Chapter Three: Survey Results

Unless otherwise indicated, the proportions and percentages presented in this chapter are our best estimates--based on the 2,005 survey respondents' responses--for the 469,047 in-theater Gulf War personnel on the ground.[1]

RECALL BIAS RESULTS

Overall, we found that the types of pesticides reported increased in the re-survey by about 13 percent, largely because of increased reporting of field-use pesticides such as No-Pest strips or sprays. Answers about use of personal pesticides, such as number of sprays used or how many times a spray was used, were stable across the surveys.

We also examined the data to see if some groups changed their answers more than other groups. We did not find strong patterns by education, rank, self-reported health status, or a number of other factors. The only group whose answers changed in statistically significant ways for all three variables were junior enlisted personnel (pay grades E-1 to E-5), who remembered more pesticides, both personal and field use. A "worst case" interpretation of these findings is that the incidence of pesticides reporting may have been underestimated in the initial survey but that the effect is not large. Thus, the results presented in this chapter could be considered conservative, in the sense that they may underestimate the fraction of the population that used each form.

PESTS AND PESTICIDES

Fly swarms were ubiquitous throughout the region--93 percent of the population experienced them. Veterans relate that there were so many flies that it was often difficult to eat. Roughly half of the population also experienced other flying insects, most notably mosquitoes. Crawling pests were widespread, with a majority of the population experiencing scorpions, spiders, and ants. Rodents were also quite frequent, with half the population reporting them.

Table 3.1 also shows that members of each service had relatively similar pest encounters. Somewhat fewer Marines and Navy personnel reported spiders, mosquitoes, and, fleas,[2] whereas more Air Force personnel reported spiders and roaches, and fewer rodents, lice, and ticks. All these differences were statistically significant. The only pests without statistically significant differences among the service groups were ants, wasps, centipedes, and "other."

^a"s.e." stands for "standard error," a commonly used statistical measure of the variability of the estimated quantity. The larger the standard error, the greater the actual (unknown) value for the population may deviate from the value estimated from the survey data. It is accepted practice to consider that the true value is highly likely to be within two standard errors of the estimated value. Thus, in this table, although we estimate that 95 percent of Army personnel saw flies, we expect the true but unknown percentage to be between 93 and 97.
^bFleas were actually very rare in this region. Service members often referred to "sand fleas," but there is no such species. The pests were probably phlebotomine sand flies, which are so small they could have been confused with fleas when they bite.

Although not shown in Table 3.1, there were also statistically significant differences for some pests among personnel in urban locations and in food service occupations. In particular, personnel in urban locations were more likely to have seen roaches (28 percent) and wasps (14 percent) and less likely to have seen scorpions (55 percent) and "other" pests (27 percent). Food service personnel were more likely to report flies (97 percent) and scorpions (82 percent), and less likely to report ants (36 percent).

Table 3.2 shows the percentage of personnel who reported each form of

personal-use pesticide and Table 3.3 shows the percentage who reported each form of field-use pesticide. Use of personal pesticides clearly differed by service more than the reports of pests. The only form that was not significantly different was the flea collar. In particular, Air Force personnel were less likely to use all forms of personal pesticides, and Army personnel much more likely to use powders. Over one-third of the population did not use any personal pesticides.

Table 3.3 shows that statistically detectable differences between field-use pesticides occurred between services for all forms of field-use pesticides except pellets, No-Pest strips, and "other." In particular, the Marine Corps/Navy showed much lower observed use of aerosols, whereas the Air Force observed a higher use of other sprays and a lower observed use of liquids. About one-half of the total population did not use or observe any field pesticides.

PERSONAL PESTICIDE TABULATIONS

Personal-Use Pesticide Tabulations by Form

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers or because of rounding.

^aThe columns labeled "Times/day" indicate the number of times per day for the days used.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers or because of rounding.

^aThe columns labeled "Times/day" indicate the number of times per day for the days used.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers or because of rounding.

^aThe columns labeled "Times/day" indicate the number of times per day for the days used.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers or because of rounding.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers or because of rounding.

• The estimated percentage of the population who used a particular form in total and by service;

• For those who used a particular form, the estimated average frequency of use, in terms of the average number of times per month and the average number of times per day, and the percentiles of use;

• For those who used a particular form, an estimated percentage breakdown of the source of pesticides;

• For those who used a particular form, the reasons they stopped using the pesticides;

• For those who used a particular form, the estimated percentage who reported experiencing side effects and the percentage who stopped or reduced using the pesticide because of side effects;

• The estimated average number of products used; and

• Where the products were used.

Many personnel reported acquiring their sprays via "military issue," although they were also frequently acquired from other sources such as a military PX, via mail from friends and relatives, and from fellow U.S. soldiers. Two-thirds of the respondents using sprays reported applying them to both body and uniform, another third to just the uniform, and fewer to just the body.[5] About 10 percent of the population using sprays believed that they experienced side effects from a spray, and about one-half of these individuals stopped using the spray because of the perceived side effects.

Liquids and lotions (Tables 3.5 and 3.6) were the next most frequently reported forms. About one-quarter of Army and Marine Corps/Navy personnel reported their use. Use in the Air Force differed, with about 10 percent using liquids and about 20 percent using lotions. In all services, the average among those who used them was about thirty times per month, but about twice per day for both forms on the days they were used. The maximum use was about six times per day for the liquids and slightly higher for lotions.

Both forms were principally military issue, with most of the balance coming from fellow U.S. soldiers.[6] As might be expected, lotions were used mainly on the body, with a much smaller percentage of soldiers using them on both body and uniform. Liquids were used on the body by about one-half of the population, on the body and uniform by about one-third, and on only the uniform by about 5 percent of the population. As with the sprays, roughly 10 percent of the population perceived side effects from each of these forms, and about one-half of these stopped using the form because of the perceived side effects.[7]

Fewer than 10 percent of the population used personal powders,[8] as Table 3.7 shows. These personnel were largely Army, with only 1 to 2 percent of personnel in the other services using powders. Those who used powders used them on average about once per day; the 95th percentile for those who used powders was about three times per day, and the maximum reported use was five times per day. Compared to sprays, liquids, and lotions, powders were more likely to lead to perceived side effects, and more powder users stopped their use because of the perceived side effects.[9]

From Table 3.8, we find that about 3 percent of the population in the Army and Marine Corps/Navy and only about 1 percent in the Air Force used flea or tick collars. Among those who wore flea or tick collars, the median user wore the collar every day. Unlike the other forms of pesticides, most flea or tick collars were acquired either directly from the United States or from a military PX.[10] A large majority who used flea or tick collars wore them over clothes or shoes, although about 20 percent wore them directly on the skin.

The AFPMB issued a message specifying that flea or tick collars are not safe for human use (AFPMB, 1990). Although this message was transmitted in early December 1990, we found that personnel continued to wear these collars through the end of the survey period, July 1991.

Table 3.9 compiles the remaining "other" products reported, which comprise mainly various stick forms of pesticides. These were used by only a small fraction of the population, mostly in the Air Force.

Personal-Use Pesticide Tabulations by Active Ingredient

• The percentage of the population who used an active ingredient in total and by service; and

• Among those who used a product with the active ingredient, the average frequency of use, in terms of the average number of times per month and the average number of times per day, and the percentiles of usage.

NOTE: Forms included sprays, liquids, lotions, other.
^aThe columns labeled "Times/day" indicate the number of times per day for the days used.

NOTE: The columns labeled "Times/day" indicate the number of times per day for the days used.

NOTE: Forms included sprays, liquids, lotions.
^aThe columns labeled "Times/day" indicate the number of times per day for the days used.

We find that DEET was the most common personal pesticide active ingredient (see Table 3.10). DEET was used by half of the Army and Marine Corps/Navy population and slightly more than a third of the Air Force population.[11] Among those who used DEET, on average it was used about three times a day for 15 days a month in the Army, and twice a day in the Marine Corps/Navy and Air Force for similar periods. The 95th percentile for frequency of use among those who used DEET was about seven times a day, although the maximum was 16 times a day (for the Army and Marine Corps/Navy; slightly less for the Air Force).

As shown in Table 3.11, permethrin was used by about 6 percent of the ODS/DS population, with a slightly higher percentage in the Army, and a slightly lower percentage in the other services.[12] Frequency of use among those who used permethrin averaged almost 30 times a month, although it tended to be used twice a day for two weeks out of four. The 95th percentile for frequency for those who used permethrin was four times a day, although the maximum was 14 times a day for the Army; it was slightly less for the Marine Corps/Navy, and for the Air Force the maximum was only four times per day.[13]

As Table 3.12 shows, sulfur was used by about 3 percent of the population--slightly higher in the Army and slightly lower in the Marine Corps/Navy and Air Force. Among those who used them, sulfur products were used on average slightly more than 30 times a month, except in the Air Force where they were used only 20 times a month. When sulfur was used, it was used about twice a day on average. At the 95th percentile, it was used about three times a day, although the maximum daily usage was ten times a day in the Army and six times a day in the Marine Corps/Navy and Air Force.

FIELD-USE PESTICIDE TABULATIONS BY FORM

We expected that field-use pesticides would be both underreported and reported with sparse information because, unlike the personal-use pesticides, field-use pesticides were generally not applied by the survey respondent. Thus, the respondent could generally provide information only on observed applications, which is most likely a subset of all the applications that occurred.

Tables 3.13 through 3.19 contain our best estimates of use from the survey data for the entire in-theater Gulf War population. The tables list, for those who used or observed the use of a particular field pesticide form:

• The estimated percentage for the total population and by service;

• For those who used a particular form, the estimated average frequency of use in terms of the average number of times per month, and the percentiles of use for number of times per month;

• An estimated percentage breakdown of the source of pesticides;

• The reasons they stopped using the pesticide(s);

• The percentage who reported experiencing side effects; for those who experienced side effects, the percentage who stopped or reduced use of the pesticide;

• The average number of products used or observed; and

• Where the products were used or observed to be used.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers
^aLime was also used to control pests. However, since it is not a pesticide, we removed it from these tabulations whenever we could identify it.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers

	Total GW Population (n = 30,530)				Army (n = 21,912)				Marines/Navy (n = 4,604)				Air Force (n = 4,014)
	<1	~1	>1	?	<1	~1	>1	?	<1	~1	>1	?	<1	~1	>1	?
1	45(7)	13(4)	5(2)	37(6)	51(9)	15(6)	4(2)	31(8)	29(9)	6(4)	12(6)	53(10)	30(8)	12(6)	6(3)	52(9)
2	13(4)	16(5)	8(2)	63(6)	13(5)	20(7)	4(2)	62(8)	11(6)	2(2)	19(8)	69(9)	18(6)	9(6)	13(6)	61(8)
3	46(9)	26(8)	27(8)	<1(<1)	49(13)	26(10)	25(11)	0	11(5)	40(13)	49(13)	0	78(10)	7(7)	8(6)	7(7)
4	19(8)	1(1)	1(1)	79(8)	22(11)	1(1)	0	77(11)	12(9)	3(3)	0	85(10)	12(5)	1(1)	6(6)	82(7)
5	6(2)	11(3)	30(6)	53(6)	5(2)	10(4)	30(8)	55(8)	3(3)	19(8)	27(9)	51(10)	18(7)	8(4)	28(8)	46(9)
6	2(1)	9(4)	2(1)	87(4)	1(1)	12(6)	0	87(6)	1(1)	0	11(7)	89(7)	9(5)	7(5)	4(4)	80(7)

NOTES The numbers in the first column identify the location used. 1 = sleeping quarters; 2 = mess halls and other designated eating areas; 3 = other eating areas; 4 = work area; 5 = latrine; 6 = other. Column headings identify the density. <1 = less than 1 per 10 x 10 feet; ~1 = about 1 per 10 x 10 feet; > 1 = more than 1 per 10 x 10 feet; ? = don't know.
^aDensity is calculated as 10 ft by 10 ft of space.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers or because of rounding.

NOTE: Columns may sum to more than 100 percent because respondents could give multiple answers

The most frequently used or observed field pesticides were aerosols; Table 3.13 shows that slightly less than one-third in each service reported using aerosols in the field. Whereas a smaller percentage of the population used or observed aerosols, those who reported them indicated heavier use in the sleeping, eating, working, and other areas. "Other" sprays in Table 3.14 were similarly ubiquitous and used or observed by about 20 percent of the population. This is the only form in which the fraction of Air Force personnel using or observing the form is significantly higher than in the other services. "Other" sprays were sprays applied by a hand-held sprayer, a truck sprayer, or a plane sprayer.

Of those who saw or used sprayers, we estimate that 40 percent were hand-held sprayers (by service: 47 percent in the Army, 36 percent in the Marine Corps/

Navy, and 29 percent in the Air Force), 69 percent were truck sprayers (by service: 60 percent in the Army, 71 percent in the Marine Corps/Navy, and 81 percent in the Air Force), and 4 percent were plane sprayers (by service: 4 percent in the Army, 10 percent in the Marine Corps/Navy, and 1 percent in the Air Force).[14] The percentages sum to more than 100 because respondents could give answers for up to three sprayers.

Tables 3.15 and 3.16 show that slightly more than 10 percent of the population used or observed pellets, crystals, granules, and field powders, respectively. Field liquids were used or observed by about 4 percent of the population (Table 3.17). On average, among those who reported them, these forms were used or observed about 20 times per month, with the 95th percentile at about 60 times per month. All of these forms were predominantly military issue, used by U.S. military troops, and observed or used roughly equally in all areas.

Table 3.18 demonstrates that about 7 percent of the population used or observed No-Pest strips. It also shows that in most of the locations (sleeping, eating, work, latrine, and "other"), the density of No-Pest strips was less than or about equal to the recommended density of one per 100 square feet of floor area.[15] However, 30 percent of latrines and other eating areas (non-mess hall and other areas not designated specifically for eating) exceeded this standard. Other eating areas had an average density of about two per 100 square feet and latrines had an average density of about four per 100 square feet.

[1]The in-theater Gulf War population consists of all Army and Marine Corps personnel located in Saudi Arabia, Kuwait, Bahrain; all Air Force personnel located in Saudi Arabia, Kuwait, Bahrain, Qatar, the United Arab Emirates, and Oman; and Navy personnel in units that can be identified as being ashore in Saudi Arabia, Kuwait, and Bahrain. Details of the statistical methodology are contained in Appendix C.

[2]Fleas were actually very unlikely in this region. While service members often referred to "sand fleas," there is no such species. The pests were probably phlebotomine sand flies, which are so small they could have been confused with fleas when they bite.

[3]Frequency of use is the combined frequency from all sprays reported.

[4]A 95th percentile for frequency of use at six times a day means that 5 percent of the population applied sprays at least six times a day.

[5]The only personal-use spray available in the military supply system during ODS/DS was a permethrin product intended for use on uniforms only. Thus, sprays acquired as "military issue" and used on the body represent either a recall error on the part of the respondent, a definition of "military issue" more liberal than just the "military supply system," or a misuse of the pesticide. See Chapter Five on possible misuse of pesticides.

[6]Gambel et al. (1998) found that more than 60 percent of the soldiers used commercial repellents. Our results show a much higher use of military issue products. This differential may be due to a restricted availability of commercial products during ODS/DS, recall error, perhaps the use of the response "military issue" to mean more than just the military supply system, or some combination of all of these and other factors.

[7]The survey did not solicit specific details about side effects.

[8]Survey respondents did report powders that did not contain pesticides, such as talcum powder. Data on these nonpesticides were removed as much as possible.

[9] The survey did not solicit specific details about side effects.

[10]Despite the fact that 6 percent of the population indicated that their flea collars were military issue, these products are not now, nor have they ever been, available through the military supply system. This may reflect respondents' perception of military issue as encompassing more than the military supply system. For example, one respondent stated, "flea collars [were] used by [the] whole unit (22 people) and these collars were U.S. military issued."

[11]These results are consistent with reports on the use of insect repellents by Army personnel in surveys conducted by Gambel et al. (1998). Gambel et al. found that 56.5 percent of the personnel they surveyed who were deployed in Bosnia applied insect repellents to the skin. However, our results show higher use than Gambel et al. found from their survey of about 200 Army personnel deployed to Kuwait in 1994 during Operation Vigilant Warrior, where only 26.2 percent of the personnel said they used insect repellents on the skin. Yet, in a similar survey of soldiers deployed to Haiti, they found that 94.3 percent of the personnel used insect repellents on their skin.

[12]These results are also consistent with those of Gambel et al. (1998). In their surveys, they found that only 7.6 percent of troops deploying to Kuwait in 1994, Haiti in 1995, and Bosnia in 1996 treated their uniforms before deployment. They found that only 13.1 percent treated their uniforms while in Kuwait, 29.4 percent while in Haiti, and 18.9 percent while in Bosnia.

[13]The only permethrin product available from the military supply system at the time of ODS/DS was an aerosol for treating uniforms. With proper application, that product should have lasted in a uniform for about six weeks or six launderings. Thus, the average level of reported use was well in excess of the recommended amount. This result may indicate misclassification or misuse (or both) of the pesticide. See Chapter Five on possible misuse of pesticides.

[14]Despite respondents' perceptions that aerial application of pesticides occurred, no known applications were made by any of the services during the Gulf War.

[15]The recommended density is one per 1,000 cubic feet, which the survey simplified by assuming a fixed ten-foot ceiling height, in which case the recommended density reduces to one per 100 square feet of floor space.