Diuretic Comparison Project (DCP)

• Time to major cardiovascular event [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  The Primary outcome measure will be time to a major cardiovascular event, defined as a composite outcome comprised of the first occurrence after randomization of any of the following: stroke, myocardial infarction, urgent coronary revascularization because of unstable angina, hospitalization for acute congestive heart failure, non-cancer death.
  
  

• Time to event for each component of the composite primary outcome and additional cardiovascular events [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  The Secondary outcomes will include time to event for (1) each of the 5 components of the composite primary outcome, (2) all deaths, (3) the composite outcome substituting all deaths for non-cancer deaths, (4) "possibly vascular deaths," (5) the composite outcome substituting "possibly vascular deaths" for non-cancer deaths, (6) any revascularization of any artery.
  
  
• All Deaths [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  All Deaths
  
  
• Urgent revascularization because of unstable angina [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  Urgent revascularization because of unstable angina
  
  
• Hospitalization for acute congestive heart failure [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  Hospitalization for acute congestive heart failure
  
  
• Non-cancer death [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  Non-cancer death
  
  
• The primary composite outcome substituting 'all deaths' for 'non-cancer deaths' [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  This secondary outcome measure will be time to a major cardiovascular event, defined as a composite outcome comprised of the first occurrence after randomization of any of the following: stroke, myocardial infarction, urgent coronary revascularization because of unstable angina, hospitalization for acute congestive heart failure, all deaths.
  
  
• Possibly Vascular Deaths [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  Defined as all deaths caused by vascular diseases, diabetes, external causes, and unknown causes.
  
  
• The primary composite outcome substituting 'possibly vascular deaths' for 'non-cancer deaths' [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  This secondary outcome measure will be time to a major cardiovascular event, defined as a composite outcome comprised of the first occurrence after randomization of any of the following: stroke, myocardial infarction, urgent coronary revascularization because of unstable angina, hospitalization for acute congestive heart failure, possibly vascular deaths. 'Possibly vascular deaths' are defined as all deaths caused by vascular diseases, diabetes, external causes, and unknown causes.
  
  
• Any revascularization of any artery [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  Any revascularization of any artery
  
  
• Erectile dysfunction (ED) [ Time Frame: Randomization to time to event; average follow-up 3 years ]
  Defined as first prescription for phosphodiesterase Type 5 (PDE5) inhibitor or referral for ED
  
  

Over 1 million Veterans are prescribed a thiazide-type diuretic each year; over 95% receive hydrochlorothiazide, and fewer than 2.5% receive chlorthalidone. Both medications are thiazide-type diuretics that have been used for more than 50 years and are considered first-line treatment for hypertension. Indirect evidence has been accumulating, however, that chlorthalidone may be more effective than hydrochlorothiazide at preventing cardiovascular events. This will be the first randomized head-to-head comparison of the effectiveness of these two drugs. The study plans to enroll 13,500 Veterans over 3 years and follow them on average for 3 years, resulting in a total study duration of 4.5 years. Patients currently prescribed hydrochlorothiazide will be randomized to either continue taking hydrochlorothiazide or to receive an equivalent dose of chlorthalidone. The unique 'point of care' or 'clinically integrated' study design will identify, enroll and follow subjects using the electronic medical record system and national VA and non-VA databases. The primary outcome is an event composite consisting of: stroke, myocardial infarction, non-cancer death, urgent revascularization, and hospitalization for acute congestive heart failure. All patient care, including the study drug, will continue to be managed by the primary care provider.

Primary Care Providers will also be included as participants in this research study. Adding providers as subjects will allow us to study the implementation of the point of care protocol design in addition to the original research question. The study team will collect data on diuretic management within the study and may contact providers by phone or email to learn reasons for declining a particular patient or discontinuing a new chlorthalidone order or an ongoing diuretic prescription.

If cardiovascular events are reduced by even a small amount by chlorthalidone, the public health effect will be considerable because of the large number of patients who take diuretics. A randomized trial, now feasible due to the investigators' efficient and inexpensive design, will provide evidence needed to better inform practice throughout the VA.