Injury and Illness Surveillance of U.S. Agricultural Workers

Short abstract

This study provides the National Institute for Occupational Safety and Health with an assessment of the feasibility and desirability of carrying out specific actions recommended by an independent panel regarding program relevance and impact.

Abstract

Agriculture remains one of America's oldest and most valued industries, but is also one of the most hazardous, with farmworkers experiencing high rates of injuries and illnesses. The National Institute for Occupational Safety and Health's (NIOSH) Agriculture, Forestry and Fishing (AgFF) Program provides leadership to prevent harm to workers in the three named sectors. Since its inception, the AgFF Program has spearheaded numerous surveillance initiatives to understand the magnitude of injuries and illnesses among agricultural worker populations, identify vulnerable groups, and evaluate the effectiveness of intervention measures.

In 2012, the program underwent a review from an independent panel convened to evaluate progress in program relevance and impact. While the panel offered a number of recommendations for improving surveillance, it provided little guidance on how to prioritize and achieve the recommendations. In 2015, NIOSH asked the RAND Corporation to assess options for action in response to panel recommendations. The goal of this study is to provide NIOSH with a practical assessment of the feasibility and desirability of carrying out actions to meet surveillance-related panel recommendations, given current AgFF Program resources and priorities.

The authors conducted literature reviews and targeted interviews to detail how actions could be implemented and identify barriers to their achievement. For each action, the authors applied criteria relevant to assessing feasibility (costs, partnership engagement, information availability and accessibility, policy barriers, timelines) and desirability (relevance to program priorities, information quality, and impact). Overall, the study identifies a number of actions that balanced both feasibility and desirability for NIOSH to consider as it determines the direction of the AgFF Program.

Agriculture remains one of America's oldest and most valued industries, but it is also one of the most hazardous. According to the Bureau of Labor Statistics (BLS) Survey of Occupational Injuries and Illnesses, in 2015, almost 36,000 farmworkers experienced an injury event (5.8 per 100 full-time workers) (BLS, 2017a). An estimated 1,900 farmworkers suffered an illness event (31.8 per 10,000 full-time workers). These rates were higher than those experienced by workers in other hazardous occupations, such as fishing and hunting (4.2 injuries per 100 full-time workers), forestry and logging (2.2 injuries per 100 full-time workers and 5.6 illnesses per 10,000 full-time workers), mining (1.4 injuries per 100 full-time workers and 7.5 illnesses per 10,000 full-time workers), and construction (3.4 injuries per 100 full-time workers and 8.6 illnesses per 10,000 full-time workers) (BLS, 2017a). In addition, about 400 fatalities (8.3 percent of total deaths across all industries)¹ occurred while working on farms, compared with 23 (0.5 percent) and 120 (2.5 percent) fatalities in fishing and hunting and mining worker populations, respectively (BLS, 2017b). Children working on farms are a particularly vulnerable population for agriculture-related illness, injury, and death. In 2012, an estimated 7,780 household youth experienced injuries on farms (National Children's Center for Rural and Agricultural Health and Safety, 2014). These statistics reveal a stark reality: The small population of workers directly responsible for feeding, clothing, and fueling the nation are shouldering a heavy public health burden.

The National Institute for Occupational Safety and Health's (NIOSH's) Agriculture, Forestry and Fishing (AgFF) Program provides leadership to prevent harm to workers in the three named sectors (NIOSH, 2017). Since its inception, the AgFF Program has spearheaded numerous surveillance initiatives to understand the magnitude of injuries and illnesses in agricultural worker populations, identify vulnerable groups, and evaluate the effectiveness of interventions. In 2012, the program underwent a review from an independent panel convened to evaluate progress in program relevance and impact (NIOSH, 2012). The panel report included several recommendations for NIOSH to improve agricultural injury and illness surveillance. In certain instances, the panel suggested specific activities to perform to improve surveillance, include vulnerable populations, and extend NIOSH research to address emerging health issues faced by farmworkers. However, the panel report itself was unclear regarding how NIOSH should prioritize recommendations and actually implement activities. In 2015, NIOSH contracted with the RAND Corporation to assess options for action in response to panel recommendations. The goal of this study is to provide NIOSH with an assessment of the feasibility and desirability of carrying out actions to meet surveillance-related recommendations given current AgFF Program resources and priorities.

Many possible actions for meeting recommendations were contained in the panel report, and we conducted literature reviews and targeted interviews to detail how actions could be implemented and identify barriers to their achievement. Table 1 describes the surveillance-related recommendations and actions. For each action, we determined whether it could be achieved through (1) an extramural funding mechanism, (2) partnership engagement or partner capabilities, and (3) direct action by NIOSH.

Table 1.

Surveillance-Related Recommendations and Actions Inferred from the 2012 Review Report

Recommendation	Action
R1: Develop standard definitions to help improve knowledge and understanding of populations at risk of worksite exposures	1.1: Develop concise definitions of populations at occupational exposure risk, and profile them by agriculture commodity specialization, demographic factors, work organization patterns, worksite tasks across various enterprise types, and knowledge or use of emerging technologies.
R2: Focus surveillance efforts on very large and nonfamily farms	2.1: Conduct an assessment of data sources for providing information on very large farms and nonfamily farms.
R3: Expand the scope of knowledge on hired farmworker risk for occupational illness and injury	3.1: Collaborate with the National Agricultural Worker's Survey to field supplementary modules on occupational health outcomes among hired farmworkers and perform specific analyses on variability of risk.
R4: Improve surveillance coverage and knowledge of child agricultural worker populations	4.1: Improve the validity of the Childhood Agricultural Injury Survey by extending the sample to include labor aggregators.
	4.2: Collaborate with the National Agricultural Worker's Survey to field a supplementary module targeted at children who work as hired farmworkers on crop farms.
R5: Establish a comprehensive understanding of data sources relevant to NIOSH agricultural surveillance-related programming	5.1: Reevaluate the surveillance mechanisms of USDA's NASS surveys for usefulness in NIOSH activities.
	5.2: Identify and evaluate the potential of existing data sources (including those currently or recently used by NIOSH) for illness and injury surveillance of agricultural workers.
R6: Reconsider analytical and reporting approaches for conveying child agricultural worker safety and health risks	6.1: Identify feasible, desirable alternatives to five-year age groupings for child agricultural workers (<18 years old).
	6.2: Stratify data pertaining to children into at least three divisions: those employed within agricultural enterprises, those self-employed, and unpaid family workers contributing effort within agricultural enterprises.
R7: Identify context in reports and analyses authored by NIOSH or regional centers	7.1: When reporting on surveillance activities, NIOSH and Ag Centers should provide information on agricultural enterprises, known operational worksite characteristics, and other useful descriptors that provide context for understanding occupational cohorts at risk.
R8: Address emerging issues in animal agricultural production practices	8.1: Establish partnerships to monitor zoonotic diseases in animal and human populations.
	8.2: The National Children's Center should monitor developments in alternative animal production systems and develop guidelines for children and youth workers within these systems.

NOTES: Ag Center = Agricultural Safety and Health Center; NASS = National Agriculture Statistics Service; USDA = U.S. Department of Agriculture.

We also constructed two versions of actions achievable through an extramural funding mechanism: an ideal version and a satisfactory version. The ideal version refers to our estimation of the types of tasks and time frames needed to either fully achieve the action or ensure that implementation results in meaningful impacts on improvement of farmworker health and safety. The satisfactory version refers to tasks and time frames that may result in lesser (compared with the ideal), but still acceptable, impacts. Different costs and timelines were estimated based on the two versions.

We applied criteria to the actions that were relevant for assessing feasibility (costs and other resources, partnership engagement, information availability and accessibility, policy barriers, and timelines) and desirability (relevance to program priorities, information quality, and action impacts). For all criteria except costs and resources, we constructed an ordinal five-point rating scale based on defined attributes of the criterion relevant to the accomplishment of an action. Low values (1–2) reflected adverse or unfavorable conditions for an action, as contrasted with high values (4–5), which reflected favorable conditions for action achievement. Because we lacked information on planned allocation of resources within the AgFF Program, for the costs and resources criterion, we presented general estimates and a descriptive assessment of implications. For each action, we calculated average scores across criteria for three conditions: achievability (average of all feasibility and desirability criteria), feasibility (average of the feasibility criteria), and desirability (average of the desirability criteria).

Overall, we find that direct action and extramural funding mechanisms tend to show higher achievability and feasibility than do actions that depend on partner capabilities for implementation. Actions that depend on partner engagement, resources, or capabilities may have lengthier timelines for establishing or building partnerships and negotiating agreements (including data use and data-sharing). In addition, there is uncertainty about the availability and accessibility of information that could be obtained through such partnerships. When we focused on desirability, however, the results are more mixed. Top actions are associated with all three implementation mechanisms.

To prioritize actions, we first excluded actions achievable through direct action. These actions all revolve around NIOSH performing new analyses or issuing new reports with existing data; NIOSH AgFF Program leadership indicated that, given resource constraints, these kinds of activities would only be done on an as-needed basis or on request. Because NIOSH already has a plan for addressing these actions, we performed a ranking exercise to determine the relative achievability, feasibility, and desirability of the remaining actions. Table 2 presents the average criteria rating scores ranked by average achievability score. The table also displays estimated action costs per year and proposed duration. Based on these rankings, we determined the following action categories:

Table 2.

Average Criteria Rating Scores for Actions (excluding the direct action mechanism)

Action	Mechanism	Cost per Year*	Duration (years)	Achievability Score	Feasibility Score	Desirability Score
1.1: Develop definitions and taxonomy	EXF	$150K	1	4.00	4.20	3.67
		$170K	1
5.2: Evaluate sources of surveillance data	EXF	$175K	2	4.00	3.60	4.67
		$160K	5
3.1: Field the National Agricultural Worker's Survey and hired worker survey	PART	$330K	3	3.88	3.40	4.67
4.1: Improve validity of the Childhood Agricultural Injury Survey	PART	$1 million	1	3.50	3.20	4.00
8.2: Monitor alternative animal production systems through the National Children's Center	PART	$175K	2	3.38	3.20	3.67
		$170K	3
4.2: Field the National Agricultural Worker's Survey and child injury survey	PART	$330K	6	3.13	3.00	3.33
8.1: Partner for zoonotic disease	PART	Unknown	Uncertain	3.13	3.00	3.33

NOTES: EXF = extramural funding; PART = partnership.

^*The two costs reflect the satisfactory version of an action (upper row) and the ideal (lower row).

• high priority (high and moderate scores in combination): 1.1, 5.2

• moderate priority (moderate scores on both or high/low combination): 3.1, 4.1

• low priority (low on both feasibility and desirability scores): 4.2, 8.1, 8.2.

The results of our assessment suggest that NIOSH should prioritize two actions based on their feasibility and desirability:

• 1.1: Develop concise definitions of populations at occupational exposure risk, and profile them by AgFF commodity specialization, demographic factors, work organization patterns, worksite tasks across various enterprise types, and knowledge or use of emerging technologies.

• 5.2: Identify and evaluate the potential of existing data sources (including those currently or recently used by NIOSH) for illness and injury surveillance of agricultural workers (includes actions 2.1 and 5.1).

When estimated total costs are factored into the assessment, action 1.1 exhibits better overall value, but we note that action 5.2 is considered highly desirable. Action 1.1, however, falls on the lower end of the cost spectrum while also exhibiting high feasibility and moderate desirability. Both actions are achievable through extramural funding mechanisms, including contracts, cooperative agreements, or grants. For action 5.2, our assessment suggests that cooperative agreements would allow for strong collaboration between NIOSH and research partners while also offering flexibility to carry out complex projects that may yield unexpected findings. We considered action 1.1, however, to be well suited to a contract mechanism because it is a clearly defined project that could be carried out in a relatively short time frame.