Overweight and obesity are major health challenges because of their high prevalence, causal relationship with serious medical complications, and economic impact (1). The risk of developing many diseases, including type 2 diabetes, increases linearly with body mass index (2–6). Obesity prevention strategies are needed that target multiple levels of the ecologic framework across multiple settings, including the worksite. Using the worksite as a venue for health promotion is promising, because most adults spend approximately half of their waking day in their work environment (6). Research suggests that environmental and policy strategies for addressing energy balance (ie, caloric intake and energy expenditure through physical activity [PA]) in the workplace are effective (7–9). Use of worksite programs to improve employee health has been recommended by the American Cancer Society, the Centers for Disease Control and Prevention, and multiple state governments. Occupational settings take advantage of a captive population and may have existing facilities, social support, convenience, and communication mechanisms in place (10).

Targeting work environments for energy balance includes using policies, programs, and organizational practices to influence behavior. Example work environments include onsite facilities such as gymnasiums, lockers, showers, accessible stairways, and healthy vending options. Policies and programs include subsidized external gymnasium memberships; incentives to bicycle, walk, or use public transportation for the commute to and from work; and group services such as onsite yoga and health fairs (11). By facilitating access to inexpensive healthy food, exercise facilities, and a culture accepting of nonsedentary work breaks, worksites can become sites for health promotion via a healthy energy balance (6). Although tools are available for assessing worksite environments and policies in place for PA and healthy eating (HE), no review has documented the content and measurement properties of these tools. Such a review of worksite energy measurement tools could serve as a guide for researchers, practitioners, and worksites in selecting among existing tools and understanding methodologic gaps to guide potential development of new instruments. The purpose of this review was to identify and assess the state of science concerning available measurement instruments related to worksite environment and policy supports for workplace energy balance.

The literature review was completed in May 2014, using PubMed, OVID, MedLine, Web of Science, and the Registry of Measures from the National Collaborative on Childhood Obesity. We also searched sources of gray literature, including Google Scholar and state health departments. Search terms were key words for worksites, energy balance, and measurement: (work OR worksite OR workplace OR employer OR job) AND (physical activity OR physical fitness OR diet OR exercise OR obesity OR active commuting) AND (evaluation OR monitor* OR survey OR questionnaire OR data collection). Titles of applicable results were screened for their relevance to the assessment of worksite environment and policy measurement, tool development, and worksite interventions targeting PA and HE.

The search was restricted to articles published in English from 1991 through 2013. Abstracts were scanned and accepted if related to 1 or more of the following criteria designed to capture the presence or absence of worksite supports and policies associated with employee PA and HE (eg, presence of an onsite gymnasium, incentives to use public transportation to and from work): 1) studies describing measurement properties of a specific instrument, 2) descriptive studies of environmental and policy supports among a sample of employees or worksites, and 3) cross-sectional or intervention studies that used a specified instrument or explicitly stated the items used to systematically assess worksite environment and policies and their potential associations with PA and HE. Full-text articles were scanned when the information from abstracts was insufficient to make a conclusion about inclusion. Abstracts were excluded if they focused solely on the development or implementation or both of worksite health promotion programs and, thus, were not related to measuring current supports and policies. Moreover, abstracts were rejected if they did not emphasize policy or environmental supports in a nonhome-based worksite. Finally, full-text articles and their reference lists were scanned for references that cited the development of a specific worksite tool, survey, or checklist on policies and environmental supports related to PA and HE. The instruments used among articles that met inclusion criteria were abstracted. Each instrument was categorized on the basis of 1 of 4 administration options: employee or self-report, manager report, observational, or multiple modes. Measurement properties, including reliability and validity, were documented.

The final component of the review involved classifying each unique instrument item into an item inventory. Items were first classified by the general health topic they addressed: PA, HE, or both (healthy eating and physical activity [HEPA]). Next, items were classified by the general worksite strategy being assessed, referred to as the primary domain. These strategies are based on the ecological model, the Guide to Community Preventive Services, and research by Kahn et al (12,13) and include promotions and programs (eg, informational media), organizational policies and practices (eg, incentives), internal physical environment (eg, access to healthy food and PA options), internal social environment (eg, role models), and external environment (eg, worksite neighborhood options for HE and PA). Primary domains were further disaggregated into subdomains by using constant comparison to classify the PA (19 subdomains) and HE (19 subdomains) strategies (Table 1). Interrater agreement for classifying the instrument items was 85% among 3 raters.

Seventeen worksite instruments were identified that included items about worksite environment and policies related to PA, HE, or both and met inclusion criteria. The administration modes of the 17 instruments varied (n = 9 self-report; n = 5 manager report, n = 1 observational; and n = 2 using multiple modes) as did the total number of HE and PA items per instrument (range, 10–226) (Table 2). More items were related to PA than to HE. Nine instruments included both PA and HE items, 7 instruments had only PA items, and only 1 included solely HE items related to worksite environment and policy supports. Of the 17 instruments, 14 reported reliability, of which 8 reported generally high interrater results (Table 2). Five instruments reported various validity measures including content, face, predictive, and construct validity results. Health promotion experts provided substantial guidance in development of the instruments, and significant correlations were found for workplace environmental sections within the instruments. The item inventory indicated that the most common health topic was PA (PA and HEPA) (64% of all items [n = 669]). HE (HE and HEPA) consisted of 369 items, or 36%.

As a venue for delivering HE and PA efforts, worksites provide a channel for reaching the large segment of the population that is employed (147 million as of November 2014, according to the US Bureau of Labor Statistics) (6,10). Moreover, measuring environmental and policy supports for PA and HE in the workplace is an important component in assessing and addressing the factors related to overweight and obesity (14). This review of worksite measures identified various data collection instruments and highlights several matters that require further consideration and attention for future research.

The results of the item inventory highlight both extensive and deficient domain coverage for both PA- and HE-related items. Overall, the primary domains of promotion and programs, organizational policies and practices, and internal physical environment had the greatest coverage among HE and PA items. The primary domain of internal social environment had few items for either HE or PA. We also found several administration modes used, most instruments being self-report. Only 1 instrument was observational (WWAT), although several used multiple methods. With 14 of the 17 instruments relying on either employee or manager self-report, the state of worksite PA and HE measurement is susceptible to respondent and social desirability bias. Regarding measurement properties, most instruments (14 of 17) reported high reliability results, mostly interrater measures. Validity was assessed for 5 instruments, with emphasis on content validation.

There was variety in the content gaps of the measures reviewed. Overall, there were few documented measures about HE in and around the workplace. Most HE measures focused on onsite cafeteria and vending options but neglected external environments (eg, healthy options within a 10-minute walk), organizational policies (eg, healthy snacks at meetings and events), and the social environment. The promotion and programs domain contains 8 measures with items related to informational media and 7 with classes or education (both subdomains); however, only 2 of 10 instruments included any items on assessments, testing, evaluation, and HE. Provided that a full-time employee spends at least 8 hours per day at the worksite — therefore, at least 1 meal is consumed at or near work during most working days — the gaps in HE measures is an important finding that deserves further attention. Exploring the diverse aspects of food environments near workplaces, rather than solely assessing onsite cafeteria and vending options, would be beneficial.

Of the 5 domains, internal social environment was included in the fewest HE- and PA-related instruments. Social environments, including role models, champions, and support, are highly associated with PA and obesity (15,16). Among the subdomains, specialized instruments (ie, Office Environment and Sitting Scale [20], Kaczynski et al [22], and the WWAT [30]) had minimal, if any, coverage. Also, despite including more than 100 unique items, CHEW had minimal coverage for the HE subdomains (only 9 of 19 subdomains covered) (Appendix).

Performing this review did have challenges and limitations. Forcing instrument items into domains and especially subdomains presented some difficulties in operationalizing the specific items. Items could also fit into more than 1 subdomain. The process of developing the subdomains was iterative; new items forced ever greater specificity in the naming and operationalization of the 38 subdomains. However, the specificity of selected subdomains — such as walkability, which can include land use mix, aesthetics, and sidewalks, compared with stairway access, which only refers to the presence of stairs — still varies greatly. We were systematic and prescriptive in our literature search for worksite measures, but this may not be an exhaustive list of worksite instruments, especially those present in the gray literature. Finally, Carnethon and colleagues (17) suggest that efforts moving forward must not only focus on PA but also reduce sedentary behaviors at worksites, and this can be accomplished via policies and designs. Future worksite measurements must do a better job of including sedentary behaviors in their instruments.

This review provides a concise guide for employers to existing worksite measures on PA and HE, both for selecting appropriate assessment instruments for the worksite and as a means to introduce new policies and programs to support healthy workers. For example, employers can administer health risk appraisals in combination with organizational health promotion checklists that have been developed. This approach would provide information to the employee and employer where there may be overlap or gaps between worksite supports and health risks and benefits. Social and physical environments in and around the workplace should be designed to be conducive to recommended healthy behaviors (18). In addition, optimal environmental modifications should promote healthy behaviors while simultaneously minimizing the physical, organizational, and occupational risk in the work environment.