Disease-Related Costs Published in The Middle East and North Africa Region: Systematic Review and Analysis of Transferability

• Published in: PharmacoEconomics Vol. 40; no. 6; pp. 587 - 599
• Main Authors: Zrubka, Zsombor, Péntek, Márta, Mhanna, Lea, Abu-Zahra, Teebah, Mahdi-Abid, Mohamed, Fgaier, Meriem, El-Dahiyat, Faris, Al-Abdulkarim, Hana, Drummond, Michael, Gulácsi, László
• Format: Journal Article
• Language: English
• Published: Cham Springer International Publishing 01.06.2022; Springer; Springer Nature B.V
• Subjects: Africa, Northern; Comparative analysis; Cost estimates; Cost of Illness; Data Collection; Decision making; Disease; Economic analysis; GDP; Gross Domestic Product; Health Administration; Health care expenditures; Health Economics; Humans; International aspects; Macroeconomics; Medical care, Cost of; Medical research; Medicine; Medicine & Public Health; Middle East; Per capita; Pharmacoeconomics and Health Outcomes; Public Health; Quality of Life Research; Systematic Review; Africa, Northern; Middle East; Northern Africa; Egypt; North Africa; United Arab Emirates; Libya; Bahrain; Syria; Oman; Saudi Arabia; Lebanon; Morocco; Qatar; Algeria; MENA countries; Iraq; Kuwait; Tunisia; Yemen
• ISSN: 1170-7690; 1179-2027; 1179-2027
• DOI: 10.1007/s40273-022-01146-6

Background In the Middle East and North Africa (MENA) the scarcity of local cost data is a key barrier to conducting health economic evaluations. We systematically reviewed reports of disease-related costs from MENA and analysed their transferability within the region. Methods We searched PubMed and included full text English papers that reported disease-related costs from the local populations of Algeria, Bahrain, Egypt, Iraq, Jordan, Saudi Arabia, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Syria, Tunisia, United Arab Emirates and Yemen between 1995 and 2019. Screening, study selection and data extraction were done in duplicate. Study-related variables, costing methods, all costs and their characteristics were extracted and analysed via descriptive methods. From multi-country studies of MENA employing homogenous costing methods, we estimated the ratio (cost transfer coefficient) between the relative differences in direct medical costs and macroeconomic indicators via robust regression. We predicted each cost via the estimated cost transfer formula and evaluated prediction error between true and predicted (transferred) costs. Results The search yielded 1646 records, 206 full text papers and 3525 costs from 84 diagnoses. Transferability was analysed involving 144 direct medical costs from eight multi-country studies. Adjusting the average of available foreign costs by 0.28 times the relative difference in GDP per capita provided the most accurate estimates. The correlation between true and predicted costs was 0.96; 68% of predicted costs fell in the true ± 50% range. Predictions were more accurate for costs from studies that involved the largest number of countries, for countries outside the Gulf region and for drug costs versus unit or disease costs. Conclusion The estimated cost transfer formula allows the prediction of missing costs in MENA if only GDP per capita is available for adjustment to the local setting. Input costs for the formula should be collected from multiple sources and match the decision situation. Plain Language Summary In the Middle East and North Africa (MENA) scarce local cost data hinder health economic evaluations. This systematic review summarized disease-related costs from 17 countries (Algeria, Bahrain, Egypt, Iraq, Jordan, Saudi Arabia, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Syria, Tunisia, United Arab Emirates and Yemen). Eight studies applied the same costing method across multiple countries. We used these data to estimate a formula for transferring costs between countries. We assumed that costs vary proportionally with gross domestic product per capita (GDP). Most accurate cost predictions were provided when relative cost differences were set to 0.28-times the relative differences in GDP per capita. The correlation between transferred and true costs was very high. Still, only 68% of transferred costs fell in the true ± 50% range. Cost estimates were more accurate if costs were transferred from multiple countries. Also, estimates were more accurate for countries outside the Gulf region and for drug costs when compared to unit- or disease costs.