S weden, 2003
Lara Payne,*† Malin Arneborn,†
Anders Tegnell,†‡ and Johan Giesecke†
Tularemia cases have been reported in Sweden since
1931, but no cyclical patterns can be identified. In 2003, the
largest outbreak of tularemia since 1967 occurred, involv-
ing 698 cases. Increased reports were received from
tularemia-nonendemic areas. Causal factors for an out-
break year and associated geographic distribution are not
a high prevalence of illness and death in humans (1) has
led to its inclusion in the ranks of potential bioterrorism
agents (2). However, endemic forms of a less virulent sub-
species of the agent also exist in parts of the Northern
Hemisphere. Bioterrorism concerns further the need for
surveillance of tularemia-endemic areas of the world, both
to learn more about the disease and to differentiate natural
from deliberate outbreaks (3). Sweden has had reported
cases of tularemia since 1931, with outbreaks of variable
magnitude, but with no cyclical patterns or trends (4). In
2003, 698 cases of tularemia were reported, the highest
number since 1967 (Figure 1); this outbreak was larger
than those usually observed (100–500 cases). Increased
numbers of cases were also reported as acquired outside
the identified tularemia-endemic region, similar to the sit-
uation in the 2000 outbreak (5). This article describes the
epidemiology of cases in 2003.
he ability of Francisella tularensis, the bacterial
pathogen of tularemia, to infect at low levels and cause
Tularemia has been a notifiable disease in Sweden
since 1968. Clinicians report diagnoses of tularemia, with
or without laboratory evidence, to the county medical offi-
cer (CMO) and the Department of Epidemiology, Swedish
Institute for Infectious Disease Control (EPI/SMI).
Reports from regional hospital microbiology laboratories
of positive cases of tularemia are also sent to SMI and
CMOs and are matched to clinical reports, on the basis of
a unique national personal identifying number. For this
analysis, we included all cases reported to SMI and regis-
tered in the database between January 1 and December 31,
Of the 698 cases reported in 2003, 591 were diagnosed
in 2003, of which 567 were reported to have been acquired
in Sweden (8 cases in Finland, 1 in Turkey, 15 not known).
A sharp peak of cases was observed in August, with cases
tapering off until December (Table 1). More male patients
(322, 57%) were reported than female patients, and the 45-
to 64-year age group was the most affected in both sexes
Of 522 cases that had a known route of infection, ani-
mal contact or insect bite was most common (n = 475,
91.0%), and 23 cases (4.4%) were reported to have been
acquired by inhaling contaminated dust or other material.
The remaining reports were of infection acquired by
another route (n = 22, 4.2%) or associated with work
(n = 2, 0.4%). Possible exposure risks of farm or outdoor
work were mentioned in an additional 25 cases, berry or
mushroom picking in 5 cases, and outdoor pursuits, such
as golf or fishing, in 6 cases. Clinical information about a
tularemia case is not systematically collected at the nation-
al level, but the oropharyngeal form of tularemia was men-
tioned in 1 case. This patient had acquired infection in the
northern coastal area of Sweden.
1440 Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 11, No. 9, September 2005
*European Programme for Intervention Epidemiology Training,
Solna, Sweden; †Swedish Institute for Infectious Disease Control,
Solna, Sweden; and ‡National Board of Health and Welfare,
Figure 1. Number of tularemia cases reported in Sweden by year
Overall, reports were received concerning infections
acquired in 15 of the 21 counties in Sweden (Figure 2).
Only 22% of patients in 2003 seemed to have been infect-
ed in what is considered the disease-endemic area of cen-
tral Sweden, and as many as 67% were infected in border
or disease-emerging areas (5). The main observations from
the largest outbreak of tularemia in Sweden since 1967
were that cases occurred earlier in the year, in greater num-
bers outside the disease-endemic area in Sweden, affected
those 45–64 years of age most frequently, and were main-
ly acquired through animal contact or insect bites.
We expected that the reporting of tularemia cases for
surveillance would be fairly complete because tularemia is
a notifiable disease in Sweden, and the public is interested
in the infection. An assessment of reporting completeness
to SMI identified 98.5%
84.6%–99.6%) for tularemia reports (1998–2002) (6).
However, mild or even undiagnosed cases are likely to be
missed by the reporting system.
The appearance of cases in midsummer is earlier than
in the 2000 outbreak (Table 1). Previous outbreaks have
usually began in late summer or early autumn (4). Since
>90% of cases reported infection through insect bites or
animal contact, this seasonal shift may be linked to climat-
ic or ecologic factors in a particular year. Cases are regis-
tered in the SMI database by date of reporting and not date
of diagnosis. However, median delay between date of sam-
ple collection and date of reporting for the period 1998 to
2002 was 11 days (interquartile range 8–19 days) (7).
Therefore, the seasonality observed in the epidemic curve
of 2003 cases was likely not greatly affected by possible
Responses regarding the route of infection may be
biased because many persons in Sweden associate
tularemia with mosquito bites. Nonetheless, the seasonal
distribution of cases observed would support this as a route
of infection. The age and sex distribution did not differ
from those of the last large outbreak in 2000 and likely
reflect the age and sex distribution of persons working out-
doors in farms or gardens in rural areas.
Sporadic tularemia cases outside of the disease-endem-
ic north-central region and northern coastal region have
been recorded since 1931 in Sweden. As in the 2000 out-
break, cases were reported from counties that previously
had very few reports (4). However, even more reports were
received in 2003 than in 2000 of infections acquired outside
the tularemia-endemic area (427 [78%] of 544 cases in
2003 vs. 227 [61%] of 370 in 2000, chi-square test = 31.79,
p<0.001; Figure 2). The geographic distribution of
tularemia seems to be changing. Public awareness and
Tularemia Outbreak, Sweden, 2003
Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 11, No. 9, September 2005 1441
Figure 2. Tularemia cases by county of probable infection, Sweden. A) Areas in analysis by Eliasson et al. B) 1981–2001 cases (N =
1,566). C) 2000 outbreak (N = 384). D) 2003 outbreak (N = 567). White areas indicate no reports.
changes in behavior for seeking medical care could con-
tribute to such a change, but no evidence suggests that
these factors differ from those in previous years.
The factors affecting the likelihood of an outbreak year
remain unknown, principally because the natural reservoir
of infection has yet to be identified (8). After the outbreak
in Sweden in 2000, a case-control study found significant
associations between acquiring tularemia and being bitten
by a mosquito, doing farm work, and owning a cat (5). The
role of ticks (9) and mosquitoes as potential vectors for
tularemia needs further investigation. Despite the interest
and awareness of tularemia in Sweden, much remains to be
understood about the dynamics of this infection among
reservoir, vector, and human populations.
Ms Payne is an epidemiologist and fellow in the European
Programme for Intervention Epidemiology Training, based at
the Swedish Institute for Infectious Disease Control. Her
research interests include vectorborne disease epidemiology and
1. Dennis D, Inglesby TV, Henderson DA, Bartlett JG, Ascher MS,
Eitzen E, et al. Tularemia as a biological weapon—medical and pub-
lic health management. JAMA. 2001;285:2763–72.
2. World Health Organization. Public health response to biological and
chemical weapons. WHO guidance. 2nd ed. Geneva: The
3. Grunow R, Finke E-J. A procedure for differentiating between the
intentional release of biological warfare agents and natural outbreaks
of disease: its use in analyzing the tularemia outbreak in Kosovo in
1999 and 2000. Clin Microbiol Infect. 2002;8:510–21.
4. Tarnvik A, Sandström G, Sjöstedt A. Epidemiological analysis of
tularemia in Sweden 1931–1993. FEMS Immunol Med Microbiol.
5. Eliasson H, Lindbäck J, Nuorti JP, Arneborn M, Giesecke J, Tegnell
A. The 2000 tularemia outbreak: a case-control study of risk factors
in disease-endemic and emergent areas, Sweden. Emerg Infect Dis.
6. Jansson A, Arrneborn M, Ekddahl K. Sensitivity of the Swedish statu-
tory surveillance system for communicable diseases 1998–2002,
assessed by the capture-recapture method. Epidemiol Infect.
7. Jansson A, Arneborn M, Skarlund K, Ekdahl K. Timeliness of case
reporting in the Swedish statutory surveillance of communicable dis-
eases 1998–2002. Scand J Infect Dis. 2004;36:865–72.
8. Tärnvik A, Priebe HS, Grunow R. Tularemia in Europe: an epidemi-
ological overview. Scand J Infect Dis. 2004;36:350–5.
9. Parola P, Raoult D. Ticks and tickborne bacterial diseases in humans:
an emerging infectious threat. Clin Infect Dis. 2001;32:897–28.
Address for correspondence: Lara Payne, Department of Epidemiology,
Swedish Institute for Infectious Disease Control, Tomtebodavägen 19A,
171 82 Solna, Sweden; fax: 46-8-300-626; email: Lara.Payne@ smi.ki.se
1442Emerging Infectious Diseases • www.cdc.gov/eid • Vol. 11, No. 9, September 2005
EID Covers Online
Image created by Curt Wommack