Gillian Maguire's List: STATISTICAL BIAS

• In an accompanying editorial, two British researchers went further, calling the notion that newer agents are more effective or safer than older drugs "spurious." 

   

  "As a group they are no more efficacious, do not improve specific symptoms, have no clearly different side-effect profiles than the first-generation antipsychotics, and are less cost effective," wrote Peter Tyrer, M.D., of Imperial College London, and Tim Kendall, M.D., Ph.D., of the National Collaborating Centre for Mental Health in London.
• They did find that four second-generation agents were significantly more effective than older drugs. They were clozapine (Clozaril), olanzapine (Zyprexa), risperidone (Risperdal), and amisulpride (Solian, not approved in the U.S.).

• More than half the studies they did include had a different type of bias, Drs. Tyrer and Kendall said -- they used haloperidol as the comparator. 

   

  Haloperidol has a high rate of extrapyramidal effects compared with other first-generation drugs. 

   

  The result, they suggested, was to make the newer drugs' side effect profile look better than if other comparators had been tested instead.

1 more annotation...

• Massive unpublished negative study bias goes on; extrapolating from that study, we would expect that this study, which only consists of about 10% unpublished (and presumably mostly negative) data, overestimates benefits with study drugs versus other drugs by some percentage; perhaps there are up to 30-50% more studies out there not included in this review.
• Meta-analysis done on an industrial scale often obscures instead of clarifying. The methods are platinum, hence a publication in Lancet, but the results are dross. The process has been likened to statistical alchemy by some researchers who have warned against this, which is why it is generally accepted that a single randomized study that is large is much more valid than adding up multiple small studies in a meta-analysis.