Thirty two RCTs (n=18,481, range 21 to 5,005) were included in the review; seven were cluster RCTs. Study quality was moderate: 70% of studies adequately reported randomisation; 58% reported allocation concealment; and 43% reported blinding. Sensitivity analysis that pooled only good-quality RCTs did not substantially alter the results.
Nurse interventions that used a treatment algorithm (14 RCTs):
There was a greater reduction in blood pressure from baseline in patients who received treatment guided by an algorithm compared with usual care (WMD -8.2mmHg, 95% CI -11.5 to -4.9; four RCTs), but no statistically significant difference between groups in achievement of study blood pressure targets (three RCTs).
Nurse prescribing (nine RCTs):
There was a greater reduction in blood pressure from baseline with nurse prescribing compared to usual care (systolic WMD -8.9mmHg, 95% CI -12.5 to -5.3; three RCTs and diastolic WMD -4.0, 95% CI -5.3 to -2.7; four RCTs), but no statistically significant difference in achievement of study blood pressure target (two RCTs).
Telephone monitoring of blood pressure by nurses (seven RCTs):
There was a statistically significant decrease in diastolic blood pressure from baseline (WMD -2.1, 95 %CI -4.1 to -0.3; three RCTs), but not for other blood pressure measurements in patients who received telephone monitoring of blood pressure by nurses compared to usual care. There was a statistically significant benefit in achievement of study blood pressure targets with telephone monitoring (RR 1.24, 95 % CI 1.08 to 1.43; three RCTs).
Community monitoring (eight RCTs):
Interventions delivered outside healthcare settings (such as community centres and work settings) showed a statistically significant reduction in blood pressure from baseline compared to usual care (systolic WMD -4.8mmHg, 95% CI -7.0 to -2.7; four RCTs and diastolic WMD -3.5, 95% CI -4.5 to -2.5; four RCTs).
Nurse-led clinics (14 RCTs):
In primary care clinics there was a statistically greater reduction in blood pressure for nurse-led clinics compared with usual care (systolic WMD -3.5mmHg, 95 % CI -5.9 to -1.1; six RCTs and diastolic WMD -1.9mmHg, 95 % CI -3.4 to -0.5; six RCTs). There were no statistically significant differences between treatment groups in achievement of study blood pressure targets (two RCTs) and there was some evidence of statistical heterogeneity (I2=72%). In secondary care clinics there were no statistical differences in blood pressure or achievement of blood pressure targets (I2=65%).
Findings for subgroup analyses by ethnic group were reported in the review.