The Global Hidradenitis Suppurativa Atlas Methodology: Combining Global Proportions in a Pooled Analysis

INTRODUCTION: Data concerning the global burden of hidradenitis suppurativa (HS) are limited. Reported prevalence estimates vary between 0.0003% and 4.1%, and data from various geographical regions are still to be collected. Previously reported prevalences have been limited by the methodological approach and source of data. This has resulted in great heterogeneity as prevalence data from physician-diagnosed cases poorly match those of self-reported apparent HS disease. METHODS: The Global Hidradenitis Suppurativa Atlas (GHiSA) introduces an innovative approach to determine the global prevalence of HS. This approach involves using a previously validated questionnaire to screen apparently healthy adults accompanying a patient to a non-dermatological outpatient clinic visit in a hospital or a private/family medicine clinic. The screening questionnaire (i.e., the index test) is combined with a subsequent physician-based in-person validation (i.e., the reference standard) of the participants who screen positive. Approximately ten percent of the screen-negative participants are also clinically assessed to verify the diagnostic precision of the test. The local prevalence (pi) will be estimated from each country that submits the number of patients who are HS positive according to the index test and clinical examination (n), and the corresponding total number of observations (N). CONCLUSION: The GHiSA Global Prevalence studies are currently running simultaneously in 58 countries across six continents (Africa, Europe, Australia, North America, South America, and Asia). The goal of the combined global proportion is the generation of a single summary (i.e., proportional meta-analysis), which will be done after a logit transformation and synthesized using a random-effects model. The novel standardization of the Global Prevalence Studies conducted through GHiSA enables direct international comparisons, which were previously not possible due to substantial heterogeneity in past HS prevalence studies.