Antibacterial resistance in urinary samples from long-term care-facility and community-dwelling older people in Gauteng, South Africa

OBJECTIVE: Older populations (≥ 60 years) are frequent users of health care services, often vulnerable to infections and antimicrobial resistance (AMR). Urinary tract infection is often diagnosed in this cohort based on non-specific or atypical symptoms, yet empirical therapy seldom considers the potential influence of residential setting on resistance patterns. This study therefore aimed to quantify and compare the in vitro AMR profiles of urinary pathogens from these two settings and assessed temporal resistance trends to inform antimicrobial stewardship priorities. METHODS: A retrospective analysis was conducted using anonymised microbiology data from Ampath Laboratories in Gauteng. Urine cultures collected from older adults (≥60 years) between 2015 and 2023 were included. The five most common uropathogens were analysed. Resistance differences between long-term care facility (LTCF) and community samples were assessed using univariate and multivariable logistic regression models adjusting for age, sex, inpatient status, and sampling year. Temporal trends were evaluated using Kendall tau-b. RESULTS: A total of 50 704 urine cultures were analysed, of which 18.81% were from LTCFs. Escherichia coli and P. mirabilis showed consistently higher resistance in LTCF isolates. In E. coli, LTCF residence was associated with increased resistance to ciprofloxacin (adjusted OR 1.26; 95% CI 1.18-1.34), nitrofurantoin (1.55; 1.38-1.75), ceftriaxone (1.24; 1.14-1.34), amoxicillin/clavulanic acid (1.11; 1.05-1.19), and co-trimoxazole (1.10; 1.03-1.18). P. mirabilis showed similar patterns, with higher resistance to ciprofloxacin (2.16), gentamicin (1.64), fosfomycin (1.72) and co-trimoxazole (1.49). Trend analysis showed increasing fosfomycin resistance from E. coli isolates in both LTCFs (3.74% to 6.18%) and community settings (2.75% to 3.74%) and rising ceftriaxone resistance in community isolates (13.32% to 17.60%). CONCLUSIONS: LTCF residence was independently associated with higher AMR in several common urinary pathogens. These findings suggest that empirical urinary tract infection regimens commonly used in community-dwelling older adults may be inadequate for LTCF residents. Setting-specific antibiograms, improved diagnostic stewardship, and strengthened antimicrobial stewardship interventions are needed to support appropriate prescribing. Further studies incorporating patient-level clinical factors are warranted to refine empirical treatment recommendations for older adults across care settings.