Employment by Occupation and Industry Research Guide

Introduction

Work is a significant social determinant of health (SDOH). It is estimated that, over a lifetime, about one-fifth of waking hours are spent working.¹ While work can be an essential source of income,² health insurance,³ and social engagement,⁴ it can also harm health. Workers face risks of workplace injuries,⁵ fatalities,⁶ exposure to toxic chemicals,⁷ and work environments that may have negative psychological impacts.⁸ The effect of work on health is evident in the frontline role of workplaces in many of the most significant health challenges of recent years. A substantial component of the risk of COVID-19 was workplace exposure.⁹ Outdoor workers face some of the most acute impacts of climate change.¹⁰ The workplace's impact on health outcomes extends beyond workers themselves. Household members can be exposed to risks brought home by others.¹¹ The availability of paid sick leave¹² and parental leave¹³ substantially affects the time and support family members can provide to one another. Work also intersects with many SDOH, including educational attainment and income.¹⁴

For these reasons, work-related factors significantly contribute to community health. Different studies have examined how differences in community employment characteristics have driven variations in community health. These studies have examined a variety of health outcomes, including mortality¹⁵ , life expectancy¹⁶ , coronary heart disease¹⁷ , drug overdoses¹⁸ , COVID-19¹⁹,²⁰ and child abuse/neglect²¹ . Studies have also examined employment-related health predictors, including: the built environment²² , automation²³ , and unionization.²⁴,²⁵

Understanding geographic differences in employment across occupations and industries can inform public health efforts. By analyzing workers' characteristics, work-related risk factors can be identified for public health actions. Knowing where workers in a community are usually employed allows for targeted education and messaging campaigns to those industries. 

This guide provides information on obtaining community-level employment data by industry and occupation. Table 1 presents some examples of data sources that have this type of data. This guide will focus on using the American Community Survey, which has information about employment at the state, county, city/town, ZIP code, and census tract levels. It highlights theoretical frameworks for describing how work impacts health, the different industry and occupation classification systems (Census Industry and Occupation, Standard Occupational Classification, North American Industry Classification System), and the broad and detailed levels at which industry and occupation can be classified. The guide explains how raw occupation and industry data can be categorized using these systems for researchers gathering primary employment data. It also describes how occupation and industry information can be used for health studies, focusing on methods for matching occupation and industry data with job exposure matrices to estimate community-level job exposures to physical, chemical, and psychosocial hazards.

Table 1. Examples of publicly available data sources with information about occupation, industry, and/or other work-related variables and geography

Data Source	Description	Link
American Community Survey	Data about occupation and industry and other work-related variables by state, county, city/town, ZIP code, and census tract.	About: https://www.census.gov/programs-surveys/acs.html To access data: https://data.census.gov/
Occupational Employment and Wage Statistics	Occupation and industry employment data at the national, state and Metropolitan or Non- Metropolitan Area.	https://www.bls.gov/oes/
Current Employment Statistics	Non-farm employment data about industries at the national, state, and Metro Area level. Includes information about	https://www.bls.gov/ces/
Local Area Unemployment Statistics	Information about employment and unemployment and other labor force data by census regions/divisions, states, counties, metropolitan areas, and some cities. Does not present by occupation and industry.	https://www.bls.gov/lau/
National Health Interview Survey	Restricted use data has information about state of residents for respondents. Some survey years have occupation and industry data. Other occupational related variables are asked in certain years.	https://www.cdc.gov/nchs/nhis/index.html
National Vital Statistics System Mortality Data	National Vital Statistics System Mortality Data	https://www.cdc.gov/nchs/nvss/deaths.htm
Longitudinal Employer-Household Dynamics	Includes various employment related indicators including job mobility, and veteran employment. Information is available at the national, state, metropolitan/micropolitan areas, and county levels.	https://lehd.ces.census.gov/data/

Theoretical Frameworks for the Impact of Work on Health

Work can influence health in a variety of ways, which is why it is valuable to understand the work-related characteristics of different geographic areas. In particular, work-related exposures can directly impact health through physical, chemical, and psychological exposures.²⁶ Occupation is also a measure of socioeconomic status. Occupations with higher prestige can be associated with better health outcomes.²⁷ Different scales have been used to assign levels of prestige and status to job titles, such as the National Statistics Socio-economic Classification used in the United Kingdom. Work sits at the crossroads of other measures of socioeconomic status. ²⁸ Educational attainment is an important determinant of the types of jobs people hold, and jobs are an important source of resources, including income, insurance, and benefits such as paid leave and retirement support.²⁹

Some specific frameworks that have been used to examine the health impact of work on health have included the Demand-Control model, which conceives of the health of workers being impacted by the demands of a job (how much work a worker is performing) and control/decision latitude (the amount of control the worker has over how the job is done).³⁰ Jobs with high strain (low control and high demands) have been found to be associated with different health outcomes, including depression³¹ and heart disease.³² The model has been supplemented by the inclusion of job support as a potential protective factor.³³ Other frameworks applied to understanding the role of work in health include the Job Demand-Resources model, which suggests that occupational stress is component of the demands of the job and the resources available to complete the job, ³⁴ Effort-Reward Model, which holds that the health impacts of work are related to the imbalance between the perceived efforts invested into the work and the perceived rewards ³⁵ , and Organizational Justice, which conceives of how workers health and well-being is impacted by fairness with respect to pay, policies and other work-related components.³⁶ Increasingly, the impact of other components of work on health has been examined, including job precarity—insecurity in employment itself or the type of employment ³⁷ , changes in the nature of employment, especially increased employment in alternative work or contingent work arrangements such as contract-based work, temporary jobs, and gig jobs ³⁸ , and the role of artificial intelligence in the workplace. ³⁹ With climate change, work-related heat exposure is a growing concern among workers. The Massachusetts Department of Public Health has developed a matrix for identifying occupations more likely to involve outdoor work, which can be matched with employment data to identify geographic areas where workers are employed in occupations with a higher likelihood of working outdoors.⁴⁰

Occupation and Industry Classification

Occupation and industry are two distinct concepts related to work and are also relevant to studying the impact of work on health. In simple terms, occupation refers to what a person doesin their work, while industry refers to the type of firm they work for. ⁴¹ For example, someone’s occupation can be a nurse (a Healthcare Practitioner), but they could work in different industries, such as a hospital (part of the Healthcare and Social Assistance Industry) or a school (part of the Educational Services industry).

Many jobs exist across different industries. For example, janitors, who are part of the Building and Grounds Cleaning and Maintenance Occupations, can work across many industries.  Many data sources that include industry and occupation information collect free-text descriptions of occupations and industries, then use them to assign standard industry and occupation codes. For example, in the 2024 National Health Interview Survey (NHIS), the following questions were asked to determine the occupation and industry of employed participants:

• Industry: What kind of business or industry is/was this?
• Occupation: What kind of work are/were you doing? ⁴²

The participants' verbal responses are then documented. These word-for-word responses are classified using standard occupational and industry coding systems.

The NHIS classified occupations according to the Bureau of Labor Statistics’ Standard Occupational Classification (SOC)⁴³ and industry according to the North American Industry Classification System (NAICS).⁴⁴ Other standardized occupation and industry classification systems include the United States Census Industry and Occupation codes⁴⁵ , the International Standard Classification of Occupations (ISCO)⁴⁶ , and the International Standard Industrial Classification.⁴⁷ All of these classification systems follow a hierarchical approach, with ‘broad’ categories followed by more detailed categories that fall within them. For example, the 2018 edition of the BLS SOC has 23 ‘major groups’ and 98 minor groups that follow within the major groups. These minor groups are then broken into 459 broad occupations. Finally, these broad occupations are broken down into 867 detailed occupations. These categories have numeric classifications, with the more detailed categories having more digits.⁴⁸ To give an example, a detailed BLS SOC occupation is Health and Safety Engineers, Except Mining Safety Engineers and Inspectors (17-2111), which falls into a broader category, Industrial Engineers, Including Health and Safety (17-2110), which itself falls under, Engineers (17-2000). The broadest classification in this case is Architecture and Engineering Occupations (17-0000).⁴⁹

It is beyond the scope of this guide to discuss the advantages and disadvantages of these different classification systems. When working with occupation or industry data that has already been classified, it is important to be familiar with the classification system that was used. This includes the system itself and its iterations, as each iteration introduces new occupations/industries and categories. Systems differ in the names of occupations and industries, the level of detail, and their categorization. In some instances, the system will match closely for broader classifications, but differ with more detailed classifications. For example, as stated on the BLS Website, referring to the differences between the SOC and NAICS Systems compared to the Census Occupation and Industry Systems: “The Census classifications use the same basic structure as the SOC and NAICS, but are generally less detailed.”⁵⁰ There are often guides to “crosswalk” one classification system to another. ⁵¹

How to classify occupations and industry

If you are working with raw occupation and industry data that has not been classified, you will need to classify the data into one of these systems for analysis. For example, when collecting survey data, many individuals will not be familiar with the exact classification of their job or industry, so respondents are often asked to enter whatever they feel is the best name for this occupation and/or industry (as with the NHIS discussed above). Similarly, death certificates allow free text entry for information about the decedent’s industry and occupation. If you are working with raw death certificate data, this industry and occupation information would need to be coded to standard categories. The process of coding raw occupation and industry information into standard categories can be time-consuming when done manually, especially with large volumes of data. The time-consuming nature of this process stems from the need to review each record individually and double-check that it is consistent with the classification performed. It also requires detailed familiarity with which occupations and industries fall into which category.

Fortunately, the National Institute for Occupational Safety and Health (NIOSH) has developed an automated coding system to classify occupations, known as the Industry and Occupation Computerized Coding System (NIOCCS). Using this system, raw occupation and industry information can be assigned to standard industry and occupation classifications.⁵² These classifications can then be reviewed and cross-walked to other systems if needed.

Obtaining employment and occupation information from the American Community Survey.

The United States Census consistently releases employment data by occupation and industry based on the American Community Survey (ACS). The ACS data is gathered annually from a representative sample of the U.S. population. Occupation data can be found at https://data.census.gov/ by searching for "occupation,” "industry," or other related terms. Specific data tables, including the ones listed below, are shown in Table 2.  In this example, I will focus on S2405, which uses broad occupation codes, and briefly discuss B24124. These principles apply to all tables containing occupation information.

Table 2. Example of tables available from the census with information about the occupation, industry, and other employment characteristics of communities

Table Number	Table Name	Work-related variable(s)
S2401	Occupation by Sex for the Civilian Employed Population 16 Years and Over	2018 Standard Occupational Classification (broad)
S2405	Industry by Occupation for the Civilian Employed Population 16 Years and Over	2022 North American Industry Classification System (broad), 2018 Standard Occupational Classification (very broad)
B24124	Detailed Occupation for the Full-Time, Year-Round Civilian Employed Population 16 Years and Over	2018 Standard Occupational Classification (detailed)
S2406	Occupation by Class of Worker for the Civilian Employed Population 16 Years and Over	2018 Standard Occupational Classification (very broad), Class of Worker (Private, self-employed, government, etc.)
S2411	Occupation by Sex and Median Earnings in the Past 12 Months (in 2024 Inflation-Adjusted Dollars) for the Civilian Employed Population 16 Years and Over	2018 Standard Occupational Classification (broad)
S2404	Industry by Sex for the Full-Time, Year-Round Civilian Employed Population 16 Years and Over	2022 North American Industry Classification System (broad)
S2403	Industry by Sex for the Civilian Employed Population 16 Years and Over	2022 North American Industry Classification System (broad), Class of Worker (Private, self-employed, government, etc.)
S2407	Industry by Class of Worker for the Civilian Employed Population 16 Years and Over	2022 North American Industry Classification System (broad)
S2413	Industry by Sex and Median Earnings in the Past 12 Months (in 2024 Inflation-Adjusted Dollars) for the Civilian Employed Population 16 Years and Over	2022 North American Industry Classification System (broad)
S2301	Employment Status	Labor force participation, employment, unemployment
B08303	Travel Time to Work	Travel time to work (in minutes)
S2408	Class of Worker by Sex for the Civilian Employed Population 16 Years and Over	Class of Worker (Private, self-employed, government, etc.)

The general table shows the number of employed people in broad occupational categories, along with standard errors. The specific sample focuses on civilian employed individuals aged 16 and older. Estimates are also provided separately for males and females. For both genders, percentages are included with the estimates. Other statistics, including for age and race/ethnicity, can be obtained in other tables.
Obtaining employment and occupation information at different geographic levels.

Default tables often provide estimates for the entire country. To access data at different geographic levels, the "Geos" option is available under “More Tools”. This section offers many geographic options, including:

• 5-digit zip code
• Census Tract
• Congressional district
• County
• State

Specific geographies can be searched for after this selection is made. It is also possible to select multiple geographic areas simultaneously, such as all counties in a single state.

When multiple geographic areas are selected simultaneously, each area appears as a separate column. When many geographic areas are chosen, the results might not display in the browser. Instead, a file containing the information can be downloaded. When selecting data for small geographic areas, it is important to consider the estimates' margins of error. If these margins are large, it is recommended to use five-year estimates. The tables generally default to one-year estimates for the most recent year with data available, but you can also select additional years and five-year average estimates. Because the five-year average estimates are based on more years of data, they generally have smaller margins of error. In some cases, estimates for certain geographic areas might not be available for 1-year estimates but are available for 5-year estimates.

Specific geographies can be searched for after this selection is made. It is also possible to select multiple geographic areas simultaneously, such as all counties in a single state. When multiple geographic areas are selected simultaneously, each area appears as a separate column. When many geographic areas are chosen, the results might not display in the browser. Instead, a file containing the information can be downloaded. When selecting data for small geographic areas, it is important to consider the estimates' margins of error. If these margins are large, it is recommended to use five-year estimates. The tables generally default to one-year estimates for the most recent year with data available, but you can also select additional years and five-year average estimates. Because the five-year average estimates are based on more years of data, they generally have smaller margins of error. In some cases, estimates for certain geographic areas might not be available for 1-year estimates but are available for 5-year estimates.

Occupational Exposure Information

Occupational information for different regions is essential for understanding the work-related exposures residents face in those areas. Several publicly available data sources provide estimates of occupational exposures,often through Job Exposure Matrices. These estimates can be linked to occupation data downloaded from the census website or other sources, along with geography, to create maps of occupational exposures at the geographic level. Below, I demonstrate how this can be done using data from the Occupational Employment and Wage Statistics (O*Net) database, followed by a list of other sources of work-related exposure data (https://www.onetonline.org). 

O*Net collects data on various jobs by surveying workers about their work characteristics. This information can help estimate occupational exposures. One O*Net variable measures work schedules, indicating how regular they are across occupations(higher values indicate less regular work, such as seasonal jobs; lower values indicate more regular schedules). An Excel file is available that lists work schedule values for 4-digit SOC occupation codes. These values can be matched to the percentages of workers in those occupations at different geographic levels,obtained from the Census or other data sources. Before matching, it is important to ensure that occupation codes across data sources are consistent. If the data sources are not consistent, cross-walking should be performed.Weighted averages for each geographic area can be calculated by multiplying the occupation’s schedule value by the number of workers in that occupation with in the area. These products are then summed and divided by the total number of workers across all occupations. This process allows comparison of schedule regularity across geographic areas. Table 3 presents data sources for job exposure information that can be used to perform this type of analysis.

Table 3. Sources for job exposure information that can be used to perform this type of analyses

Data Source	Variable available	Occupation coding	Website
Occupational Exposure Network (O*Net)	High time pressure; decision impact, decision frequency; decision freedom; structured work; repetitive task; machine pacing; conflict situations; deal with unpleasant/angry people; competitive workplace; irregular work schedule; long work hours	4 Digit SOC Codes	https://www.onetonline.org/
Canadian job-exposure matrix (CANJEM)	Chemical exposures including Clean and antimicrobial agents; engine emissions; organic solvents; Polycyclic aromatic hydrocarbons; Diesel engine emissions; Alipathic aldehydes; Alkanes; Carbon monoxide; Formaldehyde; Mononuclear aromatic carbons	Unique job titles	https://www.canjem.ca
Bureau of Labor Statistics Survey of Occupational Injuries and Illnesses	Occupational Injuries and Illnesses including specific information about injury (event, nature, body part, source, etc.)	Broad and detailed SOC and Codes	https://www.bls.gov/opub/hom/soii/concepts.htm
Bureau of Labor Statistics National Compensation Survey	Work-related benefits such as leave and insurance	Broad SOC and NAICS codes	https://www.bls.gov/ebs/data.htm
Killed at Work: U.S. Worker Memorial Database	Work-related fatalities	Direct information about the city/town where the fatality occurred	[https://nationalcosh.org/fatality-database] (https://nationalcosh.org/fatality-database)

Conclusion

Work exposures and experiences can affect health in many ways. Understanding these exposures within communities is a powerful way to assess and improve community health. This guide provides an overview of methods for understanding work-related health issues at the community level. This information can help community health organizations identify intervention targets and outreach strategies, as the workplace can serve as a key site for engagement.Researchers can also use these methods for ecological studies of community health relationships. As work continues to evolve with the shift to a service-based economy, the rise of the gig economy, and the increased use of AI, the health impacts of work will continue to change. This guide can help communities understand how these shifts influence their health.  

References

¹ Thompson, K. (2022, November 27). What percentageof your life will you spend at work?. ReviseSociology.https://revisesociology.com/2016/08/16/percentage-life-work/

² King, M. D., & Scherer, Z. (2025, July 9). NewInteractive data tool shows sources of personal income for hundreds of economicand demographic combinations. Census.gov.https://www.census.gov/library/stories/2025/04/income-sources.html

³ Bunch, L. N., & Ketema, H. (2025b, September 11).Health insurance coverage in the United States: 2024. Census.gov.https://www.census.gov/library/publications/2025/demo/p60-288.html

⁴ Durrah, O. (2023). Do we need friendship in theworkplace? The effect on innovative behavior and mediating role ofpsychological safety. Current Psychology, 42(32),28597-28610.

⁵ U.S. Bureau of Labor Statistics. (2026a, January 22).Employer-reported workplace injuries and illnesses, 2023-2024.https://www.bls.gov/news.release/osh.nr0.htm

⁶ U.S. Bureau of Labor Statistics. (2026e, February19). Census of fatal occupational injuries summary, 2024.https://www.bls.gov/news.release/cfoi.nr0.htm

⁷ Centers for Disease Control and Prevention NationalInstitute for Occupational Safety and Health. (2020, February 18). Pocketguide to chemical hazards. https://www.cdc.gov/niosh/npg/default.html

⁸ Centers for Disease Control and Prevention NationalInstitute for Occupational Safety and Health. (2022, April 28). Psychosocialhazards. https://www.cdc.gov/niosh/learning/safetyculturehc/module-2/8.html

⁹ Purkayastha,D., Vanroelen, C., Bircan, T., Vantyghem, M. A., & Gantelet Adsera, C.(2021). Work, health and Covid‑19: a literature review. ETUI ResearchPaper-Report.

¹⁰ Guo, X.,Nkwopara, C., Peters, C. E., Villeneuve, P. J., & McLeod, C. B. (2025).Impact of heat exposure on workers’ health and safety: a scoping review. Occupationaland Environmental Medicine, 82(11), 563-570.

¹¹ Jones,R. M., & Burstyn, I. (2018). A conceptual model for take-home workplaceexposures. Journal of occupational and environmental hygiene, 15(1),D8-D11.

¹² Vander Weerdt, C., Stoddard‐Dare, P., & DeRigne,L. (2023). Is paid sick leave bad for business? A systematic review. AmericanJournal of Industrial Medicine, 66(6), 429-440.

¹³ Nandi, A., Jahagirdar, D., Dimitris, M. C.,Labrecque, J. A., Strumpf, E. C., Kaufman, J. S., ... & Heymann, S. J.(2018). The impact of parental and medical leave policies on socioeconomic andhealth outcomes in OECD countries: a systematic review of the empiricalliterature. The Milbank Quarterly, 96(3), 434-471.

¹⁴ Robert Wood Johnson Foundation. (2009, September).Education matters for health.

¹⁵ Zhang,X., Gutin, I., Monnat, S., & Montez, J. K. (2025). Changes in theassociation between county industrial composition and working-age mortalityfrom 2000 to 2022. SSM-Population Health, 101849.

¹⁶ Schnake-Mahl,A. S., Mullachery, P. H., Purtle, J., Li, R., Roux, A. V. D., & Bilal, U.(2022). Heterogeneity in disparities in life expectancy across US metropolitanareas. Epidemiology, 33(6), 890-899.

¹⁷ Michimi,A., Ellis-Griffith, G., Nagy, C., & Peterson, T. (2013). Coronary heartdisease prevalence and occupational structure in US metropolitan areas: amultilevel analysis. Health & Place, 21, 192-204.

¹⁸Oh, S., Cano, M., & Kim, Y. (2024). County‐level industrial composition of the labor force and drug overdose mortality rates in the United States in 2018–2021. American journal of industrial medicine, 67(7), 636-645.

¹⁹ Le, D.,& Hawkins, D. (2021). Variation in employment in healthcare occupations andcounty-level differences in COVID-19 cases in the United States ofAmerica. Journal of Occupational and Environmental Medicine, 63(7),629-631.

²⁰ Das, P.,Igoe, M., Lenhart, S., Luong, L., Lanzas, C., Lloyd, A. L., & Odoi, A.(2022). Geographic disparities and determinants of COVID-19 incidence risk inthe greater St. Louis Area, Missouri (United States). Plos one, 17(9),e0274899.

²¹ Keeney,A. J., Beseler, C. L., & Ingold, S. S. (2023). County-level analysis onoccupation and ecological determinants of child abuse and neglect ratesemploying elastic net regression. Child Abuse & Neglect, 137,106029.

²² Kim, B.,Spoer, B. R., Titus, A. R., Chen, A., Thurston, G. D., Gourevitch, M. N., &Thorpe, L. E. (2023). Life expectancy and built environments in the US: amultilevel analysis. American journal of preventive medicine, 64(4),468-476.

²³ Patel,P. C., Devaraj, S., Hicks, M. J., & Wornell, E. J. (2018). County-level jobautomation risk and health: Evidence from the United States. SocialScience & Medicine, 202, 54-60.

²⁴ Eisenberg‐Guyot,J., Mooney, S. J., Hagopian, A., Barrington, W. E., & Hajat, A. (2020).Solidarity and disparity: declining labor union density and changing racial andeducational mortality inequities in the United States. American Journalof Industrial Medicine, 63(3), 218-231.

²⁵ DeFina,R., & Hannon, L. (2019). De-unionization and drug death rates. Socialcurrents, 6(1), 4-13.

²⁶ Levy, B. S., Wegman, D. H., Baron, S., Sokas, R. K.,& McStowe, H. L. (2018). Occupational and Environmental Health:Recognizing and preventing disease and injury. Oxford University Press.

²⁷ Fujishiro, K., Xu, J., & Gong, F. (2010). Whatdoes “occupation” represent as an indicator of socioeconomic status?: Exploringoccupational prestige and health. Social science & medicine, 71(12),2100-2107.

²⁸ Office for National Statistics. (2023, November 28). NationalStatistics Socio-economic classification (NS-SEC) variable: Census 2021.National Statistics Socio-economic Classification (NS-SEC) variable: Census2021 - Office for National Statistics.https://www.ons.gov.uk/census/census2021dictionary/variablesbytopic/labourmarketvariablescensus2021/nationalstatisticssocio-economicclassificationns-sec

²⁹Robert Wood Johnson Foundation. (2009, September). Education matters forhealth. 

³⁰ Karasek Jr, R. A. (1979). Job demands, job decisionlatitude, and mental strain: Implications for job redesign. Administrativescience quarterly, 285-308.

³¹ Madsen, I. E., Nyberg, S. T., Hanson, L. M., Ferrie,J. E., Ahola, K., Alfredsson, L., ... & IPD-Work Consortium. (2017). Jobstrain as a risk factor for clinical depression: systematic review andmeta-analysis with additional individual participant data. Psychologicalmedicine, 47(8), 1342-1356.

³² Xu, S., Huang, Y., Xiao, J., Zhu, W., Wang, L., Tang,H., ... & Liu, T. (2015). The association between job strain and coronaryheart disease: a meta-analysis of prospective cohort studies. Annals ofmedicine, 47(6), 512-518.

³³ De Lange, A. H., Taris, T. W., Kompier, M. A.,Houtman, I. L., & Bongers, P. M. (2003). " The very best of themillennium": longitudinal research and the demand-control-(support)model. Journal of occupational health psychology, 8(4),282.

³⁴ Bakker, A. B., & Demerouti, E. (2007). The jobdemands‐resources model: State of the art. Journal of managerialpsychology, 22(3), 309-328.

³⁵ Siegrist,J. (2017). The effort–reward imbalance model. The handbook of stressand health: A guide to research and practice, 24-35.

³⁶ Greenberg, J. (1987). A taxonomy of organizationaljustice theories. Academy of Management review, 12(1),9-22.

³⁷ Bodin, T., Çağlayan, Ç., Garde, A. H., Gnesi, M.,Jonsson, J., Kiran, S., ... & Vanroelen, C. (2020). Precarious employmentin occupational health–an OMEGA-NET working group position paper. Scandinavianjournal of work, environment & health, 46(3), 321-329.

³⁸ Katz, L. F., & Krueger, A. B. (2019). The riseand nature of alternative work arrangements in the United States,1995–2015. ILR review, 72(2), 382-416.

³⁹ Descatha, A., Hawkins, D., Sembajwe, G., Aloui, S.,Fadel, E., Gagliardi, D., ... & Kang, S. K. (2026). Artificial Intelligenceand occupational health: global umbrella review of applications andlimitations. Safety and Health at Work.

⁴⁰ Occupational Health Surveillance Program. (2025). AmongMassachusetts Workers, Who is More Likely to Work Outdoors?. MassachusettsDepartment of Public Health.https://www.mass.gov/doc/among-massachusetts-workers-who-is-more-likely-to-work-outdoors/download

⁴¹ Forehand, J., & Reagan, S. What is thedifference between industries and occupations?. New Mexico Department ofWorkplace Solutions.https://www.dws.state.nm.us/Portals/0/DM/LMI/What_Is_the_Difference_Between_Industries_and_Occupations.pdf

⁴² Centers for Disease Control and Prevention. (n.d.). 2024National Health Interview Survey (NHIS) questionnaire.https://ftp.cdc.gov/pub/Health_Statistics/NCHS/Survey_Questionnaires/NHIS/2024/EnglishQuest.pdf

⁴³ U.S. Bureau of Labor Statistics. (n.d.). StandardOccupational Classification. https://www.bls.gov/soc/

⁴⁴ U.S. Census Bureau. (2026). North AmericanIndustry Classification System . https://www.census.gov/naics/

⁴⁵ US Census Bureau. (2025, June 23). Industry andoccupation classification.https://www.census.gov/programs-surveys/cps/technical-documentation/methodology/industry-and-occupation-classification.html

⁴⁶ International Labour Organization. (2023, October 3).International Standard Classification of Occupations (ISCO). ILOSTAT.https://ilostat.ilo.org/methods/concepts-and-definitions/classification-occupation/

⁴⁷ United Nations. (n.d.). ISIC.https://unstats.un.org/unsd/classifications/Econ/ISIC.cshtml

⁴⁸ U.S. Bureau of Labor Statistics. (2018). StandardOccupational Classification Manual United States, 2018.https://www.bls.gov/soc/2018/soc_2018_manual.pdf

⁴⁹ U.S. Bureau of Labor Statistics. (2024, April 3). Listof SOC occupations. https://www.bls.gov/oes/2023/may/oes_stru.htm

⁵⁰ U.S. Bureau of Labor Statistics. (2018a). Historicalcomparability of occupation and industry data from the current populationsurvey. https://www.bls.gov/cps/cpsoccind.htm

⁵¹ US Census Bureau. (2025a, May 6). Industry andoccupation code lists & crosswalks.https://www.census.gov/topics/employment/industry-occupation/guidance/code-lists.html

⁵² Centers for Disease Control and Prevention NationalInstitute for Occupational Safety and Health. (2022b, December 13). AboutNIOCCS. Centers for Disease Control and Prevention.https://csams.cdc.gov/nioccs/About.aspx