The RAND survey determined that 7% of servicemembers used dichlorvos, 20% resin strips or witnessed others using them (Table 8). Six percent of the PM interviews cited use of dichlorvos resin strips (Table 13). The dichlorvos formulation used was similar to those widely available to the American public for many years. Additionally, similar formulations are still on the market. Servicemembers would hang the strips at indoor locations to suppress populations of various flying insects. The only relevant exposure route for dichlorvos released from resin strips is inhalation. The strips are provided in cardboard containers which normally minimize or prevent dermal contact.

1.  Application Scenarios

Investigators did not address application exposure separately, because it would have been inconsequential in comparison to post-application exposure. Likewise, investigators did not evaluate skin contact because any dose received would have been inconsequential. While an applicator may have had skin contact with the resin strip while inserting the strip in its holder during application, the skin surface area exposed (finger tips) and the exposure time (a few seconds per strip) would not have allowed for a consequential dose.

2.  Post-Application Scenarios

Table 51 presents the assumptions for post-application exposure to dichlorvos, 20% resin strips. Investigators presumed exposure to dichlorvos to have taken place inside tents and buildings, consistent with its intended use and with the results of the Survey (Table 11) and PM interviews (Table 13). The label^[222] specifies an application rate of one strip per 1,000 ft³. Investigators assumed this application rate for all air modeling mainly because it is specified on the label, although it is known that other application rates were used, based on the RAND survey (Table 11). The variable used in air modeling which determined the result by exposure level was ventilation rate; the higher the ventilation rate, the lower the estimated air concentration.

3.  Air Modeling for Dichlorvos

EPA’s Office of Pesticide Programs has conducted risk assessments for dichlorvos, including an assessment of residential exposure to resin strips. They have published comments from the Scientific Advisory Panel which reviewed some of OPP's work. In the July 1998 report, OPP presents a dichlorvos release rate from a single pest strip.^[223] Based on a strip containing 20 grams of dichlorvos which is completely released over a 56-day period (24 h/d), the average release rate is 14.9 mg/h. Investigators used this emission rate as the basis for exposure assessment. The OPP report also presents data from a 1973 study in which the application rate (the room volume being treated by one strip) ranged from 720 to 6,790 ft³/strip, with an average of 1,833 ft³/strip. This range brackets the label-specified application rate of 1,000 ft³/strip for a currently available pest strip (approximately the same size as those in the 1973 study).

If one assumes that servicemembers used the pest strips at a rate of 1,000 ft³/strip, then the indoor air concentration theoretically will not change with the size of the space being treated, since the number of strips installed increases in proportion to the indoor air volume. Therefore, one can determine the indoor air concentration for a single strip for a nominal 1,000 ft³ space, and this value will represent any space with the same ventilation characteristics (air exchange rate and mixing factor).

Investigators used a standard box model equation for the calculations of indoor concentrations. Complete mixing was assumed, and the calculations were conducted with and without a consideration of decay to assess the effect on the results. One can develop the box model equation from the same mass balance considerations as described for permethrin, and employ the same differential equation, with slightly different inputs:

V(dC/dt) = E + C_aIV – CIV – KCV

where,

C = concentration (mg/m³)
C_a = ambient (outdoor) concentration (mg/m³)
E = emission rate (mg/h)
I = air changes per hour in room
V = room volume (m³)
t = time (h)
K = decay rate (h^-1)

This equation has the following general solution:

C = [1/(I+K)][(E/V) + (C_a)(I)][1 – exp{-(I+K)(t)}] + C_o exp{-(I+K)(t)}

where:

C_o = initial concentration in room (mg/m³)

Consistent with EPA draft guidance for conducting residential exposure assessments, investigators assumed contributions from the outdoors were negligible (C_a = 0). With this assumption, the equation for concentration within the room simplifies to:

C = [1/(I+K)][E/V][1 – exp{-(I+K)(t)}] + C_o exp{-(I+K)(t)}

Dichlorvos in the vapor phase is degraded in the atmosphere through reactions with photochemically produced hydroxyl radicals, and this reaction has an estimated half-life of 13.6 hours.^[224] So one would expect the concentration of dichlorvos in a room to drop by 50% over 13.6 hours in the absence of other effects (such as new emissions and air turnover). Investigators examined the effect of this decay in some preliminary calculations and determined that decay would be insignificant. Therefore, subsequent calculations were conducted with K = 0 (i.e., assuming dichlorvos was unreactive).

Returning to the general box model solution and setting C_o = 0 and K = 0, the expression for concentration becomes:

C = [1/I][E/V][1 – exp{-(I)(t)}]

As time increases, the concentration will asymptotically approach an equilibrium concentration given by:

C = [1/I][E/V]

One can obtain an average concentration over the time interval from t₁ to t₂ by integrating the concentration equation above over this interval and dividing by the duration of the interval yielding:

C_avg = [1/(t₂-t₁)][1/I][E/V][(t₂-t₁) + (1/I)[exp{-(I)(t₂)}-exp{-(I)(t₁)}]]

The average concentration over the interval from t = 0 to T (i.e., for the first T hours) then becomes:

C_avg = [1/T][1/I][E/V][T + (1/I)[exp{-(I)(T)}-1]]

Investigators assumed ventilation rates of 2, 4, and 6 air changes per hour for the high, medium, and low indoor exposure scenarios, respectively. Investigators believe these rates to be reasonable given reports that strong winds readily penetrated the structures under consideration. Furthermore, investigators believe these air exchange rates to be representative for tents used during deployment in a hot, potentially windy environment, where flaps might frequently be open.

The box model equations were used to estimate equilibrium concentrations and average concentrations over the first 16 hours of exposure following the installation of the resin strip. The estimated equilibrium concentrations of dichlorvos in the tent are listed in Table 52.

Figure 12 shows time plots of dichlorvos concentrations in the tent for the first 3 hours following the installation of the resin strip. Concentrations start at zero and rapidly approach the corresponding equilibrium concentrations. Table 53 presents a summary of the calculation of estimated concentrations.

Figure 12. Time plots of resin strip (dichlorvos) air concentrations inside tent

Although EPA recommends the use of the MCCEM in conducting high-end exposure assessments, investigators used the simple box model for the same reasons discussed under permethrin.

In the OPP report,^[225] four different methods were used to estimate residential inhalation exposure to dichlorvos from pest strips. Three of these methods were based on dichlorvos air concentrations measured in homes using pest strips in a 1973 study.^[226] The fourth method was based on a calculated dichlorvos emission rate (same one used in this analysis) and the use of a computerized EPA indoor air model, the MCCEM. EPA’s fourth method is most similar to the approach used in this analysis. The MCCEM model calculates indoor air concentrations in various rooms within a structure, based on assumed fresh air infiltration rates and air migration patterns between rooms. MCCEM performs calculations similar to the box model, except that it allows numerous rooms (boxes) to be evaluated and can evaluate concentrations at different time intervals with different emission rates (although EPA held the emission rate constant in its scenarios).

EPA’s highest estimated dichlorvos exposure using the MCCEM method was 0.079 mg/kg/d for adult males over a 56-day exposure period. This daily dose is equivalent to an average air concentration of 0.43 mg/m³, based on a body weight of 70 kg, 15 h/d spent indoors and an inhalation rate of 12.9 m³/d while indoors. The house modeled by EPA had an intermediate air exchange rate (3.57 air changes per hour) and a lower application rate (one pest strip per 1,900 ft³) compared with the scenarios modeled here. The air concentrations investigators estimated for GP tents, 0.09-0.26 mg/m³, are similar to EPA’s estimate.

The 1973 indoor air measurement study used as the basis for three of EPA’s exposure estimation methods showed lower dichlorvos air concentrations than those estimated with the models discussed above.^[227] The highest measured dichlorvos concentration was 0.11 mg/m³, measured on the first day after installation of the strips. This is comparable to the concentrations estimated here for the low and medium exposure scenarios. The estimated concentrations for the high exposure scenario are somewhat higher than those measured. However, the usage rate of resin strips reported by EPA (an average of one strip per 1,833 ft³ and a range of between one strip per 720 ft³/strip and one strip per 6,790 ft³) is generally smaller than the rate of one strip per 1,000 ft³ investigators assumed in the modeling analysis presented here.

Investigators did not use the 1973 measurement study as the basis of these exposure estimates for the following reasons:

• application and ventilation rates are unknown;

• investigators considered the sampling durations too short to be representative (20 min);

• accuracy of results are unknown and suspect, since the air measurement method is poorly documented and no quality control data are presented; and

• pesticide active ingredient recovery from environmental samples can be variable and low.

For these reasons, investigators expected the 1973 measurement results would underestimate actual concentrations somewhat for the present exposure assessment.

4.  Dichlorvos Doses – Post Application

Table 54 presents doses potentially resulting from post-application exposure to dichlorvos, 20% resin strip. The only dose listed, and the only one of concern, is the potential dose rate for inhalation (PDR_I). Investigators did not calculate lifetime average daily doses (LADDs) because OPP does not currently provide a carcinogenic slope factor for dichlorvos by any exposure route (see Section B.4,   Toxicity Assessment).

| First Page | Prev Page | Next Page |