onPare nt-Child Re la�onshipChil dren’s Health

Physical Healtho CVD Risko Slee po Diabetes riskMenta l Healtho Ps ychological distress

Per ceived Perf orma nceAbsentee ismJob Sa� sfac�onTurn over

Figure 1 Summary of organizational intervention components, goals and outcomes.

56 E.E. Kossek et al.

systems change in the level of work—family conflict emanat-ing from the work environment of an entire work site or workunit. STAR targeted multiple levels of change, includingelements related to the

� organization (work—family culture, leader and memberbehaviors, and the structure of work);� work teams and units (cross-training); and

� individual (leader behaviors and self-monitoring).

STAR was rolled out in a randomized fashion in half of 30nursing homes in a large long-term health and specializedcare company (called Leef), and several dozen teams in theIT unit of a Fortune 500 company (called Tomo). STAR includ-ed common components in each industry: (1) participatoryface-to-face sessions with staff and managers; (2) participa-tory face-to-face sessions for only managers and supervisors;(3) on-the-job activities for all employees to reinforce learn-ing from sessions; and (4) manager-only computer-basedtraining and behavioral self-monitoring.

The researchers and consultants worked together to pre-pare a facilitators’ guide for STAR participatory sessions, usingsemi-structured scripts as well as interactive activities. Thesessions encouraged supervisors and employees (either jointlyor separately) to reflect on current practices and identifystrategies to increase supervisor support, increase work-timecontrol and results orientation, and reduce work—family con-flict while continuing to meet or exceed business goals.

A computer-based training protocol and supervisor self-monitoring protocol were developed for the FSSB and per-formance behavioral training. STAR materials are availableon a public website www.workfamilyhealthnetwork.org.

Adaptive design strategy across occupational contexts.The intervention followed similar goals in each industry, yetwas adapted and customized to modify content, timing, andsequencing. This was challenging because the different inter-vention features had been developed in unique contexts. Forexample, how could the ROWE intervention, which wasdeveloped for white-collar corporate professionals, beadapted to hourly workers in a 24-7 patient-centered worksystem? How should the FSSB training, developed largely in anhourly workforce setting, be adapted for a professional ITcontext? What does work-time control look like for lower-level hourly workers with place-bound jobs, compared withITworkers who have high connectivity via cell phones and theInternet?

Intervention stages. Figure 2 describes the three stagesused to organize the delivery of STAR activities. Stage 1 ispreparing for the change, which focused on the organiza-tion’s top management and leadership team. This includedmanager-only participatory sessions, computer-based super-visor training, and a first trial of self-monitoring supportivebehaviors. Stage 2 focused on setting the change in motionusing joint manager and employee participation. These ses-sions reinforced the concepts of support and control. Man-agers and employees were encouraged to put these concepts

Figure 2 Intervention stages and activities distributed over 4 months. (a, Only in information technology industry; b, Only in healthcare industry.)

Work, family and health change initiatives 57

into practice within their units or teams. For example,employees discussed how work procedures might be alteredto increase support for work—life and still ensure the workgets done. Stage 3 involved activities to sustain change byhaving all participants discuss what people had tried to dodifferently, their successes and challenges, and their plansfor next steps. Below we draw on this comprehensive casestudy, and share seven emergent design principles (Table 1).We share how they map to STAR in order to encourage appliedresearchers and scholars to select relevant guideposts fortheir work—family and health change initiatives.

WORK—FAMILY AND HEALTH INTERVENTIONDESIGN PRINCIPLES

Principle 1: Take a systemic and primary preventionapproach to organizational change management. A primaryprevention approach focuses on eliminating the work hazardin the work context before it impacts health, yet relativelyfew work—family interventions follow this approach. Orga-nizational-focused interventions seek to modify work struc-ture to reduce exposure to or eliminate job stress orconditions that promote unhealthful work—life attitudesand behaviors. This implies implementing system-wide inter-ventions to change the work context to reduce work—familystressors before they occur. One example is altering workpractices across the company to give employees greatercontrol over their work schedules. Another illustration istraining managers and coworkers to provide greater supportfor family and personal life. Clarifying and improving jobdesign to focus on quality and critical tasks and to improveprocesses, and to reduce unnecessary workload is a thirdpossible change.

Yet rather than making flexibility and work—life support anormal way of doing business, most companies target policiesand stress programs toward individuals with salient work—lifeneeds such as young mothers. This can marginalize users.Such an approach does not prevent job-induced work—familyconflict and can actually result in stigmatization and back-lash, undermining change efforts intent.

Individual approaches can be effective if they are linked tomulti-level systemic organizational change strategies. Psy-chologist Semmer’s recent review of 90 studies on job stressinterventions classified change goals as modifying aspects ofthe organization (O); the individual employee (I); or bothfactors (OI). They found that interventions targeting bothorganizational and individual levels were highest in effec-tiveness.

Alignment can also refer to integrating informal organiza-tional culture and norms with formal structure, policies andpractices. Systemic alignment can also apply to balance inthe actual content of the intervention. We found a dualagenda was needed in content, a simultaneous focus onimprovement in work and family support to improve thewhole employee’s life system.

Unfortunately, many work stress interventions try toimprove work—family issues or health, but overlook perfor-mance issues. Yet clear and consistent and well-definedperformance expectations can enhance work—family andpersonal life relationships, by making work-time more effec-tive. This personally benefits workers by allowing them to

focus on work tasks that matter most, reducing stress, over-time and conflicts.

Principle 1 applied: The Work, Family and Health Networkintervention is comprised of components that encourage theprevention of work—family conflict by changing the structureof work for all employees in a team or unit. A core message ofSTAR was that it is appropriate to systematically look at theway work is done — including when, where, and how work isperformed. The goal is to evaluate what changes mightincrease employees’ effectiveness on the job while alsoallowing them to meet personal commitments and take careof themselves.

For example, many employees in the IT organization set-ting had long commutes to the office (more than an hour eachway), which severely cut into both their work and familytime. STAR’s systematic fostering of higher schedule controlenabled employees to alter their start and stop times in orderto commute to work at times other than rush hours, decreas-ing their travel time to and from the office. Increased flex-ibility in the form of remote work also enabled IT employeesto sometimes avoid a commute altogether by working fromhome or other non-work sites.

An example of a systemic approach to improving job andpersonal performance involved shifting night conferencecalls for the IT employees who worked with others offshoreacross time zones. Before STAR, when they had conferencecalls during the night or early morning hours, employees werestill expected to be in the office by 9 a.m. to work atraditional day. Many described themselves as too tired toaccomplish much. STAR enabled employees to vary schedulesto adapt to global work demands, by working in optimal timesand places.

A holistic prevention approach also moves away from onlytargeting work redesign for individuals with work—familyneeds — a reactive approach that could stigmatize them.STAR was designed to improve the entire work site or group.STAR’s primary prevention approach to decreasing work—lifestress changed the overall work environment through mod-ifications in the structure of jobs for all workers.

For example, rather than only letting employees withchild care needs have access to adjust starting and endingwork schedules, all employees with modifiable jobs weregiven the option to explore flexible start and end times totheir work day. In the healthcare setting, a nursing assistantmight be able to change her shift so she works 7:15—3:15instead of 6:45—2:45. This could be achieved as long as theperson working overnight can stay 30 minutes later and theperson starting a shift in the afternoon is willing to start andend her shift 30 minutes later as well. In the IT office setting,this would look different — sometimes with a range of startingand ending times, rather than a 9—5 (or more likely, 8—6:30)work day.

In both organizational settings, STAR targeted the entirework environment to enable all employees to be more easilyable to adapt their schedules to fit diverse personal needs(getting a child off to school, doctor’s appointment), whilealso ensuring that their work is accomplished in a timely way.

Principle 2: Identify theoretically based interventioningredients targeting the reduction of work—family con-flict. Given that most work family initiatives are adoptedwithout first analyzing job factors inducing work—familyconflict, we suggest interventions should be designed to

58 E.E. Kossek et al.

focus on altering key pathways related to work—family con-flict that link to well-being and health. They are:

Control over work time and schedule, and results orien-tation. Control over work time (or schedule control) refers toindividual autonomy over when and sometimes where work isconducted. When applied to work—family issues, increasingschedule and work time control enhances employees’ abil-ities to rearrange work and family roles to reduce work—family conflicts. A results orientation involved eliminatinginefficient work processes and giving clear performanceexpectations and coaching.

Supervisor social support for work and family. Researchsuggests that higher social support from supervisors is aresource that consistently buffers against conflicts betweenwork and family demands. Family-supportive supervisors arethose who appreciate and have an understanding for employ-ees’ demands outside of the work domain and accommodateemployees’ efforts to seek balance between these roles.Supervisors’ actions affect the extent to which employeesare buffered from work—family conflict by supporting themto better manage work and family roles in order to besuccessful in each. Reviews by Ellen Kossek and colleaguesfound that family-specific supportive supervision has agreater impact on work—family conflict than generally sup-portive supervision. They also found that employees’ percep-tions of their own supervisors’ family supportivenessdetermines whether they are likely to perceive their com-panies as family supportive, and is more important thanoffering underused work—family policies on the books.

Principle 2 applied. The pilot studies provided evidencethat work—family interventions must target three key ingre-dients. They are (1) increasing control over work time; (2)increasing supervisor social support for family and job effec-tiveness; and (3) improved work design processes and culturesupporting results orientation and removal of low value work.

STAR targeted control and support for hourly nursingassistants by leading to changes such as the initiation of amaster schedule for the holiday season or summer vacations.Rather than having a scheduler make up a master workschedule without employee input, all workers were allowedto put in their first choice for time off.

In the IT workforce, teleworking became more widelyavailable and self-managed as long as the employee’s workoutput was maintained. The individual also had to be acces-sible and attend key meetings. Sometimes control led to newlife arrangements. Rather than take an unpaid leave, oneemployee was able to telework from another state while herelderly mother was recovering from health problems.

A Tomo employee’s remarks illustrate the importance ofhaving an intervention target time control in job design andwork culture:

I think if people feel like they’re more control of their lifeand control of their time, it allows them to basicallysay. . .if I want to take off at 3:00 today, I can do that. Idon’t have to send a note to my boss saying I’m cutting outearly today at 3:00 because I’m going to go. . .watch mykids’. . .activity this afternoon or I’ve got to leave earlybecause I’ve got a doctor’s appointment.

Supervisors and co-workers also increased verbal socialsupport for the meshing of work and personal time use.Employees were encouraged to be more open about personal

needs and their families and ask for and work out support toexercise, attend school conferences or simply stay homewhen sick.

In both settings, work processes were also improvedtoward results orientation. In the nursing homes, employeesbegan to cross-train and help each other out and had com-munication boards on each floor to improve informationsharing. In the IT settings, face time at unproductive meet-ings was reduced.

Principle 3: Include formative research and identifycomparative evaluation groups. While the popular pressoffers employers a multitude of work—life policy options,we’ve noted there is very little research on these as effectivechange management interventions. The only known evalua-tion of work—family interventions using a randomized con-trolled trial (RCT) design is the FSSB pilot study informingSTAR. Since many work—family interventions — such as flex-time and telework — are implemented with little use ofscientific experimental design principles, or pilot work toidentify barriers to implementation, formative research willdetermine if the intervention is tapping into change targets.It also allows for the identification of meaningful evaluationgroups for scientific evaluation of intervention effectiveness.

Principle 3 applied: STAR was based on formative workevaluating within and across industry comparisons. Forinstance, the idea of changing where work is done was notrelevant to most nursing home employees. Yet how work isdone became more of a focus for STAR’s largely hourly work-force. STAR targeted how employees interact with managersand with other employees. There was emphasis on processissues and working more productively and efficientlytogether, rather than on virtual working, unlike IT.

Even within the IT workforce, it was also important tocompare employees in STAR with those who continued in thetraditional work environment across each function. Forma-tive research, such as interviews with executives and inputfrom an internal study advisory group, revealed that com-pany insiders might write off results if all STAR groupshappened to be in one function. For example, if STAR wasdone only with developers rather than those who tested andsupported applications once they were launched, it mightlimit acceptance. The WFRN team addressed this by rando-mizing groups using an adaptive randomization scheme. Thisensured that there would be some STAR groups comprised ofdevelopers, some with testers, and some in a support func-tion. This example reveals the importance of knowing thecomparison groups that will be the most helpful to insiders asthey learn from the study.

For both sites, the WFRN pilot studies assessed the com-parative effectiveness of the intervention componentsbefore full implementation. This provided evidence of lin-kages to work—family conflict reduction and improved healthand work outcomes.

Principle 4: Use interdisciplinary design teams. Mostwork—family and occupational health interventions aregrounded in one primary discipline and its core philosophicalapproach to deliberate change. But an interdisciplinaryapproach is more likely to produce an effective intervention,since it frequently integrates knowledge from different dis-ciplines that is necessary for change to occur. Interdisciplin-ary work fosters greater synthesis between a psychosocial,disease-oriented focus on preventing workplace influences

Work, family and health change initiatives 59

on stress, and approaches that facilitate or promote lowerstress.

Alleviating workplace sources of occupational stress canbe synergistic with designing workplaces to foster healthyworkers on and off the job. An example would be interven-tions that include both health promotion and health protec-tion, as Harvard researcher Gloria Sorensen and colleaguesargue. Given discipline-specific biases/approaches and thefact that enacting workplace change to enhance work, familyand health is a socially complex phenomenon, interventionteams that utilize multi-disciplinary knowledge in design aremore likely to integrate multiple strands of change: preven-tive and facilitative, and individual and organizationalchange targets.

Principle 4 applied: The multiple perspectives from theWFRN interdisciplinary team and the different occupationalsettings resulted in a holistic, comprehensive approach to thechange initiative. For example, sociologists on the teaminitially emphasized increasing schedule and time control,while organizational behavior scholars focused on enhancingleadership supportive behaviors. Those in public healthemphasized the importance of collecting data addressinghealth risks. STAR included all these elements.

Further, most change efforts focus on the work or thefamily realm. STAR also focused on work—family and healthfrom not only the employment perspective but also thefamily perspective, with data collected on spouses, children,employees and managers for a complete picture of changeeffects.

Principle 5: Bottom up participation with strong man-agement support for the intervention. Interventions shouldbe both bottom up (employee initiated and supported) andtop down (top management initiated and supported), yetmost work—life interventions are largely one or the other.Generally, work—life policies and initiatives are typically notviewed as change-oriented participatory endeavors. Effec-tive designs should combine high employee participationwith top down senior management buy-in and support forthe change. Getting management buy-in to conduct theintervention during work time and having leaders attendjoint meetings with employees are two ways to promotebottom up and top down support.

High involvement of employees in change processes hasemerged as a best practice. It helps to ensure that theintervention is focused on issues that are most relevant toworkers. It also makes it more likely the intervention will beaccepted by members and integrated into the organizationalculture. High involvement also should reduce undermining ofchange during implementation. Such approaches empoweremployees to define the most pressing workplace problems,give input to create change strategies, and experiment withchanges they see as beneficial.

Principle 5 applied: STAR facilitated sessions encouragedemployees to develop and implement alternative approachesto getting work done and culturally interacting. For example,STAR participatory sessions included a role-play exercise todemonstrate how to ‘‘eradicate Sludge’’ — the negative toxiclanguage often heard in the workplace judging how someoneis spending their time. For example, ‘‘I wish I could come inlate.’’ Sludge can also be phrased in ways that seem positive,honoring folks for spending inordinate amounts of time onwork projects. For instance, when people are given awards,

what is frequently said are things like ‘‘he slept under hiscube.’’ Or ‘‘he spent only one weekend this month with hisfamily.’’ The message conveyed is the more time put in on thejob, the more valuable the employee.

Session participants came to recognize how the tone,manner, and language used with coworkers can negativelyimpact the work environment and, as a consequence, canimpact the well-being of employees at work and at home.Both managers and employees participated in these sessions,so this culture change in how people interact permeatedwhole teams. The strategy of asking: ‘‘Is there something Ican help you with?’’ or ‘‘Is there something you need?’’ inresponse to Sludge-like comments, was used by managers toemployees, coworkers to one another and employees tomanagers. Employee involvement continued, in many ofthe sites and work groups, after the sessions ended, andmany teams put STAR discussions on their regular staff meet-ings for several months after training. In the ITworkforce thisincluded discussions about their experiments working differ-ent hours or remotely and coordinating how they might cutback on unproductive meetings. In the nursing homes, com-munication boards were put on some floors to help fosterbetter collaboration.

At the same time, STAR required on-site management toencourage employee session attendance and also was open todifferent approaches encouraged by employees. Supportfrom the participating organizations was critical, as sessionstook place during work time, and sometimes led to overtimecosts in the nursing home environment.

Principle 6: Interventions should be customizable andadaptive in design. Interventions must be customized andadaptive in implementation to ensure the content, deliveryand examples fit the key issues of a particular workforce ororganization. Even if the key intervention ingredientsincluded address work—family conflict (i.e., control, sup-port, culture and job redesign) and are the same acrossworkforces, how they are illustrated and explained can bemodified to increase identification with the change effort.Consider how an intervention might need to be customizedfor lower wage service and high tech professional work-forces. An intervention targeting increased supervisor sup-port and schedule control for a grocery store worker willallow the employee to be able to call into and miss workwhen their child is sick and not have it count against the storediscipline and absenteeism policy. In contrast, an interven-tion intended to increase control and support for salariedemployees might emphasize piloting a new understandingthat individuals may choose to telework several days a weekwhen they can work just as efficiently or more efficientlythat way.

While this notion of customization may seem to be com-mon sense for some, the idea of adapting the interventionsacross sites may be novel in the management and scientificworlds, when comparative analysis and common measure-ment are often goals. In other words, adaptation across sitesdoes not fit neatly into scientific paradigms that reinforcereplication. It also does not fit easily into mandated nation-wide or global business strategies where common changemanagement and benchmarking are being done to leverageperformance indicators.

Most occupational health and work—family interventionsare not sufficiently customized to address variation in work

60 E.E. Kossek et al.

processes, human resource strategies, workforce types, andjob demands. It can seem less time intensive and easier to rollout the same program exactly the same across a firm. Yet thiscan weaken the potential for lasting organizational change.

Principle 6 applied: STAR was customized and adapted toinclude content that was relevant to participants and thenature of work and industries. An example is in the type of jobredesign and change focus of the intervention principles ineach industry. We’ve note the ITemployees could be empow-ered to work wherever and however they want (mostly). Yethealth care employees’ focus was more centered on controlover processes and how work is done or schedules are made.

Customization was also used when discussing the meaningof ‘‘workplace flexibility’’ in training. Examples that areapplicable to employees who can work remotely or flex theirhours informally do not apply to a workforce that is respon-sible for the 24-7 care of people. Health care workers, forexample, must abide by legal requirements determining howmany workers need to be in a building at a given time toensure safety. As one Leef employee commented: ‘‘Health-care workers can’t really go outside and sit under a tree for20 minutes and read a book. . . It’s just different you know. . .’’Since a health care worker in a nursing home often cannottake time off during a shift without impacting coverage.Workers were educated to implement any changes in away that was ‘‘safe, legal, and cost-neutral.’’

Customization also warranted careful consideration ofexamples used and approaches to facilitating sessions. Forexample, pictures and examples from IT or health care wereused in presentations and session content. For the IT workforce, discussions of coordinating with offshore staff (locatedprimarily in India) were incorporated, as part of discussingthe timing of work.

Intervention delivery was also customized, so in the ITwork force, virtual training sessions and the company intra-net were used. This allowed employees working in differentlocations to participate in the same intervention conversa-tions as their managers and peers. For the health care workforce, where most did not have computer access, posterswere used for employee activities and notification of sessionschedules. Since work was 24-7, scheduled sessions were heldduring multiple shifts.

Principle 7: Plan and devote resources to promoteintervention transfer to the work environment. It is impor-tant to include resources in intervention design to ensuretransfer of the changes specified in formal interventionactivities to workers’ jobs. This is a critical step that manyintervention studies and consultants overlook, making it hardto sustain quality of work—life change experiments. This gapis likely because the transfer of the intervention is essentiallya knowledge transfer process. It is a social activity where theresearch knowledge needs to be incorporated by the orga-nization and its members as relevant to its mission. While theresearchers and facilitators may believe that the work—family intervention targeting a healthier workplace shouldbe a priority, workers and management may not understandhow to incorporate this into routines or consider work—familyconflict.

Principle 7 applied: To encourage incorporating the ideasfrom STAR into the way work is done over the long term, theIntervention included structured ‘‘homework’’ activitiesbetween sessions. In the case of the health care industry,a check-in meeting was held about four weeks after the lastfacilitated session, and was facilitated by the nursing homeadministrator. This session was also attended by managersand some key employees (‘‘champions’’ of the Interventioninitiative). It was critical that someone internal facilitate thistransfer meeting, not an outside consultant, so that thegroup felt ownership of the changes that were being imple-mented.

There were also exercises at critical times during theintervention delivery that help workers transfer newlylearned knowledge and skills to their jobs and their day-to-day routines. Self-monitoring activities motivated trans-fer by using goal setting, followed by repeated self-observa-tion, evaluation, and recording of behaviors. Self-observedgaps between actual behaviors and goals (or social norms)activated psychological and behavioral motivational pro-cesses. Employees and managers were also encouraged tohold additional self-led meetings in between formal inter-vention events, during which they shared perspectives, dis-cussed why the intervention was needed and checked in onperceptions of how it was going.

FUTURE CHALLENGES

Intervention scholarship in real-world settings is hard to do,but critical to understanding the working conditions thatreduce work—family conflict and promote health. As UrieBronfenbrenner said, to really understand something, try tochange it. We believe that future research and practiceshould build on and expand upon these principles and exam-ples from the WFRN case study.

Future research should also include evaluations of theactual intervention process. That is, information on theintegrity of intervention implementation and on how theunfolding process of intervention delivery might shapechange trajectories and outcomes. Collecting data on parti-cipation rates and how the intervention is presented andenacted provides important information for evaluating inter-vention fidelity and effectiveness. This will allow for com-parative effectiveness analysis of which interventioncomponents are more effective.

Scholars and practitioners need to shift the lens from afocus on individual strategies to reduce work—family conflictafter it occurs toward prevention-focused organizationalchange initiatives to reduce work—family conflict in theworkplace. Such an approach moves work—life initiativesfrom the margins to the mainstream of current organizationalpractice.

Work, family and health change initiatives 61

SELECTED BIBLIOGRAPHY

For a thorough exploration of the effects of work—familyinitiatives on work—family conflict and business outcomes,see chapter 7 of The Academy of Management Annals, firstpublished on August 1, 2008, entitled ‘‘Getting There fromHere: Research on the Effects of Work—Family Initiatives onWork—Family Conflict and Business Outcomes,’’ written byErin L. Kelly, Ellen Ernst Kossek, Leslie B. Hammer, MaryDurham, Jeremy Bray, Kelly Chermack, Lauren A. Murphy,and Dan Kaskubar. This review of over 150 peer-reviewedstudies from a number of disciplines summarizes the litera-ture and identifies promising avenues for research and con-ceptualization. Also help in suggesting ways to bettertranslate work—family research to practice is an article byE.E. Kossek, B.B. Baltes & R.A. Matthews, ‘‘How Work—Family Research Can Finally Have an Impact In the Work-place, ‘‘Industrial and Organizational Psychology: Perspec-tives on Science and Practice,’’ 2011, 4, 352—369.

To inform the study described in this paper, we drew onempirical Work, Family & Health Network studies and find-ings. The following articles provide information informing thedevelopment and implementation of the current interven-tion. For a description of the impacts of a team-level flex-ibility initiative focused on results, see Phyllis Moen, Wen Fanand Erin L. Kelly, ‘‘Team-Level Flexibility, Work-Home Spil-lover, and Health Behavior,’’ Social Science and Medicine,2013, 84, 69—79, which examined the impacts of a team-level flexibility initiative (ROWE — Results Only Work Envir-onment) on changes in the work-home spillover and healthbehavior of employees at a large U.S. corporation. The issueof schedule control was explored in Erin L. Kelly, Phyllis Moen,and Eric Tranby, ‘‘Changing Workplaces to Reduce Work—Family Conflict: Schedule Control in a White-Collar Organiza-tion,’’ American Sociological Review, 2011, 76(2), 265—290,which considered work—family conflicts — common and con-sequential for employees, their families, and work organiza-tions. Analyses demonstrate that the workplace initiativepositively affects the work—family interface, primarily byincreasing employees’ schedule control.

To better understand the importance of supportive super-visor behavior and the impact it has, see E.E. Kossek, S.Pichler, T. Bodner and L.B. Hammer, ‘‘Workplace SocialSupport and Work—Family Conflict: A Meta-Analysis Clarify-ing the Influence of General and Work—Family SpecificSupervisor and Organizational Support,’’ Personnel Psychol-ogy, 2011, 64(2), 289—313. Other informative and relevantpublications on work—life initiatives include Kossek, S.Lewis, and L. Hammer, ‘‘Work—Life Initiatives and Organiza-tional Change: Overcoming Mixed Messages to Move From theMargin to the Mainstream,’’ Human Relations, 2010, 63(1),3—19.

To better understand how to design and evaluate a ran-domized field intervention to improve leader behaviors tosupport family see the 2011 article by Leslie Hammer, EllenKossek, Todd Bodner, Kent Anger and Kristi Zimmerman called‘‘Clarifying Work—Family Intervention Processes: The Rolesof Work—Family Conflict and Family Supportive SupervisorBehaviors,’’ Journal of Applied Psychology, 96 (1), 134—150.

For a thorough discussion of behavioral self-monitoringmethods and the effects of behavioral self-monitoring imple-mentations in the workplace, see R. Olson and J. Winchester,‘‘Behavioral Self-Monitoring of Safety and Productivity in theWorkplace: A Methodological Primer and Quantitative Litera-ture Review,’’ Journal of Organizational Behavior Manage-ment, 2008, 28, 9—75.

For information on the development, theoretical princi-ples, and effectiveness of the computer-based training pro-gram used to train supervisors in the intervention, see W.K.Anger, D.S. Rohlman, J. Kirkpatrick, R.R. Reed, C.A. Lundeenand D.A. Eckerman, ‘‘cTRAIN: A Computer-Aided TrainingSystem Developed in SuperCard for Teaching Skills UsingBehavioral Education Principles,’’ Behavioral Research Meth-ods, Instruments, & Computers, 2001, 33, 277—281.

For a more in-depth exploration of work-family policies,see the chapter in the Handbook of Occupational Health andWellness, R.J. Gatchel and I.Z. Schultz (Eds.) (New York:Springer, 2013), written by Rosalind. B. King, Georgia Kar-untzos, Lynne M. Casper, Phyllis Moen, Kelly D. Davis, LisaBerkman, Mary Durham, and Ellen Ernst Kossek entitled‘‘Work—Family Balance Issues and Work—Leave Policies.’’Also relevant is the chapter by Ellen Kossek and Brian Dis-telberg published in 2009 in Nan Crouter’s and Alan Booth’sedited book Work—Life Policies, entitled ‘‘Work and FamilyEmployment Policy for a Transformed Workforce: Trends andThemes’’ (Washington, DC: Urban Institute Press), 1—51.

For reviews of flexible schedule policies and links toproductivity and health, see ‘‘Flexible Work Scheduling,’’by Ellen Kossek and Jesse Michel in Handbook of Industrial-Organizational Psychology, edited by Sheldon Zedeck (Wash.,DC: American Psychological Association, 2011), vol. 1, 535—572, and the 2013 report Leveraging Workplace Flexibility:Fostering Engagement and Productivity, published by theSociety for Human Resource Management Foundation.

For a review of behavioral model training, see the articleby P.J. Taylor, D. Rus-Eft & D.W.L. Chan entitled ‘‘A Meta-Analytic Review of Behavior Modeling Training,’’ Journal ofApplied Psychology, 2005, 90, 692—709.

For a review of job stress interventions see an article byN.K. Semmer, ‘‘Job Stress Interventions and the Organizationof Work,’’ Scandinavian Journal of Work, Environment, &Health, 2006, 32, 515—527.

62 E.E. Kossek et al.

Ellen Ernst Kossek holds the Basil S. Turner Chair of Management at Krannert School of Management and is theinaugural research director of the Susan Bulkeley Butler Center at Purdue University. Elected the second presidentof the Work Family Researchers Network, and an APA and SIOP Fellow, she has won awards for advancing researchand practice on gender and diversity in organizations (Professor of Management, Krannert School of Management,4005 Rawls Hall, Purdue University, West Lafayette, IN 47907, United States. Tel.: +1 5173880952; e-mail:ekossek@purdue.edu).

Leslie B. Hammer is a professor of psychology at Portland State University and director of the Center for Work—Family Stress, Safety, and Health. Her research focuses on ways in which organizations can help reduce work andfamily stress and improve positive spillover among employees by facilitating formal and informal workplacesupports (Portland State University, United States).

Erin L. Kelly is a professor of sociology at the University of Minnesota and an affiliate of the Minnesota PopulationCenter. Her research on employers’ flexibility initiatives, family leaves, childcare benefits, and diversity programshas appeared in American Sociological Review, American Journal of Sociology, Law & Society Review, andelsewhere (University of Minnesota, United States).

Phyllis Moen holds a McKnight Presidential Endowed Chair and is a professor of sociology at the University ofMinnesota after twenty-five years at Cornell University. She has published numerous books and articles on careers,work organizations, health, well-being, gender, policies, and families, as they intersect and change over the lifecourse (University of Minnesota, United States).

Work, family and health change initiatives 63