Please spare us from another systematic review (Kaduszkiewicz et al BMJ 2005; 331: 321-327) that adds absolutely nothing to the sum of human knowledge.
The hubris of this group knows no bounds, they admit to being out of step with virtually every randomised controlled trial and meta analysis so far published and yet confidently assert that they have it right and the rest of the world has somehow overlooked the obvious.
This review is timely coming as the UK's National Institute for Health and Clinical Excellence (NICE) delays issuing its guidance on the only drug treatments for Alzheimer’s disease. However, even NICE acknowledges that these drugs are clinically effective, it is their cost effectiveness that it disputes.
This analysis flies in the face of expert opinion and the experiences of people with dementia and their carers. The main complaint about previous studies seems to be the choice of the last observation carried forward (LOCF) as an end point in studies with non-random dropouts as if no-one undertaking studies had ever thought of that.
The analysis of the data on the trial for which I was responsible (Wilkinson et al 2001) used the LOCF but with the active/placebo difference calculated at the time of dropout, not from the placebo decline at endpoint to minimise the exaggeration of the effect size in a deteriorating disease. This tendentious article displays a lack of understanding not only of the disease but also of the quality control in pivotal trials.
This is exemplified by far too much being made of the theoretical problem of the LOCF analysis when the effects were often as great or greater with the observed case or completer analyses which do not impute missing values.
That every last detail is not included in the published papers does not mean it did not happen or was not considered. One must lift one’s eyes to the horizon occasionally; remember if it swims, quacks and flies like a duck I am afraid it probably is a duck no matter what theoretical criticisms one has of the observer's viewpoint.
Clinical practice is not about the mean changes in test scores but about individual responses. Means and confidence intervals in good RCT's are necessary to give us the confidence to use something in clinical practice but they are not the answer, experience tells us whether it is useful.
To paraphrase Oscar Wilde this kind of review panders to the cynics who know the price of everything but the value of nothing.
This attitude was reflected in the recent draft guidance from NICE suggesting the three drugs reviewed in this paper and memantine are not worth prescribing in the NHS in England and Wales. If this stance is also taken by the similar committee in Germany the Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen (IQWIG) it will be a triumph for the English and German Generals and a complete betrayal of those suffering in the trenches yet again. The kind of irresponsible pseudoscience demonstrated in this paper only fuels this perverse zeitgeist.
Rapid Response:
Evidence for cholinesterase inhibitors in AD
Please spare us from another systematic review (Kaduszkiewicz et al BMJ 2005; 331: 321-327) that adds absolutely nothing to the sum of human knowledge.
The hubris of this group knows no bounds, they admit to being out of step with virtually every randomised controlled trial and meta analysis so far published and yet confidently assert that they have it right and the rest of the world has somehow overlooked the obvious.
This review is timely coming as the UK's National Institute for Health and Clinical Excellence (NICE) delays issuing its guidance on the only drug treatments for Alzheimer’s disease. However, even NICE acknowledges that these drugs are clinically effective, it is their cost effectiveness that it disputes.
This analysis flies in the face of expert opinion and the experiences of people with dementia and their carers. The main complaint about previous studies seems to be the choice of the last observation carried forward (LOCF) as an end point in studies with non-random dropouts as if no-one undertaking studies had ever thought of that.
The analysis of the data on the trial for which I was responsible (Wilkinson et al 2001) used the LOCF but with the active/placebo difference calculated at the time of dropout, not from the placebo decline at endpoint to minimise the exaggeration of the effect size in a deteriorating disease. This tendentious article displays a lack of understanding not only of the disease but also of the quality control in pivotal trials.
This is exemplified by far too much being made of the theoretical problem of the LOCF analysis when the effects were often as great or greater with the observed case or completer analyses which do not impute missing values.
That every last detail is not included in the published papers does not mean it did not happen or was not considered. One must lift one’s eyes to the horizon occasionally; remember if it swims, quacks and flies like a duck I am afraid it probably is a duck no matter what theoretical criticisms one has of the observer's viewpoint.
Clinical practice is not about the mean changes in test scores but about individual responses. Means and confidence intervals in good RCT's are necessary to give us the confidence to use something in clinical practice but they are not the answer, experience tells us whether it is useful.
To paraphrase Oscar Wilde this kind of review panders to the cynics who know the price of everything but the value of nothing.
This attitude was reflected in the recent draft guidance from NICE suggesting the three drugs reviewed in this paper and memantine are not worth prescribing in the NHS in England and Wales. If this stance is also taken by the similar committee in Germany the Institut für Qualität und Wirtschaftlichkeit im Gesundheitswesen (IQWIG) it will be a triumph for the English and German Generals and a complete betrayal of those suffering in the trenches yet again. The kind of irresponsible pseudoscience demonstrated in this paper only fuels this perverse zeitgeist.
Competing interests:
None declared
Competing interests: No competing interests