It is generally accepted that Professor Ignaz Semmelweis was the first to identify the mode of transmission of puerperal sepsis. However no appropriate statistical analysis of Semmelweis's data supporting his theory has been reported. Mean annual percent maternal mortality rates for the Allgemeines Krankenhaus and Dublin Maternity Hospitals (1784–1858) were analysed. The introduction of pathological anatomy at the Allgemeines Krankenhaus in 1823 was associated with increased mortality. After 1840 maternal mortality was higher in Clinic 1 which was staffed by male obstetricians and medical students who, unlike the midwives in Clinic 2, attended autopsies. The introduction of chlorine washing of the male clinicians’ hands in Clinic 1 by Semmelweis in 1847 reduced mortality, whereas the cessation of handwashing after Semmelweis left Vienna in 1850 was associated with increased mortality. This statistical analysis supports Semmelweis's hypothesis that ‘the cadaveric particles adhering to the hand had … caused the preponderant mortality in the first Clinic’.

The efficacy of vaccine when time since exposure is prolonged (more than 1 week from onset of illness in the index case) is unknown, but is likely to be significantly lower than human normal immunoglobulin (HNIG). We estimated the number of additional secondary cases that may occur through giving vaccine instead of HNIG to contacts of cases of hepatitis A who are identified more than 1 week after onset in the index case. This was calculated for different levels of vaccine efficacy, assuming HNIG efficacy to be 80–90%. The number of households that need to be treated to prevent one secondary case was calculated using estimates of secondary attack ratios (AR). If more than 1 week has elapsed from onset of illness in the index case, for an average household size of 2·3 people, a vaccine efficacy of 50% and an AR of 10–25%, 8–26 households would need to be treated with vaccine before one additional secondary case would be observed. As UK public health professionals manage around one hepatitis A case per month, it would take from 8 months to over 2 years for them to observe one additional case amongst contacts using vaccine rather than HNIG. It is unlikely that an average practitioner would notice if vaccine were 30% less effective than HNIG. Public health practice and advice to patients and contacts should be based on evidence as well as experience.

Capture–recapture analysis has been used to evaluate infectious disease surveillance. Violation of the underlying assumptions can jeopardize the validity of the capture–recapture estimates and a tool is needed for cross-validation. We re-examined 19 datasets of log-linear model capture–recapture studies on infectious disease incidence using three truncated models for incomplete count data as alternative population estimators. The truncated models yield comparable estimates to independent log-linear capture–recapture models and to parsimonious log-linear models when the number of patients is limited, or the ratio between patients registered once and twice is between 0·5 and 1·5. Compared to saturated log-linear models the truncated models produce considerably lower and often more plausible estimates. We conclude that for estimating infectious disease incidence independent and parsimonious three-source log-linear capture–recapture models are preferable but truncated models can be used as a heuristic tool to identify possible failure in log-linear models, especially when saturated log-linear models are selected.

The aim of this study was to investigate the burden of disease associated with gastroenteric viruses (rotavirus, norovirus, sapovirus, astrovirus and enteric adenovirus) using structured surveillance of children aged <6 years in the community. Faecal samples were collected between 2000 and 2003 from 685 children with symptoms of gastroenteritis. The children comprised three groups; 223 in the structured surveillance cohort, 203 in a community cohort and 259 in a cohort of hospitalized children. All samples were tested for the presence of viral pathogens using molecular methods. Questionnaires were sent to the parents/carers of the children recruited to the structured surveillance cohort in order to collect data that would allow an estimation of the severity of illness by means of the Vesikari score, and of the cost associated with gastrointestinal disease in this age group. A viral aetiological agent was detected in 53·5% of samples tested. Rotavirus was the most common pathogen found in all three cohorts followed by norovirus and enteric adenoviruses. Multiple viruses were found in 8% of the samples, and commonly involved rotavirus and any other virus. G1P[8] was the most commonly detected rotavirus strain and there was no significant difference in the distribution of rotavirus genotypes among the three cohorts. Analysis of the questionnaires indicated that rotavirus infections were likely to be more severe than any other virus infection, and children from whom a viral pathogen was identified were more likely to require rehydration therapy.

Rotavirus infections are the main cause of gastroenteritis in infants and children and it is expected that by the age of 5 years, nearly every child will have experienced at least one episode of rotavirus gastroenteritis. While severe cases are hospitalized, milder disease is either treated at home or by the GP, and as such the true prevalence of rotavirus infection in the community, and the burden of disease, is unknown. This paper reports the results of a cost-of-illness study which was conducted alongside a structured community surveillance study. Forty-eight percent of our sample was found to have rotavirus acute gastroenteritis; and the average total cost of a child presenting with rotavirus gastroenteritis ranged between £59 and £143 per episode, depending on the perspective. Given the prevalence and severity of the disease, the estimated burden of rotavirus gastroenteritis to society is £11.5 million per year.

Rotavirus is the most common cause of gastroenteritis in children aged <5 years old, two new vaccines have recently been developed which can prevent associated morbidity and mortality. While apparently safe and efficacious, it is also important to establish whether rotavirus immunization is cost effective. A decision analytical model which employs data from a review of published evidence is used to determine the cost effectiveness of a rotavirus vaccine. The results suggest that some of the health sector costs, and all of the societal costs, of rotavirus gastroenteritis in children can be avoided by an immunization programme. The additional cost to the health sector may be considered worthwhile if there is a sufficient improvement in the quality-of-life of children and parents affected by gastroenteritis; this study did not find any evidence of research which has measured the utility gains from vaccination.

This study compares the seasonality of rotavirus diarrhoeal hospital admissions and its relationship to climatic factors across three Australian cities. Weekly admission of rotavirus diarrhoea (1993–2003) in children aged <5 years and weekly average temperature and relative humidity for each city were modelled using a log-linear model with a cubic trend and season. Interactions were included to test for differences in the effect of temperature and humidity between seasons and between cities. Admissions of rotavirus diarrhoea peaked in winter and spring and were lowest in summer. Higher temperature and humidity in the previous week were associated with a decrease in rotavirus diarrhoeal admissions in three cities. The effects of both temperature and humidity on rotavirus admissions in Brisbane differed across seasons. Strategies to combat outbreaks of rotavirus diarrhoea should take climatic factors and seasonal effects into consideration to plan for the excess seasonal hospital admissions.

Improving the efficiency of outbreak investigation in restaurants is critical to reducing outbreak-associated illness and improving prevention strategies. Because clinical characteristics of outbreaks are usually available before results of laboratory testing, we examined their use for determining contributing factors in outbreaks caused by restaurants. All confirmed foodborne outbreaks reported to the Centers for Disease Control and Prevention (CDC) from 1982 to 1997 were reviewed. Clinical profiles were developed based on outbreak characteristics. We compared the percentage of contributing factors by known agent and clinical profile to their occurrence in outbreaks of unclassified aetiology. In total, 2246 foodborne outbreaks were included: 697 (31%) with known aetiology and 1549 (69%) with aetiology undetermined. Salmonella accounted for 65% of outbreaks with a known aetiology. Norovirus-like clinical profiles were noted in 54% of outbreaks with undetermined aetiology. Improper holding times and temperatures were associated with outbreaks caused by Clostridium perfringens, Bacillus cereus, Staphylococcus aureus, and Salmonella, and also with outbreaks of undetermined aetiology that fitted diarrhoea-toxin and vomiting-toxin clinical profiles. Poor personal hygiene was associated with norovirus, Shigella, and Salmonella, and also with outbreaks that fitted norovirus-like and vomiting-toxin clinical profiles. Contributing factors were similar for outbreaks with known aetiology and for those where aetiology was assigned by corresponding clinical profile. Rapidly categorizing outbreaks by clinical profile, before results of laboratory testing are available, can help identification of factors which contributed to the occurrence of the outbreak and will promote timely and efficient outbreak investigations.

To determine if a prediction of epidemic cholera using climate data can be made, we performed autoregression analysis using the data recorded in Dhaka City, Bangladesh over a 20-year period (1983–2002) comparing the number of children aged <10 years who were infected with Vibrio cholerae O1 to the maximum and minimum temperatures and rainfall. We formulated a simple autoregression model that predicts the monthly number of patients using earlier climate variables. The monthly number of patients predicted by this model agreed well with the actual monthly number of patients where the Pearson's correlation coefficient was 0·95. Arbitrarily defined, 39·4% of the predicted numbers during the study period were within 0·8–1·2 times the observed numbers. This prediction model uses the climate data recorded 2–4 months before. Therefore, our approach may be a good basis for establishing a practical early warning system for epidemic cholera.

Safe drinking water and hygiene are essential to reducing Kenya's diarrhoeal disease burden. A school-based safe water and hygiene intervention in Kenya was evaluated to assess its impact on students' knowledge and parents' adoption of safe water and hygiene practices. We surveyed 390 students from nine schools and their parents at baseline and conducted a final evaluation of 363 students and their parents. From baseline to final evaluation, improvement was seen in students' knowledge of correct water treatment procedure (21–65%, P<0·01) and knowing when to wash their hands. At final evaluation, 14% of parents reported currently treating their water, compared with 6% at baseline (P<0·01). From 2004 to 2005, school absenteeism in the September–November term decreased in nine project schools by 35% and increased in nine neighbouring comparison schools by 5%. This novel programme shows promise for reducing school absenteeism and promoting water and hygiene interventions in the home.

This study estimated the effect of discontinued use of antimicrobial growth promoters (duAGPs) on the risk of antibiotic treatment for diarrhoea, arthritis, pneumonia, unthriving and miscellaneous disorders in Danish pig farms. The estimation was done in a case-crossover study comparing: (1) the proportion of days per farm where treatment was performed (PDT) and (2) the proportion of pigs treated per day per farm at days where treatment was performed (PPT) before and after duAGPs at 68 farrow-to-finish farms. The farms were selected using a two-stage (veterinarian/farm) convenience sampling. On average, during the first year after duAGPs there was a significant increase in the risk of antibiotic treatment for diarrhoea (PDT: OR 2·5, 95% CI 1·7–3·8; PPT: OR 1·6, 95% CI 1·1–2·2). However, the effect varied among farms – some farms experienced substantial problems, while others experienced few problems after duAGPs. No effect was identified for the risk of treatment for other diseases.

A national, population-based laboratory surveillance of bloodstream infections (BSI) in Finland was performed. Blood-culturing rates were determined from data from clinical microbiology laboratories and trends in rates were evaluated using Poisson regression. During 1995–2002, 51 510 cases of BSI were notified; the annual incidence increased from 104 to145 cases/100 000 (40%). Rates increased in all age groups but persons aged ⩾75 years accounted for 28% of cases and showed the largest rate increase. Escherichia coli, coagulase-negative staphylococci, Staphylococcus aureus and Streptococcus pneumoniae accounted for 58% of isolates and their relative proportions were unchanged over time. The annual blood-culturing rate increased by one-third during the study period but the number of BSI detected per blood cultures remained unchanged. Regional BSI incidence was significantly associated with blood-culturing rates. We conclude that the increase in BSI rates may have been due to more frequent blood culturing but was not associated with changes in the reporting system or aetiology of BSI.

The prevalence, incidence and outcomes of haemolytic uraemic syndrome (HUS) and thrombotic thrombocytopaenic purpura (TTP) are not well established in adults or children from prospective studies. We sought to identify both outcomes and current management strategies using prospective, national surveillance of HUS and TTP, from 2003 to 2005 inclusive. We also investigated the links between these disorders and factors implicated in the aetiology of HUS and TTP including infections, chemotherapy, and immunosuppression. Most cases of HUS were caused by verocytotoxin-producing Escherichia coli (VTEC), of which serotype O157 predominated, although other serotypes were identified. The list of predisposing factors for TTP was more varied although use of immunosuppressive agents and severe sepsis, were the most frequent precipitants. The study demonstrates that while differentiating between HUS and TTP is sometimes difficult, in most cases the two syndromes have quite different predisposing factors and clinical parameters, enabling clinical and epidemiological profiling for these disorders.

The purpose of this study was to compare the occurrence of tuberculosis (TB) and the outcome of treatment between TB patients living in urban and rural areas. Cases of TB reported from 2001 to 2003 in England and Wales were assigned to a rural or urban area classification. The outcome of interest, non-completion of treatment, was investigated to determine the odds ratio for urban vs. rural residence. The effects of age, sex, ethnicity, place of birth, time since arrival in the United Kingdom, disease site, isoniazid resistance and previous diagnosis were adjusted for by multivariable logistic regression. Crude odds ratios showed a significantly higher level of treatment non-completion in rural areas. These results became non-significant (OR 1·02, 95% CI 0·83–1·26, P=0·82) after adjusting for the confounding effects of ethnic group and age. In England and Wales residence in a rural location is not an independent determinant of TB treatment outcome failure.

Viral infections are associated with pulmonary exacerbations in children with cystic fibrosis (CF), but few studies have addressed the frequency in adults. This paper investigates the frequency and impact of viral infections in adults with CF receiving intravenous antibiotics. Pre- and post-treatment spirometry, inflammatory markers and antibody titres against influenza A, influenza B, adenovirus, respiratory syncytial virus, Mycoplasma pneumoniae, Chlamydia psittaci, and Coxiella burnetti were analysed over a 10-year period. Non-bacterial infections were identified in 5·1% of 3156 courses of treatment. The annual incidence of admissions per patient associated with viral infection was 4·9%. The presence of viral infection in association with a pulmonary exacerbation did not adversely affect lung function or inflammatory markers in the short term. Adults with CF have a lower incidence of respiratory viral infections associated with pulmonary exacerbations requiring intravenous antibiotics compared to children and infants with CF.

Using the enumeration district (ED) block level this study looked at the incidence of dengue fever and dengue haemorrhagic fever (DF/DHF) within the Songkhla municipality in Thailand. Each of the 146 blocks in this area were considered as study units and surveyed for their environmental characteristics. A total of 287 cases of DH/DHF occurring in the year 1998 were selected for this study and the location of their homes mapped. Clustering analysis showed point clustering of the homes (P<0·0001) which was probably due to high density habitation, without any actual prevalence of case clustering. There was no evidence of clustering of the ED blocks with an incidence of DF/DHF (P=0·32). DF/DHF incidence for each block was strongly associated with the percentages of shop-houses, brick-made houses and houses with poor garbage disposal (all P<0·01). DF/DHF control should be emphasized for the areas which have a predominance of these housing types.