Survival bias in observational studies on age-related neurodegenerative diseases

Applications are invited from graduates with a BSc (First or Upper Second) or MSc (Distinction), or equivalent, to work within the Wolfson Institute of Preventive Medicine. This 3 year studentship will commence in Spring 2020 and will be based at the Charterhouse Square Campus. This is an exciting opportunity for a graduate from disciplines related to epidemiology, statistics, and behavioural sciences.

1. Background to the project

Parkinson’s disease (PD) is the second most common degenerative disease of the brain. The number of people diagnosed is growing faster than any other neurological condition and is expected to affect 14 million people worldwide by 2040. Several intriguing inverse associations have been observed between traditional causes of ill-health (such as smoking, alcohol, high body mass index and type 2 diabetes mellitus) and a decreased risk of PD. Whether the ‘protective’ effects of these factors against PD are true or an artefact remains to be determined. The role that survival bias plays in epidemiological studies with age-related outcomes, such as PD and dementia, has been poorly studied. Many of the apparent protective effects observed may in fact result from bias due to premature mortality and death before development and diagnosis of PD or dementia. This phenomenon has not been systematically studied but has great relevance to efforts to reduce the burden of neurodegenerative diseases.

2. Proposed research and techniques

The successful PhD candidate will start the project by reviewing the evidence for protective effects of (usually harmful) traits on risk of PD and dementia, before reviewing existing statistical methods and approaches to studying the effect of, and allowing for, survival bias, such as competing risk analyses, and further developing statistical methods and approaches to study the role and effect of survival bias in observational studies. The successful candidate will work with colleagues within the Preventive Neurology Unit in the Wolfson Institute of Preventive Medicine, who work with several datasets including primary care records, UK Biobank data and together with colleagues at UCL run the PREDICT-PD study (predictpd.com) which is currently recruiting 10,000 participants with the aim of developing statistical models using simple tests to predict those who are at high risk of PD before symptoms appear. The successful candidate may have opportunities to work with these data.

3. Project timeline

Months 0-6 – Literature review on protective effects of traits on risk of PD and dementia
Months 6-30 – Review of existing statistical methodologies and development of new methodologies to study the role of survivor bias
Months 30-36 – Thesis write-up

4. Eligibility

This project requires a highly skilled student with excellent communication skills, and preferably with a qualification in medical statistics or significant research experience.


Informal enquiries can be made to via email: Jonathan Bestwick [Email Address Removed]

How to apply
Your application should consist of a CV and contact details of two academic referees. You must also include a personal statement (1,000 words maximum) describing your suitability for the selected project including how your research experience and interests relate to the project.

Please submit your application to: Patrick Mullan ([Email Address Removed]).