Background: Irrational antibiotic use is an important factor for development and spread of resistance to currently used antibiotics. This study was carried out to assess antibiotic prescribing practices among cases diagnosed as malaria at three hospitals in Moshi Municipality in northern Tanzania.

Methods: This was a cross sectional, retrospective study that included patients files from Kilimanjaro Christian Medical Centre (KCMC), Mawenzi Regional Hospital and St Joseph Hospital. Patient files whose primary provisional diagnosis was malaria were analysed using a convenient sampling method. Variables of interest were the types of medications prescribed, whether or not a laboratory test was requested and treatment was initiated before laboratory reports.

Results: A total of 250 patients’ files were included in the analysis (KCMC=62.8%; Mawenzi=23.2%; St. Joseph=14.0%). In 232 (92.8%) prescriptions made in the three hospitals, laboratory tests were requested to confirm diagnoses. Among laboratory tests requested, 89.2% were blood slides for microscopic detection of malaria parasites, 3.01% malaria rapid diagnostic tests and 3.01% other tests. The majority of prescriptions across all three hospitals (KCMC=86.4%; Mawenzi=91.4%; St. Joseph= 72.4%; X2=7.787). Clinicians at Mawenzi were more likely to start treatment before laboratory findings than their counterparts at KCMC and St Joseph hospitals (X 2=7.787, p≤0.05). A significantly higher number of prescriptions made before laboratory findings were observed at KCMC than Mawenzi and St. Joseph hospitals (X2=7.787, p<0.05). Prescriptions from KCMC were more likely to include at least one type of antibiotic than in the other two facilities. Over one third (KCMC=34.0%; St. Joseph=42.1%; Mawenzi=38.1%) of the prescriptions made contained at least one type of an antibiotic. There was a strong association between health facilities and antibiotics prescription in which KCMC prescribed antibiotics at the highest rate while Mawenzi Regional Hospital prescribed antibiotics at the lowest rates (X2=29.234, p<0.001).

Conclusion: Antibiotics are prescribed at a high rate among provisionally diagnosed malaria cases before availability of laboratory results. Efforts should be made to improve laboratory services in terms of trained personnel and equipment to reduce irrational use of antibiotics in provisionally diagnosed malaria cases.