Strategies to Improve Antimicrobial Utilization with a Special Focus on Developing Countries

1Antimicrobial resistance (AMR) is a high-priority area across all countries, increasing morbidity, mortality and costs, with several studies documenting the clinical and economic impact of AMR. There is considerable resistance to commonly prescribed antibiotics among low- and middle-income countries (LMICs) including Bangladesh, with resistance rates continuing to rise across countries. Rising AMR rates need to be urgently addressed, with the World Bank (2017) believing that even in a low-AMR scenario, the loss on world output due to AMR could exceed US$1 trillion annually after 2030, and potentially up to US$3.4 trillion annually, unless addressed, equivalent to 3.8% of annual Gross Domestic Product (GDP)—with the costs of AMR appreciably exceeding those of the antibiotics prescribed.

2Rising AMR rates are due to a number of factors including rising antimicrobial utilization especially in LMICs, with the overuse of antimicrobials linked with increasing AMR. Alongside this, the availability of substandard, spurious or falsified antibiotics, high rates of self-medication especially in LMICs often for self-limiting diseases, the extensive use of antimicrobials to prevent and treat diseases in animals considering that their manure is often used in food production and for aquaculture, water pollution, lack of hand hygiene and proper sanitation as well as travelling between countries enhances AMR rates. In their recent paper, Van Boeckel et al. (2019) found nearly a 3-fold increase in bacteria showing resistance above 50% to antimicrobials in chickens and pigs in LMICs between 2000 and 2018, with resistance rates expected to rise unless addressed.

3Low vaccination rates against disease have also enhanced antimicrobial use and AMR, with current concerns that vaccination rates are being further compromised by the COVID-19 pandemic, with its resultant impact on patient movement and available facilities as a result of lockdown and on other measures, exacerbated by vaccine hesitancy generally alongside misinformation. We are though seeing approaches to combat this including mobile vaccination clinics.

4The general overuse of antimicrobials has been worsened by the COVID-19 pandemic, with concerns with differential diagnosis between infections such as tuberculosis (TB) and bacterial upper respiratory tract infections (URTIs) from influenza, coughs, fever and COVID-19. Published studies suggest that over 70% of patients with COVID-19 received antibiotics, including broad-spectrum antibiotics, even when not clinically indicated. Typically, less than 10% of COVID-19 patients appear to have bacterial or fungal infections, with some studies suggesting that only 3.2% of COVID-19 infections are co-infections warranting antibiotics. In their recent meta-analysis, Langford et al. (2021) found that overall three-quarters of patients with COVID-19 were prescribed antibiotics, which is significantly higher than the estimated prevalence of bacterial co-infections. Consequently, the authors concluded that unnecessary antibiotic use is high in patients with COVID-19, which will enhance AMR rates.

5Concerns with rising AMR rates and the consequences have resulted in international, regional, national, and local strategies to try and reduce the rises. National initiatives include National Action Plans (NAPs) under the direction of the WHO, building on its Global Action Plan. We are already seeing countries assess their progress of NAPs against agreed targets, and this will continue.

6Alongside this, there are concerns with multi-resistant microorganisms and the implications for patients with TB, human immunodeficiency virus (HIV), malaria, and typhoid, as well as those with co-infections. Early treatment for patients with co-infections with HIV and drug-resistant TB is essential to reduce the significant risk of death. There is also ongoing research to better predict patients with MDR-TB and improve future care. Treatment centers, especially in rural areas, are important to monitor treatment progress and provide access to medicines, to enhance successful management and reduce future cases.

7Overall, we are aware that coordinated activities among all key stakeholders can address over-utilization of antibiotics. Multiple demand-side measures including education of all key stakeholders, prescribing restrictions, and follow-up in pharmacies stabilized or reduced the utilization of antibiotics in former Soviet Union republics including Azerbaijan, the Republic of Srpska, Bosnia and Herzegovina, and Slovenia versus Poland that failed to instigate such measures. In Poland, there was a small but statistically significant average annual increase in total antibiotic consumption between 2007 and 2016.

8We are also aware that there are concerns with high prevalence rates of hospital-associated infections (HAIs) in a number of LMICs, with major challenges surrounding the appropriate use of antibiotics to reduce surgical site infections (SSIs). The instigation of ASPs in hospitals, including undertaking point prevalence surveys (PPS) to identify areas for quality improvement programs, can improve future antibiotic use and reduce AMR. There are though manpower and resource challenges with implementing ASPs in hospitals in LMICs. Having said this, successful ASPs have been introduced in hospitals in LMICs to improve antimicrobial prescribing providing guidance to others. Successful programs have also been introduced in ambulatory care in LMICs to improve prescribing and dispensing of antibiotics for essentially viral infections such as acute respiratory infections (ARIs) and including URTIs. We are also seeing new technologies being developed, including Apps, to provide prescribing advice and other information to physicians and patients to reduce inappropriate use of antibiotics. However, there are concerns with variable adherence to guidelines, possibly exacerbated by guidelines not always considering local resistance patterns, as well as other activities such as community engagement to improve patient care in practice.

9We would like to build on these issues and challenges to discuss potential ways to decrease future AMR rates among LMICs. We have chosen LMICs as they currently have the greatest increase in antimicrobial use, the highest rate of self-purchasing of antibiotics, as well as the highest rate of antimicrobial use in hospitals. In addition, mortality rates from AMR are likely to be highest among LMICs, including those in Africa and Asia, by 2050 unless addressed. Effectively tackling AMR is crucial for the future of all countries, exacerbated by few new antibiotics currently being developed; however, this is counterbalanced to some extent by developments in vaccine technology. We hope our review paper will consolidate current knowledge in key areas to provide guidance on potential ways forward among all key stakeholders across LMICs to reduce AMR rates, especially where there are continued concerns.

10We undertook a narrative review of published papers to document potential ways to enhance appropriate prescribing and dispensing of antimicrobials. We purposely did not undertake a systematic review as an appreciable number of these have already been conducted on different aspects to improve future prescribing and dispensing of antimicrobials. In addition, we did not document the number of publications by country and sector as we are aware that some countries are more prolific in their publications than others, and such analyses may bias the actual situation. We did though divide initiatives up into those directed at hospitals compared with those in ambulatory care. We are aware that issues such as self-purchasing of antibiotics are especially important in LMICs as up to 100% of pharmacies in these countries dispense antibiotics without a prescription despite legislation which can account for up to 93% to 100% of all dispensed antibiotics. This because our principal aim was to provide possible future guidance to all key stakeholder groups within LMICs for potential debate based on the considerable experiences of the co-authors from multiple LMICs. We have successfully used this approach before to stimulate debate in other priority healthcare areas to provide future guidance.

11The World Bank classification was used to break countries down into LMICs or upper income countries where applicable. Similar to previous studies conducted by the co-authors, upper-income countries will only be commented upon where pertinent. This is because we are aware that there are different practices and beliefs, as well as a greater availability of healthcare professionals and resources, in higher- versus middle- and low-income countries. In addition, in LMICs, the costs of medicines can account for up to 60% or more of total healthcare expenditure, much of which will be out of pocket, with potentially catastrophic consequences for families when members become ill.

12There can also be long waiting times for patients to see physicians in primary healthcare centers (PHCs) in LMICs, long distances to travel to PHCs, and pharmacists are typically open later in the day, resulting in patients often seeking their advice as alternatives to physicians. This is helped by the fact that community pharmacists are typically trusted by patients and less expensive than physicians, with patients not having to pay physician fees when the need arises. However, we are aware that there can be concerns with the knowledge of pharmacists and their assistants regarding antibiotics and AMR.

13When documenting interventions that have been undertaken to enhance appropriate prescribing and dispensing of antimicrobials, we will break these down into the 4Es to enhance understanding and comparisons, namely Education, Engineering, Economics and Enforcement. Education includes developing guidelines or formularies, with adherence to well-constructed guidelines increasingly seen as demonstrating good quality care. This also includes educational and monitoring activities surrounding the WHO AWaRe list of antibiotics. Instigating Drug and Therapeutic Committees (DTCs) and ASPs in hospitals to provide educational activities to enhance the rational use of medicines including antimicrobials is also classified as education. Engineering includes organizational or managerial interventions which incorporate prescribing and quality targets. These include the percentage of antibiotics prescribed according to agreed guidance and the WHO AWaRe list, the percentage of patients not prescribed prolonged antibiotics to prevent SSIs as well as the percentage of patients reaching agreed target blood pressure levels. Economics includes financial incentives to physicians, pharmacists, patients or hospitals, for instance providing financial incentives to hospitals to improve their patient safety records and/or reduce payments where there are concerns, incentivizing physicians for attaining agreed prescribing targets including adherence to guidelines, or fining pharmacists for illegally dispensing an antibiotic without a prescription. Enforcement includes regulations by law such as banning the dispensing of antibiotics in pharmacies without a prescription, compulsory generic substitution as well as national policies outlining the need for DTCs in hospitals and how their activities should be monitored.

14No ethical approval was required for this review as we were not dealing with patients. This is in line with previous similar papers published by the co-authors.

15There are a number of key issues that need to be addressed to enhance appropriate antimicrobial use in patients and reduce AMR. We will break our findings down into key areas within hospitals and subsequently ambulatory care before concluding with guidance on potential activities that can be undertaken among all key stakeholder groups to reduce AMR.

16Activities that can be undertaken in hospitals include PPS studies to gain baseline data of current utilization patterns to guide future quality improvement programs as part of ASPs and other activities. Initiatives in ambulatory care center around activities to improve prescribing and dispensing of antimicrobials especially for self-limiting infections such as viral URTIs. These are especially important since, as mentioned, we are aware that up to 100% of pharmacies in LMICs dispense antibiotics without a prescription despite legislation, and this can account for up to 93% to 100% of all dispensed antibiotics. Having said this, HAIs are considered as the most adverse event in healthcare delivery.

17The first consideration within hospitals is to ensure that activities to improve antimicrobial prescribing are given a high priority if this is not already the case. This includes initiatives such as instigating active infection, prevention and control committees (IPC committees), which can be part of antimicrobial stewardship groups. The latter can be part of IPC committees or attached to Drug and Therapeutic Committees (DTCs). In addition, PPS studies can be undertaken and ASPs indicated to improve future antimicrobial use in hospitals (Table 1).

18However, there can be issues with the instigation of IPC committees and their effectiveness in hospitals (Table 1), concerns about the knowledge of ASPs among clinicians in LMICs, concerns with resource issues with establishing active ASPs including manpower, as well as a lack of institutional policies in LMICs. These issues can be exacerbated by a lack of clear commitment nationally despite the instigation of NAPs, which needs to be addressed to improve future antimicrobial use.

19Figure 1 incorporates the core elements of any ASP, with Kakkar et al. (2020) discussing the potential composition of personnel within antimicrobial stewardship groups in LMICs including physicians, pharmacists, microbiologists, infection control personnel as well as hospital administrators. Ideally, there should be a single focal point, with leaders of ASPs typically being either physicians or pharmacists, with pharmacists increasingly helping to drive forward ASPs in LMICs. ASPs can also be part of DTC activities if needed in some hospitals.

20Box 1 contains a number of indicators that have been used to assess the quality of antimicrobial prescribing within hospitals.

21Availability, ready access and subsequent compliance to local or regional treatment guidelines for key infections. This includes the availability of antibiograms within hospitals to guide empiric use whilst waiting for culture and sensitivity test (CST) findings as well as a potential list of restricted antibiotics depending on physician experienceWhether antibiotics prescribed are included in the hospital formulary/national essential medicine listsRationale (indication) for antimicrobial use documented in the patient’s notesWhether therapy was empiric or targeted based on CST findingsDocumentation of the dose in the patient’s notesDocumentation of a stop/review date for given antibiotics in the patient’s notes. This can include the rate of de-escalation to a narrower-spectrum antibiotic where appropriateDocumentation of any missed doses of antibioticsExtent of IV to oral switching (based on the stability of the gastrointestinal tract)Duration of surgical prophylaxisChanges in antibiotic use over time in terms of Defined Daily Doses/100(0) bed days and/or days of therapy/patient-daysWhether active antimicrobial stewardship groups and/or drug and therapeutic committees/Infection Prevention and Control groups in the hospital

22Availability, ready access and subsequent compliance to local or regional treatment guidelines for key infections. This includes the availability of antibiograms within hospitals to guide empiric use whilst waiting for culture and sensitivity test (CST) findings as well as a potential list of restricted antibiotics depending on physician experience

23Whether antibiotics prescribed are included in the hospital formulary/national essential medicine lists

24Rationale (indication) for antimicrobial use documented in the patient’s notes

25Whether therapy was empiric or targeted based on CST findings

26Documentation of the dose in the patient’s notes

27Documentation of a stop/review date for given antibiotics in the patient’s notes. This can include the rate of de-escalation to a narrower-spectrum antibiotic where appropriate

28Documentation of any missed doses of antibiotics

29Extent of IV to oral switching (based on the stability of the gastrointestinal tract)

30Duration of surgical prophylaxis

31Changes in antibiotic use over time in terms of Defined Daily Doses/100(0) bed days and/or days of therapy/patient-days

32Whether active antimicrobial stewardship groups and/or drug and therapeutic committees/Infection Prevention and Control groups in the hospital

33Multiple activities to reduce HAIs include encouraging greater hand hygiene, regular reviews surrounding the use and care of catheters, instigating appropriate strategies to reduce ventilator-associated pneumonia including subgluteal space suctioning before each position change, chlorhexidine gluconate bathing as well as active monitoring of agreed guidelines to reduce HAIs. A number of initiatives have been undertaken to improve hand hygiene in LMICs including those orchestrated by the WHO and others providing direction.

34Multiple activities can also be undertaken to improve antibiotic utilization to reduce SSIs among LMICs. Recent publications have documented that the number of patients receiving their first dose of antibiotics within 60 min of the first incision, with a second dose administered for longer procedures, can be as low as 6.7% of surgical patients in LMICs. This is a concern as studies have shown the risks of SSIs are almost five times higher when antibiotics are administered more than two hours prior to the first incision and almost doubled when administered after the first incision. There are also concerns that extending prophylaxis beyond one day, which is frequent among LMICs with for instance a mean up to 8.7 days post-operatively in Nigeria, does not appear to improve outcomes whilst concurrently increasing the potential for acute kidney injury, Clostridium difficile infections, and AMR as well as increasing costs. Vicentini et al. (2019) in their published study found that adequate antibiotic choices and length of administration were associated with significantly reduced risks of SSIs, with a number of approaches successfully undertaken in LMICs to improve antibiotic use surrounding the prevention of SSIs.

35ASPs as well as other interventions can improve overall utilization of antimicrobials in hospitals as well as reduce costs alongwith improved hand hygiene and other activities to reduce HAIs and SSIs (Table 1). There are a number of areas that can be initial targets for ASP programs to gain their acceptance among key stakeholder groups in hospitals to address barriers. These include developing guidelines for specific indications and monitoring subsequent usage against these guidelines, as well as developing prompts for physicians to regularly review antibiotic prescriptions.

36Table 2 summarizes some of the activities that have been undertaken in LMICs to improve antimicrobial utilization in hospitals, which build on reviews in predominantly high-income countries including the use of computerized decision support systems. This includes (Table 2) activities surrounding SSIs and their impact to provide future guidance. Activities principally centre on educational initiatives, with educational initiatives known to enhance understanding of ASPs in LMICs given concerns about knowledge and activities among students and physicians in a number of LMICs.

37We are aware that ASPs have been successfully instigated in Accident and Emergency Departments in hospitals. However, this typically applies to higher-income countries.

38Other strategies that antimicrobial stewardship groups and DTCs can undertake in hospitals is to develop effective approaches to address antimicrobial shortages, with medicine shortages being an increasing concern across countries. Shortages of antimicrobials can lead to administering less effective, more toxic, and costlier antimicrobials as well as potentially increasing AMR rates. One way to address this is for key personnel in hospitals, including those who are part of ASTs, to agree in advance therapeutic interchange strategies. This can include consultations between prescribers and hospital pharmacists before alternative antibiotics are administered.

39Emergency procurement procedures should also be in place in hospitals alongside regular reviews of stock levels in wards, exacerbated by the current COVID-19 pandemic. Available resources and funding is also essential to reduce the possibility of antimicrobial shortages in hospitals. The Bamako Initiative in sub-Saharan Africa is one such initiative adopted by Ministers of Health to increase access to medicines including antimicrobials through revolving drug funds. However, sustaining earlier gains with this initiative remains a challenge, and alternative drug financing schemes are essential to ensure physicians and patients can access appropriate antimicrobials in hospital to improve outcomes and reduce length of stay.

40We will principally concentrate on ARIs as these are the most common infections seen in ambulatory care in LMICs and where the highest rate of inappropriate prescribing and dispensing takes place. Suboptimal management of ARIs in LMICs is exacerbated by a number of issues including social-cultural issues, diagnostic uncertainty, clinical competency and misconceptions as well as from commercial, patient and time pressures.

41Consequently, there are a number of strategies that can be undertaken in ambulatory care across LMICs to address current issues to improve future antimicrobial utilization. These can principally be divided into those impacting on the prescribing of antimicrobials by physicians for what are essentially viral infections, those strategies that can reduce inappropriate dispensing of antibiotics without a prescription as well as those that can address concerns with multi-resistant organisms for infections such as malaria and TB. In addition, there are a number of strategies that can be implemented to improve vaccination rates including during pandemics.

42Improving antimicrobial utilization in ambulatory care for ARIs is especially important in the current pandemic, with, as mentioned, published studies suggesting that over 70% of patients with COVID-19 are receiving antibiotics, including broad-spectrum antibiotics, even when not clinically indicated, with perhaps as low as 3.2% of infections in ambulatory care warranting antibiotics.

43A number of initiatives have been undertaken across LMICs to improve physician prescribing of antibiotics for ARIs. Table 3 summarizes examples of the many multiple activities that have been undertaken in LMICs to improve antimicrobial prescribing especially for ARIs, principally in education, and their impact.

44Among LMICs, there is an absolute imperative to develop actionable quality and patient safety indicators that better represent current prescribing compared with the traditional WHO/INRUD criteria. This reflects the rise in the number of patients in LMICs, especially sub-Saharan Africa, with both infectious and non-infectious communicable diseases (NCDs) as well as combinations of infectious diseases such as HIV and TB and combinations of infectious diseases and NCDs. Consequently, there is a need to develop more specific quality and patient safety indicators for LMICs to improve future antimicrobial prescribing. This can be achieved using agreed and common coding of data to enable coordinated and transferable learning between countries including patients attending ambulatory care clinics in sub-Saharan Africa. This will be the subject of future research activities among the group.

45These activities build on recent studies, including those undertaken in sub-Saharan Africa, suggesting that adherence to guidelines is a better indicator of the quality of prescribing in ambulatory care in LMICs compared with the current WHO/INRUD criteria. Having said that, it is important that there is only one national reference guideline since contradictions can arise when there are multiple guidelines within a country. It is also essential that any guidelines produced must be readily available in multiple media formats including online and downloadable, as well as in print. National guidelines also need to be combined with educational inputs and auditing of prescribing practices to further improve future prescribing.

46Regular updates of any guidelines, especially online, are also essential to improve adherence rates, and any updates need to be rapidly disseminated. This is being achieved among a number of sub-Saharan African countries through mobile technologies, providing guidance to other LMICs.

47We are aware though that over prescribing of antibiotics can be appreciably more common in telemedicine consultations than in face-to-face visits. Consequently, these consultations also need to be part of any future quality initiatives.

48There are concerns with the extent of inappropriate dispensing of antimicrobials in community pharmacies and drugs stories across LMICs, especially with, as mentioned, such utilization accounting for up to 93% to 100% of all dispensed antibiotics.

49These concerns can be exacerbated by variable knowledge regarding antibiotics and AMR among community pharmacists. Having said this, there are key issues in rural settings in LMICs that need to be considered by Governments and others when reviewing strategies to reduce inappropriate dispensing of antimicrobials. These include the fact that pharmacists maybe the only healthcare professional available and they are trusted. There can also be major issues with affordability among patients in LMICs. Meeting both physician fees and the cost of medicines can have catastrophic consequences for individuals and families in LMICs. Trained pharmacists, assisted by guidelines, can reduce inappropriate self-purchasing of antibiotics in LMICs as well as be leaders of ambulatory care ASPs in their countries to improve future antibiotic utilization especially for infections that are essentially viral.

50Strategies to successfully reduce inappropriate self-purchasing of antimicrobials range from having well-educated pharmacists offering advice (Education) to patients as seen in Kenya, regulations by law restricting the purchasing of antimicrobials without a prescription coupled with fines (Economics and Enforcement), or a combination of these as seen in the Republic of Srpska (Education, Economics and Enforcement).

51The extent of ‘enforcement’ including the potential for fines or closure of pharmacies is vital as this impacts on the outcome of this intervention (Table 4). For instance, in Venezuela, the Government implemented policies to limit self-purchasing of antibiotics of three antibiotic groups. However, there were no public awareness campaigns and ‘Enforcement’ was only via government publications with no follow up. As a result, there was no decrease in antibiotic utilization levels but, in fact, the opposite with an increase. Similarly, in Colombia, whilst initial enforcement of the law in 2005 had a modest impact on overall sales in the first three years (−1.00 DDDs/1000 inhabitants per day), a follow-up study five years after implementation found that 80.3% of pharmacies were still not complying with the law due to lax monitoring, prompting calls for greater enforcement. This contrasts with an appreciable reduction in antibiotics dispensed without a prescription in a number of countries following multiple interventions, e.g., Saudi Arabia (Table 4).

52We see similar examples in other disease areas where the intensity of follow-up of prescribing restrictions (enforcement) appreciably impacts on their effectiveness and outcome in practice.

53However, it can be difficult in practice in a number of LMICs to enforce regulations banning self-purchasing of antibiotics due to manpower issues as well as shortages of antimicrobials within public healthcare facilities. Such activities may also be counter-productive in rural areas in LMICs and we have seen that trained pharmacists can help reduce inappropriate dispensing of antibiotics.

54Monitoring pharmacy dispensing activities through mobile technologies as well as potentially implementing information technology (IT) surveillance systems to track antibiotics through the supply chain are also potential strategies to reduce inappropriate dispensing.

55Adoption of such tactics may mean re-looking at pharmacists’ knowledge regarding AMR and its causes, with the potential for increased training if needed starting in pharmacy school and continuing post-qualification. In addition, potentially ensuring through regulation and other mechanisms that any antibiotic dispensed for ARIs and other common conditions are principally from the Access group of antibiotics and not from the WHO ‘Watch’ or ‘Reserve’ list.

56The impact of programs to reduce inappropriate prescribing and dispensing of antimicrobials for essentially viral infections such as ARIs can be enhanced by concurrently targeting patients and the public. This is because multiple studies have shown that some people can put considerable pressure on healthcare professionals to prescribe and dispense antibiotics for ARIs. Knowledge about antibiotics and AMR can vary though among patients and the public, and they can have strong beliefs about the effectiveness of antibiotics for them even for viral infections, which adds to the pressure to prescribe or dispense antimicrobials for ARIs.

57Similar to the situation with physicians and pharmacists, multifaceted approaches are needed to change attitudes and behaviors among patients and caregivers. However, such programs have had variable degrees of success for countries including both high-income countries and LMICs. Published studies, particularly among high-income countries, suggest that a number factors can enhance the success of any antibiotic awareness campaign. These include (i) the use of carefully designed simple key messages; (ii) targeting a wide audience including patients and their families; (iii) designing key messages for all key stakeholder groups for completeness; (iv) using mass media including social media to raise awareness; and (v) repeating key messages especially for new parents and others.

58Two systematic reviews have suggested that campaigns among patients can significantly reduce antibiotic utilization. McDonagh et al. (2018) found that parent education reduced antibiotic prescribing for ARIs by 21%. Cross et al. (2017), mainly including studies from high-income countries, also found that the majority of studies reported reductions in antibiotic utilization. The reduction was greater than 14% in the prescribing of antibiotics for RTIs, up to a maximum reduction of 30%.

59However, the impact of patient education initiatives in LMICs have not yet been fully elucidated, nor their costs and cost-effectiveness. Consequently, caution is still needed before fully embarking on such campaigns in LMICs unless these are part of research programs.

60There are concerns that patients with TB living in rural areas, and without regular supply of medicines, will have poorer treatment outcomes. Potential reasons for this include insufficient funds for transport to treatment centers despite the medicines often being provided free of charge alongwith missing out on regular reviews of medicine use addressing concerns including compliance with prescribed medicines, similar to the situation with NCDs. Poorer outcomes and lack of adherence to treatments will enhance resistance rates.

61The overall burden of TB has been declining at an annual average of 2%; however, the number of patients with drug-resistant TB (DR-TB) is increasing. Aminoglycosides, such as kanamycin, and some polypeptides, such as capreomycin, are used to treat DR-TB. However, whilst aminoglycosides are inhibitors of prokaryotic protein synthesis at commonly accepted therapeutic concentrations, they can affect the protein synthesis of cells at larger concentrations, leading to toxicity such as ototoxicity, vestibulotoxicity, and nephrotoxicity. Hearing loss though can now be monitored using an ototoxicity grading system within a mobile app to assist health professionals in assessing patients for ototoxicity. Subsequently, monitoring progression of hearing loss to improve planning for any changes in care. Such developments are likely to continue to improve the future use of aminoglycosides.

62There are also concerns with rising resistance rates to antimicrobials used to treat malaria since for instance when resistance of P. falciparum to sulfadoxine-pyrimethamine was greater than 10%, the risk ratio for malaria infections was 5.9 times higher compared to a period of time with low resistance rates, and the risk of death from malaria 10.8 times higher.

63Developing resistance to treatments for HIV also increases mortality with studies from sub-Saharan Africa indicating that when pre-treatment drug-resistance levels are evident but below 10%, AMR will be responsible for an estimated 710,000 AIDS deaths and 380,000 new infections by 2030. This increases to an estimated 890,000 AIDS deaths and 450,000 new infections by 2030 attributable to HIV when resistance rates are above 10%.

64Strategies to address concerns with resistance to treatments for HIV, malaria and TB include initiatives to ensure regular access to treatment facilities, including expanding these in rural areas given concerns with travel costs and time off work, as well as regular access to treatments ideally provided free of charge. Alongside this, regular monitoring of adherence rates knowing adherence can be a major issue in some patients. Strategies to improve adherence rates include patient and staff education addressing pertinent issues of anxiety and depression with their condition, encouraging clinic visits, addressing socioeconomic factors where possible as well as increasing the use of digital technologies including mobile phones and video-observed therapy.

65Concurrent with this, we are aware of groups in South Africa undertaking research on HIV reservoirs and using locally discovered potent, broadly neutralizing antibodies to neutralize HIV in the reservoir to render patients non-infectious as a cure strategy. Recent promising data from primates indicate the potential value of this approach and clinical trials are being planned.

66We are also aware of ongoing research that provides an expansive dataset of compounds that could be redirected for antimalarial development and also point towards proteins that can be targeted in multiple parasite life cycle stages to improve the management of patients with malaria resistant to current treatments. We will be monitoring these developments and their implications.

67Reduced vaccine uptake including vaccine hesitancy fueled by misinformation is a concern in view of the impact on AMR, morbidity and mortality (Table 5).

68This increasing recognition of the role of vaccines in reducing antimicrobial use and AMR is reflected in the recent WHO Action Framework to enhance current vaccination rates, encourage greater knowledge and dissemination on the role of vaccines to reduce AMR as well as encourage the development of new vaccines to further reduce AMR.

69Addressing mis- and dis-information surrounding vaccines is critical to reduce future morbidity and mortality arising from pandemics. We are already seeing social media companies such as Facebook and Twitter tightening up on accounts that repeatedly publish misinformation about COVID-19, building on activities among some African countries to fine companies for spreading misinformation, and such activities are likely to continue. This builds on activities by the WHO to join forces with Governments to address the level of misinformation surrounding COVID-19 as an example as well as Governments funding projects to specifically address misinformation surrounding vaccines for COVID-19 to enhance their uptake. Inclusivity is essential to engage with community groups and patient organizations, and ensure that clear public health messaging is delivered that is sensitive to age, culture, socioeconomic context and education, and to build population resilience against AMR behaviors.

70Potential ways forward, building on the activities of the WHO and Governments, especially with proposed medicines for COVID-19 including hydroxychloroquine, lopinavir/ritonavir and remdesivir that failed to live up to their preceived promise, include instigating a culture of evidence-based medicine among all key stakeholder groups starting in universities and continuing post-qualification. Such activities would have helped reduce shortages and price rises of hydroxychloroquine and antibiotics arising from the initial studies as well as help reduce inappropriate prescribing of antibiotics.

71Governments and others working with pertinent patient organizations within countries also need to proactively address misinformation, especially if, as mentioned, this increases future morbidity, mortality and costs, as seen with misinformation regarding hydroxychloroquine and unwarranted fears about vaccines. Transparent communication and appropriate public health messaging, in collaboration with patient groups and community leaders, are key for increasing the trust of the community and improving vaccination rates where concerns.

72Suggested strategies to improve the utilization of antimicrobials among LMICs and thereby reduce AMR have been divided into short- and long-term initiatives to provide future guidance (Table 6). These build on the potential for expanding the use of electronic health records and electronic prescribing as well as digital technologies, including clinical decision support systems (CDSS), to improve appropriate use of antibiotics across countries and sectors in LMICs. Such developments should be welcomed given the increasing complexity of medical decision making and, as mentioned, the high percentage of patients especially in sub-Saharan Africa with co-morbidities including HIV, malaria and TB, as well as co-morbidities with both infectious and non-infectious diseases.

73Effectively addressing inappropriate antibiotic prescribing is particularly challenging since the majority of antibiotic prescribing in LMICs is undertaken by general clinicians and their assistants who are not experts in infection management and with often influence from pharmaceutical companies in the absence of continual professional development. This is compounded across Africa due to high workloads as well as limited consultation times and available resources, driven mainly by a shortage of healthcare professionals and the high prevalence of infectious diseases. Recent evidence is promising though about the feasibility and effectiveness of digital tools to improve future antibiotic prescribing and dispensing.

74AMR continues to be a high priority across countries in view of its impact on morbidity, mortality and costs, driven by the overuse of antimicrobials across all sectors including in animals and for the treatment of infectious diseases in patients. The overuse of antimicrobials in humans has been exacerbated by the recent COVID-19 pandemic, where, as mentioned, typically less than 10% of COVID-19 patients have bacterial or fungal infections, with as low as 3.2% of COVID-19 infections warranting antibiotics. There are similar concerns with current low vaccination rates in a number of countries including current vaccine hesitancy surrounding the COVID-19 vaccine alongwith misinformation surrounding treatments for COVID-19, all of which negatively impact on costs and outcomes.

75Despite issues of resources including personnel, a number of activities can be undertaken in LMICs to improve future prescribing of antibiotics. These center around the instigation of ASPs in hospitals including PPS studies to identify areas for quality improvement, with a number of quality indicators now developed to monitor the appropriateness of future prescribing (Box 1). Strategies to improve future antibiotic prescribing in hospitals include those to reduce inappropriate timing and duration of antibiotic prescribing to prevent SSIs. Table 1 and Table 2 document a number of examples of initiatives that have been successfully instigated in hospitals in LMICs to improve future antibiotic prescribing, thereby providing exemplars to others.

76Successful strategies have also been implemented among physicians in LMIC to reduce inappropriate prescribing of antibiotics for essentially viral infections (Table 3), with patients playing a key role. Patients and pharmacists are also important to reduce inappropriate dispensing of antibiotics without a prescription (Table 4). However, recognizing that in rural LMICs, especially where there are high patient co-payment levels, community pharmacists maybe the only healthcare professional accessible and available. Community pharmacists are often more accessible than ambulatory care physicians in LMICs, and there is no need for physician co-pays, which combined with the cost of treatments can be catastrophic for some families.

77Healthcare professionals, including community pharmacists, along with patient organizations have a key role to play to encourage evidence-based approaches to healthcare, thereby reducing the impact of misinformation. Misinformation has been a real concern during the current COVID-19 pandemic resulting in for instance reduced uptake of the vaccines and the implications for this and other vaccines (Table 5), with a number of strategies identified to address this (Section 3.2.5).

78A number of strategies and initiatives have also been identified to reduce levels of multi-drug resistance organisms to HIV, malaria, and TB, given the resultant impact on morbidity and mortality. Key areas include the provision of treatments free of charge, ready access to healthcare facilities as well as initiatives to enhance adherence to prescribed treatments.

79Potential strategies for all key stakeholders were subsequently consolidated into short-, medium- and long-term activities to provide future direction (Table 6). The key is Government commitment through NAPs and other activities to drive forward future initiatives. We will be monitoring the situation given continued concerns with AMR and its resultant impact on mortality as well as GDP.

80We are aware of a number of limitations with this review paper. These include the fact that we did not undertake a systematic review for the reasons discussed. We also did not break the number of examples down by country or care setting as our objective was to provide pertinent examples to guide others. In addition, we are aware that some developing countries are more pro-active with their research than others, potentially biasing the findings. Despite these limitations, we believe that our findings and suggestions are robust, providing future direction.

81In conclusion, reducing AMR should be a high priority across countries given its clinical and economic impact. Reducing AMR rates requires multiple coordinated activities across sectors driven by Governments and others. This is essential given limited new antimicrobials being developed, although compensated to some extent by developments in vaccine technologies. This will also require strategies to address high rates of vaccine hesitance that exist in a number of countries as seen in the recent COVID-19 pandemic. A coordinated approach including all key stakeholder groups is also essential to minimize misinformation and maximize the impact of future interventions to reduce AMR rates.