Page 239 - PhD GT

10.5 An alternative approach to the design and analysis
From the results of regression analysis it would seem that, of those variables measured in the study, more were found to influence the change in use of the substance than the change in dependence on the substance. This begs a number of questions: is there a different design that might better be able to test the study hypotheses regarding the nature of change in dependence? What other measures might have been useful in elucidating the nature of change in dependence through this sort of analysis? Is the repeated finding of dependence as a predictor of change in dependence more than a statistical artefact? Certainly the higher the initial dependence score the greater the scope there is for change.
That change in dependence occurs over time can be said to have been established. The influences on such change remain less clear; further examination of the data collected in the present study with emphasis on changes in use of the substance and the relationship between use and dependence may shed more light on the course of change in dependence. Of particular importance might be the question of the direction of the influence between use and dependence. In a different design, it would be possible to investigate the relationship between use and dependence by tracking both more closely over time and investigating the questions of whether it is the case that as dependence develops it is driven by use and as it declines it drives use. It may be that the influence of other variables is mediated through use of the substance in the ways suggested above in the discussion of the findings. Treatment was shown to predict changes in use which in turn are associated with changes in dependence. Is the conclusion in either case that use, because of its relative transparency and conceptual simplicity should remain the target of treatment?
10.5.1 Coping
That there is a relationship between coping and dependence seems to have been established. The more dependence there is overall, the less coping there is. The two things change and the relationship between them changes. The timing of the respective changes was not identified by the methodology used in this thesis. The design of the analysis was based upon the assumption that coping would influence the level of dependence but it may be that it was the level of dependence which determined the nature of coping, or that the two variables were separately influenced. It was difficult to establish whether change in coping drove change in dependence or that the two changes happened simultaneously. In order further to investigate this relationship, measures of coping, use and dependence might be collected at regular intervals, say one month intervals over a period of
234