Study design and setting
Unmatched 1:1 case control study was conducted in Argoba district, which is one of the districts in south Wollo Zone, Amhara region, from June 21 to July 7, 2016. The district is divided in to 7 kebele (the smallest administrative division) with a total population of 39,873. The altitudes of the district were 300-2100 Meter above sea level (90% Kola, 5% Woina Dega and 5% Dega) with annual rain fall of 500-750 mmHg and the temperature of 29-39.50 degree centigrade. It has 2 health centers and 7 health posts with 12 health extension workers which are currently on service (Fig 1).
Population and sample
All individuals in the district were considered as a source population. The required sample size was determined by using Epi-Info version 7 Statistical software using power approach formula, considering the 95% confidence level and 80% power (15). However, there were a total of 75 cases during the study period. 75 case and 75 controls were included in the study. Cases were taken from treatment registration book/line listing and neighborhoods to the cases were taken as control, lottery method was applied to select individual participants in the selected household for interview.
Data collection and variable of the study
Data was collected by using structured questionnaire by two health extension workers and two laboratory technicians under the supervision of the principal investigator and two BSC nurses. The questionnaire used in the study was developed for this particular study, and it was developed by reviewing literatures. One day training was given for data collectors and supervisors to maintain the data quality. Review of weekly Integrated Disease surveillance and Response (IDSR) report at different level (district health office and Health facilities), visiting the affected kebele, house to house visits were made to search for additional cases. Both Rapid Diagnostic Test (RDT) and Microscopic laboratory diagnosis test were also performed. A line listing of the cases were done.
Health extension workers used RDTS to identify confirmed malaria cases at health post and community level during outbreak investigation. Laboratory technicians used microscope to detect malaria parasites at Medina Health Center of Argoba district. Additionally, RDTS were also used in this health center during the interruption of electric power.
Data was also collected about the mosquito breeding sites from the district health office and health facilities. Cases/patients and controls were interviewed about presence of mosquito breeding sites around/near to their home. Similarly, observation on these potential mosquito breeding sites was made by health extension workers.
The dependent variable, presence of malaria (cases), was defined as an acute febrile illness with a peripheral blood smear positive for malaria parasite or a positive rapid antigen test by the RDTs. Whereas, controls were defined as when there were no signs and symptoms of acute febrile illness one month prior to the data collection. The independent variables were socio-demographic and economic factors (age, sex, occupation, educational status, and religion), behavioral factors (ITN utilization, Dichlorodiphenyltrichloroethane (DDT) sprays of households, sleeping area during night) and human and environmental factors.
Confirm the diagnosis and verify the existence of the outbreak
Previous five years malaria data was reviewed from Argoba district health office and health facility to set epidemic threshold level. However, due to incompleteness of the previous five years data, last year’s (2015/16) weekly malaria cases report was used to set epidemic threshold level by comparing weekly data with similar week of 2016/17, an epidemic curve was constructed. An increase above the weekly threshold was observed, it implies an epidemic has been occurred. Blood sample were taken from 20 fever cases and tested by microscopy.
Receiving rumor and prepared for field
An early warning alarm was reported on June 19, 2016 from Argoba district. Following the request from the district health office and analysis of routine surveillance data, Zonal PHEM case team decided to investigate/ confirm the outbreak and to take intervention. Checklist/semi-structured questioner was prepared and necessary logistics were collected from zonal health department. The team were arrived to district on June 20, 2016 and discussed with district health office and RRT about the reported malaria outbreak rumor.
Laboratory methods
Laboratory technicians used microscope to detect malaria parasites at Medina Health Center of Argoba district. Additionally, RDTS were also used in this health center during the interruption of electric power. Health extension workers used RDTS to identify confirmed malaria cases at health post and Community level during outbreak investigation.
Environmental assessment
Data was collected about the mosquito breeding sites from the district health office and health facilities. Cases/patients and controls were interviewed about presence of mosquito breeding sites in their compound and near to home. Similarly, observation of these potential mosquito breeding sites and presence of anopheles larvae in stagnant water was conducted.
Data processing and analysis
Data was entered using Epi-info version 7 statistical software and exported to Statistical Package for Social Science (SPSS) version 20 for analysis. Descriptive and summary statistics were conducted to describe study variables. The Magnitude of outbreak was described by person, place and time. Variables with p-value less than 0.05 in the multivariable logistic regression analysis were considered as statistically significant. Odds ratio with 95% confidence level was used to measure the presence and strength of the association between the malaria case and contributing factors. Hosmer and Lemeshow goodness of fit test was conducted to test the model fitness.
The ethical approval was waived by the institutional review board of the University of Gondar. Formal letter of cooperation was written from Argoba district health office.