A step-wise risk assessment, using previously
accrued survey data, was used to estimate the risk of
acquiring cryptosporidiosis or giardiosis by consumption
of mung bean sprouts in Norway. The model is
simple and highlights the assumptions necessary, due
to data gaps, in making these estimates. As further
data become available the model can be expanded and
Table 4
Areas of uncertainty in risk assessment and where further data, or refined data, may improve risk assessment estimate
Further data required Reason for lack of data/why data may improve risk assessment
Prevalence and concentration of
contamination on mung bean sprouts
Limited survey size (only one supplier) with room for improvement in analytical method.
Genotype of oocysts/cysts detected No genotyping undertaken in survey; different infectivity patterns may be associated with
different genotypes. For this analysis, worst possible case scenario was used; i.e. that all
parasites were infectious to humans.
Viability of oocysts/cysts detected No viability assessment undertaken in survey; viability will affect likelihood of causing infection.
If all parasites non-viable, then risk of infection eliminated. Oocysts are known to be susceptible
to dehydration (Robertson et al., 1992), and therefore may not have survived seed drying and
storage procedures prior to sprouting. For this analysis, worst possible case scenario was used;
i.e. that all parasites were infectious.
Extent of mung bean sprout
consumption in Norway
Estimates made indirectly from some data on bean sprout production in Norway. Data from food
surveys may provide more accurate estimates.
Composition of consumer groups in
Norway in terms of risk.
Different consumer groups have different risks of infection/illness, as follows: Lowest risk group
(healthy children (non-infants) and adults with intact, fully functional immune systems and
previous exposure to these parasites). Low–medium risk group (healthy children (non-infants)
and adults with intact, fully functional immune systems and no previous exposure to these
parasites). Medium–high risk group (infants, elderly, pregnant, other dslightlyT
immunocompromised groups—may or may not have been previously exposed to these parasites).
High risk group (seriously immunocompromised (e.g. AIDS patients, transplant patients, etc.)
with no previous exposure to these parasites).
Mung bean sprout use in Norway This produce is widely associated with oriental cuisine. It is probable that a proportion
is used in oriental restaurant cuisine, potentially increasing population exposure. Bean sprouts
are frequently eaten raw (in salads, sandwiches), or, if cooked, only lightly (e.g. brief stir frying
in oriental cookery) or added as garnish to soup after cooking, so most cooking processes are
unlikely to impact greatly on parasite survival. However, this should be investigated. In
some cases, consumers may wash sprouts dat homeT. This may result in reduction in the extent
of contamination. However, as bean sprout production includes many washing steps and
parasites are detected on product for sale, home rinsing is unlikely to be highly effective.
L.J. Robertson et al. / International Journal of Food Microbiology 98 (2005) 291–300 299
refined to include these, and to provide more accurate
and realistic estimates. Finally, although there is
uncertainty, the estimates of risk are non-trivial and
as such would warrant some degree of attention by
public health authorities to better understand the
potential for risk from this source and subsequently
refine the analysis in order to better manage the risk