Longitudinal trajectories of smoking, alcohol and other health behaviours and relationship with clinical features in multiple sclerosis
- Research Opportunity
- PhD, Masters by Research, Honours
- Number of Honour Places Available
- Medicine and Radiology
- Royal Melbourne Hospital
|Dr Steve Simpson, Jr.||firstname.lastname@example.org||Personal web page|
|A/Prof Tracey Weilandemail@example.com||Personal web page|
Multiple sclerosis (MS) is a progressive, autoimmune, demyelinating condition of the central nervous system, manifesting in sensory, motor and/or cognitive dysfunction. Given its onset is typically in the prime years of life – often in the 20s – it has devastating impacts on the quality of life and independent living of the patients so affected.
While it is well-known that more healthy lifestyle behaviours are associated with better clinical course, there is a significant potential for reverse causality in these relationships, particularly behaviours like physical activity and diet, as well as other behaviours like alcohol and smoking. What is lacking is a large-sample study of the long-term trajectories in health-related behaviours and how these trajectories relate to clinical outcomes.
This project will make use of advanced statistical techniques, particularly latent class analysis, to assess trajectories of health behaviours over time in the HOLISM longitudinal cohort (n=2,644 at baseline, n=1,401 at 2.5-yr follow-up, n=952 at 5-yr follow-up), assess the characteristics, clinical and demographic with membership in these trajectory groups, and correlate these trajectory groups with clinical outcomes -including relapse rate, disability as measured by the Patient Determined Disease Steps and fatigue as measured by the Fatigue Severity Scale - and quality of life as measured by the MSQOL-54. The results of this work will be invaluable for medical practitioners and carers seeking to improve the health behaviours of people with MS and the population more broadly.
The primary analysis method to be employed here is latent class analysis, which will glean the groupings of individuals’ health-related behaviours over time, which may range from consistently healthy, consistently non-healthy, improving, declining, static, or some mixture of these. Other analysis methods to be employed include linear regression, multilevel mixed-effects linear regression, log-binomial regression, and/or Poisson regression, including univariable and multivariable models, as well as potential inter-group assessment of interaction by sex, age, MS course, or others as appropriate.
Faculty Research Themes
School Research Themes
PhD, Masters by Research, Honours
Students who are interested in joining this project will need to consider their elegibility as well as other requirements before contacting the supervisor of this research
For further information about this research, please contact a supervisor.
Research NodeRoyal Melbourne Hospital
MDHS Research library
Explore by researcher, school, project or topic.