Search

Varicella zoster virus (VZV) is a highly infectious member of the alpha-herpesvirus family that is specific to humans presenting initially as varicella (chickenpox, CP) upon primary infection. It is so infectious that in an unvaccinated population the annual incidence of CP is considered roughly equivalent to the populations birth cohort. While CP is generally considered a relatively benign childhood disease resulting in a characteristic rash accompanied by fever, infection can result in hospitalisation and even death due to secondary infections such as pneumonia. Prior to VZV vaccination, it was estimated that CP infection resulted in 1,500 hospitalisations and five deaths annually in Australia. While cases only have CP once, in all cases that recover there is the lifetime risk of further disease as the VZV virus is not eliminated but rather enters a latency period residing in the hosts dorsal root (spinal) ganglia with the ability to re-emerge as herpes zoster (HZ or shingles) usually associated with aging or immunocompromise. The burden of disease with HZ is considered to be greater than that of CP with the risk of repeated infection and a high proportion of HZ cases suffering from complications, most notably post-herpetic neuralgia (PHN). PHN is an enduring dermatomal pain that is estimated to occur in over 60% of HZ cases and can last over one year in about 5% of cases. It has been estimated that prior to VZV vaccination, HZ re-emergence resulted in 10 deaths per year and over 3,000 hospitalisations annually in Australia. To combat this disease the Australian Nation Immunisation Program currently recommends and funds the provision of a single dose of VZV vaccine in combination with the measles, mumps and rubella vaccine (i.e., MMRV) at 18 months of age, as well as a live attenuated HZ vaccine for person aged 70 to 79 years. However, several Australian studies have examined the effectiveness of VZV vaccination on CP and HZ and concluded that a two-dose VZV regimen may be more effective. Therefore, primary purpose of this study was to determine what is the most effective (and ideally cost-effective) vaccination strategy for reducing VZV disease (i.e., both CP and HZ) in the Australian population. [...]

In chapter 1, using a Galton-Watson branching-process analysis, we characterize differences in measles transmission by estimating the association between genotype and the reproduction number among post-elimination California measles cases. Genotype B3 is found to be a significant predictor of transmissibility.In chapter 2, we determine whether data from FDA-cleared wired and Bluetooth smart thermometers sold by a San Francisco-based company aid influenza forecasting efforts. We compare this smart thermometer data to regional influenza and ILI surveilance data from the California Department of Public Health. We evaluated the correlation between the regional California surveillance data and smart thermometer data, tested the hypothesis that smart thermometer readings and symptom reports provide regionally specific predictions, and determined whether smart thermometer and mobile application improved disease forecasts. Our results are consistent with the hypothesis that smart thermometer readings and symptom reports reflect underlying disease transmission in California. Data from such cloud-based devices could supplement syndromic influenza surveillance data.In chapter 3, we examine whether changes in varicella transmissibility may have occurred following the change from one- to two-dose vaccination scheduling in 2007. Following the change in ACIP recommendations for varicella vaccination in 2007, the median outbreak size decreased. However, while the number of outbreaks has continued to decrease following 2008, we do not find evidence that the distribution of sizes has changed since 2008. Using insights from branching process models assuming both subcritical and supercritical transmission and with and without depletion of susceptibles, we cannot rule out that varicella transmission is supercritical in school-based settings.

Varicella Zoster Virus (VZV) firstly manifests as varicella infection early in life, and then it may reactivate in older age causing herpes zoster (HZ). It is believed that re-exposure to varicella infection boosts against reactivation (exogenous boosting dynamic) delaying or suppressing HZ disease. In this study, we examine the effect that demographic changes could have had on boosting, by reducing VZV re-exposure, and consequently on VZV epidemiological trends in Australia over the 20th century. We developed a dynamic age-structured mathematical model informed with historical Australian birth and death rates to reproduce varicella and HZ incidence over time. Furthermore we analysed how different assumptions on duration of boosting and HZ recurrence affect the HZ incidence trends. The model is calibrated against age-specific varicella seroprevalence data collected between 1997 and 1999 and varicella and HZ hospitalisation data from AIHW collected between 1993 and 2009. We show that the reduction of varicella incidence follows the overall decline in birth rate. In the period 1950-2000, this reduction in varicella circulation, together with the aging population, produced a 19.1% increase in HZ crude incidence. However, following age-standardisation, HZ trends showed an increase of 3.3% over the same period. We found that demographic changes may be partially responsible for the rise in HZ incidence observed prior to and at the beginning of the varicella immunisation program. In order to explain and project more realistic disease incidence trends, future mathematical models should account for demographic changes and present age-standardised results.

The varicella-zoster virus (VZV) causes chickenpox or varicella, a disease primarily in children, and Herpes Zoster (HZ) or zoster or shingles, a disease that affects adults. A 2-dose Varicella vaccination is recommended in the United States, the first dose at age 15-18 months and the second dose at 4 to 6 years. In this study, we used multinomial logistic regression to analysis data from the 2011 National Immunization Survey-Teen (NIS-Teen) to identify factors that have a significant impact on the number of doses (0-dose, 1-dose, or 2-dose) a teen will have. We evaluate Varicella vaccination coverage stratified by Census region and assessed factors independently associated with varicella vaccination.

Abstract: Objectives: This thesis consists of 3 partial vaccine economic evaluations. The objectives were to: 1) estimate the effectiveness of Canada’s universal varicella childhood immunization strategy on varicella-related hospitalizations, 2) measure the economic impact of Alberta’s universal childhood pneumococcal immunization program (Prevnar 7), and 3) generate a general model estimating the cost of vaccine research, development, and deployment (RDD). Methods: For the first objective, rate ratios were calculated based on pre- and post- cases hospitalization across 10 provinces, and 6 age categories. For the second objective, an ex post cost analysis was conducted, comparing the costs associated with pneumococcal both before and after the introduction of universal Prevnar 7 immunization. The third objective was met by modifying cost models from the drug development literature. Results: For the first objective, there was evidence of significant declines in varicella across all provinces in Canada, across most age categories less than 40, suggesting some evidence of herd immunity effect. For the second objective, the costs averted in Alberta as a result of decreased cases of pneumonia were $9.2 million and $1.8 million if serotype replacement is taken into account. For the third objective, Rotarix was used as a case example to demonstrate the vaccine costing model. It is found that it cost an estimated $2.7 billion to develop Rotarix vaccine, and $3.7 billion after consideration of risk premiums. Conclusions: For objective one, the Canadian immunization strategy has been successful in reducing varicella-related hospitalizations across the country, and across ages. For the second objective, costs have been averted in Alberta as a result of decreased cases of pneumonia due to PCV7. However, the relationship between the vaccine and serotype replacement could erode some of these benefits. For the third objective, a general model has been created that can estimate the cost of vaccine RDD. This model can be used to estimate cost of RDD for any vaccine in which the cost of clinical testing is known.

Many infectious diseases are spread through interactions between susceptible and infectious individuals. Keeping track of where each exposure to the disease took place, when it took place, and which individuals were involved in the exposure can give public health officials important information that they may use to formulate their interventions. Further, knowing which individuals in the population are at the highest risk of becoming infected with the disease may prove to be a useful tool for public health officials trying to curtail the spread of the disease. Epidemiological models are needed to allow epidemiologists to study the population dynamics of transmission of infectious agents and the potential impact of infectious disease control programs. While many agent-based computational epidemiological models exist in the literature, they focus on the spread of disease rather than exposure risk. These models are designed to simulate very large populations, representing individuals as agents, and using random experiments and probabilities in an attempt to more realistically guide the course of the modeled disease outbreak. The work presented in this thesis focuses on tracking exposure risk to chickenpox in an elementary school setting. This setting is chosen due to the high level of detailed information realistically available to school administrators regarding individuals' schedules and movements. Using an agent-based approach, contacts between individuals are tracked and analyzed with respect to both individuals and locations. The results are then analyzed using a combination of tools from computer science and geographic information science.