Advertisement

Physical Activity and Body Mass Index

The Contribution of Age and Workplace Characteristics

      Background

      The workplace is an important domain for adults, and many effective interventions targeting physical activity and weight reduction have been implemented in the workplace. However, the U.S. workforce is aging, and few studies have examined the relationship of BMI, physical activity, and age as they relate to workplace characteristics.

      Purpose

      This paper reports on the distribution of physical activity and BMI by age in a population of hospital-based healthcare workers and investigates the relationships among workplace characteristics, physical activity, and BMI.

      Methods

      Data from a survey of patient care workers in two large academic hospitals in the Boston area were collected in late 2009 and analyzed in early 2013.

      Results

      In multivariate models, workers reporting greater decision latitude (OR=1.02, 95% CI=1.01, 1.03) and job flexibility (OR=1.05, 95% CI=1.01, 1.10) reported greater physical activity. Overweight and obesity increased with age (p<0.01), even after adjusting for workplace characteristics. Sleep deficiency (OR=1.56, 95% CI=1.15, 2.12) and workplace harassment (OR=1.62, 95% CI=1.20, 2.18) were also associated with obesity.

      Conclusions

      These findings underscore the persistent impact of the work environment for workers of all ages. Based on these results, programs or policies aimed at improving the work environment, especially decision latitude, job flexibility, and workplace harassment should be included in the design of worksite-based health promotion interventions targeting physical activity or obesity.

      Introduction

      Evidence-based cancer-prevention strategies lie largely in the realm of public health and behavioral intervention.
      • Colditz G.A.
      • Wolin K.Y.
      • Gehlert S.
      Applying what we know to accelerate cancer prevention.
      Two strong risk factors for cancer are physical inactivity and obesity.
      • Colditz G.A.
      • Wolin K.Y.
      • Gehlert S.
      Applying what we know to accelerate cancer prevention.
      • Thune I.
      • Furberg A.S.
      Physical activity and cancer risk: dose–response and cancer, all sites and site-specific.
      • Ligibel J.
      Lifestyle factors in cancer survivorship.
      • Pischon T.
      • Nothlings U.
      • Boeing H.
      Obesity and cancer.
      For both physical activity and obesity, there is increasing evidence that targeting environmental and contextual factors can strengthen the impact of behavioral interventions to reduce cancer risk.
      • Colditz G.A.
      • Wolin K.Y.
      • Gehlert S.
      Applying what we know to accelerate cancer prevention.
      • Kahn L.K.
      • Sobush K.
      • Keener D.
      • et al.
      Recommended community strategies and measurements to prevent obesity in the U.S.
      • Salinardi T.C.
      • Batra P.
      • Urban L.E.
      • et al.
      Lifestyle intervention reduces body weight and improves cardiometabolic risk factors in worksites.
      • Verweij L.M.
      • Proper K.I.
      • Weel A.N.H.
      • et al.
      The application of an occupational health guideline reduces sedentary behavior and increases fruit intake at work: results from an RCT.
      One context in which behavioral interventions have been successfully implemented is the workplace. Given that approximately 64% of adults are employed and spend an average of 34 hours per week at work, the workplace remains an important domain for adults.

      Bureau of Labor Statistics, U.S. Department of Labor. News release, January 2013. www.bls.gov/bls/newsrels.htm.

      The workplace can have an important effect on worker health, both positive and negative. For example, adverse health effects can result from work overload, excessive demands, role conflict, job strain, shift work, and inflexible schedules.
      • Taylor S.E.
      • Repetti R.L.
      • Seeman T.
      Health psychology: what is an unhealthy environment and how does it get under the skin?.
      • Sorensen G.
      • Stoddard A.M.
      • Stoffel S.
      • et al.
      The role of the work context in multiple wellness outcomes for hospital patient care workers.
      • Sorensen G.
      • Stoddard A.
      • Dubowitz T.
      • Barbeau E.M.
      • Berkman L.F.
      • Peterson K.E.
      The influence of social context on changes in fruit and vegetable consumption: results of the healthy directions studies.
      Conversely, work may also have a positive effect on health by providing income, access to health care, linkages to social networks, and access to health promotion programs.
      Few studies have examined the relationship between BMI and physical activity by age, in relation to workplace characteristics. This gap in the literature is important to address, as the median age of the U.S. labor force continues to rise and employers will increasingly rely on older workers.
      • Caban A.J.
      • Lee D.J.
      • Fleming L.E.
      • Gomez-Marin O.
      • LeBlanc W.
      • Pitman T.
      Obesity in U.S. workers: the National Health Interview Survey, 1986–2002.
      This trend is especially striking in the healthcare industry, as fewer young people enter this field and about 70% of current workers will retire in 20–25 years.
      • Caban-Martinez A.J.
      • Lee D.J.
      • Fleming L.E.
      • et al.
      Leisure-time physical activity levels of the U.S. workforce.
      Further, patient care work is physically and psychologically demanding, involves shift work, and puts workers at high risk for musculoskeletal injury. These factors likely affect cancer risk–related behaviors, as patient care workers are at higher risk for both obesity and physical inactivity.
      • Sorensen G.
      • Stoddard A.
      • Dubowitz T.
      • Barbeau E.M.
      • Berkman L.F.
      • Peterson K.E.
      The influence of social context on changes in fruit and vegetable consumption: results of the healthy directions studies.
      • Loeppke R.R.
      • Schill A.L.
      • Chosewood L.C.
      • et al.
      Advancing workplace health protection and promotion for an aging workforce.
      • Silverstein M.
      Meeting the challenges of an aging workforce.
      Further, this impact may grow stronger as workers age, as the body’s natural resiliency to physical and psychosocial stressors decreases with age.
      • Caban A.J.
      • Lee D.J.
      • Fleming L.E.
      • Gomez-Marin O.
      • LeBlanc W.
      • Pitman T.
      Obesity in U.S. workers: the National Health Interview Survey, 1986–2002.
      • Jones M.K.
      • Latreille P.L.
      • Sloane P.J.
      • Staneva A.V.
      Work-related health risks in Europe: are older workers more vulnerable?.
      This paper presents findings from a study of patient care workers (including registered nurses, licensed practical nurses, and patient care associates). The purposes of this paper are to (1) assess the extent to which the distribution of physical activity and BMI differ among workers older than 45 years, compared to those younger than 45; (2) investigate the relationships of workplace characteristics to physical activity and BMI; and (3) test the interaction between select workplace characteristics and cancer risk behaviors to determine if associations vary by age.
      Finally, as both physical inactivity and obesity are associated with long-term health outcomes other than cancer, such as cardiovascular disease and diabetes,
      • Knowler W.C.
      • Barrett-Connor E.
      • Fowler S.E.
      • et al.
      Reduction of the incidence of type 2 diabetes with lifestyle intervention or metformin.
      Writing Group of the PREMIER Collaborative Research Group
      Effects of comprehensive lifestyle modification on blood pressure control: main results of the PREMIER Clinical Trial.
      it is anticipated that these study results will be relevant not only to cancer researchers but also to those interested in other chronic conditions.

      Methods

      The Be Well Work Well (BWWW) study is a research study conducted by the Harvard School of Public Health Center for Work, Health, and Well-Being. The data presented here are drawn from the first BWWW survey, a cross-sectional survey of patient care workers administered in two large academic teaching hospitals in the Boston area in late 2009.
      • Sorensen G.
      • Stoddard A.M.
      • Stoffel S.
      • et al.
      The role of the work context in multiple wellness outcomes for hospital patient care workers.
      • Buxton O.M.
      • Hopcia K.
      • Sembajwe G.
      • et al.
      Relationship of sleep deficiency to perceived pain and physical disability in hospital patient care workers.
      This project was approved by the applicable IRB for protection of human subjects.

      Sample

      Eligible staff were identified through the hospitals’ human resources database. The sampling frame included all benefits-eligible staff employed in Patient Care Services who provided direct patient care between May 30 and August 22, 2009. Additional eligibility criteria included the following: being employed between October 1, 2008, and September 30, 2009; working on a patient care unit (e.g., adult medical/surgical, adult ICU, pediatric/neonatal ICU); and working at least 20 hours per week. Staff who were assigned to the “float” unit, considered allied healthcare professionals (physical therapy, occupational therapy); worked in environmental services; worked on physical medicine units; had an absence of more than 12 weeks; worked per diem; or worked as a traveling or contract nurse were excluded. To obtain the sample, 2000 workers were randomly selected from 7019 eligible workers and invited, via e-mail, to participate in an online survey as previously described.
      • Sorensen G.
      • Stoddard A.M.
      • Stoffel S.
      • et al.
      The role of the work context in multiple wellness outcomes for hospital patient care workers.
      A total of 1572 workers completed the survey. The response rate was 79%.

      Measures

      Outcomes

      Physical activity was measured using an adapted version of the CDC Behavioral Risk Factor and Surveillance System Physical Activity measure.
      CDC
      Behavioral Risk Factor Surveillance System survey questionnaire.
      Respondents were asked about their participation in both vigorous and moderate physical activities outside of work.
      • Sorensen G.
      • Stoddard A.M.
      • Stoffel S.
      • et al.
      The role of the work context in multiple wellness outcomes for hospital patient care workers.
      Adequate physical activity was defined as reporting at least 30 minutes of moderate or vigorous activity on at least 5 days a week or at least 20 minutes of vigorous activity on at least 3 days a week.

      USDHHS. Healthy People 2010. 2008 physical activity guidelines for adults. www.health.gov/PAGuidelines/factsheetprof.aspx.

      Body mass index was assessed by self-reported height and weight, and was computed by dividing weight (in kilograms) by height squared (in meters). Participants were classified as normal weight (<25); overweight (25–<30); or obese (≥30).

      Independent variables

      Age was assessed using employee record data and reflects respondents’ age on January 1, 2009. Sleep Deficiency was operationalized using questions adapted from the Pittsburgh Sleep Quality Index.
      • Buysse D.
      • Reynolds C.
      • Monk T.
      • Berman S.
      • Kupfer D.
      The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research.
      Sleep duration was measured as the number of hours slept per night. Insomnia symptoms were assessed by asking about difficulty falling asleep and waking during the night. Insufficient sleep was assessed by asking about feeling rested on waking.
      • Buxton O.M.
      • Quintiliani L.M.
      • Yang M.H.
      • et al.
      Association of sleep adequacy with more healthful food choices and positive workplace experiences among motor freight workers.
      Sleep deficiency (yes versus no) was defined as the presence of short sleep duration (<6 hours/day); insomnia symptoms; or insufficient sleep.

      Sociodemographic control variables

      Gender, race/ethnicity, and education were assessed using standard measures.

      Workplace characteristics

      Occupation included the following categories: staff nurse, patient care associate, and other (e.g., operations coordinator). Work shift was categorized as “regular days,” “regular evenings,” and “other.” Hours worked was determined by reports of hours worked in a typical week. Job tenure, or years employed by current employer, was extracted from the employee database.
      Psychological job demands were assessed using a five-item version of the Job Content Questionnaire.
      • Karasek R.
      • Theorell T.
      Healthy work: stress, productivity, and the reconstruction of working life.
      Responses were summed, with a higher score representing greater job demands (response categories: strongly disagree=1 to strongly agree=5, scale range=12–48). The decision latitude scale was made up of three items that assessed decision authority and five items that assessed skill discretion. The decision latitude scale is a weighted sum of decision authority and skill discretion; a higher score reflects more decision latitude (response categories: strongly disagree=1 to strongly agree=5, scale range=24–96).
      To assess job flexibility, three questions assessing how often respondents changed the shift they work to accommodate family or personal needs were combined (to a more desirable shift, to a less desirable shift, and shift length; response categories: never=1 to always=5). A higher score reflects a more flexible job situation (range=3–15).
      Both the supervisor and coworker support scales were adapted from the Job Content Questionnaire.
      • Karasek R.
      • Theorell T.
      Healthy work: stress, productivity, and the reconstruction of working life.
      Supervisor support reflects supervisor help, support, and appreciation of work achievements. Coworker support indicates the extent to which coworkers are helpful and supportive. Responses were summed (scale range for supervisor support=3–15 and for coworker support=2–10; response categories: strongly agree=5 to strongly disagree=1). For both scales, a higher score reflects greater support.
      Harassment at work was assessed by asking how often in the previous 12 months someone at work yelled/screamed at, made hostile/offensive gestures to, swore at, talked down to, or treated the respondent poorly.
      • Krieger N.
      • Waterman P.D.
      • Hartman C.
      • et al.
      Social hazards on the job: workplace abuse, sexual harassment, and racial discrimination—a study of black, Latino, and White low-income women and men workers in the U.S.
      Respondents were coded as experiencing workplace harassment if he/she responded “more than once” to any question (yes versus no).
      People-oriented culture was measuring using four items assessing cooperative working relationships, open and trusting communication, and staff involvement in decision making (response categories: strongly disagree=1 to strongly agree=5).
      • Amick B.C.
      • Habeck R.V.
      • Hunt A.
      • et al.
      Measuring the impact of organizational behaviors on work disability prevention and management.
      Responses were averaged. A higher score reflects a more positive culture (range=1–5).
      To assess the degree of understaffing, respondents were asked how often there were enough nurses and patient care workers, there was sufficient administrative support, and there was enough time to discuss patient care problems (response categories: always=1 to never=5). Responses were summed. A higher score reflects a greater degree of understaffing (range=4–20).
      Ergonomic practices was assessed with questions regarding the extent to which the respondent’s work was designed to reduce lifting, pushing, pulling, bending, stooping and reaching, and to what extent ergonomic factors are considered in work design (response categories: strongly disagree=1 to strongly agree=5). Responses were averaged. A higher score reflects greater consideration of ergonomics in work design (range=1–5).
      Positive workplace safety practices was measured using items that inquired about unsafe working conditions, housekeeping, ramifications for breaking safety rules, supervisory response to unsafe behaviors, and supervisor safety training (response categories: strongly disagree=1 to strongly agree=5).
      • Amick B.C.
      • Habeck R.V.
      • Hunt A.
      • et al.
      Measuring the impact of organizational behaviors on work disability prevention and management.
      Responses were averaged. A higher score reflects better working conditions (range=1–5).

      Statistical Analyses

      Bivariate relationships were assessed between each predictor and each outcome, using chi-square, t test, or ANOVA tests as appropriate. Predictors were included in the multivariate models if p<0.2, using logistic regression for physical activity and multinomial logistic regression for BMI. If p<0.05, the predictor was left in model. Age remained in each model, regardless of significance. Interactions between workplace characteristics variables and age were tested if the workplace characteristic was significantly related to the outcome in the multivariate final model and there was a theoretical justification for doing so. All analyses were conducted using SAS, version 9.3, in January–February 2013.

      Results

      Sample Characteristics

      The sample was mostly female, predominantly white, and educated. There were many older workers in the sample, 28% were age 45–54 and 16% were age ≥55 years (Table 1).
      Table 1Sociodemographic and workplace characteristics among participants of the Be Well Work Well Study (N=1572)
      Characteristicsn (%) or M (SD)
      Adequate physical activity825 (54.3)
      BMI
       Normal (<25)698 (49.2)
       Overweight (25–<30)418 (29.4)
       Obese (≥30)304 (21.4)
      Age (years)
       21–34513 (32.6)
       35–44367 (23.3)
       45–54436 (27.7)
       ≥55256 (16.3)
      Presence of sleep deficiency963 (63.4)
      JOB CHARACTERISTICS
      Occupation
       Staff nurse1103 (70.5)
       Patient care associate127 (8.1)
       Other occupation335 (21.4)
      Shift
       Regular days469 (29.9)
       Regular evenings158 (10.1)
       Other shifts939 (60.0)
      Hours worked per week
       Part time (<34)535 (34.2)
       Full time (35–44)961 (61.4)
       Overtime (>44)70 (4.5)
      Tenure with current employer (years)
       <5555 (35.3)
       5–9407 (25.9)
       ≥10610 (38.8)
      Psychological demands (M [SD])35.9 (5.17)
      Decision latitude (M [SD])71.7 (9.67)
      WORKPLACE CHARACTERISTICS
      Harassment at work (M [SD])913 (58.1)
      Job flexibility (M [SD])6.1 (2.82)
      Supervisor support (M [SD])10.6 (2.98)
      Co-worker support (M [SD])8.0 (1.49)
      People-oriented culture (M [SD])3.6 (0.75)
      Understaffing (M [SD])9.1 (2.81)
      Ergonomic practices (M [SD])3.1 (0.83)
      Positive safety practices (M [SD])3.7 (0.66)
      SOCIODEMOGRAPHICS
      Female gender1369 (90.5)
      Race/ethnicity
       White1185 (79.1)
       Hispanic65 (4.3)
       Black159 (10.6)
       Mixed race/other89 (5.9)
      Education
       Grade 12/GED or less78 (5.2)
       1–3 years of college or technical school360 (23.9)
       4-year college degree (graduate)803 (53.4)
       Any graduate school264 (17.5)
      GED, General Educational Development

      Bivariate Analyses

      Inadequate physical activity was associated with sleep deficiency (p<0.05); working as a patient care associate (p<0.01); less decision latitude (p<0.01); less job flexibility (p<0.01); lower coworker support (p<0.01); and reporting lower levels of people-oriented culture (p<0.01; Table 2).
      Table 2Bivariate associations with physical activity among participants of the Be Well Work Well Study, n (%) unless otherwise noted
      Adequate physical activity (n=825)Inadequate physical activity (n=693)p-value
      p-values for continuous variables were based on t-tests; p-values for categorical variables were based on χ2.
      Age (years)0.051
       21–34291 (58.3)208 (41.7)
       35–44198 (55.5)159 (44.5)
       45–54207 (49.3)213 (50.7)
       ≥55129 (53.3)113 (46.7)
      Sleep deficiency0.024
       No320 (58.5)227 (41.5)
       Yes499 (52.5)452 (47.5)
      JOB CHARACTERISTICS
      Occupation<0.001
       Staff nurse611 (57.3)455 (42.7)
       Patient care associate46 (38.3)74 (61.7)
       Other occupation166 (50.9)160 (49.1)
      Shift0.094
       Regular days243 (53.8)209 (46.2)
       Regular evenings72 (46.8)82 (53.2)
       Others508 (56.1)398 (43.9)
      Hours worked per week0.094
       Part time (<34)303 (58.0)219 (42.0)
       Full time (35–44)487 (52.8)436 (47.2)
       Overtime (>44)33 (48.5)35 (51.5)
      Tenure with current employer (years)0.562
       <5301 (56.1)236 (43.9)
       5–9212 (52.6)191 (47.4)
       ≥10312 (54.0)266 (46.0)
      Psychological demands (M [SD])36.1 (5.18)35.8 (5.15)0.267
      Decision latitude (M [SD])72.7 (9.15)70.5 (10.00)<0.001
      WORKPLACE CHARACTERISTICS
      Harassment at work0.271
       No354 (56.0)278 (44.0)
       Yes471 (53.2)415 (46.8)
      Job flexibility (M [SD])6.4 (2.79)5.8 (2.81)0.001
      Supervisor support (M [SD])10.8 (3.03)10.5 (2.90)0.059
      Coworker support (M [SD])8.1 (1.43)7.8 (1.53)<0.001
      People-oriented culture (M [SD])3.6 (0.72)3.5 (0.78)0.005
      Understaffing (M [SD])9.0 (2.79)9.2 (2.82)0.236
      Ergonomic practices (M [SD])3.1 (0.84)3.2 (0.82)0.126
      Positive safety practices (M [SD])3.8 (0.65)3.7 (0.68)0.418
      SOCIODEMOGRAPHICS
      Gender0.362
       Male83 (58.5)59 (41.5)
       Female734 (54.4)614 (45.5)
      Race/ethnicity<0.001
       White697 (59.4)477 (40.6)
       Hispanic25 (38.5)40 (61.5)
       Black56 (36.4)98 (63.6)
       Mixed race/other34 (39.5)52 (60.5)
      Education0.001
       Grade 12/GED or less27 (35.5)49 (64.5)
       1–3 years of college or technical school181 (51.9)168 (48.1)
       4-year college degree (graduate)466 (58.5)331 (41.5)
       Any graduate school140 (53.4)122 (46.6)
      GED, General Educational Development
      a p-values for continuous variables were based on t-tests; p-values for categorical variables were based on χ2.
      Sleep was related to BMI, with a higher risk of sleep deficiency for those who were obese (p<0.01). Obesity was associated with working more hours per week (p<0.02); having a longer tenure at one’s job (p<0.01); experiencing harassment (p=0.04); less job flexibility (p=0.05); and reporting a less people-oriented culture (p<0.01; Table 3).
      Table 3Bivariate associations with BMI among participants of the Be Well Work Well Study, n (%) unless otherwise noted
      Normal (n=698)Overweight (n=418)Obese (n=304)p-value
      p-values for continuous variables were based on ANOVA tests; p-values for categorical variables were based on χ2.
      Age (years)<0.001
       21–34303 (64.3)105 (22.3)63 (13.4)
       35–44153 (46.2)95 (28.7)83 (25.1)
       45–54152 (38.8)142 (36.2)98 (25.0)
       ≥5590 (39.8)76 (33.6)60 (26.5)
      Sleep deficiency0.002
       No272 (52.5)161 (31.1)85 (16.4)
       Yes423 (47.1)257 (28.6)219 (24.4)
      JOB CHARACTERISTICS
      Occupation<0.001
       Staff nurse542 (53.3)284 (27.9)191 (18.8)
       Patient care associate31 (32.3)31 (32.3)34 (35.4)
       Other occupation122 (40.3)103 (34.0)78 (25.7)
      Shift0.880
       Regular days198 (47.1)129 (30.7)93 (22.1)
       Regular evenings65 (47.8)42 (30.9)29 (21.3)
       Others431 (50.2)247 (28.8)181 (21.1)
      Hours worked per week0.016
       Part time (<34)254 (51.0)152 (30.5)92 (18.5)
       Full time (35–44)422 (49.3)243 (28.4)191 (22.3)
       Overtime (>44)19 (30.6)23 (37.1)20 (32.3)
      Tenure with current employer (years)<0.001
       <5300 (59.9)122 (24.4)79 (15.8)
       5–9174 (46.2)114 (30.2)89 (23.6)
       ≥10224 (41.3)182 (33.6)136 (25.1)
      Psychological demands (M [SD])36.1 (5.21)36.0 (5.33)35.7 (4.97)0.591
      Decision latitude (M [SD])71.9 (9.47)71.7 (9.81)71.9 (9.55)0.951
      WORKPLACE CHARACTERISTICS
      Harassment at work0.041
       No303 (52.2)172 (29.6)106 (18.2)
       Yes395 (47.1)246 (29.3)198 (23.6)
      Job flexibility (M [SD])6.4 (2.80)6.1 (2.87)5.8 (2.85)0.052
      Supervisor support (M [SD])10.7 (2.91)10.5 (3.02)10.6 (3.02)0.661
      Coworker support (M [SD])8.1 (1.44)8.0 (1.47)7.9 (1.51)0.260
      People-oriented culture (M [SD])3.7 (0.71)3.6 (0.77)3.5 (0.78)0.006
      Understaffing (M [SD])9.1 (2.77)9.1 (2.70)9.2 (2.90)0.836
      Ergonomic practices (M [SD])3.1 (0.84)3.1 (0.83)3.2 (0.84)0.875
      Positive safety practices (M [SD])3.8 (0.62)3.7 (0.71)3.7 (0.69)0.469
      SOCIODEMOGRAPHICS
      Gender<0.001
       Male46 (33.3)60 (43.5)32 (23.2)
       Female650 (50.8)358 (28.0)272 (21.3)
      Race/ethnicity<0.001
       White590 (51.8)326 (28.6)223 (19.6)
       Hispanic18 (30.5)21 (35.6)20 (33.9)
       Black40 (29.4)45 (33.1)51 (37.5)
       Mixed race/other47 (60.3)23 (29.5)8 (10.3)
      Education<0.001
       Grade 12/GED or less22 (34.5)17 (26.6)25 (39.1)
       1–3 years of college or technical school129 (39.2)118 (35.9)82 (24.9)
       4-year college degree (graduate)416 (54.2)205 (26.2)146 (19.0)
       Any graduate school128 (50.6)78 (30.8)47 (18.6)
      GED, General Educational Development
      a p-values for continuous variables were based on ANOVA tests; p-values for categorical variables were based on χ2.

      Multivariate Analyses

      After controlling for sociodemographic characteristics and workplace characteristics, age was no longer significantly associated with physical activity (p=0.17). However, a greater amount of decision latitude was associated with a greater likelihood of achieving adequate physical activity (OR=1.02, 95% CI=1.01, 1.03). Similarly, greater job flexibility was associated with increased likelihood of achieving adequate physical activity (OR=1.05, 95% CI=1.01, 1.10; Table 4).
      Table 4Predictors of adequate physical activity among participants of the Be Well Work Well Study
      Model is adjusted for the effect of race/ethnicity
      OR (95% CI)p-value
      p-values found using Wald χ2 test for Type 3 effects
      Age (years)0.174
       21–34 (ref)1.00
       35–440.81 (0.59, 1.11)
       45–540.71 (0.52, 0.97)
       ≥550.89 (0.59, 1.34)
      Decision latitude1.02 (1.01, 1.03)0.001
      Job flexibility1.05 (1.01, 1.10)0.029
      a Model is adjusted for the effect of race/ethnicity
      b p-values found using Wald χ2 test for Type 3 effects
      Risk of overweight and obesity increased with age (p<0.01). Respondents who reported sleep deficiency had a 1.56 greater odds (95% CI=1.15, 2.12) of obesity compared to those who did not report sleep deficiency. Those who reported harassment at work had a 1.62 greater odds (95% CI=1.20, 2.18) of obesity compared to those who did not report such experiences (Table 5).
      Table 5Predictors of overweight and obese BMI among participants of the Be Well Work Well Study,
      Model is adjusted for the effects of gender and race/ethnicity.
      OR (95% CI)
      BMI (3-category) multivariates (final model)Overweight vs normal BMIObese vs normal BMIp-value
      p-values found using Wald χ2 test for Type 3 effects
      Age (years)<0.001
       21–34 (ref)11
       35–441.871.32, 2.642.651.78, 3.94
       45–542.972.14, 4.133.722.52, 5.48
       ≥552.831.92, 4.173.792.43, 5.90
      Sleep deficiency0.007
       No (ref)11
       Yes0.970.75, 1.251.561.15, 2.12
      Experienced abuse/harassment more than once0.005
       No (ref)11
       Yes1.250.97, 1.621.621.20, 2.18
      a Model is adjusted for the effects of gender and race/ethnicity.
      b p-values found using Wald χ2 test for Type 3 effects
      To determine if the relationship of workplace characteristics to either BMI or physical activity differed by age group, interaction terms were added, one at a time, to the final model. Interaction terms that were assessed included, for BMI, harassment at work by age and sleep deficiency by age; and for physical activity, job latitude by age, and staff flexibility by age. None of the interaction terms reached significance when included in the multivariate models.

      Discussion

      This paper examined the relationships of two cancer-related risks—overweight/obesity and inadequate physical activity—with age, and expressly explored the role of workplace characteristics in these relationships. The findings indicated that risk of overweight and obesity increased with age even when controlling for workplace characteristics. In addition, sleep deficiency and experiences of workplace harassment remained significantly associated with risk of being obese, even when controlling for age. The association between physical activity and age did not remain significant in multivariate analyses. However, getting enough physical activity was associated with job decision latitude and job flexibility. These findings underscore the persistent impact of the work environment for workers of all ages.

      Body Mass Index and Age

      In this model, the relationship between age and BMI changed somewhat after including other predictors in the model, such that the age gradient (increasing age predicting increased BMI) leveled off somewhat after age 45—for both overweight and obese individuals. It is possible that a “healthy worker survivor effect” underlies this finding, as overweight workers with weight-related comorbidities may leave the workforce earlier than normal weight workers. Previous research
      • Jones M.K.
      • Latreille P.L.
      • Sloane P.J.
      • Staneva A.V.
      Work-related health risks in Europe: are older workers more vulnerable?.
      has demonstrated the importance of the healthy worker effect when examining the relationship between age and health among workers. This may have considerable effect in patient care, which is physically demanding and requires long hours standing.
      Further, these results indicate that experiencing harassment at work was the most important workplace characteristic associated with obesity. Previous research has linked workplace harassment to other health issues, including psychological distress, elevated blood pressure, and likelihood of injury.
      • Krieger N.
      • Kaddour A.
      • Koenen K.
      • et al.
      Occupational, social, and relationship hazards and psychological distress among low-income workers: implications of the “inverse hazard law.”.
      • Krieger N.
      • Chen J.T.
      • Waterman P.D.
      • et al.
      The inverse hazard law: blood pressure, sexual harassment, racial discrimination, workplace abuse and occupational exposures in U.S. low-income black, white and Latino workers.
      • Sabbath E.
      • Hurtado D.
      • Okechukwu C.A.
      • et al.
      Occupational injury among hospital patient-care workers: what is the association with non-physical workplace violence?.
      However, though it is possible that workplace harassment may lead to deleterious health outcomes, such as being overweight, it is also possible that those who are overweight/obese are more likely to experience workplace harassment, given the stigmatization of obesity. In addition, there is evidence that suggests that the healthcare workplace can be a psychologically and emotionally hostile environment.
      • Croft R.K.
      • Cash P.A.
      Deconstructing contributing factors to bullying and lateral violence in nursing using a postcolonial feminist lens.
      In addition to harassment, healthcare workers may experience bullying, intimidation, and assault from their coworkers,
      • Krieger N.
      • Chen J.T.
      • Waterman P.D.
      • et al.
      The inverse hazard law: blood pressure, sexual harassment, racial discrimination, workplace abuse and occupational exposures in U.S. low-income black, white and Latino workers.
      • Hutchinson M.
      • Vickers M.H.
      • Wilkes L.
      • Jackson D.
      A typology of bullying behaviours: the experiences of Australian nurses.
      and, like victims of workplace harassment, those who have been the targets of this type of behavior may experience health consequences, including severe psychological trauma, depression, anxiety, post-traumatic stress disorder, and physical illness.
      • Hutchinson M.
      • Wilkes L.
      • Jackson D.
      • Vickers M.H.
      Integrating individual, work group and organizational factors: testing a multidimensional model of bullying in the nursing workplace.
      In this occupational setting, where rotating shifts and long shifts are common, sleep deficiency may be seen, at least in part, as an occupational risk. These results indicate that, even after controlling for age, sleep deficiency was associated with an increased risk of obesity. There is a well-established link between sleep and BMI.
      • Cappuccio F.P.
      • Taggart F.M.
      • Kandala N.B.
      • et al.
      Meta-analysis of short sleep duration and obesity in children and adults.
      Some, but not all, prospective studies indicate that short sleep causes more weight gain over time, as sleep loss affects the hormones that affect appetite regulation.
      • Knutson K.L.
      • Van Cauter E.
      Associations between sleep loss and increased risk of obesity and diabetes.
      • Patel S.R.
      • Hu F.B.
      Short sleep duration and weight gain: a systematic review.
      In addition, there is evidence that the relationship between sleep and increased BMI grows weaker with increasing age.
      • Patel S.R.
      • Hu F.B.
      Short sleep duration and weight gain: a systematic review.
      However, the data did not support this; rather, the relationship between sleep and BMI was similar across age groups. This finding may be due to the somewhat restricted age distribution in the sample.

      Physical Activity and Age

      It was surprising that after adjusting for race/ethnicity and workplace characteristics, physical activity was no longer associated with age. It is well accepted that as individuals age, they become less physically active. However, these results indicate that physical activity was more strongly related to workplace characteristics. This relationship may exist because those with greater latitude in their jobs and more flexibility in their work schedules may find it easier to engage in regular physical activity, regardless of the age. These findings underscore the importance of attending to these core characteristics of the work environment across the age spectrum of workers.

      Limitations

      It is important to note that this study employed a cross-sectional study design, and determining temporal sequence or causality is not possible. In addition, the measures relied predominately on self-report, and accordingly were subject to recall and social desirability biases. There may be unknown confounders that were not measured or considered in this analysis; however, many known or suspected confounders of the relationship between age and health behavior were included. These findings are based on a study of patient care workers in two large teaching hospitals, most of whom were women; the characteristics of the sample place limits the generalizability of the findings. Finally, although BMI is related to cancer risk, a recent meta-analysis found little evidence for linkages between BMI and all-cause mortality.
      • Flegal K.M.
      • Kit B.K.
      • Orpana H.
      • Graubard B.I.
      Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis.
      Nonetheless, this study includes many strengths, including a high response rate (79%) and the use of multiple, validated indicators of work experiences.

      Implications

      Targeting work environment in physical activity promotion, for all workers

      One important finding of this study is that the work environment has comparable impact on worker physical activity, regardless of age. Therefore, workplace-based physical activity interventions should target workers of all ages. In addition, these findings indicate that flexibility in scheduling and latitude in determining job tasks and timing is important for maintaining physical activity. Thus, it may be beneficial to consider these facets of the work environment when designing physical activity interventions.

      Targeting obesity prevention and weight control, among older workers

      Based on the finding that older workers are at higher risk for overweight and obesity, even when holding workplace characteristics constant, it may be advantageous to design and test worksite interventions that emphasize obesity prevention and weight control among older workers. In addition, these findings indicate that obesity risk is associated with sleep deficiency. In a previous study using the same sample of healthcare workers, Buxton and colleagues
      • Buxton O.M.
      • Hopcia K.
      • Sembajwe G.
      • et al.
      Relationship of sleep deficiency to perceived pain and physical disability in hospital patient care workers.
      found that sleep deficiency was associated with both coworker and supervisor support. Thus, interventions that target social support in the workplace may also help improve sleep among workers, which may then affect worker BMI.

      Targeting work environment in obesity prevention, for all workers

      Finally, these results indicated that the work environment was a contributor to obesity risk. Aside from the occupationally and psychologically damaging effects of harassment within the workplace, these results suggest an association with obesity. Thus, when addressing overweight and obesity within the workplace, it may be beneficial to consider workplace harassment. One way of targeting workplace harassment suggested in the nursing literature is to build social support and encourage individuals in the workplace to share responsibility for negative behavior.
      • Hutchinson M.
      Restorative approaches to workplace bullying: educating nurses towards shared responsibility.

      Acknowledgments

      This research was supported by a grant from the National Institute for Occupational Safety and Health ( U19 OH008861 ) for the Harvard School of Public Health Center for Work, Health and Well-being.
      The publication of this supplement was made possible through the CDC and the Association for Prevention Teaching and Research (APTR) Cooperative Agreement No. 1 U360E000005-01. The findings and conclusions in this report are those of the authors and do not necessarily represent the official position of the CDC or the APTR.
      This study would not have been accomplished without the participation of Partners HealthCare System and leadership from Dennis Colling, Sree Chaguturu, and Kurt Westerman. The authors would like to thank Partners Occupational Health Services including Marlene Freeley for her guidance, as well as Elizabeth Taylor, Elizabeth Tucker O’Day, and Terry Orechia. We also thank individuals at each of the hospitals, including Jeanette Ives Erickson and Jacqueline Somerville in Patient Care Services leadership, and Jeff Davis and Julie Celano in Human Resources. Additionally, we wish to thank Charlene Feilteau, Mimi O’Connor, Margaret Shaw, Eddie Tan, and Shari Weingarten for assistance with supporting databases. We also thank Chris Kenwood of NERI for his statistical and programming support, Evan McEwing, Project Director, Lorraine Wallace, Senior Project Director, and Linnea Benson-Whelan for her assistance with the production of this manuscript.
      No financial disclosures were reported by the authors of this paper.
      Conflicts of Interest and Sources of Funding: There are no conflicts of interest to report, but in the interest of full disclosure, Dr. Nelson receives her funding from Harvard School of Public Health (Harvard-Liberty Postdoctoral Fellowship), Dr. Wagner receives his funding from CDC/National Institute for Occupational Safety and Health (NIOSH), Dr. Caban-Martinez from the NIOSH, Grant U19 OH008861; the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) , Grant T32 AR055885 ; and the Clinical Orthopedic and Musculoskeletal Education and Training (COMET) Program at Brigham and Women’s Hospital, Harvard Medical School and Harvard School of Public Health. Dr. Buxton has received two investigator-initiated grants from Sepracor Inc (now Sunovion; ESRC-0004 and ESRC-0977, ClinicalTrials.gov Identifiers NCT00555750, NCT00900159), and two investigator-initiated grants from Cephalon Inc (now Teva; ClinicalTrials.gov Identifier: NCT00895570). OMB received Speaker’s Bureau, CME and non-CME lecture honoraria, and an unrestricted educational grant from Takeda Pharmaceuticals North America. OMB serves as a consultant and expert witness for Dinsmore LLC, consulting fees for serving on the Scientific Advisory Board of Matsutani America, and consulting fees from the Wake Forest University Medical Center (North Carolina). OMB received speaking fees and/or travel support for speaking from American Academy of Craniofacial Pain, National Heart, Lung, Blood Institute, National Institute of Diabetes and Digestive and Kidney Diseases, National Postdoctoral Association, Oklahoma State University, Oregon Health Sciences University, SUNY Downstate Medical Center, American Diabetes Association, and New York University. Mr. Kenwood from NIOSH, Grant: 6U19OH008861, Dr. Sabbath from the John D. and Catherine T. MacArthur Foundation Network on an Aging Society and NIA (Grant 5R01AG040248-02), Dr. Hashimoto is an employee of Partners Healthcare System, Dr. Hopcia is funded by the University of Illinois at Chicago, Dr. Allen is supported by Cooperative Agreement Number U48DP001946 from the CDC and the National Cancer Institute, and Dr. Sorensen by NIOSH (Grant 2U19 OH008861) and the National Cancer Institute (Grant 5 K05 CA108663-06 ).

      References

        • Colditz G.A.
        • Wolin K.Y.
        • Gehlert S.
        Applying what we know to accelerate cancer prevention.
        Sci Transl Med. 2012; 4 (127rv124)
        • Thune I.
        • Furberg A.S.
        Physical activity and cancer risk: dose–response and cancer, all sites and site-specific.
        Med Sci Sports Exerc. 2001; 33: S530-S550
        • Ligibel J.
        Lifestyle factors in cancer survivorship.
        J Clin Oncol. 2012; 30: 3697-3704
        • Pischon T.
        • Nothlings U.
        • Boeing H.
        Obesity and cancer.
        Proc Nutr Soc. 2008; 67: 128-145
        • Kahn L.K.
        • Sobush K.
        • Keener D.
        • et al.
        Recommended community strategies and measurements to prevent obesity in the U.S.
        Morb Mortal Wkly Rep. 2009; 58: 1-30
        • Salinardi T.C.
        • Batra P.
        • Urban L.E.
        • et al.
        Lifestyle intervention reduces body weight and improves cardiometabolic risk factors in worksites.
        Am J Clin Nutr. 2013; 97: 667-676
        • Verweij L.M.
        • Proper K.I.
        • Weel A.N.H.
        • et al.
        The application of an occupational health guideline reduces sedentary behavior and increases fruit intake at work: results from an RCT.
        Occup Environ Med. 2012; 69: 500-507
      1. Bureau of Labor Statistics, U.S. Department of Labor. News release, January 2013. www.bls.gov/bls/newsrels.htm.

        • Taylor S.E.
        • Repetti R.L.
        • Seeman T.
        Health psychology: what is an unhealthy environment and how does it get under the skin?.
        Annu Rev Psychol. 1997; 48: 411-447
        • Sorensen G.
        • Stoddard A.M.
        • Stoffel S.
        • et al.
        The role of the work context in multiple wellness outcomes for hospital patient care workers.
        J Occup Environ Med. 2011; 53: 899-910
        • Sorensen G.
        • Stoddard A.
        • Dubowitz T.
        • Barbeau E.M.
        • Berkman L.F.
        • Peterson K.E.
        The influence of social context on changes in fruit and vegetable consumption: results of the healthy directions studies.
        Am J Public Health. 2007; 97: 1216-1227
        • Caban A.J.
        • Lee D.J.
        • Fleming L.E.
        • Gomez-Marin O.
        • LeBlanc W.
        • Pitman T.
        Obesity in U.S. workers: the National Health Interview Survey, 1986–2002.
        Am J Public Health. 2005; 95: 1614-1622
        • Caban-Martinez A.J.
        • Lee D.J.
        • Fleming L.E.
        • et al.
        Leisure-time physical activity levels of the U.S. workforce.
        Prev Med. 2007; 44: 432-436
        • Loeppke R.R.
        • Schill A.L.
        • Chosewood L.C.
        • et al.
        Advancing workplace health protection and promotion for an aging workforce.
        J Occup Environ Med. 2013; 55: 500-506
        • Silverstein M.
        Meeting the challenges of an aging workforce.
        Am J Ind Med. 2008; 51: 269-280
        • Jones M.K.
        • Latreille P.L.
        • Sloane P.J.
        • Staneva A.V.
        Work-related health risks in Europe: are older workers more vulnerable?.
        Soc Sci Med. 2013; 88: 18-29
        • Knowler W.C.
        • Barrett-Connor E.
        • Fowler S.E.
        • et al.
        Reduction of the incidence of type 2 diabetes with lifestyle intervention or metformin.
        N Engl J Med. 2002; 346: 393-403
        • Writing Group of the PREMIER Collaborative Research Group
        Effects of comprehensive lifestyle modification on blood pressure control: main results of the PREMIER Clinical Trial.
        JAMA. 2003; 289: 2083-2093
        • Buxton O.M.
        • Hopcia K.
        • Sembajwe G.
        • et al.
        Relationship of sleep deficiency to perceived pain and physical disability in hospital patient care workers.
        J Occup Environ Med. 2012; 54: 851-858
        • CDC
        Behavioral Risk Factor Surveillance System survey questionnaire.
        USDHHS, CDC, Atlanta GA2009
      2. USDHHS. Healthy People 2010. 2008 physical activity guidelines for adults. www.health.gov/PAGuidelines/factsheetprof.aspx.

        • Buysse D.
        • Reynolds C.
        • Monk T.
        • Berman S.
        • Kupfer D.
        The Pittsburgh Sleep Quality Index: a new instrument for psychiatric practice and research.
        Psychiatry Res. 1998; 28: 193-213
        • Buxton O.M.
        • Quintiliani L.M.
        • Yang M.H.
        • et al.
        Association of sleep adequacy with more healthful food choices and positive workplace experiences among motor freight workers.
        Am J Public Health. 2009; 99: S636-S643
        • Karasek R.
        • Theorell T.
        Healthy work: stress, productivity, and the reconstruction of working life.
        Basic Books, New York NY1990
        • Krieger N.
        • Waterman P.D.
        • Hartman C.
        • et al.
        Social hazards on the job: workplace abuse, sexual harassment, and racial discrimination—a study of black, Latino, and White low-income women and men workers in the U.S.
        Int J Health Serv. 2006; 36: 51-85
        • Amick B.C.
        • Habeck R.V.
        • Hunt A.
        • et al.
        Measuring the impact of organizational behaviors on work disability prevention and management.
        J Occup Rehabil. 2000; 10: 21-38
        • Krieger N.
        • Kaddour A.
        • Koenen K.
        • et al.
        Occupational, social, and relationship hazards and psychological distress among low-income workers: implications of the “inverse hazard law.”.
        J Epidemiol Community Health. 2010; 65: 260-272
        • Krieger N.
        • Chen J.T.
        • Waterman P.D.
        • et al.
        The inverse hazard law: blood pressure, sexual harassment, racial discrimination, workplace abuse and occupational exposures in U.S. low-income black, white and Latino workers.
        Soc Sci Med. 2008; 67: 1970-1981
        • Sabbath E.
        • Hurtado D.
        • Okechukwu C.A.
        • et al.
        Occupational injury among hospital patient-care workers: what is the association with non-physical workplace violence?.
        Am J Ind Med. 2013 Oct 22; ([Epub ahead of print])
        • Croft R.K.
        • Cash P.A.
        Deconstructing contributing factors to bullying and lateral violence in nursing using a postcolonial feminist lens.
        Contemp Nurse. 2012; 42: 226-242
        • Hutchinson M.
        • Vickers M.H.
        • Wilkes L.
        • Jackson D.
        A typology of bullying behaviours: the experiences of Australian nurses.
        J Clin Nurs. 2010; 19: 2319-2328
        • Hutchinson M.
        • Wilkes L.
        • Jackson D.
        • Vickers M.H.
        Integrating individual, work group and organizational factors: testing a multidimensional model of bullying in the nursing workplace.
        J Nurs Manag. 2010; 18: 173-181
        • Cappuccio F.P.
        • Taggart F.M.
        • Kandala N.B.
        • et al.
        Meta-analysis of short sleep duration and obesity in children and adults.
        Sleep. 2008; 31: 619-626
        • Knutson K.L.
        • Van Cauter E.
        Associations between sleep loss and increased risk of obesity and diabetes.
        Ann N Y Acad Sci. 2008; 1129: 287-304
        • Patel S.R.
        • Hu F.B.
        Short sleep duration and weight gain: a systematic review.
        Obesity (Silver Spring). 2008; 16: 643-653
      3. CDC. U.S. physical activities statistics. www.cdc.gov/nccdphp/dnpa/physical/stats/index.htm.

        • Flegal K.M.
        • Kit B.K.
        • Orpana H.
        • Graubard B.I.
        Association of all-cause mortality with overweight and obesity using standard body mass index categories: a systematic review and meta-analysis.
        JAMA. 2013; 309: 71-82
        • Hutchinson M.
        Restorative approaches to workplace bullying: educating nurses towards shared responsibility.
        Contemp Nurse. 2009; 32: 147-155