ABSTRACT
BACKGROUND: The magnitude of the 'white coat effect', the alerting rise in blood pressure, is greater for doctors than nurses. This could bias interpretation of studies on nurse-led care in hypertension, and risks overestimating or overtreating high blood pressure by doctors in clinical practice. AIM: To quantify differences between blood pressure measurements made by doctors and nurses. DESIGN AND SETTING: Systematic review and meta-analysis using searches of MEDLINE, CENTRAL, CINAHL, Embase, journal collections, and conference abstracts. METHOD: Studies in adults reporting mean blood pressures measured by doctors and nurses at the same visit were selected, and mean blood pressures extracted, by two reviewers. Study risk of bias was assessed using modified Cochrane criteria. Outcomes were pooled across studies using random effects meta-analysis. RESULTS: In total, 15 studies (11 hypertensive; four mixed hypertensive and normotensive populations) were included from 1899 unique citations. Compared with doctors' measurements, nurse-measured blood pressures were lower (weighted mean differences: systolic -7.0 [95% confidence interval {CI} = -4.7 to -9.2] mmHg, diastolic -3.8 [95% CI = -2.2 to -5.4] mmHg). For studies at low risk of bias, differences were lower: systolic -4.6 (95% CI = -1.9 to -7.3) mmHg; diastolic -1.7 (95% CI = -0.1 to -3.2) mmHg. White coat hypertension was diagnosed more frequently based on doctors' than on nurses' readings: relative risk 1.6 (95% CI =1.2 to 2.1). CONCLUSIONS: The white coat effect is smaller for blood pressure measurements made by nurses than by doctors. This systematic difference has implications for hypertension diagnosis and management. Caution is required in pooling data from studies using both nurse- and doctor-measured blood pressures.