A total of 122 human and animal Salmonella Typhimurium DT104 isolates and 6 epidemiologically related DT104b isolates from human and animal products were analysed by pulsed-field gel electrophoresis (PFGE). Genomic DNA was subjected to macrorestriction with three enzymes, SpeI, SfiI and XbaI. A total of 14 restriction fragment length polymorphism (RFLP) profiles were identified when the PFGE patterns from the three enzymes were combined. The majority of isolates (81·2%) exhibited the same RFLP profile. Six animal DT104 isolates, susceptible to enrofloxacin and resistant to naladixic acid, were identified from the antibiotic susceptibility test. Four of these isolates had a different PFGE profile from the common RFLP. In addition, 4 of the 6 isolates were geographically clustered in one region. It was concluded that there was one predominant strain of S. Typhimurium DT104 in Ireland and that the potential and selection pressures for emergence of fluoroquinolone-resistant isolates were present.

Resident and transient Escherichia coli strains from the colonic microflora of 13 Swedish schoolgirls were analysed for carriage of genes encoding a range of adhesins (P, type 1 and S fimbriae, Dr haemagglutinin and three varieties of the P fimbrial papG adhesin) and other virulence traits (K1 and K5 capsule, haemolysin and aerobactin) using multiplex PCR. Forty-four percent of the resident clones carried genes for P fimbriae, K1 or K5 capsule, and aerobactin, compared with only 3% of transient clones (P<0·0001). The P-fimbriated clones most often had the class II variety of the P-fimbrial adhesin gene papG and this adhesin was significantly associated with persistence of a strain. S fimbriae and type 1 fimbriae were equally common in resident and transient strains. The results indicate that not only P fimbriae, but also, certain capsules and the ability to produce the siderophore aerobactin might contribute to persistence of E. coli in the large intestine.

The carriage rate of a range of virulence genes was compared between resident and transient Escherichia coli strains obtained from the rectal flora of 22 home-delivered Pakistani infants 0–6 months old. Genes for the following virulence factors were assessed using multiplex PCR: P, type 1 and S fimbriae, three P fimbrial adhesin varieties, Dr haemagglutinin, K1 and K5 capsule, haemolysin and aerobactin. The E. coli strains examined here differed from those previously obtained from hosts in Western Europe in a lower prevalence of genes for P, S and type 1 fimbriae, K1 capsule and haemolysin. Nevertheless, genes for P fimbriae, the class II variety of papG adhesin, and aerobactin were significantly more common among resident than transient strains, as previously observed in a Swedish study. The results suggest that P fimbriae and aerobactin, previously implicated as virulence factors for urinary tract infection, might contribute to persistence of E. coli in the normal intestinal microflora.

The emergence of type 6B Streptococcus pneumoniae resistant to five antibiotics (penicillin, chloramphenicol, trimethoprim–sulphamethoxazole, erythromycin and tetracycline) in both the Northern Territory and Queensland prompted an investigation of the genetic relatedness and patterns of migration of the isolates. Pulsed field gel electrophoresis of genomic DNA of 74 multiple drug-resistant (MDR) isolates cultured in both regions between August 1988 and June 1997 showed that 100% of MDR isolates from the Northern Territory and 96% of MDR strains from Queensland were genetically indistinguishable or closely related to the index strain. None of a further 65 type 6B isolates that were resistant to one or two, or susceptible to all of the above antibiotics, were clonally related to the MDR pneumococci. The geographical distribution of the MDR type 6B clone increased over time. The index strain, first isolated in Darwin in August 1988, was identified in Brisbane, 2900km distant, less than 4 years later and subsequently in other Queensland centres. Surveillance programmes are important to monitor the emergence and spread of potentially invasive MDR pneumococcal clones in countries that are well serviced by air and road transport.

The Northern Territory of Australia has had historically very high incidence rates of invasive Haemophilus influenzae type b disease in children less than 5 years of age, with the burden of disease greatest among Aboriginal infants less than 12 months. This study documents the impact of conjugate Hib vaccines introduced in 1993. Immunization rates were monitored using an existing immunization register, and case finding was done retrospectively using hospital and laboratory records. Following the vaccine introduction, the incidence fell abruptly to a seventh of its pre-vaccination level, in both Aboriginal and non-Aboriginal children. The effectiveness of PRP-OMPC (PedvaxHIBTM) was 97·5% and the overall effectiveness of the vaccination programme was 86·3%. The study shows Hib immunization as an effective intervention while discussing continuing needs for Hib control in high risk populations. It also illustrates the benefit of immunization registers in the evaluation of immunization programmes and assessment of vaccine effectiveness.

Time trends and geographical variation of Helicobacter pylori antibodies in Finland were investigated by enzyme immunoassay in 20- to 34-year-old randomly selected females from six localities during 1969–73 (n = 375), and 15- to 45-year-old females representing nine communities and four geographical areas in 1983 (n = 882) and 1995 (n = 842). In the six communities investigated at three different time points, the overall prevalence declined from 38 to 12%, with an emphasis on the latter 12 years. The regionally varying rate of decrease in helicobacter prevalence changed the pre-existing geographical variation, leaving northern Finland with the highest rate. A 10%-units higher local helicobacter prevalence seemed to predict a 23% (95% CI 3–44%) higher gastric cancer incidence 20 years later. The overall decline in helicobacter seropositivity is consistent with earlier reports from Finland and other developed countries, and supports the cohort theory as an explanation for the age-related increase in H. pylori seroprevalence.

Though it is recognized that the extent of ‘clustering’ of isolates from tuberculosis cases in a given population is related to the amount of disease attributable to recent transmission, the relationship between the two statistics is poorly understood. Given age-dependent risks of disease and the fact that a long study (e.g. spanning several years) is more likely to identify transmission-linked cases than a shorter study, both measures, and thus the relationship between them, probably depend strongly on the ages of the cases ascertained and study duration. The contribution of these factors is explored in this paper using an age-structured model which describes the introduction and transmission of M. tuberculosis strains with different DNA fingerprint patterns in The Netherlands during this century, assuming that the number of individuals contacted by each case varies between cases and that DNA fingerprint patterns change over time through random mutations, as observed in several studies.

Model predictions of clustering in different age groups and over different time periods between 1993 and 1997 compare well against those observed. According to the model, the proportion of young cases with onset in a given time period who were ‘clustered’ underestimated the proportion of disease attributable to recent transmission in this age group (by up to 25% in males); for older individuals, clustering overestimated this proportion. These under- and overestimates decreased and increased respectively as the time period over which the cases were ascertained increased. These results have important implications for the interpretation of estimates of the proportion of disease attributable to recent transmission, based on ‘clustering’ statistics, as are being derived from studies of the molecular epidemiology of tuberculosis in many populations.

Objective: To identify risk factors for infectious intestinal disease (IID) due to rotavirus group A in children aged under 16 years.

Methods: Case-control study of cases of IID with rotavirus infection presenting to general practitioners (GPs) or occurring in community cohorts, and matched controls.

Results: There were 139 matched pairs. In children under 16 years the following risk factors were significantly associated with rotavirus IID: living in rented council housing (adjusted OR=3·78, P=0·022), accommodation with more than five rooms (OR=0·72, P=0·002), contact with someone ill with IID (OR=3·45, P<0·001). Some foods were associated with decreased risk. In infants, bottle feeding with or without breast feeding was associated with increased risk (OR=9·06, P<0·05).

Conclusions: Contact with persons with IID, living in rented council housing and accommodation with fewer rooms, were significant risk factors for sporadic rotavirus IID in children whereas breast feeding is protective in infants.

Objectives: To study the association between community influenza activity and acute hospital admissions for pneumonia and influenza among elderly persons.

Design: Multiple regression analysis of acute hospital admissions against community influenza activity, air temperature and seasonal and long-term trends.

Setting: Three English health districts: 1987–95.

Subjects: Persons aged [ges ]65 years.

Outcome measures: Acute hospital admissions for pneumonia and influenza (ICD9: 480–487); excess hospital admissions during epidemic periods.

Results: The final regression model explained 70% of the total variation in hospital admissions for pneumonia and influenza, including 14% due to community influenza activity. However, most variation was explained by long-term and seasonal changes unrelated to influenza. In the large influenza epidemic of 1989/90 a typical health district (500000 total population) experienced 56 excess admissions for pneumonia and influenza attributable to epidemic influenza among persons aged [ges ]65 years, requiring 672 additional bed-days. However the figure varied widely between seasons and over the whole study period, the average winter excess was 17·5 admissions per health district, requiring an additional 210 bed-days.

Conclusions: Influenza epidemics exert a variable impact on acute hospital admissions for pneumonia and influenza among elderly persons, which in the past have been poorly quantified. Although the absolute numbers of excess admissions is modest, their impact on bed availability may be considerable because of the duration of hospital stay in elderly persons.

We conducted an epidemiological study of a Japanese encephalitis (JE) outbreak in the southwestern part of Nepal in 1997. A high density of JE infections was found and it was estimated that 27·9% the total population were infected with JE virus in the study area. The fatality rate was 13·2% and there was no difference in the fatality rate between males and females over 5 years old. However, the case fatality rate was 2·1 times higher in females than in males (14·6% vs. 6·9%) among children under 5 years of age. Fifty-three blood samples were collected from suspected JE cases during the epidemic period in 1998. Findings for JE specific IgM revealed that clinical diagnoses of JE were serologically confirmed in an average 78% (70–93%) of patients in three collaborating hospitals. These studies demonstrated that JE was highly prevalent in the area and clinical diagnoses were reliable. Effective preventive measures should be taken against this vaccine-preventable disease.

A cohort of 214 drug addicts with serum hepatitis and a cohort of 193 hepatitis patients without drug addiction were examined in respect of death rates, causes of death and a number of risk factors for reduced survival. The death rate was significantly higher among the drug addicts than among non-addicts. The annual mortality rate was 1·5% in the drug addict group and 0·7% in the non-addict group. The highest relative risk of death was 860 for female drug addicts in age group 15–24 compared to females of the same age in the general population. The most prevalent cause of death in the drug addict group was drug overdose (53%), whereas in the other group 66% died from various somatic diseases. Hepatitis or complications of viral hepatitis played no role as cause of death among the drug addicts, and infections as a whole were also responsible for very few deaths. For male drug addicts, imprisonment before admission and leaving hospital without the doctors' permission were risk factors for early death.

The prevalence of blood-borne viruses in injecting drug users (IDUs) in Tayside, Scotland was determined by testing serum samples from IDUs who underwent attributable HIV antibody testing during 1993–7. The prevalence of antibodies to HIV was 29/802, (3·6%); to hepatitis C virus (HCV) 451/691, (65·3%); and to human T-cell leukaemia/lymphoma viruses type 1 and 2 (HTLV) 0/679, (0·0%). The prevalence of HIV and HCV antibodies were higher in subjects over the age of 25 (P=0·03 and P=0·001, respectively). During 1993–7 the prevalence of HCV fell only in younger female IDUs (P<0·01). HIV prevalence has declined dramatically since 1985, when a rate of 40% was recorded in similar populations. Harm reduction measures have failed to control HCV the spread of infection among IDUs in Tayside, as indicated by the high proportion of antibody positive IDUs, particularly males under the age of 25. Future studies should address the nature and effective reduction of continuing risk taking among IDUs in Tayside.

Enzyme immunoassay (EIA) was used to detect antibodies to Trichomonas vaginalis in sera from Zimbabwe. The EIA showed a sensitivity of 95 and 94% when compared with vaginal swab culture among women attending a family planning clinic (FPC) and female commercial sex workers (CSW) respectively. The specificity was 85 and 77% in the two groups. Culture-negative FPC women were sub-divided into high risk or low risk of exposure to trichomoniasis. The seroprevalence was 10% (6/61) among low risk women, 21% (10/48) among high risk women and 23% (9/39) among culture negative CSW. The EIA was positive in 46% (18/39) men with genital discharge but only 5% (2/37) healthy blood donors. None of 31 sera from prepubescent children was positive. The EIA may be useful for community surveys of trichomoniasis. Because T. vaginalis is a common sexually transmitted disease, the test may indicate behaviour that increases the risk of STD transmission.

Deterministic models were employed to investigate the biology of Trypanosoma evansi infection in the Indonesian buffalo. Models were fitted to two age-structured data sets of infection. The Susceptible–Infected–Susceptible (SIS) model was the best supported description of this infection, although the results of the analysis depended on the serological test used: the Tr7 Ag-ELISA was judged the most reliable indicator of infection. Estimated forces of infection increase with age from 1·2 to 2·0 acquisitions per buffalo per year. The buffaloes would clear infection in an estimated mean time period of 16·8 months (95% CIs: 12·5–25·9 months) since acquisition, either by drug treatment by owners or self-cure. A general discussion on the role of immunity in protozoan infections includes consideration that the fitted SIS model would be consistent with strain-specific immunity. The model may become a useful tool for the evaluation of control programmes.

We describe a major outbreak of pseudorabies virus (PRV) in a sow herd in which the sows were vaccinated simultaneously three times a year with a vaccine containing Bartha strain. Also in the associated rearing herd in which the gilts were vaccinated twice an outbreak of PRV occurred. The outbreak was analysed with mathematical models, statistical methods and Monte-Carlo simulation. Under the assumption that the outbreak started with one introduction of virus the reproduction ratio Rind – as a measure of transmission of PRV between individuals – in the sow herd was estimated with a Generalized Linear Model to be 1·6. Also under the assumption of one introduction of virus Rind in the rearing herd was estimated with a martingale estimator to be 1·7. Both estimates were significantly larger than 1. Mathematical analysis showed that heterogeneity in the sow herd, because of the presence of not-optimally immunized replacement sows could not be the only cause of the observed outbreak in the sow herd. With Monte-Carlo simulations, the duration of an outbreak after a single introduction of virus and Rind = 1·6 did not mimic the data and thus the hypothesis of a single introduction with Rind = 1·6 could also be rejected and Rind is thus, not necessarily above 1. Moreover, with statistical analysis, endemicity in the combination of herds as a cause for the observed outbreak could be rejected. Endemicity in the rearing herd alone could not be excluded. Therefore, multiple introductions from outside and most probably from the rearing herd were possibly the cause of the observed outbreak(s). The implications for eradication of pseudorabies virus were discussed.

Shiga toxin-producing Escherichia coli (STEC) O157 was investigated with respect to its halotolerance and whether it can survive in marine water. STEC O157 could multiply in a medium containing 5% NaCl and in sterilized marine water, and could survive in unsterilized marine water for at least 15 days. On the basis of these results, we postulated that STEC O157 may survive in natural marine water, and attempted to isolate the bacterium and Shiga toxin gene (stx) from marine water in Japan. The stx, comprising stx1 and stx2, was detected from marine water samples by PCR. STEC and other stx-positive bacteria, however, could not be isolated from these samples in this study. These results indicate that stx-positive bacteria may survive in marine water and suggest the necessity of a survey.

An outbreak of E. coli O157 infection occurred in the Highland Region of Scotland in the summer of 1999. The source of the outbreak was traced to an untreated private water supply. All six cases identified arose in visitors to the area, and most had very limited exposure to the contaminated water. Permanent residents on the same supply were unaffected. The E. coli O157 isolates from the water, sheep faeces collected from around the source and the human stool samples were indistinguishable using pulsed field gel electrophoresis. Previously reported outbreaks of E. coli O157 linked to potable water supplies have resulted from structural or treatment failures, which allowed faecal contamination of source water. Here, contamination of the water supply and subsequent human infection was due to the use of an untreated, unprotected private water source in a rural area where animals grazed freely.

In epidemiological studies of infectious disease, researchers often rely on specific cues of the host, such as clinical signs, as surrogate indicators of pathogen presence. A selection bias would manifest if the specific visual cues used in sampling for the pathogen were not representative of the full range of signs caused by the strains of that pathogen. In our molecular epidemiological studies of Escherichia coli associated with avian cellulitis in broilers, we collect carcasses at the processing plant based on visual cues of lesion morphology. Therefore, the objectives of this study were to: (1) explore the potential impacts of selection bias in an application of infectious disease epidemiology, and (2) utilize a validation protocol to assess the potential for selection bias in our molecular epidemiological studies of E. coli and avian cellulitis. In two different trials, E. coli DNA fingerprints were compared between birds that our observers collected and the birds that the observers missed. Using Fisher's exact tests and simulation models, we determined that the isolates collected by the observers were not significantly different from the isolates missed by the observers (P>0·60 in both trials). Our method of selecting birds suspected of having cellulitis did not significantly bias our inferences about the population of E. coli associated with cellulitis in the flock. We encourage more investigators to critically assess the relationship of the sample to the target population in epidemiological studies of infectious disease.

Outbreaks of Legionnaire's disease present a public health challenge especially because fatal outcomes still remain frequent. The aim of this study was to describe the abundance and epidemiology of Legionellaceae in the human-made environment. Water was sampled from hot-water taps in private and public buildings across the area of Göttingen, Germany, including distant suburbs. Following isolation, we used polymerase chain reaction in order to generate strain specific banding profiles of legionella isolates. In total, 70 buildings were examined. Of these 18 (26%) had the bacterium in at least one water sample. Legionella pneumophila serogroups 1, 4, 5 and 6 could be identified in the water samples. Most of the buildings were colonized solely by one distinct strain, as proven by PCR. In three cases equal patterns were found in separate buildings. There were two buildings in this study where isolates with different serogroups were found at the same time.