Study Heterogeneity and Estimation of Prevalence of Primary Aldosteronism: A Systematic Review and Meta-Regression Analysis

Abstract

CONTEXT: For health care planning and allocation of resources, realistic estimation of the prevalence of primary aldosteronism is necessary. Reported prevalences of primary aldosteronism are highly variable, possibly due to study heterogeneity. OBJECTIVE: Our objective was to identify and explain heterogeneity in studies that aimed to establish the prevalence of primary aldosteronism in hypertensive patients. DATA SOURCES: PubMed, EMBASE, Web of Science, Cochrane Library, and reference lists from January 1, 1990, to January 31, 2015, were used as data sources. STUDY SELECTION: Description of an adult hypertensive patient population with confirmed diagnosis of primary aldosteronism was included in this study. DATA EXTRACTION: Dual extraction and quality assessment were the forms of data extraction. DATA SYNTHESIS: Thirty-nine studies provided data on 42 510 patients (nine studies, 5896 patients from primary care). Prevalence estimates varied from 3.2% to 12.7% in primary care and from 1% to 29.8% in referral centers. Heterogeneity was too high to establish point estimates (I(2) = 57.6% in primary care; 97.1% in referral centers). Meta-regression analysis showed higher prevalences in studies 1) published after 2000, 2) from Australia, 3) aimed at assessing prevalence of secondary hypertension, 4) that were retrospective, 5) that selected consecutive patients, and 6) not using a screening test. All studies had minor or major flaws. CONCLUSIONS: This study demonstrates that it is pointless to claim low or high prevalence of primary aldosteronism based on published reports. Because of the significant impact of a diagnosis of primary aldosteronism on health care resources and the necessary facilities, our findings urge for a prevalence study whose design takes into account the factors identified in the meta-regression analysis.