Background: Researchers increasingly use social contact data to inform models for infectious disease spread with the aim of guiding effective policies about disease prevention and control. In this article, we undertake a systematic review of the study design, statistical analyses, and outcomes of the many social contact surveys that have been published. Methods: We systematically searched PubMed and Web of Science for articles regarding social contact surveys. We followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guidelines as closely as possible. Results: In total, we identified 64 social contact surveys, with more than 80% of the surveys conducted in high-income countries. Study settings included general population (58%), schools or universities (37%), and health care/conference/research institutes (5%). The largest number of studies did not focus on a specific age group (38%), whereas others focused on adults (32%) or children (19%). Retrospective (45%) and prospective (41%) designs were used most often with 6% using both for comparison purposes. The definition of a contact varied among surveys, e.g., a nonphysical contact may require conversation, close proximity, or both. We identified age, time schedule (e.g., weekday/weekend), and household size as relevant determinants of contact patterns across a large number of studies. Conclusions: We found that the overall features of the contact patterns were remarkably robust across several countries, and irrespective of the study details. By considering the most common approach in each aspect of design (e.g., sampling schemes, data collection, definition of contact), we could identify recommendations for future contact data surveys that may be used to facilitate comparison between studies.

BackgroundNatural immunity to Streptococcus pneumoniae is thought to be induced by exposure to S. pneumoniae or cross-reactive antigens. No longitudi

This paper analyses Streptococcus pneumoniae transmission dynamics in households using longitudinal data on pneumococcal (Pnc) carriage in the United Kingdom. Ten consecutive swabs were taken at 4-week intervals from all members of 121 households. The family status is derived from the observed Pnc carriage status of each family member. Transition matrices are built for each family size and composition containing the observed frequency of transitions between family statuses over a 28-day interval. A density-dependent transmission model is fitted to derive maximum-likelihood estimates of the duration of carriage and acquisition rates from the community and from infected individuals within the household. Parameter values are estimated for children (<5 years) and adults (5+ years). The duration of carriage is longer in children <5 years of age than in older family members (51 vs. 19 days). Children are 3–4 times more likely than adults to acquire Pnc infection from the community. Transmission rates within the household suggest that adults are more infectious but less susceptible than children. Transmission within the household is most important in large families. The proportion of household-acquired infection ranges from 29 to 46% in households of three persons to 38–50% in larger households. Evidence of density-dependent within-household transmission is found, although the strength of this relationship is not clear from the model estimates.

A 10-month longitudinal household study of pre-school children and their families was undertaken with monthly visits collecting epidemiological data and nasopharyngeal swabs in Hertfordshire, England from 2001 to 2002. Pneumococcal culture was with standard methods. In total, 121 families (489 individuals) took part. Mean prevalence of carriage ranged from 52% for age groups 0–2 years, 45% for 3–4 years, 21% for 5–17 years and 8% for [ges ]18 years. Carriage occurred more than once in 86% of children aged 0–2 years compared to 36% of those aged [ges ]18 years. The most prevalent serotypes in the 0–2 years age group were 6B followed by 19F, 23F, 6A and 14. Young children were responsible for the majority of introductions of new serotypes into a household. Erythromycin resistance (alone or in combination) occurred in 10% of samples and penicillin non-susceptibility in 3·7%. Overall the recently licensed 7-valent conjugate vaccine (PCV) would protect against 64% of serotypes with no intra-serogroup cross protection and 82% with such protection. Nasopharyngeal carriage of S. pneumoniae is common in a UK setting in the pre-conjugate vaccine era. PCV would protect against a large proportion of carriage isolates. However, the impact of vaccination on non-vaccine serotypes will need to be monitored.

Knowledge of the determinants of infectious disease transmission is a public health priority as it allows the design of optimal control strategies for endemic or emerging infections. We analyse a detailed dataset on contact patterns across five European countries and use available serological profiles for varicella and parvovirus B19 infections to identify the types of contact that may be most relevant for transmission. We show that models informed by contact data fit well the observed serological profiles of both infections. We find that intimate types of contacts explain the pattern of acquisition of serological markers by age better than other types of social contacts. We observe similar patterns in each of the countries analysed, suggesting that there are consistent biological mechanisms at work.