	

	UNITED STATES ENVIRONMENTAL PROTECTION AGENCY

	WASHINGTON, D.C.  20460

OFFICE OF PREVENTION, PESTICIDES AND TOXIC SUBSTANCES

	

April 10, 2007

MEMORANDUM

SUBJECT:	Review of Fenoxaprop Incident Reports

		DP Barcode D338156 Chemical #128701

FROM:	Monica Hawkins, M.P.H., Environmental Health Scientist

		Chemistry and Exposure Branch

		Health Effects Division (7509P)

		Hans Allender, Ph.D., Statistician

		Chemistry and Exposure Branch

		Health Effects Division (7509P)

THRU:	David J. Miller, Chief

		Chemistry and Exposure Branch 

		Health Effects Division (7509P)

TO:		Kylie Rothwell, Chemical Review Manager

		Special Review and Reregistration Division (7508P)

	

BACKGROUND

			

To generate this review, the Office of Pesticides (OPP) consulted the
following databases for the poisoning incidents on the active ingredient
Fenoxaprop. The purpose of the database searches is to identify
potential patterns of the extent and severity of the health effects
attributed to Fenoxaprop exposure.

1)  Poison Control Centers – OPP purchases American Association of
Poison Control Center (AAPCC) data covering the years 1993 through 2005
for all reported pesticides incidents. The last acquisition of data
covering 2004 and 2005 took place in late 2006. Most of the national
Poison Control Centers (PCCs) participate in a national data collection
system, previously called The Toxic Exposure Surveillance System (TESS).
PCC reports includes 61 centers at hospital and universities that cover
100% of the United States population. Datasets incorporate an extensive
quality assurance process including a standardized computer-based data
collection. Some PCCs cover multiple States; others cover a metropolitan
area, or an entire State. PCCs provide telephone consultation for
individual consumers and health care providers on suspected poisonings,
involving drugs, household products, pesticides, etc.; data collection
to aid pesticide regulatory programs is secondary to this service. An
elaborate, computer assisted protocol is followed by specially trained
medical staff, and reports assess confirmed, possible, probable, or
definite pesticide exposures. Approximately 50% of base reports include
medical follow-up and clinical outcome are determined, e.g., cases
resolved by phone consultation with PCC specialist, patients seen in
health care facility, patients hospitalized, and patients admitted to
Intensive Care Unit (ICU) and in some cases death. PCCs receive calls 24
hours a day throughout the entire year, and they handle non-occupational
as well as occupational calls. Discussions are underway to improve
collection of OPP-specific occupational data. Because mainly consumers
initiate PCC calls, this database is not currently a complete source of
occupational poisoning incidents.

2)  OPP Incident Data System (IDS) This database includes reports of
incidents from various sources, including mandatory Federal Insecticide,
Fungicide and Rodenticide Act (FIFRA) Section 6 (a) (2) reports from
registrants, other federal and state health and environmental agencies
and individual consumers.  The IDS has been in existence since 1992. 
Reports submitted to the Incident Data System represent anecdotal
reports or allegations only, unless otherwise stated in this report. 
Typically, OPP does not draw conclusions implicating the pesticide as a
cause of any of the reported health effects.  Nevertheless, sometimes
with enough cases and/or documentation, patterns and risk mitigation
measures may be suggested.

3) National Institute of Occupational Safety and Health Sentinel Event
Notification System for Occupational Risks (NIOSH/SENSOR) performs
standardized surveillance in twelve states from 1998 through 2006. 
States included in this reporting system are Arizona, California,
Florida, Iowa, Louisiana, Michigan, North Carolina, New York, New
Mexico, Oregon, Texas, and Washington. A number of other States can and
do report, periodically and on request for special circumstances,
because they participate in the NIOSH/ SENSOR network as part of the
Council of State and Territorial Epidemiologists (CSTE).  Reporting is
variable from state-to-state because of the dissimilar cooperation from
different sources of reporting (e.g., workers compensation, Poison
Control Centers, emergency departments and hospitals, enforcement
investigations, private physicians, etc.).  Therefore, OPP does not
characterize these reports as estimating the total magnitude of
poisoning for a State.  The focus of OSHA/SENSOR database is on
occupationally related cases. However, the information collected on each
case is standardized and categorized according to the certainty of the
information collected and the severity of the case. Detailed exposure
circumstances may be available if the State provided follow-up case
reports, as in serious cases ending in death.

The purpose of this review is to summarize available pesticide poisoning
incident data for the active ingredient Fenoxaprop. This report contains
3 sections: 

1- Cases reported in the Poison Control Center Database from 1993 to
2005

2- Cases reported in the Incident Data System from 1999 to the present,
and

3- Cases reported in the NIOSH system from 1998 to 2003.

1.	  SEQ CHAPTER \h \r 1 Poison Control Center Data – 1993-2005

This section discusses results from the Poison Control Center’s Toxic
Exposure Surveillance System (TESS) from the years 1993 through 2005 and
reflects data stratified by population: occupational, non-occupational,
and children. The children class is five years of age or less; this
definition includes children about to become six years old, or up to 72
months old. Cases involving exposures to multiple products and cases
with unrelated medical outcome are excluded.  Also excluded are
intentional exposures, i.e., suicide attempts. The tables incorporated
in this section contain summaries of acute pesticide poisoning incidents
resulting from exposure to Fenoxaprop.

The following tables evaluate the frequency of poisoning incidents for
Fenoxaprop compared to the composite of all pesticides for which the PCC
received a non-excluded incident report.  The tables report the
frequency of events by health severity category (i.e.; all symptoms,
moderate, and major) and by level of health care received.  A
comparative ratio provides a simple measure of the relative frequency of
reported health effects by severity category.  Knowledge of the ratios
of symptoms for a single chemical (or a group of chemicals) provides a
relative measure of the public health impact of the acute pesticide
events.  In addition, a Likelihood Ratio test shows whether the compound
under study is significantly different from the average of all other
pesticides.  An (s) next to the ratio indicates the proportions are
significantly different.   

For a detailed explanation of the tables, see Attachment 2.

Occupational Cases.  Only five cases of exposure to Fenoxaprop were
reported for the occupational class in the 13-year span of PCC data
available. Only three of these cases presented minor symptoms with no
case presenting moderate or major symptoms. 

Non-Occupational cases.  The following table 1 summarizes this class.

Table 1. Relative Frequency of Acute Poisoning Incidents by Symptom
Severity: Non-Occupational Cases 

	Severity of outcome

(outcome determined)	Total

exposed	Health care 

provided

Denominator numbers	6	14	2

Measures	SYM1	MOD2	MAJ3	HCF4	HOSP5	ICU6

Numerator numbers	4	2	1	2	0	0

Fenoxaprop percents	66.67%	33.33%	16.67%	14.29%	0.00%	0.00%

All Pesticides percents	64.72%	10.61%	0.45%	15.45%	7.94%	3.02%

Ratio of  Fenoxaprop

/All pesticides	1.03 

P =0.92	3.14

P = 0.07	37.04(S)

P = 0.00 (*)	0.92

P = 0.90	0.00

P = 0.67	0.00

P =0.80



1) Percent of cases with ANY symptom in relation to the total of cases
followed; 2) Percent with moderate symptoms in relation to the total of
cases followed; 3) Percent with major or fatal outcome in relation to
the total of cases followed; 4) Percent of case seen in a Health Care
Facility (HCF) in relation to total exposures; 5) Percent of cases seen
in a Hospital in relation to cases seen in a HCF; 6) percent of cases
taken to an Intensive Care Unit (ICU) in relation to cases seen in a
HCF.

 (*) Sample size not big enough for a reliable test.

Children Cases 

In the children class, only four exposures to Fenoxaprop were reported
by PCC from 1993 to 2005. None of the cases was symptomatic or required
a visit to a health care facility.

Table 2. Relative Frequency of Acute Poisoning Incidents by Symptom
Severity: Combination of All Cases Regardless Class 

	Severity of outcome

(outcome determined)	Total

exposed	Health care 

provided

Denominator numbers	9	23	5

Measures	SYM1	MOD2	MAJ3	HCF4	HOSP5	ICU6

Numerator numbers	7	2	1	5	0	0

Fenoxaprop percents	77.78%	22.22%	11.11%	21.74%	0.00%	0.00%

All Pesticides percents	42.69%	6.02%	0.29%	16.01%	6.05%	2.16%

Ratio of  Fenoxaprop

/All pesticides	1.82(S)

P =0.03	3.69 (S)

P = 0.04	38.31 (S)

P = 0.00 (*)	1.36

P = 0.45	0.00

P = 0.22	0.00

P =0.47



1) Percent of cases with ANY symptom in relation to the total of cases
followed; 2) Percent with moderate symptoms in relation to the total of
cases followed; 3) Percent with major or fatal outcome in relation to
the total of cases followed; 4) Percent of case seen in a Health Care
Facility (HCF) in relation to total exposures; 5) Percent of cases seen
in a Hospital in relation to cases seen in a HCF; 6) percent of cases
taken to an Intensive Care Unit (ICU) in relation to cases seen in a
HCF.

 (*) Sample size not big enough for a reliable test.

Table 3 provides a summary by year, and Figure 1 shows a graphic summary
for yearly exposures cases, symptomatic cases, and cases seen in a
health care facility.  

Table 3. Summary of Cases by Year

Summary of Cases by Year

Year	Symptomatic

Cases	Moderate

Cases	Major

Cases	Cases 

Followed	Total Exposed	HCF

Cases	Hospital

Cases	ICU

Cases

1993	2	0	0	2	2	2	0	0

1994	0	0	0	0	1	0	0	0

1995	1	0	0	1	1	0	0	0

1996	1	0	0	1	1	0	0	0

1997	0	0	0	0	0	0	0	0

1998	0	0	0	0	4	0	0	0

1999	0	0	0	0	1	1	0	0

2000	1	1	1	1	4	2	0	0

2001	1	0	0	2	2	0	0	0

2002	1	1	0	1	4	0	0	0

2003	0	0	0	0	1	0	0	0

2004	0	0	0	0	0	0	0	0

2005	0	0	0	1	2	0	0	0

Total	7	2	1	9	23	5	0	0



Figure 1 Total exposure, symptomatic cases, and cases sent to HCF by
year

Analysis of Results

The main characteristic of exposures to Fenoxaprop is its low rate of
exposure with 23 exposures in the13-year period and only 7 symptomatic
cases in the entire population. Exposures for the classes Occupational
and Children are almost non-existent with five and four cases each. The
non-occupational class presents the most cases with 14 cases with one
case reported as major health effect; however, this case did not present
any long-term effect and doctors released the individual quickly. 

No trend in total exposure, symptomatic cases, and cases seen in a HCF
is apparent for the 13 year-span of data collected on Fenoxaprop, with
no case send to a HCF since 2001. Calculations show an average of about
1.73 exposures per year, 0.54 symptomatic cases per year, and 0.38 cases
per year seen in a heath care facility.

2. Cases reported in the Incident Data System from 1999 to the present

Three cases were reported in the IDS. For a detailed description, see
attachment 1.

3. NIOSH SENSOR

Out of 5,899 reported cases from 1998 to 2003, no case was reported in
the SENSOR database involving Fenoxaprop.

Study Summary

The summary findings for the period 1993 to 2005 for Fenoxaprop are:

Fenoxaprop has a low incidence of cases in the population, with only 23
exposures and seven symptomatic cases in the 13 year-span of data
collected.

 Fenoxaprop data shows very low averages of about 1.73 exposures/year,
0.54 symptomatic cases/year, and 0.38 cases/year seen in a heath care
facility. 

No apparent annual trend is apparent in the 13 year-span, in addition,
no case has visited a HCF since 2001.

Recommendations

Based on the incident data reviewed from the above sources, OPP suggests
no mitigation actions for Fenoxaprop at this time.

Attachment 1.

Cases from the Incident Data System for Fenoxaprop

Incident#2796-222

	A pesticide incident occurred in 1994, when the product drifted from a
field into gardens.  An individual reported dampness on his face.  No
further information on the disposition of the case was reported.

Incident#3037-102

	A pesticide incident occurred in 1994, when a pilot died after their
crop duster crashed.  The incident was not related to exposure of the
product.  No further information on the disposition of the case was
reported.

Incident#7155-66

	A pesticide incident occurred in 1996, when a parcel service worker
moved a box of the product and he put it down some of the powder got
into his face.  The worker was asymptomatic.  No further information on
the disposition of the case was reported. 

Attachment 2.

Table Description

The following is a general description of the table components that
compares the pesticide of interest with the composite average of all
pesticides. The table produces ratios in several health and exposure
measurements.

Table 1 General Explanation of Comparison of the Pesticide of Interest
with the Composite Average of all Pesticides.

	Severity of Outcome

(Outcome determined)	Total

Exposed	Health Care 

Provided

Denominator numbers	25,549	68,005	18,084

Measures	SYM	MOD	MAJ	HCF	HOSP	ICU

Numerator numbers	578	83	9	18,084	302	219

Pesticide “X” percents	2.26%	0.32%	0.03%	26.59%	1.67%	1.21%

All Pesticides percents	21.72%	1.42%	0.12%	15.68%	4.34%	1.75%

Ratio of  “X”

to average of all pesticides	0.10(S)

P =0.00	0.23 (S)

P = 0.00	0.25 (S)

P = 0.01	1.69 (S)

P = 0.00	0.38 (S)

P = 0.00	0.69 

P =0.51



A) The firs row, or title row, divides the table in three sub-tables
“Severity of Outcome”, “Total Exposed”, and “Health Care
Provided”

B) The second row “Denominator numbers” presents the numbers
extracted from the PCC database for the pesticide of interest “X”
that will serve as denominators in the calculation of the percents, and
they are:

In the center column, total amount of individuals exposed to pesticide
“X” (n = 68,005) or “Total Exposed”. These are cases collected
by the Poison Control Center for the period of interest regardless of
outcome. 

In the left column under “Severity of Outcome”, is the number of
cases followed, (n = 25,546). These cases have an outcome determined
“Outcome determined” and these cases may  or may not have symptoms.

At the right of the table, cases that visited a Health Care Facility (n
= 18084) “Health Care Provided”.

C) The third row “Measures” describes the clinical outcome and level
of care required for the exposed cases: 

“SYM” are those cases that were symptomatic; 

“MOD” cases are those that were classified with moderate symptoms; 

“MAJ” cases were cases classified as major symptoms or fatal
outcome; 

“HCF” are cases that went to a Health Care Facility; 

“HOSP” are cases that went to a hospital; and 

“ICU” are cases that went to an Intensive Care Unit.

D) The fourth row “Numerator numbers” is actually the number of
cases for pesticide “X” that presented the outcome describe in the
“Measures” row. The numbers in this row are inclusive. This is, a
case classified as “Major”, is first classified as
“Symptomatic”, and also classified as “Moderate”; so a
“Major” case will count on the three measurements. The same scenario
applies to the “Health Care Provided” section; a case classified as
“ICU” also counts as a case that went to a “HCF” and to a
“Hospital”. In addition, a case classified as “Hospital” will
count also in “HCF” but not on “ICU”

E) The fifth row “Pesticide “X” percents” gives the percentages
on each measure. These percentages are calculated by dividing the
numerator number by the denominator number (from each major section of
the Table 1) and multiplying by 100. For example, to calculate SYM, MOD,
and MAJ, the numbers 578, 83 and 9 are divided by 25,549 on the first
row, producing the respective percents 2.26%, 0.32%, and 0.03%.  In a
similar fashion 18,084 is divided by 68,005 for the percentage of cases
seen in a HCF of 26.59%; and for the last section of the fifth row, 302
and 219 are divided by 18,084 to find the percentages, 1.67% and 1.21%, 
of cases that went to a “Hospital” and “ICU”. 

F) The sixth row “All Pesticides Percents” are percentages that were
previously calculated using all pesticide cases available in the PCC
database with the exception of cases involving exposures to multiple
products, cases with unrelated medical outcome, and cases where the
exposure was intentional. These exclusions make the comparison more
meaningful for US EPA regulatory program purposes. The percentages serve
as the baseline for the comparison with a single pesticide or a group.
These percentages are a unique characteristic for the “class” and
“period” of the data assembled in the table, and are completely
independent of the previously calculated values for pesticide “X”.

 

	G) The last row “Ratio of “X” to average of all pesticides” is
obtained by dividing the percentages of the “X” compound by the
“all pesticides” percentages, to obtain a dimensionless number or
ratio. This ratio provides an idea of the relative frequency of the
compound “X”. For example, a ratio of 1 (one) indicates that the
percentages are the same for the compound “X” and the composite of
all pesticides; a ratio, say of 2 (two), indicates that the chemical
under study produces twice the effect; and a ratio of 0.5 indicates that
the compound has half of the activity of the composite average. These
ratios provide a quick overview of the relative toxicity of the
chemical.

H) Also, the “p” value that results from a Likelihood ratio test is
entered in the ratio cell and when significant (p < 0.05) then an (S) is
entered next to the ratio to mark statistical significance. A cell
marked with (S) means that the percentages that produce the ratios are
“statistically” different and not due to chance. The percentage of
pesticide “X” could be higher or lower than the percentage produce
by the composite as indicated by the ratio. However, when mark by an (S)
the difference is significant. For example, a ratio = 1.6 (S) means the
effect produced by the compound is higher that the effect produced by
the composite and statistically different; a ratio = 0.7 (S) signifies
that pesticide “X” has a lower effect than the composite and is
statistically different. On the other hand, a ratio = 1.2, with no (S),
means that the percentage, although higher, is not statistically
d晩敦敲瑮映潲⁭桴⁥潣灭獯瑩⹥‍̍഍ഄ̍഍ഄ഍഍഍
഍഍഍഍഍

഍

഍഍഍

഍

഍഍഍഍ഁ഍഍഍഍഍഍഍഍഍഍഍഍഍഍഍഍഍഍഍
഍഍഍഍഍഍഍഍഍഍഍഍഍

