On Tuesday 26 June 2007 15:59, Ken Harvey wrote: > Having said that, I don't agree that studies such as this are only an > "academic wank".
Seriously, Ken: if I investigate a given number of drugs that (one would assume) are prescribed to ill people ... .. one would expect that the outcome as a whole compared to their matched peers (except for the illness and severity thereof) would be worse (unless the treatment is a perfect cure without any adverse effects) Thus, all they do is stating the obvious - namely people suffering from a condition treated with certain medicatons are more likely to suffer adverse outcome than people not receiving the medication and possibly *not suffering the condition* requiring the medication In order not to qualify as "academic wank" one would expect that they undergo the trouble and comparing those who were prescribed the "safer alternatives" for comparable indications and severities of illness and co-morbidities etc - THEN we would get valuable insight! Or, even better - if they would investigate what happens when we cease that medication in a properly randomized fashion in those patients - do those who were ceased fare better than those who continue? THAT would give us good insight too, and would be likely to change prescription behaviour because THAT is data that can impress a clinician. But such study cannot be done sitting comfortably in an armchair simply by ruminating over conveniently available computerized data. Horst _______________________________________________ Gpcg_talk mailing list [email protected] http://ozdocit.org/cgi-bin/mailman/listinfo/gpcg_talk
