The importance of supplementary immunisation activities to prevent measles outbreaks during the COVID-19 pandemic in Kenya.

BACKGROUND: The COVID-19 pandemic has disrupted routine measles immunisation and supplementary immunisation activities (SIAs) in most countries including Kenya. We assessed the risk of measles outbreaks during the pandemic in Kenya as a case study for the African Region. METHODS: Combining measles serological data, local contact patterns, and vaccination coverage into a cohort model, we predicted the age-adjusted population immunity in Kenya and estimated the probability of outbreaks when contact-reducing COVID-19 interventions are lifted. We considered various scenarios for reduced measles vaccination coverage from April 2020. RESULTS: In February 2020, when a scheduled SIA was postponed, population immunity was close to the herd immunity threshold and the probability of a large outbreak was 34% (8-54). As the COVID-19 contact restrictions are nearly fully eased, from December 2020, the probability of a large measles outbreak will increase to 38% (19-54), 46% (30-59), and 54% (43-64) assuming a 15%, 50%, and 100% reduction in measles vaccination coverage. By December 2021, this risk increases further to 43% (25-56), 54% (43-63), and 67% (59-72) for the same coverage scenarios respectively. However, the increased risk of a measles outbreak following the lifting of all restrictions can be overcome by conducting a SIA with ≥ 95% coverage in under-fives. CONCLUSION: While contact restrictions sufficient for SAR-CoV-2 control temporarily reduce measles transmissibility and the risk of an outbreak from a measles immunity gap, this risk rises rapidly once these restrictions are lifted. Implementing delayed SIAs will be critical for prevention of measles outbreaks given the roll-back of contact restrictions in Kenya.

Influence of age, severity of infection, and co-infection on the duration of respiratory syncytial virus (RSV) shedding.

RSV is the most important viral cause of pneumonia and bronchiolitis in children worldwide and has been associated with significant disease burden. With the renewed interest in RSV vaccines, we provide realistic estimates on duration, and influencing factors on RSV shedding which are required to better understand the impact of vaccination on the virus transmission dynamics. The data arise from a prospective study of 47 households (493 individuals) in rural Kenya, followed through a 6-month period of an RSV seasonal outbreak. Deep nasopharyngeal swabs were collected twice each week from all household members, irrespective of symptoms, and tested for RSV by multiplex PCR. The RSV G gene was sequenced. A total of 205 RSV infection episodes were detected in 179 individuals from 40 different households. The infection data were interval censored and assuming a random event time between observations, the average duration of virus shedding was 11·2 (95% confidence interval 10·1-12·3) days. The shedding durations were longer than previous estimates (3·9-7·4 days) based on immunofluorescence antigen detection or viral culture, and were shown to be strongly associated with age, severity of infection, and revealed potential interaction with other respiratory viruses. These findings are key to our understanding of the spread of this important virus and are relevant in the design of control programmes.

The source of respiratory syncytial virus infection in infants: a household cohort study in rural Kenya.

BACKGROUND: Respiratory syncytial virus (RSV) vaccine development for direct protection of young infants faces substantial obstacles. Assessing the potential of indirect protection using different strategies, such as targeting older children or mothers, requires knowledge of the source of infection to the infants. METHODS: We undertook a prospective study in rural Kenya. Households with a child born after the preceding RSV epidemic and ≥ 1 elder sibling were recruited. Nasopharyngeal swab samples were collected every 3-4 days irrespective of symptoms from all household members throughout the RSV season of 2009-2010 and tested for RSV using molecular techniques. RESULTS: From 451 participants in 44 households a total of 15 396 nasopharyngeal swab samples were samples were collected, representing 86% of planned sampling. RSV was detected in 37 households (84%) and 173 participants (38%) and 28 study infants (64%). The infants acquired infection from within (15 infants; 54%) or outside (9 infants; 32%) the household; in 4 households the source of infant infection was inconclusive. Older children were index case patients for 11 (73%) of the within-household infant infections, and 10 of these 11 children were attending school. CONCLUSION: We demonstrate that school-going siblings frequently introduce RSV into households, leading to infection in infants.

Kinetics of the neutralizing antibody response to respiratory syncytial virus infections in a birth cohort.

The kinetics of respiratory syncytial virus (RSV) neutralizing antibodies following birth, primary and secondary infections are poorly defined. The aims of the study were to measure and compare neutralizing antibody responses at different time points in a birth cohort followed-up over three RSV epidemics. Rural Kenyan children, recruited at birth between 2002 and 2003, were monitored for RSV infection over three epidemic seasons. Cord and 3-monthly sera, and acute and convalescent sera following RSV infection, were assayed in 28 children by plaque reduction neutralization test (PRNT). Relative to the neutralizing antibody titers of pre-exposure control sera (1.8 log10 PRNT), antibody titers following primary infection were (i) no different in sera collected between 0 and 0.4 months post-infection (1.9 log10 PRNT, P=0.146), (ii) higher in sera collected between 0.5 and 0.9 (2.8 log10 PRNT, P<0.0001), 1.0-1.9 (2.5 log10 PRNT, P<0.0001), and 2.0-2.9 (2.3 log10 PRNT, P<0.001) months post-infection, and (iii) no different in sera collected at between 3.0 and 3.9 months post-infection (2.0 log10 PRNT, P=0.052). The early serum neutralizing response to secondary infection (3.02 log10 PRNT) was significantly greater than the early primary response (1.9 log10 PRNT, P<0.0001). Variation in population-level virus transmission corresponded with changes in the mean cohort-level neutralizing titers. It is concluded that following primary RSV infection the neutralizing antibody response declines to pre-infection levels rapidly (~3 months) which may facilitate repeat infection. The kinetics of the aggregate levels of acquired antibody reflect seasonal RSV occurrence, age, and infection history.

The natural history of respiratory syncytial virus in a birth cohort: the influence of age and previous infection on reinfection and disease.

This study aimed to quantify the effect of age, time since last infection, and infection history on the rate of respiratory syncytial virus infection and the effect of age and infection history on the risk of respiratory syncytial virus disease. A birth cohort of 635 children in Kilifi, Kenya, was monitored for respiratory syncytial virus infections from January 31, 2002, to April 22, 2005. Predictors of infection were examined by Cox regression and disease risk by binomial regression. A total of 598 respiratory syncytial virus infections were identified (411 primary, 187 repeat), with 409 determined by antigen assay and 189 by antibody alone (using a "most pragmatic" serologic definition). The incidence decreased by 70% following a primary infection (adjusted hazard ratio = 0.30, 95% confidence interval: 0.21, 0.42; P < 0.001) and by 59% following a secondary infection (hazard ratio = 0.41, 95% confidence interval: 0.22, 0.73; P = 0.003), for a period lasting 6 months. Relative to the age group <6 months, all ages exhibited a higher incidence of infection. A lower risk of severe disease following infection was independently associated with increasing age (P < 0.001) but not reinfection. In conclusion, observed respiratory syncytial virus incidence was lowest in the first 6 months of life, immunity to reinfection was partial and short lived, and disease risk was age related.

Molecular epidemiology of human rhinovirus infections in Kilifi, coastal Kenya.

This study reports pediatric surveillance over 3 years for human rhinovirus (HRV) at the District Hospital of Kilifi, coastal Kenya. Nasopharyngeal samples were collected from children presenting at outpatient clinic with no signs of acute respiratory infection, or with signs of upper respiratory tract infection, and from children admitted to the hospital with lower respiratory tract infection. Samples were screened by real-time reverse transcriptase polymerase chain reaction (real-time RT-PCR) and classified further to species by nucleotide sequencing of the VP4/VP2 junction. Of 441 HRV positives by real-time RT-PCR, 332 were classified to species, with 47% (155) being HRV-A, 5% (18) HRV-B, and 48% (159) HRV-C. There was no clear seasonal pattern of occurrence for any species. The species were present in similar proportions in the inpatient and outpatient sample sets, and no significant association between species distribution and the severity of lower respiratory tract infection in the inpatients could be determined. HRV sequence analysis revealed multiple but separate clusters in circulation particularly for HRV-A and HRV-C. Most HRV-C clusters were distinct from reference sequences downloaded from GenBank. In contrast, most HRV-A and HRV-B sequences clustered with either known serotypes or strains from elsewhere within Africa and other regions of the world. This first molecular epidemiological study of HRV in the region defines species distribution in accord with reports from elsewhere in the world, shows considerable strain diversity and does not identify an association between any species and disease severity.

Epidemiology of COVID-19 infections on routine polymerase chain reaction (PCR) and serology testing in Coastal Kenya.

Background: There are limited studies in Africa describing the epidemiology, clinical characteristics and serostatus of individuals tested for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) infection. We tested routine samples from the Coastal part of Kenya between 17 th March 2020 and 30 th June 2021. Methods: SARS-CoV-2 infections identified using reverse transcription polymerase chain reaction (RT-PCR) and clinical surveillance data at the point of sample collection were used to classify as either symptomatic or asymptomatic. IgG antibodies were measured in sera samples, using a well validated in-house enzyme-linked immunosorbent assay (ELISA). Results: Mombasa accounted for 56.2% of all the 99,694 naso-pharyngeal/oro-pharyngeal swabs tested, and males constituted the majority tested (73.4%). A total of 7737 (7.7%) individuals were SARS-CoV-2 positive by RT-PCR. The majority (i.e., 92.4%) of the RT-PCR positive individuals were asymptomatic. Testing was dominated by mass screening and travellers, and even at health facility level 91.6% of tests were from individuals without symptoms. Out of the 97,124 tests from asymptomatic individuals 7,149 (7%) were positive and of the 2,568 symptomatic individuals 588 (23%) were positive. In total, 2458 serum samples were submitted with paired naso-pharyngeal/oro-pharyngeal samples and 45% of the RT-PCR positive samples and 20% of the RT-PCR negative samples were paired with positive serum samples. Symptomatic individuals had significantly higher antibody levels than asymptomatic individuals and become RT-PCR negative on repeat testing earlier than asymptomatic individuals. Conclusions: In conclusion, the majority of SARS-CoV-2 infections identified by routine testing in Coastal Kenya were asymptomatic. This reflects the testing practice of health services in Kenya, but also implies that asymptomatic infection is very common in the population. Symptomatic infection may be less common, or it may be that individuals do not present for testing when they have symptoms.

Hepatitis-B virus endemicity: heterogeneity, catastrophic dynamics and control.

Hepatitis-B virus infection is globally ubiquitous, but its distribution is very heterogeneous, with prevalence of serological markers in various nations ranging from less than 1% to more than 90%. We propose an explanation for this diversity using a mathematical model of hepatitis-B virus transmission dynamics that shows, for the first time, 'catastrophic' behavior using realistic epidemiological processes and parameters. Our major conclusion is that the prevalence of infection is largely determined by a feedback mechanism that relates the rate of transmission, average age at infection and age-related probability of developing carriage following infection. Using the model we identify possible, highly non-linear, consequences of chemotherapy and immunization interventions, for which the starting prevalence of carriers is the most influential, predictive quantity. Taken together, our results demand a re-evaluation of public health policy towards hepatitis-B.

Quantifying previous SARS-CoV-2 infection through mixture modelling of antibody levels.

As countries decide on vaccination strategies and how to ease movement restrictions, estimating the proportion of the population previously infected with SARS-CoV-2 is important for predicting the future burden of COVID-19. This proportion is usually estimated from serosurvey data in two steps: first the proportion above a threshold antibody level is calculated, then the crude estimate is adjusted using external estimates of sensitivity and specificity. A drawback of this approach is that the PCR-confirmed cases used to estimate the sensitivity of the threshold may not be representative of cases in the wider population-e.g., they may be more recently infected and more severely symptomatic. Mixture modelling offers an alternative approach that does not require external data from PCR-confirmed cases. Here we illustrate the bias in the standard threshold-based approach by comparing both approaches using data from several Kenyan serosurveys. We show that the mixture model analysis produces estimates of previous infection that are often substantially higher than the standard threshold analysis.

Rotavirus within day care centres in Oxfordshire, UK: characterization of partial immunity.

Repeated measures data for rotavirus infection in children within 14 day care centres (DCCs) in the Oxfordshire area, UK, are used to explore aspects of rotavirus transmission and immunity. A biologically realistic model for the transmission of infection is presented as a set of probability models suitable for application to the data. Two transition events are modelled separately: incidence and recovery. The complexity of the underlying mechanistic model is reflected in the choice of the fixed variables in the probability models. Parameter estimation was carried out using a Bayesian Markov chain Monte Carlo method. We use the parameter estimates obtained to build a profile of the natural history of rotavirus reinfection in an individual child. We infer that rotavirus transmission in children in DCCs is dependent on the DCC prevalence, with symptomatic infection of longer duration, but no more infectious per day of infectious period, than asymptomatic infection. There was evidence that a recent previous infection reduces the risk of disease and, to a lesser extent, reinfection, but not duration of infection. The results provide evidence that partial immunity to rotavirus infection develops over several time scales.

Evaluating the performance of tools used to call minority variants from whole genome short-read data.

Background: High-throughput whole genome sequencing facilitates investigation of minority virus sub-populations from virus positive samples. Minority variants are useful in understanding within and between host diversity, population dynamics and can potentially assist in elucidating person-person transmission pathways. Several minority variant callers have been developed to describe low frequency sub-populations from whole genome sequence data. These callers differ based on bioinformatics and statistical methods used to discriminate sequencing errors from low-frequency variants. Methods: We evaluated the diagnostic performance and concordance between published minority variant callers used in identifying minority variants from whole-genome sequence data from virus samples. We used the ART-Illumina read simulation tool to generate three artificial short-read datasets of varying coverage and error profiles from an RSV reference genome. The datasets were spiked with nucleotide variants at predetermined positions and frequencies. Variants were called using FreeBayes, LoFreq, Vardict, and VarScan2. The variant callers' agreement in identifying known variants was quantified using two measures; concordance accuracy and the inter-caller concordance. Results: The variant callers reported differences in identifying minority variants from the datasets. Concordance accuracy and inter-caller concordance were positively correlated with sample coverage. FreeBayes identified the majority of variants although it was characterised by variable sensitivity and precision in addition to a high false positive rate relative to the other minority variant callers and which varied with sample coverage. LoFreq was the most conservative caller. Conclusions: We conducted a performance and concordance evaluation of four minority variant calling tools used to identify and quantify low frequency variants. Inconsistency in the quality of sequenced samples impacts on sensitivity and accuracy of minority variant callers. Our study suggests that combining at least three tools when identifying minority variants is useful in filtering errors when calling low frequency variants.

Molecular characterization of rotavirus group A strains circulating prior to vaccine introduction in rural coastal Kenya, 2002-2013.

Background: Kenya introduced the monovalent Rotarix® rotavirus group A (RVA) vaccine nationally in mid-2014.  Long-term surveillance data is important prior to wide-scale vaccine use to assess the impact on disease and to investigate the occurrence of heterotypic strains arising through immune selection. This report presents baseline data on RVA genotype circulation patterns and intra-genotype genetic diversity over a 7-year period in the pre-vaccine era in Kilifi, Kenya, from 2002 to 2004 and from 2010 to 2013. Methods: A total of 745 RVA strains identified in children admitted with acute gastroenteritis to a referral hospital in Coastal Kenya, were sequenced using the di-deoxy sequencing method in the VP4 and VP7 genomic segments (encoding P and G proteins, respectively). Sequencing successfully generated 569 (76%) and 572 (77%) consensus sequences for the VP4 and VP7 genes respectively. G and P genotypes were determined by use of BLAST and the online RotaC v2 RVA classification tool. Results: The most common GP combination was G1P[8] (51%), similar to the Rotarix® strain, followed by G9P[8] (15%) , G8P[4] (14%) and G2P[4] (5%).  Unusual GP combinations-G1P[4], G2P[8], G3P[4,6], G8P[8,14], and G12P[4,6,8]-were observed at frequencies of <5%. Phylogenetic analysis showed that the infections were caused by both locally persistent strains as evidenced by divergence of local strains occurring over multiple seasons from the global ones, and newly introduced strains, which were closely related to global strains. The circulating RVA diversity showed temporal fluctuations both season by season and over the longer-term. None of the unusual strains increased in frequency over the observation period.   Conclusions: The circulating RVA diversity showed temporal fluctuations with several unusual strains recorded, which rarely caused major outbreaks.  These data will be useful in interpreting genotype patterns observed in the region during the vaccine era.

Human rhinovirus spatial-temporal epidemiology in rural coastal Kenya, 2015-2016, observed through outpatient surveillance.

Background: Human rhinovirus (HRV) is the predominant cause of upper respiratory tract infections, resulting in a significant public health burden. The virus circulates as many different types (168), each generating strong homologous, but weak heterotypic, immunity. The influence of these features on transmission patterns of HRV in the community is understudied. Methods: Nasopharyngeal swabs were collected from patients with symptoms of acute respiratory infection (ARI) at nine out-patient facilities across a Health and Demographic Surveillance System between December 2015 and November 2016. HRV was diagnosed by real-time RT-PCR, and the VP4/VP2 genomic region of the positive samples sequenced. Phylogenetic analysis was used to determine the HRV types. Classification models and G-test statistic were used to investigate HRV type spatial distribution. Demographic characteristics and clinical features of ARI were also compared. Results: Of 5,744 NPS samples collected, HRV was detected in 1057 (18.4%), of which 817 (77.3%) were successfully sequenced. HRV species A, B and C were identified in 360 (44.1%), 67 (8.2%) and 390 (47.7%) samples, respectively. In total, 87 types were determined: 39, 10 and 38 occurred within species A, B and C, respectively. HRV types presented heterogeneous temporal patterns of persistence. Spatially, identical types occurred over a wide distance at similar times, but there was statistically significant evidence for clustering of types between health facilities in close proximity or linked by major road networks. Conclusion: This study records a high prevalence of HRV in out-patient presentations exhibiting high type diversity. Patterns of occurrence suggest frequent and independent community invasion of different types. Temporal differences of persistence between types may reflect variation in type-specific population immunity. Spatial patterns suggest either rapid spread or multiple invasions of the same type, but evidence of similar types amongst close health facilities, or along road systems, indicate type partitioning structured by local spread.

Erratum: Whole genome analysis of local Kenyan and global sequences unravels the epidemiological and molecular evolutionary dynamics of RSV genotype ON1 strains.

[This corrects the article DOI: 10.1093/ve/vey027.][This corrects the article DOI: 10.1093/ve/vey027.].

Whole genome analysis of local Kenyan and global sequences unravels the epidemiological and molecular evolutionary dynamics of RSV genotype ON1 strains.

The respiratory syncytial virus (RSV) group A variant with the 72-nucleotide duplication in the G gene, genotype ON1, was first detected in Kilifi in 2012 and has almost completely replaced circulating genotype GA2 strains. This replacement suggests some fitness advantage of ON1 over the GA2 viruses in Kilifi, and might be accompanied by important genomic substitutions in ON1 viruses. Close observation of such a new virus genotype introduction over time provides an opportunity to better understand the transmission and evolutionary dynamics of the pathogen. We have generated and analysed 184 RSV-A whole-genome sequences (WGSs) from Kilifi (Kenya) collected between 2011 and 2016, the first ON1 genomes from Africa and the largest collection globally from a single location. Phylogenetic analysis indicates that RSV-A circulation in this coastal Kenya location is characterized by multiple introductions of viral lineages from diverse origins but with varied success in local transmission. We identified signature amino acid substitutions between ON1 and GA2 viruses' surface proteins (G and F), polymerase (L), and matrix M2-1 proteins, some of which were positively selected, and thereby provide an enhanced picture of RSV-A diversity. Furthermore, five of the eleven RSV open reading frames (ORFs) (G, F, L, N, and P) formed distinct phylogenetic clusters for the two genotypes. This might suggest that coding regions outside of the most frequently studied G ORF also play a role in the adaptation of RSV to host populations, with the alternative possibility that some of the substitutions are neutral and provide no selective advantage. Our analysis provides insight into the epidemiological processes that define RSV spread, highlights the genetic substitutions that characterize emerging strains, and demonstrates the utility of large-scale WGS in molecular epidemiological studies.

Understanding the transmission dynamics of respiratory syncytial virus using multiple time series and nested models.

The nature and role of re-infection and partial immunity are likely to be important determinants of the transmission dynamics of human respiratory syncytial virus (hRSV). We propose a single model structure that captures four possible host responses to infection and subsequent reinfection: partial susceptibility, altered infection duration, reduced infectiousness and temporary immunity (which might be partial). The magnitude of these responses is determined by four homotopy parameters, and by setting some of these parameters to extreme values we generate a set of eight nested, deterministic transmission models. In order to investigate hRSV transmission dynamics, we applied these models to incidence data from eight international locations. Seasonality is included as cyclic variation in transmission. Parameters associated with the natural history of the infection were assumed to be independent of geographic location, while others, such as those associated with seasonality, were assumed location specific. Models incorporating either of the two extreme assumptions for immunity (none or solid and lifelong) were unable to reproduce the observed dynamics. Model fits with either waning or partial immunity to disease or both were visually comparable. The best fitting structure was a lifelong partial immunity to both disease and infection. Observed patterns were reproduced by stochastic simulations using the parameter values estimated from the deterministic models.

Vaccination in pulses: a strategy for global eradication of measles and polio?

Recent American successes against poliomyelitis and measles have been attributed to repeated 'pulse' vaccination campaigns. Whilst logistic and economic constraints will be crucial, a deeper epidemiological understanding of the mechanism, strengths and weaknesses of pulse vaccination will optimize the chances of success elsewhere in the world.

Age- and sex-specific HIV-1 prevalence in the urban community setting of Addis Ababa, Ethiopia.

OBJECTIVE: To estimate the age and sex-specific prevalence of HIV infection in the population of Addis Ababa, Ethiopia. DESIGN: Two-stage cluster sampling of the population aged 0-49 years of Addis Ababa, using kebeles (urban dwelling associations) as clusters. METHODS: The sera used for this study were collected in an earlier study (1994) on the rate of acquisition of antibodies against measles, rubella, and hepatitis B. After separate approvals were obtained from the institutional ethics committees, sera were tested by enzyme-linked immunosorbent assay confirmed by Western blot. Age- and sex-specific HIV prevalence rates were estimated. The prevalence of HIV in men and women over 15 years of age was compared by calculating age-standardized HIV prevalence, using the age distribution of the census population as the standard. A time-dependent catalytic model was used to obtain crude estimates of HIV incidence from age-prevalence data. RESULTS: A total of 3853 sera were available for analysis. The prevalence of HIV in adults was 6.0% [95% confidence interval (CI), 4.5-7.4%] for men and 6.9% (95% CI, 5.3-8.5%) for women, with peak prevalence in the 25-29 year age group of 16.3 and 11.8%, respectively. After standardization for age using the direct method, the HIV prevalence ratio comparing adult men with women was 0.97:1 (95% CI, 0.70:1 - 1.35:1). Three children aged less than 5 years were HIV-positive. The prevalence of HIV among adults ranged from 0-21.3% in different clusters, indicating the heterogeneity of the spread of HIV in the city. HIV prevalence estimates among the antenatal clinic patients of Addis Ababa in 1996 far exceeded the estimates obtained during the community survey, particularly in the youngest age group (15-24 years). Estimates of HIV incidence (per susceptible person per annum) for the age group 16-22 years ranged from 1.3-2.25% for men and from 2.1-2.4% for women. CONCLUSION: By 1994, a substantial proportion of the adult population of Addis Ababa was infected with HIV. Promotion of behavioural changes and the control of sexually transmitted diseases should be strongly supported to limit the spread of the HIV epidemic in Ethiopia.

Who mixes with whom? A method to determine the contact patterns of adults that may lead to the spread of airborne infections.

Although mixing patterns are thought to be important determinants of the spread of airborne infectious diseases, to our knowledge, there have been no attempts to directly quantify them for humans. We report on a preliminary study to identify such mixing patterns. A sample of 92 adults were asked to detail the individuals with whom they had conversed over the period of one, randomly assigned, day. Sixty-five (71%) completed the questionnaire, providing their age, the age of their contacts and the social context in which the contacts took place. The data were analysed using multilevel modelling. The study identified, and allowed the quantification of, contact patterns within this sample that may be of epidemiological significance. For example, the degree of assortativeness of mixing with respect to age was dependent not only on the age of participants but the number of contacts made. Estimates of the relative magnitude of contact rates between different social settings were made, with implications for outbreak potential. Simple questionnaire modifications are suggested which would yield information on the structure and dynamics of social networks and the intensity of contacts. Surveys of this nature may enable the quantification of who acquires infection from whom and from where.

The influence of age on the development of the hepatitis B carrier state.

The relation between the age at infection with hepatitis B virus (HBV) and the development of the carrier state is examined by using data from a number of published and unpublished surveys. A remarkably consistent relation was found. Infants infected perinatally (within the first 6 months of life) were found to have a high probability of becoming carriers (0.885; 95% C.L. 0.84-0.93). Over the infant and early childhood age classes there was found to be a sharp decrease in the proportion of infections which lead to the carrier state. By adulthood (over 15 years) the probability of developing the carrier status was found to be about 0.1. A model was fitted to the data by using maximum likelihood, which provides a good empirical description of the observed data and can be used to predict the expected probability of developing the carrier state given the age at infection. It is postulated that, as a result of this rapid decline in the probability of becoming a carrier during early childhood, a mass childhood immunization campaign, which will tend to postpone the average age at infection in the unvaccinated community, will have a disproportionately large impact on the rate of generation of new carriers.