Journal of Adolescent Health 56 (2015) 672e677

www.jahonline.org

Original article

Effects of a Workplace Intervention on Sleep in Employees’ Children

Susan M. McHale, Ph.D. a,*, Katie M. Lawson, Ph.D. a, Kelly D. Davis, Ph.D. a,Lynne Casper, Ph.D. b, Erin L. Kelly, Ph.D. c, and Orfeu Buxton, Ph.D. d,e, f,gaDepartment of Human Development and Family Studies, The Pennsylvania State University, University Park, PennsylvaniabDepartment of Sociology, University of Southern California, Los Angeles, CaliforniacDepartment of Sociology, University of Minnesota, Minneapolis, MinnesotadDepartment of Biobehavioral Health, The Pennsylvania State University, University Park, PennsylvaniaeDivision of Sleep Medicine, Department of Medicine, Brigham and Women’s Hospital, Boston, MassachusettsfCenter for Population and Development Studies, Harvard Chan School of Public Health, Cambridge, MassachusettsgDivision of Sleep Medicine, Harvard Medical School, Boston, Massachusetts

Article history: Received November 15, 2014; Accepted February 13, 2015Keywords: Adolescent sleep; Contextual influences on sleep; Parental employment; Workplace; Intervention

A B S T R A C TIMPLICATIONS AND

Purpose: The implications of sleep patterns for adolescent health are well established, but weknow less about larger contextual influences on youth sleep. We focused on parents’ workplaceexperiences as extrafamilial forces that may affect youth sleep.Methods: In a group-randomized trial focused on employee work groups in the information tech-nology division of a Fortune 500 company, we tested whether a workplace intervention improvedsleep latency, duration, night-to-night variability in duration, and quality of sleep of employees’offspring, aged 9e17 years. The intervention was aimed at promoting employees’ schedule controland supervisor support for personal and family life to decrease employees’workefamily conflict andthereby promote the health of employees, their families, and the work organization. Analysesfocused on 93 parenteadolescent dyads (57 dyads in the intervention and 46 in the comparisongroup) that completed baseline and 12-month follow-up home interviews and a series of telephonediary interviews that were conducted on eight consecutive evenings at each wave.Results: Intent-to-treat analyses of the diary interview data revealed main effects of the inter-vention on youth’s sleep latency, night-to-night variability in sleep duration, and sleep quality, butnot sleep duration.Conclusions: The intervention focused on parents’ work conditions, not on their parenting orparentechild relationships, attesting to the role of larger contextual influences on youth sleep andthe importance of parents’ work experiences in the health of their children.

� 2015 Society for Adolescent Health and Medicine. All rights reserved.

Conflicts of Interest: Buxton reports no direct conflicts of interest but in the in-terest of full disclosure, reports (last 3 yrs) investigator-initiated research grantsupport from Sepracor (now Sunovion, ESRC-0997); and consulting fees fromDinsmore LLC (expert witness testimony) and Matsutani America (scientificadvisory board) outside of the submitted work. All other authors have no conflictsof interest or disclosures to report. This work was produced without commercialfinancial support, and does not involve off-label or investigational use.Disclaimer: The contents of this publication are solely the responsibilityof the authors and do not necessarily represent the official views of these

institutes and offices. The National InDisease Control and Prevention, and otive role in study design, conduct, rearticle.* Address correspondence to: Susan M.

Development and Family Studies, 114 Heversity Park, PA 16802.

E-mail address: mchale@psu.edu (S.M.

1054-139X/� 2015 Society for Adolescent Health and Medicine. All rights reserved.http://dx.doi.org/10.1016/j.jadohealth.2015.02.014

CONTRIBUTION

Sleep is linked to youthhealth, but we know lessabout socialeecologicalinfluences on sleep pat-terns. A workplace inter-vention to reduceemployees’ workefamilyconflict had positive ef-fects on regularity of ado-lescents’ nighttime sleepduration, sleep quality,and time to fall asleep, butnot sleep duration.

stitutes of Health, the Centers forther funders did not play a direc-porting, or decision to submit the

McHale, Ph.D., Department of Humannderson, Penn State University, Uni-

McHale).

S.M. McHale et al. / Journal of Adolescent Health 56 (2015) 672e677 673

Adolescence is a period of dramatic physical, cognitive, andsocioemotional developmental changes, including in youth sleeppatterns. Both psychosocial factors such as youth’s increasinginvolvement in the world beyond home and biological factorssuch as circadian time, for example, underlie changes in ado-lescents’ preferences for going to bed and getting up at latertimes [1]. Nonetheless, sleep remains critical for adolescents’health and development [1]. Indeed, accumulating evidencedocuments the significance of sleep patterns for adolescents’well-being in domains ranging from psychological and socialadjustment and health risk behaviors to school performance andobesity [2e9]. U.S. adolescents, however, experience greatersleep deprivation than either children or adults [10]dalmost 2hours less than the recommended average of 9 hours per night[11]. And, national data suggest that sleep deprivation increasesacross adolescence. [12].

Although research has examined the characteristics andhealth implications of youth sleep, there remain gaps in theliterature about the social contextual determinants of healthfulsleep [13]. Prior work highlights demographic factors (e.g.,ethnicity, socioeconomic status), but a focus on such status var-iables does not provide insights about malleable processes andconditions to target for intervention. Some work also shows thatfamily processes such as parental warmth are linked to youthsleep [13]. Correlational research designs, however, limit con-clusions about the causal links between the social ecology andyouth sleep because unmeasured third variables may explainpatterns of association.

We grounded our work in an ecological model, which holdsthat youth are embedded within a system of nested contextsranging from more proximal (e.g., family) to distal (e.g., societalinstitutions) influences [14]. Using a randomized controlled fieldtrial design, we tested whether an experimental intervention,aimed at reducing employees’ workefamily conflict, improvedsleep in employee-parents’ adolescent-aged offspring. Sleep wasassessed in terms of the duration, night-to-night variability induration, latency, and quality of youth’s sleep.

From an ecological perspective, youth health is influenced bythe microsystems of everyday life, such as family and school, butin addition, by contexts in which youth do not directly partici-pate. Such exosystem influences include, for example, theirteachers’ family lives and their parents’ workplace conditions[14]. Consistent with ecological tenets, the workehome re-sources model [15] posits that parents’ work experiences cancross over to negatively affect their children’s health by depletingparents’ personal resources, such as positive mood and timeneeded for monitoring and promoting children’s healthful dailyroutines. Parents’work experiences can also enhance family roleperformance and foster children’s well-being when those expe-riences provide parents with personal resources, such as controlover their work schedules, that allow time for parentalresponsibilities.

As noted, research on social contextual correlates of youthsleep focuses on demographic characteristics, such as familysocioeconomic status and parents’marital status. Reviewing thisliterature, Hale et al. [13] concluded that these status character-istics may mark social/psychological stressors such as financialhardship and family conflict, which serve as the mechanismslinking demographic factors and youth sleep patterns. Researchon family dynamics is consistent with this conclusion, showinglinks between both marital and parentechild conflict andyouth sleep [16,17]. By contrast, positive parental involvement,

including parentechild shared time, monitoring, and appro-priate limit-setting, may promote healthful sleep [13,17,18].

Prior research also documents associations between parents’work experiences and the same kinds of parenting behaviors thathave been linked to youth sleep patterns. For example, parents’job demands were related to less parentechild shared time andwarmth but more conflict [19e21], and a negative social climateat work was correlated with negative parentechild interactions[22]. In contrast, employees’ schedule flexibility was related tomore parentechild shared time and, in turn, greater warmth[21], and employees’ positive interactions with supervisors wereassociated with greater parental warmth [23]. In the UnitedStates, limited public policy means that employers are left todevelop programs and practices that support working families[24,25]. Although past decades have seen new family-friendlyworkplace policies, there are few systematic data on the effec-tiveness of those policies for improving employee health, and weknow almost nothing about whether and how family-orientedwork policies benefit the physical health of employees’ chil-dren [26].

This study used data from a field test of the STAR (Support,Transform, Achieve Results) workplace intervention program toexamine the role of parents’ work experiences in theiradolescent-aged offspring’s sleep patterns. STAR was designed toreduce employees’ workefamily conflict by promoting job re-sources in two domains: supervisor support for employees’personal and family lives and employees’ perceived control overtheir work schedules. Importantly, for the purposes of this study,the intervention did not target parenting practices, althoughprior findings on the links between supervisor support andparental warmth and schedule control and parental involvementsuggested that STAR effects might spillover to affect the sameparenting behaviors that have been implicated in youth sleep.

STAR was implemented in the information technology (IT)division of a U.S. Fortune 500 company over a 3-month periodand included training sessions for managers to learn about theintervention and strategies to support employees’ personal andfamily lives while maintaining high levels of work performance.The supervisor support training also included a self-paced,computer-based training followed by real-time self-monitoringof managers’ supportive behaviors via an iPod Touch (Apple Inc.,Cupertino, CA) with an alarm reminder to log support behaviors.In addition, STAR involved 8 hours of work group participatorytraining sessions (four sessions) for managers and employees.Highly scripted sessions focused on targeted areas for change(e.g., attitudes that more hours spent at the office reflectedgreater commitment or productivity). The sessions were highlyinteractive and aimed at identifying new work practices thatwould focus employees’ time and attention on key work resultsrather than face time. The intervention is described in detail, andprogram materials are available online [27].

The first analyses of the effects of STAR established that, at the6-month follow-up, the intervention had predicted positive ef-fects: Employees who were randomly assigned to the interven-tion reported more schedule control and supervisor support forfamily and personal life and less conflict between work andfamily responsibilities than did those in the usual practice (UP)condition. Additional analyses indicated that STAR employeesalmost doubled their hours of work at home and were morelikely to describe their schedules as “variable” at follow-up; theyalso exhibited more time adequacy for activities with familymembers and time spent with their children [28]. Their greater

Table 1Participant demographic information (n ¼ 93), mean (standard deviation) ornumber (%)

Intervention(N ¼ 57)

Usual practice(N ¼ 36)

Employee measuresAge (years) 45.58 (6.16) 43.56 (4.71)Gender (% female) 28 (49.12) 14 (38.89)Education (% college graduate) 46 (80.70) 28 (77.78)Marital status (% married/cohabiting) 49 (85.96) 34 (94.44)Number of children living in household 1.93 (.90) 2.25 (1.20)Work hours (hours/week) 46.84 (6.10) 45.78 (5.54)Income (% more than 100,000/year) 41 (77.36) 22 (64.71)Tenure (years) 13.64 (7.66) 12.32 (5.74)Diary days completed 7.82 (.47) 7.78 (.48)

Youth measuresAge 12.93 (1.99) 13.31 (2.40)Gender (% female) 29 (50.88) 17 (47.22)Diary days completed 7.86 (.40) 7.88 (.42)

The results of independent samples t tests and chi-squared tests revealed nosignificant differences between the intervention and usual practice groups onthese measures.

S.M. McHale et al. / Journal of Adolescent Health 56 (2015) 672e677674

availability may mean that parents are more knowledgeableabout their children’s activities and are better able to orchestratefamily routines that promote healthful sleep patterns in youth.

The current analyses built on this initial evaluation of theintervention to test its effects on the sleep patterns of theadolescent-aged offspring of employee-parents. We measuredfour domains of youth sleep that have been linked in priorresearch to youth physical, psychological, and behavioral health:sleep duration (hours of sleep per night), night-to-night variationin sleep duration (to mark consistent sleep routines), sleep onsetlatency (reflecting difficulty falling asleep), and perceived sleepquality [1,2,10,17]. Based on the tenets of the workehome re-sources model [15] and our prior research [28], we hypothesizedthat, by the 12-month follow-up, youth whose parents hadparticipated in STAR would exhibit more positive sleep patternsthan youth whose parents were randomly assigned to UP. Find-ings that healthful sleepmay decline across adolescence as youthbecome more autonomous and involved in activities beyond thehome [8] led us to expect that intervention effects might mani-fest in the form of buffering age-related declines in healthfulsleep patterns.

Methods

The data came from the baseline and 12-month follow-upwaves of a group-randomized field experiment designed totest the effects of a workplace intervention on the health andwell-being of employees, their families, and their work or-ganization. The research team partnered with a Fortune 500company, pseudonym TOMO, to recruit study participantsfrom its IT division. After the baseline data collection, teamsof employees who worked together and/or reported to thesame supervisors (N ¼ 56 teams) were assigned to theintervention or UP condition. Given the differing sizes andfunctions of the work teams, we used a modified biased coinrandomization approach for group randomization [27] to yielda balance across the intervention and UP conditions in jobfunction, team size, and executive (vice president) leader.Importantly, for our purposes here, employee work groups,not the youth who were the focus of these analyses, were theunit of group assignment.

STAR was introduced by the organization’s IT executives as acompany-sponsored pilot program. Our research team andoutside consultants jointly developed the intervention, custom-izing the materials for the targeted IT work force. Four groupfacilitators delivered the STAR intervention to supervisors andemployees. The evaluation study was introduced separately andnot directly linked to STAR, and a separate group of research staff,blind to participants’ group assignments, was responsible fordata collection.

Participants

The sample of parenteyouth dyads was drawn from the largersample of employees who participated in workplace interviews.Of the 823 employees who completed workplace interviews, 222(26.97%) were a parent of a child aged 9e17 years who lived athome for at least 4 days a week and thus eligible for the familycomponent of our study. Of the 147 parenteyouth dyads thatcompleted the baseline home interview, 93 completed at leastthree of the eight daily telephone diary interviews at bothbaseline and the 12-month follow-up and were the focus of the

present analyses. Tests for differential attrition (t and chi-squaredtests) revealed no differences between those who remainedversus those who left the study at 12 months as a function ofdemographic or work characteristics (i.e., age, gender, income,ethnicity, marital status, number of children in the household,job tenure) or the target intervention variables (i.e., schedulecontrol, family-supportive supervisor behaviors, work-to-familyconflict). Eligible (N ¼ 222) and participating (N ¼ 147) fam-ilies differed only in income, ethnicity, and youth age (partici-pants had lower salaries, were less likely to be minorities, andyouth were older).

At baseline, average annual income of participating familiesfell in the range of $80,000e$90,000, and most parents (80.80%)had a bachelor’s degree or more education. In addition, 83.2% ofyouth had married parents. Most (67.00%) were white, non-Hispanic, with smaller percentages of Asian/Pacific Islander(20.00%), Hispanic (9.60%), black, non-Hispanic (1.60%), andmultiracial (1.60%) youth. Parents (47.00% female) averaged45.05 years of age (standard deviation [SD] ¼ 6.03), 12.73 yearsemployed by the company (SD ¼ 6.45), and 46.45 hours of workper week (SD ¼ 5.94), although all worked daytime shifts. Youthparticipants (52.78% female) averaged 13.34 years of age (SD ¼2.30). There were no differences on any of these measures foryouth whose parents were assigned to the STAR (n ¼ 57) versusthe UP (n ¼ 36) conditions, and no differential attrition acrossgroups (Table 1).

Procedures

Trained interviewers conducted interviews with employeesat the worksite and with employees and their children at theirhomes at baseline and the12 month follow-up. During thelatter, parents provided consent and youth assented to partic-ipation. After the home interviews, in eight, consecutive,nightly phone calls averaging 15 minutes, youth reported ontheir experiences during the 24 hours before the call. The datacollection centers’ institutional review boards approved theprocedures. Families received $150 for participation at baselineand $200 for participation at the 12-month follow-up tele-phone diary assessment.

Table 2Means (standard deviations [SDs]) and Pearson Correlations for Sleep Measures[1] at baseline (above diagonal) and 12-month follow-up (below diagonal)

Baseline 12 months Duration Variability Latency Quality

Mean (SD) Mean (SD)

Duration 8.99 (1.55) 8.94 (1.74) .72** �.27** .20* .13Variability 1.09 (.49) 1.34 (.59) �.05 .33** .01 �.17Latency 1.52 (.83) 1.50 (.87) .28** �.14 .64** �.25*

Quality 3.53 (.62) 3.51 (.55) �.00 �.02 �.40** .53**

Stability coefficients (bolded) on diagonal.Duration measured in hours, variability measured as the within-person standarddeviation of the duration scores across days, latency rated on a four-point scale(1 ¼ about 15 minutes; 4 ¼ more than 60 minutes), and quality rated on a four-point scale (1 ¼ very badly; 4 ¼ very well).*p < .05; **p < .01.

S.M. McHale et al. / Journal of Adolescent Health 56 (2015) 672e677 675

Measures

We adapted items from the Pittsburg Sleep Quality Index [29]for the diary interviews. During each nightly interview, youthwere asked four questions about the prior night: “What time didyou go to bed?” “What time did you wake up this morning?”“How long did it take you to fall asleep?” and “How well did yousleep last night?” Bedtime and wakeup time were recorded inmilitary time, and the interval between them was calculated toindex sleep duration (hours/day). Night-to-night sleep durationvariability was the within-person standard deviation of theduration scores across the diary days, with high scores signifyinggreater variation [30]. Sleep latency was coded on a four-pointordinal scale (1 ¼ about 15 minutes; 2 ¼ between 15 and30 minutes; 3¼ 30 to 60 minutes, 4¼more than an hour). Youthrated their sleep quality on each call using a four-point ordinalscale (1 ¼ very badly; 4 ¼ very well). More than 90% of youthcompleted all eight calls, and all completed at least six calls.

Results

To account for the clustered design (occasions within persons),we conducted mixed effect, multilevel models using SAS ProcMixed (SAS Institute, Cary, NC). To test intervention effects on thesleep duration, latency and quality dependent measures, the twowaves of data (baseline and 12-month follow-up) were stacked(i.e., 16 rows per person, 8 days for each wave). The analysis ofsleep variability required only one within-individual measure,

Table 3Mixed model results (coefficients and standard errors) of intervention effects on yout

Sleep duration Sle

CovariatesDay in study .01 (.02)Youth gender .30 (.15)* .0Youth age �.27 (.04)*** .0Wave mean school days �1.03 (.18)*** �.1

Intervention effectIntercept 8.82 (.16)*** .9Wave �.07 (.12) .4Condition �.10 (.17) .2Wave 3 condition L.07 (.15) L.2

Sleep duration measured in hours, variability measured as the within-person standard(1 ¼ about 15 minutes; 4 ¼ more than an hour), and quality rated on a four-point scIntervention effects are bolded. Wave coded as baseline ¼ 0, 12-month follow-up ¼ 1Youth gender coded as 0 ¼ female, 1 ¼ male.*p < .05; **p < .01; ***p < .001.

wave.Wave (0¼ baseline, 1¼12-month follow-up) and conditionvariables (0 ¼ UP, 1 ¼ STAR) were the primary predictors; theestimate obtained for their interaction indicated whether thechange in youth sleep outcomes from the baseline to 12-monthfollow-up differed for the UP versus STAR groups. Given therelatively small sample size, to limit the number of factors, we firsttested for links between potential covariates and between cova-riates and sleep measures, including parent income, ethnicity, andthe workplace intervention targets. Given nonsignificant effects,these factors were excluded from the final models. The finalmodels included youth age and gender (0 ¼ female), day in study(0 ¼ Day 1, 7 ¼ Day 8), and the percentage of the eight diary daysat each wave that the youth attended school. The latter was anindex of the timing of the eight-day diary.

Descriptive data are shown in Table 2 and results in Table 3.Analyses revealed no effects for sleep duration, but predictedeffects of the intervention emerged for the other measures: (1)youth whose parents were assigned to the STAR interventionshowed no change in sleep duration variability, but youth in theUP group increased in night-to-night variability in sleep durationfrom the baseline to the 12-month follow-up, effect size ¼ .23;(2) STAR youth exhibited decreases in latency of sleep onset, butyouth in the UP group exhibited increases in the time it tookthem to fall asleep, effect size¼ .15; and (3) STAR youth exhibitedno change in sleep quality, but youth in the UP group exhibiteddeclines, effect size ¼ .13.

Discussion

Our findings showed that a workplace intervention,designed to reduce employees’ workefamily conflict throughincreasing employees’ schedule control and supervisor supportfor work and personal life, had corollary effects on the sleep ofemployees’ adolescent-aged offspring. The results suggest thatthe intervention served to buffer youth from age-related de-clines in healthful sleep patterns. Prior research on socialcontextual factors in youth sleep suggested that markers offamily stress are associated with less healthful sleep patterns[9,13]. Correlational designs, however, do not allow for stronginferences about the role of the social context in youth sleep,because unmeasured third variables can account for both socialinfluences and youth health. Using a group-randomizedexperimental design, this study documented the significanceof social contextual influences on youth sleep by demonstrating

h sleep duration, variability, latency, and quality [1].

ep variability Sleep latency Sleep quality

d �.02 (.01)* �.01 (.01)5 (.08) .13 (.13) �.01 (.08)7 (.02)** �.05 (.03) �.02 (.02)1 (.13) .35 (.10)*** .07 (.07)

6 (.09)*** 1.48 (.13)*** 3.69 (.08)***

1 (.10)*** .16 (.05)** �.10 (.04)*

1 (.11) �.01 (.14) �.22 (.08)**

6 (.13)* L.27 (.06)*** .15 (.05)**

deviation of the duration scores across days, latency rated on a four-point scaleale (1 ¼ very badly; 4 ¼ very well).. Condition coded as 0 ¼ usual practice, 1 ¼ Support, Transform, Achieve Results.

S.M. McHale et al. / Journal of Adolescent Health 56 (2015) 672e677676

that experimentally-induced changes in the exosystem can leadto more healthful sleep in adolescence. Our workgroup randomassignment meant that youth, themselves, were not randomlyassigned to the intervention, limiting the causal inferences thatcan be drawn from this study. That group differences in changesover time in youth sleep were evident despite the fact that theworkplace intervention targeted neither parenting nor parent-youth relationships, however, attests to the significance ofparents’ work experiences in youth health.

Taken together, our results have several implications for anunderstanding of youth sleep and adolescent health, moregenerally. First, the findings were consistent with the tenets of anecological model, which holds that youth are embedded within asystem of nested contexts and highlights the significance of forcesbeyond youth’s immediate settings in their well-being anddevelopment [14]. Our findings document the power of exosysteminfluencesdcontexts outside youth’s own everyday experi-encesdto affect youth sleep, and they alert practitioners to takethese more distal and sometimes malleable influences into ac-count in efforts to promote youth health and health behaviors.Second, although impediments to healthful sleep may changeacross adolescence as youth become increasingly autonomous andinvolved in the world beyond home [13], in this study, exosysteminfluences emanating from parents’work conditions were evidentinto late adolescence. Our findings are thus consistent with a bodyof research on adolescent development that highlights thecontinued importance of parents and family life across thisdevelopmental period [31]. Finally, our results are congruent witha body of correlational research linking parents’ work conditionsand youthwell-being, but which to date, has not focused onyouthsleep or health. As noted, the STAR program included no compo-nents focused on parenting practices and behaviors, underscoringthe potentially powerful effects of parents’ work experiences ontheir children. The significant effects of STAR should also beviewed in the light of mixed findings from tests of youth-orientedpsychoeducational interventions for promoting healthful sleep.Although increases in youth knowledge have been documented,program effects on sleep behaviors have been limited [32,33].

An important direction for research on workplace effects is totest whether resources such as schedule control and supervisorsupport have effects on the sleep patterns of youth whose par-ents are employed in other kinds of industries. In particular, ourfindings are limited to employees with relatively high incomesand education and should be replicated in less-advantagedsamples. The small sample size precluded tests of moderatingfactors, including youth age, gender, and seasonality differencesin youth sleep, other directions for future research. In addition,our study is limited by its reliance on youth self-reports of theirsleep duration and timing. The measure of sleep durationcaptured time in bed, or sleep opportunity, and thus may haveoverestimated actual sleep duration. Indeed, average sleepduration in this samplewas longer than typically reported, whichmay also reflect our focus on a relatively advantaged sample ofyouth. Research on a more socioeconomically diverse group ofyouth using objective measures such as wrist actigraphy wouldprovide stronger evidence about social contextual effects onyouth sleep. Of particular interest is whether interventions likeSTAR have stronger impacts on youth who are at higher risk forsleep problems.

Another step is to identify the proximal processes throughwhich workplace policies and programs like STAR can affectyouth sleep. As noted, our prior research showed that STAR

increased two workplace resources, schedule control and su-pervisor support and reduced employees’ work-family conflict.Furthermore, STAR positively impacted employees’ reports oftime adequacy and time spent with their children [28]. Byproviding employees with more control over their work sched-ules, STAR parents may have been able to align their time athome to fit their children’s schedules and needs. Although wedid not detect intervention effects on sleep duration, findingsthat youth whose parents participated in STAR did not show thesame increase in night-to-night variation in sleep duration asyouth with parents in the UP group are consistent with the ideathat parental involvement, which provides opportunities formonitoring and developmentally appropriate limit-setting, canpromote regular sleep routines during a developmental periodwhen school, work, and peer activities may otherwise promotetheir decline. Another mechanism throughwhich STARmay havehad its impact is through its documented effects on employees’workefamily conflict. Prior research highlights the role of familystressors in youth sleep, including through its effects on youth’semotional security [13,34]. Our findings are consistent with theidea that reducing parents’ work-related stress can have positiveimpacts on their children’s sleep, including sleep quality and thetime it takes them to fall asleep.

At the most general level, our findings speak to the impor-tance of looking beyond the immediate settings of youth’s dailylives for influences in the larger environment that have animpact on their health and health behaviors. As such, the resultsare consistent with an ecological perspective, which highlightsthe embeddedness of youth in a larger, multilayered system ofinteracting influences [14]. From this perspective, interventionsthat target only individual adolescents may not be effective ifchanges to behavior and health fail to take into accountpowerful and potentially competing influences in the largerecology. An ecological perspective also opens up new oppor-tunities to intervene at points in the system that are mostmalleable and that may have the broadest impact. Consistentwith the workehome model [15], our findings suggest thatproviding parents with workplace resources that reduce theirexperiences of workefamily conflict may alter the family sys-tem inways that support and promote youth health. At the mostgeneral level, the significant effects of the STAR intervention onyouth sleep attest to the power of parents’ workplace condi-tions to affect the health of their children.

Acknowledgments

Special acknowledgement goes to Extramural Staff ScienceCollaborator, Rosalind Berkowitz King, Ph.D. and Lynne Casper,Ph.D. for design of the original Workplace, Family, Health andWell-Being Network Initiative. We also wish to express ourgratitude to the worksites, employers, and employees whoparticipated in this research and to Rosalie Ammerman andEmily Fidler for assistance in preparing this article.

Funding Sources

This research was conducted as part of the Work, Family andHealth Network (www.WorkFamilyHealthNetwork.org), whichwas funded by a cooperative agreement through the NationalInstitutes of Health and the Centers for Disease Control andPrevention: Eunice Kennedy Shriver National Institute of ChildHealth and Human Development (grants U01HD051217,

S.M. McHale et al. / Journal of Adolescent Health 56 (2015) 672e677 677

U01HD051218, U01HD051256, U01HD051276), National Instituteon Aging (grant U01AG027669), Office of Behavioral and ScienceSciences Research, and National Institute for Occupational Safetyand Health (grant U01OH008788, U01HD059773). Grants fromthe William T. Grant Foundation (#9844), Alfred P. Sloan Foun-dation, and the Administration for Children and Families haveprovided additional funding.

References

[1] Carskadon M, Tarokh L. Developmental changes in circadian timing andsleep: Adolescence and emerging adulthood. In: Wolfson A, Montgomery-Downs H, eds. The Oxford Handbook of Infant, Child, and Adolescent Sleepand Behavior. New York: Oxford; 2013:70e7.

[2] Fuligni A, Hardway C. Daily variations in adolescents’ sleep, activities, andpsychological well-being. J Res Adolesc 2006;16:353e78.

[3] Lemola S, Raikkonen K, Scheier M, et al. Sleep quantity, quality and opti-mism in children. J Sleep Res 2011;20:12e20.

[4] Peach HD, Gaultney JF. Sleep, impulse control, and sensation-seekingpredict delinquent behavior in adolescents, emerging adults, and adults.J Adolesc Health 2013;53:293e9.

[5] Institute of Medicine. Sleep disorders and sleep deprivation: An unmetpublic health problem. Washington, DC: The National Academies Press;2006.

[6] Meijer AM, Habekothé RT, Van DenWittenboer GLH. Time in bed, quality ofsleep and school functioning of children. J Sleep Res 2000;9:145e53.

[7] Tatone-Tokuda F, Dubois L, Ramsay T, et al. Sex differences in the associ-ation between sleep duration, diet, and body mass index: A birth cohortstudy. J Sleep Res 2012;21:448e60.

[8] Snell EK, Adam EK, Duncan G. Sleep and the body mass index and over-weight status of children and adolescents. Child Dev 2007;78:309e23.

[9] Silva GE, Goodwin JL, Parthasarathy S, et al. Longitudinal association be-tween short sleep, body weight, and emotional and learning problems inHispanic and Caucasian children. Sleep 2011;34:1197e205.

[10] Wolfson AR, Carskadon MA. Understanding adolescents’ sleep patterns andschool performance: A critical appraisal. Sleep Med Rev 2003;7:491e506.

[11] Knutson KL, Lauderdale DS. Sociodemographic and behavioral predictors ofbed time and wake time among US adolescents aged 15 to 17 years.J Pediatr 2009;154:426e30.

[12] Eaton DK, McKnight-Eily LR, Lowry R, et al. Prevalence of insufficient,borderline, and optimal hours of sleep among high school studentsdUnitedStates, 2007. J Adolesc Health 2010;46:399e401.

[13] Hale L, Parente V, Phillips GK. Social determinants of children’s sleep. In:Wolfson AR, Montgomery-Downs HE, eds. The Oxford Handbook of Infant,Child, and Adolescent Sleep and Behavior. New York: Oxford;2013:99e112.

[14] Bronfenbrenner U, Crouter AC. Evolution of environmental models indevelopmental research. In: Kessen W, Mussen PH, eds. Handbook of ChildPsychology. New York: Wiley; 1983:357e414.

[15] ten Brummelhuis LL, Bakker AB. A resource perspective on the work-homeinterface: The work-home resource model. Am Psychol 2012;67:545e56.

[16] El-Sheikh M, Buckhalt JA, Mize J, Acebo C. Marital conflict and disruption ofchildren’s sleep. Child Dev 2006;77:31e43.

[17] McHale SM, Kim J, Kan ML, Updegraff K. Sleep in Mexican American ado-lescents: Social, ecological and well-being correlates. J Youth Adolesc 2011;40:666e79.

[18] Adam EK, Snell EK, Pendry P. Sleep timing and quantity in ecological andfamily context: A nationally representative time-diary study. J Fam Psychol2007;21:4e19.

[19] Milkie MA, Mattingly MJ, Nomaguchi KM, et al. The time squeeze: Parentalstatuses and feelings about time with children. J Marriage Fam 2004;66:739e61.

[20] Ransford CR, Crouter AC, McHale SM. Implications of work pressure andsupervisor support for fathers’, mothers’ and adolescents’ relationshipsand well-being in dual-earner families. Community Work Fam 2008;11:37e60.

[21] Roeters A, Van Der Lippe T, Kluwer ES. Work characteristics and parent-child relationship quality: The mediating role of temporal involvement.J Marriage Fam 2010;72:1317e28.

[22] Repetti RL. Short-term and long-term processes linking job stressors tofather-child interaction. Soc Dev 1994;3:1e15.

[23] Gassman-Pines A. Associations of low-income working mothers’ daily in-teractions with supervisors and mother-child interactions. J Marriage Fam2011;73:67e76.

[24] Kelly EL. Discrimination against caregivers? Gendered family re-sponsibilities, employer practices, and work rewards. In: Nielsen LB,Nelson RL, eds. The Handbook of Employment Discrimination Research.New York: Kluwer Academic Publishers; 2005:341e62.

[25] Waldfogel J. Work and family research: A public policy perspective. In:Bianchi SM, Casper LM, King RB, eds. Work, Family, Health, and Well-Being.Mahwah, NJ: Erlbaum; 2005:227e38.

[26] Kossek EE, Lewis S, Hammer LB. Work-life initiatives in organizationalchange: Overcoming mixed messages to move from the margin to themainstream. Hum Relat 2010;63:3e19.

[27] Bray JW, Kelly EL, Hammer LB, et al. An integrative, multilevel, and trans-disciplinary research approach to challenges of work, family, and health. ResTriangle Park, NC: RTI Press. RTI Press publication no. MR-0024-1303. Avail-able at: http://www.rti.org/publications/rtipress.cfm?pubid¼20777; 2013.Accessed July 24, 2014.

[28] Kelly EL, Moen P, Oakes JM, et al. Changing work and work-family conflict:Evidence from the work, family, and health network. Am Sociol Rev 2014;79:485e516.

[29] Buysse DJ, Reynolds CF, Monk TH, et al. The Pittsburgh sleep quality index:A new instrument for psychiatric practice and research. Psychiatry Res1989;28:193e213.

[30] Ram N, Gerstorf D, Lindenberger U, Smith J. Developmental change andintraindividual variability: Relating cognitive aging to cognitive plasticity,cardiovascular lability, and emotional diversity. Psychol Aging 2011;26:363e71.

[31] Collins WA, Laursen B. Parent-adolescent relationships and influences. In:Damon W, Lerner RL, eds. Handbook of Adolescent Psychology. Hoboken,NJ: Wiley; 2004:331e62.

[32] Cain N, Gradisar M, Moseley L. A motivational school-based interventionfor adolescent sleep problems. Sleep Med 2011;12:246e51.

[33] Moseley L, Gradisar M. Evaluation of a school-based intervention foradolescent sleep problems. Sleep 2009;32:334e41.

[34] El-Sheikh M, Buckhalt JA, Cummings ME, Keller P. Sleep disruptions andemotional insecurity are pathways of risk for children. J Child PsycholPsychiatry 2007;48:88e96.