Record Detail Back
THERAPEUTIC STRATEGIES IN DEMENTIA
- Share to:
One approach to answering this question and perhaps the most conventional in an ‘evidence-based medicine’ medical culture is by conducting meta-analyses.
In terms of ‘level of evidence’, the meta-analysis is considered strongest [1]. However, whilst achieving much statistically by aggregating trials and synthesizing data to achieve greater precision of outcome estimates, the meta-analysis usually achieves only a very nar- row conclusion. Also, by aggregating data from a variety of sources, meta-anlayses can miss out particular effects in subsample populations. Finally, because meta-analysis combines data from different sources, the meta-analysis is only as reliable as the data from which it is drawn.
If the condition being treated has a single (usually biological) outcome, then both trials and subsequent meta-analyses of trials will achieve the primary aim of meta-analyses, i.e. to answer ‘does the drug work?’. For example, if the disease is due to an infection, then a meta- analysis of a new antimicrobial would combine trials looking at the efficacy of the drug in removing the underlying bacteria. This could be tested by the return of negative cultures after a certain duration of treatment. In this case, clear, relevant biological outcomes and a simple trial design are used, making interpretation of the meta-analysis straightforward.
How do we define a positive outcome in treating Alzheimer’s disease (AD) though? Is it to be cognitive, functional, behavioural, quality of life, biological, mortality, avoidance of institutionalization, reduced carer stress, global clinical impression or neuropsychiatric? It is clear that in a complex illness such as AD, a positive outcome means different things to dif- ferent parties. This efficacy vs. effectiveness [2] dilemma (Figure 1.1) has been underpinning the unenviable challenge that groups like the UK’s National Institute for Health and Clinical Excellence (NICE) have been facing in describing whether AD treatments are ‘effective’ [3] and this is articulated in the chapter by Colin Green in this book. A further criticism of AD clinical trials has been levelled (usually without offering any sensible solution) which criti- cizes the design of the trials themselves and their analysis. The main thrust of the collective criticism is that analysis of data from conditions that are associated with inevitable decline would favour treatments that cause high dropout early in the trial if the standard ‘last observation carried forward’ method of imputing data is used. There are also criticisms of trials that last only several months in a condition that clinically usually lasts up to 10 years.
C. W. Ritchie, D. Ames, C. L. Masters, J. Cummings - Personal Name
1st Edtion
978 1 84692 566 5
NONE
THERAPEUTIC STRATEGIES IN DEMENTIA
Management
English
Atlas Medical Publishing Ltd
2007
USA
1-373
LOADING LIST...
LOADING LIST...