Modelling of disease progression in acute stroke by simultaneously using the NIH stroke scale, the Scandinavian stroke scale and the Barthel index
Kristin E. Karlsson(1), Justin Wilkins(1,3), Mats O. Karlsson(1) and E. Niclas Jonsson(1,2)
(1) Dep. of Pharmaceutical Biosciences, Uppsala University, Uppsala, Sweden, (2) Hoffman-La Roche Ltd, Basel, Switzerland, (3) Novartis Pharma AG, Basel, Switzerland
Objectives: The aim of this study was to develop a disease progression model jointly for three stroke scales; the NIH stroke scale, the Scandinavian stroke scale and Barthel index. The three stroke scales are assessing neurological and/or functional deficit in stroke patients.
Methods: Scores assessed on three occasions over a 90-day period in 772 acute stroke patients were used to model the time course of recovery using NONMEM VI. The patients were from the control arm of a double-blind, multinational, multicenter, placebo-controlled investigation of the efficacy of a novel acute stroke compound. At each measurement occasion, three discrete events, on each of the scales, were possible: attainment of a maximum score on the scale (full recovery), improvement or decline, or dropout [1]. Each of these possible events had a probability and a score change magnitude associated with it. A joint function for the dropout was used since the definition of dropout was withdrawal from the study, i.e. a dropout event will occur on all scales simultaneously. To accommodate the non-monotonic nature of these transitions, it was necessary to develop a strategy that considered both the longitudinal, continuous aspects and the probabilistic characteristics of the data simultaneously. Time-related variables, including baseline score, previous score, and time since the previous observation, were considered as predictors, as well as demographic covariates such as age.
Results: A model using the information from three stroke scales was developed which included a joint dropout model and correlation between the interindividual variabilities in the three scales.
Conclusions: In stroke studies several stroke scales are often used, e.g. one neurological and one functional assessment scale. With simultaneous modelling of these scales it is possible to combine information from the scales regarding for example correlations between interindividual variabilities between different scales. What is also important is that a simultaneous model enables the use of joint, i.e. scale independent, functions such as a dropout model. However, the scales can be used as predictors of the dropout event. This kind of model could utilise data across different scales collected in different trials.
Reference:
[1] Jonsson F et al. A longitudinal model for non-monotonic clinical assessment scale data. J Pharmacokinet Pharmacodyn. 2005 Dec;32(5-6):795-815.
