Estimation of vaccine effectiveness using the screening method

The screening method provides a simple and rapid way of estimating vaccine effectiveness. This paper discusses the validity of the screening method with particular reference to bias and precision. Methods for correcting confounding, adjusting for covariates and over-dispersion, and deriving confidence limits are discussed in a modelling framework. The methods are illustrated using data on measles and pertussis vaccines.

The distribution of time to extinction in subcritical branching processes: applications to outbreaks of infectious disease

We consider the distribution of the number of generations to extinction in subcritical branching processes, with particular emphasis on applications to the spread of infectious diseases. We derive the generation distributions for processes with Bernoulli, geometric and Poisson offspring, and discuss some of their distributional and inferential properties. We present applications to the spread of infection in highly vaccinated populations, outbreaks of enteric fever, and person-to-person transmission of human monkeypox.

Statistical Models for Biosurveillance of Multiple Organisms

Objective

To look at the diversity of the patterns displayed by a range of organisms, and to seek a simple family of models that adequately describes all organisms, rather than a well-fitting model for any particular organism.

Introduction

There has been much research on statistical methods of prospective outbreak detection that are aimed at identifying unusual clusters of one syndrome or disease, and some work on multivariate surveillance methods (1). In England and Wales, automated laboratory surveillance of infectious diseases has been undertaken since the early 1990’s. The statistical methodology of this automated system is described in (2). However, there has been little research on outbreak detection methods that are suited to large, multiple surveillance systems involving thousands of different organisms.

Methods

We obtained twenty years’ data on weekly counts of all infectious disease organisms reported to the UK’s Health Protection Agency. We summarized the mean frequencies, trends and seasonality of each organism using log-linear models. To identify a simple family of models which adequately represents all organisms, the Poisson model, the quasi-Poisson model and the negative binomial model were investigated (3,4). Formal goodness-of-fit tests were not used as they can be unreliable with sparse data. Adequacy of the models was empirically studied using the relationships between the mean, variance and skewness. For this purpose, each data series was first subdivided into 41 half-years and de-seasonalized.

Results

Trends and seasonality were summarized by plotting the distribution of estimated linear trend parameters for 2250 organisms, and modal seasonal period for 2254 organisms, including those organisms for which the seasonal effect is statistically significant.

Relationships between mean and variance were summarized as given in Figure 1.

Similar plots were used to summarize the relationships between mean and skewness.

Conclusions

Statistical outbreak detection models must be able to cope with seasonality and trends. The data analyses suggest that the great majority of organisms can adequately – though far from perfectly – be represented by a statistical model in which the variance is proportional to the mean, such as the quasi-Poisson or negative binomial models.

Measures of disassortativeness and their application to directly transmitted infections

We propose a measure of disassortativeness to summarize contact patterns relevant to the transmission of directly transmitted infections. We discuss the properties of this measure, describe standardization relative to homogeneous mixing, and generalize it to multivariate contact structures. We explore some of its properties and apply our methods to serological surveys of close contact infections and surveys of self-reported social contacts obtained in several European countries.

Epidemiological studies of the non-specific effects of vaccines: II--methodological issues in the design and analysis of cohort studies

We review sources of bias which can affect non-randomized cohort studies of non-specific effects of vaccines on child mortality. Using examples from the literature on non-specific effects, we describe different sources of selection and information bias, and, where possible, outline analysis strategies to mitigate or eliminate such biases.

Hepatitis B vaccination and first central nervous system demyelinating events: reanalysis of a case-control study using the self-controlled case series method

The hypothesis that hepatitis B vaccination is a risk factor for multiple sclerosis has been discussed at length. The data from an earlier case-control study were reanalyzed using the self-controlled case series method. Using the matched cases from the case-control study, we found a relative incidence of 1.68, 95% CI (0.77-3.68) for the 0-60-day post-vaccination risk period; this compares to an odds ratio of 1.8, 95% CI (0.7-4.6). When an additional 53 unmatched cases not used in the case-control study were included, the relative incidence was 1.35, 95% CI (0.66-2.79). Our results throw further light on the methodological aspects of the case series method. We recommend that, when case-control studies of vaccination and adverse events are planned, case series analyses based on the cases are also undertaken when appropriate.

Matrix models for childhood infections: a Bayesian approach with applications to rubella and mumps

Mathematical modelling is an established tool for planning and monitoring vaccination programmes. However, the matrices describing contact rates are based on subjective choices, which have a large impact on results. This paper reviews published models and obtains prior model probabilities based on publication frequency and expert opinion. Using serological survey data on rubella and mumps, Bayesian methods of model choice are applied to select the most plausible models. Estimates of the basic reproduction number R0 are derived, taking into account model uncertainty and individual heterogeneity in contact rates. Twenty-two models are documented, for which publication frequency and expert opinion are negatively correlated. Using the expert prior with individual heterogeneity, R0=6.1 [95% credible region (CR) 4.3-9.2] for rubella and R0=19.3 (95% CR 4.0-31.5) for mumps. The posterior modes are insensitive to the prior for rubella but not for mumps. Overall, assortative models with individual heterogeneity are recommended.

Does concurrent prescription of selective serotonin reuptake inhibitors and non-steroidal anti-inflammatory drugs substantially increase the risk of upper gastrointestinal bleeding?

BACKGROUND

A 15-fold increased risk of gastrointestinal bleeding has been reported with concurrent use of selective serotonin reuptake inhibitors and non-steroidal anti-inflammatory drugs. Recent guidance cautions against concurrent prescription, particularly in older people.

AIM

To quantify the risk of gastrointestinal bleeding associated with current exposure to non-steroidal anti-inflammatory drugs, selective serotonin reuptake inhibitors, and both drugs concurrently.

METHODS

We conducted a case-control analysis of 11,261 cases with upper gastrointestinal bleeding and 53,156 controls matched by gender, age and general practice from computerized primary care data. We coupled this with self-controlled case series analysis.

RESULTS

Both drugs were associated with a twofold increased risk of gastrointestinal bleeding (odds ratio =2.38, 95% confidence interval 2.08-2.72 for selective serotonin reuptake inhibitors and odds ratio = 2.15, 95% confidence interval 2.02-2.28 for non-steroidal anti-inflammatory drugs). This increased risk was marginally higher for concurrent prescription (odds ratio = 2.93, 95% confidence interval 2.25-3.82). The self-controlled analysis showed a greater incidence rate ratio for gastrointestinal bleeding with non-steroidal anti-inflammatory drugs (2.71, 95% confidence interval 2.51-2.91) and lower incidence rate ratio with selective serotonin reuptake inhibitors (1.71, 95% confidence interval 1.48-1.98). The incidence rate ratio when both drugs were combined was 3.25, 95% confidence interval 1.95-5.42. Estimates were similar after restricting to people over 80 years of age. Increased risk of gastrointestinal bleeding was not specifically related to class of non-steroidal anti-inflammatory drugs and was similar when we looked at tricyclic anti-depressants.

CONCLUSIONS

Our study suggests that the risk of gastrointestinal bleeding is not substantially increased when non-steroidal anti-inflammatory drugs and selective serotonin reuptake inhibitors are prescribed together, compared with their use alone.

Parvovirus B19 viraemia in Dutch blood donors

Blood, donated by asymptomatic donors, may contain and transmit parvovirus B19. To investigate the dynamics of parvovirus viraemia in asymptomatic blood donors, we studied the amounts of parvovirus DNA in pools of donor plasma, the prevalence of parvovirus antibodies among blood donors in relation to age, and the seasonal and year-to-year variation of the incidence of parvovirus infection in The Netherlands. The incidence of parvovirus infection follows a seasonal cycle and a cycle of several years. Among Dutch blood donors the incidence was estimated to be 0.56% per year. Forty seven out of 100 pools of 5000 plasma donations tested positive for parvovirus DNA. We inferred that the course of viraemia in asymptomatic donors shows a short peak (> 10(9) copies parvovirus DNA/ml), followed by viraemia below 10(6) copies/ml for about 2 weeks.

Estimation of infectious disease parameters from serological survey data: the impact of regular epidemics

Serological surveys are a useful source of information about epidemiological parameters for infectious diseases. In particular they may be used to estimate contact rates, forces of infection, the reproduction number and the critical vaccination threshold. However, these estimation methods require the assumption that the infection is in endemic equilibrium. Such equilibria seldom exist in practice: for example, many common infections of childhood exhibit regular epidemic cycles. In this paper, we investigate whether ignoring such cycles produces biased estimates. We apply the methods to data on mumps and rubella in the U.K. prior to the introduction of the combined measles, mumps, rubella (MMR) vaccine. We conclude that past epidemics have only a marginal effect on estimates, and that standard methods that do not adjust for regular epidemics are valid.

Infections with Varying Contact Rates: Application to Varicella

We develop methods for the analysis of infectious disease data when age-specific contact rates vary over time. Our methods are valid when contact rates vary slowly on the time scale of the infection process, and are applicable to a variety of data types including serial seroprevalence surveys and case reports. The methods exploit approximate endemic equilibria, and require numerical solution of an associated integral equation in age and time. We also estimate summary statistics such as time-dependent analogs of the basic reproduction number and critical immunization threshold. We illustrate the methods with data on varicella (chickenpox) in the United Kingdom.

Control without separate controls: evaluation of vaccine safety using case-only methods

Epidemiological methods involving only cases are reviewed in the context of vaccine safety studies. These methods include some ecological methods, case-coverage methods, case-crossover and self-controlled case series methods. The properties of the methods are described using examples from the literature. It is argued that such methods, and in particular the self-controlled case series method, are powerful epidemiologic tools meriting the same attention as more traditional cohort and case-control methods.

Estimation of effective reproduction numbers for infectious diseases using serological survey data

The effective reproduction number of an infection, denoted Re, may be used to monitor the impact of a vaccination programme. If Re is maintained below 1, then sustained endemic transmission of the infection cannot occur. In this paper we discuss methods for estimating Re from serological survey data, allowing for age and individual heterogeneity. We describe semi-parametric and parametric models, and obtain an upper bound on Re when vaccine coverage and efficacy are not known. The methods are illustrated using data on mumps and rubella in England and Wales.

On vaccine efficacy and reproduction numbers

We consider the impact of a vaccination programme on the transmission potential of the infection in large populations. We define a measure of vaccine efficacy against transmission which combines the possibly random effect of the vaccine on individual susceptibility and infectiousness. This definition extends some previous work in this area to arbitrarily heterogeneous populations with one level of mixing, but leads us to question the usefulness of the concept of vaccine efficacy against infectiousness. We derive relationships between vaccine efficacy against transmission, vaccine coverage and reproduction numbers, which generalize existing results. In particular we show that the projected reproduction number RV does not depend on the details of the vaccine model, only on its overall effect on transmission. Explicit expressions for RV and the basic reproduction number R0 are obtained in a variety of settings. We define a measure of projected effectiveness of a vaccination programme PE=1-(RV/R0) and investigate its relationship with efficacy against transmission and vaccine coverage. We also study the effective reproduction number Re(t) at time t. Monitoring Re(t) over time is an important aspect of programme surveillance. Programme effectiveness PE is less sensitive than RV or the critical vaccination threshold to model assumptions. On the other hand Re(t) depends on the details of the vaccine model.

Branching process models for surveillance of infectious diseases controlled by mass vaccination

Mass vaccination programmes aim to maintain the effective reproduction number R of an infection below unity. We describe methods for monitoring the value of R using surveillance data. The models are based on branching processes in which R is identified with the offspring mean. We derive unconditional likelihoods for the offspring mean using data on outbreak size and outbreak duration. We also discuss Bayesian methods, implemented by Metropolis-Hastings sampling. We investigate by simulation the validity of the models with respect to depletion of susceptibles and under-ascertainment of cases. The methods are illustrated using surveillance data on measles in the USA.

Deaths from variant Creutzfeldt-Jakob disease in the UK

In 2002, 17 people died from variant CJD (vCJD) in the UK, compared with 20 in 2001 and 28 in 2000. We analysed data for deaths from vCJD since 1995 and estimated the underlying trend in mortality. The trend had a quadratic component (p=0.005), suggesting that the increase was not exponential, and that the previously increasing trend is slowing down. The death rate peaked in 2000. These findings are encouraging, but mortality might increase again in the future.

Lack of association between intussusception and oral polio vaccine in Cuban children

Two exploratory investigations found an increased risk of intussusception after oral polio vaccine (OPV). A large, national, population-based study was undertaken in Cuba to investigate a possible association. Three hundred and thirty-five cases of intussusception in children under 2 years of age occurring in 1995-2000 were identified and their OPV records retrieved. The relative incidence (RI) of intussusception in defined periods up to 42 days after OPV in children under 1 year was estimated using the self-controlled case series method, controlling for age and season. The RI was not significantly raised in any of the time intervals examined within the 0-42 day period after OPV. For the period 0-42 days as a whole the RI was 1.11, 95% CI 0.74-1.67. This study does not support the hypothesis that OPV causes intussusception.

Interval estimation for Poisson capture-recapture models in epidemiology

Capture-recapture studies in epidemiology are frequently undertaken to adjust underlying rates for underreporting. This contrasts with applications in ecology, in which the aim is to estimate population size. In consequence, the Poisson model provides a more appropriate framework than the multinomial in some epidemiological applications. A method for constructing profile Poisson likelihood interval estimates using log-linear modelling, avoiding the need for constrained optimization, is described for several relevant parameters, including the unobserved disease rate, the total disease rate, and the case ascertainment probability. Asymptotic properties of these estimates are studied and contrasted with those derived from the multinomial model. The method is illustrated with a published data set on spina bifida.

Changes in age related seroprevalence of antibody to varicella zoster virus: impact on vaccine strategy

AIM

To study changes in the seroprevalence of varicella zoster virus (VZV) antibody over the past 25 years with a view to determining the target age group for any future vaccination strategy.

METHODS

Stored sera collected from different age groups over a period of 25 years were tested by a commercial VZV IgG enzyme immunoassay at a four year time interval. Data were analysed by logistic regression to investigate the evidence for changes in incidence and hence seroprevalence over that period.

RESULTS

There was a significant rise in VZV antibody prevalence in the 1-4 year age group during the study period.

CONCLUSIONS

A universal childhood VZV vaccination strategy will need to take account of the increase in incidence of VZV infection in children under the age of 4 years; hence, the suggested target age would be between 12 and 18 months---soon after the disappearance of maternal antibody.

Within-subject exposure dependency in case-crossover studies

In the case-crossover design, only cases are sampled and the hazard ratio is estimated from within-subject comparisons of exposures at the event time and in M control periods prior to the event. We consider the effect of within-subject dependence of exposures in successive time periods. We show that estimates obtained from the conditional logistic model are biased. This bias disappears if the distribution of exposures in the M+1 successive time intervals is exchangeable. In contrast, the Mantel-Haenszel estimator for the odds ratio is approximately unbiased provided that exposures are stationary. Suitable methods of analysis of case-crossover designs using maximum likelihood may be derived from cohort rather than case-control models.

MMR and autism: further evidence against a causal association

The hypothesis that MMR vaccines cause autism was first raised by reports of cases in which developmental regression occurred soon after MMR vaccination. A previous study found no evidence to support this hypothesis. It has recently been suggested that MMR vaccine might cause autism, but that the induction interval need not be short. The data from the earlier study were reanalysed to test this second hypothesis. Our results do not support this hypothesis, and provide further evidence against a causal association between MMR vaccination and autism.

Idiopathic thrombocytopenic purpura and MMR vaccine

A CAUSAL ASSOCIATION BETWEEN MEASLES: mumps-rubella (MMR) vaccine and idiopathic thrombocytopenic purpura (ITP) was confirmed using immunisation/hospital admission record linkage. The absolute risk within six weeks of immunisation was 1 in 22 300 doses, with two of every three cases occurring in the six week post-immunisation period being caused by MMR. Children with ITP before MMR had no vaccine associated recurrences.

Vaccine trials

This article reviews some of the issues involved in evaluating vaccines in humans. Vaccine trials are required for licensure and are essential for demonstrating a vaccine's safety and protective efficacy. The formal framework of phase I, II, and III trials is described, with particular emphasis on the choice of hypotheses, trial design, and biases that arise in the context of vaccine trials. However, some aspects of a vaccine's performance cannot be evaluated in clinical trials owing to their relatively small size. Thus, vaccine evaluation must continue after licensure, for example, to evaluate the vaccine with respect to rare reactions, duration of protection, and ecological effects. The article reviews some of the methods commonly used for post-licensure studies of vaccine efficacy and safety.

Incidence of variant Creutzfeldt-Jakob disease in the UK

The number of deaths from variant CJD (vCJD) in the UK increased in the last quarter of 1998, although numbers were lower in subsequent quarters. We analysed the numbers of definite and probable (living and dead) vCJD cases since 1994 to assess trends in incidence. We estimated that the number of onsets increased by 23% per year for 1994-2000 (p=0.004), and that deaths increased by 33% for 1995-2000 (p=0.005). The absolute number of cases in the UK is still low, but such an increase should be a matter of concern.

Molecular epidemiology of a large outbreak of hepatitis B linked to autohaemotherapy

BACKGROUND

Unregulated skin-piercing procedures potentially facilitate the transmission of bloodborne pathogens. In February, 1998, a patient who had recently received autohaemotherapy at an alternative medicine clinic in the UK was diagnosed with acute hepatitis B. The autohaemotherapy procedure involved the drawing of 1 mL of the patient's blood, mixing with saline, and reinjection of the autologous blood mixture. We investigated the extent of hepatitis B virus (HBV) infection in patients and staff of the clinic.

METHODS

Patients who had attended the clinic between January, 1997, and February, 1998, were tested for serological markers of HBV, and for HBV DNA by PCR. HBV DNA was sequenced to assess the relatedness of the virus identified in the cases. We analysed the number and dates of visits with regard to HBV status.

FINDINGS

Serum samples were received from 352 patients and four staff members. Serological evidence of exposure to HBV was found in 57 (16%). Of the 33 patients and staff who were positive for hepatitis B surface antigen, 30 (91%) showed complete nucleotide identity in the DNA segments derived from the surface and core genes. Five patients with linked infection had markers of chronic hepatitis B, and one of these was regarded as the likely source of the outbreak. The attack rate was associated with the number of visits (p<0.0001) and the week of visit (p=0.011). Contaminated saline in a repeatedly used bottle was the probable vehicle of transmission.

INTERPRETATION

We have described a large community-based outbreak of hepatitis B due to transmission by a single HBV variant. Our findings emphasise the continuing risk of transmission of bloodborne viruses in all health-care settings where skin-piercing procedures are used.

Residuals for proportional hazards models with interval-censored survival data

We develop diagnostic tools for use with proportional hazards models for interval-censored survival data. We propose counterparts to the Cox-Snell, Lagakos (or martingale), deviance, and Schoenfeld residuals. Many of the properties of these residuals carry over to the interval-censored case. In particular, the interval-censored versions of the Lagakos and Schoenfeld residuals may be derived as components of suitable score statistics. The Lagakos residuals may be used to check regression relationships, while the Schoenfeld residuals can help to detect nonproportional hazards in semiparametric models. The methods apply to parametric models and to the semiparametric model with discrete observation times.

Epidemiology of transmissible diseases after elimination

Elimination of an infectious disease is often understood to mean the total absence of cases in a population. This situation can occur only if the entire population is immune as a result of either natural disease or vaccination. However, this costly and unrealistic scenario is not necessary to ensure elimination, more appropriately defined as a situation in which sustained transmission cannot occur and secondary spread from importations of disease will end naturally, without intervention. The authors describe the size and duration of outbreaks caused by imported infections after indigenous transmission has been eliminated. They show that the status of the elimination process can be monitored by assessing the proportion of cases imported and the distribution of outbreak sizes. Measles in Canada, the United States, and the United Kingdom provides a good example of the relevance of these criteria. Surveillance of the size and duration of these outbreaks enables maintenance of elimination to be monitored.

Outbreak of aseptic meningitis associated with mass vaccination with a urabe-containing measles-mumps-rubella vaccine: implications for immunization programs

A mass immunization campaign with a Urabe-containing measles-mumps-rubella vaccine was carried out in 1997 in the city of Salvador, northeastern Brazil, with a target population of children aged 1-11 years. There was an outbreak of aseptic meningitis following the mass campaign. Cases of aseptic meningitis were ascertained through data collected from the records of children admitted to the local referral hospital for infectious diseases between March and October of 1997, using previously defined eligibility criteria. Vaccination histories were obtained through home visits or telephone calls. Eighty-seven cases fulfilled the study criteria. Of those, 58 cases were diagnosed after the vaccination campaign. An elevated risk of aseptic meningitis was observed 3 weeks after Brazil's national vaccination day compared with the risk in the prevaccination period (relative risk = 14.3; 95% confidence interval: 7.9, 25.7). This result was confirmed by a case series analysis (relative risk = 30.4; 95% confidence interval: 11.5, 80.8). The estimated risk of aseptic meningitis was 1 in 14,000 doses. This study confirms a link between measles-mumps-rubella vaccination and aseptic meningitis. The authors discuss the implications of this for the organization and planning of mass immunization campaigns.

Ecological analysis of ethnic differences in relation between tuberculosis and poverty

OBJECTIVE

To examine the effect of ethnicity on the relation between tuberculosis and deprivation.

DESIGN

Retrospective ecological study comparing incidence of tuberculosis in white and south Asian residents of the 39 electoral wards in Birmingham with ethnic specific indices of deprivation.

SETTING

Birmingham, 1989-93.

SUBJECTS

1516 notified cases of tuberculosis.

MAIN OUTCOME MEASURES

Rates of tuberculosis and measures of deprivation.

RESULTS

Univariate analysis showed significant associations of tuberculosis rates for the whole population with several indices of deprivation (P<0.01) and with the proportion of the population of south Asian origin (P<0.01). All deprivation covariates were positively associated with each other but on multiple regression, higher level of overcrowding was independently associated with tuberculosis rates. For the white population, overcrowding was associated with tuberculosis rates independently of other variables (P=0.0036). No relation with deprivation was found for the south Asian population in either single or multivariable analyses.

CONCLUSIONS

Poverty is significantly associated with tuberculosis in the white population, but no such relation exists for those of Asian ethnicity. These findings suggest that causal factors, and therefore potential interventions, will also differ by ethnic group.

Autism and measles, mumps, and rubella vaccine: no epidemiological evidence for a causal association

BACKGROUND

We undertook an epidemiological study to investigate whether measles, mumps, and rubella (MMR) vaccine may be causally associated with autism.

METHODS

Children with autism born since 1979 were identified from special needs/disability registers and special schools in eight North Thames health districts, UK. Information from clinical records was linked to immunisation data held on the child health computing system. We looked for evidence of a change in trend in incidence or age at diagnosis associated with the introduction of MMR vaccination to the UK in 1988. Clustering of onsets within defined postvaccination periods was investigated by the case-series method.

FINDINGS

We identified 498 cases of autism (261 of core autism, 166 of atypical autism, and 71 of Asperger's syndrome). In 293 cases the diagnosis could be confirmed by the criteria of the International Classification of Diseases, tenth revision (ICD10: 214 [82%] core autism, 52 [31%] atypical autism, 27 [38%] Asperger's syndrome). There was a steady increase in cases by year of birth with no sudden "step-up" or change in the trend line after the introduction of MMR vaccination. There was no difference in age at diagnosis between the cases vaccinated before or after 18 months of age and those never vaccinated. There was no temporal association between onset of autism within 1 or 2 years after vaccination with MMR (relative incidence compared with control period 0.94 [95% CI 0.60-1.47] and 1.09 [0.79-1.52]). Developmental regression was not clustered in the months after vaccination (relative incidence within 2 months and 4 months after MMR vaccination 0.92 [0.38-2.21] and 1.00 [0.52-1.95]). No significant temporal clustering for age at onset of parental concern was seen for cases of core autism or atypical autism with the exception of a single interval within 6 months of MMR vaccination. This appeared to be an artifact related to the difficulty of defining precisely the onset of symptoms in this disorder.

INTERPRETATION

Our analyses do not support a causal association between MMR vaccine and autism. If such an association occurs, it is so rare that it could not be identified in this large regional sample.

Interval-censored survival data with informative examination times: parametric models and approximate inference

We develop parametric methods for analysing interval-censored data when examination and survival times are not independent. The hazard function is modelled by introducing individual frailties related to the frequency of examinations. Model parameters may be obtained by direct maximization of the marginal log-likelihood. We develop a simpler approximate method in which the frailties are estimated by empirical Bayes. The two approaches are equivalent asymptotically as the number of examinations on each individual increases. Simulations suggest that the approximate method is adequate for estimating regression parameters even when the number of examinations on each individual is small. The methods are used to estimate age and period effects on HIV incidence in a cohort of repeat attenders at genito-urinary clinics in London, U.K.

The prevalence of hepatitis B infection in adults in England and Wales

Cost effectiveness analyses of alternative hepatitis B vaccination programmes in England and Wales require a robust estimate of the lifetime risk of carriage. To this end, we report the prevalence of infection in 3781 anonymized individuals aged 15-44 years whose sera were submitted in 1996 to 16 microbiology laboratories in England and Wales. One hundred and forty-six individuals (3.9%) were confirmed as anti HBc positive, including 14 chronic carriers (0.37%). The prevalence of infection and carriage was higher in samples collected in London and increased with age. No increased risk of infection was seen in sera from genito-urinary (GUM) clinics. Only 15 sera positive for hepatitis B were also positive for hepatitis C. Our results confirm the low prevalence of hepatitis B in England and Wales, are consistent with previous estimates of carriage and suggest that many infections were acquired while resident outside the UK. Future prevalence studies should determine the country of birth and other risk factors for each individual in order to confirm these findings.

Pre-AIDS mortality in HIV-infected individuals in England, Wales and Northern Ireland, 1982-1996

OBJECTIVE

To assess pre-AIDS mortality in HIV-infected patients in England, Wales and Northern Ireland during the period 1982-1996.

DESIGN

Surveillance data on pre-AIDS and AIDS deaths reported to the PHLS-AIDS Centre were analysed.

METHODS

Pre-AIDS mortality was estimated as the proportion of pre-AIDS deaths among all deaths in HIV-infected people. Trends over time in the number of pre-AIDS and AIDS deaths were compared using Poisson regression with logarithmic link. Causes of pre-AIDS deaths were recorded. Comparisons were made between the pre-AIDS and the AIDS death groups by logistic regression for: age, location of residence at death, year of death and risk exposure.

RESULTS

Four-hundred and sixty-eight pre-AIDS deaths and 8574 AIDS deaths were identified. Pre-AIDS mortality accounted for 5.0% of HIV-related deaths. Trends over time in the number of pre-AIDS and AIDS deaths were not significantly different (P=0.11). Reported causes of pre-AIDS death included pneumonia (92), liver disease (62), septicaemia (51), malignancies (49), suicide (45), cardiopulmonary causes (46), haemorrhage (42), overdose (24) and accidental causes (24). Factors positively associated with pre-AIDS death were injecting drug use, haemophilia and blood transfusion, residence outside the Thames regions, and death at an older age.

CONCLUSIONS

Pre-AIDS mortality represents a substantial proportion of HIV mortality, particularly where injecting drug use is a frequent route of HIV transmission.

Hospital admissions attributable to rotavirus infection in England and Wales

Laboratory reports and data on hospital admissions were used to estimate the number of hospitalizations due to group A rotavirus infection in England and Wales. Between January 1990 and December 1994, there were 75,059 laboratory reports of rotavirus infection, and 66,062 of these were in children <5 years old; rotavirus represented 39% of all pathogens identified in fecal specimens from this age group. Between April 1993 and March 1994, 1904 hospital admissions coded as "infectious intestinal disease" and 2354 coded as "noninfective gastroenteritis" occurred in children <5 in the North Thames region (a health authority representing 13% of the population in England and Wales). By modeling admission and laboratory reporting data, it was estimated that 54% of hospitalizations for intestinal infectious disease and 34% for noninfective gastroenteritis were attributable to rotavirus. By extrapolation of the North Thames data, it was estimated that 17,810 rotavirus-related hospitalizations (5/1000 children <5 years old) occurred in England and Wales during the same period. Effective vaccines have the potential to substantially reduce the number of hospital admissions due to group A rotavirus infection.

Case series analysis of adverse reactions to vaccines: a comparative evaluation

A modified cohort method has been proposed for estimating the relative incidence of rare adverse reactions after vaccination. The method requires only a sample of the cases, thus avoiding the need for following large population cohorts or selecting controls. This case series method has statistical power equivalent to that of the full cohort method when the risk periods after vaccination are short and vaccine coverage is high. The method also eliminates confounding by variables associated with both the outcome and avoidance of vaccination. In this paper, the cohort, case-control, and case series methods are reviewed, and their underlying assumptions and performances are compared. Theoretical results are illustrated using data on febrile convulsions after measles-mumps-rubella vaccination in the United Kingdom.

Interval censored survival data: a generalized linear modelling approach

A method is described for weak parametric modelling of arbitrarily interval censored survival data using generalized linear models. The method makes use of an associated Bernoulli model, with standard errors based on the observed information matrix. Three types of models are discussed: additive and multiplicative hazard models with piecewise constant baseline hazard, and a proportional hazards model with discrete baseline survivor function. These models may be fitted in the statistical package GLIM.

Feasibility study for identifying adverse events attributable to vaccination by record linkage

To investigate the feasibility of using a record linkage method for identifying vaccine attributable adverse events, computerized hospital admissions and vaccination records from South East Kent district were linked and checked for accuracy. Records for 90% of children under 2 years of age admitted to hospital over a 2-year period were matched with vaccination records using a computer algorithm based on name, date of birth, sex, and post-code supplemented by visual inspection. Relative to this gold standard, matching on date of birth, sex and postcode alone had a sensitivity of 60% and an incorrect match rate of 0.2% after matches to more than one vaccine recipient were excluded. Manual checking of a sample of admissions showed that only 4% had been assigned incorrect International Classification of Disease (ICD) codes. Routine record linkage of ICD admission codes to vaccination records therefore yields data of good quality which may be used for surveillance purposes.

Relative incidence estimation from case series for vaccine safety evaluation

A method is described for estimating the relative incidence of clinical events in defined time intervals after vaccination compared to a control period using only data on cases. The method is derived from a Poisson cohort model by conditioning on the occurrence of an event and on vaccination histories. Methods of analysis for event-dependent vaccination histories and survival data are discussed. Asymptotic arguments suggest that the method retains high efficiency relative to the full cohort analysis under conditions which commonly apply to studies of vaccine safety.