Developing And Applying Biologically Plausible Statistical Models For Normal And Non-normal Family Data
Funder
National Health and Medical Research Council
Funding Amount
$339,700.00
Summary
Although molecular and computing advances have enabled more detailed investigations of inherited diseases and the ability to fit realistic statistical models to these data, limitations still exist when analysing family data. Often only basic statistical analyses are performed, due to the lack of understanding of complexities within the data and-or inability of researchers to fit appropriate statistical models. These factors have hampered the search for genes and environmental factors influencing ....Although molecular and computing advances have enabled more detailed investigations of inherited diseases and the ability to fit realistic statistical models to these data, limitations still exist when analysing family data. Often only basic statistical analyses are performed, due to the lack of understanding of complexities within the data and-or inability of researchers to fit appropriate statistical models. These factors have hampered the search for genes and environmental factors influencing common diseases. This project aims to develop novel, biologically realistic statistical models for investigation of common, complex diseases, such as heart disease and cancer, in families. These models will incorporate both measured and unmeasured genetic and environmental factors, and will be applicable to both normally distributed and non-normally distributed traits. Model fitting will use computer-intensive simulation techniques. Application of the models to data from two large pre-existing studies of international renown, the Victorian Family Heart Study and the Australian Prostate Cancer Family Study, will enable a better understanding of the genetic and environmental factors influencing heart disease and cancer. The models will also be applicable to many other studies of diseases which use data from families, and allow more accurate and useful information to be obtained from data. Software will also be made freely available to other researchers. This will ultimately translate into better outcomes from familial genetic research, and eventually, better prevention, detection, and treatment of the diseases.Read moreRead less
Statistical Analyses Of Breast Cancer Risks For Australian BRCA1 And BRCA2 Mutation Carriers
Funder
National Health and Medical Research Council
Funding Amount
$424,628.00
Summary
About 10 years ago two genes, called BRCA1 and BRCA2, were discovered. The normal function of these genes is to prevent breast and other cancers from developing. All people have two copies of each gene, one inherited from their mother and one from their father. Women who have inherited a fault in one copy are at increased risk of breast and ovarian cancer. There has been considerable controversy about what their actual cancer risks are, especially about how those risks might depend on their age. ....About 10 years ago two genes, called BRCA1 and BRCA2, were discovered. The normal function of these genes is to prevent breast and other cancers from developing. All people have two copies of each gene, one inherited from their mother and one from their father. Women who have inherited a fault in one copy are at increased risk of breast and ovarian cancer. There has been considerable controversy about what their actual cancer risks are, especially about how those risks might depend on their age. We have already conducted studies on this and have developed the necessary statistical methods to address these issues by analysing data from the families in which there are faulty genes. In this study we propose to use two large Australian studies, one of families with multiple-cases of breast cancer (Kathleen Cuningham Consortium for Research on Familial Breast Cancer; kConFab) and the other of the families of women with breast cancer chosen, irrespective of their family cancer histories, through the Victorian and NSW Cancer Registries (Australian Breast Cancer Family Study; ABCFS). A large amount of work has already been conducted to identify these families and test them for faults in BRCA1 and BRCA2. There are over 350 families who carry faults, making this one of the largest studies of its type in the world. We will check the cancer histories of these families and determine which members have, or are likely to have, inherited a faulty gene. We will then estimate the breast and ovarian cancer risks accurately, and with much more precision, than has been done previously. We will also use these large datasets to develop a simple method to identify which Australian women are most likely to carry a fault in BRCA1 or BRCA2, based on their personal and family cancer histories. This study will assist genetic counsellors inform Australian women who consider mutation testing for BRCA1 and BRCA2 about their cancer risks, and help make breast cancer genetics more cost effective.Read moreRead less
Novel Statistical Methods For The Analysis Of Meausred Genetic And Environmental Risk Factors In Twin Studies
Funder
National Health and Medical Research Council
Funding Amount
$478,314.00
Summary
Studies on twins are an important way to determine whether the risk of disease is likely to be influenced by genetic factors but have traditionally focussed on unmeasured factors. New epidemiological studies measure thousands of genetic variants on many participants. This project will extend methods for analysing data within and between twin pairs to determine whether risk factors are likely to be causal and therefore should be the subject of further designed studies based on intervention.
Analysing Genetic And Environmental Risk Factors And Their Interactions For Common Cancers And Cardiovascular Disease
Funder
National Health and Medical Research Council
Funding Amount
$129,937.00
Summary
The statistical models for analysing cancer and cardiovascular risk factor family data are important for understanding the genetic and environmental aetiology of these diseases, but complicated by the different levels of correlations between relatives in a family. The conventional assumption of independence in observations is invalid in these situations. We intend to develop, test, implement and distribute a comprehensive suite of new statistical methods designed specifically to assistant molecu ....The statistical models for analysing cancer and cardiovascular risk factor family data are important for understanding the genetic and environmental aetiology of these diseases, but complicated by the different levels of correlations between relatives in a family. The conventional assumption of independence in observations is invalid in these situations. We intend to develop, test, implement and distribute a comprehensive suite of new statistical methods designed specifically to assistant molecular geneticists and genetic epidemiologists undertake informative and meaningful analyses of the measured and latent genetic and environmental risk factors and their possible interactions. The two associate investigators, Prof John Hopper and Prof Stephen Harrap, will bring their respective genetic epidemiological and biometric statistical expertise and their prestigious family data resources to this project. With the suite of flexible statistical models and analyses, we will further our knowledge about genetic and environmental risk factors and their interactions of common cancers and major gene effects for cardiovascular phenotypes. Simulation studies will help us understand some phenomena accounted in the research but cannot be replicated in reality and assess the efficiency of the statistical methods and credibility of our analysis results independently. Statistical programs developed in this project can also be used in other genetic and epidemiological studies (e.g. diabetes, epilepsy) where such high-level statistical tools are not yet available.Read moreRead less
Strategies For Handling Missing Data In The Development, Validation And Implementation Of Clinical Risk Prediction Tools
Funder
National Health and Medical Research Council
Funding Amount
$451,692.00
Summary
Tools that predict the future outcome of disease are common. Missing data is a problem in studies that develop and validate such tools and affects their validity because simple approaches to dealing with missing data are biased. We will develop statistical methodology in this area and compare the performance of this and other methodologies. Alongside this methodological work we will re-assess existing prediction tools and develop new tools in the areas of cardiac surgery and kidney disease.
Analysis Of HIV Virologic Response-rebound Data: Prognostic Indicators Of Post-HAART Viral Control
Funder
National Health and Medical Research Council
Funding Amount
$144,000.00
Summary
The introduction of potent anti-retroviral therapy into standard clinical management of HIV infected individuals has been associated with high rates of reduction in plasma viral loads over short periods of time. However, there remains considerable variation in the degree of longer-term viral control as a result of viral resistance, toxicity, timing of treatment initiation and choice of drug regimen. In particular, the most appropriate time to initiate treatment remains clouded, with the need to ....The introduction of potent anti-retroviral therapy into standard clinical management of HIV infected individuals has been associated with high rates of reduction in plasma viral loads over short periods of time. However, there remains considerable variation in the degree of longer-term viral control as a result of viral resistance, toxicity, timing of treatment initiation and choice of drug regimen. In particular, the most appropriate time to initiate treatment remains clouded, with the need to initiate treatment sufficiently early in order to avoid irreversible damage balanced by the problems of potential viral resistance or toxicity if started too soon. Determination of factors which will assist practitioners to optimise the timing of treatment initiation remains a high priority. Our aim in this project is to develop and study the use of novel statistical mixed-effects models designed to analyse factors associated with visit-time viral load data following commencement of therapy, taking account of the entire follow-up profiles of responses over time. The project involves both theoretical and empirical analyses of the estimation and inferential properties of the mixed-model method in conjunction with comprehensive analyses of prognostic factors associated with post-treatment virologic control in patients from the Western Australian HIV Cohort Study. These include demographic, virologic, immunologic, adherence and host genetic factors. The statistical methods developed will have wide applicability and add significantly to the suite of procedures available for the analysis of longitudinal response data.Read moreRead less
Statistical Methods To Assist The Control Of Communicable Diseases, And Their Application
Funder
National Health and Medical Research Council
Funding Amount
$211,527.00
Summary
A range of programs, such as vaccination schedules and intervention in outbreaks, is in place to reduce our burden of illness from infectious diseases. This project aims to develop new methods, based on models and associated statistical analyses, to help ensure that our control programs are based on the best available evidence. There are five specific themes: 1. Developed methods for predicting major outbreaks of diseases and appy them to Australian data on measles, pertussis and Ross River viru ....A range of programs, such as vaccination schedules and intervention in outbreaks, is in place to reduce our burden of illness from infectious diseases. This project aims to develop new methods, based on models and associated statistical analyses, to help ensure that our control programs are based on the best available evidence. There are five specific themes: 1. Developed methods for predicting major outbreaks of diseases and appy them to Australian data on measles, pertussis and Ross River virus. The new methods will enable timely intervention to reduce the population risk from these diseases, and to guide the proposed elimination of measles from Australia. 2. The current concept of vaccine efficacy, which is central to the epidemiological assessment of vaccines, performs poorly when the vaccine provides only partial protection, when immunity wanes over time and when infected vaccinees have a lower infectivity. A concept of vaccine efficacy will be developed that overcomes these weaknesses. This concept will find widespread use in epidemiology. 3. Data available from 80 outbreak investigations of transmissible diseases will be analysed to assess the relative effectiveness of various intervention steps. 4. A comprehensive model, and associated computer software, will be developed as a tool for assessing the relative merits of different vaccination schedules. It will be applied to address Australian policy issues for vaccination against pertussis, chickenpox and rubella. 5. The method of backprojection for estimating trends in HIV infection and predicting AIDS incidence will be enhanced to incorporate newly available data in which HIV and AIDS diagnoses are linked. This will enable a precise assessment of infection incidence among heterosexuals and other exposure categories with relatively few cases, to ensure that HIV does not spread more widely.Read moreRead less
Design And Analysis Of Interrupted Time Series Studies In Health Care Research: Resolution Of Methodological Issues
Funder
National Health and Medical Research Council
Funding Amount
$307,125.00
Summary
An interrupted time series (ITS) study involves a population observed on multiple occasions before and after the implementation of an intervention program. However, methods for statistical analysis and designing such studies have not been well developed and many statistical analyses of such studies are flawed. This proposal will investigate appropriate methods for design and analysis, and develop guidelines and software for its implementation by health researchers.
Diagnostics For Mixture Regression Models: Applications To Public Health
Funder
National Health and Medical Research Council
Funding Amount
$128,250.00
Summary
In many public health studies, finite mixture regression models are often used to analyse data arising from heterogeneous populations. It is important to assess the stability of parameter estimates and the validity of statistical inferences when the underlying assumptions appear to be violated, but appropriate diagnostics are lacking in the literature. This research aims to develop effective diagnostic methods for assessing the adequacy of mixture regression models and the sensitivity of accompa ....In many public health studies, finite mixture regression models are often used to analyse data arising from heterogeneous populations. It is important to assess the stability of parameter estimates and the validity of statistical inferences when the underlying assumptions appear to be violated, but appropriate diagnostics are lacking in the literature. This research aims to develop effective diagnostic methods for assessing the adequacy of mixture regression models and the sensitivity of accompanying test statistics. The methodology developed will enable health care professionals to focus on substantive issues and to draw accurate and valid conclusions inferred from correlated and over-dispersed outcomes. In the presence of anomalous observations, the influence diagnostics can provide insights into the source of heterogeneity and the apparent over-dispersion, while accommodating the inherent correlation due to the longitudinal study design or nested data structure. Significance of the research lies in its scientific novelty and the breadth of its practical applications. The benefits to public health will accrue both nationally and internationally. For the empirical studies that motivated and are linked to this research, evaluation of health outcomes has significant implications in the prevention and control of recurrent urinary tract infections, hospital strategic planning, and post-stroke care and rehabilitation management. Moreover, appropriate assessment of a physical activity intervention for older adults is pertinent to falls prevention and reduction of musculoskeletal disorders among sedentary seniors.Read moreRead less
Improved Ways To Study The Effect Of Transient Exposures On The Risk Of An Illness, With Application To Flying And DVT
Funder
National Health and Medical Research Council
Funding Amount
$212,250.00
Summary
Improved methods of analysis will be developed to estimate the extent to which certain short-term activities trigger a particular illness and to determine who is most at risk. The new methods of analysis have many potential applications, including study of the effect of periods of intense exercise or intense alcohol consumption on the risk of a heart attack. Here we apply them to study the effect of air travel on illness due to blood clots in a vein (deep vein thrombosis, DVT), and the factors t ....Improved methods of analysis will be developed to estimate the extent to which certain short-term activities trigger a particular illness and to determine who is most at risk. The new methods of analysis have many potential applications, including study of the effect of periods of intense exercise or intense alcohol consumption on the risk of a heart attack. Here we apply them to study the effect of air travel on illness due to blood clots in a vein (deep vein thrombosis, DVT), and the factors that put individuals at greatest risk. The extra understanding that our improved analysis of the data provides will accelerate research into ways to minimise the risk of flight-induced DVT and implementation of preventative measures in the travel industry. This has the potential to prevent many cases, and deaths, because the number of flights taken, globally, by individuals in a year is now in the billions. Specifically, dependence of flight-induced DVT on age, sex, being pregnant, recent fractures and having certain cancers will identify individuals at greatest risk, while dependence on the duration of the flight will identify flights that present greater risk. The development of these new methods of analysis will make a lasting contribution to public health research, because they can be used to study many short-term-activity - illnesses combinations. The methods will see increasing applications because the type of data they rely on, namely the complete history of exposures over time periods, will increasingly become available as electronic recording of activities becomes more common place. For example, electronic records of flights are now almost universal and bookings at squash courts and other sporting venues are increasingly recorded electronically. The computer software to apply the new methods of analysis will be made available to other researchers, to promote studies of this type.Read moreRead less