Malaria pattern observed in the highland fringe of Butajira, Southern Ethiopia: a ten-year retrospective analysis from parasitological and metrological data.

BACKGROUND

Studying the magnitude of highland malaria is necessary to implement effective control measures in highland fringes of Ethiopia. Since repeated epidemics were reported in Butajira, this study hypothesized autochthonous transmission in the highland fringes of Butajira. Thus we aimed to determine the malaria occurrence and its association with meteorological variables in the highland fringe of Butajira, Southern Ethiopia from parasitological and metrological data.

METHODS

Retrospective monthly malaria case data was collected from monthly outpatient morbidity reports of the Butajira Health Center, for January 2000 to December 2009. Monthly total rainfall and average temperature (maximum and minimum), which was recorded in Butajira weather station, was obtained from National Meteorological Agency, for the same period. Spearman correlation coefficient was used to quantify the strength of linear relationships between meteorological variables and malaria cases. The effect of each independent variable on malaria cases was assessed using multiple linear regression.

RESULTS

During the study period, a fluctuating trend of malaria cases was observed with (62.5%) dominancy. The highest occurrence of malaria was recorded in 2003, 2004, 2008 and 2009 and the least was recorded in 2000-2002. None of the meteorological variables was positively correlated with monthly malaria cases at zero months lag. However, minimum temperature was positively correlated with total malaria cases, and with one month lag. Correlation analysis showed that all of the meteorological variables, except maximum temperature at one month lag, were not significant with total monthly malaria cases and each species of malaria (P-value >0.05) at zero and one month lag effect. Nevertheless, after adjusting for the effect of other variables the linear regression analysis indicated that cumulative monthly rainfall (Beta= -0.24; 95% CI: -0.47, -0.02) at zero months lag and maximum temperature (Beta= -12.13; 95% CI: -23.43, -0.82) at one month lag had a significant negative effect on the total malaria count.

CONCLUSION

There was no significant association between malaria occurrences and meteorological variables between January 2000 and December 2009; therefore non-climatic factors together with climatic variables should be assessed to know the spread and intensity of malaria in the highland fringe of Butajira. This report also warrants the Ministry of Health to include highland areas in its current malaria controlling campaign so as to address those non-endemic foci of the country.