Antibiotic resistance: implications of hospital practices for public health : a study from Hanoi, Vietnam

Abstract: Background: Antibiotic resistance is a global challenge, requiring urgent attention. Inefficient hospital infection control and hospital dissemination of antibiotic residues and antibiotic resistant bacteria contribute to the spread of antibiotic resistance. Overall aim: To explore hospital practices in relation to the spread of antibiotic resistance in a rural and an urban hospital in Hanoi, Vietnam. Methods: The thesis consists of one qualitative (II) and three quantitative studies (I,III&IV). In Paper I, a cross-sectional study, questionnaires consisting of items on knowledge and practices of infection control were collected from 339 hospital staff. For analysis, total knowledge or practice score ranged from 0-15. In Paper II, individual interviews and focus group discussions were conducted with a total of 50 doctors, nurses and cleaning workers. A semi-structured guide on infection control was used. Content analysis was applied. In Papers III & IV, wastewater samples were collected every month over one year in the two hospitals; quantities of antibiotics used were also collected in the rural hospital. High-performance liquid chromatography-tandem mass spectrometry was used to determine antibiotic concentrations; standard disk diffusion and E-test were applied for antibiotic susceptibility testing in 265 Escherichia coli (E. coli) isolates from the water samples; polymerase chain reactions were used to detect antibiotic resistance genes. Results: The majority of hospital staff showed good knowledge of infection control and had good or adequate practice scores. Median knowledge scores were 11.8 (6.8-13.9) and 12 (1.4- 14.5) (p=0.17); median practice scores were 11.4 (4.7-15.0) and 12.4 (1.0-15.0) in the rural and the urban hospitals respectively (p=0.003). Cleaning workers had lower knowledge as well as practice scores than doctors and nurses (I). The staff acknowledged poor infection control practices during interviews and groups discussions. They pointed out various difficulties but were not aware of the situation of healthcare-associated infections in their hospitals (II). In hospital wastewater samples, studied antibiotics were present both before and after wastewater treatment: 70.5µg/L before treatment and 34.0µg/L after treatment per month in the rural hospital; 93.5µg/L before treatment and 32.4µg/L after treatment per month in the urban hospital. A significant correlation with the quantities used was found for ciprofloxacin (r = 0.78; p = 0.01) and metronidazole (r = 0.99; p < 0.001) (III). Resistance to at least one of the studied antibiotics was detected in 83% of E. coli isolates; multidrug resistance was found in 32%. The highest resistance prevalence was found for co-trimoxazole (70%). Forty-three percent of isolates were ESBL-producing, with the blaTEM gene being more common than blaCTX-M. Co-harbouring of the blaCTX-M, blaTEM and qepA genes was found in 46% of isolates resistant to ciprofloxacin (IV). Conclusions: Hospital staff was generally knowledgeable about infection control, but was not aware of the situation in their own hospital. Although practice scores were good or adequate for the majority of the staff, in fact, the practices seem to remain poor. Antibiotic residues were present in hospital wastewater both before and after wastewater treatment at concentrations that can promote the development of antibiotic resistance. Antibiotic-resistant E. coli along with genes coding for cephalosporin resistance blaCTX-M and blaTEM, and a gene coding for ciprofloxacin resistance qepA were highly prevalent among the isolates.