Introduction
Studies clearly show that using bed nets that have been factory-impregnated with long-lasting insecticides (LLINs) is an effective malaria management tactic. This was shown perhaps most dramatically by Fegan and others,1 who observed a 44% decrease in malaria mortality of children in Kenya with the use of LLINs. Lengeler and deSavigny2 said of the Fegan and others1 study: “With this work, the use of insecticide-treated bed nets is confirmed as a major child-survival intervention.” Consequently, LLINs have become a key component of malaria management programs worldwide.3
Much attention has been devoted to how best to distribute the LLINs. Teklehaimanot and others4 argue for full distribution and universal access to LLINs to achieve community-wide protection against malaria. Furthermore, they suggest that, by giving enough bed nets to cover every sleeping space, the reservoir of infection is replaced by the nets' community effect, thereby significantly reducing malaria transmission.4
Despite the ability of LLINs to protect people from malaria and despite the ambitious plans for their widespread use, the health risks from the LLINs themselves have not been adequately investigated and reported in the peer-reviewed science literature. A scientifically rigorous, objective, and transparent examination of the health risks associated with LLINs is needed to provide risk–benefit considerations of this promising technology.
A brief review of risk assessments associated with bed nets highlights some important issues and information gaps. There are only two peer-reviewed risk assessments of insecticide-treated bed nets in the literature.5,6 Both assessments used deterministic values (i.e., single-point inputs) and relied on surrogate data to estimate risks from deltamethrin5 and permethrin, deltamethrin, λ-cyhalothrin, α-cypermethrin, and cyfluthrin.6 Barlow and others5 assessed risks from field-treated bed nets that rely on soaking the nets in an insecticide-treated solution. Macedo and others6 assessed risks to deployed military forces, not civilian populations with various age groups of men, women, and children. Estimates of inhalation exposure to the insecticides from respiration under the bed nets were based on Bomann,7 who used only cyfluthrin in an indoor study. Estimates of dermal exposure to the insecticides used removal of insecticide applied to carpeting as a surrogate for how much insecticide might be removed from netting. In addition, the World Health Organization (WHO) Pesticide Evaluation Scheme (WHOPES) has summarized safety assessments for their recommended bed nets,8–13 but these also are based on deterministic surrogate inputs.
Because the previous risk assessments are deterministic and use uncertain surrogate values, a probabilistic risk assessment is needed to estimate the range of possible risk values and better understand uncertainty and sensitivity of the exposure model. In this paper, we use a probabilistic risk assessment approach to estimate the risks to Africans from exposure to the newer factory-impregnated LLINs such as Olyset Net (Sumitomo Chemical Co., London, United Kingdom), PermaNet 2.0 (Vestergaard Frandsen, Lausanne, Switzerland), DawaPlus 2.0 (Tana Netting, Bangkok, Thailand), Yorkool (Yorkool International Co., Tianjin, P. R. China), DuraNet (Clarke, Roselle, IL), Netprotect (Bestnet Europe, London, United Kingdom), and Interceptor (BASF, Ludwigshafen, Germany), which are fully recommended or have interim recommendations by the WHO for malaria management.
Materials and Methods
Insecticides, body weights, and sleep times.
The insecticides used to factory-impregnate bed nets currently being used include permethrin (Olyset Net),10,14 deltamethrin (PermaNet 2.0, Netprotect, Yorkool, and DawaPlus 2.0),10,11,15,16 and α-cypermethin (Interceptor and DuraNet).12,15 The target dose of each insecticide is the amount of insecticide applied to the bed nets, and in the case of Olyset Net, it is the average application concentration that is measurable on the net's surface.14 We assumed that all nets were between 14 and 15 m25,7,12,14,15,17,18 (Table 1). To be conservative, we assumed that the insecticide applied to the LLINs did not degrade over time because of repeated washing or other factors. The insecticides in LLINs degrade over time, but the degradation can be highly variable.11
Long-lasting insecticide-impregnated bed nets used in the risk assessment
| Bed net brand name | Insecticide active ingredient | Application rate (mg/m2) |
|---|---|---|
| Olyset Net | permethrin | 30* |
| Netprotect | deltamethrin | 63 |
| PermaNet 2.0; Yorkool | deltamethrin | 55 |
| DawaPlus 2.0; DuraNet | deltamethrin/α-cypermethrin | 80/261 |
| Interceptor | α-cypermethrin | 200 |
Average concentration of permethrin on the bed net's surface.
Exposures to LLINs were estimated using several different age groups to incorporate different body weights and sleeping behaviors. These age groups were separated by gender and included 17 groups between the ages of 0 and 70 years. We estimated exposure and risk for both males and females, but in this paper, we are only presenting data for females; because of their lighter body weight, they are more sensitive receptors than males.
Body weights for African children were obtained for girls ages 3.5–10.4 years (N = 651) from rural South Africa and were extrapolated for all bed net users within defined age groups.19 Body weights were obtained for women between the ages of 15 and 64 years (N = 544) from South Africa.20 Body weights for girls ages 11 to 14 years were obtained from Portier and others.21 Because there were no weight data for African girls in that age group, we used data from Portier and others,21 who estimated the age-specific body weight distributions for US residents using data collected by the National Health and Nutrition Examination Survey in four surveys over the past 24 years. Body-weight distributions for infant and toddler girls (ages 0 to 3 years) were log-normal and also were obtained from Portier and others21 (Table 2). Body-weight distributions for all of the above groups were log-normal (Table 2).
Female body weights for age groups used in the exposure assessment
| Group | Mean (SD) in kg | Ref. |
|---|---|---|
| Infants and toddlers (0–3 years) | 9.1 (1.24) | 21 |
| Children (3.5–10.4 years) | 19 | |
| 3.5 | 12.5 (2.30) | |
| 4.5 | 14.6 (2.90) | |
| 5.6 | 16.1 (2.00) | |
| 6.5 | 17.6 (2.70) | |
| 7.5 | 19.1 (2.90) | |
| 8.4 | 20.2 (3.30) | |
| 9.4 | 22.0 (3.30) | |
| 10.4 | 21.8 (5.10) | |
| Youth (11–14 years) | 36.16 (7.12) | 21 |
| Adults (15–64 years) | 20 | |
| 15–24 | 62.2 (12.10) | |
| 25–34 | 69.1 (15.60) | |
| 35–44 | 77.8 (18.10) | |
| 45–54 | 78.6 (13.80) | |
| 55–70 | 77.3 (12.30) |
Sleep durations (the amount of time the person was exposed to the net per 24-hour day) were assumed to range between 6 and 12 hours. Infants and toddlers, defined as individuals between the ages of 0 and 3 years, were assumed to sleep an average of 11 hours, with a uniformly distributed range between 10 and 12 hours. Children in the age groups between 3 and 10 years were assumed to sleep an average of 10 hours distributed uniformly between 8 and 12 hours. Individuals in the age groups greater than 10 years were assumed to sleep an average 8 hours a day distributed uniformly between 6 and 10 hours (Table 3).
Values and distributions for probabilistic input variables used in the exposure assessment
| Input variable | Group | Value | Units | Distribution | Ref. |
|---|---|---|---|---|---|
| Sleep duration | Infants and toddlers (0–3 years) | 10–12 | hour | Uniform | |
| Sleep duration | Children (3.5–10.4 years) | 8–10 | hour | ||
| Sleep duration | Youth (11–14 years) | 6–10 | hour | ||
| Sleep duration | Adults (15–70 years) | 6–10 | hour | ||
| Respiration rate | Infants and toddlers (0–3 years) | 0.17 (0.02) | m3/hour | Lognormal | 22, 23 |
| Respiration rate | Children (3.5–10.4 years) | 0.24 (0.02) | m3/hour | ||
| Respiration rate | Youth (11–18 years) | 0.35 (0.04) | m3/hour | ||
| Respiration rate | Adults (19–30 years) | 0.33 (0.03) | m3/hour | ||
| Respiration rate | Adults (31–60 years) | 0.32 (0.03) | m3/hour | ||
| Respiration rate | Adults (61+ years) | 0.30 (0.03) | m3/hour | ||
| Air concentration of insecticide under the bed net | All | 0.02–0.06 | μg/m3 | Uniform | 7 |
| Body surface area in contact with bed net | All | 10–60 | % | Uniform | |
| Transfer coefficient from dry net to skin | All | 0.49–2.5 (0.49 most likely) | % | Triangular | 5, 17 |
| Saliva extraction factor | All | 25–75 (50 most likely) | % | Triangular | 26 |
| Surface area of child's hand | Infants and toddlers (0–3 years); child (3.5–10 years) | 0.003–0.004; 0.008–0.01 | m2 | Uniform | 16 |
Inhalation.
Dermal.
Oral.
Potential exposures from inhalation, dermal, and oral routes were summed to estimate potential total systemic exposures (also known as total body burden). We assumed that individuals in each age group slept under the LLINs each night for extended periods of time, and therefore, all exposures were considered chronic. We summed total potential exposures for each age group as well as lifetime adjusted daily dose (LADD) by partitioning exposure per age group into proportions of total lifespan and then summing the proportions into a 70-year lifespan.
Risk characterization.
The Q* value for permethrin is 9.567 × 10−3 mg/kg per day−1, a value that, when multiplied by the potential exposure, gives an estimate of numbers of cancer cases above a background level. Assumptions for exposure duration were 365 days of exposure over an entire lifetime of 70 years for all cancer and non-cancer risk estimates. Lifetime adjusted cancer risk was estimated by partitioning exposure per age group into proportions of total lifespan (LADD) and then summing the proportions into a 70-year lifespan.
Probabilistic analysis.
We used Monte Carlo simulation (Crystal Ball 7.3; Decisioneering, Denver, CO) to evaluate the RQ and input variables used to calculate the RQ. Probabilities of occurrence of RQ values were determined by sampling from the statistical distribution of each input variable used to calculate the RQs. Each of the input variables was sampled so that each input variable's distribution shape was reproduced. Then, the variability for each input was propagated into the output of the model so that the model output reflected the probability of values that could occur.
Analyses were performed using 20,000 iterations to calculate percentile values and other statistics for our estimates of risk. Sensitivity analysis was performed to examine which of the input variables contributed the most variability to the output variability of the model.
Results
Exposures to all of the insecticide active ingredients were by far greatest for the oral route in the child age groups (0–10 years). Oral exposure, primarily through sucking directly on the bed net, resulted in > 90% of the total exposure. Depending on the child age group, dermal and inhalation exposure only resulted in 0.9–7% and 0.07% of the total exposure, respectively. For youth > 10 years and adults, dermal and inhalation exposure was 99.7% and 0.3% of total exposure, respectively. As discussed above, we assumed no oral exposure for youth > 10 years or adults.
Risk quotients at the 50th and 90th percentiles for non-cancer risks were < 1.0 (i.e., exposure was less than toxicity threshold) for lifetime adjusted risk and all youth and adult age groups. Risk quotients at the 50th and 90th percentiles for lifetime adjusted risk were 0.33 and 0.42 for DuraNet, 0.27 and 0.34 for Interceptor, 0.19 and 0.25 for Netprotect, 0.24 and 0.45 for DawaPlus, 0.17 and 0.22 for PermaNet and Yorkool, and 0.02 and 0.03 for Olyset Net, respectively. Risk quotients for infants and toddlers (0–3 years) and child groups from 3 to 10 years were ≥ 1.0 for specific bed nets (Table 4). In particular, the LLINs treated with deltamethrin or α-cypermethrin had the largest RQs, but those quotients were generally < 4.0 at the 90th percentile.
Risk quotient percentiles for selected age groups
| Bed net (active ingredient) | Age group | |||||
|---|---|---|---|---|---|---|
| Infants and toddlers (0–3 years) | Child (3–4 years) | Child (4–10 years) | ||||
| 50% | 90% | 50% | 90% | 50% | 90% | |
| DuraNet (α-cypermethrin) | 3.04 | 4.09 | 2.80 | 3.18 | 1.31–1.96 | 1.77–2.76 |
| Interceptor (α-cypermethrin) | 2.50 | 3.36 | 1.87 | 2.60 | 1.07–1.87 | 1.44–2.60 |
| Netprotect (deltamethrin) | 1.58 | 2.13 | 1.22 | 1.70 | 0.72–1.06 | 0.96–1.49 |
| PermaNet 2.0, Yorkool (deltamethrin) | 1.38 | 1.86 | 1.07 | 1.48 | < 1 | 0.84–1.30 |
| DawaPlus 2.0 (deltamethrin) | 2.00 | 2.69 | 1.55 | 2.16 | 0.91–1.34 | 1.22–1.87 |
| Olyset Net (permethrin) | < 1 | < 1 | < 1 | < 1 | < 1 | < 1 |
Estimates of lifetime adjusted cancer risks to the population were 8.62 and 10.9 additional cancer cases above background levels per 1,000,000 people at the 50th and 90th percentiles of exposure, respectively (Table 5). When exposure to 0–10 year olds was excluded from the analysis (thus removing the large exposure from oral intake), the population cancer risks were 1.26 and 1.99 additional cancer cases above background levels per 1,000,000 people at the 50th and 90th percentiles of exposure, respectively.
Mean, 50th percentile, and 90th percentile cancer risk estimate for Olyset Net
| Age group | Percentile | Mean | |
|---|---|---|---|
| 50% | 90% | ||
| All ages | 8.6 × 10−6 | 10.9 × 10−6 | 8.7 × 10−6 |
| Excluding 0–10 years | 1.3 × 10−6 | 1.9 × 10−6 | 1.3 × 10−6 |
Sensitivity analysis (percent contribution of the input variable to the output variance) for all insecticides
| Input variable | Age group | ||
|---|---|---|---|
| Infants and toddlers (0–3 years) | Child (3–10 years) | All other age groups (10–70 years) | |
| Saliva extraction factor | 72% | 52–75% | – |
| Body weight | 28% | 24–46% | < 5% |
| Body surface area in contact with bed net | < 1% | < 1% | 54–99% |
| Dry transfer rate | < 1% | < 1% | 0–44% |
Sensitivity analysis of the exposure model for cancer risks revealed that the saliva extraction factor contributed 77% of the variance to the output variance followed by the body surface area in contact with the bed net (14.7%) and infant/toddler weight (5.7%). When saliva extraction factor was fixed at a deterministic value of 50%, uncertainty of the body surface area in contact with the net and infant/toddler weight contributed 67.6% and 23.5% of the variance to the output, respectively. When exposure to 0–10 year olds was excluded from the analysis, 99% of the variance to the output was the result of the uncertainty in the body surface area in contact with the bed net.
Sensitivity analysis of the exposure model for non-cancer risks revealed that, for children < 10 years old, saliva extraction factor and body weight contributed 52–75% and 28–46% of the variance to the output, respectively (Table 6). When saliva extraction factor was fixed at a deterministic value of 50%, infant/toddler weight contributed 99% of the variance to the output. For all other age groups, body surface area in contact with the net and the transfer coefficient from the dry net to the skin contributed 54–99% and 0–44% of the variance to the output, respectively.
Discussion
Our risk assessment provides regulatory decision makers with guidance in the event that the RQs or cancer risk estimates exceed regulatory levels of concern. The results reveal that inhalation exposure is negligible compared with dermal and oral exposure, and additional resources expended to better characterize this route of exposure for regulatory reasons would be questionable. The sensitivity analysis reported here identifies three primary input variables that could be refined to meaningfully improve future risk assessments: (1) saliva extraction factor, (2) body surface area in contact with the bed net, and (3) transfer coefficient from the dry bed net to the skin.
To reduce uncertainty associated with our risk estimates, saliva extraction factor would need to be better understood for childhood exposures. Currently, only one study has examined potential saliva extraction by removing insecticides placed onto aluminum foil with sponges wetted with human saliva and similar analogs.26 Because of the large potential exposure compared with inhalation and dermal exposures, child oral exposure to the bed net through hand to mouth contact and directly sucking on the net needs to be assessed. This would entail studies at night during sleep to determine to what extent infants and children engage in these oral behaviors.
To reduce uncertainty for ages > 10 years, the body surface area in contact with the net and the transfer coefficient from the dry net to the skin would need to be better understood. These input variables currently are highly uncertain. To be conservative, we assumed that each night over an entire lifetime, 10–60% of the person's unclothed body would contact the bed net. Provided that the bed net is used properly, the contact should be much less. However, based on the results of our assessment, it would be worthwhile to generate data on contact amount and frequency between skin and the bed net during sleep. This would directly address a large area of uncertainty in the exposure model.
Except for permethrin, which was fixed at 0.49%, we assumed the transfer coefficient from the dry net to the skin was a triangular distribution, with a minimum and most likely value of 0.49% and a high value of 2.5%. We used 0.49% for the most likely value, because the other pyrethroid insecticides are similar to permethrin. However, because there is no publicly available data on transfer coefficients for α-cypermethrin or deltamethrin when factory-impregnated onto fabric, we used the more conservative triangular distribution, with 2.5% as the maximum value. The 2.5% value is from carpet treated with cyfluthrin, another pyrethroid, but the treatment was not from a factory-impregnated process. Instead, it was from a spray used directly on the carpet and then allowed to dry.
Although probabilistic, our analysis used many conservative exposure values, and even at low percentile values, it likely overestimates the risks. For example, we assumed that all children grabbed 15 × 15 cm (225 cm2) of bed net and sucked on it each night over the first 10 years of their life. If the assumption is refined so that each child from 0 to 3 years does this only every third night, the non-cancer risk would be reduced by 66%. In addition, we assumed that the concentration of insecticide factory-impregnated on the LLIN would not degrade over time through numerous washings or other factors.
Our risk assessment can be used for new LLINs with different insecticide active ingredients or formulations. Although our focus in this paper was the risks presented by the LLINs when used in malaria management in Africa, our risk assessment model also can be used to make risk–benefit decisions for use of LLINs for many insect vector-borne diseases and nuisance situations. Elsewhere, we have conducted and reported the results of risk assessments that address other insect vector management tactics, such as insecticide-impregnated clothing,6 biological control,31 repellents,32 and outdoor space applications of insecticides.6,33–39 These studies, with the LLIN risk assessment presented here, can provide regulatory authorities and others with information for making improved risk–benefit decisions about personal protective measures and insect management tactics.
ACKNOWLEDGMENTS:
This study was supported by Montana State University and the Montana Agricultural Experiment Station.
- 1.↑
Fegan G, Noor AM, Akhwale WS, Cousens S, Snow RW, 2007. Effect of expanded insecticide-treated bednet coverage on child survival in rural Kenya: a longitudinal study. Lancet 370: 1035–1039.
- 2.↑
Lengeler C, deSavigny D, 2007. Programme diversity is key to the success of insecticide-treated bednets. Lancet 370: 1009–1010.
- 3.↑
Ghebreyesus TA, Lynch MC, Coll-Seck AW, 2008. The Global Malaria Action Plan for a Malaria-Free World. Geneva, Switzerland: Roll Back Malaria Partnership, 274.
- 4.↑
Teklehaimanot A, Sachs JD, Curtis C, 2007. Malaria control needs mass distribution of insecticidal bednets. Lancet 369: 2143–2146.
- 5.↑
Barlow SM, Sullivan FM, Lines J, 2001. Risk assessment of the use of deltamethrin on bednets for the prevention of malaria. Food Chem Toxicol 39: 407–422.
- 6.↑
Macedo PA, Peterson RKD, Davis RS, 2007. Risk assessments for exposure of deployed military personnel to insecticides and personal protective measures used for disease-vector management. J Toxicol Environ Health A 70: 1758–1771.
- 7.↑
Bomann W, 1995. How Safe Are Pyrethroids-Treated Mosquito Nets? An Evaluation Based on the Example of Solfac EW 050. Leverkusen, Germany: Bayer AG Animal Health Division.
- 8.↑
World Health Organization, 2004. A Generic Risk Assessment Model for Insecticide Treatment and Subsequent Use of Mosquito Nets. Geneva, Switzerland: World Health Organization.
- 9.
World Health Organization, 2007. Insecticide Treated Mosquito Nets: A Position Statement on ITNs. Geneva, Switzerland: World Health Organization, Global Malaria Programme.
- 10.↑
World Health Organization, 2009. Report of the Thirteenth WHOPES Working Group Meeting: Review of Olyset LN, DawaPlus 2.0 LN, and Tianjin Yorkool LN. Geneva, Switzerland: World Health Organization.
- 11.↑
World Health Organization, 2009. Report of the Twelfth WHOPES Working Group Meeting: Review of Bioflash GR, PermaNet 2.0, PermaNet 3.0, PermaNet 2.5, and Lambda-Cyhalothrin LN. Geneva, Switzerland: World Health Organization.
- 12.↑
World Health Organization, 2007. Report of the Tenth WHOPES Working Group Meeting: Review of Spinosad 0.5% GR & 12%SC, Lambda-Cyhalothrin 10%CS, K-O Tab 1-2-3, and Interceptor. Geneva, Switzerland: World Health Organization.
- 13.↑
World Health Organization, 2004. WHO Specifications and Evaluations for Public Health Pesticides: Long-Lasting (Incorporated into Filaments) Insecticidal Net. Geneva, Switzerland: World Health Organization.
- 14.↑
World Health Organization, 2001. Report of the Fifth WHOPES Working Group Meeting: Review of Olyset Net and Bifenthrin 10% WP. Geneva, Switzerland: World Health Organization.
- 15.↑
World Health Organization, 2008. Report of the Eleventh WHOPES Working Group Meeting: Review of Spinosad 7.48% DT, Netprotect, DuraNet, Dawaplus, and Icon Maxx. Geneva, Switzerland: World Health Organization.
- 16.↑
World Health Organization, 2004. Report of the Seventh WHOPES Working Group Meeting: Review of Vectobac WG, PermaNet, and Gokilaht-S 5EC. Geneva, Switzerland: World Health Organization.
- 17.↑
Najera JA, Zaim M, 2002. Malaria Vector Control. Decision Making Criteria and Procedures for Judicious Use of Insecticides. Geneva, Switzerland: World Health Organization.
- 18.↑
Zaim M, Aitio A, Nakashima N, 2000. Safety of pyrethroid-treated mosquito nets. Med Vet Entomol 14: 1–5.
- 19.↑
Monyeki KD, Cameron N, Getz B, 2000. Growth and nutritional status of rural South African children 3–10 years old: The Ellisras Growth Study. Am J Hum Biol 12: 42–49.
- 20.↑
Steyn K, Bourne L, Jooste P, Fourie JM, Rossouw K, Lombard C, 1998. Anthropometric profile of a black population of the Cape Peninsula in South Africa. East Afr Med J 75: 35–40.
- 21.↑
Portier K, Tolson JK, Roberts SM, 2007. Body weight distributions for risk assessment. Risk Anal 27: 11–26.
- 22.↑
United States Environmental Protection Agency, 1997. Exposure Factors Handbook. Vol I. General Factors. Washington, DC: US Environmental Protection Agency.
- 23.↑
United States Environmental Protection Agency, 2008. Child-Specific Exposure Factors Handbook. Washington, DC: US Environmental Protection Agency.
- 24.↑
United States Environmental Protection Agency, 2002. Office of Research and Development NCfEA, ed. Child-Specific Exposure Factors Handbook (Interim Report). Washington, DC: US Environmental Protection Agency.
- 25.↑
Snodgrass HL, 1992. Permethrin transfer from treated cloth to the skin surface: potential for exposure in humans. J Toxicol Environ Health 35: 91–105.
- 26.↑
Camann DE, Majumdar TK, Geno PW, 1995. Determination of Pesticide Removal Efficiencies from Human Hand Wiped with Gause Moistened with Three Salivary Fluids (Experiment I.3). San Antonio, TX: ManTech.
- 27.↑
World Health Organization, 2010. WHO Specifications and Evaluations for Public Health Pesticides: Permethrin. Geneva, Switzerland: World Health Organization.
- 28.↑
World Health Organization, 2008. WHO Specifications and Evaluations for Public Health Pesticides: Deltamethrin. Geneva, Switzerland: World Health Organization.
- 29.↑
World Health Organization, 2009. WHO Specifications and Evaluations for Public Health Pesticides: Alpha-Cypermethrin. Geneva, Switzerland: World Health Organization.
- 30.↑
United States Environmental Protection Agency, 2006. Reregistration Eligibility Decision (RED) for Permethrin. Washington, DC: US Environmental Protection Agency, 1–95.
- 31.↑
Schleier JJ III, Sing SE, Peterson RKD, 2008. Regional ecological risk assessment for the introduction of Gambusia affinis (western mosquitofish) into Montana watersheds. Biol Invasions 10: 1277–1287.
- 32.↑
Antwi F, Shama LM, Peterson RKD, 2008. Risk assessments for the insect repellents DEET and picaridin. Regul Toxicol Pharmacol 51: 31–36.
- 33.↑
Peterson RKD, Macedo PA, Davis RS, 2006. A human-health risk assessment for West Nile virus and insecticides used in mosquito management. Environ Health Perspect 114: 366–372.
- 34.
Macedo PA, Schleier JJ III, Reed M, Kelley K, Goodman GW, Brown DA, Peterson RKD, 2010. Evaluation of efficacy and human health risk of aerial ultra-low volume applications of pyrethrins and piperonyl butoxide for adult mosquito management in response to West Nile virus activity in Sacramento County, California. J Am Mosq Control Assoc 26: 57–66.
- 35.
Schleier JJ III, Peterson RKD, 2010. Toxicity and risk of permethrin and naled to non-target insects after adult mosquito management. Ecotoxicology 16: 1140–1146.
- 36.
Peterson RKD, 2010. Mosquito management and risk. Wing Beats 21: 28–31.
- 37.
Schleier JJ III, Macedo PA, Davis RS, Shama LM, Peterson RKD, 2009. A two-dimensional probabilistic acute human-health risk assessment of insecticide exposure after adult mosquito management. Stoch Environ Res Risk Assess 23: 555–563.
- 38.
Schleier JJ III, Peterson RKD, Macedo PA, Brown DA, 2008. Environmental concentrations, fate, and risk assessment of pyrethrins and piperonyl butoxide after aerial ultralow-volume applications for adult mosquito management. Environ Toxicol Chem 27: 1063–1068.
- 39.↑
Davis RS, Peterson RKD, Macedo PA, 2007. An ecological risk assessment for insecticides used in adult mosquito management. Integr Environ Assess Manag 3: 373–382.