Effectiveness of Mailed Patient Activation Letters for Blood Pressure Control: A Randomized Quality Improvement Trial

Uncontrolled hypertension (HTN) is a leading preventable cause of death. Interventions are needed that activate patients and motivate them to work with clinicians to control their blood pressure (BP).

To test whether mailing patients a letter including information about their hypertension and a summary of computerized algorithm-generated medication adjustment suggestions improves BP control processes and outcomes.

Randomized quality improvement trial.

We identified patients receiving primary care at a large academic medical center with diagnosed HTN and uncontrolled BP (> 140 mmHg systolic or > 90 mmHg diastolic) at both of their last 2 visits.

Participants were randomized into three groups. The BP Activate Letter group received a letter containing algorithm-generated BP medication adjustment suggestions, and a recommendation to discuss these suggestions with their provider; the Control Letter group received a letter that recommended they talk to their provider about their HTN, without specific medication suggestions; and the No Letter group received no mailed outreach.

The primary outcome was time to occurrence of either a BP medication intensification or documented achievement of BP control to < 140/< 90 mmHg using EHR data extracted 6 months after letters were mailed.

The primary outcome, which was time to medication intensification or achievement of BP control, did not occur more frequently in the BP Activate Letter group (hazard ratio = 0.86; 95% confidence interval [CI]: 0.65 to 1.14), or in the Control Letter group (0.78; 0.59 to 1.03) compared to the No Letter group, and we saw no evidence of significant improvement in any secondary outcome or subgroup. Time to medication intensification appeared to be significantly longer in the Control Letter compared to the No Letter group (0.50; 0.30 to 0.85).

Mailing patients one letter with computerized BP medication adjustment suggestions to consider did not lead to effective patient activation.