Retrospective antimicrobial consumption surveillance at health facility level in Dodoma Region, Tanzania

Antimicrobial resistance (AMR) is a major threat in sub-Saharan Africa (SSA), but assessments of antimicrobial consumption (AMC) are limited. This study aimed to investigate regional AMC and resistance patterns in a representative area of Tanzania and to introduce a method for determining AMC in low-resource settings.

We conducted a retrospective study using prescription data collected over 5 years (2013-2017) from multiple hospitals and selected primary health facilities in the Dodoma Region of Tanzania. The study employed the WHO's Anatomical Therapeutic Chemical (ATC) classification and Daily Defined Dose (DDD) methodology to quantify antimicrobial use. Outpatient prescription records that met our inclusion criteria were analysed, while incomplete records were excluded. Sensitivity testing for frequently prescribed antimicrobials was performed against representative gram-negative and gram-positive bacteria, with resistance expressed as minimum inhibitory concentrations and resistance percentages.

This study was conducted across several high-volume healthcare facilities in the Dodoma Region, encompassing both hospital and primary care settings. The dataset comprises outpatient prescription records from these facilities, representing a significant proportion of the regional healthcare usage. The selected facilities were chosen based on their substantial catchment populations to maximise data volume and relevance.

The primary outcome measure was the DDD per 1000 inhabitants per day (DID) for various antimicrobial classes. Secondary outcomes included the prevalence of specific drugs, such as amoxicillin and erythromycin, and their corresponding resistance profiles. Resistance data were quantitatively analysed, with particular attention given to penicillinase-sensitive penicillins and their resistance rates among gram-negative and Gram-positive bacteria.

Analysis revealed that single penicillins, particularly amoxicillin, dominated prescriptions, accounting for 25% to 60% of hospitals and 13% to 29% of primary health centres. Erythromycin was prescribed in 9.4% to 25.1% of cases across facilities. The overall AMC in the region ranged from 36.7 to 50.2 DID during the study period, with consumption patterns showing an initial increase of 4.3% from 2013 to 2014, a subsequent 29.0% decrease from 2014 to 2015, followed by a 34.1% increase from 2015 to 2016 and a further 37.9% increase from 2016 to 2017. Resistance testing demonstrated that penicillinase-sensitive penicillins exhibited an average resistance rate of 87.3%, with gram-negative and gram-positive bacteria showing resistance levels of 90.1% (±8%) and 83.6% (±8%), respectively, indicating a statistically significant association (p<0.05) between high consumption and elevated resistance.

Our findings reveal that AMC in the Dodoma Region is high and is linked to significant resistance against commonly used agents. This study presents a robust method for monitoring AMC and resistance in a resource-efficient manner, offering potential applicability in similar settings across SSA. The data, though representing only a fraction of total antimicrobial use, underscore the urgent need for targeted antimicrobial stewardship interventions, particularly at the primary healthcare level. Future research should further investigate demographic influences and evaluate intervention strategies to mitigate AMR effectively.