Abstract INTRODUCTION The burden of central nervous system tumors is unevenly distributed globally. In Africa, this burden is further compounded by a lack of cancer-focused health system strengthening. The authors sought to quantify these system-level challenges with a novel index. METHODS This prototype index was developed from publicly available data provided by the World Health Organization’s Global Health Observatory, the World Bank Group’s Country Database, and the Institute For Health Metrics and Evaluation’s Global Burden of Disease database. Metrics were selected based on their ability to represent health system components including workforce, service delivery, infrastructure, information management, health financing, and governance. The metrics were normalized either using standardization or min-max scaling. Next, they were weighted iteratively based on importance, relative impact, and expert judgment. Also, the weighted metrics were grouped into facilitators or obstructors. The difference between obstructors and facilitators for each country was percentile ranked to obtain the index score between 0 and 100. Finally, sensitivity analysis was conducted to test best- and worst-case scenarios based on uncertainty values. The t-test and ANOVA were used to evaluate association between country subregion/income and the index. RESULTS Data was found for 51 African countries. Ethiopia (98), Madagascar (96), and Mali (94) had the highest ANAD Index score whereas Gabon (0), Seychelles (2), and Mauritius (4) had the lowest scores. North African countries had lower ANAD Index scores than sub-Saharan African countries (mean (SD) 19 (4) vs. 54 (4), p=0.005). Also, low-income countries had higher ANAD Index scores than lower-middle- and upper-middle-income countries (72 (19) vs. 37 (27) vs. 9 (12), p<0.001). CONCLUSION This prototype will be further tested for face, construct, content, and criterion validity. The authors anticipate that expanding this metric globally will improve its validity while facilitating advocacy and allowing for impact assessment at the system level.
Read full abstract