Mr Stephen Nash

Research Fellow


Keppel Street
United Kingdom

020 7927 2111

I am a statistician; I completed my MSc in Medical Statistics at the School in 2013, and have subsequently worked at King's College (the Institute of Psychiatry) and UCL (the Cancer Trials Centre).

Prior to 2012 I worked for seven years as a communications and information manager for various charities providing support to people in prison, their families, and people seeking to resettle in the community after a custodial sentence.


Department of Infectious Disease Epidemiology
Faculty of Epidemiology and Population Health


Centre for Statistical Methodology


I am currently employed to support Module Organisers prepare for online teaching. I have worked on three modules: STEPH (an introductory statistics module for ~250 students), Regression, and General Linear Models. The latter two are in the Medical Statistics department, and are taken by around 30 students on the Medical Statistics MSc.

Previously (2015-2019) I was a module co-leader on the Statistics for Epidemiology and Population Health (STEPH) module, which runs from October to December each year. My fellow co-leaders were Liz Allen and David Macleod.


My current role is for education only. Previously (2015-2020) I divided my research time between two strands of work:

a) I provided statistical advice to the MRC Ugandan Virus Research Institute. In particular, I was the trial statistician for CHAPS, a novel study using an ex vivo HIV challenge of resected foreskin tissue to assess effecicy of PrEP. The trial is taking place in four sites across three countries (Uganda, South Africa, Zimbabwe).

b) I did methodological research. I have co-written two Stata commands: clan and slopepower, both available from the SSC server. The former performs cluster level analysis of a CRT, and can be used for binary, count, and Normal data. The latter is a sample size calculator for use in situations where the outcome is a rate over time, and you have previous data, which the program uses to estimate the covariance matrix, and hence allow more efficient trials to be planned.