Coccidioidomycosis outbreak among inmate wildland firefighters: California, 2017

1 |. INTRODUCTION

Coccidioidomycosis (Valley fever) is an infectious disease endemic to the southwestern United States caused by spore inhalation of the soil-dwelling fungus Coccidioides spp. Approximately 97% of the US cases are reported in Arizona and California.¹ Although many infections are asymptomatic, approximately 40% lead to an influenza-like illness or pneumonia 1–3 weeks after exposure. Coccidioidomycosis can result in hospitalization and, in rare instances, the disease can disseminate or be fatal.²

In California, 7658 incident cases (19.3/100,000 population) were reported in 2017, the highest annual incidence in California since coccidioidomycosis became reportable in 1995.³ Outdoor workers in high-incidence coccidioidomycosis areas are at increased risk, particularly when job tasks involve soil disturbance (e.g., excavation work) or when conditions are windy or dusty; outbreaks have been reported in these populations, particularly among construction workers.^4–10 In 2017, a total of 599 wildfires were reported in the five California counties with the highest coccidioidomycosis rates.¹¹ The California Department of Forestry and Fire Protection (CAL FIRE) and the California Department of Corrections and Rehabilitation (CDCR) jointly operate a Conservation Camp program for the approximately 3000–4000 state prison inmates who work as wildland firefighters.^12,13 These inmates are essential to wildland fire response, perform most of their work using hand tools, and are known as hand crews.^13,14

In July 2017, five inmate firefighters became ill after responding to Wildfire A, a 4-day wildfire in one of the five California counties with coccidioidomycosis rates >50 per 100,000 in 2017; all later received coccidioidomycosis laboratory confirmation. On August 17, the California Correctional Health Care Services, which provides health care for state prison inmates, notified the California Department of Public Health (CDPH) and requested assistance. We investigated to identify additional cases and risk factors and possible prevention measures for coccidioidomycosis among inmate wildland firefighters.

2 |. MATERIALS AND METHODS

3. |. RESULTS

4 |. DISCUSSION

After a wildfire in an area with known Coccidioides proliferation, a coccidioidomycosis outbreak occurred among hand crew firefighters performing dirt-disturbing activities, some of whom had severe complications, including meningitis and respiratory failure. We found specific work practices associated with illness and a widespread lack of coccidioidomycosis training. In California, as both length and severity of the fire season is likely increasing,¹⁵ coccidioidomycosis risk among wildland firefighter hand crews performing similar work is likely to continue. Thus, providing wildland firefighters with a better understanding of their coccidioidomycosis risk and how to reduce risk is imperative.

Previous coccidioidomycosis outbreaks among California wildland firefighters have been reported. One occurred among 22 wards of a California Youth Authority facility that responded to a wildfire in Kern County in July 2000; job duties consisted of wards following bulldozers and putting out hotspots, and fire captains reported large amounts of dust in the air.¹⁶ Another study reported that the Coccidioides immitis positivity rate was 6% among Los Angeles County juvenile offenders placed in 10 work camps responding to wildfires during a 5-month period in 1965.¹⁷ Notably, both of these populations were correctional institution inmates, and most CAL FIRE hand crews are composed of inmate firefighters. Noninmate CAL FIRE wildland firefighters usually work in crews that primarily use fire engines, bulldozers, or aerial resources (e.g., helicopters), so the risks may be different, although they may also use hand tools at times.¹⁴ These outbreaks among wildland firefighters are likely due to substantial dirt-disturbance, which has been demonstrated in other coccidioidomycosis outbreaks occurring primarily among construction workers. The 2017 outbreak that we describe here occurred on a wildfire in a remote area where Coccidioides proliferates, with steep terrain not easily accessible by heavy equipment. Firefighter hand crews were essential to fire containment and control, demonstrating the unique risks associated with their job tasks.

Our investigation is the first to assess specific job tasks that may put wildland firefighter hand crews at increased risk for coccidioidomycosis. We found that being in a dust cloud or storm and conducting the dirt-disturbing activities of tossing dirt in the air and cutting fire line with a McLeod tool, and not the shovel or the Pulaski, were associated with illness. Tossing dirt in the air may be performed while digging fire lines or extinguishing hotspots with various hand tools. If wildland firefighters are cautioned against this practice or advised to stay upwind of soil disturbance, exposure to soil and dust containing Coccidioides might be decreased. The McLeod tool is used to scrape the fire containment line to expose soil after vegetation is removed by other hand tools, and possibly generates more dust than other tools (e.g., Pulaski). Wildland firefighters should be aware of the risk and understand how to reduce dust exposure when using the McLeod tool or other tools in a similar manner. Being in a dust cloud or storm may indicate the presence of airborne dust that was generated by wildland firefighting activities, wind, or both. However, we were unable to further distinguish the specific cause of the dusty conditions reported. This finding is similar to other outbreak investigations among solar farm construction workers, where frequently being in a dust cloud or storm was the strongest predictor of coccidioidomycosis.⁵

Despite their occupational risks for exposure to Coccidioides, inmate wildland firefighters did not wear respiratory protection during firefighting activities nor were they instructed to do so. Most did not remove dusty clothing and shoes or shower immediately upon returning to the fire camp; these practices could help prevent further exposure to Coccidioides after the work shift ends. Further, most hand crew firefighters were unaware that their jobs could increase coccidioidomycosis risk and received no training about recognizing symptoms or how to prevent exposure through dust control.

Our investigation had multiple limitations. First, because of logistical challenges during an active wildfire season, surveys were mailed 4–6 months after Wildfire A, which could have resulted in missed cases because of our inability to reach some parolees, difficulties recalling symptoms and exposures, and potential recall bias if case-patients remembered their exposures differently than controls. Second, there is a possibility for selection bias, if those with knowledge of coccidioidomycosis or those who were symptomatic were more likely to respond, which could result in an overestimation of the attack rate. It is also possible that individuals who were ill were less likely to respond or that those with mild or asymptomatic illness were not classified as clinical cases, which could result in an underestimation of the attack rate. We used a highly specific case definition to increase the likelihood that clinical cases were true coccidioidomycosis cases, and we required controls to be asymptomatic, resulting in the exclusion of symptomatic noncases from the case–control analysis. Sensitivity analyses that included these individuals as controls biased associations toward the null, likely due to outcome misclassification. It is possible that some of these individuals were actually cases with mild symptoms or with missing symptom onset date; those who met the onset date criteria were additionally classified as case-patients in further sensitivity analyses, where only the association with frequently being in a dust cloud or storm remained. The response rate among parolees was lower than among inmates; however, there is no reason to believe that parolees and inmates had different exposures or outcomes. Logistically, serologically testing all firefighters was not feasible. Because of the limited number of cases, estimates lacked statistical precision, which precluded our ability to perform more robust multivariable analyses.

As a result of this investigation, CDPH issued multiple prevention recommendations for wildland firefighters using hand tools, taking into consideration the industrial hygiene hierarchy of controls.¹⁸ Eliminating or substituting Coccidioides is not possible, so engineering controls would be the most effective strategy, followed by administrative controls. Personal protective equipment (i.e., respirators) is considered the least effective control measure and should not be used in place of other more effective control strategies. Agencies deploying wildland firefighters to areas where Coccidioides proliferates should minimize dust exposure when possible by wetting soil near areas of active digging (engineering control), instructing firefighters to stay upwind while performing firefighting activities (administrative control), and minimizing hand crew use when tasks could be performed by other personnel (administrative control) using heavy machinery in which operators are protected by enclosed cabs with regularly maintained air conditioning and filtration systems (engineering control). Vehicles transporting firefighters to and from fires on dirt roads should have windows closed and air conditioning on recirculation mode (engineering control). Firefighters should be made aware that dust on their clothes, shoes, and skin could contain Coccidioides and may continue to be a risk after returning to the fire camp. Employers should implement procedures that support proper storage and cleaning of dusty gear and changing into clean clothes as soon as possible. Additionally, agencies should locate fire camps and tents upwind and away from dust sources (administrative control).

CAL FIRE reported having a respiratory protection program for civilian wildland firefighters that included medical clearance, fit testing, and training to use particulate respirators, and having respirators available at fire camps. However, at the time of this investigation, the agency did not have a policy requiring or encouraging respirator use during active firefighting for protection against Coccidioides, and CDCR inmate wildland firefighters were not included in a respiratory protection program.¹⁹ CDPH recommended that CAL FIRE explore developing a policy on respiratory protection use by civilian and inmate wildland firefighters in high-incidence coccidioidomycosis areas. We suggested a pilot test to evaluate the feasibility of respirator use when wildland firefighters were away from active burning but performing specific soil-disturbing tasks (e.g., using the McLeod tool, digging) or in dusty areas; to our knowledge, this recommendation has not been pursued. Firefighting agencies have legitimate concerns regarding the use of respiratory protection on wildfires, including the strenuous and emergency nature of the work, potential flammability, and need for compatibility with other protective equipment worn. Given the multiple particulate respiratory hazards to which wildland firefighters may be exposed, we support future research and product development efforts that would make progress toward the feasible and effective use of respiratory protection in wildland firefighting.

CDPH recommends that all wildland firefighters working in areas where Coccidioides proliferates receive comprehensive coccidioidomycosis education at the time of initial employment, annually, and during safety deployment briefings, and that all trainings should be equivalent for inmate and civilian wildland firefighters. Trainings should highlight potential coccidioidomycosis risk, particularly when performing or in the vicinity of dirt disturbance and should educate wildland firefighters about high-incidence coccidioidomycosis areas and how to reduce dust exposure. Coccidioidomycosis is often underdiagnosed, in part, because of a lack of provider awareness, and patients who have knowledge of the illness may receive a diagnosis earlier.²⁰ Wildland firefighters should be also be trained on coccidioidomycosis signs and symptoms and how to report symptoms to supervisors so that they may be promptly evaluated by a healthcare provider. In California, providers must report any cases to the local health jurisdiction, and employers are required to report any hospitalized patients to California Division of Occupational Safety and Health.

5 |. CONCLUSIONS

Wildland firefighters performing dust-generating activities, such as those identified in this investigation, in areas where Coccidioides proliferates are at increased risk for coccidioidomycosis. This is concerning given that more wildland firefighters may be needed to respond to the increasingly severe wildfires in the United States, and limited information may be provided to these workers, especially if they are deployed from an area where coccidioidomycosis is uncommon. Illness risk factors identified in this investigation included being in a dust cloud or storm and specific dirt-disturbing job tasks, including digging a fire line with a McLeod tool and tossing dirt in the air. Despite challenges inherent in implementing control measures on active fires, agencies deploying wildland firefighters to high-incidence areas should train them on coccidioidomycosis symptoms, risk for illness from wildland firefighting, and risk reduction. Employers should limit dust exposure when possible and implement respiratory protection when feasible.