PulseExploreJournal ClubResearchersJournals
Instagram
HomeJournal ClubExplore
Synapse
⌘+K
Synapse
March 31, 2026Journal of General Internal Medicine

Estimating the Hawthorne Effect in Real-World Blood Pressure Control Trials: An Analysis of the BP Home Trial

View Full Paper
Ask AI
Bookmark
Share

Why the study?

Results from blood pressure control interventional trials must be interpreted cautiously, motivating an analysis to deconstruct observed drops in systolic blood pressure into components such as increased adherence, regression to the mean, and initiation of new medications.

Does enrollment in a research study and self-measurement of blood pressure reduce systolic blood pressure through a Hawthorne effect in patients with uncontrolled blood pressure?

Population

Smartphone-owning patients with uncontrolled BP > 145 mmHg wishing to lower BP by > 10 mmHg

Comparison

One of two devices for self-measurement of BP

Design

Secondary analysis of a pragmatic randomized controlled trial

Follow-up

Up to 24 months

Key result

Enrollment in a blood pressure trial resulted in an immediate systolic blood pressure drop of -4.2 mmHg (95% CI, -5.1 to -3.3; p<0.001) due to increased adherence to pre-existing medications.

Authors

MRMax RosenVFValy FontilMMMadelaine Faulkner Modrow

Discussion

Loading...

Member takes

Overview

BP trial SBP reductions may partly reflect adherence gains and regression to the mean; supports adjusted analyses in future intervention studies.

Key Points

  • Analyze the components contributing to observed systolic blood pressure reductions in a BP control trial.
  • Secondary analysis of BP Home trial data.
  • Participants were smartphone owners with uncontrolled BP, randomly assigned to BP monitoring devices.
  • Systolic BP trajectories were analyzed using linear mixed models.
  • Average immediate drop in SBP of -4.2 mmHg (95% CI, -5.1 to -3.3; p<0.001) due to increased medication adherence.
  • New medication class initiation led to an SBP drop of -4.1 mmHg (95% CI, -5.3 to -2.8; p<0.001).
  • Average expected regression to the mean was -3.9 mmHg (95% CI, -4.4 to -3.3; p<0.001).

Study Design

Type

RCT

Randomization

randomly assigned

Structured PICO

Does enrollment in a research study and self-measurement of blood pressure reduce systolic blood pressure through a Hawthorne effect in patients with uncontrolled blood pressure?

P
Population
Patients owning a smartphone who reported uncontrolled blood pressure at their last clinic visit (> 145 mmHg) and a desire to lower their blood pressure by > 10 mmHg.
I
Intervention
One of two devices for self-measurement of blood pressure, followed for up to 24 months via electronic health records.
C
Comparator
The other device for self-measurement of blood pressure.
O
Outcome
Electronic health record (EHR)-recorded office systolic blood pressure (SBP).surrogate

Main Result

Effect estimate: Mean drop -4.2 mmHg (95% CI -5.1 to -3.3)

p-value: p=<0.001

A significant portion of blood pressure reductions observed in clinical trials may be attributable to the Hawthorne effect (increased adherence due to study enrollment) and regression to the mean, rather than the specific intervention being tested.

Cite This Study

Rosen et al. (2026) conducted an RCT in Uncontrolled blood pressure. Self-measurement of BP devices was evaluated on EHR-recorded office SBP (Mean drop -4.2 mmHg, 95% CI -5.1 to -3.3, p=<0.001). Enrollment in a blood pressure trial resulted in an immediate systolic blood pressure drop of -4.2 mmHg (95% CI, -5.1 to -3.3; p<0.001) due to increased adherence to pre-existing medications.

synapsesocial.com/papers/6a025c8eedf6f48138594762https://doi.org/10.1007/s11606-026-10249-1
View Full Paper
Ask AI
Bookmark
Share