U.S. and European guidelines endorse the use of any of the following medications for the first-line treatment of hypertension, unless contraindicated: thiazide or thiazide-like diuretics, angiotensin-converting enzyme (ACE) inhibitors, angiotensin receptor blockers, and calcium channel blockers. A systematic, large-scale study was designed to compare the effectiveness and safety of these antihypertensive medications across six administrative claims and three electronic health record databases from four countries. The researchers used a retrospective, comparative new-user design with each database source.

Data from nearly 4.9 million patients were analyzed. ACE inhibitors were found to be the most frequently initiated antihypertensives (48%), followed by thiazide or thiazide-like diuretics (17%), dihydropyridine calcium channel blockers (16%), angiotensin receptor blockers (15%), and nondihydropyridine calcium channel blockers (3%). More than half of comparisons showed no differences in effectiveness among medication classes. However, the researchers found a 15% lower event rate for three primary effectiveness outcomes-acute myocardial infarction, heart failure, and stroke-for thiazide or thiazide-like diuretics compared with ACE inhibitors. There were no significant differences in these outcomes between thiazide or thiazide-like diuretics and either angiotensin receptor blockers or dihydropyridine calcium channel blockers, and no differences between ACE inhibitors, angiotensin receptor blockers, and dihydropyridine calcium channel blockers. Nondihydropyridine calcium channel blockers, however, were found to be inferior to the other medications. The safety profile of thiazide or thiazide-like diuretics, according to the authors, was "markedly better" than that of ACE inhibitors.

The researchers conclude that initiating treatment with a thiazide or thiazide-like diuretic instead of an ACE inhibitor could have a substantial effect on public health. They also note some limitations of their study. Some of the databases lacked blood pressure measurements, for example, and baseline blood pressure readings might influence medication choice. In addition, they note that reliability and reproducibility may be limited with cardiovascular observational research.