Ashenafi Assefa^1,2*, Ahmed Ali², Wakgari Deressa², Heven Sime¹, Hussien Mohammed¹, Amha Kebede³, Hiwot Teka⁴, Hiwot Solomon⁵, Kevin Gurrala⁶,  Brian Matei⁶, Richard Maud⁷, Ipsita Sinha⁷, Jackie Cook⁸, Ruth Ashton⁹, Glenn Wilson¹⁰, Brian Wakeman⁶, Ya Ping Shi⁶, Lorenz Von Seidlein⁷, Chris Drakeley⁸, Eric Rogier⁶, Jimee Hwang^11,12

¹Ethiopian Public Health Institute, Addis Ababa, Ethiopia

²Addis Ababa University, Addis Ababa, Ethiopia

³African Society for Laboratory Medicine, Addis Ababa, Ethiopia

⁴U.S. President’s Malaria Initiative, United States Agency for International Development, Addis Ababa, Ethiopia

⁵Ethiopian Federal Ministry of Health, Addis Ababa, Ethiopia

⁶Malaria Branch, Division of Parasitic Diseases and Malaria, Centers for Disease Control and Prevention, Atlanta, GA

⁷Mahidol Oxford Research Unit, Mahidol University, Bangkok, Thailand

⁸London School of Hygiene and Tropical Medicine, UK

⁹Center for Applied Malaria Research and Evaluation, Tulane School of Public Health and Tropical Medicine, New Orleans, LA

¹⁰University of Southern Denmark, Odense, Denmark

¹¹U.S. President’s Malaria Initiative, Malaria Branch, Division of Parasitic Diseases and Malaria, Centers for Disease Control and Prevention, Atlanta, GA

¹²Global Health Group, University of California San Francisco, CA

*Corresponding author

Abstract

Background

Measures of malaria burden using conventional diagnostic methods in cross-sectional household surveys may incompletely describe the burden of malaria transmission in low transmission settings where few infections are detected. This study describes the burden and pattern of malaria transmission in Ethiopia using serological estimates derived from a nationwide household survey completed in 2015.

Methods

Dried blood spot (DBS) samples were collected during a Malaria Indicator Survey in 2015 representing malarious areas of Ethiopia.  Samples were analysed using bead-based multiplex assays for IgG antibodies to six Plasmodium antigens; four human malaria species-specific merozoite surface protein 1 19kD antigens (MSP-1), and apical membrane antigen 1 (AMA-1) for P. falciparum and P. vivax. Seroconversion rate was estimated using a reversible catalytic model fitted with maximum likelihood methods.  Models were stratified by age group, elevation and administrative regions.

Results

Of 9,801 DBS samples, 91.3 (8,944) had valid serological results for analysis. The mean age of participants was 14.9 years (95% confidence interval (CI): 14.5-15.3); 53.3% (4,748)  participants were female. National seroprevalence for antibodies to P. falciparum was 35.7% (95% CI: 34.6-36.7) and antibodies to P. vivax was 23.8% (95% CI: 22.9-24.7). Estimated seroprevalence for P. malariae and P. ovale were 3.2% (95% CI: 6.6-7.7) and 2.6% (95% CI: 2.2-2.9), respectively. Seroprevalence estimates were significantly higher at lower elevations (<2000m) compared to higher elevations (2000-2500m) for both P. falciparum [Odds Ratio (OR) 1.7 (95% CI: 1.5-1.9), p=0.001], however evidences were weak for P. vivax [OR 1.2 (95% CI: 1.0-1.4)p=0.037]. Among administrative regions, P. falciparum seroprevalence ranged from 64.1% (95% CI: 60.6-67.6) for Gambella to 11.1% (95% CI: 9.0-13.7) in Somali Region. P. vivax seroprevalence ranged from 39.4% (95% CI: 36.1-42.7) in Amhara to 3.9% (95% CI: 2.7-55.9) in Somali Region. Models fitted to measure seroconversion rates show variation nationally, by elevation with variability by regions, antigen type and within species.

Conclusion

Using multiplex serology assays for IgG antibody detection, this study explored the cumulative malaria burden and regional dynamics in Ethiopia and the use of multiplex assay on multiple antigens in a low transmission setting as a more sensitive biomarker.

Key word: Multiplex Serology, seroprevalence, Malaria, Ethiopia