Natural infection with the influenza virus is believed to generate cross-protective immunity across both types and subtypes. However, less is known about the persistence of this immunity and thus the susceptibility of individuals to repeat infection. We used 13 years (2005–2017) of surveillance data from Queensland, Australia, to describe the incidence and distribution of repeat influenza infections. Consecutive infections that occurred within 14 days of prior infection were considered a mixed infection; those that occurred more than 14 days later were considered separate (repeat) infections. Kaplan-Meier plots were used to investigate the probability of reinfection over time and the Prentice, Williams and Peterson extension of the Cox proportional hazards model was used to assess the association of age and gender with reinfection. Among the 188 392 notifications received during 2005–2017, 6165 were consecutively notified for the same individual (3.3% of notifications), and 2958 were mixed infections (1.6%). Overall, the probability of reinfection was low: the cumulative incidence was <1% after one year, 4.6% after five years, and 9.6% after ten years. The majority of consecutive infections were the result of two type A infections (43%) and were most common among females (adjusted hazard ratio (aHR): 1.15, 95% confidence interval (CI) 1.09–1.21), children aged less than 5 years (relative to adults aged 18–64 years aHR: 1.58, 95% CI 1.47–1.70) and older adults aged at least 65 years (aHR: 1.35; 95% CI 1.24–1.47). Our study suggests consecutive infections are possible but rare. These findings have implications for our understanding of population immunity to influenza.

Host density is a key regulatory factor in parasite transmission. The goldfish (Carassius auratus)-Gyrodactylus kobayashii model was used to investigate effects of host density on population growth of gyrodactylids. A donor fish infected by five gravid gyrodactylids was mixed with 11 parasite-free goldfish at five host densities. There was a significant positive correlation between host density and mean abundance of G. kobayashii throughout the 58-day experiment. During early infection (days 15–24), mean abundance in medium high (0.5 fish L−1) and high host density groups (1 and 2 fish L−1) was significantly higher than that in the low host density groups (0.125 and 0.25 fish L−1). At high host density, prevalence increased more rapidly, and the peak prevalence was higher. Fitting of an exponential growth model showed that the population growth rate of the parasite increased with host density. A hypothesis was proposed that higher host density contributed to increased reinfection of detached gyrodactylids. A reinfection experiment was designed to test this hypothesis. Both mean abundance and prevalence at a host density of 1 fish L−1 were significantly higher than those at 0.25 fish L−1 on days 1 and 3, which suggested that more reinfections of G. kobyashii occurred at the higher host density. Density-dependent transmission during the early infection was an important contributor of population growth of G. kobayashii, as well as density-dependent reinfection of the detached gyrodactylids.

Clostridium difficile, the most common cause of hospital-associated diarrhoea in developed countries, presents major public health challenges. The high clinical and economic burden from C. difficile infection (CDI) relates to the high frequency of recurrent infections caused by either the same or different strains of C. difficile. An interval of 8 weeks after index infection is commonly used to classify recurrent CDI episodes. We assessed strains of C. difficile in a sample of patients with recurrent CDI in Western Australia from October 2011 to July 2017. The performance of different intervals between initial and subsequent episodes of CDI was investigated. Of 4612 patients with CDI, 1471 (32%) were identified with recurrence. PCR ribotyping data were available for initial and recurrent episodes for 551 patients. Relapse (recurrence with same ribotype (RT) as index episode) was found in 350 (64%) patients and reinfection (recurrence with new RT) in 201 (36%) patients. Our analysis indicates that 8- and 20-week intervals failed to adequately distinguish reinfection from relapse. In addition, living in a non-metropolitan area modified the effect of age on the risk of relapse. Where molecular epidemiological data are not available, we suggest that applying an 8-week interval to define recurrent CDI requires more consideration.

Although high hepatitis C virus (HCV) prevalence has been observed in people who inject drugs (PWID) for decades, research suggests incidence is falling. We examined whether PWIDs’ use of opioid substitution therapy (OST) and their needle-and-syringe sharing behaviour explained HCV incidence. We assessed HCV incidence in 235 PWID in Melbourne, Australia, and performed discrete-time survival with needle-sharing and OST status as independent variables. HCV infection, reinfection and combined infection/reinfection incidences were 7·6 [95% confidence interval (CI) 4·8–11·9], 12·4 (95% CI 9·1–17·0) and 9·7 (95% CI 7·4–12·6) per 100 person-years, respectively. Needle-sharing was significantly associated with higher incidence of naive HCV infection [hazard ratio (HR) 4·9, 95% CI 1·3–17·7] but not reinfection (HR 1·85, 95% CI 0·79–4·32); however, a cross-model test suggested this difference was sample specific. Past month use of OST had non-significant protective effects against naive HCV infection and reinfection. Our data confirm previous evidence of greatly reduced HCV incidence in PWID, but not the significant protective effect of OST on HCV incidence detected in recent studies. Our findings reinforce the need for greater access to HCV testing and prevention services to accelerate the decline in incidence, and HCV treatment, management and support to limit reinfection.

Children encounter repeated respiratory tract infections during their early life. We conducted a prospective clinical and serological follow-up study to estimate the respiratory syncytial virus (RSV) primary infection and reinfection rates in early childhood. Sera were collected from 291 healthy children at the ages of 13, 24 and 36 months and antibody levels against RSV antigens were determined by enzyme immunoassay. The RT–PCR method was also used for identifying the possible presence of RSV in symptomatic patients. At ages 1, 2 and 3 years, 37%, 68% and 86%, respectively, of studied children were seropositive for RSV. In children seropositive at age 1 year, RSV reinfection rate was at least 37%. Only one of reinfected children showed evidence for a third reinfection by age 3 years. Of children who turned RSV seropositive between ages 1 and 2 years, the reinfection rate was 32% during the third year of life. The mean antibody levels at primary infection were very similar in all age groups. The average decrease of antibody levels was 25–30% within a year. In 66 cases RSV infection was identified by RT–PCR. RSV infection rate in early childhood is 86% and reinfection rate is around 35%. This prospective serological follow-up study also provided evidence for the presence of RSV infections in children that did not show clinical signs warranting RSV RNA detection.

We performed a study to determine rates of reinfection in three groups followed for 2 years after successful treatment: American Indian/Alaska Native (AI/AN) persons living in urban (group 1) and rural (group 2) communities, and urban Alaska non-Native persons (group 3). We enrolled adults diagnosed with H. pylori infection based on a positive urea breath test (13C-UBT). After successful treatment was documented at 2 months, we tested each patient by 13C-UBT at 4, 6, 12 and 24 months. At each visit, participants were asked about medication use, illnesses and risk factors for reinfection. We followed 229 persons for 2 years or until they became reinfected. H. pylori reinfection occurred in 36 persons; cumulative reinfection rates were 14·5%, 22·1%, and 12·0% for groups 1, 2, and 3, respectively. Study participants who became reinfected were more likely to have peptic ulcer disease (P = 0·02), low education level (P = 0·04), or have a higher proportion of household members infected with H. pylori compared to participants who did not become reinfected (P = 0·03). Among all three groups, reinfection occurred at rates higher than those reported for other US populations (<5% at 2 years); rural AI/AN individuals appear to be at highest risk for reinfection.

Reinfection pattern among 6- to 20-year-old subjects was studied over 24 months in two Mauritanian villages of intenseSchistosoma haematobium infection after a targeted chemotherapy with praziquantel involving the whole community. Subjects received treatment according to the presence of haematuria/proteinuria and this indirect screening technique was able to identify 98–100% of the heavily infected subjects (50 + eggs/10 ml). The two villages differed with respect to their characteristics, quality of follow-up and reinfection pattern. The post-treatment 6-month cumulative incidence during the two transmission periods following the chemotherapy, estimated from a subset of 116 subjects, was 18·0% and 20·5%. Reinfection rates were higher among males (Cox-Mantel: P = 0·0015), among children 6–10 years of age than older (P = 0·0078) and among subjects with more than 50 eggs/10 ml of urine before treatment than subjects with a lower egg output (P = 0·009). A Cox proportional hazard regression model was fitted and confirmed that gender, age and pretreatment level of infection were predictors of the rate of reinfection but that there was no interaction between these predictors.

Reinfection with hookworm (Necator americanus) following chemotherapy was studied over 2 years in a rural village in Madang Province, Papua New Guinea. The prevalence of hookworm infection had returned to pre-treatment levels after 2 years, and the geometric mean hookworm burden had returned to 58 % of the pre-treatment value. The rate of acquisition of adult worms was independent of host age, and was estimated as a geometric mean of 2·9–3·3 worms/host/year (arithmetic mean 7·9–8·9 worms/host/year). There was significant predisposition to hookworm infection; the strength of this predisposition did not vary significantly between age or sex classes.

This paper reports a longitudinal study, conducted in 1989/90, of 1893 children aged 5 to 14 years in 9 poor urban areas of the city of Salvador (population 2·44 million), capital of Bahia State in northeast Brazil. Stool examinations were performed to measure nematode infection and reinfection 9 months after treatment, and an extensive questionnaire was applied to collect information on each child and on the conditions of the household. Comparison of areas with different levels of infrastructure showed the following trends as the level of community sanitation improved: clustering of cases by household became more significant, predisposition of individuals to reinfection and to heavy infection became more marked, and infections with different species were increasingly aggregated in the same individuals. These results suggest that sewerage and drainage can significantly reduce transmission of intestinal nematode infections in the public domain, but that other measures are required to control transmission within the household.

Chagas' disease, caused by Trypanosoma cruzi, affects approximately 20 million people. There are 3 stages in the disease: acute, intermediate and chronic, the diversity and severity of the symptoms range from a mild electrocardiographic alteration to sudden death. We have previously demonstrated that when reinfections were carried out in the acute phase they produce greater cardiac damage. The aim of the present work was to investigate whether T. cruzi reinfected mice present electrocardiographic abnormalities that could be characteristic and only achieved after reinfections. Of the mice reinfected during the acute phase 100% showed abnormalities from days 90 post-infection, with a predominance of auricle ventricle blocks (67–71%). All the mice reinfected during the chronic infection showed electrocardiographic alteration after 30 days post-first reinfection. Of the mice infected, without reinfection, 60% exhibited electrocardiographic dysfunction at 90 days post-infection. Our results demonstrated that when the host was reinfected in the acute phase, more serious electrocardiographic alterations were developed than when the reinfections were carried out in the chronic stage. Sudden death described in some chagasic patients, might be related to some of the findings described here.