BPLTTC

Systematic reviews of observational studies have quantified regression to the mean in the general population and showed that usual BP has a much narrower distribution than baseline BP. This project quantifies the contribution of regression to the mean to long-term BP changes in clinical trials, and assesses treatment effects by control group usual BP.

The largest tabular meta-analysis of BP trials has recently suggested similar proportional benefits of BP lowering at different baseline trial-level BP for a range of CVD outcomes but uncertainty remains whether this is true for all patient subgroups. Furthermore, tabular meta-analyses are unable stratify by baseline BP at an individual level and to control for competing events. A detailed IPD meta-analysis will provide a more in depth examination of BP lowering at different baseline levels results and the different proportional effects on CVD outcomes. Furthermore, this will help to ascertain whether certain drug classes have particularly strong positive or negative effects in specific patient subgroups.

Although some BP lowering trials have shown that more intensive lowering is associated higher risk of SAEs as less intensive lower. However, this has not been investigated in detail across the whole spectrums of BP levels and by other important patient features. A detailed IPD meta-analysis will help to address the question of safety of BP lowering overall, by BP levels at baseline, age and disease status.

Observational evidence suggests BP lowering is associated with a lower risk of non-vascular mortality. However, whether this is causal or due to event misclassification is uncertain. Conflicting meta-analyses have suggested angiotensin receptor blockers increase the risk of cancer and have no effect on cancer. A large-scale IPD meta-analysis will help to provide important evidence on BP lowering effects on cancer risk overall and by drug classes.

Observational studies have shown a clear and strong association between elevated BP and future risk of diabetes. Meta-analysis of randomised trials, on the other hand, have indicated that beta-blocker and diuretic use is associated with an increased risk of new onset diabetes and that RAS inhibitor use is associated with a decreased risk of new onset diabetes. However, it is unclear if BP lowering (with either RAS inhibitor or CCB) reduces the risk of new onset diabetes or if prevention of incident diabetes is a class effect.

Some observational analyses have suggested that initiation of BP lowering therapy is associated with an increased risk of fractures. However, these analyses are susceptible to confounding (sicker elderly patients being initiated on BP lowering medication). In contrast, an analysis of HYVET indicated that BP lowering in the elderly reduced the rate of fractures. An IPD meta-analysis will investigate the effect of antihypertensive therapy on falls/fractures in more detail.

Observational evidence suggests elevated BP is associated with an increased risk of dementia. However, a meta-analysis of randomised trials found no effect of BP lowering on risk of dementia, but was underpowered to detect a difference. In this project, we explore the effect of BP lowering on risk of dementia.

Elevated BP is a risk factor for incident atrial fibrillation (AF). However, in a tabular meta-analysis, BP lowering was not associated with a decreased risk of AF outside heart failure patients. This meta-analysis was limited by the small number of incident AF events in non-heart disease populations. Use of individual patient data with more AF events will allow for a reliable determination of whether BP lowering reduces the risk of AF.