Table 3.

Summary of studies identified

Author, year	Study location	Study design	Study population (sector)	Aims of the study	Intervention	Summary of results
Boyles et al., 201335	Two general medical wards in an academic teaching hospital, Groote Schuur Hospital in the Western Cape	Observational	General medical patients who had been prescribed antibiotics (public)	Introduction of a two-component intervention aimed at reducing antibiotic consumption for better patient outcomes	Antibiotic prescription chart and a weekly antibiotic stewardship ward round	592.0 DDDs were prescribed per 1000 inpatient days (control period) versus 475.8 DDDs per 1000 inpatient days (intervention period) There was a 19.6% decrease in antibiotic volume following the intervention There was a 35% cost reduction in the pharmacy’s antibiotic budget An increase in laboratory tests was noted due to requests for procalcitonin levels There was no difference in inpatient mortality or 1 month readmission rates following the intervention
Boyles et al., 201734	Academic teaching hospital, Groote Schuur Hospital, in the Western Cape	Observational	General medical patients who had been prescribed antibiotics (public)	To report, over 4 years: (i) antibiotic consumption; and (ii) cost of a public hospital ASP in SA	A comprehensive ASP consisting of online education, an antibiotic prescription chart and weekly antibiotic stewardship ward rounds	Total antibiotic consumption fell from 1046 DDDs per 1000 patient days in 2011 (control period) to 868 DDDs per 1000 inpatient days by 2013 and remained at similar levels for the next 2 years following introduction of the ASP Reductions in IV antibiotic use were noted, particularly for ceftriaxone Cost savings on antibiotics over 4 years adjusted for inflation totalled ZAR 3.2 million Laboratory tests and costs increased (total increased cost = ZAR 0.4 million) There was no difference in inpatient mortality or 30 day readmission rates following the interventions
Brink et al., 201632	47 private hospitals operated by the hospital group Netcare Ltd. in seven of the nine South African provinces	Observational	Netcare hospital patients (private)	Assess the implementation of an ASP in a setting with limited infectious disease resources	A pharmacist-driven, prospective audit and feedback strategy for AMS	116 662 patients receiving antibiotics at 47 hospitals during 104 weeks of standardized measurement and feedback were reviewed 7934 pharmacist interventions were recorded An estimated 1 in 15 prescriptions required intervention 3116 (39%) of 7934 pharmacist interventions dealt with excessive antibiotic duration ASP led to a decrease in average antibiotic DDDs per 100 patient days from 101.38 (95% CI 93.05–109.72) in the pre-intervention phase to 83.04 (74.87–91.22) in the post-intervention phase (P < 0.0001)
Brink et al., 201725	34 private hospitals operated by the hospital group Netcare Ltd. in seven of the nine South African provinces	Observational	Netcare hospital patients (private)	To implement an improvement model for PAP	A pharmacist-driven, prospective audit and feedback strategy involving change management and improvement principles	70 weeks of standardized measurements and feedback was conducted 24 206 surgical cases were reviewed Significant improvement (P < 0.0001) in compliance with all process measures from 66.8% (95% CI 64.8–68.7) to 83.3% (95% CI 80.8–85.8)
Bronkhorst et al., 201436	Steve Biko Academic Hospital, Gauteng	Quantitative, cross-sectional, operational, prospective audit	All patients admitted to surgical and trauma ICU wards (public)	To describe the contributions of a clinical pharmacist in a surgical and trauma ICU as evidence for the necessity of an appointment of a full-time clinical pharmacist in the ICU	Implementation of AMS and drafting and implementation of antimicrobial guidelines in the ICU through the introduction of clinical pharmacy (assessment of prescribing patterns, drug-related interventions and time needed to provide clinical pharmaceutical care)	Of the total 181 interventions suggested, 70% were accepted and implemented 15.5% of interventions dealt with untreated medical conditions and, where indicated, 13.8% on inappropriate therapy or course and 11% on inappropriate doses or dosing frequency 28% of the 250 h that the pharmacist spent on the ward was dedicated to pharmaceutical care while 21% was directed toward ward rounds
Chunnilall et al., 201522	Private anonymized hospital, KZN	Quantitative, retrospective analysis	Adult ICU patients (private)	To evaluate the prescribing patterns and adherence to STGs, EML, SAMF and IDSA guidelines	Surveillance of prescribing patterns and adherence to STGs, EML, SAMF and IDSA guidelines	28.8% of patients (n = 226) received antibiotics in ICU 58.5% (n = 131) required antibiotics Of these, 70.2% were prescribed treatment consistent with the guidelines Doses were correct in 91.1% of the sample Evidence of microbiological investigations in 61.2% of patients De-escalation was noted in only 13.1% of the 70.8% of cases 41.1% of patients received an antibiotic prescription without indication There was a lack of microbiological verification in 38.8% of patients, inaccurate drug choice in 29.8% of the subset for whom antibiotics were indicated and incorrect dosing in 8.9% of the subset for whom antibiotics were indicated
du Toit, 201538	Critical care unit of an acute-care private hospital, Stellenbosch, Western Cape	Pre-and post-interventional	Critical care unit patients (private)	To identify the role of the critical care nurse in the implementation of an ASP	None	Nurses can play an important role in the implementation of an ASP and are a cost-efficient resource Nurses should be an essential part of an AMS team Additional team training regarding AMS and IPC is necessary
Gasson et al., 201827	PHC facilities, Cape Town, Western Cape	Retrospective review of medical charts	Patients attending PHC facilities (public)	To evaluate: (i) prescribing in PHC facilities in the Cape Town Metro District; (ii) adherence to current national guidelines; and (iii) reasons why prescriptions were not adherent to guidelines	None	654 cases were included An antibiotic was prescribed in 68.7% Overall guideline adherence was 45.1% There was a significant difference in adherence between facilities Healthcare professional type and patient gender had no significant effect on adherence The main reasons for non-adherence to guidelines were an undocumented diagnosis (30.5%), antibiotic not required (21.6%), incorrect dose (12.9%), incorrect drug (11.5%) and incorrect duration of therapy (9.5%)
Hoffman et al., 201731	Red Cross War Memorial Children’s Hospital, Cape Town, Western Cape	Phase 1: retrospective folder review; Phase 2: post analysis of a multicomponent education intervention	Convenience sample of the first 107 available patient folders; study population included both inpatients and outpatients less than 18 years old (public)	To conduct an MUE as a quality-improvement project on an item that met the following criteria: high expenditure, high usage, high risk or problem prone (inappropriate use)	MUE of nystatin oral drop usage; posters were designed to inform prescribers about the differential diagnoses of OC and contained an algorithm to follow when dealing with OC, based on the PHC level 2014 STG; in-service training sessions were held at various meetings to share the results of the MUE and to distribute the posters	Of the files reviewed, only 24.3% indicated oral thrush as the diagnosis 76% of nystatin prescriptions were for a condition that was not in the STGs 54% of prescriptions were for use during antibiotic therapy, 23% as prophylaxis after liver and renal transplant and 5% following caustic injury of the oesophagus
Mabila et al., 201623	VVMH in Tzaneen, Limpopo Province	Retrospective cross-sectional census study	Paediatric patients (public)	To determine antibiotic prescribing patterns amongst healthcare professionals in a paediatric ward at VVMH	None	Ampicillin (64.7%) was the most commonly prescribed antibiotic, followed by gentamicin (47.4%) The most common conditions for which the antibiotics were prescribed were bronchopneumonia, diarrhoea and dehydration 69.1% of prescriptions were in accordance with STGs 66.1% of antibiotics were prescribed empirically and did not depend on culture, i.e. in most cases no laboratory results were requested
Matsitse et al., 201724	Two correctional centres in the North West Province	Investigational descriptive study, including retrospective and prospective data	One male-only facility housing approximately 1500 inmates as well as awaiting-trial detainees; the second centre housed both male and female inmates (1400); both were PHC facilities (public)	To assess: (i) compliance with the 2008 PHC STGs/EML in the management of sexually transmitted infections; and (ii) potential factors contributing to the compliance and non-compliance with STGs	None	MUS, LAP and GUS were the three most common STIs Doxycycline (95.0%), ciprofloxacin (90.6%) and metronidazole (90.1%) were prescribed for most of the STIs Overall compliance with the 2008 PHC STGs/EML was 75.9% for MUS, 11.4% for LAP and 14.8% for GUS
Messina et al., 201826	Four South African private hospitals (Johannesburg and Pretoria), Gauteng	Retrospective chart review	Adult patients aged >18 years who were prescribed IV colistin for at least 72 h (private)	To evaluate: (i) the current utilization of colistin in four private-sector SA hospitals; and (ii) opportunities to improve the appropriate use of colistin in the future	None	There was 99.0% compliance with obtaining a culture prior to antibiotic therapy, 93.5% compliance with prescription of a loading dose and 98.5% compliance regarding prescription of colistin in combination with another agent; overall composite compliance with the six colistin stewardship process measures was 82.0% Non-compliance related to inappropriate loading and maintenance doses, lack of adjustment according to renal function and lack of de-escalation following culture and susceptibility Significantly shorter durations of treatment were noted in patients who received higher loading doses (P = 0.040) and in those who received maintenance doses of 4.5 MU twice daily versus 3 MU three times daily (P = 0.0027) Compared with patients who survived, more patients who died received the 3 MU three times daily maintenance dose (P = 0.0037; phi coefficient 0.26)
Messina et al., 201533	33 private-sector Netcare hospitals in SA	Prospective multicentre study	Netcare hospital patients (private)	To measure the change in compliance with IV antimicrobial administration within 1 h following implementation of a pharmacist-driven hang-time process-improvement protocol	Pharmacist-led interventions followed by hang-time compliance assessment consisting of five stages	32 985 patients who received new IV antibiotic orders were assessed for hang-time compliance with first doses over a 60 week period 21 069 patients appropriately received first dose IV antibiotics within 60 min after prescription order, i.e. were hang-time compliant Hang-time compliance improved following the pharmacist-led intervention from 41.2% pre-intervention Week 1 (164/398) to 78.4% post-intervention Week 60 (480/612; P < 0.0001) Pharmacists’ patient review time improved during the final 4 weeks (1680) compared to the first 4 weeks (834; P < 0.0001)
Mthethwa and Matjila, 201830	Medunsa Oral Health Centre, Gauteng	Retrospective, cross-sectional descriptive study of medical records	A portion of medical records of patients who received medication from the Medunsa Oral Health Centre (public)	To evaluate the antibiotic prescribing practices of dentists; surveillance of amount and class used and condition for which the antibiotic was prescribed	None	Antibiotics were prescribed for 65.5 % of the participants 59.1% of the prescriptions were for prophylactic use of antibiotics Of these, 65.5% were healthy patients with no history of a systematic illness 10.9% were HIV infected Routine extractions accounted for 54% of the perioperative prophylactic recipients 5.4% of the antibiotics prescribed were in the absence of a diagnosis Amoxicillin 500 mg three times daily was the most frequently prescribed antibiotic
Paruk et al., 201221	South African private and public hospitals	Three-part prospective, descriptive study that included a 1 day point prevalence study to provide a ‘snapshot’ of events in the ICU	Study population comprised public- and private-sector hospitals in SA that were included in part 1 of the National Critical Care Audit; to ensure a true South African representation, all adult and paediatric ICUs in the private and public (tertiary, regional and community level) sectors were included (private and public)	To: (i) document antibiotic prescription practices in public and private ICUs in SA; and (ii) determine their relationship to patient outcomes		Therapeutic antibiotics were initiated in 182 (73.5%) of 248 recruited patients 54.9% received an inappropriate antibiotic initially De-escalation was practiced in 33.3% and 19.7% of the public and private sector patients, respectively Antibiotic duration was inappropriate in most cases An appropriate choice of antibiotic was associated with an 11% mortality, while an inappropriate choice was associated with a 27% mortality (P = 0.01) The mortality associated with appropriate or inappropriate duration of antibiotics was 17.6% and 20.6%, respectively (P = 0.42)
Ramsamy et al., 201329	TICU at the IALCH, a tertiary/quaternary public service institution in Durban, KZN	Prospective descriptive database review	All patients admitted to the TICU, IALCH who underwent mechanical ventilation during the study period (public)	To determine: (i) the spectrum of nosocomial pathogens in a level 1 TICU based on monthly surveillance and how frequently the initial empirical choice of antimicrobials was correct; and (ii) whether ultrabroad-spectrum combination therapy (U-bSCT) was warranted and, when used, how frequently it was actually necessary	Not applicable	Of 227 patients, 106 (46.6%) had 136 culture-positive isolates These included a total of 323 pathogens (201 Gram-negative, 119 Gram-positive, 3 Candida albicans) There were 19 species of Gram-negative pathogens, of which 56% comprised Enterobacteriaceae ESBL production was found in 6 of 31 (19%) Escherichia coli and 6 of 24 (25%) Klebsiella isolates Staphylococcal species accounted for 60% of the Gram-positive isolates, of which 18 were MRSA All Candida isolates were susceptible to fluconazole Despite positive cultures, antimicrobials were not prescribed for 21 patients who had no evidence of sepsis Excluding MDR Acinetobacter isolates, there were 87 (93.5%) appropriate and 6 (6.5%) incorrect prescriptions U-bSCT was employed for 11 patients but was necessary in only 2
Rout et al., 201737	200 bed private hospital in KZN	Qualitative research approach; semi-structured interviews	Intensive/high care unit patients (private)	To understand the perceptions of AMS team members regarding the role of the ICU nurse in the AMS team	None	Participants from the different disciplines of the AMS team felt that the ICU nurse played an important role within the AMS team; four functions were identified as important and included organizational, advocacy, clinical and collaborative roles
van der Sandt et al., 201928	1650 bed public anonymized hospital;358 bed private anonymized hospital, SA	Retrospective chart reviews	Paediatric patients undergoing surgery (public and private)	To evaluate compliance with South African SAP guidelines for paediatric patients undergoing surgery in surgical subspecialties (ENT, colorectal, urology and maxillofacial)	Adherence to the SAP guidelines	224 charts were reviewed (112 from each of the private and public sectors) The majority (P = 1.000) of patients received antibiotics in line with the SAP (77.3% and 100.0%, respectively, from the public and private hospitals) 21.1% and 45.9% of patients, respectively, received antimicrobials without an indication Full compliance to all five of the criteria (correct choice of antibiotic, dose, timing of administration, redosing and duration of use) were not met by either the public or private hospitals The majority of criteria were met in the teaching hospital (three out of five conditions or 58%); there is a need for quality-improvement interventions, surveillance and implementation of SAP guidelines

VVMH, Van Velden Memorial Hospital; KZN, KwaZulu-Natal; IALCH, Inkosi Albert Luthuli Central Hospital; OC, oral candida; MUS, male urethritis syndrome; LAP, lower abdominal pain; GUS, genital ulcer syndrome; SAMF, South African Medicine Formulary; MU, million units; ENT, ear, nose and throat; TICU, trauma ICU.