In the linked systematic review and meta-analysis (doi:10.1136/bmj.d2234), Bangalore and colleagues assess cardiovascular and other outcomes associated with angiotensin receptor blockers (ARBs). They found firm evidence to refute previous concerns that these drugs increase the risk of myocardial infarction. They also found that compared with controls, ARBs reduce the risk of stroke, heart failure, and new onset diabetes.1

In balancing risk against benefit health systems commonly raise safety concerns about drugs in common use. Such concerns may stem from the clear demonstration of catastrophic risk, such as was identified for thalidomide, and result in immediate and essential withdrawal of the drug from the market. Withdrawal can also be voluntary, as for drugs that do not pose a specific risk but have no advantages over competitor products. The cyclo-oxygenase-2 selective non-steroidal anti-inflammatory drugs rofecoxib and valdecoxib, which resulted in greater use of non-selective non-steroidal anti-inflammatory drugs plus proton pump inhibitors, provide such an example.2

Some drugs that were rapidly withdrawn on safety grounds would have best been withdrawn gradually or not at all. An example of this is the precipitate 1995 withdrawal of third generation oral combined contraceptives (because of a 1.7 relative risk but …