RESEARCH

 

Impact of pharmacist-prescriber partnerships to track antibiotic prescribing in publicly funded primary care in the Cape Town metropole, South Africa: An implementation study

 

 

O van Hecke^I; Y Adegoke^II; M Allwood^III; K von Pressentin^IV; M Namane^V; C Butler^VI; M Mendelson^VII; R Coetzee^VIII

^IFRCGP, DPhil ; Centre for General Practice, Department of Public Health and Primary Care, Ghent University, Belgium
^IIMSc, PhD ; School of Public Health, Faculty of Community and Health Sciences, University of the Western Cape, Cape Town, South Africa
^IIIBSc; Nuffield Department of Primary Care Health Sciences, University of Oxford, UK
^IVMMed (Farn Med), PhD; Division of Family Medicine, Department of Family, Community and Emergency Care, Faculty of Health Sciences, University of Cape Town, South Africa
^VMPhil (Farn Med), MSc (Clin Epi) ; Division of Family Medicine, Department of Family, Community and Emergency Care, Faculty of Health Sciences, University of Cape Town, South Africa
^VIMD, FMedSci ; Nuffield Department of Primary Care Health Sciences, University of Oxford, UK
^VIIMM BS, PhD ; Division of Infectious Disease and HIV Medicine, Department of Medicine, Groote Schuur Hospital, University of Cape Town, South Africa
^VIIIPharm D; School of Public Health, Faculty of Community and Health Sciences, University of the Western Cape, Cape Town, South Africa

Correspondence

 

 

---

ABSTRACT

BACKGROUND. More than 80% of the South African (SA) population receive their care in publicly funded primary care clinics. The majority come from socioeconomically disadvantaged communities, and bear the greatest burden from infectious diseases. However, there are very limited published data on and evaluation of antibiotic prescribing linked to clinical indication. This is a major gap.
OBJECTIVES. To assess the impact of a pharmacist-prescriber partnership to track antibiotic prescribing in publicly funded primary healthcare clinics in the Cape Town metropole, SA.
METHODS. We conducted a prospective observational study across five clinics where there was a dispensing pharmacist team. At each clinic, a prescriber-pharmacist team gathered prospective antibiotic prescribing data for 'acute cough', linked to clinical indication, and provided individual prescribing feedback through each clinic's WhatsApp messaging platform about their prescribing quality (antibiotic dose, duration, frequency).
RESULTS. Eight out of every 10 patients (adults and children) were prescribed an antibiotic (n=457). In a third of patients, an antibiotic was prescribed for suspected 'community-acquired pneumonia'. The WhatsApp prescribing feedback was used in half of all visits. The proportion of pharmacy-dispensed antibiotics concordant with local guidelines in terms of prescribing quality was 95% (95% confidence interval 0.93 - 0.98), n=239). Against AWaRe (access/watch/reserve) guidance, 97% of antibiotics prescribed belonged to the 'access' group.
CONCLUSION. Although prescribing concordance with guidelines was good, a significant proportion of patients were prescribed an antibiotic for 'acute cough'. Our findings have filled a fundamental gap in the evidence base that will inform antibiotic stewardship innovations, guideline development and future interventions.

Keywords: antimicrobial stewardship; primary care, prescribing feedback

---

 

 

South Africa (SA) has a high burden of infectious diseases complicated by the disproportionate impact of HIV and tuberculosis. Antibiotic resistance affects low- and middle-income countries disproportionately.^[1] Respiratory tract infections such as acute cough form the bulk (>60%) of all antibiotic prescriptions in primary care.^[2] Frontline clinicians (doctors, nurse prescribers) must make daily decisions about antibiotic treatment without knowing the exact cause of infection or whether patients require antibiotics. The decision to prescribe an antibiotic, for example, for acute cough, is a complex behaviour that requires bespoke interventions tailored to the context.

 

The evidence gap

This upper middle-income country is responding to the national threat from antibiotic resistance through its National Strategic Framework (2018 - 2024).^[3,4] The Standard Treatment Guidelines and Essential Medicines List (EML) for SA (primary healthcare level) provide an evidence base for prescribing in primary care.^[5] However, there is limited published research and evaluation of antibiotic prescribing in primary care, especially in the public sector, because of a lack of integrated pharmacy, laboratory and clinical information systems. This is a fundamental evidence gap. Where prescribing data are available, these data are retrospective, and antibiotic prescribing data are not linked to clinical indication.^{[4, ,6, 7]} There have been no studies that have systematically documented sequential patients with acute cough attending publicly funded primary care health facilities in SA. Prospective antibiotic prescribing data are key as a benchmark for future antibiotic stewardship interventions and guideline development, to optimise the quality of antibiotic prescribing and reduce the risk of antimicrobial resistance in primary care.^[8]

 

Pharmacist-prescriber partnerships to optimise antibiotic prescribing

SA has made significant inroads into this gap by employing a multidisciplinary strategy in hospitals, where pharmacists and clinicians are engaged in antibiotic stewardship partnerships to improve antibiotic prescribing.^[910] However, this is needed in community settings, too.^[4] This will become increasingly important in the country's transition to the anticipated National Health Insurance, where pharmacists are likely to expand their role in primary healthcare services to increase access, and point-of-care tests may be expanded into the community for common infection syndromes. There is good evidence from high-income countries (HICs) that a pharmacist-prescriber model effectively decreases antibiotic prescribing, and increases guideline-adherent prescribing in community settings.^[11] However, evidence generated in HICs is not always applicable in a country like SA, which has markedly different patient populations, disease spectra, health systems and social and cultural determinants of health.^[12] This study builds on exploratory work^[12-14] to test this approach in the context of an upper middle-income country.

The aim of our study was to assess the implementation of a pharmacist-prescriber partnership (PPP) to track and optimise antibiotic prescribing for patients presenting with acute cough in publicly funded primary healthcare clinics (PHCs) in the Cape Town metropole, SA.

In order of priority, there were three study objectives. The first objective was to collect baseline observational data (demographic, clinical and prescribing data) of patients presenting with acute cough. The second objective was to to assess the proportion of antibiotics prescribed for acute cough that are concordant with local prescribing guidance when PHCs employ PPPs. The third objective was to measure the impact of PPPs on health outcomes (subsequent contacts with the health services or hospital admission within 7 days).

 

Methods

The Cape Town metropole is a peri-urban region of just under 300 km (188 miles) around Cape Town. It includes the suburbs of Khayelitsha and Mitchells Plain, with an estimated population of 4.5 million people. Based on census data,^[15] most of the population are uninsured (77%) and rely on publicly funded healthcare in the region. Access to basic services such as water, sanitation and electricity is good (>85%), but unemployment is high (>40%), with a substantial proportion of the population living in informal dwellings (20.5%). The Cape Metro region has 152 public primary healthcare facilities and 8 district hospitals. Access to publicly funded primary healthcare services, where the vast majority of the population seek their healthcare, is free of charge in SA. Obtaining antibiotics is by prescription only, although recent research has shown that privately owned pharmacies in townships in SA are dispensing antibiotics without prescription.^[16] Although internet access is available in some larger PHCs, patient medical records are paper-based only. Mobile phone connectivity coverage is excellent.

The present study is a pharmacist-driven, prospective observational study with a novel (near) real-time individual prescribing feedback mechanism across five PHCs (sites) where there is a dispensing pharmacy team. We recruited PHCs through the local family physician forum, nursing and pharmacy teams. Participation was voluntary. We staggered the introduction of sites during the study period with an onsite training visit about study procedures.

We employed a PPP to gather prospective antibiotic prescribing data for acute cough (<28 days duration) linked to antibiotic prescribing quality (antibiotic dose, duration, dose frequency) and reported clinical indication. We selected 28 days to reflect the common duration of symptoms for acute cough in 90% of cases.^[17] Study procedures, data collection and materials were piloted before the main study (start date 2 June 2022). We planned to collect data for up to 6 months, and recruit between 500 and 800 patients.

During their routine clinical practice, frontline clinicians (doctors and nurse prescribers or 'clinical nurse practitioners') identified and described patients according to a paper-based carbon-copy one-page case report form (CRF) (appendix https://www.samedical.org/file/2308). Clinicians were asked to include consecutive patients as best they could. Patient details were de-identified using a patient study identification number on the CRF. Clinicians recorded relevant medical history, presenting symptoms, diagnosis, antibiotic therapy if prescribed, and any planned follow-up. This study did not involve any change in clinical practice. CRFs were collated fortnightly, transferred to a secure electronic database (REDCap) and checked after data entry.

Patients were adults and children presenting with acute cough or, alternatively, where cough was not the most prominent symptom (e.g. fever, malaise) but where the clinician considered acute respiratory tract infection as the primary diagnosis. HIV-positive patients with acute cough were eligible, as well as patients in whom tuberculosis was suspected. This included patients with asthma and chronic obstructive pulmonary disease (COPD).

Patients were not eligible if the working diagnosis was a non-infective cause of cough (e.g. heart failure), or required immediate hospitalisation. Patients received their usual clinical care.

Pharmacists, located at each PHC site, recorded and dispensed the prescribed treatment, including antibiotic course, as usual. At the same time, they audited and reviewed the appropriateness (concordance) of the antibiotic prescribed on the study CRF according to SA Standard Treatment Guidelines and Essential Medicines List 2020 (primary healthcare level) in terms of recommended antibiotic dose, duration and frequency. The pharmacy team then notified clinicians of their assessment using the pre-existing clinic group messaging platform (i.e. WhatsApp group chat). Only the patient study ID and information about concordance were sent in a pre-approved text message within the group chat. Pharmacists were free to discuss the antibiotic prescription with the clinician through their usual means of communication.

After the initial consultation, the research team attempted to contact patients to ascertain whether they had had subsequent contact(s) with the health services and/or hospital admission within 7 days of the initial consultation. This occurred between 7 and 10 days after the initial consultation, to accommodate weekends and public holidays. Contacts were conducted through telephone assessment (three attempts over 2 consecutive days).

The study was conducted according to the guidelines of the Declaration of Helsinki, and approved by the University of Oxford (OxTREC ref. no. 504-21, 12 April 2021) and the University of the Western Cape (BMREC ref. no. BM21/10/17, 26 Nov 2021)

 

Results

Between 2 June 2022 and 30 September 2023, we enrolled 457 patients with symptoms of acute cough at five PHCs.

Baseline characteristics are shown in Table 1. There were slightly more adults (56%) than children (44%). The most common listed comorbidities were hypertension (15%), COPD (10%) and HIV (7%). Overall, 8 out of every 10 patients (84%) were prescribed an antibiotic for acute cough. This varied between sites. Out of all patients who were prescribed antibiotics (n=382) for acute cough, 57% were adults and 42% children.

 

 

The top three antibiotics prescribed were amoxicillin (63%), co-amoxiclav (13%) and phenoxymethylpenicillin (6%). In terms of quality indicator prescribing against AWaRe (access/watch/reserve) guidance, 97.6% of antibiotics prescribed belonged to the 'access' group (with amoxicillin and co-amoxiclav accounting for 92%), ceftriaxone and azithromycin accounted for one case each in the 'watch' group, and there were none in the 'reserve' group (available data n=382, Table 1)

A clinical diagnosis of 'community-acquired pneumonia' (CAP) was by far the top clinical indication (35%) for which an antibiotic was prescribed across four of the five sites (in site 5, the most common indication was acute bronchitis, Fig. 1 and Table 2). This was followed by tonsillitis (11%), viral upper respiratory tract infection (URTI) (8%), influenza (8%) and bronchitis (8%). Differences between adults and children per clinical indication are outlined in Fig. 1.

Antibiotics were prescribed for all suspected CAP (n=158). More than half of the antibiotics for CAP were prescribed by nurse prescribers compared with doctors (53% v. 35%). CAP cases were more often diagnosed in children by nurses, compared with CAP in adults, who were diagnosed more commonly by doctors (Table 3).

 

 

Half of all recorded antibiotics were prescribed by nurse prescribers (51%, n=235). Antibiotics were directly dispensed without pharmacy input in over half of these cases.

Sixty percent (140/235) of nurse-led antibiotic prescriptions were for children. This is in comparison with doctors, where 80% of antibiotic prescriptions were for adults (128/161), and where pharmacy staff had the opportunity to comment on the prescription concordance and provide feedback via the clinic WhatsApp group chat.

There was no statistical difference in antibiotic prescribing rate for acute cough conditions between nurse prescribers and doctors (85% v. 83%, p=0.48).

Pharmacy prescribing feedback using the group messaging platform was used in just under half of all consultations (n=102/208, 49%), with a similar split between nurses (n=103) and doctors (n=105), excluding those cases where nurses had directly dispensed antibiotics.

The proportion of pharmacy-dispensed antibiotics that agreed with local guidelines in terms of antibiotic dose, duration and frequency was 95% (95% confidence interval (CI) 93% - 98%, n=239).

Follow-up data (n=386) at 1 week were incomplete. The proportion of patients with an invalid or no registered telephone number was 37% (n=140). Just under 20% of patients (n=74) did not respond after three contact attempts. From available complete data at 1week follow-up (n=172), re-attendance was 3% (n=5), and no hospitalisations were recorded.

 

Discussion

Main findings

Our findings show that 80% of patients were prescribed an antibiotic for an acute cough presentation. In more than a third of these cases, an antibiotic was prescribed for a clinical diagnosis of CAP, and determined by the type of healthcare professional (nurse or doctor) whom the patient (adult or child) would routinely consult. Half (51%) of all recorded antibiotics were prescribed by nurse prescribers who directly dispensed antibiotics without pharmacy input. Where the group WhatsApp prescribing feedback was used, there was good concordance of prescribed antibiotics with local guidelines in terms of antibiotic dose, duration and frequency.

Comparison with existing literature

Encouragingly, most antibiotics prescribed were from the 'access' group of antibiotics in the AWaRe (access/watch/reserve) classification. However, there was a high rate of antibiotic prescribing, which mirrors that of earlier studies in the public sector, where the proportion of patients receiving an antibiotic prescription ranged between 69% and 78% of consultations.^[6,18,19] This high antibiotic prescribing rate is not dissimilar from other published data in the SA private sector.^[19,20] However, our prospective study now adds the presumed clinical indication for this prescription (where previously, this was absent in a third of cases).^[7]

We captured prescribing data using a multidisciplinary pharmacist-prescriber approach. This aligns with the findings of a systematic review that showed that the pharmacist-prescriber model can effectively increase guideline-adherent antibiotic prescribing (odds ratio 1.96, 95% CI 1.56 - 2.45, 10 trials).^[11] The studies included in the review were all conducted in HICs. This study is the first attempt in Africa to utilise this partnership in publicly funded primary care. In this study, the concordance of prescribed antibiotics was much higher (95%) than that in earlier published studies in the same region with different study designs ^[6,7]The Gasson et al. ^[6] study involved a point prevalence survey using a retrospective medical folder review, and found that overall antibiotic guideline adherence was 32%, and for patients presenting with respiratory symptoms, 34% (n=182). The de Vries et al.^[7] study was a peer audit and feedback intervention study, where 10 antibiotic prescriptions were randomly selected monthly for peer review. It measured adherence to seven antibiotic quality process measures, and showed improvement in the mean overall adherence, from 19% (baseline) to 47% (post-intervention). In our study, however, we only asked pharmacists to comment on the antibiotic concordance with local guidelines in terms of dose, duration and frequency.

Study strengths and limitations

Antibiotic prescribing data are not routinely available in publicly funded primary care. Where prescribing data are available, these data are often from the private sector^[21,22] and/or inpatient settings,^[9,23] retrospective^[24] and not linked to clinical indication.^[6,20] To our knowledge, this is the first study to capture prospective prescribing data for acute cough linked to clinical indication in publicly funded primary care in the region. This study builds on previous research into where and how a cross-disciplinary team could work in these specific settings to optimise antibiotic prescribing.^[12-14] Collecting this baseline information is critical to set a benchmark for antimicrobial stewardship initiatives, provide consistency and improve the quality of antibiotic prescribing for common infection syndromes.

We have included both nurse and doctor prescribers to reflect the breadth of antibiotic prescribing practice in the region, recognising that nurses see most patients.^[25] Our team-based approach to collecting these data ensured that we used limited resources advantageously, and empowered prescriber-pharmacist teams to co-design and participate in the study.

We harnessed the high mobile data connectivity across SA to provide efficient and low-cost (near) real-time individual prescribing feedback, mindful of the confines of data privacy and regulations. This feedback mechanism, therefore, has the potential to be refined and scaled up across the region.

There are several limitations, which have prompted further research questions. First, we were not able to incorporate our prescribing feedback mechanism to nurses dispensing antibiotics from their own consulting rooms. However, this has shed light on the significant proportion of antibiotics prescribed and dispensed by nurses, and provides impetus for future targeted antimicrobial stewardship initiatives.

Second, data on health outcomes at 1 week (i.e. subsequent contacts with health services or hospital admission within 7 days) were limited. In part, this was due to a mobile population with no permanent abode, multiple or obsolete mobile phone numbers, or apprehension of cold callers, e.g. debt collectors.

Third, not all the sites had the capacity to implement the quality improvement component successfully, and the same sites also struggled to collect prospective data on top of their clinical workload. We took a pragmatic approach, and therefore we were not able to verify whether clinicians collected data from consecutive patients, introducing the likelihood of sampling bias. There was also regular social unrest, which meant clinics were forced to close, and data collection was paused.

Fourth, although the antibiotic prescribing quality concordance was high, the study design did not allow us to determine whether the prescribed antibiotic was clinically appropriate. Likewise, the study design did not lend itself to making a comparison of sites that used PPP v. sites that did not use PPP.

Lastly, we accept that some prescribers may up-rate a diagnosis to justify their decision to prescribe an antibiotic. Likewise, clinicians might be wary of, for example, the high incidence of rheumatic heart disease in sub-Saharan Africa when children present with a sore throat (tonsillitis).

However, there are also a significant proportion of cases where an antibiotic would rarely be indicated, e.g. viral URTI or acute bronchitis.

Implications for practice, policy and future research

Our findings support the need to better understand the prescribing behaviour of frontline clinicians, and to explore the factors influencing this prescribing behaviour (e.g. uncertainty related to diagnosis or prognosis, preventing breakdown of clinician-patient relationship, etc.).

Clinically, our findings also support the need to better identify patients who might benefit from an antibiotic by promoting and evaluating fit-for-purpose point-of-care tests (as part of a multifaceted intervention). The proportion of patients prescribed an antibiotic for a clinical diagnosis of CAP without radiological verification was high (42%). By comparison, in the UK, this prescribing proportion is between 5 and 12% presenting to general practitioners in primary care. We accept that there is a high infectious disease burden in SA; however, the 'true' proportion of radiologically confirmed CAP is likely lower, but remains unknown. As there is limited community-based laboratory and radiological infrastructure, this provides an avenue to test the feasibility of point-of-care testing to help frontline clinicians make more informed treatment decisions at the point of care before initiating antibiotic treatment for suspected CAP.

This study collected important prospective prescribing data for acute cough. This is critical to accurately assess what problems there are in terms of access to, and excess use of, antibiotics in the region, and will inform antibiotic stewardship innovations, guideline development and future interventions.

Our findings have also prompted further research questions. For example, half of all antibiotic prescribing was performed by nurse prescribers, who primarily work according to validated treatment protocols, with appreciable protocol adherence. Mindful of the plausible risk of sampling bias, the high antibiotic prescribing rate calls into question whether such a symptom-based protocol favours an antibiotic prescription. Other plausible reasons for this high prescribing rate might include more severe illness than we anticipate, or that there are important social determinants of health that override protocol-driven treatment regimens.^[9]

In terms of quality indicator prescribing against World Health Organization AWaRe guidance, more than 95% of antibiotics prescribed belonged to the 'access' group. This highlights the limitations of the AWaRe guidance in resource-poor primary care settings, where there is a limited supply of 'watch' and 'reserve' antibiotics. Hence our findings will refocus the agenda on improving the quality of prescribing in these settings by developing quality indicators that focus on the diagnostic process, like those successfully employed in chronic diseases such as diabetes and cardiovascular diseases, to improve the appropriateness of antibiotic prescribing.

Finally, our findings will help determine the optimal sample size, research processes and recruitment strategy for a future randomised trial of the clinical- and cost-effectiveness of point-of-care testing for acute cough, which will include important embedded qualitative research methods.

Data availability. Data are available on request from the corresponding author.

Declaration. None.

Acknowledgements. The authors are grateful to Mrs Yasmina Johnson (Pharmacy Services, Western Cape Government Health, and A/Prof. Elma de Vries (family physician, Nelson Mandela University, Gqeberha, SA) for helpful suggestions in the initial study design, support and critical reading of the manuscript.

Author contributions. All authors contributed to the conception of this research. Ethical approvals, data collection, initial analysis and first draft for submission: OVH and RC. YA assisted in the original data collection and MA in data management. All authors contributed to the original submission and the final draft of this article. All authors approved the version to be published and agreed to be accountable for all aspects of the work.

Funding. The International Society for Infectious Diseases (ISID) and Pfizer Global Medical Grants Challenge: Antimicrobial Stewardship in Africa and Middle East (ref. no. 63318123, 15 September 2021).

Conflicts of interest. OVH has received consulting fees from MindGap (fees paid to institution), has participated on data safety monitoring boards or advisory boards for National Institute for Health and Care Research (NIHR)-funded trials in the UK, and has an unpaid leadership or fiduciary role in the British Society of Antimicrobial Chemotherapy.

 

References

1. Okeke IN, Laxminarayan R, Bhutta ZA, et al. Antimicrobial resistance in developing countries. Part I: Recent trends and current status. Lancet Infect Dis 2005;5(8 ):481-493. https://doi.org/10.1016/s1473-3099(05)70189-4        [ Links ]

2. National Institute for Health and Clinical Excellence. Respiratory tract infections - antibiotic prescribing: Prescribing of antibiotics for self-limiting respiratory tract infections in adults and children in primary care. London: NICE, 2008.         [ Links ]

3. Departments of Health and Agriculture Forestry and Fisheries, South Africa. Antimicrobial Resistance National Strategy Framework 2018 - 2024.         [ Links ]

4. Chigome A, Ramdas N, Skosana P, et al. A narrative review of antibiotic prescribing practices in primary care settings in South Africa and potential ways forward to reduce antimicrobial resistance. Antibiotics 2023;12(10):1540. https://doi.org/10.3390/antibiotics12101540        [ Links ]

5. National Department of Health, South Africa. Essential drugs programme. Standard Treatment Guidelines and Essential Medicines List for South Africa, primary healthcare level. Pretoria: NDoH, 2018.         [ Links ]

6. Gasson J, Blockman M, Willems B. Antibiotic prescribing practice and adherence to guidelines in primary care in the Cape Town Metro District, South Africa. S Afr Med J 2018;108:304-310        [ Links ]

7. De Vries E, Johnson Y, Willems B, et al. Improving primary care antimicrobial stewardship by implementing a peer audit and feedback intervention in Cape Town community healthcare centres. S Afr Med J 2022;112:812-818. https://doi.org/10.7196/SAMJ.2022.v112i10.16397        [ Links ]

8. Van Hecke O, Wang K, Lee JJ, et al. The implications of antibiotic resistance for patients' recovery from common infections in the community: A systematic review and meta-analysis. Clin Infect Dis 2017;65:371-382. https://doi.org/10.1093/cid/cix233        [ Links ]

9. Brink AJ, Messina AP, Feldman C, et al. Antimicrobial stewardship across 47 South African hospitals: An implementation study. Lancet Infect Dis 2016;16:1017-1025. https://doi.org/10.1016/S1473-3099(16)30012-3.         [ Links ]

10. Brink AJ, Messina AP, Feldman C, et al. From guidelines to practice: A pharmacist-driven prospective audit and feedback improvement model for peri-operative antibiotic prophylaxis in 34 South African hospitals. J Antimicrob Chemother 2017;72:1227-1234. https://doi.org/10.1093/jac/dkw523        [ Links ]

11. Saha SK, Hawes L, Mazza D. Effectiveness of interventions involving pharmacists on antibiotic prescribing by general practitioners: A systematic review and meta-analysis. J Antimicrob Chemother. 2019 May 1;74(5):1173-1181. https://doi.org/10.1093/jac/dky572        [ Links ]

12. Van Hecke O, Butler C, Mendelson M, Tonkin-Crine S. Introducing new point-of-care tests for common infections in publicly funded clinics in South Africa: A qualitative study with primary care clinicians. BMJ Open 2019;9:e029260. https://doi.org/10.1136/bmjopen-2019-029260        [ Links ]

13. Van Hecke O, Coetzee R. Establishing a pharmacist-prescriber partnership in publicly funded primary healthcare clinics to optimise antibiotic prescribing in the Western Cape: An exploratory study. S Afr Fam Pract 2020;62(1):e1-e4. https://doi.org/10.4102/safp.v62i1.5090        [ Links ]

14. Epps A, Albury C, van Hecke O. Exploring primary care clinicians' views about how best to implement a potential trial around point-of-care tests for common infections in South Africa. Diagnostics 2021;11(11):2100. https://doi.org/10.3390/diagnostics11112100        [ Links ]

15. Department of Statistics, South Africa. Census 22. Pretoria: Statistics South Africa, 2022. https://census.statssa.gov.za/#/province/1/2 (accessed 12 December 2023).         [ Links ]

16. Mokwele RN, Schellack N, Bronkhorst E, et al. Using mystery shoppers to determine practices pertaining to antibiotic dispensing without a prescription among community pharmacies in South Africa - a pilot survey. JAC Antimicrob Resist 2022;4:dlab196. https://doi.org/10.1093/jacamr/dlab196.         [ Links ]

17. Thompson M, Vodicka TA, Blair PS, et al. Duration of symptoms of respiratory tract infections in children: Systematic review. Br Med J 2013;347:f7027. https://doi.org/10.1136/bmj.f7027        [ Links ]

18. Mathibe LJ, Zwane NP. Unnecessary antimicrobial prescribing for upper respiratory tract infections in children in Pietermaritzburg, South Africa. Afr Health Sci 2020;20:1133-1142. https://doi.org/10.4314/ahs.v20i3.15        [ Links ]

19. Lagarde M, Blaauw D. Levels and determinants of overprescribing of antibiotics in the public and private primary care sectors in South Africa. BMJ Glob Health 2023;8(7):e012374. https://doi.org/10.1136/bmjgh-2023-012374        [ Links ]

20. Guma SP, Godman B, Campbell SM, Mahomed O. Determinants of the empiric use of antibiotics by general practitioners in South Africa: Observational, analytic, cross-sectional study. Antibiotics 2022;11(10):1423. https://doi.org/10.3390/antibiotics11101423        [ Links ]

21. Alabi ME, Essack SY. Antibiotic prescribing amongst South African general practitioners in private practice: An analysis of a health insurance database. JAC Antimicrob Resist 2022;4(5):dlac101. https://doi.org/10.1093/jacamr/dlac101        [ Links ]

22. Ncube NB, Solanki GC, Kredo T, Lalloo R. Antibiotic prescription patterns of South African general medical practitioners for treatment of acute bronchitis. S Afr Med J 2017;107(2):119-122. https://doi.org/10.7196/SAMJ.2017.v107i2.11276        [ Links ]

23. Mthombeni TC, Burger JR, Lubbe MS, Julyan M. Antibiotic prescribing to inpatients in Limpopo, South Africa: A multicentre point-prevalence survey. Antimicrob Resist Infect Control 2023;12(1):103. https://doi.org/10.1186/s13756-023-01306-z        [ Links ]

24. Almansoori N, Parag N. . Antibiotic prescribing patterns in emergency department at regional hospital in South Africa. Afr Health Sci 2021;21(4):1651-1661. https://doi.org/10.4314/ahs.v21i4.19.         [ Links ]

25. Mash B, Fairall L, Adejayan O, et al. A morbidity survey of South African primary care. PLoS ONE 2012;7(3):e32358. https://doi.org/10.1371/journal.pone.0032358        [ Links ]

 

 

Correspondence:
O van Hecke
oliver.vanhecke@ugent.be

Received 24 February 2024
Accepted 21 August 2024