We reviewed serological and epidemiological data relating to 1000 consecutive patients from whom specimens were submitted for estimation of bartonella antibodies, using MRL Diagnostics Bartonella IFA IgM and IgG kits. Using 289 control sera, we estimated the specificity of the kits as [ges ]99·0%. Evidence of bartonella infection was found in 16·3% of patients examined. Rates varied by patient group: 20% of patients for whom a diagnosis of cat scratch disease (CSD) was considered probable had evidence of infection, as did 10·4% of patients with ‘possible CSD’, 8·1% of patients with possible bacillary angiomatosis, 18·2% of patients with ‘culture negative’ endocarditis and 17·6% of patients with possible bartonellosis with ophthalmic involvement. An IgM response was seen in 6·6% of patients and IgG in 15·1%. Cases were more frequent among males than females (18·5% vs. 13·9%). Analysis by age showed that although rates of infection were highest in the decades 0–9 years (19·4%) and 10–19 years (20·7%), they fell only slightly in the next three decades. MRL bartonella kits appears to provide a useful and specific approach to the diagnosis of these infections.

The prevalence and characteristics of Escherichia coli and Salmonella spp. as well as counts of E. coli in raw oysters, condiments/spices, and raw oyster cocktails sampled from 72 vendors across Western Trinidad were determined. The microbial quality of the water used in the preparation of raw oysters was also investigated. Of 200 samples each of raw oysters, condiments/spices and oyster cocktails tested, 154 (77·0%), 89 (44·5%) and 154 (77·0%) respectively yielded E. coli. The differences were statistically significant (P=[les ]0·001; χ2=62·91). The mean E. coli count per g in the ready-to-eat oyster cocktail ranged from 1·5×103±2·7×103 in Couva to 8·7×106±4·9×107 in San Fernando. One hundred and forty-six (73·0%) oyster cocktails contaminated with E. coli had counts that exceeded the recommended standard of 16 per g. Of a total of 590 E. coli isolates from various sources tested, 24 (4·1%), 20 (3·4%) and 69 (11·7%) were mucoid, haemolytic and non-sorbitol fermenters respectively. Twelve (2·0%) isolates of E. coli were O157 strains, while 92 (46·0%) of 200 E. coli isolates tested belonged to enteropathogenic serogroups. Ninety (45·0%) and 73 (36·5%) of 200 water samples contained total coliforms and faecal coliforms respectively, with counts that exceeded 2·2 coliforms per 100 ml. Salmonella spp. were isolated from 7 (3·5%), 1 (0·5%) and 2 (1·0%) of 200 samples each, of raw oysters, condiments/spices and oyster cocktails respectively. Oysters pose a health risk to consumers in Trinidad, particularly from colibacillosis and salmonellosis, and the need for increased public awareness of this hazard cannot be over-emphasized.

There have been no previous longitudinal studies of otitis media conducted in non-Aboriginal Australian children. This paper describes the rate and risk factors for middle ear effusion (MEE) in children attending day care in Darwin, Australia. A prospective cohort study of 252 children under 4 years was conducted in 9 day care centres over 12 fortnights between 24 March and 15 September 1997. Tympanometry was conducted fortnightly and multivariate analysis used to determine risk factors predicting MEE. The outcome of interest was the rate of type B tympanograms per child detected in either ear at fortnightly examinations. After adjusting for clustering by child, MEE was detected on average 4·4 times in 12 fortnights (37% of all examinations conducted). Risk factors associated with presence of effusion were younger age, a family history of ear infection, previous grommets (tympanostomy tubes), ethnicity and the day care centre attended. A history of wheeze appeared protective. These effects were modest (RR 0·57–1·70). Middle ear effusion is very common in children attending day care in Darwin. This has clinical importance, since MEE during early childhood may affect optimal hearing, learning and speech development. There is little scope for modification for many of the risk factors for MEE predicted by this model. Further study of the day care environment is warranted.

Mouse virulence and the ability to adhere to, and invade cultured MDCK cells were investigated in 38 phage type reference strains of Salmonella enterica serotype Enteritidis and correlated with genomic lineage. The genomic lineage of 11 of the strains was determined in the present study; one IS200 and one ribotype pattern that had not been reported previously were observed. Log c.f.u. in the spleen 10 days post intraperitoneal (i.p.) infection with 3×103 bacteria (logVC10) varied between 2·9 and 8·7. The reference strains of PT7 and PT23 were found to be semi-rough and were of low virulence. All other strains possessed smooth LPS. Within each of the two major clonal lines, as well as among phage types outside these, both highly virulent and moderate to low virulent strains were present. While all strains of PT1, PT2 and PT8 were highly virulent, low virulent strains were detected in PT4 and PT13. The ability to adhere to, and invade MDCK cells varied between phage types (adherence between 13 and 61% of the inocula and invasion between 4 and 151% of the adherent cells). The results of the cell culture experiments did not correlate with the results of mouse virulence tests. No correlation between clonal lineage and virulence was found within S. Enteritidis. It seems most likely that some strains have lost some of the essential factors enabling this serotype to cause successful systemic infection.

The Traveller's Diarrhoea Network, by which the Infectious Disease Surveillance Center is electronically connected with two major airport quarantine stations and three infectious disease hospitals, was launched in February 1988 in Japan. The data on travellers' diarrhoea detected is reported weekly by e-mail. Two clusters of infection among travellers returning from Italy were reported by two airport quarantine stations at the end of September 1998. A total of 12 salmonella isolates from 2 clusters were examined. All were identified as Salmonella enteritidis, phage type 4 and showed identical banding patterns on pulsed-field gel electrophoresis. A case-control study showed that the scrambled eggs served at the hotel restaurant in Rome were the likely source of this outbreak. This outbreak could not have been detected promptly and investigated easily without the e-mail network. International exchange of data on travellers' diarrhoea is important for preventing and controlling food-borne illnesses infected abroad.

One question of particular importance in phase III HIV vaccine trials is the choice of efficacy measure (EM) to validly and precisely estimate the true vaccinal efficacy. Traditional EMs, based on hazard rate ratio (HRR) or cumulative incidence ratio (CIR) are time-sensitive to mode of vaccine action and population heterogeneities. Through Monte-Carlo simulation, the performance of HRR and CIR based EMs are examined across different trial designs and vaccine and population characteristics. A new EM based on log-spline hazard regression (HARE) is proposed. Given that vaccinal properties (mode of action, time-lag, waning) are unknown a priori, appropriate selection of EM is problematic, and HRR and CIR can be unreliable to estimate the true maximum efficacy of candidate products. Non-random sexual mixing can exacerbate biases in HRR and CIR. HARE can offer valid estimation across different modes of vaccine action and in presence of frailty effects, contrary to its traditional counterparts. Our simulation studies highlight the weaknesses of widely used EMs while offering guidelines for trial design and suggesting new avenues for statistical analysis.

Four cases of mumps were reported among 180 army recruits who had received MMR vaccine 16 days earlier. Mumps serology, salivary mumps IgM and PCR tests for the SH gene were performed on the 4 cases and on 5 control recruits who remained well. PCR products were sequenced and the sequences compared to those of wild type and vaccine strains of mumps. Further salivary mumps IgM tests were performed on the remaining 171 recruits. Mumps infection was confirmed in the 4 cases but not in the 5 controls. The controls had serological evidence of prior immunity. The SH gene sequence found in the 4 cases was wild type. Saliva tests identified 2 additional recruits with mumps IgM, one of whom had presented with suspected mumps 2 days before the MMR vaccine was given. Thus 6 (5 symptomatic and 1 asymptomatic) cases of mumps in army recruits recently receiving MMR vaccine were not due to the vaccine but to coincidental infection with wild-type mumps virus. The probable index case was revealed by salivary mumps IgM tests. This study highlights the importance of appropriate investigation of illness associated with MMR vaccination.

A cluster of acute melioidosis cases occurred in a remote, coastal community in tropical Western Australia. Molecular typing of Burkholderia pseudomallei isolates from culture-confirmed cases and suspected environmental sources by pulsed-field gel electrophoresis (PFGE) of XbaI chromosomal DNA digests showed that a single PFGE type was responsible for five cases of acute infection in a community of around 300 during a 5 week period. This temporal and geographical clustering of acute melioidosis cases provided a unique opportunity to investigate the environmental factors contributing to this disease. B. pseudomallei isolated from a domestic tap at the home of an asymptomatic seroconverter was indistinguishable by PFGE. Possible contributing environmental factors included an unusually acid communal water supply, unrecordable chlorine levels during the probable exposure period, a nearby earth tremor, and gusting winds during the installation of new water and electricity supplies. The possible role of the potable water supply as a source of B. pseudomallei was investigated further.

In a 5-year retrospective survey of respiratory syncytial virus (RSV) infections among hospitalized children, 1340 cases were identified of which, 98·4% were children <5 years old with a male:female ratio of 1·5:1. Most cases occurred from April to September showing a significant positive correlation with temperature and relative humidity. Community-acquired infections accounted for 92·5% of the cases with a mean hospital stay of 5 days. The estimated annual incidence of RSV infection requiring hospitalization was 2·5/1000 children <5 years old with a mortality of 0·15% among hospitalized cases. On average, 248 children were admitted each year to the 1400-bed acute regional hospital accounting for an expenditure of HK $1.94 (∼US $0.25) million for hospitalization costs which equates to an annual cost in excess of HK $6.67 (∼US $0.86) million for the whole of Hong Kong. An RSV vaccine should be a priority.

In 1997, prevalence of and risk factors for hepatitis A virus (HAV) infection were evaluated in 146 homosexual and 286 heterosexual men attending a Sexually Transmitted Disease (STD) Clinic in Rome, Italy. Total HAV antibody (anti-HAV) was detected in 60·3% of homosexuals and 62·2% of heterosexuals. After adjustment for the confounding effects of age, years of schooling, number of sexual partners, use of condoms, and history of STD, homosexuals were not found to be at increased risk of previous HAV exposure than heterosexuals (OR 1·1; 95% CI 0·7–1·9). Independent predictors of the likelihood of anti-HAV seropositivity among homosexuals and heterosexuals were: age older than 35 years and positive syphilis serology which is likely a proxy of lifestyles that increase the risk of faecal–oral infections.

These findings do not support a higher risk in homosexual men but could suggest a role for the vaccination of susceptible patients attending STD clinics.

A cross-sectional study was conducted in prisons of Cantabria (northern Spain) from June 1992 to December 1994. Inmates were asked to participate in a survey on prevalence and risk factors for monoinfections and coinfections with HIV, HBV and HCV. Crude and multiple odds ratios of risk factors were calculated (by polychotomous logistic regression). Prevalence of coinfections was higher than that of monoinfections. IDU risk factors were the main independent variables associated with monoinfections and coinfections with these agents. The strength of association increased with the degree of coinfection for IDU risk factors and penal status, e.g. duration of injecting drug use for more than 5 years yielded an adjusted OR ranging from 1·3 (95% CI: 0·4–5·1) for HBV monoinfection to 180 (95% CI: 61·0–540·0) for HIV–HBV–HCV coinfection. In comparison, sexual behaviours were less important than IDU risk factors.

In a prison in Victoria, Australia, our objectives were contact tracing of inmates and staff at risk of exposure to an identified index case; and to determine risk factors for prevalent and incident infection. Inmates and staff who were potentially exposed to the index case were screened with a Mantoux skin test and a questionnaire. Inmate movements within the prison were compared to movements of the index case. Logistic regression was used to determine risk factors for infection. The index case had smear positive, cavitating pulmonary tuberculosis (TB), which was undiagnosed for 3 months. This was the period of potential exposure. The prevalence of positive skin test reactions in 190 inmates and staff at the prison was 10%. Significant predictors of a positive skin test were being an inmate (odds ratio (OR) 15·5), older age (OR 8·3) and being born overseas (OR 10·7). Bacille Calmette Guerin (BCG) vaccination, proximity to the index case in various prison sites, duration of incarceration, number of incarcerations and number of inmates per cell were not significant. There were three recent skin test conversions from negative to positive, representing a conversion rate of 3·5%. We did not find evidence of significant transmission of TB from a single index case. The prevalence of infection in this Australian prison was lower than published rates in other countries. Better prison conditions and different demographics of prison inmates in Australia may explain these differences.

A two-dose mass vaccination programme with a combined vaccine against measles, mumps and rubella (MMR) was adopted in the Netherlands in 1987, replacing the selective schoolgirl vaccination strategy introduced in 1974. To obtain insight into the effect of mass vaccination and the population's immunity, the antibody levels against rubella were studied in the general Dutch population and in religious groups refusing vaccination. In the national sample, we observed a high prevalence (96·5%) for rubella antibodies in vaccinated cohorts as well as in the older unvaccinated cohorts. No indications of rapidly waning immunity after vaccination were found. There are indications of low virus circulation in the last few years. The very high seroprevalence in women at childbearing age is consistent with the few reported cases of congenital rubella syndrome (CRS) at present. However, individuals in the age group of 1–9 years who are not vaccinated for religious or other reasons have a considerably lower seroprevalence and thus there is a potential risk of a CRS outbreak in the future.

In a prospective study of community-dwelling people 60–90 years of age, we examined the coverage of influenza vaccine during 1992–3 and 1993–4, the efficacy of vaccination in reducing serologically-confirmed clinical episodes of influenza A during 1993, and the effect of cigarette smoking. During 1992 and 1993, influenza vaccine was given to 106/215 (49%) and 120/204 (59%) people with risk conditions, and 84/225 (37%) and 103/235 (44%) without risk conditions. Influenza vaccination and general practitioner consultations during 1992 were independent predictors of vaccination in 1993, but current smoking was a negative predictor. Of 209 unimmunized people, 8/35 (23%) smokers had clinical influenza as compared with 11/174 (6 %) non-smokers (OR 4·4, 95% CI 1·6 to 11·9). Of 371 non-smokers, 1/197 (0·5%) vaccinees had influenza as compared with 11/174 (6 %) non-vaccinees (OR 0·075, 95% CI 0·587 to 0·009). No cases of influenza occurred among 21 current smokers who were vaccinated.

Four years after the occurrence of an outbreak of hepatitis B and HIV infection among injecting drug user inmates at Her Majesty's Prison Glenochil in Scotland, a study design was developed to complete the epidemiological account of the HIV outbreak. Our aim was to identify potential cases of (1) HIV transmission not diagnosed during the original outbreak investigation and (2) the source(s) of the outbreak. Scotland's HIV positive case register was searched for matches to a soundexed list of 636 Glenochil inmates imprisoned during January–June 1993. Eight HIV infections that may have been acquired in Glenochil and four possible sources of the outbreak were identified. The second stage of follow-up – molecular epidemiological techniques used on stored sera samples from identified individuals – is described in the companion paper. Without breach of medical or prisoner confidentiality, indirect and anonymous follow-up has proved possible for the Glenochil inmates.

This study compares the timing of the rise and decline of measles-specific IgM in serum samples and in oral fluid samples. Two hundred and eighty 9-month-old infants presenting for routine measles vaccination in Addis Ababa, Ethiopia, were enrolled. Paired serum and oral fluid samples were collected before and 1, 2, 3 or 4 weeks after measles vaccination. Samples were tested by using a modified antibody-capture enzyme immunoassay. For the 321 IgM-negative pre- and post-vaccination serum samples, 317 (99%) of their corresponding oral fluid samples were IgM-negative. Among the 130 IgM-positive serum samples, 75% of their paired oral fluid samples were IgM-positive, with the percentage rising to 87% after oral fluid samples collected [ges ]3·5 weeks after vaccination were excluded. Among the post-vaccination serum samples, the percent IgM-positive peaked in week 3 and declined to 79% in week 4. For post-vaccination oral fluid samples, the percent IgM-positive peaked in weeks 2 and 3, and then declined to 43% in week 4. This modified antibody-capture enzyme immunoassay appears to detect vaccine-induced measles-specific IgM in the first 3 weeks after vaccination.

Protozoa are important enteric pathogens in patients with human immunodeficiency virus (HIV) infection. In this study the prevalence of intestinal protozoa in 154 HIV-infected patients, with or without diarrhoea, in our region (Apulia, South Italy) was evaluated between December 1993 and February 1998. In the majority of patients CD4+ T cell count was below 200/μl. The overall prevalence of intestinal protozoa was 43/154 (27·92%). Twenty-eight (43·08%) out of 65 patients with diarrhoea and 15 (16·85%) out of 89 non-diarrhoeic patients were parasitized. In particular, in the group of 65 patients with diarrhoea the following protozoa were identified: Cryptosporidium parvum in 14 (21·54%), Blastocystis hominis in 7 (10·77%), microsporidia in 6 (9·23%), Giardia lamblia in 4 (6·15%) and Isospora belli in 1 (1·54%). Three patients were Cryptosporidium parvum-microsporidia co-infected. In patients without intestinal symptoms, prevalence was 3/89 (3·37%) for Cryptosporidium parvum, 9/89 (10·11%) for Blastocystis hominis, 1/89 (1·12%) for microsporidia and 2/89 (2·25%) for Giardia lamblia. A significant (P<0·001) correlation was observed between protozoan infection and the presence of diarrhoea. In particular, Cryptosporidium parvum and microsporidia infections were significantly (P<0·001) and P=0·046, respectively) associated with diarrhoeal illness. Moreover, the majority of cases of cryptosporidiosis were first diagnosed in the periods of heaviest rainfall. Therefore, drinking water contamination may be a possible source of human infection in our area.

In a molecular investigation into the outbreak of HIV in Glenochil during the first 6 months of 1993, we previously demonstrated that 13 out of the 14 HIV positive inmates were infected with a virtually identical strain, and discounted 2 others as potential sources. Here we investigate a further 8 potential contacts and 4 potential sources which were identified in the companion paper [1]. We were able to examine viral sequence from all but one of these 12 and results have revealed them to be distinct both from each other and the original 14. Thus, despite an intensive follow-up investigation, we have been unable to identify any further HIV infections that might have been part of the 1993 outbreak. It is possible that persons who were infected at that time remain undetected; however this and the companion report strongly suggest that if this were the case the likely numbers would be few.

To investigate the feasibility of using information about the health and management of lambs on farms to predict the risk of gross abnormalities at post-mortem meat inspection, 6732 lambs from 30 different farms in Great Britain were followed through to slaughter in 1995/6. The farm-level data were collected during farm visits at the beginning of the study. Routine meat inspection findings for the lambs were obtained from the 10 participating abattoirs. The most common abnormalities found during post-mortem inspection were pneumonia/pleurisy (53% of cohorts), lungworm (40%), abscesses (30%), liver fluke (27%) and nephritis/nephrosis (27%). The farm-level risk factors associated with abnormalities at slaughter varied with the type of lesion. The most significant risk factor was the age of the lambs at slaughter. Lambs slaughtered at an older age were more likely to have an abnormality, especially pneumonia, abscesses and liver fluke. After age, environmental factors appeared to be better predictors of those cohorts that would have lesions at slaughter than health and disease control variables. However, a much larger study would be required to identify a set of farm-level factors that adequately discriminated between lambs with high and low risks of lesion at slaughter. At the end of the study, the farmers were informed of the meat inspection findings for their lambs and a third indicated that they would improve their animal husbandry as a result of the information.

Ocular disease is the commonest disabling consequence of toxoplasma infection. Incidence and lifetime risk of ocular symptoms were determined by ascertaining affected patients in a population-based, active reporting study involving ophthalmologists serving a population of 7·4 million. Eighty-seven symptomatic episodes were attributed to toxoplasma infection. Bilateral visual acuity of 6/12 or less was found in seven episodes (8%) and was likely to have been transient in most cases. Black people born in West Africa had a 100-fold higher incidence of symptoms than white people born in Britain. Only two patients reported symptoms before 10 years of age. The estimated lifetime risk of symptoms in British born individuals (52% of all episodes) was 18/100000 (95% confidence interval: 10·8–25·2). The low risk and mild symptoms in an unscreened British population indicate limited potential benefits of prenatal or postnatal screening. The late age at presentation suggests a mixed aetiology of postnatally acquired and congenital infection for which primary prevention may be appropriate, particularly among West Africans.