Fifteen RCTs were included in the review (9,111 patients; range 40 to 6,963). Of the 15 RCTs, approximately 85% reported using appropriate sequence generation methods, 35% concealed allocation, 15% blinded outcome assessors, 85% were free from reporting bias, and 95% were free from other biases. Duration of follow-up ranged from four to 24 months.
The use of pharmacist care resulted in significantly greater reductions in systolic blood pressure (WMD -6.2mmHg, 95% CI -7.8 to -4.6; 12 RCTs; Ι²=2%), diastolic blood pressure (WMD -4.5mmHg, 95% CI -6.2 to -2.8; nine RCTs; Ι²=46%), total cholesterol (WMD -15.2mg/dL, 95% CI -24.7 to -5.7; eight RCTs; Ι²=75%), low-density lipoprotein cholesterol (WMD -11.7mg/dL, 95% CI -15.8 to -7.6; nine RCTs; Ι²=41%), and body mass index (WMD -0.9kg/m², 95% CI -1.7 to -0.1; five RCTs; Ι²=92%), but not high-density lipoprotein cholesterol (six RCTs).
Results of post hoc subgroup analyses and sensitivity analyses were also reported.
Results of the test for publication bias showed that bias could be present.