Self‐Reported Pesticide Application, Access to Pesticide Labels, and QR Code Use Among Agricultural H‐2A Workers, North Carolina, 2024

ABSTRACT

Background

Despite the composition of the United States' agricultural workforce and the sector's reliance on Spanish‐speaking migrant labor, pesticide labels are largely available in English only. Currently, federal regulators are considering strategies to provide pesticide label information access in Spanish via a QR code or through other electronic methods on the pesticide container. Data on access to labels and the use of QR codes among agricultural workers are limited. We provide policy‐relevant findings from surveys collected at a large agricultural worker H‐2A visa arrival hub.

Methods

In April and May 2024, we verbally administered Spanish‐language, in‐person surveys in North Carolina (n = 160). We asked questions about pesticide use, access to labels, and QR code use as well as age, gender, and seasons worked in the United States.

Results

Descriptive analyses revealed approximately 60% of workers had not used a QR code. Among participants who had used pesticides and worked for two or more seasons (42%), approximately 30% reported not having access to the label on the container.

Conclusion

The survey findings show QR‐code‐based pesticide labels will reach a subset of workers, but substantial gaps will remain.

1. Introduction

The agricultural sector in the United States faces significant occupational health and safety challenges, including high rates of fatalities and injuries, placing agricultural work among the most hazardous professions [1]. The inherent nature of farm work, compounded by workers’ mobility and language, presents substantial barriers to providing adequate protection to the agricultural worker community. In response to concerns over occupational pesticide exposure specifically, the United States Environmental Protection Agency (US EPA) has proposed enhancing pesticide label accessibility through bilingual labels and QR codes or other electronic methods [2]. However, the success of these initiatives and their impact on workers’ health hinge on the readiness of Spanish‐speaking agricultural workers to effectively utilize such technology. Ensuring that pesticide information is both accessible and comprehensible remains critical for worker health and safety.

Data to inform such policies are hard to collect due to both the structure of the agricultural system and the characteristics of agricultural workers. The sector relies heavily on a foreign labor force, predominantly from Mexico and Central America. The National Center for Farmworker Health estimates that the United States' agricultural workforce totals approximately 2.9 million workers [3]. According to the National Agricultural Workers Survey, 68% of workers are foreign‐born, and 61% originate from Mexico. Notably, only 37% of these workers report speaking and reading English well [4]. The National Agricultural Workers Survey has historically excluded H‐2A temporary agricultural workers. There has been a notable demographic shift among agricultural workers, characterized by a trend toward settling in one place and a decrease in interstate migration. Additionally, there has been a decline in new entrants to farming since 2007 [5]. These shifts have exacerbated a labor shortage, leading to increased dependence on the H‐2A Temporary Agricultural Program, which has seen a significant rise in certified applications [5]. This program allows agricultural employers to hire workers from eligible countries to work in the United States on a temporary or seasonal basis. North Carolina (NC) is in the top five states with H‐2A positions, and seven out of every 100 H‐2A workers are working in NC [6]. A recent survey of NC agricultural workers revealed that more than 95% were from Mexico and Spanish‐speaking [7].

In this study, we use a unique data source—in‐person data collection at one of the largest H‐2A arrival hubs in the United States—to inform policy implementation by US EPA by assessing pesticide handling, access to pesticide labels, and QR code familiarity among H‐2A temporary agricultural workers in NC.

2. Methods

2.1. Design

There is no sampling frame of agricultural workers in NC. Trained bilingual data collectors attended the arrival hub and conducted a census of arriving workers until all incentives were exhausted on three dates in April and May 2024. Participants provided verbal consent and received a sun‐protective hat as an incentive. Surveys were fielded verbally in Spanish using a Qualtrics offline survey app on an iPad or laptop/tablet with a hotspot. To conduct these surveys, we had to avoid delaying the transfer of arriving workers between arrival processing and departing transportation to farms. Thus, this survey was designed to be as efficient as possible. We designed the survey to last just a few minutes; we used a split‐panel approach to limit the number of questions any participant received.

The East Carolina University and Medical Center IRB approved the research protocol under an exempt determination (#23‐001063).

2.2. Measures

The full codebook is available in our institutional repository. Briefly, we assessed QR code use using the prompt (and a visual aid in the survey) of “Some online information is available by scanning a ‘QR code’ [código QR] with your mobile phone. Here's what one looks like. Have you used a QR code on your phone before?” We separately assessed information about pesticide use and labeling. We asked about pesticide use in the past year (“Last year, did you work with any pesticides while you were in the USA?”) and, for participants with two or more seasons of work in the United States and who reported use of pesticides in the past year, we asked, “When working with pesticides last year, did you see the label on the container?”

For demographics, we assessed age, gender (the interviewer was instructed to ask if unsure), and seasons worked in the United States (“How many agricultural seasons have you worked in the USA before this one?”).

2.3. Analysis

We utilized R (v4.4.0) to perform analyses. Demographic variables, including gender, age, and seasons worked in the United States, were examined to detect any inconsistencies in the data. Additionally, we evaluated the extent of missing responses and addressed these by performing pairwise deletion during the analysis. Descriptive statistics were computed to provide an overview of the demographic characteristics and outcomes. To facilitate a detailed analysis, we categorized age and seasons worked in the United States based on their quantile values. We also performed logistic regression models to investigate the associations between demographic characteristics and the outcomes of interest. We obtained responses on QR codes and pesticide‐related information from 160 workers after removing one respondent aged 18 years who reported working 13 agricultural seasons.

3. Results

Table 1 shows participant demographics. Overall, 41% of participants had used a QR code. Among participants who had worked two or more seasons in the United States, 42% had worked with pesticides. Of those participants, 69% reported access to the pesticide label, which is where the proposed QR codes would be placed. The minimum age of respondents was 18 years, with a median of 36 (IQR: 16.3), and the number of agricultural seasons worked in the United States ranged from 0 to 35, with a median of 10 (IQR: 10.0). The distribution highlights 30% of workers fell into the less experienced group (0–6 seasons), and a significant proportion of the respondents were relatively experienced, with nearly half having worked more than 10 seasons. Table 2 provides a detailed breakdown of QR code use, pesticide use, and access to pesticide labels among workers by age and the number of seasons worked in the United States. The highest usage of QR codes was observed in the youngest age group (64%). There was a noticeable decline in QR code use with increasing age (p‐value < 0.05). Only 25% of workers aged 47+ reported using QR codes.

Table 1.

Background characteristics of surveyed agricultural workers about QR code and pesticide information at H‐2A visa arrival hub, North Carolina, 2024 (N = 160).

Participant characteristics	N	%
Sex
Male	159	99.4
Female	1	0.6
Age (years)
18–30	42	26.3
31–36	43	26.9
37–46	35	21.9
47+	40	25.0
Seasons
0–6	48	30.0
7–10	32	20.0
11–16	41	25.6
17+	39	24.4

Table 2.

Percentage of QR code use, pesticide use, and pesticide label accessibility among agricultural workers by age and seasons worked in the United States.

Characteristics	QR code use (N = 160)	Pesticide use (N = 141)	Access pesticide label (N = 58)
Age
18–30	64.3	48.2	69.2
31–36	46.5	34.2	64.3
37–46	25.7	44.1	71.4
47+	25.0	43.6	70.6
p‐value	< 0.001	0.67	0.98
Seasons
0–6	52.1	48.3a	85.7a
7–10	43.8	37.5	41.7
11–16	43.9	41.5	68.8
17+	23.1	41.0	75.0
p‐value	0.05	0.86	0.10
Overall percentage	41.2	41.8	69.0

Note: p‐values are derived from χ ² tests assessing the association between demographic characteristics (age and seasons worked) and the outcomes (QR code use, pesticide use, and access to pesticide labels). Significant p‐values (p < 0.05) indicate a statistically significant association.

^a Indicates value for 2–6 seasons.

Table 3 presents associations between characteristics and the three outcomes of QR code use, pesticide use, and access to the pesticide label. Age was inversely, significantly associated with QR code use.

Table 3.

Odds ratios (ORs) and 95% confidence intervals (CIs) from logistic regression models assessing the association between demographic characteristics and QR code use, pesticide use, and access to pesticide label among agricultural workers.

Characteristics	QR code use	Pesticide use	Access pesticide label
Age	0.89* (0.83, 0.95)	0.98 (0.93, 1.04)	0.99 (0.90, 1.09)
Seasons	1.07 (0.99, 1.15)	1.03 (0.96, 1.11)	1.03 (0.90, 1.17)

Note: * indicates significance at the traditional p < 0.05 level. Age and seasons worked in the United States were treated as continuous variables to calculate the odds ratios (ORs) and 95% confidence intervals (CIs).

4. Discussion

In‐person surveys at a large H‐2A arrival hub provide a novel data source for a population that is otherwise spread out across often remote and isolated [8] farms with substantial barriers to survey participation. The data provide unique and important information to a current regulatory science question about how to implement Spanish‐language pesticide labels. We outline the current regulatory framework and practical challenges in implementing new technologies like QR codes. A requirement for QR code‐based Spanish‐language labels would enhance the pesticide information required by the existing US EPA Worker Protection Standard (WPS). A recent study in NC found that 76.5% of agricultural workers had consistent internet access and 93.9% had a smart phone in the 2023 growing season [7]. However, as shown in our data, agricultural workers have limited experience with and access to QR codes. Further, workers’ access to pesticide labels is not universal and would therefore limit the reach and impact of such policy changes. Under the WPS regulation, there are two types of workers: pesticide handlers and agricultural workers. The latter are hired to do agricultural work but are not allowed to handle pesticides. This restriction limits their access to pesticide containers. There is one exception to this rule: agricultural workers can handle pesticide containers if they have never been opened. It is unlikely that agricultural workers would be handling pesticide containers or entering a pesticide storage area to check QR codes. Considering that a large number of workers are classified as agricultural workers by the WPS, including H‐2A workers, the proposed bilingual pesticide label QR codes on containers would not be accessible to a large segment of this population of agricultural workers. Indeed, our data showed that nearly 60% of survey respondents had not used a pesticide, and among those who had reported using a pesticide, nearly one‐third had not had access to a pesticide label. The greatest impact of the proposed policy change would be to Spanish‐speaking pesticide handlers, another WPS designation, and applicators due to their interaction with pesticide containers and labels.

Regarding the content of the label accessed by QR code, the information on the Spanish‐language labels should not only provide basic safety precautions, as proposed, but also inform pesticide handlers with practical, detailed instructions on the proper use of pesticides. Special attention should be given to the translation of the label, ensuring the translation is easy to read and technical terms are standardized across pesticide labels. Literacy levels among agricultural workers vary, with an average formal education being completion of ninth grade in the National Agricultural Workers Survey [4]. Because mixing, loading, and applying pesticides are high‐risk activities, it is essential to alert pesticide handlers and applicators not to rely solely on QR codes for pesticide label information because the effectiveness of the QR codes depends on mobile phone signal strength, which can vary across farm locations [7].

While the novel data in this study provide important information for policy implementation, there are limitations to this study. We collected data in a single state and at a single H‐2A arrival hub, and our data thus provide input from an important part of the agricultural workforce. Nonetheless, there are seasonal workers and migrant workers not on H‐2A visas working in agriculture; these workers are not represented in our data.

Future work should consider field studies and evaluations of QR codes. For example, researchers could conduct field studies to assess QR codes’ actual usage and effectiveness in real‐world settings, gather feedback from agricultural workers and pesticide handlers to refine the implementation of QR codes, and evaluate the levels of comprehension of agricultural workers and pesticide handlers of the information provided via the QR codes. Other research could examine the best strategies to enhance training programs for agricultural workers and pesticide handlers on using QR codes and interpreting the information provided.

5. Conclusions

The study uses a novel data source and concludes that while QR codes on pesticide containers would increase language access, their utility for agricultural workers is limited due to infrequent contact with pesticide containers and a lack of familiarity with QR codes. The EPA's proposed QR code system may also encounter implementation challenges among Spanish‐speaking pesticide handlers and applicators, underscoring the need for careful consideration of usability and effective enforcement measures. Based on this study, we recommend placing QR codes in locations that have internet access and that are accessible for agricultural workers (e.g., posting at the required central location), providing training in using QR codes along with mandatory WPS training, and ensuring the information available through QR codes is clear and practical. This approach aims to improve safety and reduce pesticide exposure risks. The use of these data can improve the implementation of important federal policies to protect health and promote equity.

Author Contributions

Tania Connaughton‐Espino, Joseph G. L. Lee, Catherine E. LePrevost, and Natalie D. Rivera conceptualized and designed the study. Lilibeth Andres, Tania Connaughton‐Espino, Paul Janampa, and Natalie D. Rivera acquired the data. Nowrin Nusrat analyzed the data. Lilibeth Andres, Cesar Asuaje, Tania Connaughton‐Espino, Paul Janampa, Joseph G. L. Lee, Catherine E. LePrevost, Nowrin Nusrat, and Natalie D. Rivera interpreted the data, drafted the manuscript, and revised critically for important intellectual content. All authors gave final approval of the version to be published and agree to be accountable for all aspects of the work, ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved.

Disclosure By AJIM Editor of Record

Leena Nylander‐French declares that she has no conflict of interest in the review and publication decision regarding this article.

Ethics Statement

Research was performed at North Carolina Farmworker Health Program, East Carolina University, and North Carolina State University. The East Carolina University and Medical Center IRB approved the research protocol under an exempt determination (#23‐001063).

1. Consent

Verbal consent was obtained from participants before initiating the survey.

Conflicts of Interest

The authors declare no conflicts of interest.

Data Availability Statement

The data that support the findings of this study are openly available in Dataverse at https://dataverse.unc.edu/dataset.xhtml?persistentId=doi:10.15139/S3/0PCZQR, reference number doi:10.15139/S3/0PCZQR.