Gross domestic product and health expenditure associated with incidence, 30-day fatality, and age at stroke onset: a systematic review.

BACKGROUND AND PURPOSE Differences in definitions of socioeconomic status and between study designs hinder their comparability across countries. We aimed to analyze the correlation between 3 widely used macrosocioeconomic status indicators and clinical outcomes. METHODS We selected population-based studies reporting incident stroke risk and/or 30-day case-fatality according to prespecified criteria. We used 3 macrosocioeconomic status indicators that are consistently defined by international agencies: per capita gross domestic product adjusted for purchasing power parity, total health expenditures per capita at purchasing power parity, and unemployment rate. We examined the correlation of each macrosocioeconomic status indicator with incident risk of stroke, 30-day case-fatality, proportion of hemorrhagic strokes, and age at stroke onset. RESULTS Twenty-three articles comprising 30 population-based studies fulfilled the eligibility criteria. Age-adjusted incident risk of stroke using the standardized World Health Organization World population was associated to lower per capita gross domestic product adjusted for purchasing power parity (ρ=-0.661, P=0.027, R(2)=0.32) and total health expenditures per capita at purchasing power parity (ρ=-0.623, P=0.040, R(2)=0.26). Thirty-day case-fatality rates and proportion of hemorrhagic strokes were also related to lower per capita gross domestic product adjusted for purchasing power parity and total health expenditures per capita at purchasing power parity. Moreover, stroke occurred at a younger age in populations with low per capita gross domestic product adjusted for purchasing power parity and total health expenditures per capita at purchasing power parity. There was no correlation between unemployment rates and outcome measures. CONCLUSIONS Lower per capita gross domestic product adjusted for purchasing power parity and total health expenditures per capita at purchasing power parity were associated with higher incident risk of stroke, higher case-fatality, a greater proportion of hemorrhagic strokes, and lower age at stroke onset. As a result, these macrosocioeconomic status indicators may be used as proxy measures of quality of primary prevention and acute care and considered as important factors for developing strategies aimed at improving worldwide stroke care.