Variation in the incidence of invasive pneumococcal disease across South and West England, in 1995, was measured through a survey of microbiology laboratories. A 100% response rate was achieved. The incidence by laboratory varied between 5·2 and 20·4 per 100000 catchment population (P<0·001). Adjusting for pneumococcal vaccine uptake rate in over 65 year olds, hospital admission rates, blood culture system used and for the age and sex structure of the population, did not account for this variation. When blood culture sampling rates were included in a logistic regression model, the variation between laboratories was much less and of lower statistical significance (P=0·019). Higher rates of blood culture sampling were associated with a higher incidence of invasive pneumococcal disease. Consistently high sampling should be encouraged because a higher diagnostic rate should result in more selective prescribing of antibiotics, and secondly because improved ascertainment of severe pneumococcal infections is a prerequisite for the evaluation of new pneumococcal conjugate vaccines.

One hundred and six patients suffering from severe dehydrating diarrhoea were studied of whom 36 patients were positive for Vibrio cholerae. Out of 36, 15 were positive for V. cholerae O1, 10 for V. cholerae O139 and 11 for V. cholerae non-O1 non-O139. O1 and O139 were positive for the 301-bp ctxA amplicon and 471-bp tcpA amplicon indicating that the strains possessed toxigenic capability whereas no non-O1 non-O139 strain possessed ctxA or tcpA genes. Post-admission severity of purging and amount of ORS required were less in the V. cholerae non-O1 non-O139 group (P<0·05) compared to the V. cholerae O1 and O139 groups. It appears from this study that a cholera-like clinical condition can be caused in the absence of CT as exemplified by strains of non-O1 non-O139.

An outbreak of Salmonella enterica serotype Typhimurium belonging to phage type (PT) 193 occurred in autumn 1995 and involved 83 individuals in a large area of Northern Italy (Lombardy Region). Epidemiological and microbiological investigations of strains isolated from clinical and food specimens revealed that the vehicle was a batch of salami, produced on 4 September 1995 by a local firm. The outbreak was contained when the batch was withdrawn from sale. Insufficient ripening of the salami had allowed the salmonella to survive, emphasizing the need to define criteria for the ripening process of foods such as salami.

A collection of Salmonella enterica serotype Typhimurium human strains isolated in Northern Sardinia (Italy) was examined for the insertion sequence IS200, phage type, antibiotic profile, ribotyping polymorphisms and plasmid profile. All clinical isolates studied contained from 4 to 10 copies of the IS200 element. IS200 permitted to discriminate Typhimurium strains and to identify five IS200 types, some of them circulating in Sardinia at least since 1900. Strains belonging to phage DT104 predominated and correlated with a specific IS200 pattern.

Eight cases of Salmonella senftenberg infection in infants were identified in the first half of 1995 in England, five were indistinguishable S. senftenberg strains. A case-control study showed an association between illness and consumption of one brand of baby cereal (P=0·03). The cereal manufacturer reported isolating S. senftenberg in June 1994 from an undistributed cereal batch. Outbreak strains and the cereal strain were all plasmid-free in contrast to other human isolates of S. senftenberg in the same period. Changes in the production process were implemented to prevent further contamination.

Surveillance centres should strengthen the detection and investigation of outbreaks of gastrointestinal infection in susceptible groups, especially young children. In this outbreak, the study of only five cases led to identification of the vehicle of infection. Even when few cases are reported, epidemiological investigation in conjunction with molecular typing may lead to public health action which prevents continuing or future outbreaks.

The 1994 cholera epidemic in Guinea-Bissau resulted in 15878 reported cases and 306 deaths. Early in the epidemic, although the health ministry mandated that the bodies of persons dying of cholera be disinfected, outbreaks occurred in several villages following funerals in the region of Biombo. To determine the influence of disinfection and funeral activities on cholera transmission, we analysed surveillance data and conducted a case-control study following a funeral. The attack rate during the week following funerals was higher in villages where bodies were not disinfected (risk ratio = 2·6, 95% confidence interval [CI] 1·9–3·8). Cholera was strongly associated with eating at a funeral with a non-disinfected corpse (odds ratio [OR] = 14·5, 95% CI 0·9–786) and with touching (i.e., transporting, washing) the body (OR = 36·2, 95% CI 2·6–1769). During cholera epidemics, in addition to other cholera prevention activities, health officials should inform community leaders about the risk of cholera transmission during funerals, meals should not be served at funerals, and bodies of persons dying of cholera should be disinfected.

Salmonella enterica serotype Enteritidis isolates of phage types (PTs) PT1, PT4, PT13a and PT22 derived from sporadic cases and outbreaks of food poisoning in Japan during 1994 and 1995 were analysed by pulsed-field gel electrophoresis (PFGE). While PT1 strains from 5 different outbreaks showed 14 PFGE patterns, 5 PFGE patterns were observed among PT4 isolates from 5 different outbreaks and 6 independent isolates from imported chicken. Interestingly, 8 out of 9 PT4 strains associated with foreign travel to Southeast Asia were indistinguishable in PFGE pattern from 5 independent isolates of imported chicken from England. Although both PT13a and PT22 were first reported in Japan in 1994, PT22 showed various PFGE patterns compared to PT13a which had the same pattern within an outbreak, unlike PT1. These results could indicate that multiple clonal lines of PT1 and PT22 had already spread while relatively fewer clonal lines of PT4 and PT13a might exist in Japan.

We conducted a study to evaluate risk factors for developing typhoid fever in a setting where the disease is endemic in Karachi, Pakistan. We enrolled 100 cases with blood culture-confirmed Salmonella typhi between July and October 1994 and 200 age-matched neighbourhood controls. Cases had a median age of 5·8 years. In a conditional logistic regression model, eating ice cream (Odds ratio [OR]=2·3; 95% confidence interval [CI] 1·2–4·2, attributable risk [AR]=36%), eating food from a roadside cabin during the summer months (OR=4·6, 95% CI 1·6–13·0; AR=18%), taking antimicrobials in the 2 weeks preceding the onset of symptoms (OR=5·7, 95% CI 2·3–13·9, AR=21%), and drinking water at the work-site (OR=44·0, 95% CI 2·8–680, AR=8%) were all independently associated with typhoid fever. There was no difference in the microbiological water quality of home drinking water between cases and controls. Typhoid fever in Karachi resulted from high-dose exposures from multiple sources with individual susceptibility increased by young age and prior antimicrobial use. Improving commercial food hygiene and decreasing unnecessary antimicrobial use would be expected to decrease the burden of typhoid fever.

A 16-month old female child living on an Ontario dairy farm was taken to hospital suffering from bloody diarrhoea. Escherichia coli O157[ratio ]H7 was isolated from her stool. Initial tests of well water samples were negative for E. coli by standard methods but culture of selected coliform colonies on sorbitol-MacConkey agar led to isolation of E. coli O157[ratio ]H7. E. coli O157[ratio ]H7 was also isolated from 63% of cattle on the farm. The E. coli O157[ratio ]H7 isolates from the child, the water and the cattle were phage type 14, produced verotoxins 1 and 2, and were highly related on analysis by pulsed field gel electrophoresis. The child did not have known direct contact with the cattle and did not consume unpasteurized milk. Hydrogeological investigation revealed the design and location of the well would allow manure-contaminated surface water to flow into the well. This investigation demonstrates that cattle farm well water is a potential source of E. coli O157[ratio ]H7 which may not be identified by standard screening for E. coli in water.

In 1994, 171 (27%) of all positive blood cultures in our hospital were due to Acinetobacter species. Of these, 138 cultures were considered significant, 91 (66%) were community-acquired and 47 (34%) were nosocomial. Most acinetobacter bacteraemia in children [les ]1 year old was community-acquired, while nosocomial infection was more common in children >1 year old (P=0·01). Most children [les ]5 years old were severely malnourished. The incidence of bacteraemia was lowest during the post-monsoon to early winter months. Acinetobacter bacteraemia associated mortality was twice (16%) that of all other patients (7·7%, P < 0·0005) and accounted for 4·5% of all hospital deaths during the study period. Bacteraemia caused by Acinetobacter species is an important cause of morbidity and mortality among our patient population with diarrhoeal disease.

We describe here an examination of the validity of molecular types of Campylobacter jejuni as defined by separation of SmaI-digested DNA using pulsed-field gel electrophoresis (PFGE), recently suggested as part of a molecular subtyping scheme. Thirty-four Danish strains from humans, water, poultry and cattle were assigned to one of six SmaI ‘profile groups’ (PGs), with two additional strains included as genotypically distinct controls. The interstrain relationships were reexamined by PFGE of SalI, KpnI and BamHI-digested DNA, and also by serotyping with heat-stable antigens. All outbreak-related strains were indistinguishable by all criteria, as were two sets of two randomly-isolated human strains. Two groups of indistinguishable isolates contained randomly isolated strains from more than one source (poultry, humans and/or cattle), a finding with significant epidemiological connotations. All ‘genetically identical’ strains belonged to the same serotype, whereas genetic differences were detected between strains assigned to the same SmaI PG but differing in serotype. We conclude that PFGE-based genetic fingerprinting can yield invaluable data for epidemiological studies of sporadic C. jejuni infection, but that results based on one restriction site polymorphism must be checked with another enzyme.

To investigate if there is a reservoir of Escherichia coli O157 in Norwegian cattle, faecal samples from 197 cattle herds were screened for E. coli O157 by the use of immunomagnetic separation (IMS) and PCR during the 1995 grazing season. Six E. coli O157[ratio ]H-isolates were detected in two herds, one isolate in one and five in the other. The isolates carried the stx1, stx2, and eae genes, and a 90 MDa virulence plasmid. They were toxinogenic in a Vero cell assay. From 57 other herds, 137 faecal samples were positive for stx1 and/or stx2 genes detected by PCR run directly on IMS-isolated material. Among these samples, stx2 were the most widely distributed toxin encoding genes. No difference was found among milking cows and heifers in the rate of stx1 and/or stx2 in positive samples.

In September 1994, a complaint was registered at a public health unit concerning a cheese product. In addition, public health laboratories in Ontario reported an increase in the number of isolates of Salmonella berta from patients with diarrhoeal illness. A clinical, environmental and laboratory investigation was initiated to determine the nature of this outbreak. Isolates of Salmonella berta were compared using large fragment genomic fingerprinting by pulsed-field gel electrophoresis (PFGE). By late October, 82 clinical cases had been identified including 35 confirmed, 44 suspected and 3 secondary. The investigation linked illness to consumption of an unpasteurized soft cheese product produced on a farm and sold at farmers' markets. Subtyping results of patient, cheese and chicken isolates were indistinguishable, suggesting that the cheese was contaminated by chicken carcasses during production. The outbreak illustrates the potential role of uninspected home-based food producers and of cross-contamination in the transmission of foodborne bacterial pathogens.

We assessed the quality of evidence for the use of erythromycin in preventing secondary transmission of pertussis to close contacts of primary cases. A literature search was undertaken and identified papers were reviewed critically. Thirteen original papers and 1 manuscript met the inclusion criteria for review (3 randomized controlled trials, 4 analytical studies and 7 descriptive studies). Evidence from both experimental and analytical studies showed little effect of the use of erythromycin in preventing secondary transmission. Its effect is at best modest when compared with the protection conferred by use of good quality whole cell vaccine. Three studies reported adverse events with erythromycin prophylaxis; these were mainly nausea, vomiting and abdominal pain. In countries where effective pertussis vaccines are in use, erythromycin use should be confined to close contacts of cases, particularly unimmunized children or partially immunized infants who would be most susceptible to the complications of pertussis, or adults who come into close contact with vulnerable children.

Certain strains of verotoxigenic Escherichia coli (VTEC), and in particular those belonging to serogroup O157, cause attaching and effacing (AE) lesions of the host gut mucosa during pathogenesis. The mechanisms involved with bacterial attachment and the destruction of microvilli are determined by a cluster of genes within the LEE region, which also encode five secreted proteins. Sera from patients with antibodies to the lipopolysaccharide (LPS) of E. coli O157 and other VTEC were tested for antibodies to these secreted proteins. Twenty-one of 34 (62%) sera with antibodies to the lipopolysaccharide (LPS) of E. coli O157 also contained antibodies to one or more of the secreted proteins. Five of 12 sera containing antibodies to the LPS of a range of other VTEC serogroups also contained antibodies to 1 or more of the 5 secreted proteins, as did 16 of 70 (23%) sera from patients with haemolytic uraemic syndrome (HUS), haemorrhagic colitis (HC) or diarrhoea, but without bacteriological evidence of infection with VTEC and which did not contain antibodies to VTEC serogroups O5, O115, O145, O153 or O157. The detection of serum antibodies to secreted proteins may provide additional information for interpreting the results of established lipopolysaccharide-based VTEC serology.

The prevalence of Clostridium botulinum type E gene in fish and fishery products of commercial importance in Finland was determined using a quantitative PCR analysis. The contamination level in 438 raw fish samples from intestines, surface and whole fish and 208 fish roe samples varied from 10–40% and from 4–14% respectively, depending on the fish species studied. The presence of C. botulinum in European wild freshwater fish and roe was demonstrated for the first time by isolation of the organism from PCR-positive samples. Five percent of 214 vacuum-packed and 3% of 123 air-packed fishery product samples examined at retail level were positive for the botulinum neurotoxin type E gene. A contamination level of 10% in vacuum-packed hot-smoked whitefish was detected. The results demonstrate that C. botulinum type E poses a serious health risk for those consuming fishery products from the Baltic Sea area.

In 1990, to study regional prevalences and risk factors of Helicobacter pylori infection in healthy young adult males, sera were collected from a nationwide sample of 1659 males (mean age 20·7 years) at introduction into the Air Force School for military students in Caserta, Italy. An enzyme-linked immunosorbent assay was used to detect H. pylori specific immunoglobulin G antibodies. The observed overall seropositivity rate was 17·5% (95% CI 15·7–19·4). Prevalence was higher in southern Italy and in the Italian islands as compared with northern Italy and central Italy (21·3% vs. 9·5%). Multiple logistic regression analysis showed that residence in southern areas and islands was the strongest predictor of the likelihood of H. pylori seropositivity; number of siblings in the household was marginally associated; years of father's schooling was not a significant predictor. H. pylori positive subjects were more likely positive for antibodies to hepatitis A virus infection (anti-HAV) than those H. pylori negative (35·4% vs. 24·9%; Odds Ratio 1·7, 95% CI 1·3–2·2). Adjustment for the confounding effect of sociodemographic variables weakened this association (OR 1·3, 95% CI 1·0–1·7). These findings suggest that differences in environmental conditions rather than in socioeconomic status may have played the major role in the different spread of H. pylori infection across the country.

Public health surveillance requires the monitoring of waterborne disease, but sensitive and specific detection of relevant incidents is difficult.

The Communicable Disease Surveillance Centre receives information from various sources about clusters of cases of illness in England and Wales. The reporter may suspect that water consumption or recreational water exposure is the route of infection, or subsequent investigation may raise the hypothesis that water is associated with illness.

It is difficult to prove beyond reasonable doubt that such a hypothesis is correct. Water samples from the time of exposure are seldom available, some organisms are difficult to detect and almost everyone has some exposure to water. Therefore, we have developed a method of categorizing the degree of evidence used to implicate water. The categories take into account the epidemiology, microbiology and water quality information. Thus outbreaks are classified as being associated with water either ‘strongly’, ‘probably’ or ‘possibly’.

This system allows a broad database for monitoring possible effects of water and is not confined to the few outbreaks which have been intensively investigated or have positive environmental microbiology. Thus, for reported incidents, the sensitivity of classifying it as water associated should be high but this may be at the expense of specificity, especially with the ‘possible’ association.

Following the 1993 Milwaukee cryptosporidiosis outbreak, we examined data from eight sources available during the time of the outbreak. Although there was a remarkable temporal correspondence of surveillance peaks, the most timely data involved use of systems in which personnel with existing close ties to public health programmes perceived the importance of providing information despite workload constraints associated with an outbreak. During the investigation, surveillance systems which could be easily linked with laboratory data, were flexible in adding new variables, and which demonstrated low baseline variability were most useful. Geographically fixed nursing home residents served as an ideal population with non-confounded exposures. Use of surrogate measurements of morbidity can trigger worthwhile public health responses in advance of laboratory-confirmed diagnosis and help reduce total morbidity associated with an outbreak. This report describes the relative strengths and weaknesses of these surveillance methods for community-wide waterborne illness detection and their application in outbreak decision making.

Between March and May 1996 Stenotrophomonas maltophilia was cultured from endotracheal aspirate samples from five preterm infants in a neonatal intensive care unit (NICU). Four infants were superficially colonized, but a fifth died due to S. maltophilia septicaemia. S. maltophilia was cultured from tap water from three outlets in the NICU including one with a previously unnoticed defective sink drain. Water from these outlets was used to wash the preterm infants. Environmental and clinical S. maltophilia isolates yielded identical banding patterns on random arbitrary polymorphic DNA (RAPD) PCR analysis. The outbreak was controlled by reinforcement of hand disinfection, limitation of the use of tap water for hand washing and by using sterile water to wash the preterm infants. We conclude that tap water should not be used for washing preterm infants in the NICU, unless steps are taken to prevent microbial growth in the outlets.