Acute Pesticide Poisoning Among Farm Workers

AMERICAN JOURNAL OF INDUSTRIAL MEDICINE 51:883–898 (2008)

Acute Pesticide Poisoning Among Agricultural
Workers in the United States, 1998–2005

Geoffrey M. Calvert, MD, MPH,1Ã Jennifer Karnik, MPH,1 Louise Mehler, PHD, MD,2
John Beckman, BS,3 Barbara Morrissey, MS,4 Jennifer Sievert, BA,5
Rosanna Barrett, MPH,6 Michelle Lackovic, MPH,7 Laura Mabee, BA,8
Abby Schwartz, MPH,9 Yvette Mitchell, MS,10 and Stephanie Moraga-McHaley, MS11

Background Approximately 75% of pesticide usage in the United States occurs in
agriculture. As such, agricultural workers are at greater risk of pesticide exposure than
non-agricultural workers. However, the magnitude, characteristics and trend of acute
pesticide poisoning among agricultural workers are unknown.
Methods We identified acute pesticide poisoning cases in agricultural workers between the
ages of 15 and 64 years that occurred from 1998 to 2005. The California Department of
Pesticide Regulation and the SENSOR-Pesticides program provided the cases. Acute
occupational pesticide poisoning incidence rates (IR) for those employed in agriculture were
calculated, as were incidence rate ratios (IRR) among agricultural workers relative to non-
agricultural workers.
Results Of the 3,271 cases included in the analysis, 2,334 (71%) were employed as
farmworkers. The remaining cases were employed as processing/packing plant workers
(12%), farmers (3%), and other miscellaneous agricultural workers (19%). The majority
of cases had low severity illness (N ¼ 2,848, 87%), while 402 (12%) were of medium
severity and 20 (0.6%) were of high severity. One case was fatal. Rates of illness among
various agricultural worker categories were highly variable but all, except farmers,
showed risk for agricultural workers greater than risk for non-agricultural workers by an
order of magnitude or more. Also, the rate among female agricultural workers was almost
twofold higher compared to males.
Conclusion The findings from this study suggest that acute pesticide poisoning in the
agricultural industry continues to be an important problem. These findings reinforce the
need for heightened efforts to better protect farmworkers from pesticide exposure. Am. J.
Ind. Med. 51:883–898, 2008. Published 2008 Wiley-Liss, Inc.{

KEY WORDS: pesticides; surveillance; poisoning; agriculture; farmworkers

1 9
Division of Surveillance, Hazard Evaluations and Field Studies, National Institute for Division of Environmental Health, Michigan Department of Community Health, Lansing,
Occupational Safety and Health, Centers for Disease Control and Prevention, Cincinnati, Michigan
10
Ohio Bureau of Occupational Health, New York State Department of Health,Troy, New York
2 11
Department of Pesticide Regulation, California Environmental Protection Agency, Sacra- New Mexico Occupational Health Registry, University of New Mexico, Albuquerque,
mento, California New Mexico
3
Public Health Institute, Oakland, California The findings and conclusions in this report are those of the authors and do not necessarily
4
Office of Environmental Assessments,Washington State Department of Health, Olympia, represent the views of the National Institute for Occupational Safety and Health or each
Washington author’s state agency.
5
Environmental and Injury Epidemiology and Toxicology Branch, Texas Department of *Correspondence to: Geoffrey M. Calvert, National Institute for Occupational Safety and
State Health Services, Austin,Texas Health, 4676 Columbia Parkway, R-17, Cincinnati, OH 45226. E-mail: jac6@cdc.gov
6
Florida Department of Health,Tallahassee, Florida Accepted 23 June 2008
7
Louisiana Department of Health and Hospitals, New Orleans, Louisiana DOI 10.1002/ajim.20623. Published online in Wiley InterScience
8
Office of Environmental Public Health, Oregon Department of Human Services, Portland, (www.interscience.wiley.com)
Oregon
Published 2008 Wiley-Liss, Inc.
{
This article is a US Government work and, as such, is in
the public domain in the United States of America.

884 Calvert et al.

INTRODUCTION illnesses for more than 30 years [Calvert et al., 2001]. To assess
the magnitude, characteristics and trend of acute pesticide
Pesticides are widely used in agriculture to control insects, poisoning among agricultural workers in the United States
microorganisms, fungi, weeds, and other pests. The control of since the revised WPS went into effect in 1995, an analysis of
these pests serves to increase crop yield and decrease manual data obtained from these surveillance systems was performed.
labor [Litchfield, 2005]. In 2000 and 2001, over 5 billion
pounds of pesticides were used annually throughout the world. MATERIALS AND METHODS
The United States was responsible for 24% of this total usage
[Kiely et al., 2004]. Within the US, the agricultural industry Data were obtained on individuals age 15 through 64 who
accounts for approximately 75% of the annual poundage used. developed an acute pesticide-related illness or injury and who
Farming is an essential component of our economy, but were employed in the agricultural industry when the occupa-
agricultural workers suffer elevated rates of injuries, hearing tional pesticide exposure occurred. Census 1990 industry
loss, and respiratory disease [Rust, 1990; Linaker and codes (1990 CIC) and Census 2002 industry codes (2002 CIC)
Smedley, 2002; Tak and Calvert, 2008]. Pesticides are also were used to identify cases employed in agriculture [US
an important source of injury and illness among farmers and Bureau of the Census, 1992; US Census Bureau, 2005]. The
farm workers [Calvert et al., 2004]. Previous work has agricultural industry was defined as: agricultural production,
suggested that the agricultural industry’s disproportionately excluding livestock (1990 CIC ¼ 010; 2002 CIC ¼ 0170);
high pesticide use puts farmers and farm workers at greater agricultural production, including livestock (1990 CIC ¼ 011;
risk of pesticide exposure than others [Reeves and Schafer, 2002 CIC ¼ 0180); and agricultural services (1990
2003; Calvert et al., 2004]. Farmers and farmworkers may be CIC ¼ 030; 2002 CIC ¼ 0290). All agricultural industry cases
exposed by mixing, loading and applying pesticides, or while also had their occupation coded using Census 1990 occupation
performing duties not involved with pesticide application codes (1990 COC) and Census 2002 occupation codes (2002
(e.g., weeding, harvesting, thinning, irrigating, or planting). COC) [US Bureau of the Census, 1992; US Census Bureau,
Recognizing the need for increased worker protections 2005]. Agricultural occupations included: farmworkers (1990
from pesticide exposures, the U.S. Environmental Protection COC ¼ 477, 479, 484; 2002 COC ¼ 6050, 6120, 8710, 8960);
Agency (EPA) promulgated rules in 1974 known as the farmers (1990 COC ¼ 473–476; 2002 COC ¼ 0200, 0210);
Worker Protection Standard for Agricultural Pesticides processing/packing plant workers (1990 COC ¼ 488, 699;
(WPS; 40 CFR 170) and aimed at reducing pesticide 2002 COC ¼ 6040, 7830, 7850, 8640, 8720, 8800, 8860,
exposures among agricultural workers. However, by 1992, 9640); and, other miscellaneous agricultural workers (workers
EPA estimated that hired farmworkers alone experienced up employed in agriculture but whose 1990 COC and 2002 COC
to 10,000–20,000 illnesses and injuries from pesticide did not match any of those specified for the other three
exposures each year [US EPA, 1992] and concluded that agricultural occupations). A pesticide handler was defined as
the WPS was inadequate in its requirements and scope of an individual who mixed, loaded, transported and/or applied
coverage. That year, EPA revised and expanded the WPS pesticides, or an individual who repaired or maintained
rules to include changes in labeling, coverage of more pesticide application equipment at the time of pesticide
workers and agricultural operations, prohibition of employer exposure (insufficient information was available to determine
retaliation against workers attempting to comply with the pesticide handler status for 68 individuals). This analysis
standard, and the following requirements: notification of excluded illnesses associated with non-occupational expo-
workers about pesticide applications; restriction of re-entry sures and illnesses associated with intentional (e.g., suicidal,
into pesticide-treated areas; and, provision of personal malicious intent) exposures.
protective equipment (PPE), decontamination supplies, Cases under 15 years of age and those 65 years and older
emergency assistance, and pesticide safety training. Detailed were omitted from analysis. The age range was chosen a
information on the magnitude, characteristics and trend of priori, and is considered to include the vast bulk of workers
acute pesticide poisoning since the revised WPS went into who are gainfully employed. A total of 66 cases age 65 and
effect in 1995 are unavailable. older were identified but not included in this analysis (this
The National Institute for Occupational Safety and Health represents a rate of 13/100,000 agricultural workers age 65
(NIOSH) developed the Sentinel Event Notification System and older). Furthermore, Current Population Survey (CPS)
for Occupational Risks-Pesticides (SENSOR-Pesticides) pro- data, the source of our denominator data, are unavailable on
gram [Calvert et al., 2001] to monitor risks from pesticide workers less than 15 years of age.
exposure. Data from this program are available beginning in
1998, when standardized definitions for cases and data Data Sources
elements were adopted [Calvert et al., 2001]. The California
Department of Pesticide Regulation (CDPR) has a similar Data for this analysis were obtained from CDPR and the
surveillance program that has been tracking pesticide-related SENSOR-Pesticides program. State health departments in

Pesticide Poisoning Among Agricultural Workers 885

ten states participated in the SENSOR-Pesticides program from I to IV, with I representing the most toxic and IV the
and contributed data. These ten state health departments were least toxic substances [US EPA, 1975]. The toxicity cate-
the: Arizona Department of Health, California Department of gory for each case was obtained by the relevant state agency
Public Health (CDPH), Florida Department of Health, conducting pesticide poisoning surveillance. When toxicity
Louisiana Department of Health and Hospitals, Michigan category data was not given, the category was determined
Department of Community Health, New Mexico Department by NIOSH based on standardized criteria from a dataset
of Health (through an agreement with the University of New provided by EPA. Cases exposed to more than one pesticide
Mexico), New York State Department of Health, Oregon product were assigned the toxicity category representing the
Department of Human Services, Texas Department of State pesticide product with the greatest toxicity.
Health Services, and the Washington State Department of
Health. The time frame for data availability varied according Case Definition
to state agency. The years of data availability are provided in
Table I. Each of these agencies maintains its own passive A standardized case definition is used by all participating
population-based surveillance system for acute pesticide- SENSOR-pesticides states. Cases of acute pesticide poison-
related illness or injury with occasional outreach to potential ing are included in the analyses if they were classified
reporters to stimulate reporting (e.g., contacting poison as definite, probable, possible or suspicious. A classification
control centers to encourage them to report or reviewing category is assigned to a case based on three factors: (1) the
physician reports submitted to workers’ compensation strength of evidence that a pesticide exposure occurred; (2)
insurance carriers to identify eligible cases) [Calvert et al., whether adverse health effects were observed by a healthcare
2001, 2004]. Each agency obtains case reports from many professional versus being self-reported; and (3) the presence
different sources. All require physician reporting of pesti- of sufficient evidence that the known toxicology of the agent
cide-related illness cases. Other sources of case reports vary was consistent with the observed health effects. Cases
by state and include poison control centers, state agencies exposed to pesticides for which there is limited toxicological
with jurisdiction over pesticide use (e.g., departments of data were classified as suspicious [CDC, 2001a]. CDPR uses
agriculture), and workers’ compensation claims. Because a comparable case definition [CDPR, 2006]. In this article,
each state removes any personal identifiers from the data acute pesticide poisoning and acute pesticide-related illness
prior to submission to the Centers for Disease Control and and injury are used interchangeably.
Prevention this study was exempt from consideration by the Illness severity was assigned to all cases using stand-
federal Human Subjects Review Board. ardized criteria which were based on signs and symptoms,
Once a case report is received, the state agency determines medical care received, and lost time from work [CDC,
whether the subject was symptomatic and whether the 2001b]. Low severity illness/ injury consist of illnesses and
involved chemical is a pesticide. If so, attempts are made to injuries that generally resolve without treatment and
interview the poisoned subject or their proxy to obtain details where minimal time (<3 days) is lost from work. Such cases
on the poisoning event, and any medical records are requested. typically manifest as eye, skin and/or upper respiratory
Besides identifying, classifying, and tabulating pesticide irritation. Moderate severity illness/injury consists of non-
poisoning cases, the states periodically perform in-depth life-threatening health effects that are generally systemic and
investigations of pesticide-related events, and develop inter- require medical treatment. No residual disability is detected,
ventions aimed at particular industries or pesticide hazards. and time lost from work is less than 6 days. High severity
Cases obtained from CDPR were cross-referenced with illness/injury consists of life threatening health effects that
cases from the CDPH based on age, gender, date of exposure, usually require hospitalization, involve substantial time lost
and pesticide name. Matching cases were assumed to be the from work (>5 days), and may result in permanent impair-
same individual and were counted only once. ment or disability. Death pertains to fatalities resulting from
exposure to one or more pesticides.
Information Available on Each Case
Data Analysis
Data collected for each case by the SENSOR-Pesticides
and CDPR surveillance systems include case demographics, SAS v. 9.1 was used for data management and analysis
signs and symptoms of illness or injury, illness severity, EPA [SAS Institute Inc, 2003]. Chi square statistical analyses
toxicity category, identity of implicated pesticides and the were performed on categorical data. Incidence rates (IR) for
target (e.g., crop) of their application (if any), information on acute occupational pesticide poisoning were calculated
factors that may have contributed to the pesticide exposure for those employed in agriculture. Rates were calculated
that precipitated illness, and the source of the case report. for occupational categories within agriculture, for each year
EPA evaluates the toxicity of and assigns a toxicity studied, by age group, and for three geographic regions in the
category to each pesticide product. The categories range US. The numerator represents the number of relevant cases

886 Calvert et al.

TABLE I. Data on Demographics, PesticideToxicity, Pesticide Handler, Pesticide Functional Class, and ApplicationTarget for 3,271Acute Pesticide Poisoning
Cases in the Agricultural Industry by Severity Category,1998^2005

Fatal, N High Severity, N Medium Severity, N Low Severity, N Total, N (%)a
Total 1 20 402 2,848 3,271
Age
15^17 0 0 5 19 24 (1)
18^24 0 1 82 521 604 (18)
25^34 0 2 109 786 897 (27)
35^44 0 7 103 630 740 (23)
45^54 0 3 59 358 420 (13)
55^64 1 5 28 138 172 (5)
Unknown 0 2 16 396 414 (13)
State where illness identified (years of data availability)
Arizona (1998^1999) 0 0 4 17 21 (1)
California (1998^2005) 1 10 274 2,235 2,520 (77)
Florida (1998^2005) 0 0 23 109 132 (4)
Louisiana (2001^2005) 0 4 14 27 45 (1)
Michigan (2001^2005) 0 1 7 14 22 (1)
New Mexico (2005 only) 0 0 2 10 12 (1)
NewYork (1998^2005) 0 0 6 7 13 (1)
Oregon (1998^2005) 0 0 6 37 43 (1)
Texas (1998^2005) 0 3 40 146 189 (6)
Washington (2001^2005) 0 2 26 246 274 (8)
Gender
Female 0 6 114 934 1,054 (32)
Male 1 14 288 1,886 2,189 (67)
Unknown 0 0 0 28 28 (1)
Year exposed
1998 0 2 64 358 424 (13)
1999 1 1 85 337 424 (13)
2000 0 4 64 315 383 (12)
2001 0 0 30 236 266 (8)
2002 0 0 26 576 602 (18)
2003 0 0 43 279 322 (10)
2004 0 11 35 396 442 (14)
2005 0 2 55 351 408 (12)
Toxicity categoryb
I 1 11 232 1,418 1,662 (51)
II 0 1 68 599 668 (20)
III and IV 0 6 82 792 880 (27)
Unknown 0 2 20 39 61 (2)
Pesticide handler
Yesc 1 10 190 867 1,068 (33)
No 0 10 200 1,925 2,135 (65)
Unknown 0 0 12 56 68 (2)
Pesticide functional classd
Insecticidesöall 1 10 210 1,540 1,761 (54)
Insecticides only 0 7 115 747 869 (27)
Insecticides combined 1 3 95 793 892 (27)
Fungicidesöall 1 4 90 734 829 (25)
Fungicides only 0 2 28 147 177 (5)
Fungicides combined 1 2 62 587 652 (20)


TABLE I. (Continued )

Fatal, N High Severity, N Medium Severity, N Low Severity, N Total, N (%)a
Disinfectantsöall 0 2 56 389 447 (14)
Disinfectants only 0 2 48 238 288 (9)
Disinfectants combined 0 0 8 151 159 (5)
Herbicidesöall 0 1 56 400 457 (14)
Herbicides only 0 1 42 318 361 (11)
Herbicides combined 0 0 14 82 96 (3)
Fumigantsöall 0 4 44 416 464 (14)
Fumigants only 0 4 44 416 464 (14)
Fumigants combined 0 0 0 0 0 (0)
Other 0 1 18 130 149 (5)
Application target
Fruit crops 0 3 112 1,047 1,162 (36)
Vegetable crops 0 2 45 411 458 (14)
Soil 0 2 20 316 338 (10)
Grains, grasses and fiber crops 0 1 58 201 260 (8)
Landscape/ornamental 0 1 18 159 178 (5)
Undesired plant 0 0 6 74 80 (2)
Beverage crops 0 0 7 35 42 (1)
Crops that cross categories 0 0 6 32 38 (1)
Building structure/surface/space 0 1 8 35 44 (1)
Oil crops 0 0 5 15 20 (1)
Miscellaneous field crops 0 0 8 11 19 (1)
Veterinary (livestock or domestic) 0 0 4 13 17 (1)
Other 1 2 18 182 203 (6)
Not applicable 0 4 55 161 220 (7)
Unknown 0 4 32 156 192 (6)
a
Percentages may not sum to 100 due to rounding.
b
Acute pesticide toxicity category as defined by the U.S. EPA.
c
A pesticide handler was defined as an individual who mixed, loaded, transported and/or applied pesticides, or an individual who repaired or maintained pesticide application
equipment at the time of pesticide exposure.
d
Cases may be exposed to more than one functional class. The rows labeled with ‘‘combined’’ pertain to cases exposed to more than one pesticide active ingredient, some of
which belong to the pesticide functional class specified in the row label and others belonging to other pesticide functional classes.

captured by CDPR and SENSOR-Pesticides from 1998 to were calculated to determine the risk of acute pesticide
2005. Denominator data, including employment counts and poisoning while working in agriculture. This ratio was
the hours worked estimate, were obtained from the CPS calculated by dividing the IR among agricultural workers
[BLS, 2007]. The hours worked data were used to derive full- with that among non-agricultural workers. A ratio greater
time equivalent (FTEs) estimates, with one FTE equivalent to than one suggests an increased risk in farmers or farm-
2,000 hr worked. Denominator data correspond to the states workers, while a ratio less than one suggests a decreased risk.
and time periods of data availability (Table I). Although Confidence intervals (95% CIs) were calculated for each rate
rates were calculated with the two denominator estimates ratio as described by Rothman [1986].
(employment counts and FTE estimates), the rates calculated
with FTEs as the denominator are given prominence as they RESULTS
have previously been shown to be conceptually preferable
over the use of raw employment counts [Ruser, 1998]. The From 1998 to 2005, 3,271 case reports met inclusion
comparison group consisted of all workers not employed in criteria (Table I). Of these, 1,078 (33%) were identified by the
agriculture. IR for workers employed in non-agricultural SENSOR-Pesticides program and 2,193 (67%) originated
industries were similarly calculated, with the numerator and from CDPR (527 cases were identified by both SENSOR and
denominator data obtained from the same agencies (SEN- CDPR and were included in the CDPR total only). There
SOR/CDPR and CPS, respectively) that provided the data on were 1,942 separate pesticide exposure events, 1,762 of
agricultural workers. Finally, incidence rate ratios (IRR) which (91%) involved only one ill agricultural worker. Of the

888 Calvert et al.

180 (9%) multi-victim events, the median number of ill O’Malley et al., 2005 for detailed information on the event
agricultural workers was 3 (range 2–123). The number of involving 72 workers]. The second largest incident occurred
pesticide exposure events decreased over the time period in 2004, when 121 peach harvesters became ill after exposure
studied but the average number of cases per event increased to drift from an application of methamidophos and mancozeb
(in 1998 there were 308 events with an average of 1.4 cases to a nearby potato ﬁeld. Most of these workers experienced
per event whereas in 2005 there were 209 events with an low severity illness (N ¼ 111, 92%), and the other 10 workers
average of 2.0 cases per event). (8%) experienced moderate severity illness.

Description of the Three Largest Events Incidence Rates

More than three quarters of the cases (N ¼ 2,520, 77%) Tables II and III and Figure 1 summarize IRs for
occurred in California. Among these cases, we found a small agricultural workers and non-agricultural workers from 1998
number of events that exposed large numbers of agricultural to 2005. Overall, the average annual IR among agricultural
workers. In two separate 2002 incidents, irritant vapors workers was 53.6/100,000 FTEs and 1.38/100,000 FTEs
drifted from soil treatments with metam-sodium and caused among all non-agricultural workers combined (IRR ¼ 38.9
low severity illness in 123 vineyard workers and in 72 95% CI 37.2, 40.6). Agricultural workers’ annual rates
workers at a carrot processing facility, respectively [see ﬂuctuated between 33.8/100,000 FTEs (2001) and 79.9/

TABLE II. Incidence Rates by Industry,Year of Exposure, Age Group and US Region for 3,271Acute Pesticide Poisoning Cases,1998^2005

Agricultural workers Non-agricultural workers

FTE Incidence FTE Incidence Incidence rate
Count estimatea rateb Count estimatea ratec ratiod
Year of exposure
1998 424 790,837 53.6 762 40,792,468 1.9 28.7
1999 424 781,985 54.2 656 42,040,152 1.6 34.7
2000 383 781,654 49.0 577 41,041,774 1.4 34.9
2001 266 787,481 33.8 552 49,456,474 1.1 30.3
2002 602 753,595 79.9 598 49,110,280 1.2 65.6
2003 322 756,610 42.6 694 50,151,930 1.4 30.8
2004 442 735,270 60.1 716 50,989,934 1.4 42.8
2005 408 710,851 57.4 638 53,000,554 1.2 47.7
Age group (years)
15^17 24 99,364 24.2 69 3,140,858 2.2 11.0
18^24 604 792,852 76.2 851 43,536,103 2.0 39.0
25^34 897 1,392,263 64.4 1,246 92,090,687 1.4 47.6
35^44 740 1,601,894 46.2 1,303 106,763,138 1.2 37.9
45^54 420 1,276,042 32.9 968 89,437,936 1.1 30.4
55^64 172 935,868 18.4 356 41,614,845 0.9 21.5
Unknown 414 ö ö 400 ö ö ö
US region
Weste 2,858 3,168,485 90.2 3,556 149,655,538 2.4 38.0
Southf 366 2,258,774 16.2 1,225 141,962,800 0.9 18.8
Eastg 45 671,024 7.0 412 84,965,229 0.5 13.8
Total 3,271 6,098,283 53.6 5,193 376,583,567 1.4 38.9
a
FTE, full-time equivalent.
b
Incidence rate per 100,000 FTEs. Includes agricultural workers in Arizona, California, Florida, Louisiana, Michigan, New Mexico, New York, Oregon, Texas, and Washington.
c
Incidence rate per 100,000 FTEs. Includes non-agricultural workers in Arizona, California, Florida, Louisiana, Michigan, New Mexico, New York, Oregon,Texas, and Washington.
d
Compares the rate of acute pesticide poisoning among agricultural workers for a given year with non-agricultural workers. Cases are identified by participating SENSOR-
Pesticides states and CDPR. All IRRs were significantly elevated (P < 0.0001).
e
Arizona, California, New Mexico, Oregon, and Washington.
f
Florida, Louisiana, and Texas.
g
Michigan, New York.


TABLE III. Incidence Rates by Occupation for 3,271 Acute Pesticide Poisoning Cases in the Agricultural Industry,
1998^2005

Occupation Number Percent FTE estimatea Incidence rateb
Farmworkeröall 2,334 71 3,119,402 74.8
Farmworkerömale 1,620 69 2,625,146 61.7
Farmworkeröfemale 701 30 494,256 141.8
Farmer 89 3 1,852,030 4.8
Farmerömale 80 90 1,510,632 5.3
Farmeröfemale 9 10 341,398 2.6
Processing/packing plant worker 394 12 108,646 362.6
Processing/packing plant workerömale 108 27 21,094 512.0
Processing/packing plant workeröfemale 279 71 87,552 318.7
All other agricultural occupations 454 14 1,018,205 44.6
All other agricultural occupationsömale 381 84 674,521 56.5
All other agricultural occupationsöfemale 65 14 343,684 18.9
Totalc 3,271 100 6,098,283 53.6
Totalömale 2,189 67 4,831,393 45.3
Totalöfemale 1,054 32 1,266,890 83.2
a
FTE, full-time equivalent.
b
Incidence rates per100,000 FTEs. Includes agricultural workers in Arizona, California, Florida, Louisiana, Michigan, New Mexico, New
York, Oregon, Texas, and Washington.
c
Sex was unknown for 28 cases (farmworkers ¼13, processing/packing plant worker ¼ 7, all other agricultural occupations ¼ 8).

100,000 FTEs (2002), driven primarily by the occurrence of workers was highest in the West. The rates in the West were
large California events (Fig. 1). Limiting the analysis to the largely driven by California and Washington State, where the
ﬁve states (California, Florida, New York, Oregon, Texas) rates were 100.8 and 113.0/100,000 FTEs, respectively.
that provided data for all 8 years had little effect on the plot in Information on age was available for 87% of the cases
the Figure 1. By US geographic region, the IR for agricultural (N ¼ 2,857; Table I). The median age was 33 years (range

FIGURE 1. Incidence rates for acute pesticide poisoning cases among agricultural and non-agricultural workers by year,
age15^64 years,1998 ^2005.

890 Calvert et al.

15–64), and over half of the cases were between the ages of methomyl near his body. This led investigators to suspect he
25 and 44 years (N ¼ 1,637, 57%). The IR was highest among had violated regulations by opening water-soluble methomyl
agricultural workers age 18–24 years (76.2/100,000 FTEs; packets and using the potent carbamate insecticide to
Table II). contaminate hot dogs for use as bait to kill coyotes. His
Because agricultural workers employed in the states autopsy found a small amount of methomyl in his gastric
and time periods covered in this study worked 2,173 hr contents but none in his blood. His blood and bile also
per year on average, using FTEs in the denominator produced contained a relatively large concentration of benzothiazole,
rates that were approximately 7% lower compared to when an industrial chemical and a metabolite of cyprodinil
employment counts were used in the denominator. In (a toxicity category III fungicide). The large concentration
contrast, because non-agricultural workers worked 1,935 hr of benzothiazole was suggestive of chronic exposure, or
per year on average, using FTEs in the denominator produced heavy acute exposure at least 24 hr earlier. The medical
rates that were approximately 4% higher compared to when examiner concluded that the cause of death was likely due to
employment counts were used in the denominator. The an interaction between the methomyl and cyprodinil. The
overall average annual IR among agricultural workers using source of the exposure to benzothiazole or cyprodinil was not
employment counts in the denominator was 57.6/100,000 known. Whether the exposures were accidental or intentional
workers, and was 1.33/100,000 FTEs among all non- (i.e., suicidal) could not be distinguished.
agricultural workers combined (IRR ¼ 43.2, 95% CI 41.4,
45.1). Signs and Symptoms
Occupations of the Affected Agricultural Table IV lists the signs and symptoms most often
Workers reported in this cohort. It also provides information on the
health effects among those exposed to the four pesticide
Most of the 3,271 affected agricultural workers were chemical classes most commonly involved in illness.
employed as farmworkers (N ¼ 2,334, 71%; Table III). The
394 affected processing/packing plant workers (12%)
represented a disproportionately large share of people so Pesticides Responsible for Illness
employed, while farmers (N ¼ 89, 3%) seemed less at risk
than hired agricultural workers. Most of the ‘‘other Information on the pesticides responsible for illness is
miscellaneous agricultural workers’’ were employed as pest provided in Tables I, IV and V. Insecticides (alone or in
control operators (N ¼ 255, 56%). Because CPS data for pest combination with other pesticides) were implicated in more
control operators were too sparse (e.g., in 1999 and 2001 no than half of the illnesses (N ¼ 1,761, 54%). Cholinesterase
CPS data were available for this occupation), IRs were not inhibitors (organophosphates and N-methyl carbamates)
calculated for this occupation. were prominent among the insecticides (N ¼ 892, 51%),
Gender was reported for all but 28 (1%) of the cases. particularly chlorpyrifos (N ¼ 190), methamidophos
Males predominated in each occupational category except (N ¼ 130), dimethoate (N ¼ 84), malathion (N ¼ 78), and
processing/packing plant workers. Paradoxically, IRs were diazinon (N ¼ 70). Over half of the cases (N ¼ 1,662, 51%)
higher among female than male farmers and farmworkers, were exposed to toxicity category I pesticides, the most toxic
but higher among male than female processing/packing plant category as deﬁned by EPA (Table I). We found little
workers. Females were less likely than males to be pesticide variation in illness severity by pesticide category.
handlers (females ¼ 8%, males ¼ 45%, P < 0.001). Informa-
tion on race and ethnicity was available for 727 cases (22%). Activity at Time of Exposure
A total of 502 (69%) were Hispanic, 187 (26%) were non-
Hispanic white, 12 (2%) were black, and the remaining 26 Information on activity at time of pesticide exposure
(4%) recorded various other races. was available for 3,203 (98%) of the affected workers. Of
these, 33% (N ¼ 1,068) were pesticide handlers and 67%
Severity and Description of Fatal Case (N ¼ 2,135) were doing routine work not involved with a
pesticide application. Most of the handlers (71%) were
A vast majority of the illnesses were of low severity exposed while making applications. Among the 2,135 doing
(2,848 cases, 87%), while 402 (12%) were of medium routine work, half were exposed to off-target drift of pesticide
severity and 20 (0.6%) were high severity (Table I). One case from a nearby application (N ¼ 1,068), and 35% (N ¼ 744)
was fatal. The fatal case occurred in 1999 and involved a 59- had contact with pesticide residues present on a treated
year-old Hispanic male who was employed as an irrigator and surface (e.g., plant material or treated animal).
farmworker supervisor. He was found dead in an orange Table I lists the targets to which pesticides were applied
grove in California, with packages of hotdogs and packets of in incidents that resulted in human illness. Among the


TABLE IV. Illness Characteristics by Pesticide Chemical Class for 3,271Acute Pesticide Poisoning Cases in the Agricultural Industry,1998^2005

Pesticide chemical classa,b

Alla, Cholinesterase Pyrethroids, Inorganics, Dithiocarbamates,
Signs and symptoms N ¼ 3,271 (%) inhibitors, N ¼ 892 (%) N ¼182 (%) N ¼ À567 (%) N ¼ 512 (%)
Nervous/sensory 1,743 (53) 672 (75) 120 (66) 241 (43) 237 (46)
Headache 1,268 (39) 499 (56) 94 (52) 164 (29) 185 (36)
Dizziness 708 (22) 297 (33) 39 (21) 88 (16) 85 (17)
Muscle weakness 243 (7) 126 (14) 10 (5) 29 (5) 23 (4)
Blurred vision 204 (6) 86 (10) 8 (4) 34(6) 29(6)
Paresthesias 198 (6) 76 (9) 15 (8) 31 (5) 25 (5)
Muscle pain 98 (3) 44 (5) 7 (4) 7 (1) 15 (3)
Diaphoresis 94 (3) 59 (7) 6 (3) 8(1) 7(1)
Salivation 63 (2) 48 (5) 3 (2) 2 (<1) 19 (4)
Fasciculation 47 (1) 32 (4) 3 (2) 3 (1) 3 (1)
Confusion 46 (1) 19 (2) 1 (1) 3 (1) 2 (<1)
Gastrointestinal 1,300 (40) 510 (57) 91 (50) 174 (31) 188 (37)
Nausea 1,063 (33) 438 (49) 74 (41) 131 (23) 152 (30)
Vomiting 582 (18) 261 (29) 39 (21) 73 (13) 91 (18)
Abdominal pain/cramping 371 (11) 161 (18) 15 (8) 34 (6) 75 (15)
Diarrhea 148 (5) 80 (9) 8 (4) 8 (1) 17 (3)
Ocular 1,300 (40) 272 (30) 54 (30) 243 (43) 297 (58)
Irritation/pain/inflammation 1,112 (34) 208 (23) 48 (26) 222 (39) 262 (51)
Lacrimation 443 (14) 92 (10) 14 (8) 51 (9) 166 (32)
Conjunctivitis 80 (2) 8 (1) 5 (3) 23 (4) 1 (<1)
Dermatologic 1,077 (33) 235 (26) 57 (31) 191 (34) 96 (19)
Pruritis 580 (18) 106 (12) 26 (14) 125 (22) 50 (10)
Rash 571 (17) 98 (11) 17 (9) 126 (22) 64 (13)
Erythema 349 (11) 52 (6) 14 (8) 76 (13) 27 (5)
Irritation/pain 321 (10) 81 (9) 34 (19) 48 (8) 23 (4)
Respiratory 1,074 (33) 329 (37) 56 (31) 232 (41) 152 (30)
Upper respiratory pain/irritation 645 (20) 183 (21) 35 (19) 142 (25) 103 (20)
Dyspnea 408 (12) 115 (13) 19 (10) 91 (16) 60 (12)
Cough 278 (9) 67 (8) 5 (3) 75 (13) 32 (6)
Cardiovascular 211 (6) 77 (9) 7 (4) 43 (8) 31 (6)
Chest pain 131 (4) 45 (5) 4 (2) 32 (6) 18 (4)
Tachycardia 33 (1) 17 (2) 1 (1) 2 (<1) 3 (1)
a
More than one sign/symptom may be reported by a case, and therefore the sum of the specific clinical effects may not equal the total number of system effects.
b
Cases may be exposed to more than one chemical class. Columns include cases exposed to the labeled chemical class only as well as those exposed to mixtures containing that
and other chemical classes.

fruit crops, the most common application targets were Factors That Contributed to Pesticide
small fruits (e.g., grapes; N ¼ 529, 46%), tree nuts (N ¼ 181, Exposure
16%), citrus fruits (N ¼ 175, 15%), and pome fruits
(e.g., apples; N ¼ 151, 13%). Among the most common We identified factors that contributed to pesticide
vegetable crop targets were root and tuber vegetables exposure in 1,926 (59%) of the cases (Table VI). The most
(e.g., onions and potatoes; N ¼ 185, 40%), leafy vegetables common factors identified were off-target drift (N ¼ 1,216,
(N ¼ 180, 39%) and fruiting vegetables (e.g., eggplant, 63%), early reentry into a recently treated area (N ¼ 336,
tomatoes, and peppers; N ¼ 48, 10%). Among grain, grass 17%), and use in conflict with the label (N ¼ 319, 17%). In
and fiber crops, the most common pesticide application 992 (30%) cases, no obvious contributory factors could be
targets were cotton (N ¼ 140, 54%), and cereal grains identified (e.g., restricted entry interval was observed but
(N ¼ 61, 23%). worker still became ill; wore all required PPE but still

892 Calvert et al.

TABLE V. Fifteen Most Common Active Ingredients for 3,271Acute Pesticide Poisoning Cases in the Agricultural Industry by Severity Category,1998^2005

Functional class High severity/
Active ingredient (chemical class) fatal Moderate severity Low severity N (%)a
Sulfur Insecticide and fungicide (inorganic) 2 45 421 468 (14)
Metam-sodium Fumigant (dithiocarbamate) 1 5 279 285 (9)
Glyphosate Herbicide (phosphonate) 3 25 223 251 (8)
Mancozeb Fungicide (dithiocarbamate) 1 17 184 202 (6)
Chlorpyrifos Insecticide (organophosphate/cholinesterase inhibitor) 0 33 157 190 (6)
Sodium hypochlorite Disinfectant (halogen) 2 35 149 186 (6)
Methamidophos Insecticide (organophosphate/cholinesterase inhibitor) 1 10 119 130 (4)
Abamectin Insecticide (microbial) 0 18 108 126 (4)
Imidacloprid Insecticide (neonicotinoid) 4 5 104 113 (3)
Methomyl Insecticide (N-methyl carbamate/cholinesterase inhibitor) 1 7 101 109 (3)
Myclobutanil Fungicide (triazole) 1 11 97 109 (3)
Propargite Insecticide (sulfite ester, inhibits oxidative phosphorylation) 0 21 77 98 (3)
Spinosad Insecticide (derived from Saccharopolyspora spinosa) 1 10 85 96 (3)
Methyl bromide Fumigant (halocarbon) 2 29 60 91 (3)
Dimethoate Insecticide (organophosphate/cholinesterase inhibitor) 4 6 74 84 (3)
a
Percentages do not sum to100, as not all cases are included in this table.

TABLE VI. FactorsThat Contributed to Pesticide Exposure and/or Illness for 3,271Acute Pesticide Poisoning Cases in the Agricultural Industry by Severity
Category,1998^2005

Exposure/illness factora Medium or higher severity, N (%) Low severity, N Total, N (%)
All factors combined 219 (11) 1,707 1,926 (59)
Drift 118 (10) 1,098 1,216 (37)
Early reentry 41 (12) 295 336 (10)
Use in conflict with label 40 (13) 279 319 (10)
Failure to use required equipment 19 (12) 139 158 (5)
Oral notification of pesticide application not provided 9 (6) 143 152 (5)
PPE not worn 19 (16) 98 117 (4)
Training not provided or inadequate (excludes applicators) 30 (29)c 75 105 (3)
Hazard communication or other OSHAviolation 9 (9) 86 95 (3)
Transport for care not provided 5 (6) 84 89 (3)
Application site not posted/notification posters incorrect 3(4) 81 84 (3)
Decontamination facilities inadequate 16 (26)c 46 62 (2)
Unsafe equipment/failure 8 (14) 49 57 (2)
Inadequate record keeping 7 (20) 28 35 (1)
Worker not told of health effects caused by pesticides 0 (0) 32 32 (1)
Person in treated area during application 2 (10) 19 21 (1)
Unspecified worker protection standard violation 3 (15) 17 20 (1)
PPE in poor repair 3 (23) 10 13 (<1)
FIFRA-other and unspecifiedb 6 (27)c 16 22 (1)
None identified 126 (13) 866 992 (30)
Unknown 78 (22) 275 353 (11)
a
One factor was identified for 1,279 cases. Two or more factors were identified for 647 cases.
b
Includes situations when a licensed applicator was not on site(N ¼1) and when an applicator was not properly trained or supervised (N ¼ 9).
c
The proportion with medium or higher severity among cases with the factor of interest was significantly greater than the proportion with medium or higher severity in all other
cases (P < 0.05). Those with insufficient information to identify factors (i.e., unknown category) were excluded from this analysis.


became ill; all pesticide label requirements appeared to have remaining cases including health care professionals, employ-
been followed). Compared to cases where no obvious ers, worker representatives (union, legal services), and self-
contributory factors could be identified, identification of a report. The specific number of cases reported by these other
contributory factor was not found to be significantly asso- sources is unavailable. Females were more likely to have
ciated with severity of illness (P ¼ 0.33). For 353 (11%) been reported by one of these other sources (females ¼ 41%,
cases, insufficient information was available to identify males ¼ 23%, P < 0.01), and less likely to be identified by
factors that may have contributed to the pesticide exposure. workers’ compensation (females ¼ 25%, males ¼ 39%,
Among the 2,367 cases with personal protection P < 0.01) or poison control centers (females ¼ 8%, mal-
equipment (PPE) usage information, 1,157 (49%) wore es ¼ 15%, P < 0.01). Females and males were equally likely
PPE (Table VII). Women were far less likely to wear PPE to have been reported by other government agencies (28%).
(females ¼ 27%, males ¼ 40%, P < 0.01). Pesticide handlers
were more likely to use PPE (65% overall, 66% among men DISCUSSION
and 51% among women) compared to non-handlers (21%
overall, 18% among men and 26% among women). Those It is important to conduct surveillance of acute occupa-
exposed to toxicity category II pesticides were more likely to tional pesticide poisoning to determine whether policies,
wear PPE (61%) then those exposed to toxicity category I or practices and regulations are effective in preventing hazard-
III /IV pesticides (53% and 54%, respectively). Table VII ous pesticide exposures. National estimates of hospitalized
also provides information on the health effects experienced pesticide poisonings in the 1970s and early 1980s, including
by those who used each type of protective equipment. agricultural workers, are available [Keefe et al., 1985, 1990];
Compared to those who used no protective equipment or had but to the best of our knowledge this is the first detailed multi-
unknown information on its use, those who used protective state assessment of both hospitalized and non-hospitalized
equipment were less likely to have health effects involving acute pesticide poisoning among agricultural workers. Our
the nervous, gastrointestinal and respiratory systems, but findings indicate that despite strengthening of the WPS in
were more likely to have ocular and dermatologic health 1995, agricultural workers continue to have an elevated risk
effects. for acute pesticide poisoning. The pesticide poisoning
incidence among US agricultural workers was found to be
Report Source 39 times higher than the IR found in all other industries
combined.
Case reports were received from many different sources.
The three leading sources of case reports were workers’ Improvement Compared to the 1980s
compensation (N ¼ 1,109, 34%), other government agencies
(e.g., county health departments and the state department of Although there was not a clear trend in the rates of
agriculture; N ¼ 901, 28%), and poison control centers poisoning during the time period that we studied, there is
(N ¼ 407, 12%). A variety of sources accounted for the evidence to suggest that the counts of pesticide poisoning

TABLE VII. Illness Characteristics byType of Protective Equipment Used1998^2005

Of workers who used protective equipment, the number (%) who
had signs/symptoms involving these organs/systems
Type of protective Number that used
equipment protective equipment Nervous/sensory Gastro-intestinal Ocular Respiratory Dermatologic
Any PPE 1,157 557 (48) 404 (35) 472 (41) 301 (28) 437 (38)
Air-purifying respirator 261 121 (46) 91 (35) 92 (35) 49 (19) 94 (36)
Dust mask 40 16 (40) 12 (30) 19 (48) 18 (45) 15 (38)
Chemical resistant gloves 700 306 (44) 208 (30) 293 (42) 150 (21) 254 (36)
Chemical resistant clothing 542 230 (42) 177 (33) 214 (39) 105 (19) 213 (39)
Chemical resistant boots 367 170 (46) 110 (30) 152 (41) 63 (17) 127 (35)
Cloth/leather gloves 298 192 (64) 138 (46) 92 (31) 114 (38) 140 (47)
Goggles/eye protection 488 193 (40) 126 (26) 198 (41) 80 (16) 196 (40)
Engineering controlsa 100 48 (48) 34 (34) 23 (23) 20 (20) 41 (41)
No or unknown PPE useb 0 1,186 (56) 896 (42) 828 (39) 773 (37) 640 (30)
a
Engineering controls included such things as enclosed tractor cabs or closed mixing/loading systems.
b
A total of 2,114 individuals used no protective equipment or had unknown information on use of protective equipment.

894 Calvert et al.

cases among agricultural workers have decreased since the whom 43.4% reported getting sick or having a reaction. That
1980s. Mehler et al. [1992] reported an annual average of 723 is, 1.4% of US crop workers attributed health effects such as
cases of pesticide illness or injury in California arising from skin problems (59%), eye problems (24%), nausea/vomiting
agricultural establishments from 1982 to 1990. In contrast, (30%), headache (26%), and numbness/tingling (12%) to
the California surveillance programs reported an average of pesticide exposure during the preceding 12 months. In a
315 cases per year from 1998 to 2005. The numbers are not separate NAWS question, 0.6% of all US crop workers
entirely comparable. Mehler characterized poisoning cases reported that in the last 12 months they had ‘‘received
as being agricultural if the poisoned subject was a worker and medical attention by a doctor or nurse due to pesticide
the exposure arose from an agricultural establishment. Her exposure.’’ To our knowledge, neither these nor similar
definition included non-agricultural workers (e.g., truck questions to assess the incidence of pesticide poisoning
drivers, construction workers, school employees), while we were included in NAWS surveys before or after 1999. In
included only workers employed in agriculture. comparison, we found an average annual acute occupational
pesticide poisoning IR of 0.05% among all agricultural
True Incidence Remains Uncertain: workers, and 0.07% among farmworkers.
Comparison With Data From the Acute pesticide poisoning IR were also assessed among
Bureau of Labor Statistics, the farmers participating in the Agricultural Health Study
National Agricultural Workers Survey, (AHS), a prospective cohort study sponsored by the National
and the Agricultural Health Study Institutes of Health (i.e., the National Cancer Institute and the
National Institute of Environmental Health Sciences) and
In order to put our findings in perspective, acute EPA [AHS, 2007]. The AHS cohort consists of 52,395
pesticide poisoning annually accounts for a small percentage farmers, 32,347 spouses of these farmers, and 4,916
of the total occupational illnesses experienced by agricultural commercial pesticide applicators residing in Iowa or North
workers. The 2005 Bureau of Labor Statistics (BLS) survey Carolina. In a nested case–control analysis involving 16,416
of illnesses and injuries (SOII), which excludes agricultural farmers/pesticide applicators who were interviewed by
production establishments with 10 employees or fewer, telephone in 1999–2000, 54 (0.33%) reported ‘‘an incident
reports an annual injury rate of 5.7% and an annual illness with fertilizers, weed killers, or other pesticides that caused
rate of 0.4% (3% of which involve poisonings) among an unusually high personal exposure’’ in the previous
workers in farming, forestry and fishing [Myers, 2007]. Most 12 months that resulted in physical symptoms [Bell et al.,
of these illnesses consisted of dermatitis, respiratory con- 2006]. Among these 54 individuals, only 7 (13%, or .04% of
ditions, and other conditions not specified (e.g., musculoske- the entire subcohort) sought medical care. In contrast,
letal conditions arising from cumulative trauma). We report a we found an average annual acute occupational pesticide
rate of pesticide poisoning five times higher than the SOII rate poisoning IR of 0.005% among farmers. However, as was
for all poisonings. This may indicate that pesticide poisonings observed in our study, the findings from AHS and NAWS
are concentrated among the small establishments excluded suggest that the risk of pesticide poisoning is lower among
from SOII, that under-reporting to SOII is more extreme than farmers compared to farmworkers.
SENSOR-pesticide under-ascertainment, or that SENSOR- The true incidence of pesticide poisoning among
pesticide classification standards accept a large number of agricultural workers remains uncertain. Our findings (51/
cases that BLS does not count. 100,000) fall between the low SOII estimate (less than 10/
Data from the National Agricultural Workers Survey 100,000) and the high rates elicited by NAWS interviews
(NAWS), by contrast, suggest an incidence of acute pesticide (1,400/100,000 symptomatic, 600/100,000 sought medical
poisoning among agricultural workers [US Department of care).
Labor, 2004] an order of magnitude greater than that found in
this study. NAWS is a nationally representative annual survey Limitations
of US crop workers conducted by the US Department of
Labor. In 1999, NAWS included questions to determine if One of this study’s major limitations is that under-
crop workers were poisoned by pesticides. This information reporting compromises, to varying degrees, all the surveil-
was collected in two parts. First NAWS asked the crop worker lance systems that provided the data for this analysis. Factors
if they were exposed to pesticides by ‘‘having them sprayed that contribute to under-reporting include: affected people
or blown on you,’’ ‘‘spilled on you,’’ or ‘‘when cleaning not seeking care, or consulting care providers outside the
or repairing containers or equipment used for applying or jurisdiction of surveillance programs; misdiagnosis of this
storing pesticides.’’ NAWS then asked if the crop worker uncommon condition; and health care provider neglect of
became ‘‘sick or [had] any reaction because of this incident.’’ legal requirements to report. The rates provided should be
Our analysis of these data found that 3.2% of crop workers considered low estimates of the magnitude of acute pesticide
acknowledged exposure during the previous 12 months, of poisoning among agricultural workers.


Disproportionate numbers of agricultural workers prob- several reasons, including the transient employment of many
ably are deterred from seeking health care by lack of health seasonal and migrant farmworkers, migration into and out of
insurance [US Department of Labor, 2005], unfamiliarity with the United States in a manner that is not entirely predictable,
workers’ compensation benefits or inability to qualify for and the tendency of many farmworkers to avoid government
them, and fear of job loss if they miss time from work to seek contact [Rust, 1990]. Our agricultural worker population
health care [Das et al., 2001; Arcury and Quandt, 2007], estimates were derived from the CPS, which is conducted by
as well as concerns related to immigration status. Similarly, the BLS and the United States Census Bureau. The CPS goes
a variety of interrelated problems may lead health care to great lengths to capture reliable workforce data [Bureau of
professionals to misdiagnose acute pesticide poisoning. Labor Statistics, 2002]. Nevertheless, a population seeking to
Health professionals rarely receive much training in environ- escape detection could well be under-counted, leading to
mental toxicology generally or on pesticide poisoning inflated apparent rates of illness/injury. Finally, illness rates
specifically [Schenk et al., 1996]. Consequently, they may for those known to have occupational pesticide exposure are
not collect a pesticide exposure history, which is necessary to not available because the numbers or workers exposed to
make a diagnosis of acute pesticide poisoning [Balbus et al., pesticides are unknown.
2006]. Pesticide poisoning is relatively rare in developed
countries, and its signs and symptoms often resemble those of Reasons for Higher Poisoning Rates in
more common conditions, which may be diagnosed preferen- Western States
tially. The difficulty and delays of receiving reimbursement
through workers’ compensation may also bias health care Rates of both agricultural and non-agricultural acute
providers against diagnosing and reporting pesticide poison- pesticide poisoning are higher in the western states as
ing. Even among those cases correctly diagnosed, some cases compared to the south and eastern regions of the United
may escape report to public health authorities through States. It is credible that labor-intensive Western agriculture
ignorance of the requirement (despite the fact that 30 states may impose excess risk for acute pesticide poisoning illness,
have a mandatory reporting system of occupationally related but it is also important to note that California (especially the
pesticide poisoning [Calvert et al., 2001]) or because the CDPR program) and Washington have well established,
health care professionals fear that their patients may be longstanding and experienced state-based surveillance pro-
subject to retaliation. Other cases may go unreported because grams with higher staffing levels compared to other states
many farmworkers immigrate from Mexico [US Department participating in the SENSOR-Pesticides program [Calvert
of Labor, 2005], and some poisoned Mexican farmworkers et al., 2004]. In addition, these states were much more likely
may prefer to visit physicians in Mexico where cultural and to be notified about cases through the state workers’
linguistic barriers are removed and fees are lower [US EPA, compensation system (in Washington State 76% of the cases
1992; Arcury and Quandt, 2007]. Our state-based surveil- were so identified; in California 34% of reports were
lance partners received only nine reports of pesticide provided by physicians to a workers’ compensation insur-
poisoning cases managed outside the US. ance carrier). In contrast, only two other states identified
Another limitation in our study was that information was cases through their state workers’ compensation system:
incomplete for some reported cases. Most cases lacked Oregon and Texas (7% and 4%, respectively, of cases in these
information on race and ethnicity. Missing information states were so identified).
could lead to misclassification of severity, if not all signs and
symptoms were reported, or to inappropriate exclusion of the Higher Poisoning Rates Among Female
case. More detailed information on the affected worker’s Agricultural Workers
activities, pesticide exposures and health effects might have
increased our case totals. Some cases in this report may be Female agricultural workers experienced nearly twice the
false-positives, with compatible symptoms that are coinci- risk of pesticide poisoning of male agricultural workers (Table
dental with but not caused by pesticide exposure. Finally, III), a finding that was quite unexpected. Before indulging in
information on factors that contributed to illness was speculation about possible differences in susceptibility, risk of
identified in only 1,926 (59%) cases. In many cases exposure, or rate of ascertainment, we plan to perform more
a timely and definitive investigation into the factors res- detailed analyses by geographic region, activity at time of
ponsible for exposure and illness was not possible due to exposure, pesticide, protective equipment, and severity.
insufficient investigatory resources and/or because of tardy
notification of the exposure to state authorities. Higher Poisoning Rates Among
Rates of pesticide poisoning may also be distorted by Processing/Packing Plant Workers
inaccuracy in estimates of population at risk. The size of the
agricultural worker population, including farmworkers and Processing/packing plant workers were found to have
processing/packing plant workers, is difficult to estimate for the highest acute pesticide poisoning IR compared to all other

896 Calvert et al.

agricultural occupations. Many farms are increasing the important improvements in safety. Measures to minimize
amount of food processing that is performed on site [National drift (including equipment specifications, establishment of
Research Council and Institute of Medicine, 2008]. This is buffer zones, and limitations on maximum wind speed
due to a variety of factors, including the advent of new conditions during an application) seem likely to provide the
technology, quest for improved quality control and freshness, greatest benefit. Our finding that 992 (30%) cases had no
and a desire to increase profit. The types of food processing obvious factors contributing to exposure suggests that
activities performed on farms include cleaning, sorting, pesticide regulations and label requirements may also need
packing, and cooling/freezing. Among the 394 poisoned to be enhanced. Additionally, reduced-risk pest control
processing/packing plant workers, the pesticides most measures such as integrated pest management should be
commonly responsible for illness were fumigants (N ¼ 151, adopted, which can achieve reductions in pesticide exposure
38%), disinfectants (N ¼ 151, 38%), and insecticides and misuse [National Research Council, 2000]. The high
(N ¼ 73, 19%). The fumigant exposures were commonly poisoning rates observed among processing/packing plant
related to drift from a nearby field (N ¼ 111), the disinfectant workers and the increased amount of food processing
exposures where commonly related to malfunction of performed on farms suggests that processing/packing plant
disinfecting equipment (N ¼ 74) and cleaning produce in workers should be covered under the WPS.
disinfectant solutions (N ¼ 31) and the insecticide exposures Given the limitations in this analysis, improved state-
were commonly related to pesticide residue present on the based surveillance programs for pesticide-related illness are
produce (N ¼ 37). Because the WPS only covers workers also needed. A comprehensive system needs to address the
involved in the production of agricultural plants, processing/ limitations described above including: agricultural workers
packing plant workers may not covered by WPS. and health care providers need to recognize the pesticide-
relatedness of the illness; disincentives to receiving health
Chronic Health Effects Associated With care, including lack of health insurance, must be overcome; the
Acute Pesticide Poisoning costs of evaluation and treatment of acute occupational
pesticide poisoning should be paid for by workers’ compen-
In addition to acute morbidity with it attendant costs in sation; health care providers need to make timely reports to
health care resources, and time lost from work and normal pesticide poisoning surveillance systems; and, surveillance
daily activities, acute pesticide poisoning is also associated systems need to optimize access to and use of workers’
with chronic adverse health sequelae. For example organo- compensation data, poison control center data and data from
phosphate poisoning has been found to be associated with other state agencies with jurisdiction over pesticides.
deficits in neurobehavioral and neurosensory function
[Steenland et al., 1994]. In addition, the Agricultural Health CONCLUSION
Study found that those who sought medical care for pesticide
poisoning or who experienced an incident involving Agricultural workers are at increased risk of acute
‘‘unusually high’’ pesticide exposure had an increased risk pesticide poisoning in comparison to non-agricultural work-
for chronic neurologic symptoms (Kamel et al. 2005). These ers, particularly through drift, early reentry into a treated
‘‘unusually high’’ pesticide exposures, which are labeled by area, and use in conflict with the label. The IR was almost
the authors as high pesticide exposure events (HPEE), result twofold higher in female agricultural workers compared to
in acute symptomatic illness about 50% of the time [Bell males. In addition to acute intoxication, pesticide poisoning
et al., 2006]. Those who ever experienced an HPEE also had may also lead to chronic adverse health sequelae. Improved
an increased risk for farmers lung (Hoppin et al. 2006). A compliance with and stringent enforcement of laws and
non-significant elevated risk for prostate cancer was regulations regarding pesticide applications are needed.
observed among those who had ever experienced an HPEE Alternative pest control measures such as integrated pest
(odds ratio ¼ 1.11, 95% CI ¼ 0.8, 1.6) [Alavanja et al., 2003]. management reduce the use of pesticides and therefore the
To our knowledge, the AHS has not published findings on any potential for adverse health effects.
other associations between HPEE and cancer.
ACKNOWLEDGMENTS
Recommendations
Funding to support this study was provided by NIOSH,
The most common factors that contributed to pesticide EPA, and the state agencies that contributed data.
exposure included off-target drift, early reentry into a treated
area, and use in conflict with the pesticide label. These REFERENCES
findings and the observations of other investigators [Arcury
Agricultural Health Study. Study and cohort descriptions. 2007.
et al., 2001] suggest that improved compliance with and Available: http://www.aghealth.org/study.html. [Accessed 7 February
enforcement of existing pesticide regulations could achieve 2008].


Alavanja MCR, Samanic C, Dosemeci M, Lubin J, Tarone R, Lynch CF, Keefe TJ, Savage EP, Wheeler HW. 1990. Third national study of
Knott C, Thomas K, Hoppin JA, Barker J, Coble J, Sandler DP, Blair A. hospitalized pesticide poisonings in the United States, 1977–1982. Fort
2003. Use of agricultural pesticides and prostate cancer risk in the Collins, CO: Colorado State University.
Agricultural Health Study cohort. Am J Epidemiol 157:800–814.
Kiely T, Donaldson D, Grube A. Pesticides industry sales and usage:
Arcury TA, Quandt SA. 2007. Delivery of health services to migrant and 2000 and 2001 market estimates. EPA-733-R-04-001. Washington, DC:
seasonal farmworkers. Annu Rev Public Health 28:345–363. Office of Pesticide Programs, U.S. Environmental Protection Agency.
2004. Available: http://www.epa.gov/oppbead1/pestsales/01pestsales/
Arcury TA, Quandt SA, Cravey AJ, Elmore RC, Russell GB. 2001. market_estimates2001.pdf [Accessed 7 February 2008].
Farmworker reports of pesticide safety and sanitation in the work
environment. Am J Ind Med 39:487–498. Linaker C, Smedley J. Respiratory illness in agricultural workers.
Occup Med 2002. 52:451–459.
Balbus JM, Harvey CE, McCurdy LE. 2006. Educational needs
assessment for pediatric health care providers on pesticide toxicity. Litchfield MH. 2005. Estimates of acute pesticide poisoning in
J Agromedicine 11:27–38. agricultural workers in less developed countries. Toxicol Rev 24(4):
271–278.
Bell EM, Sandler DP, Alavanja MC. 2006. High pesticide exposure
events among farmers and spouses enrolled in the Agricultural Health Mehler LN, O’Malley MA, Krieger RI. 1992. Acute pesticide morbidity
Study. J Agric Saf Health 12:101–116. and mortality: California. Rev Environ Contam Toxicol 129:51–66.
Bureau of Labor and Statistics (US). 2007. Current population survey Myers JR. 2007. National occupational surveillance for the agriculture,
1998–2004 microdata files. Washington: U.S. Department of Labor, forestry and fishing industry sector. Unpublished poster presented at the
Bureau of Labor Statistics. 2007 Council of State and Territorial Epidemiologists Annual Confer-
ence. Atlantic City, New Jersey. June 2007.
Bureau of Labor Statistics (US). Current population survey design and
methodology. Technical paper 63RV. Washington: U.S. Department of National Research Council. 2000. The future role of pesticides in US
Labor, Bureau of Labor Statistics. 2002. Available: http://www.cen- agriculture. Washington DC: National Research Council.
sus.gov/prod/2002pubs/tp63rv.pdf [Accessed 7 February 2008].
National Research Council and Institute of Medicine. Agriculture,
Calvert GM, Sanderson WT, Barnett M, Blondell JM, Mehler LN. 2001. Forestry, and Fishing Research at NIOSH. Committee to Review the
Surveillance of pesticide-related illness and injury in humans. In: NIOSH Agriculture, Forestry, and Fishing Research Program. Rpt No.
Krieger R, editor. Handbook of pesticide toxicology 2nd edition. San 3, Reviews of Research Programs of the National Institute for
Diego, CA: Academic Press. p 603–641. Occupational Safety and Health. Washington, D.C.: The National
Academies Press. 2008. Available: http://www.nap.edu/catalog.
Calvert GM, Plate DK, Das R, Rosales R, Shafey O, Thomsen C, Male php?record_id¼12088 [Accessed 1 May 2008].
D, Beckman J, Arvizu E, Lackovic M. 2004. Acute occupational
pesticide-related illness in the US, 1998–1999: Surveillance findings O’Malley M, Barry T, Ibarra M, Verder-Carlos M, Mehler L. 2005.
from the SENSOR-Pesticides program. Am J Ind Med 45:14–23. Illnesses related to shank application of metam-sodium, Arvin,
California, July 2002. J Agromedicine 10:27–42.
CDC. Case definition for acute pesticide-related illness and injury cases
reportable to the national public health surveillance system. Cincinnati, Reeves M, Schafer KS. 2003. Greater risks, fewer rights: U.S.
OH: Centers for Disease Control and Prevention. National Institute for farmworkers and pesticides. Int J Occup Environ Health 9:30–39.
Occupational Safety and Health. 2001a. Available: http://www.cdc.gov/
niosh/topics/pesticides/pdfs/casedef2003_revapr2005.pdf [Accessed 7 Rothman KJ. 1986. Modern epidemiology. Boston: Little, Brown.
February 2008]. p 164–172.

CDC. Severity index for use in state-based surveillance of acute- Ruser JW. 1998. Denominator choice in the calculation of workplace
pesticide related illness and injury. Cincinnati, OH: Centers for Disease fatality rates. Am J Ind Med 33:151–156.
Control and Prevention. National Institute for Occupational Safety and Rust GS. 1990. Health status of migrant farmworkers: A literature
Health. 2001b. Available: IR http://www.cdc.gov/niosh/topics/pesti- review and commentary. Am J Public Health 80:1213–1217.
cides/pdfs/pest-sevindexv6.pdf [Accessed 7 February 2008].
SAS Institute, Inc. 2003. SAS: Version 9.1 for Windows. Cary, North
CDPR. Preventing pesticide illness: California’s pesticide illness Carolina: SAS Institute Inc.
surveillance program. Sacramento, CA: California Environmental
Protection Agency. Department of Pesticide Regulation. 2006. http:// Schenk M, Popp SM, Neale AV, Demers RY. 1996. Environmental
www.cdpr.ca.gov/docs/whs/pisp/brochure.pdf [Accessed 7 February medicine content in medical school curricula. Acad Med 71:499–
2008]. 501.
Das R, Steege A, Baron S, Beckman J, Harrison R. 2001. Pesticide- Steenland K, Jenkins B, Ames RG, O’Malley M, Chrislip D, Russo J.
related illness among migrant farm workers in the United States. Int J 1994. Chronic neurologic sequelae to organophosphate pesticide
Occup Environ Health 7:303–312. poisoning. Am J Public Health 84:731–736.
Hoppin JA, Umbach DM, Kullman GJ, Henneberger PK, London SJ, Tak S, Calvert GM. 2008. Hearing difficulty attributable to employment
Alavanja MCR, Sandler DP. 2006. Pesticides and other agricultural by industry and occupation: An analysis of the National Health
factors associated with self-reported farmer’s lung among farm Interview Survey—United States, 1997 to 2003. J Occup Environ Med
residents in the Agricultural Health Study. Occup Environ Med 50:46–56.
64:334–342.
U.S. Bureau of the Census. 1992. 1990 census of population and
Kamel F, Engel L, Gladen BC, Hoppin JA, Alavanja MCR, Sandler DP. housing. Alphabetical index of industries and occupations. Washington,
2005. Neurologic symptoms in licensed private pesticide applicators in DC: U.S. Department of Commerce, Bureau of the Census.
the agricultural health study. Environ Health Perspect 113:877–882.
U.S. Census Bureau. Industry and Occupation 2002. Washington, D.C.:
Keefe TJ, Savage EP, Munn S, Wheeler HW. 1985. Evaluation of United States Census Bureau. 2005. Available: http://www.census.
epidemiologic factors from two national studies of hospitalized gov/hhes/www/ioindex/ioindex02/view02.html [Accessed 7 February
pesticide poisonings, USA. Fort Collins, CO: Colorado State University. 2008].

898 Calvert et al.

US Department of Labor. The National Agricultural Workers Survey. US EPA. General Provisions. Precautionary statements for physical and
Washington, D.C.: US Department of Labor; 2004. Available: http:// chemical hazards. 40 CFR 156.78. Washington, D.C. U.S. Environ-
www.doleta.gov/agworker/naws.cfm [Accessed 7 February 2008]. mental Protection Agency; 1975. Available: http://a257.g.akamaitech.
net/7/257/2422/08aug20051500/edocket.access.gpo.gov/cfr_2005/
US Department of Labor. Monetary bonuses and insurance beneﬁts. In: julqtr/pdf/40cfr156.70.pdf [Accessed 7 February 2008].
Findings from the National Agricultural Workers Survey (NAWS)
2001–2002. A demographic and employment proﬁle of United States US EPA. Regulatory impact analysis of worker protection standard for
farm workers. Research report No. 9. Washington, D.C.:US Department agricultural pesticides. Washington, D.C. U.S. Environmental Protec-
of Labor; 2005. Available: http://www.doleta.gov/agworker/report9/ tion Agency; 1992. Available: http://www.aei-brookings.org/admin/
toc.cfm [Accessed 7 February 2008]. authorpdfs/page.php?id¼821 [Accessed 7 February 2008].

Acute Pesticide Poisoning Among Farm Workers

More Related Content

What's hot (10)

Similar to Acute Pesticide Poisoning Among Farm Workers (20)

More from Z3P (20)

Recently uploaded (20)

Acute Pesticide Poisoning Among Farm Workers