Re: Meta-analysis of side effects of statins shows need for trial transparency
The results of your metanalysis may be considerably flawed, because of lack of external validity of the data on statins recorded in clinical trials.The majority of clinical trials with statins have been performed in highly selected populations.
In the HPS (Heart Protection Study), 32,145 patients with the inclusion criteria participated in a run in phase, but 11,609 (36%) were excluded because of lack of effect on cholesterol, increase in liver enzymes, increase in CPK, or increase of creatinine, or also because "the patient had little probability of complying with the treatment during 5 years".
In the MEGA trial (Management of Elevated Cholesterol in the Primary Prevention Group of Adult Japanese), 15,210 patients entered a 4-week run in phase, but 7,201 (48%) were excluded for similar reasons, and only 8,009 finally participated in the trial.
In the JUPITER trial (Justification for the Use of Statins in Prevention: an Intervention Trial Evaluating Rosuvastatin), 89,890 were initially screened, but 78% were excluded for unclear reasons, and only 19,323 were considered for randomisation. Then, an additional 1,521 were excluded during a run in phase because of poor treatment adherence, and only 17,802 patients (19.8%) were finally randomised.
In view of these massive exclusions, we conclude that your data lack external validity, and that no conclusions on their incidence and severity can be drawn from these trials with a pre-randomisation run in phase.
Competing interests: We do not have any financial competing interest.