High prevalence of antimicrobial resistance in rural Burkina Faso : assessment of risk factors for prevention and control
(2025)
Files
Thesisdocument_DanielValia.pdf
Open Access - Adobe PDF
- 5.22 MB
Details
- Authors
- Supervisors
- Abstract
- In Burkina Faso, from 2012 to 2018, extended-spectrum β-lactamase-Producing Escherichia coli (ESBL-EC) and Klebsiella pneumoniae (ESBL-KP) have been increasingly isolated from community-acquired invasive infections. This increase could be reflecting a high and increasing prevalence of faecal colonisation with these resistant bacteria in the community, as far as gut colonisation was identified as precursor to bloodstream infections. The overall objective of this thesis was therefore to assess the extent of these resistant bacteria in the community in rural Burkina Faso and to understand associated factors, in order to inform tailored infection prevention and control (IPC) interventions. Our data showed an estimated prevalence of faecal colonisation with ESBL-EC and ESBL-KP at 61.3% in rural Burkina Faso. This prevalence was higher during the rainy season compared to the dry season (70.2% vs 53.6%, p<0.001) and higher among study participants reporting not washing hands with soap before meals compared to those who did (62.5 vs 49.0%, p<0.001). In both bacteria, blaCTX-M-15 was the most prevalent (47.3% in E. coli and 19.9%; in K. pneumoniae) β-Lactamase genes. Plasmid-mediated quinolone resistance (PMQR) genes as qnr (48.1% in E. coli and 81.1% in K. pneumoniae), aac(6’)-ib-cr (21.2% in E. coli and 18.9% in K. pneumoniae) as well as OqxAB (5.8% in E. coli, and 78.4% in K. pneumoniae) were found along with β-Lactamase genes. In patients with severe acute febrile illness attending the Nanoro district hospital, 39.5% reported pre-hospital antibiotic use. This pre-hospital antibiotic use was significantly higher among patients referred from primary healthcare centers than among those who self-referred (54.0% vs 26.7%, p<0.001). Among all pre-hospital antibiotic use reported (424), Watch antibiotics were more frequently reported by referrals compared to self-referred patients (42.2% vs 28.1%, p=0.004). The investigations to understand the role of different healthcare providers and knowledge of antibiotics, showed that 33.5% healthcare were seeking outside healthcare facilities, including informal medicine vendors (47.7%), self-medication with left-over medicines kept at home (26.5%), medicine vendors in formal pharmacies (16.4%), traditional healers (9.4%) and only the latters (traditional healers) were not antibiotic dispensers. Reported reasons for seeking healthcare outside healthcare facilities included financial limitation, proximity to informal drug vendors, long waiting times at healthcare facilities and health professionals’ non-empathetic attitudes towards their patients. Antibiotics knowledge (only for illnesses of bacterial origin) was limited among healthcare professionals, very limited among medicines vendors in formal pharmacies, and non-existent among informal medicine vendors and the general community. While investigating antibiotic use by clinical presentation across all healthcare providers, we found that outpatient antibiotic use was more frequent after health center visits (54.8%, of which 16.5% Watch, n = 1249) than after visits to pharmacies (26.2%, 16.3% Watch, n = 328) and informal medicine vendors (26.9%, 50.0% Watch, n = 349). Across all healthcare providers, patients presenting with clinical presentations for which antibiotics were not recommended such as malaria, rhinopharyngitis, bronchitis, gastroenteritis, pain and wound were dispensed (Watch) antibiotics. Compliance with WHO’s AWaRe Antibiotic Book could have averted at least 68.4% of all Watch antibiotic use in outpatients at health centers. Community-wide, 2.9 DDD (95% CI 1.9–3.9) were used per 1000 adult inhabitants per day, health centers representing 89.7% of it. We concluded that the challenge of controlling antimicrobial resistance in such a setting should be multifaceted and combine both tailored antimicrobial stewardship (AMS) across all healthcare providers and the community to reduce antibiotic selective pressure and community-based hygiene interventions to break the cycle of transmission in order to mitigate and/or prevent spread. Antimicrobial stewardship program should be particularly intensified in health centers and should include dedicated education and awareness on AMR for healthcare workers, improved diagnostic tools to differentiate bacterial from non-bacterial infections, patient management algorithms based on the latest WHO recommendations for antibiotic prescription. At formal pharmacies, regulation on antibiotic sales should be strengthened, in combination with regular AMR awareness activities and monitoring to mitigate over-the-counter dispensing. At informal medicine vendors, AMR awareness programs should help self- restriction of Watch antibiotic sales. At community level, awareness activities should include risk behaviours leading to emergence and spread of resistant microorganisms in the community. Regarding community-based hygiene interventions, improving hand hygiene practices and enhancing sanitation can be effective steps toward mitigating the burden of antimicrobial resistance. Our studies were based on rigorous methodology. However, we are aware of some limitations that we have tried to overcome whenever possible. Regarding all questions on antibiotic use the last three months before the day of survey in the community, community members surveyed might have used medicine, not knowing whether it was an antibiotic or not. To mitigate response biases, we provided each fieldworker with any type of antibiotic available from the range of healthcare providers in the study area (Pharmacies in healthcare facilities, private pharmacies, informal medicine vendors). Once in the household, in case medicine use the last three months was reported, fieldworkers should ask the interviewed person to identify which one(s) among the batch of antibiotics they acknowledge having used. As well, selection of isolates for molecular characterization might have led to the absence of less phenotypically expressed genes. To mitigate this, we first divided all isolates into their different expressed phenotypic group (seven groups) and performed random selection among each group. In pre-hospital antibiotic use evaluation through patients attending the Nanoro district hospital, data was collected via a survey capturing use during the two weeks before consultation at the hospital, potentially underestimating actual use. Whenever possible, reported antibiotic use was verified from referral forms, patient medical files (healthcare booklet), and antibiotic packaging or blisters. During investigations on places of healthcare seeking the last three months, Participants may struggle to accurately recall where they sought healthcare, especially if multiple visits occurred. Minor illnesses visits may have been forgotten or misreported. Respondents may have also reported seeking care at formal healthcare facilities rather than informal or traditional healers, even if they used both, due to perceived judgment. Likewise, a severe illness or a visit closer to the time of the survey may be more easily recalled than earlier visits, which may lead to over reporting of recent experiences. To mitigate responses biases, we used a more or less short recall period (e.g. three months instead of six months). For each participant, field workers had to ask about symptoms or illnesses presented and for each of these illnesses, ask thereafter all the providers visited. Field workers also have to provide memory aid by recalling all healthcare providers in the surrounding. Visits were also crosschecked with medical records whenever possible. In the investigation of antibiotic use by clinical presentation across all healthcare providers, reported prevalence, means and proportions were corrected for clustering by healthcare provider, in two strata (Nanoro and Nazoanga), except for informal medicine stores, for which neither store nor health area were recorded to ensure confidentiality. Despite these limitations that we have tried to overcome as much as possible, this thesis reports robust findings that provide a full picture of population-level antimicrobial resistance in a rural setting of low-and Middle-income countries and the interconnected factors that may be underlying it, in order to set up tailored infection prevention and control interventions.
- Affiliations
Citations
Valia, D. (2025). High prevalence of antimicrobial resistance in rural Burkina Faso : assessment of risk factors for prevention and control. https://hdl.handle.net/2078.5/243567