Background/Aims We investigated the ability of composite scores developed using statistical models to differentiate progressive cognitive deterioration in Huntington's disease (HD) from natural decline in healthy controls.

Methods Using data from TRACK-HD the optimal combination of 24-month changes in quantitative cognitive measures to differentiate early stage HD individuals from controls was determined using logistic regression. Composite scores were calculated from the parameters of the model. Linear regression models were used to calculate effect sizes (ES) quantifying the difference in longitudinal change in the composite over 24 months between premanifest and early stage HD groups and controls respectively. This ES was compared with those for individual cognitive outcomes and other measures used in HD research. 0.632 bootstrap methodology was used to allow for biases which result from developing and testing models in the same sample.

Results The composite score gave an ES for the difference in rate of 24-month change between early HD and controls of 1.14 (95% CI 0.90 to 1.39): larger than that for any of the individual cognitive outcomes and those for the UHDRS TFC and TMS. Additionally the composite gave a statistically significant difference in the rate of change in premanifest HD compared to controls over 24-months (ES: 0.24; 95% CI 0.04 to 0.44), despite none of the individual cognitive outcomes having statistically significant ES over this period.

Conclusions Composite scores developed using appropriate statistical modelling techniques have the potential to materially reduce required sample sizes for randomised controlled trials.