The Centre for Biostatistics at The University of Manchester holds half-day seminars each March and October focussed around a designated theme. Our next seminar takes place on Wednesday 18th March on 'Issues in Joint Modelling', and features three excellent speakers. All are welcome to attend. The seminar is free, but registration is required for catering purposes. Please register in advance at: https://www.eventbrite.co.uk/e/centre-for-biostatistics-seminar-on-issues-in-joint-modelling-registration-15927725255 ------------------------------------------------------------------------------------------------------------------------------------ Date: Wednesday 18th March 2015 Time: 14.00-17.00 Venue: Manchester Dental Education Centre (MANDEC), Higher Cambridge Street, Manchester, M15 6FH Theme: Issues in Joint Modelling Programme: 14:00-14:50 "River Blindness, Eyeworm and a Calibration Problem" Professor Peter Diggle (Lancaster University and University of Liverpool) Onchocerciasis (River blindness) is a major public health problem in the wet tropical regions of the world, including most of sub-Saharan Africa. A multi-national programme to control the disease by mass administration of a protective drug has been very successful, with more than 60 million treatments to date over 19 countries. Howefver, the programme has been hampered by the recognition that people heavily infected with the Loa loa (eyeworm) parasite. Before the drug is administered in a community, it is relatively easy to estimate the prevalence of eyeworm infection, much harder under field conditions to estimate how many people are heavily infected, defined as carrying more than 8,000 parasites per ml of blood. This leads to the following prediction problem: given a sample of n people, of whom Y are infected with eyeworm, how many people in the community from which the sample is drawn are carrying more than 8,000 parasites per ml of blood? To address this problem we first develop a model for the variation in parasite count between individuals within a village, then explore how the parameters of this distribution vary between villages. The model will be used to calculate the predictive probability that an individual in a community will be carrying more than 8,000 parasites per ml of blood, given an estimate of community-level prevalence and any other relevant information in the form of covariates and/or prevalence estimates from neighbouring villages. 14:50-15:40 "Joint modelling of lung function and survival in cystic fibrosis" Dr Jessica Barrett (University of Cambridge) Cystic fibrosis is a genetic disorder which affects the lungs, pancreas, liver and intestine. Breathing difficulty is a serious symptom, and lung function is an important predictor of long-term survival. We explore the association between lung function and survival in cystic fibrosis patients using joint modelling. We analyse data from the UK Cystic Fibrosis Registry between 1999 and 2010 using discrete time methods. We compare various specifications of a shared random effects model which links the lung function and survival processes. 15:40-16:00 Refreshments 16:00-16:50 "Investigating the effect of drug titration on the relative effects of AEDs on treatment failure" Dr Ruwanthi Kolamunnage-Dona (University of Liverpool) Time to withdrawal of a randomised drug or addition of another (treatment failure) has been recommended by the International League Against Epilepsy to be one of the primary endpoints for clinical trials of anti-epileptic drugs (AEDs). Patients may decide to switch to an alternative AED because of inadequate seizure control (ISC) or to withdraw from a treatment because of unacceptable adverse effects (UAE). Overall analysis of treatment failures may miss differential effects of AEDs on the reasons for withdrawal, which may differ in terms of their relative importance for patients. The SANAD (Standard And New Antiepileptic Drugs) study is the largest trial of AEDs to date. The primary analysis of patients with partial epilepsy concluded that the newer drug lamotrigine was preferable in terms of treatment failure to carbamazepine which had been the standard for many years. Subsequent correspondence after the trial results were published indicated that some readers were concerned that differential titration rates may have been to the disadvantage of carbamazepine. An AED that is titrated more quickly may bring benefits in terms of seizure control but be more likely to cause adverse effects. This criticism has also been levelled at previous AED trials. We investigate the effect of drug titration on the relative effects of the two AEDs on treatment failure. In order to compare the two AEDs after adjusting for titration rate, we extend the standard joint model to allow for competing causes of treatment failure (ISC and UAE). Titration of AEDs is a difficult and controversial topic that impinges on the design and interpretation of AED trials. Investigating the association between titration rate and drug withdrawal through joint modelling offers a way forward. You may leave the list at any time by sending the command SIGNOFF allstat to [log in to unmask], leaving the subject line blank.