Research overview
The text below is the source article’s original abstract under a verified reuse license. cc by
BackgroundAlzheimer's disease and related dementias (ADRD) and Parkinson's disease (PD), pose growing global health challenges. Socio-demographic and economic development acts paradoxically, complicating the process that determines how governments worldwide designate policies and allocate resources for healthcare.MethodsWe extracted data on ADRD and PD in 204 countries from the Global Burden of Disease 2019 database. Health disparities were estimated using the slope index of inequality (SII), and concentration index (CIX) based on the socio-demographic index. Estimated annual percentage changes (EAPCs) were employed to evaluate temporal trends.ResultsGlobally, the SII increased from 255.4 [95% confidence interval (CI), 215.2 to 295.5)] in 1990 to 559.3 (95% CI, 497.2 to 621.3) in 2019 for ADRD, and grew from 66.0 (95% CI, 54.9 to 77.2) in 1990 to 132.5 (95% CI, 118.1 to 147.0) in 2019 for PD; CIX rose from 33.7 (95% CI, 25.8 to 41.6) in 1990 to 36.9 (95% CI, 27.8 to 46.1) in 2019 for ADRD, and expanded from 22.2 (95% CI, 21.3 to 23.0) in 1990 to 29.0 (95% CI, 27.8 to 30.3) in 2019 for PD. Age-standardized disability-adjusted life years displayed considerable upward trends for ADRD [EAPC = 0.43 (95% CI, 0.27 to 0.59)] and PD [0.34 (95% CI, 0.29 to 0.38)].ConclusionsGlobally, the burden of ADRD and PD continues to increase with growing health disparities. Variations in health inequalities and the impact of socioeconomic development on disease trends underscored the need for targeted policies and strategies, with heightened awareness, preventive measures, and active management of risk factors.
Scientific assessment
Research question
Assess the question within the original study’s scope.
Methods & sample
Check the source article for study design and sample details.
Key findings
An editorially approved findings summary is not yet published for this record.
Limitations
Generalizability, bias and clinical meaning require separate assessment.
Methods, samples and clinical conclusions are not inferred from bibliographic metadata.