Use this URL to cite or link to this record in EThOS:
Title: The analysis of multiple correlated outcome measures in randomised controlled trials
Author: Vickerstaff, Victoria
ISNI:       0000 0004 8508 031X
Awarding Body: UCL (University College London)
Current Institution: University College London (University of London)
Date of Award: 2019
Availability of Full Text:
Access from EThOS:
Full text unavailable from EThOS. Please try the link below.
Access from Institution:
Multiple primary outcomes are sometimes collected and analysed in randomised controlled trials (RCTs), and are used in favour of a single outcome. By collecting multiple primary outcomes, it is possible to fully evaluate the effect that an intervention has for a given disease process. A simple approach to analysing multiple outcomes is to consider each outcome separately, however, this approach does not account for any pairwise correlations between the outcomes. Any cases with missing values must be ignored, unless an additional imputation step is performed. Alternatively, multivariate methods that explicitly model the pairwise correlations between the outcomes may be more efficient when some of the outcomes have missing values. When analysing multiple outcomes in a trial, it is important to control the family wise error rate (FWER), which is the probability of finding at least one false positive result. A common approach is to adjust the p-values for each statistical test. It is also important to consider the power to detect the true effects of the intervention. In this thesis, I present an overview of the relevant methods that could be used to analyse multiple outcomes in RCTs, including methods based on multivariate multilevel models. I perform simulation studies to provide guidance on which methods should be used to adjust for multiple comparisons in the sample size calculation, and which methods should be used for the analysis when the multiple primary outcomes are correlated. Additionally, I use simulation studies to investigate the differences in the power obtained when using multivariate models compared to analysing the outcomes separately using univariate models. Different simulation scenarios were constructed by varying the number of outcomes, the type of outcomes, the degree of correlations between the outcomes and the proportions and mechanisms of missing data.
Supervisor: Omar, R. Z. ; Ambler, G. Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available