Problem Statement

History and Characterization

Francisella tularensis is highly infectious facultative intracellular bacterium which causes Tularemia. It is a gram-negative bacterium that is a member of the Y-subdivision of Proteobacteria and its closest human pathogen relatives are Coxiella burnetti and Legionella pnuemophila . There are four recognized genetically diverse subspecies of F. tularensis, each of which have different localities in the world and diverse lethal doses in humans (See table II). Characteristics native to this organism such as low infective dose and simple aerosolizability have established F. tularensis as a threat in the CDC’s top list of biological threat agents.

Table II.

Tularemia pic 2.png

The main target of Francisella tularensis during its infective life cycle is to repliacate in macrophages, which can produce up to a 1,000-fold replication and production of harmful proteins . Though the mechanism by which Francisella replicates and survives within macrophages by way of phagosome, very few specific virulence factors have been determined other than a lipopolysaccharide and capsule in each of the subspecies. Since the early 1950s, fear of the use of F. tularensis, a pathogen much more infection than Bacillus anthracis, as a biological weapon has existed throughout the world. Not only did the Japanese germ-warfare program examine the possibility of intentionally causing Tularemia through a release of F. tularensis in humans, it has been alleged that offensive biological warfare programs in the former Soviet Union and the United States previously considered the weaponization of this pathogen during World War II . The potential effects of a release of F. tularensis in a population would inevitably be devastating. According to the WHO; the city and 1 million people would require preventive antibiotic treatment for at least 10 days. Even if exposed individuals were treated with antibiotics within 48 hours, 10% (25,000) of people would require hospitalization and 1% (2,500) would die.

In order to develop a plan of risk assessment, variables such as morbidity, mortality, strain virulence and method of pathogen delivery to an affected population must be considered.

Hazard Identification

Pathogenesis and Clinical Symptoms

Francisella tularensis infects humans through the skin, mucous membranes, gastrointestinal tract, and lungs. It is a facultative intracellular bacterium that multiples within macrophages. If not treated, the bacteria can multiply into skin or mucous membranes, spread to the regional lymph nodes and further multiply, and may disseminate to body . Symptoms usually appear 3 to 5 day after exposure to the bacteria, but can take as long as 14 days (CDC, 2008).

The primary clinical forms of tularemia vary in severity and presentation according to virulence of infecting organism, dose, and transmission route or site of inoculation (Table 1). The most frequent form of clinical manifestation is the ulceroglanular or granular. A recent epidemiology and molecular study carried out in USA with human isolates of F. tularensis recovered from 1964 through 2004 demonstrated a significant frequency of the respiratory and typhoidal forms.

Table 1. Clinical Symptoms and Routes of Acquisition

Form Symptoms Frequency Route of Acquisition
Ulceroglandular or glandular Skin ulcer, lymph node enlargement, fever, chills, headache 75-85% 65%* Vector-borne and direct contact
Oculoglandular Fever, conjunctivitis, photophobia, glaucomatous lesions <1% 4%* Touching the eye with contaminated fingers or possibly from infective dust
Oropharyngeal Ulcerative-exudative stomatitis, pharyngitis, lymphadenitis, vomiting, diarrhea 1%* Ingesting contaminated food or water
Respiratory Cough, chest pain, increased repiratory rate, fever, nausea, vomiting 17%* Inhaling contaminated dust or laboratory-acquired infection
Typhoidal Fever, weight loss, diarrhea and pain 12% Unknown (probably oral or respiratory)

Population Affected

Tularemia occurs mainly among previously healthy people and in all ages. Hayes et al (2002) in a report of national surveillance for the period of 1990 to 2000 detected that the average annual incidence of tularemia was highest in persons 5-9 years (rate: 0.10x105) and in person aged 75 years or greater (rate: 0.14x105). Staples et al (2006) studying cases of tularemia from the period of 1964 to 2004 found that patients with F. tularensis type A infections were significantly younger than patients with type B infection (median age 38 years x 50 years).

Tularemia has a higher incidence in males and seems that it is attributed to their more frequent outdoor professional and leisure activities (Eliasson et al, 2002). Since the cell-mediated immunity is mandatory to lost resistance against tularemia, a fulminant course of the disease may be expected in patients T-cell deficiency, such as lymphoma, HIV disease and in those undergoing corticosteroid or cytostatic treatment


Tularemia is not transmitted from human to human (WHO 2007). Tularemia has been reported in many countries of the northern hemisphere; none in the southern hemisphere (Table II.). It has been reported in all states in the United States, with the exception of Hawaii. Most cases occur in the south-central and western states. The majority of cases occur in rural areas. According to Dennis et al. (2001), the worldwide incidence of tularemia is unknown and the disease is probably under-reported. The cases in the United States have dropped dramatically from several thousand per year before 1950 to less than 200 per year in the 1990s . Cases occur more frequently in June through September, because disease transmitted by insects is more common. The United States cases are often occur sporadically and in small clusters, whereas, in Europe, waterborne, arthropod-borne, and airborne outbreaks that have affected hundreds of people have been reported.

Table II.

Tularemia pic 2.png

Exposure Assessment

In order to quantify and examine the risk posed to a community of people by a single infected rabbit in the water supply, it was first necessary to establish parameters. Assuming that the rabbit excretes between 15-60 g of feces per day with 1000000 CFU/g feces , the minimum concentration in the unchlorinated 7618.8 m3 water tank, ultimately producing a maximum of 7.875 CFU/L in the final water holding tank prior to distribution . After exiting the tank, the contaminated water will have two main ways of infecting the population at large, ingestion and inhalation. The average person in the UK, where the scenario is located, is approximately 140L . In order to estimate the average ingestions, a lognormal distribution of intake rate (1L/day) was assumed.

The two major exposure routes considered were in showering and drinking. In showering, aerosolization of F. tularensis and subsequent inhalation may occur. Ingestion can occur in drinking from a contaminated water source, where direct infection may occur in humans as well. In order to consider the most detrimental routes of exposure, dermal contact, which can produce infection, was considered but not modeled. This assumption is consistent with past Tularemia studies. Additional defined parameters were an exposure time of 24 hours, assuming that the situation was recognized and proper management measures were taken, and a completely diminished risk of infection when cooking with contaminated water because of the instability of Francisella tularensis at high temperatures. For inhalational exposure, the average inhalation rate of human adults was used as well as a lognormal distribution for time in shower, for which the average is 7 min/day . After a shower, the concentration of organisms in the air was assumed to be 5e-4, proportional to the concentration of organisms in the water for aerosolization purposes.

Dose Response

In order hypothesize the potential effects of an intentional release of F. tularensis into a public water supply; the potential for zoonotic release through an infected rabbit to a human population was used. For the purposes of this risk assessment, an average size adult rabbit purposely infected with Tularemia, type A, subspecies tularensis, strain M was put into a 7618.8 m3 water supply tank that supplies water to 100,000 homes with a mean of four inhabitants per household. The dose response data was modeled after an ongoing study at Michigan State University and was obtained through personal communication. For the inhalation model, an exponential model with risk parameter 0.0776 was used. This model was obtained through testing with aerosol exposures in rhesus monkeys. It was chosen because it had the highest risk of all other considered models and it was a conservation estimate given the lose dose. For the ingestion model, we chose a beta-Poisson model with parameters α= 0.281 and N50= 8473.8 obtained from animal study data on strain M based on concentrations in the lung. (See attached Dose-Response Data, Fig. 1)

Figure 1.

Tularemia pic 1.png

Risk Characterization

After running a Monte Carlo analysis accounting for all established parameters and distributions, it was determined that the risk of developing an infection through inhalation is highly unlikely, with an individual risk of approximately .000000001 in a community of 400,000 people. On the other hand, infection with F. tularensis through ingestion proved to be much more likely, with a mean risk of 695 out of 400,000 individuals. Clearly, the risk of infection is within ingestion and not in inhalational exposure. Considering the previously published literature, the finding of this risk assessment are consistent with outcomes establishing ingestion as the main cause of illness in populations affected by water contaminated with Francisella tularensis.

Risk Management & Communication

The major risk factors for contracting tularemia are contact with infected or contaminated animals, exposure to arthropod vectors (e.g., fleas, ticks, and mosquitoes), and infection from contaminated waterways and substrates (e.g., soil, sediment, mud). People that work or recreate in places where tularemia is endemic are at greater risk. F. tularensis may survive for extended periods in a cold, moist environment. Dennis et al. (2001) expect a short half-life due to desiccation, solar radiation, oxidation and other environmental factors, and a very limited risk from secondary dispersal. Awareness and knowledge about tularemia are low in the United States ; simple procedures can be carried out to prevent tularemia. To prevent specific exposure, latex or rubber waterproof gloves and a mask should be worn if handling dead or sick animals. Immediately following contact with suspect animals, water, or soil, washing skin with soap and water or alcohol gel should occur. According to Dennis et al. (2001), using insect repellant containing DEET on skin, treating clothing with repellent containing permethrin, cooking food thoroughly, and ensuring drinking water comes from a safe source are other preventative actions against tularemia. Sick animals should be isolated from humans and from each other. Surfaces should be disinfected by a bactericidal commercial product or 10 percent bleach solution . It is expected that standard levels of chlorine in municipal water sources should protect against waterborne infection. Personal protective clothing should be worn by health care workers, physicians, and emergency response workers when tularemia is suspected. Lab personnel should take special precautions and should be alerted to possible tularemia samples ahead of arrival. Routine diagnostic procedures can be performed in biological safety level 2 (BSL-2) conditions, further examination should be done at a BSL-3 level lab (e.g., at a state public health laboratory) .

In running this risk assessment it has become apparent that data from animal models as well as models including cutaneous exposure are necessary to further work concerning F. tularensis. In order to test for the presence of Tularemia in a water supply, it would be necessary to have a rapid detection method, such as Polymerase Chain Reaction (PCR). A 1MDS charged filter in the final holding tank will capture samples of the water, which can then be routinely tested for the presence of pathogenic organisms. Testing of the samples by bacterial DNA/RNA extraction will then quickly indicate the presence of a pathogen in the water allowing for prompt response by local Public Health Authorities and the F.B.I. in the case of an intentional release. For quantification purposes, quantitative PCR methods will provide information necessary to respond efficiently and effectively. Though it is known that Tularemia is a disease cause by a highly infectious and virulent bacterium, little is known about the actual pathogenic mechanism of Francisella tularensis.


Abril, C. “Rapid diagnosis and qualification of Francisella tularensis in organs of naturally infected common squirrel monkeys (Saimiri sciureus). Veterinary Microbiology. 127 (2008): 203-208

Bollin, "L. Pnuemophila In aerosols." Appl. Environmental Microbiology 50(1985): 1123-1130.

Bolin. C. Personal Communication. Michigan State University. 2008.

Burmaster. D.E. “A Lognormal Distribution for Time Spent Showering” Risk Analysis 3 (1987): 35-38

Berdal, B. P., R. Mehl, H. Haaheim, M. Loksa, R. Grunow, J. Burans, C. Morgan, and H. Meyer. 2000. Field detection of Francisella tularensis. Scand. J. Infect. Dis. 32:287–291.

Dennis, David. "Tularemia as a Biological Weapon." JAMA 285(2001): 2763-2770.

Ellis, J. et al. Tularemia. Clinical Microbiology Review, 15 (4): 631-646, 2002.

Eliasson, H. et al. The 2000 tularemia outbreak: a case control study of risk factors disease-endemic and emergent areas, Sweden. Emerging Infectious Diseases, *: 956-960, 2000..

Forsman M, Sandström G, Sjöstedt A(1994). Analysis of 16S ribosomal DNA sequences of Francisella strains and utilization for determination of the phylogeny of the genus and for identification of strains by PCR. International Journal of Systematic Bacteriology, 44:38–46.

Greco, D.. "A Waterborne Tularemia Outbreak." European Journal of Epidemiology 3(1987): 35-38.

Hayes, E. et al. Tularemia – United Staes, 1999-2000. MMWR, March, 2008.

Helvaci, S., S. Gedikoglu, H. Akalin, and H. B. Oral. 2000. Tularemia in Bursa, Turkey: 205 cases in ten years. Eur. J. Epidemiol. 16:271–276.

Jo, W.K. et al. “Routes of Chloroform Exposure and Body Burden from Showering with Chlorinated Tap Water”. Risk Analysis.May 2006. 10(4) pp 575-580.

Oyston, C.F.. "Tularaemia: bioterrorism defence renews interest in Francisella tularensis." Nature Reviews Microbiology 2(2004): 967-978.

Paerregaard, A. F Espersen, OM Jensen, M Skurnik - “Interactions between Yersinia enterocolitica and rabbit ileal mucus: growth, adhesion, penetration,” Infection and Immunity, 1991 59(1):253-60.

"Security of Supply 2006-2007 report" Sjöstedt A (2005). Francisella. In: Garrity G et al., eds. The Proteobacteria, Part B, Bergey’s Manual of Systematic Bacteriology, New York, NY, Springer, 200–210

Mark A. Suckow et al. The Laboratory Rabbit (Plastic Comb). Book 1997.

Stewart, S. J. 1996. Tularemia: association with hunting and farming. FEMS Immunol. Med. Microbiol. 13:197–199

Tarnvik, A., G. Sandstrom, and A. Sjostedt. 1996. Epidemiological analysis of tularemia in Sweden, 1931–1993. FEMS Immunol. Med. Microbiol. 13:201–204.

TE Ford - Microbiological safety of drinking water: United States and global perspectives. - Environmental Health Perspectives, 1999.

Titball, R.W.. "Francisella tularensis; an overview." ASM News 69(2003): 558

WHO. WHO Guidelines on Tularemia. , 2007. 241 p.

Wildlife Center of Virginia. 2007.