Efficacy and effectiveness of valsartan/amlodipine and valsartan/amlodipine/hydrochlorothiazide in hypertension: randomized controlled versus observational studies

Objective: The aim of this post-hoc analysis was to compare the results from randomized controlled trials (RCTs) and real-world evidence (RWE) studies of valsartan/amlodipine (Val/Aml) and valsartan/amlodipine/hydrochlorothiazide (Val/Aml/HCTZ) in patients with uncontrolled hypertension (>140/90mmHg). Methods: Data was pooled from 15 RCTs (N=5542) and 8 RWE studies (N=1397) for Val/Aml; and 2 RCTs (N=804) and 5 RWE studies (N=9380) for Val/Aml/HCTZ. Patients who received Val/Aml (80/5, 160/5, 160/10, 320/5, or 320/10mg), Val/Aml/HCTZ (160/5/12.5, 160/5/25, 160/10/12.5, 160/10/25, or 320/10/25mg) or placebo were considered for this analysis. Only patients with both baseline and follow-up assessment within 60-90 days after baseline had been included in the analysis. Patients with missing values were excluded from the analysis. Using fitted linear mixed-effects model and random factors, treatment interactions and study design with mean sitting systolic blood pressure (msSBP), diastolic BP (msDBP) and pulse pressure (msPP) reductions from baseline to Week 8-12 of treatment were compared. Results: Baseline demographics and patient characteristics were comparable between RCT and RWE datasets and within Val/Aml and Val/Aml/HCTZ treatment groups. In both RCT and RWE studies, least-squares mean (LSM) reduction in msSBP/msDBP and msPP from baseline were significant (p<.05) across all dosages. The efficacy of Val/Aml in RCTs was statistically significantly greater than in RWE studies for msSBP/msDBP ( 23.1/ 13.8 vs. 17.9/-9.1mmHg) but the difference was non-significant for msPP (-8.6 vs. -9.3mmHg; p = .77). For Val/Aml/HCTZ, no direct comparison was available but a similar trend was observed. The difference observed for msSBP and msDBP may be due to routine practice setting, larger populations may have more confounders and different behaviors towards treatment adherence. Conclusion: These findings demonstrate that the efficacy of Val/Aml and Val/Aml/HCTZ in RCTs was more pronounced compared with their effectiveness in RWE studies in different ethnic populations although the overall benefit was not different.