Clinico-epidemiological determinants for empirical antibiotic prescriptions in suspected urinary tract infection cases attending peripheral hospitals of West Bengal, India- a cross sectional analysis

Primary and Secondary care hospitals in India do not have any microbiological diagnostic facilities, hence clinicians in these hospitals primarily depend on clinical and epidemiological features of patients for their empiric antibiotic treatment decision. In this study we aimed to identify clinical and epidemiological factors that determines the prescription of any antibiotic, multiple antibiotic and antibiotic as per WHO AWaRe classifications by the physicians for suspected Urinary Tract Infection (UTI) cases in primary and secondary care hospitals in South 24 Pargana district, West Bengal, India. In this facility based cross sectional study, we collected 375 prescriptions of suspected UTI cases at their first clinical encounter in the current episode from two primary care and one secondary care hospitals in south 24 pargana districts of West Bengal. We have counted the number of antibiotics prescribed and their WHO AWaRe group. We considered antibiotic prescribed as outcome variables and all the clinical symptoms, age, gender, comorbidity, health tier, number of non-antibiotic drugs prescribed as independent variables. Unadjusted an Adjusted Odds Ratio with 95% Confidence Intervals were used respectively as univariate and multivariate (Binary and Multinomial Logistic Regression) methods. Overall, any antibiotic was prescribed in 226 (60%) out of 375 prescriptions. Urinary urgency was associated with prescription of any antibiotic. (Adjusted OR – 2.45, 95% CI – 1.24–4.85). Inpatient (Adjusted OR – 32.61, 95% CI – 8.36 -127.17) and recurrent episodes (Adjusted OR – 3.39, 95% CI – 1.01–11.50) were strongly associated with prescription of multiple antibiotics. Number of prescribed non antibiotics drugs ≥ 4 and presence of fever showed association with both any antibiotic and multiple antibiotic prescription only in univariate analysis. Dysuria and increased urinary frequency were associated with choice of a Watch antibiotics as compared to Access group antibiotic. We concluded that clinical determinants of antibiotic prescriptions are not consistent rather prescriber practices driven by contextual determinants and perceived severity are consistently associated with prescription of antibiotic. Hence to rationalize the use of antibiotics in these peripheral settings, initiation of customized antimicrobial stewardship grounded on microbiological diagnosis need to be urgently considered.