Yellow fever (YF), an acute infectious disease, is endemic in the north and central-west of Brazil. This disease can be prevented by the use of a vaccine. In Brazil, four fatal adverse events have been associated with the YF vaccine used in the country (17DD vaccine). We briefly describe the last two fatalities, and estimate the risk of 17DD-associated fatal adverse events under different epidemiological scenarios. Controversies regarding the appropriate denominator that enters the estimation of risk serve as a motivation for each proposed scenario. The statistical procedures used show optimum behaviour when assessing the risk of rare events. Risk estimates vary from 0·043 (95% CI 0·017–0·110) to 2·131 (95% CI 0·109–12·071) fatalities per million doses administered. The robust estimates of the risk of fatal adverse events we present constitute an important element in future risk–benefit analysis and point to the need for good quality vaccine coverage and adverse-events surveillance data to assess the risk of vaccination. Although vaccination of YF endemic regions is necessary to maintain low disease prevalence, preventive administration of YF vaccine to the entire population should be cautiously analysed.

An individual-based stochastic simulation model was constructed to study the epidemiology of Haemophilus influenzae type b (Hib) transmission, immunity and invasive disease. Embedded in a demographic model, the transmission model of Hib carriage employs the most important social mixing patterns with three types of contact sites (family, day-care group, and school class). The model includes immunity against invasive Hib disease, initiated and boosted by Hib carriage and cross-reactive bacterial encounters. The model reproduces the observed age patterns in Hib carriage and disease in Finland before large-scale use of the Hib conjugate vaccines. The model was used to investigate characteristics of Hib transmission. The analysis emphasizes transmission between children and adults in families while pointing out the importance of pre-school and school-aged children in maintaining Hib circulation. Carriage in these age groups is thus identified as being essential to target for sustained effects of interventions by vaccination.

We used a structured population model to study factors determining the magnitude of indirect protection in Haemophilus influenzae type b (Hib) vaccination. On a simulation platform mimicking the population of Finland, a Hib transmission and immunity model, including cross-reactive bacterial encounters, was formulated. Utilizing different vaccination coverages and vaccine types we could study how fast the incidence of Hib disease declined due to direct and indirect vaccination effects. With the Finnish vaccination schedule we could reproduce the observed disappearance of Hib cases. Our results show that an indirect effect was already significant with a relatively low vaccine coverage, even with a vaccine only partly reducing carriage acquisition. This suggests that the vaccination schedule and vaccine to be used should be chosen to result, in addition to immunological memory, in high antibody concentrations, sufficient to reduce carriage, the latter being the main factor behind successful elimination of transmission and disease.

Giardiasis surveillance data as well as drinking water, socioeconomic and land-use data were used in spatial regression models to investigate determinants of the geographic distribution of endemic giardiasis in southern Ontario. Higher giardiasis rates were observed in areas using surface water [rate ratio (RR) 2·36, 95% CI 1·38–4·05] and in rural areas (RR 1·79, 95% CI 1·32–2·37). Lower rates were observed in areas using filtered water (RR 0·55, 95% CI 0·42–0·94) and in those with high median income (RR 0·62, 95% CI 0·42–0·92). Chlorination of drinking water, cattle density and intensity of manure application on farmland were not significant determinants. The study shows that waterborne transmission plays an important role in giardiasis distribution in southern Ontario and that well-collected routine surveillance data could be useful for investigation of disease determinants and identification of high-risk communities. This information is useful in guiding decisions on control strategies.

The objective of this study was to assess the serological association between previous Chlamydia trachomatis infection and subfertility in a general population sample. A nested case (n=493)-control (n=986) study in a population-based birth cohort consisting of 12058 live births from the year 1966 was conducted. The analysis was restricted to those 6007 cohort members who replied to a postal inquiry and participated in a health examination including blood samples at the age of 31 years. The presence of C. trachomatis-specific serum IgG antibodies was screened by a synthetic peptide-based enzyme-linked immunosorbent assay. All the positive sera were further tested by the microimmunofluorescence method using immunotype pools and individual immunotypes of C. trachomatis as antigens. An association was found between the detection of immunotype-specific C. trachomatis antibodies and subfertility both in men and women. The results of the present study confirm the serological association between past C. trachomatis infections and subfertility in male or female partners of the couple in the population-based sample.

Vibrio cholerae is an important human pathogen and the cause of cholera. Since genetic variation and antibiotic resistance of strains have implications for effective treatment of the disease, we examined the genetic diversity and antibiotic resistance profile in 92 clinical strains (serogroup O1) and 56 environmental strains (O1 antigen, 42 strains; non-O1 antigen, 14 strains) isolated in Brazil between 1991 and 1999. Clinical and environmental O1 strains showed greater drug resistance compared to environmental non-O1 strains. Nearly all clinical O1 strains were resistant to one or more antibiotics while half of the environmental O1 and non-O1 strains were resistant to one or more antibiotics. No plasmids or class 1 integrons were detected in the strains by PCR analysis. Multilocus enzyme electrophoresis analysis (MLEE) suggests most of the O1 strains belong to a single (South American) clone that is related but different to seventh-pandemic strains isolated from other parts of the world. Our results show that there is a close genetic relationship between clinical and environmental O1 strains and that many serogroups and the environment can be a reservoir for antibiotic resistance.

Questionnaire surveys were implemented to study the incidence and physicians' knowledge of Vibrio vulnificus infections in Japan. Registered emergency physicians were selected by stratified random sampling for a questionnaire survey. A total of 235 out of 386 physicians (61%) responded to the questionnaire and 12 V. vulnificus septicaemia cases were reported from 10 respondents. The annual estimated number of V. vulnificus septicaemia was calculated as 425 (95% CI 238–752). The study also revealed that only 15·7% (95% CI 11·3–21·0) of responding physicians had a basic knowledge of V. vulnificus infection. Education for both physicians and people in the high-risk group for developing the infection (e.g. immunocompromised, chronic liver disease) will be necessary for the prevention, early diagnosis and appropriate treatment of the disease.