Background Antibiotic resistance is an emerging problem caused due to antibiotic use. In countries with high rates of infectious diseases, antibiotic resistance is a frequent cause of mortality. The aim was to analyse antibiotic prescribing practices between 2008 and 2017 in a teaching (TH) and a non-teaching (NTH) hospital, as typical hospitals of low- and middle-income countries, and to compare antibiotic prescribing for severe infectious indications for which empiric antibiotic treatment is recommended. Methods Data from adult patients registered at two Indian private-sector hospitals with one of the following indications: epiglottitis, pneumonia, peritonitis, pyelonephritis, cellulitis, erysipelas, septic arthritis, endocarditis, meningitis or sepsis; were included and analysed. Antibiotic prescription data was analyzed using the World Health Organization’s (WHO) Anatomical Therapeutic Chemical classification system and the Defined Daily Doses. Chi-square and linear regression were used to compare the data between groups. Time series analyses were conducted using linear regression. P-values <0.05 were considered significant. Results In total, 3,766 patients were included, 2,504 inpatients in the NTH and 1,262 in the TH, of which 92% and 89% patients, respectively, were prescribed antibiotics. Sixty-one percent of total prescriptions in the TH and 40% in the NTH comprised the access category of antibiotics (i.e. the first-choice of treatment according to the WHO). The WHO’s second-choice of treatment, the watch category, comprised 29% and 40% of total prescriptions in the TH and NTH, respectively. Prescribing of fixed-dose combinations (FDCs) of antibiotics was significantly higher in the NTH (18%) than in the TH (8%, P <0.05). Prescribing of watch antibiotics and FDCs increased significantly in both hospitals between 2008 and 2017 among patients with pneumonia, cellulitis and peritonitis (P <0.05). Conclusions Prescribing of watch antibiotics and FDCs of antibiotics increased over time at both hospitals, indicating under prescribing of access antibiotics and more prescribing of second-choice antibiotics. The results can be used to highlight the areas of improvement in similar settings. Implementing diagnostic routines and local prescribing guidelines could improve the prescribing practices.