Diagnostic test accuracy of outpatient blood pressure measurement to detect hypertension in older people
Abstract
Introduction
Older people presenting to outpatient (OP) clinics require blood pressure (BP) measurement to consider morbidity contributing to falls and long-term cardiovascular risk. However, BP measurements taken in the outpatient setting (OBPs) may not correlate with BP patterns obtained using ambulatory measurements (ABPMs). ABPM is considered a more accurate method of diagnosing hypertension than OBP in the general population. Using OBPs for clinical decision-making in isolation risks either under or over-diagnosis of hypertension. This study aimed to investigate the diagnostic test accuracy of OBP to detect hypertension in a cohort of older people.
Method
We retrospectively evaluated consecutive attendances of older people to a multidisciplinary outpatient clinic over a period of four months. We obtained demographic and clinical data from electronic records and compared individual OBPs with subsequent ABPMs. We defined hypertension using NICE criteria and followed the “Standards for Reporting of Diagnostic Accuracy” (STARD) to categorise data.
Results
Among 208 consecutive OP attendances, we identified 194 patients with OBPs, of whom 65 subsequently underwent ambulatory blood pressure monitoring (ABPM). The mean age of attendees was 76 years, 65% were female, and 49% had a history of hypertension. 121(62%) patients were hypertensive on OBP – of the 35 who progressed to ABPM, only 14 were hypertensive (a positive predictive value (PPV) of 40%). Seventeen patients without hypertension on OBP had ABPM, of whom three were hypertensive on ABPM (negative predictive value (NPV) of 82%).
Conclusion
Older people who are normotensive on OBP are likely also to be normotensive on ABPM. However, the majority who appear to be hypertensive on OBP are not found to be hypertensive on ABPM. We conclude that OBP has low diagnostic accuracy in this population, risking significant overdiagnosis of hypertension. Obtaining ABPM for older people with hypertension on OBP is likely to aid clinical decision-making.
Comments
The data showing that 21 out…
The data showing that 21 out of 35 patients (60%) with high OBP were actually normotensive on ABPM is quite striking. Does this suggest that 'White Coat Hypertension' is the primary driver of overdiagnosis in this older population, and how can we practically increase ABPM access to avoid unnecessary medication?
Thank you for your comment…
Thank you for your comment. The ‘white coat effect’ is definitely a key factor in the misdiagnosis of hypertension and this effect is more common in older people. There are also lots of individual patient factors which on the day of blood pressure measurement may affect the result and not represent their blood pressure control accurately. In terms of improving ABPM access, there may be a need for more of a push towards using ABPM before starting or adjusting medication but the practicalities of this are difficult and dependent on local services.
WCH
I agree with Conor that this discrepancy is most likely due to WCH. We hope that this evidence could be used to justify increased provision of ABPM given the likelihood of cost savings on antihypertensives and (potentially) admissions due to falls (although clearly this is extrapolation). Practically we hope it will give practitioners more confidence in the utility of ABPM, including in primary care. Previous studies have suggested that hospital grade ABPM is superior to patient managed home BP measurements