Bayesian modeling to unmask and predict influenza A/H1N1pdm dynamics in London.

The tracking and projection of emerging epidemics is hindered by the disconnect between apparent epidemic dynamics, discernible from noisy and incomplete surveillance data, and the underlying, imperfectly observed, system. Behavior changes compound this, altering both true dynamics and reporting patterns, particularly for diseases with nonspecific symptoms, such as influenza. We disentangle these effects to unravel the hidden dynamics of the 2009 influenza A/H1N1pdm pandemic in London, where surveillance suggests an unusual dominant peak in the summer. We embed an age-structured model into a bayesian synthesis of multiple evidence sources to reveal substantial changes in contact patterns and health-seeking behavior throughout the epidemic, uncovering two similar infection waves, despite large differences in the reported levels of disease. We show how this approach, which allows for real-time learning about model parameters as the epidemic progresses, is also able to provide a sequence of nested projections that are capable of accurately reflecting the epidemic evolution.

Dynamic models of pneumococcal carriage and the impact of the Heptavalent Pneumococcal Conjugate Vaccine on invasive pneumococcal disease.

BACKGROUND: The 7-valent pneumococcal conjugate vaccine has been introduced in national immunisation programmes of most industrialised countries and recently in two African GAVI eligible countries (Rwanda and The Gambia). However the long term effects of PCV are still unclear, as beneficial direct and herd immunity effects might be countered by serotype replacement. METHOD: A dynamic, age-structured, compartmental model of Streptococcus pneumoniae transmission was developed to predict the potential impact of PCV7 on the incidence of invasive disease accounting for both herd immunity and serotype replacement effects. The model was parameterised using epidemiological data from England and Wales and pre and post-vaccination surveillance data from the US. RESULTS: Model projections showed that serotype replacement plays a crucial role in determining the overall effect of a PCV7 vaccination programme and could reduce, negate or outweigh its beneficial impact. However, using the estimate of the competition parameter derived from the US post-vaccination experience, an infant vaccination programme would prevent 39,000 IPD cases in the 20 years after PCV7 introduction in the UK. Adding a catch-up campaign for under 2 or under 5 year olds would provide a further reduction of 1,200 or 3,300 IPD cases respectively, mostly in the first few years of the programme. CONCLUSIONS: This analysis suggests that a PCV vaccination programme would eradicate vaccine serotypes from circulation. However, the increase in carriage of non-vaccine serotypes, and the consequent increase in invasive disease, could reduce, negate or outweigh the benefit. These results are sensitive to changes in the protective effect of the vaccine, and, most importantly, to the level of competition between vaccine and non-vaccine types. The techniques developed here can be used to assess the introduction of vaccination programmes in developing countries and provide the basis for cost-effectiveness analyses.

Interpreting changes in measles genotype: the contribution of chance, migration and vaccine coverage.

BACKGROUND: In some populations, complete shifts in the genotype of the strain of measles circulating in the population have been observed, with given genotypes being replaced by new genotypes. Studies have postulated that such shifts may be attributable to differences between the fitness of the new and the old genotypes. METHODS: We developed a stochastic model of the transmission dynamics of measles, simulating the effects of different levels of migration, vaccination coverage and importation of new genotypes on patterns in the persistence and replacement of indigenous genotypes. RESULTS: The analyses illustrate that complete replacement in the genotype of the strain circulating in populations may occur because of chance. This occurred in >50% of model simulations, for levels of vaccination coverage and numbers of imported cases per year which are compatible with those observed in several Western European populations (>80% and >3 per million per year respectively) and for the given assumptions in the model. CONCLUSION: The interpretation of genotypic data, which are increasingly being collected in surveillance programmes, needs to take account of the underlying vaccination coverage and the level of the importation rate of measles cases into the population.

The potential for measles transmission in England.

BACKGROUND: Since the schools vaccination campaign in 1994, measles has been eliminated from England. Maintaining elimination requires low susceptibility levels to keep the effective reproduction number R below 1. Since 1995, however, MMR coverage in two year old children has decreased by more than 10%. METHODS: Quarterly MMR coverage data for children aged two and five years resident in each district health authority in England were used to estimate susceptibility to measles by age. The effective reproduction numbers for each district and strategic health authority were calculated and possible outbreak sizes estimated. RESULTS: In 2004/05, about 1.9 million school children and 300,000 pre-school children were recorded as incompletely vaccinated against measles in England, including more than 800,000 children completely unvaccinated. Based on this, approximately 1.3 million children aged 2-17 years were susceptible to measles. In 14 of the 99 districts, the level of susceptibility is sufficiently high for R to exceed 1, indicating the potential for sustained measles transmission. Eleven of these districts are in London. Our model suggests that the potential exists for an outbreak of up to 100,000 cases. These results are sensitive to the accuracy of reported vaccination coverage data. CONCLUSION: Our analysis identified several districts with the potential for sustaining measles transmission. Many London areas remain at high risk even allowing for considerable under-reporting of coverage. Primary care trusts should ensure that accurate systems are in place to identify unimmunised children and to offer catch-up immunisation for those not up to date for MMR.

Delaying the international spread of pandemic influenza.

BACKGROUND: The recent emergence of hypervirulent subtypes of avian influenza has underlined the potentially devastating effects of pandemic influenza. Were such a virus to acquire the ability to spread efficiently between humans, control would almost certainly be hampered by limited vaccine supplies unless global spread could be substantially delayed. Moreover, the large increases that have occurred in international air travel might be expected to lead to more rapid global dissemination than in previous pandemics. METHODS AND FINDINGS: To evaluate the potential of local control measures and travel restrictions to impede global dissemination, we developed stochastic models of the international spread of influenza based on extensions of coupled epidemic transmission models. These models have been shown to be capable of accurately forecasting local and global spread of epidemic and pandemic influenza. We show that under most scenarios restrictions on air travel are likely to be of surprisingly little value in delaying epidemics, unless almost all travel ceases very soon after epidemics are detected. CONCLUSIONS: Interventions to reduce local transmission of influenza are likely to be more effective at reducing the rate of global spread and less vulnerable to implementation delays than air travel restrictions. Nevertheless, under the most plausible scenarios, achievable delays are small compared with the time needed to accumulate substantial vaccine stocks.

Developing a realistic sexual network model of chlamydia transmission in Britain.

BACKGROUND: A national chlamydia screening programme is currently being rolled out in the UK and other countries. However, much of the epidemiology remains poorly understood. In this paper we present a stochastic, individual based, dynamic sexual network model of chlamydia transmission and its parameterisation. Mathematical models provide a theoretical framework for understanding the key epidemiological features of chlamydia: sexual behaviour, health care seeking and transmission dynamics. RESULTS: The model parameters were estimated either directly or by systematic fitting to a variety of appropriate data sources. The fitted model was representative of sexual behaviour, chlamydia epidemiology and health care use in England. We were able to recapture the observed age distribution of chlamydia prevalence. CONCLUSION: Estimating parameters for models of sexual behaviour and transmission of chlamydia is complex. Most of the parameter values are highly correlated, highly variable and there is little empirical evidence to inform estimates. We used a novel approach to estimate the rate of active treatment seeking, by combining data sources, which improved the credibility of the model results. The model structure is flexible and is broadly applicable to other developed world settings and provides a practical tool for public health decision makers.

Reconstructing a spatially heterogeneous epidemic: Characterising the geographic spread of 2009 A/H1N1pdm infection in England.

Understanding how the geographic distribution of and movements within a population influence the spatial spread of infections is crucial for the design of interventions to curb transmission. Existing knowledge is typically based on results from simulation studies whereas analyses of real data remain sparse. The main difficulty in quantifying the spatial pattern of disease spread is the paucity of available data together with the challenge of incorporating optimally the limited information into models of disease transmission. To address this challenge the role of routine migration on the spatial pattern of infection during the epidemic of 2009 pandemic influenza in England is investigated here through two modelling approaches: parallel-region models, where epidemics in different regions are assumed to occur in isolation with shared characteristics; and meta-region models where inter-region transmission is expressed as a function of the commuter flux between regions. Results highlight that the significantly less computationally demanding parallel-region approach is sufficiently flexible to capture the underlying dynamics. This suggests that inter-region movement is either inaccurately characterized by the available commuting data or insignificant once its initial impact on transmission has subsided.

Potential overestimation of HPV vaccine impact due to unmasking of non-vaccine types: quantification using a multi-type mathematical model.

INTRODUCTION: Estimates of human papillomavirus (HPV) vaccine impact in clinical trials and modelling studies rely on DNA tests of cytology or biopsy specimens to determine the HPV type responsible for a cervical lesion. DNA of several oncogenic HPV types may be detectable in a specimen. However, only one type may be responsible for a particular cervical lesion. Misattribution of the causal HPV type for a particular abnormality may give rise to an apparent increase in disease due to non-vaccine HPV types following vaccination ("unmasking"). METHODS: To investigate the existence and magnitude of unmasking, we analysed data from residual cytology and biopsy specimens in English women aged 20-64 years old using a stochastic type-specific individual-based model of HPV infection, progression and disease. The model parameters were calibrated to data on the prevalence of HPV DNA and cytological lesion of different grades, and used to assign causal HPV types to cervical lesions. The difference between the prevalence of all disease due to non-vaccine HPV types, and disease due to non-vaccine HPV types in the absence of vaccine HPV types, was then estimated. RESULTS: There could be an apparent maximum increase of 3-10% in long-term cervical cancer incidence due to non-vaccine HPV types following vaccination. CONCLUSION: Unmasking may be an important phenomenon in HPV post-vaccination epidemiology, in the same way that has been observed following pneumococcal conjugate vaccination.

The cost-effectiveness of varicella and combined varicella and herpes zoster vaccination programmes in the United Kingdom.

BACKGROUND: Despite the existence of varicella vaccine, many developed countries have not introduced it into their national schedules, partly because of concerns about whether herpes zoster (HZ, shingles) will increase due to a lack of exogenous boosting. The magnitude of any increase in zoster that might occur is dependent on rates at which adults and children mix - something that has only recently been quantified - and could be reduced by simultaneously vaccinating older individuals against shingles. This study is the first to assess the cost-effectiveness of combined varicella and zoster vaccination options and compare this to alternative programmes. METHODS AND FINDINGS: The cost-effectiveness of various options for the use of varicella-zoster virus (VZV) containing vaccines was explored using a transmission dynamic model. Underlying contact rates are estimated from a contemporary survey of social mixing patterns, and uncertainty in these derived from bootstrapping the original sample. The model was calibrated to UK data on varicella and zoster incidence. Other parameters were taken from the literature. UK guidance on perspective and discount rates were followed. The results of the incremental cost-effectiveness analysis suggest that a combined policy is cost-effective. However, the cost-effectiveness of this policy (and indeed the childhood two-dose policy) is influenced by projected benefits that accrue many decades (80-100 years or more) after the start of vaccination. If the programme is evaluated over shorter time frames, then it would be unlikely to be deemed cost-effective, and may result in declines in population health, due to a projected rise in the incidence of HZ. The findings are also sensitive to a number of parameters that are inaccurately quantified, such as the risk of HZ in varicella vaccine responders. CONCLUSIONS: Policy makers should be aware of the potential negative benefits in the first 30-50 years after introduction of a childhood varicella vaccine. This can only be partly mitigated by the introduction of a herpes zoster vaccine. They have to decide how they value the potential benefits beyond this time to consider childhood vaccination cost effective.

Mathematical modelling long-term effects of replacing Prevnar7 with Prevnar13 on invasive pneumococcal diseases in England and Wales.

INTRODUCTION: England and Wales recently replaced the 7-valent pneumococcal conjugate vaccine (PCV7) with its 13-valent equivalent (PCV13), partly based on projections from mathematical models of the long-term impact of such a switch compared to ceasing pneumococcal conjugate vaccination altogether. METHODS: A compartmental deterministic model was used to estimate parameters governing transmission of infection and competition between different groups of pneumococcal serotypes prior to the introduction of PCV13. The best-fitting parameters were used in an individual based model to describe pneumococcal transmission dynamics and effects of various options for the vaccination programme change in England and Wales. A number of scenarios were conducted using (i) different assumptions about the number of invasive pneumococcal disease cases adjusted for the increasing trend in disease incidence prior to PCV7 introduction in England and Wales, and (ii) a range of values representing serotype replacement induced by vaccination of the additional six serotypes in PCV13. RESULTS: Most of the scenarios considered suggest that ceasing pneumococcal conjugate vaccine use would cause an increase in invasive pneumococcal disease incidence, while replacing PCV7 with PCV13 would cause an overall decrease. However, the size of this reduction largely depends on the level of competition induced by the additional serotypes in PCV13. The model estimates that over 20 years of PCV13 vaccination, around 5000-62000 IPD cases could be prevented compared to stopping pneumococcal conjugate vaccination altogether. CONCLUSION: Despite inevitable uncertainty around serotype replacement effects following introduction of PCV13, the model suggests a reduction in overall invasive pneumococcal disease incidence in all cases. Our results provide useful evidence on the benefits of PCV13 to countries replacing or considering replacing PCV7 with PCV13, as well as data that can be used to evaluate the cost-effectiveness of such a switch.

Modelling the impact of a combined varicella and zoster vaccination programme on the epidemiology of varicella zoster virus in England.

This study updates previous work on modelling the incidence of varicella and Herpes Zoster (HZ) following the introduction of childhood vaccination. The updated model includes new data on age-specific contact patterns, as well as data on the efficacy of zoster vaccination in the elderly and allows for HZ among vaccinees. The current study also looks at two-dose varicella childhood programmes, and assesses the combined impact of varicella vaccination in childhood and zoster vaccination of the elderly. The results suggest that a two-dose schedule is likely to reduce the incidence of varicella to very low levels, provided first dose coverage is around 90% and second dose coverage is in excess of 70%. Single dose varicella vaccination programmes are expected to result in large numbers of breakthrough cases. Childhood vaccination is expected to increase the incidence of zoster for more than 40 years after introduction of the programme, the magnitude of this increase being influenced primarily by the duration of boosting following exposure to the varicella zoster virus. Though this increase in zoster incidence can be partly offset by vaccination of the elderly, the effectiveness of this combined strategy is limited, as much of the increase occurs in those adults too young to be vaccinated. Childhood vaccination at intermediate levels of coverage (70% and 60% for first and second dose coverage respectively) is expected to lead to an increase in adult varicella. At high coverage (90% and 80% coverage) this is unlikely to be the case. These results will be used to inform a cost-effectiveness analysis of combined varicella and zoster vaccination programmes.

What types of contacts are important for the spread of infections?: using contact survey data to explore European mixing patterns.

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.

Economic evaluation of infant and adolescent hepatitis B vaccination in the UK.

A Markov model of hepatitis B virus (HBV) disease progression in the UK estimated that 81% of predicted HBV-associated morbidity and mortality could be prevented by universal infant vaccination at a cost of approximately £ 260,000 per QALY gained. Universal adolescent vaccination would be less effective (45% prevented) and less cost-effective (£ 493,000 per QALY gained). Higher HBV incidence rates in males and intermediate/high risk ethnic populations meant it was approximately 3 times more cost-effective to vaccinate these groups. At current vaccine costs a selective infant vaccination programme, based on vaccinating intermediate/high risk ethnic populations would not be considered cost effective. The threshold cost per vaccinated child at which the programme would be considered cost-effective was investigated. Universal infant vaccination would be cost-effective if the average cost of vaccinating each child against HBV, including vaccine and administration costs of all doses, was less than £ 4.09. Given the low cost of vaccination required to make a universal programme cost-effective the most feasible policy in the UK would be to use a suitably priced combined vaccine that included the other antigens in the current infant vaccination schedule.

7-Valent pneumococcal conjugate vaccination in England and Wales: is it still beneficial despite high levels of serotype replacement?

BACKGROUND: The UK introduced the 7-valent pneumococcal conjugate vaccine (PCV7) into the national vaccination program in September 2006. Previous modelling assumed that the likely impact of PCV7 on invasive pneumococcal disease (IPD) would be similar to the US experience with PCV7. However, recent surveillance data show a more rapid replacement of PCV7 IPD cases by non-PCV7 IPD cases than was seen in the US. METHODS AND FINDINGS: A previous model of pneumococcal vaccination was re-parameterised using data on vaccine coverage and IPD from England and Wales between 2006 and 2009. Disease incidence was adjusted for the increasing trend in reported IPD cases prior to vaccination. Using this data we estimated that individuals carrying PCV7 serotypes have much higher protection (96%;95% CI 72%-100%) against acquisition of NVT carriage than the 15% previously estimated from US data, which leads to greater replacement. However, even with this level of replacement, the annual number of IPD cases may be 560 (95% CI, -100 to 1230) lower ten years after vaccine introduction compared to what it may have been without vaccination. A particularly marked fall of 39% in children under 15 years by 2015/6 is predicted. CONCLUSION: Our model suggests that PCV7 vaccination could result in a decrease in overall invasive pneumococcal disease, particularly in children, even in an environment of rapid replacement with non-PCV7 serotypes within 5 years of vaccine introduction at high coverage.

Estimating progression rates for human papillomavirus infection from epidemiological data.

A Markov model was constructed in order to estimate type-specific rates of cervical lesion progression and regression in women with high-risk human papillomavirus (HPV). The model was fitted to age- and type-specific data regarding the HPV DNA and cytological status of women undergoing cervical screening in a recent screening trial, as well as cervical cancer incidence. It incorporates different assumptions about the way lesions regress, the accuracy of cytological screening, the specificity of HPV DNA testing, and the age-specific prevalence of HPV infection. Combinations of assumptions generate 162 scenarios for squamous cell carcinomas and 54 scenarios for adenocarcinomas. Simulating an unscreened cohort of women infected with high-risk HPV indicates that the probability of an infection continuing to persist and to develop into invasive cancer depends on the length of time it has already persisted. The scenarios and parameter sets that produce the best fit to available epidemiological data provide a basis for modeling the natural history of HPV infection and disease.

Transmission dynamic modelling of the impact of human papillomavirus vaccination in the United Kingdom.

Many countries are considering vaccination against human papillomavirus (HPV). However, the long-term impact of vaccination is difficult to predict due to uncertainty about the prevalence of HPV infection, pattern of sexual partnerships, progression of cervical neoplasias, accuracy of screening as well as the duration of infectiousness and immunity. Dynamic models of human papillomavirus (HPV) transmission were developed to describe the infection spread and development of cervical neoplasia, cervical cancer (squamous cell and adenocarcinoma) and anogenital warts. Using different combinations of assumptions, 9900 scenarios were created. Each scenario was then fitted to epidemiological data and the best-fitting scenarios used to predict the impact of vaccination. Results suggest that vaccinating 12-year-old girls at 80% coverage will result in a 38-82% reduction in cervical cancer incidence and 44-100% reduction in anogenital warts incidence after 60 years of an ongoing vaccination programme if vaccine protection lasts 20 years on average. The marginal benefit of vaccinating boys depends on the degree of protection achieved by vaccinating girls.

Estimating the impact of childhood influenza vaccination programmes in England and Wales.

There is increasing interest in routine vaccination of children against influenza. We use an age-structured model to demonstrate that the long-term incidence of influenza A could decrease by 11-21% in the overall population by vaccinating individuals aged 6 to <24 months, and by 22-38% and 65-97% through targeting those aged 6 to <60 months and 6 months to 16 years, respectively. The corresponding reductions predicted for influenza B were 25-35%, 44-69% and 85-96%, respectively. These results are sensitive to assumptions about contact patterns and several parameters, including the vaccine efficacy among those aged <24 months, require further study. Consistently high levels of vaccination coverage among pre-school children has the potential to bring benefits to both those vaccinated and the community.

Pneumococcal carriage in United Kingdom families: estimating serotype-specific transmission parameters from longitudinal data.

Repeated observations of pneumococcal infection in 121 United Kingdom families (October 2001-July 2002) were used to explore the transmission properties of five highly prevalent pneumococcal serotypes (6A, 6B, 14, 19F, 23F). A family-based Markov model was developed, and maximum likelihood estimates were produced for model parameters. The authors found higher community acquisition rates among preschool children for all serotypes and higher within-household transmission for 6A and 14. Significant differences in the spontaneous clearance rate were estimated between age categories and serotypes, with 6B being carried for almost 4 months in children. Different mechanisms of competition between serotypes were investigated, and a complete exclusion model (i.e., the resident strain cannot be outcompeted by challengers) was discarded in favor of a competing mechanism that leaves a resident serotype partially or fully susceptible to challengers. Large variation was found in the challenging strength, which was low for 19F and 23F and high for 6A and 6B. Serotype 6B was the only one characterized by high resistance capacity. Only small differences in the transmission characteristics were found when vaccine and nonvaccine serotypes were grouped, suggesting that a serotype-specific analysis is needed to detect distinctive serotype behavior.

Dynamic models of meningococcal carriage, disease, and the impact of serogroup C conjugate vaccination.

Much interest has surrounded the use of conjugate vaccines in recent years, with the development of vaccines against disease caused by Haemophilus influenzae type b, Neisseria meningitidis, and Streptococcus pneumoniae. These vaccines offer the potential for safe and effective disease control, but some questions remain, particularly regarding the duration and mechanisms of protection and the longer-term impact of vaccination on carriage. In this paper, the authors use data on immunization with serogroup C meningococcal conjugate vaccines in England and Wales to develop and apply a mathematical model to investigate the direct and indirect (herd immunity) effects of a conjugate vaccine program. A realistic, age-structured, dynamic model was developed and parameterized and was fitted to epidemiologic data from England and Wales. The effects of a range of vaccine strategies, including hypothetical scenarios, were investigated. The basic reproduction number was estimated to be 1.36. Catch-up vaccination targeting teenagers generated substantial herd immunity and was important in controlling disease rapidly. The results were sensitive to changes in the assumptions regarding the method of vaccine action, particularly duration of protection and efficacy of vaccination against carriage acquisition. This model can be used to help predict the potential impact of vaccine strategies both in the United Kingdom and elsewhere.