Acute Illnesses and Injuries Related to Total Release Foggers — 10 States, 2007–2015

Ruiling Liu, Ph.D; Walter A. Alarcon, MD; Geoffrey M. Calvert, MD; Kathleen G. Aubin, MSPH; John Beckman; Karen R. Cummings, MPH; Lucia S. Graham, PhD; Sheila A. Higgins, MPH; Prakash Mulay, MBBS; Ketki Patel, PhD, MD; Joanne B. Prado, MPH; Abby Schwartz, MPH; Derry Stover, MPH; Justin Waltz, MPH


Morbidity and Mortality Weekly Report. 2018;67(4):125-130. 

In This Article

Abstract and Introduction


Total release foggers (TRFs) (also known as "bug bombs") are pesticide products often used indoors to kill insects. After an earlier report found that TRFs pose a risk for acute illness,[1] the Environmental Protection Agency required improved labels on TRFs manufactured after September 2012.[2] To examine the early impact of relabeling, the magnitude and characteristics of acute TRF-related illness were evaluated for the period 2007–2015. A total of 3,222 TRF-related illnesses were identified in 10 participating states, based on three data sources: Sentinel Event Notification System for Occupational Risk–Pesticides (SENSOR) programs, the California Department of Pesticide Regulation (CDPR) program, and poison control centers (PCCs) in Florida, Texas, and Washington. No statistically significant decline in the overall TRF-illness incidence rate was found. Failure to vacate treated premises during application was the most commonly reported cause of exposure. To reduce TRF-related illness, integrated pest management strategies[3] need to be adopted, as well as better communication about the hazards and proper uses of TRFs. Redesigning TRFs to prevent sudden, unexpected activation might also be useful.

Acute TRF-related illnesses were identified from the SENSOR programs in 10 participating states (2007–2015),* CDPR (2007–2014), and PCCs in Florida, Texas, and Washington (2007–2015). Complete PCC data were unavailable from the other seven states with SENSOR programs. Cases meeting all of the following criteria were included: exposure to TRFs with known active ingredients, at least two signs or symptoms related to or possibly related to TRF exposure, and no involvement of suicide or intentional harm to others. A total of 3,222 unique cases were identified.§ Cases were categorized as definite, probable, or possible based on case-level evidence. The magnitude, trends, and characteristics of acute TRF-related illnesses were assessed. Incidence rates were calculated using U.S. Census standard population estimates as denominators.[4] Poisson regression analyses were conducted to estimate incidence rate ratios (IRR) during 2013 (the first full year after label improvement when many TRF products on store shelves likely still had the old labels) and 2014–2015 (the period after label improvement when most TRF products likely had new labels) compared with 2007–2012 (the period before label improvements) for all cases and by reported causes of exposure, controlling for state to adjust for discordance in missing data across states. Stepwise logistic regression analysis was conducted to explore reported causes associated with more severe illness** (high or moderate versus low severity), adjusting for age, sex, and preexisting health conditions.

Overall, 3,573 cases were identified, including 1,843 from the SENSOR and CDPR programs and 1,730 from PCCs in Florida, Texas, and Washington (Table 1); 351 cases were identified from both the SENSOR programs and PCCs in Florida, Texas, or Washington, resulting in a total of 3,222 unique cases. Among cases from the SENSOR and CDPR programs, 87% were reported to the programs by PCCs; 6% were classified as definite, 20% as probable, and 74% as possible. After combining unique cases from the three data sources, the overall incidence rates in the 10 states during 2007–2012, 2013, and 2014–2015 were 27.0, 26.3, and 29.5 per 10 million population, respectively. The adjusted incidence rate did not change in 2013 or 2014–2015, compared with 2007–2012 (Table 2).

Five percent of cases occurred in children aged 0–5 years and 14% in adults aged ≥60 years (Table 1); the median age was 40 years. Approximately 56% occurred in females; 92% of exposures happened in private residences, and 91% were not work-related. Respiratory signs and symptoms (cough, upper respiratory pain or irritation, and dyspnea) and gastrointestinal signs and symptoms (vomiting, nausea, and abdominal pain or cramping) were the most commonly reported. Severity was classified as low, moderate, and high for 78%, 21%, and 0.7% of the illnesses, respectively. Four (0.1%) cases were fatal. Approximately 93% of cases involved exposure to the TRF active ingredients pyrethroid (78%) or pyrethrin (24%). The most commonly reported causes of exposure were failure to vacate treated premises during application, early reentry into treated premises, inability to vacate treated premises before TRF discharge, and inadequate ventilation of treated premises; approximately 4% of cases were caused by TRF discharge by children aged <13 years (Table 1). Incidence rates associated with failure to vacate premises during application increased during 2014–2015 compared with 2007–2012 (adjusted IRR = 1.39, p = 0.002), whereas rates related to excessive fogger use (i.e., using more foggers than necessary) decreased (adjusted IRR = 0.43, p = 0.001) (Table 2). Moderate or high severity illness were more common among males, persons aged >60 years, those with preexisting asthma, and those who failed to vacate premises during application, or who were exposed to excessive TRFs (Table 3).

*Under the SENSOR-Pesticides program, CDC provides cooperative agreement funding and technical support to state health departments to conduct surveillance for acute, occupational, pesticide-related illness and injury. Funding support is also provided by the Environmental Protection Agency. In 2017, a total of 13 states participated in this program. Data were available for this study for the period 2007–2013 in Oregon; 2007–2014 in California, Nebraska, and New York; and 2007–2015 in Florida, Louisiana, Michigan, North Carolina, Texas, and Washington.
In California, two programs identify cases of acute pesticide-related illness and injury; one is located at the California Department for Public Health (CDPH) and participates in the SENSOR program; the other is the Pesticide Illness Surveillance Program (PISP), administered by CDPR. PISP operates similarly to the SENSOR program, but the case definition and the variables used to characterize cases differ between the two programs ( Although PISP does not formally participate in the SENSOR program, both programs collaborate on joint activities. CDPH collects only work-related cases of acute pesticide-related illness and injury; PISP collects data for work-related and nonwork-related acute pesticide-related illness and injury. To ensure California cases were counted only once, CDPH cross-referenced its cases with those from PISP using name, date of illness and injury and, if available, Social Security number and date of birth.
§7,441 persons with TRF exposure were identified, and 3,222 unique cases remained after exclusions. The following persons were excluded: fewer than two signs or symptoms reported (3,638), suicide or intentional harm to others (24), and exposed to TRF products with unknown active ingredients (193). Also, 13 cases were reported by both the California SENSOR program and CDPR, and 351 by both SENSOR programs and PCCs in Florida, Texas, or Washington.
In the SENSOR program, cases are defined as definite (objective evidence is available to confirm exposure and health effects), probable (a combination of objective and self-reported data), and possible (self-reported exposure and health effects data) ( Cases from CDPR are categorized as definite (both physical and medical evidence documenting exposure and consequent health effects), probable (limited and circumstantial evidence supporting a relationship to pesticide exposure) and possible (health effects correspond generally to the reported exposure, but evidence is not available to support a relationship). Case categorization was not available for cases obtained only from PCCs in Florida, Texas, and Washington.
**Illness and injury severity for SENSOR and CDPR cases was categorized into four groups using the following standardized criteria: low severity (the condition usually resolves without treatment and <3 days are lost from work); moderate severity (the condition is not life-threatening but requires medical treatment; no residual impairment is expected, and time lost from work is ≤5 days); high severity (the condition is life threatening, requires hospitalization, often has >5 days lost from work, and might result in permanent impairment); and fatal ( For cases from PCCs in Florida, Texas, and Washington, case severity was based on the medical outcomes reported. Those cases reported with "death," "major effect," "moderate effect" and "minor effect" were classified as death, high severity, moderate severity, and low severity, respectively. PCC cases reported with "not followed, minimal clinical effects possible (no more than minor effect possible)" or "unable to follow, judged as a potentially toxic exposure" were also classified as low severity, unless, for those with "unable to follow, judged as a potentially toxic exposure," if the call to the PCC arose from a health care facility and the case had at least two moderate or high severity signs or symptoms (, then the case was classified as moderate severity.