Environmental Engineering Reference
In-Depth Information
are based on values described by Havelaar
et al. (2000a) for illness in the general population
and illness reported to general practitioners. The
severity weight (0.28) and duration (365 days)
for complicated campylobacteriosis is based on
the mean value for Guillain-Barr
ยด
syndrome - a
disease of the peripheral nerves and occasional
complication of campylobacter infection (adapted
from Havelaar et al. 2000a), with the incidence
based on the campylobacteriosis hospitalization
rate of 0.5% reported byMead et al. (1999). Recent
mortality statistics (2001-2005) from the UK
indicate that fatalities from campylobacteriosis
occur in people aged over 65. An age-related
case-fatality rate of 0.0083% for the UK was esti-
mated, with a mean age of death of 80 (ONS 2007)
and a resultant loss of 8.3 years of life (based on
additional life expectancies for males and females
aged 80 of 7.57 years and 9.03 years respectively
(GAD 2007).
pathogens are usually those that present a worst
case combination (NRMMC 2006) of high occur-
rence, relatively high concentrations inflood com-
ponents and high pathogenicity.
Based on an examination of the literature relat-
ing to flood water quality (Smith et al. 2007,
Table 1) the most frequently identified bacterial
and protozoan pathogens are Campylobacter spp.,
Salmonella spp., Cryptosporidium spp. and Giar-
dia spp. Campylobacters are the most commonly
reported cause of bacterial gastroenteritis in the
UK, with an estimated incidence of 8.7/1000
population (Adak et al. 2002) and, on this basis,
were chosen as the bacterial representative. Cryp-
tosporidium infection is more prevalent than
Giardia infection (estimated incidence in the UK
of 8.1/10,000 and 5.4/10,000 respectively: Adak
et al. 2002). Cryptosporidium has also been more
frequently associated with waterborne illness
than Giardia in the UK, and was chosen as the
protozoan representative.
The choice for virus was less clear cut. Noro-
viruses are the most common cause of viral gas-
troenteritis in the UK, with an estimated annual
incidence rate of 12.1/1000 (Adak et al. 2002) but
there is no dose-response relationship for this
virus. Rotaviruses are the second most common
cause of viral gastroenteritis and a dose-response
relationship has been established (Haas and Eisen-
berg 2001), but concentrations cannot be deter-
mined by cell culture. A composite virus was
chosen based on concentrations of adenovirus
(which can be determined by cell culture) with
the dose-response characteristics of rotavirus.
Cryptosporidium spp.
Infection with Cryptosporidium spp. in devel-
oped countries is believed to result in illness in
the immunocompetent population in 71% of
cases, while infection in the immunocompro-
mised population is thought to lead to illness in
virtually 100%of cases (Havelaar et al. 2000b). The
severity weights and illness duration are 0.054 and
6 days, respectively, for illness in the normal pop-
ulation, and 0.13 and 47 days, respectively, in the
immunocompromised population. A case-fatality
rate of 0.0158% has been estimated from UK data
onage-related incidence, knownunder-reportingof
non-fatal cases and population statistics (FSA2000;
Adak et al. 2002; HPA 2007a; ONS 2007). The
mean age at death from cryptosporidiosis is
63.7 years (this figure was assumed to be 65).
The effect of HIV/AIDS on mortality from
cryptosporidiosis infection is unclear. A study
reported in 1987 showed that cryptosporidiosis
in patients with HIV/AIDS had a case-fatality
rate that was significantly higher (p
Hazard characterization
The reference pathogens have been outlined
above; this section summarizes the key character-
istics of the pathogens in terms of QMRA and
DALY quantification.
Campylobacter spp.
Infection with Campylobacter spp. proceeds to
clinical illness in 30% of cases (WHO 2004). The
severity weight (0.086) and duration (6 days)
for uncomplicated cases of campylobacteriosis
<
0.01) than
the case-fatality rate for patients without
reported HIV/AIDS (Navin and Hardy 1987).
Furthermore, Ruisin et al.
(2000) suggested
