Probabilistic models in infectious disease epidemiology
and risk analysis of food production chains
Researchers: Kari Auranen, Jukka Ranta, Mervi Eerola, Elja
The work of the group is focused on dynamical modelling
of the spread of infection and pathogenic contamination in human and animal
populations. Common features of the methodology include data augmentation,
latent variables, hidden Markov models, and extensive use of Markov chain
Monte Carlo simulation in Bayesian hierarchical modelling. The work is mostly
joint research with the Department of Vaccines in the National Public Health
Institute (KTL) and the Department of Risk Assessment in the
National Veterinary and Food Research Institute
A major part of the collaboration with KTL has dealt with bacterial
infections. Their dynamics is characterised by transmission via
subclinical (asymptomatic) infections, temporary immunity against
re-infection and relatively rare clinical (observed) diseases. Specific
applications include development of Bayesian estimation methods for point
process models of transmission of recurrent infections: spread of
Haemophilus influenzae type b (Hib) bacteria in a family environment,
transmission of Streptococcus pneumoniae (Pnc) within families, and
studies on the time course from Pnc infection until disease, modified by
immunological responses, as well as the time course of a single outbreak
of meningococcal epidemics in a military garrison.
Auranen is currently leading one of the workpackages in an EU
research project (Pnc Euro, www.ktl.fi/pnceuro/ ). Specific aims in the
project include development and applications of data analysis in
multilevel transmission models, studies of Pnc transmission using
stochastic simulation models, and assessment of underlying Pnc disease
rates in surveillance systems with non-perfect sensitivity and
specificity. In addition, Auranen is involved in the analysis of data from
two major pneumoccocal vaccine trials run by KTL: FinOM (acute otitis
media in Finnish children) and ARIVAC (pneumonia in Filipino children).
In addition to meningococcal outbreaks, Ranta has previously
spatial distribution of childhood diabetes incidence and the detection
probability of poliovirus in an environmental sampling scheme. His current
work concentrates on risk analysis of health hazards in animal and food
production. Specifically, the spread of salmonella contamination in a
production chain from farm to fork has been modelled both for broiler meat
and beef meat. The models combine expert knowledge with data about
hierarchically structured production processes. In addition to modelling
the current situation, a special interest is in modelling the effect of
interventions under different possible scenarios by means of predictive
distributions. Occasionally, Ranta has taken part in jointly organized
short courses by giving lectures on probabilistic models for risk
Eerola has recently studied possibilities to use expert opinions
Bayesian information processing in the risk assessment of food production
chains where typically very little, or no measured data at all, is
available. Eerola has also been developing multilevel models to evaluate
protective immunity of vaccines by the aid of animal models instead of
large scale population trials.
Auranen K., Ranta J., Takala A.K. and Arjas E. (1996) A statistical
of transmission of Hib in a family. Statistics in Medicine 15: 2235-2252.
Ranta J., Pitkäniemi J., Karvonen M., Virtala E., Rusanen
J., Colpaert A.,
Naukkarinen A. and the DiMe study group (1996) Detection of overall
space-time clustering in a non-uniformly distributed population.
Statistics in Medicine 15: 2561-2572.
Ranta J., Mäkelä P.H., Takala A.K., and Arjas E. (1999)
course of meningococcal disease outbreaks in closed subpopulations.
Epidemiology and Infection 123: 359-371.
Auranen K., Eichner M., Käyhty H., Takala A.K. and Arjas
E. (1999) A
hierarchical Bayesian model to predict the duration of immunity against
Hib. Biometrics 55: 1306-1313.
Eerola, M., Mannila, H., Salmenkivi, M.: Frailty factors and
time-dependent hazards in modelling ear infections by BASSIST. In:
Proceedings of 13th Symposium of Computational Statistics in Bristol (R.
Payne and P. Green, eds.), Physica-Verlag: Heidelberg, 287-292, 1998.
Auranen K. (2000) Backcalculating the age-specific incidence
recurrent subclinical Haemophilus influenzae type b infection. Statistics
in Medicine 19: 281-296.
Auranen K. (1999) On Bayesian modelling of recurrent infections.
thesis. University of Helsinki. Rolf Nevanlinna Institute Reasearch
Reports A26. Yliopistopaino, Helsinki. ISBN 952-9528-52-3. ISSN 0787-8338.
Ranta J. and Penttinen A. (2000) Probabilistic small area risk
using GIS-based data: a case study on Finnish childhood diabetes.
Statistics in Medicine 19: 2345-2359.
Auranen K., Arjas E., Leino T., and Takala A. K. (2000) Transmission
pneumococcal carriage in families: a latent Markov process model for
binary data. JASA 95: 1044-1053.
Ranta J., Rytkönen M., Karvonen M. Näkökulmia
alueelliseen terveystutkimukseen ja tilastollisiin menetelmiin. (A few
aspects on spatially referenced data, health geographics and statistical
methods) Sosiaalilääketieteellinen aikakauslehti (Journal of Social
Medicine) 1999; 36: 285-293.
O'Neill, P., Balding, D., Becker, N., Eerola, M., Mollison,
of Infectious Disease Data from Household Outbreaks by Markov Chain Monte
Carlo Methods. Journal of the Royal Statistical Society (2000), C 49:
Edwards, J, van der Heijden, O., Eerola, M., Gay, N. (2000)
Rubella in Europe. Epidemiology and Infection 125: 617-634.
Leino, T. and Auranen, K. and Jokinen, J. and Leinonen, M and
P. and Takala, A.K. (2001) Pneumococcal carriage in children during their
first two years: important role of family exposure. Pediatric Infectious
Disease Journal, ISSN 0891-3668.20: 1022-1027.
Leino, T and Auranen, K. (2002) Voidaanko rokotuspäätösten
väestöstä ennustaa?,Duodecim 118: 57-61.
Puumalainen, T. and Zeta-Capeding, M. R. and Käyhty, H.
and Lucero, M. G.
and Auranen, K. and Leroy, O. and Nohynek, H. (2001)
Antibody response to an elevenvalent pneumococcal conjugate vaccine in
Filipino infants", The Pediatric Infectious Disease Journal, accepted
Mäkelä, P. H. and Käyhty, H. and Leino, T. and
and Peltola, H. and Lindholm, N and Eskola, J. (2001) Long term
persistence of immunity after conjugate vaccines, analysed with Hib
vaccine as the mode, submitted.
Leino, T., Auranen, K., Mäkelä, P. H., Käyhty,
H. and Takala, A. K.
Dynamics of natural immunity caused by subclinical infections; a case
study on Haemophilus influenzae type b (Hib)', to appear in Epidemiology
Ranta, J. and Hovi, T and Arjas, E. (2001) Poliovirus surveillance
examining sewage water specimens; studies on detection probability using
simulation models. Risk Analysis 21: 1087-1096.
Rytkönen, M and Ranta, J and Tuomilehto, J and Karvonen,
Bayesian analysis of geographical variation in the incidence of type I
diabetes in Finland",
Diabetologia 44 (Suppl. 3): B 37--B 44.
Kokki, E and Ranta, J. and Penttinen, A. and Pukkala, E. and
(2001) Small area estimation of incidence of cancer around a known source
of exposure with fine resolution data.
Occupational & Environmental Medicine 58 No. 5: 315-320.
Ranta, J. (2001) On probabilistic models for surveillance and
of disease incidence with latent processes: case studies on meningococcal
outbreaks, childhood diabetes and poliomyelitis. Ph.D. thesis. University
of Helsinki. Rolf Nevanlinna Institute Reasearch Reports A34.
Yliopistopaino, Helsinki. ISBN 952-9528-61-2 (nid.), 952-9528-62-0
( PDF). ISSN 0787-8338.
Ranta, J. and Maijala, R. (2001) A probabilistic transmission
model of salmonella in the primary broiler production chain. Risk Analysis,
accepted for publication.
Index | Research |Staff