Do these data apply to me? Examining the applicability of trials assessing strategies for optimal management of blood pressure to older patients in UK primary care

Conference:

SAPC ASM 2019 - Exeter

Talk Code:

1B.7a

Presenter:

Mark Lown

Twitter:

@jamessheppard48

Co-authors:

James Sheppard, Jenni Burt, Mark Lown, Eleanor Temple, Rebecca Lowe, Hannah Ashby, Oliver Todd, Bethany Diment, Sarah Oliver, Julie Allen, Gary Ford, Carl Heneghan, FD Richard Hobbs, Sue Jowett, Paul Little, Jonathan Mant, Jill Mollison, Rupert Payne, Marney Williams, Ly-Mee Yu, Richard McManus

Author institutions:

University of Oxford, University of Cambridge, University of Southampton, University of Leeds, University of Leeds, University of Bristol

Problem

There is debate as to what extent older patients (≥80 years) should be treated for high blood pressure. Existing trials show that blood pressure lowering in this population is effective at preventing stroke and heart failure but also results in an increased risk of adverse events. However, it has been suggested that these studies enrolled healthier patients, who are less representative of the general population and more likely to benefit from treatment. This study aimed to compare the characteristics of patients eligible for three blood pressure management trials and assess the likelihood of eligibility for each trial based on common characteristics of older patients.

Approach

Cross-sectional study of data extracted from the medical records of 15,376 patients aged ≥80 years, registered to 24 general practices in the south of England. Anonymised patient data relating to the eligibility criteria for two previous medication intensification trials (HYVET, SPRINT) and one medication reduction trial (OPTiMISE) were extracted. Patients eligible for each trial were defined according to criteria specified in each trial protocol. Descriptive statistics were used to define the characteristics of each trial population. A logistic regression model was constructed to estimate predictors of eligibility for each trial, with practice included as a random effect.

Findings

Approximately 268 (1.7%), 5,290 (34.4%) and 3,940 (25.6%) patients were eligible for HYVET, SPRINT and OPTiMISE trials respectively. There was little overlap in eligibility for each trial (1.0% were eligible for HYVET and SPRINT; 0% were eligible for HYVET and OPTiMISE; 10.2% were eligible for SPRINT and OPTiMISE). Patients eligible for OPTiMISE were comparable to the general population in terms of frailty (eFI 0.12 [OPTiMISE] vs 0.11 [general population]), but had more morbidities (4 vs 3) and cardiovascular medications prescribed (4 vs 2). Patients in HYVET and SPRINT were less frail, multi-morbid and prescribed less cardiovascular medications. Overall, increasing frailty and a history of cardiovascular disease reduced the likelihood of being eligible for any trial.

Consequences

Patients eligible for OPTiMISE appear to best represent the population aged ≥80 years attending UK primary care. Increasing frailty and/or multi-morbidity reduce the likelihood of eligibility for all three blood pressure trials. Caution should be exercised when applying the results from randomised controlled trials to management of blood pressure in frail and multi-morbid patients.

Submitted by:

James Sheppard

Funding acknowledgement:

This work was funded by the NIHR Oxford CLAHRC and School for Primary Care Research