Blood pressure pills cut major cardiovascular events including heart attacks and strokes by 11%, but only for patients who took at least 80% of their doses

Blood pressure treatment lowered the rate of major cardiovascular events, a group that includes heart attacks and strokes, by 11% among trial participants who took at least 80% of their assigned medication, according to pooled data on 91,339 people from nine randomised trials. Among participants who took less than that, the researchers found no statistically significant reduction. The analysis was presented on Aug. 30 at the European Society of Cardiology’s annual congress by Qianqian Yang of the University of Oxford.

Stay ahead on AI: Join 600,000+ professionals who read The Deep View, the daily newsletter that keeps you up to date with artificial intelligence.

Subscribe to The Deep View →

We may earn a commission if you sign up.

Yang’s summary was blunt: drugs don’t work in patients who don’t take them. She said patients with lower adherence derived little or no cardiovascular benefit. The point of running the question through trial data, she explained, was that earlier evidence on missed doses came from observational studies that could be confounded by other factors, such as sicker or less engaged patients being the ones who skip tablets.

An Adherence-Stratified Look at Nine Randomised Trials

The work comes from the Blood Pressure Lowering Treatment Trialists’ Collaboration, led from Oxford by Professor Kazem Rahimi and Dr. Milad Nazarzadeh, and is an individual participant data meta-analysis, per the European Society of Cardiology’s press release of Aug. 27. Each of the nine trials compared a blood pressure regimen with placebo or a less intensive regimen. The release lists funding from the British Heart Foundation and discloses scholarship support to Yang from the China Scholarship Council and the China Oxford Scholarship Fund. ScienceDaily’s repost of the release appeared on Oct. 25, about two months after the congress session.

What kind of analysis is this? Not a per-protocol trial result in the usual sense; the release never uses that term. Participants were sorted after the fact by how much of the assigned treatment they took, with 80% as the cut-off, and the intervention was then compared with the comparator inside each adherence group. According to HCPLive’s account of the presentation, adherence was updated at each follow-up visit, blood pressure effects were estimated with linear mixed-effects models, clinical events with trial-stratified Cox models, and the authors said randomisation was preserved within adherence strata. Even so, patients are never randomised to adherence, so residual confounding between strata cannot be ruled out, a general caution the summaries do not list among limitations.

The 11% Reduction, the Intervals and the Blood Pressure Gap

The composite outcome covered stroke, heart attack, ischaemic heart disease, and heart failure causing hospitalisation or death. HCPLive reported a hazard ratio of 0.89 (95% confidence interval 0.86 to 0.92) in the higher-adherence group, the 11% reduction, against 1.01 (0.92 to 1.10) in the lower-adherence group, with a P value for interaction of 0.04. The analysis counted 15,460 major cardiovascular events and 9,591 deaths over a median follow-up of 4.7 years. Those intervals and totals appear in the trade-press account; the ESC release itself gives only the 11% and the absence of a significant effect.

Blood pressure itself shows the same split. Systolic pressure fell by 5.2 mmHg (95% confidence interval 5.0 to 5.4) in the higher-adherence group and by 3.0 mmHg (2.7 to 3.3) in the lower-adherence group, compared with the comparator arms. A 3.0 mmHg fall is not nothing, so “no benefit” is a statement about the event count not reaching significance, not proof that partial adherence does literally nothing. HealthDay’s Aug. 31 coverage carried the same finding as lower adherence being associated with no benefit for preventing cardiovascular events.

Adherence slid even inside trials, where participants are monitored. The share classed as higher adherence fell from 88% in year one to 79% by year five, per the ESC release. HCPLive broke that out by arm: 88% on the intervention and 87% on the comparator at one year, 80% and 77% at five years. The release adds that an estimated 1.4 billion adults aged 30 to 79 live with hypertension and that roughly half of patients do not take prescribed medicines as directed, so real-world adherence is probably lower than these figures.

The investigators and the ESC press adherence, not a new drug

Mahfoud said non-adherence has been neglected for years as a cause of uncontrolled hypertension. Yang described the finding as evidence that poor adherence to blood pressure medication is associated with worse outcomes, and noted that a decline in adherence over time was apparent even in the structured setting of trials.

Professor Felix Mahfoud, chair of the ESC Communication Committee, said assessment of adherence should become a routine part of hypertension care. According to News-Medical’s report, Yang’s group also called for better patient education and open, non-judgmental conversations about the barriers to taking medication, plus simpler dosing schedules, single-pill combinations and longer-acting drugs that limit the effect of a missed dose.

The limits are those of a congress presentation. The ESC release links only to the session page, not to a peer-reviewed paper, the 80% line is a binary cut-off, and the intervals above come second-hand from trade coverage. The adherence data came from nine of the collaboration’s trials; HCPLive reports the wider analysis covered 115,389 participants from 12 trials. The finding is best read as strong evidence that benefit depends on actually taking the drug, not as a prediction for any individual patient.

This article was produced with the assistance of AI and reviewed by Morning Overview editors prior to publication.

Stay ahead on AI: Subscribe to The Deep View →

We may earn a commission if you sign up.