Goodness-of-fit Testing Of Log-link Models For Categorical Outcome Data
Funder
National Health and Medical Research Council
Funding Amount
$260,863.00
Summary
Information about the health consequences of exposure to causal factors is obtained from mathematical models of observed data. Incorrect inferences are possible if the model does not adequately represent the data. Relative risk models are recommended for observations over time on a cohort of subjects, but it is not known how best to assess the adequacy of such models. This project will assess the performance of summary measures of goodness-of-fit when applied to relative risk models.
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.
Evaluation And Improvement Of The Implementation Of The Intention To Treat Model In Controlled Trials Of Psychotherapies
Funder
National Health and Medical Research Council
Funding Amount
$409,000.00
Summary
Randomized controlled trials (RCTs) are the best way to determine whether patients benefit from a new treatment. In these trials patients are randomly assigned to the new, active treatment, or to a placebo or existing treatment. The groups are compared at the end of the trial. RCTs may be mounted for psychotherapy and educational programs as well as for new drugs and other medical procedures. A major problem for RCTs concerns their statistical analysis when some participants drop out before the ....Randomized controlled trials (RCTs) are the best way to determine whether patients benefit from a new treatment. In these trials patients are randomly assigned to the new, active treatment, or to a placebo or existing treatment. The groups are compared at the end of the trial. RCTs may be mounted for psychotherapy and educational programs as well as for new drugs and other medical procedures. A major problem for RCTs concerns their statistical analysis when some participants drop out before the end of the trial. Dropout is common in trials. Participants may drop out because they feel no benefit from the treatment, dislike side effects, or even because they have recovered quickly. Thus, to compare the groups remaining at the end of trial may introduce serious bias. The Intention to Treat (ITT) principle which has been widely adopted states that outcomes from all patients who enter a trial should be compared at its end. To achieve this, the last available observation for a participant who withdraws is often 'carried forward' to the end of the trial. While currently believed to be conservative, there is evidence that this approach is not always optimal. This project will examine the way in which dropout is treated in trials of two common psychiatric conditions: depression and anxiety disorders. The project will also undertake simulation research to investigate which of a number of modern methods of data analysis yield the most accurate results when participants drop out, and how changes in the design of trials might improve accuracy. The project is important because it will enable researchers to improve the conduct of trials in the future. Erroneous conclusions drawn from RCTs stand to condemn those suffering from disorders to ineffective treatment and to lead to the premature abandonment of potentially useful interventions which are falsely claimed to lack efficacy.Read moreRead less
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
Hierarchical Finite Mixture Modelling Of Health Outcomes: A Risk-adjusted Random Effects Approach
Funder
National Health and Medical Research Council
Funding Amount
$117,000.00
Summary
In medical and health studies, finite mixture regression models have been used to analyze data arising from heterogeneous populations. Traditionally, the application of mixture models is mainly concerned with finite normal mixtures. Recent computational advances and methodological developments have enhanced the extension of the method to non-normal finite mixtures, such as the modelling of discrete responses in finite mixture of generalized linear models and overlapping phases of failure time da ....In medical and health studies, finite mixture regression models have been used to analyze data arising from heterogeneous populations. Traditionally, the application of mixture models is mainly concerned with finite normal mixtures. Recent computational advances and methodological developments have enhanced the extension of the method to non-normal finite mixtures, such as the modelling of discrete responses in finite mixture of generalized linear models and overlapping phases of failure time data in the context of survival analysis. However, due to the hierarchical study design or the data collection procedure, the inherent correlation structure and-or clustering effects present may contribute to extra variations and violation of the independence assumption, resulting in spurious associations and misleading inferences based on the finite mixture model. This project aims to present a unified approach to accommodate both heterogeneity and dependency of observations, by incorporating random effects into finite mixture regression models. The new methodology will provide an integrated framework to analyze heterogeneous and correlated health outcomes. Three empirical studies are considered, namely, evaluation of an occupational injury reduction intervention, length of hospital stay modeling, and analysis of survival times of patients after cardiac surgery. The long term benefits to bioscience are accurate and valid conclusions inferred from medical and health studies, as well as the correct identification of high-risk subgroups. For the three application areas of this project, the improved analyses will specifically enable the evaluation of a participatory ergonomics intervention, the assessment of hospital efficiency and factors influencing length of hospitalization, and the determination of effectiveness of treatments prescribed pre- and post- operation, respectively.Read moreRead less
Effects Of Anti-smoking Advertising, Tobacco-related Press Coverage And Tobacco Control Policies On Smoking Behaviour
Funder
National Health and Medical Research Council
Funding Amount
$358,696.00
Summary
Tobacco control programs invest significant resources into the production, development, and placement of anti-smoking advertisements. Evaluation of anti-smoking advertising campaigns is made difficult by the fact that they operate within the broader social context of changing tobacco control policies, tobacco marketing, and other forms of relevant paid and unpaid media. Public health advocates spend considerable effort attempting to obtain favourable news coverage about tobacco control issues, y ....Tobacco control programs invest significant resources into the production, development, and placement of anti-smoking advertisements. Evaluation of anti-smoking advertising campaigns is made difficult by the fact that they operate within the broader social context of changing tobacco control policies, tobacco marketing, and other forms of relevant paid and unpaid media. Public health advocates spend considerable effort attempting to obtain favourable news coverage about tobacco control issues, yet virtually no research has systematically related the volume and nature of news coverage on tobacco issues to smoking behaviour change. This study will exploit the existing variation in tobacco control acitvity between states and over time to study effects of these two forms of media and tobacco control policies on youth and adult smoking behaviour. The study will merge reliable measures of exposure to televised anti-smoking advertising, press coverage about tobacco issues and detailed quantified measures of other tobacco control policies, with monthly surveys of smoking prevalence and consumption from 1991 to 2005. This unique monthly survey database has tracked smoking behaviour changes over time and between states using a standard method over the course of a 15-year period. Analysis of the data will determine the most optimum levels of anti-smoking advertised required to reduce smoking, the potential influence of news reporting as a way of educating people to reduce smoking and effects of tobacco policies, such as advertising bans, smoke-free policies and youth access policies on smoking behaviour. By assessing the effectiveness of anti-smoking advertising and other tobacco policies in changing smoking behaviour, this study will guide public health policy and funding decisions related to tobacco control.Read moreRead less
Exploring The Contributions Of Individual-, Area- And Service- Level Factors To Indigenous Health Outcomes
Funder
National Health and Medical Research Council
Funding Amount
$486,919.00
Summary
We will use linked hospital and death data, and multilevel models, to estimate the contributions of individual-, area- and service-level factors to inequalities in the outcomes of hospital care between Indigenous and non-Indigenous people. Factors investigated will include socioeconomic status, remoteness, access to primary care services, and hospital characteristics. Conditions explored will include heart disease, stroke, diabetes, chronic obstructive pulmonary disease and asthma.
Adaptations Of Methods For Estimation Of Familial Correlation In Age At Onset Of Disease
Funder
National Health and Medical Research Council
Funding Amount
$146,250.00
Summary
Chronic diseases such as coronary heart disease, breast cancer, prostate cancer and non-insulin dependent diabetes are responsible for a significant burden of ill-health in society. Studies of familial aggregation are important in determining the relative magnitude of genetic and lifestyle-environment factors associated with chronic diseases, and in identifying individuals and families at high risk, even in the absence of conventional risk factors. The findings have implications for health promo ....Chronic diseases such as coronary heart disease, breast cancer, prostate cancer and non-insulin dependent diabetes are responsible for a significant burden of ill-health in society. Studies of familial aggregation are important in determining the relative magnitude of genetic and lifestyle-environment factors associated with chronic diseases, and in identifying individuals and families at high risk, even in the absence of conventional risk factors. The findings have implications for health promotion in the general population and disease prevention in those identified to be at high risk. An outstanding characteristics of these studies is that many participants had an event at some unknown time before the entry. This project aims at adapting current methods to properly account for events before entry and to provide estimates of familial aggregation between parents, between children and between parents and children in the same model. Furthermore, it provides freely available software for proper familial analyses which have not had any feasible numerical methods (or software). In addition, it provides estimates of familial aggregation of coronary heart disease in Busselton families which has longer follow-up than most other studies. This project is very cost effective as the Busselton Health Study consists of a series of cross-sectional surveys since 1968 and currently has hospital morbidity and death follow-up from medical record linkage until the end of 1997.Read moreRead less
New Computational Methods For The Analysis Of Whole-genome Data
Funder
National Health and Medical Research Council
Funding Amount
$151,516.00
Summary
A complete understanding of the mechanisms underlying common diseases can only be achieved if all pathways at which genetic variation contributes to disease risk are identified. Most available methods to identify such predisposing genetic variation are adequately powered only when analysing data for many thousands of samples. We will develop more powerful statistical methods that can increase our ability to identify disease genes from large-scale association studies.
The Impact Of Outdoor Aeroallergen Exposure On Asthma Exacerbations In Children And Adolescents
Funder
National Health and Medical Research Council
Funding Amount
$473,924.00
Summary
Asthma is a chronic condition usually diagnosed in childhood and an important public health concern. We do not fully understand what triggers an asthma attack, although outdoor pollen and moulds may be important. This project will establish the relative importance of pollen and moulds in triggering asthma attacks among Australian children. It will fill gaps in our knowledge of environmental triggers of asthma. Such knowledge will improve asthma management and ultimately public health.