Background
Antimicrobial resistance (AMR) is a global public health problem associated with approximately 5 million deaths in 2019.1 Studies cite that countries with lower social development status often have higher mortality rates related to AMR. 1,Reference Harun, Sumon and Hasan2 Unfortunately, the COVID-19 pandemic has negatively impacted the efforts to control AMR in hospitals. 3–Reference Ghosh, Bornman and Zafer6 Actions to halt the emergence of AMR, including the prevention of infection and the proper use of antimicrobials, should be reinforced in all healthcare settings. Reference Aslam, Khurshid and Arshad7–10
Since the first global AMR report by the World Health Organization was released in 2014, several health agencies have released strategies that include the implementation of antimicrobial stewardship programs (ASPs). ASPs are included as an important strategy within health institutions to promote the adequate use of antimicrobials, ultimately, slowing the emergence of AMR. Reference Majumder, Rahman and Cohall11 Developing countries risk AMR for its high consumption of antimicrobials, moreover, another major obstacle are their lack of structured ASPs. Reference Gebretekle, Haile Mariam and Abebe12–Reference Menezes, Santos and Costa16 Furthermore, the operation mode of ASPs often varies due to differences in available resources, workforce, technology, local medical culture, and insufficient support from hospital administration. Reference Howard, Pulcini and H17
The CDC, Centers for Disease Control and Prevention, suggests that pharmacists are the ideal professional to lead implementation efforts to improve the use of antibiotics. Studies have shown the importance of infectious diseases (IDs) pharmacists in the success of the ASPs. Reference Bessesen, Ma and Clegg18–Reference Heil, Kuti, Bearden and Gallagher20 However, the small number of certified ID specialists limit their engagement in these programs. Clinical pharmacists could play a key role in managing the use of antimicrobials, as they are directly involved in dispensing and monitoring these drugs. However, in reality, in many countries, especially in low- and middle-income countries (LMICs), their role in antimicrobial management is not as developed. Reference Gebretekle, Haile Mariam and Abebe12,Reference Sakeena, Bennett and McLachlan21 Given this scenario, the aim of our review is to identify and describe pharmacist’s interventions in LMICs on ASPs.
Methods
Systematic literature review and inclusion and exclusion criteria
This review was conducted according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) statement. Reference Moher, Liberati, Tetzlaff and Altman22 The steps to choose articles relevant for this review were: identification of the theme, development of the guiding research question, definition of inclusion and exclusion criteria, and choice of databases for the search. Then, we evaluated the abstracts, selected them according to their eligibility and contributions to the topic, interpreted and compiled the results, and synthesized the data that were described in the discussion stage. The guiding research questions were: “What are the interventions led by pharmacists in antimicrobial stewardship programs in LMICs?” The established inclusion criteria were studies performed in inpatient settings and studies that focused on the role of pharmacists in hospitals located in LMICs with emphasis on pharmacist-led interventions on ASP. To classify countries in LMICs, we used the World Bank classification as a reference. 23 Letters to the editor, commentaries, conference abstracts, and review articles were excluded.
Search strategy
The literature searches were performed in Medical Literature Analysis and Retrieval System Online (MEDLINE/PubMed). Considering that in the last decade, different health agencies have published on the importance of implementing ASPs, this review covers publications from January 1, 2012, to November 4, 2023, using the following keywords: “antimicrobial stewardship program or antibiotic stewardship,” “pharmacist or pharmacist-led or pharmacist-driven or clinical pharmacist,” “low- and middle-income countries or developing countries.” The terms were combined with each other through the Boolean operators “AND” and/or “OR.” We also reviewed the reference lists of retrieved articles to identify studies that were not identified from the preliminary literature searches.
Data abstraction and quality assessment
Titles and abstracts of all articles were screened to assess whether they met inclusion criteria. Abstract screening was performed by one reviewer (TAM). Two reviewers (TAM and GRJ) independently abstracted data from each article which included: country, population, study design, study period (months), primary objectives, interventions carried out by pharmacists, evaluated antimicrobials (dosage, indication, duration of treatment, de-escalation, therapeutic duplicity, adherence to guidelines), and final outcomes, such as AMR, costs, and length of stay. Reviewers settled disagreements by consensus. We assessed the risk of bias using the Downs and Black scale. Reference Downs and Black24 Quality analysis was performed independently, and discrepancies were resolved after consensus among reviewers.
Results
In total, 522 citations published between January 2012 and October 2023 were identified in the database search. After screening by title and abstracts, 452 articles were excluded. Of the 70 articles remaining for full reading, 50 articles were excluded, and 4 additional studies were added by screening the references of the articles. In sum, a total of 24 articles met the inclusion criteria (Figure 1).
Figure 1. PRISMA flow chart showing the review selection process.
Characteristics of included studies
We included twenty-four studies in the final review Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25–Reference Zhou48 (Table 1). All studies were non-randomized; most of them were quasi-experimental studies Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29–Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Holguín, Amariles, Ospina, Pinzón and Granados33,Reference Jantarathaneewat, Camins and Apisarnthanarak34,Reference Kuruvilla, Madhan and Puttaswamy36,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38,Reference Messina, van den Bergh and Goff40,Reference Wang, Zhou, Liu, Liu, Liu and Li43–Reference Yu, Liu and Qu45,Reference Zhou48 , five were prospective cohort studies Reference van den Bergh, Messina and Goff27,Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Li, Cheng and Zhang37,Reference Zhang, Qian and Zhang46,Reference Zhang, Li and He47 , three were retrospective observational studies Reference Cheng, Dang and Li28,Reference El Masri, Haddad and Saad39,Reference Wang, Wang and Yu42 , and two were pre-post intervention. Reference Kerr, Sefah and Essah35,Reference Nampoothiri, Sudhir and Joseph41
Table 1. Summary of studies on Antimicrobial Stewardship Programs (ASPs) with pharmacist-led or pharmacist-driven classified by country or region
Note. D&B, Downs and Black; DDD, defined daily dose; DOT, days of therapy; CPC, clinical pharmacist consultation; AMR, antimicrobial resistance; MRSA, methicillin-resistant Staphylococcus aureus; E. coli, Escherichia coli; ASP, antimicrobial stewardship program; ICU, intensive care unit; CAP, community acquired pneumonia; CwPAMS, Commonwealth Partnerships for Antimicrobial Stewardship; LOS, length of stay; ID pharmacists, pharmacists specialists in infectious diseases; nonspecialist, pharmacists nonspecialists in infectious diseases.
*Assessment by the authors as the study design was not specified in the reference.
Nearly half of the studies included in our review were conducted in China (10 studies) Reference Cheng, Dang and Li28,Reference Du, Li and Wang30,Reference Li, Cheng and Zhang37,Reference Wang, Wang and Yu42–Reference Zhou48 , two in India Reference Kuruvilla, Madhan and Puttaswamy36,Reference Nampoothiri, Sudhir and Joseph41 , and two in Thailand Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Jantarathaneewat, Camins and Apisarnthanarak34 . Five studies were performed in Africa Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference van den Bergh, Messina and Goff27,Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Kerr, Sefah and Essah35,Reference Messina, van den Bergh and Goff40 , three studies in Latin America Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29,Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Holguín, Amariles, Ospina, Pinzón and Granados33 , and two in the Middle East Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38,Reference El Masri, Haddad and Saad39 . Most of the studies were performed between 2019 and 2022 Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference van den Bergh, Messina and Goff27,Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29–Reference Kerr, Sefah and Essah35,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38,Reference El Masri, Haddad and Saad39,Reference Nampoothiri, Sudhir and Joseph41–Reference Xu, Huang and Yu44,Reference Zhang, Qian and Zhang46–Reference Zhou48 , three in 2023 Reference Cheng, Dang and Li28,Reference Kuruvilla, Madhan and Puttaswamy36,Reference Yu, Liu and Qu45 , two in 2015 Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Messina, van den Bergh and Goff40 and one in 2017 Reference Li, Cheng and Zhang37 (Figure 2).
Figure 2. Distribution of articles by year of publication and geographic region Sankey Diagram.
The study duration varied from four to seventy-seven months. In sixteen studies, the pharmacists implementing the interventions were not specialized in ID. Reference Abubakar, Syed Sulaiman and Adesiyun26–Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Kerr, Sefah and Essah35–Reference Messina, van den Bergh and Goff40,Reference Xu, Huang and Yu44,Reference Yu, Liu and Qu45,Reference Zhang, Li and He47,Reference Zhou48 Six studies described ID pharmacists implementing the interventions Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Holguín, Amariles, Ospina, Pinzón and Granados33,Reference Nampoothiri, Sudhir and Joseph41–Reference Wang, Zhou, Liu, Liu, Liu and Li43 , while two studies had both nonspecialists and ID pharmacists. Reference Jantarathaneewat, Camins and Apisarnthanarak34,Reference Zhang, Qian and Zhang46 All included studies assessed patients admitted to tertiary care hospitals. Ten studies were conducted in medical and surgical wards, five in ward units and intensive care units, two in intensive care units, and seven did not specify the department. Most studies evaluated the consumption of broad-spectrum antimicrobials Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference Cheng, Dang and Li28,Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29,Reference Gebretekle, Haile Mariam and Abebe Taye31–Reference Jantarathaneewat, Camins and Apisarnthanarak34,Reference Kuruvilla, Madhan and Puttaswamy36–Reference El Masri, Haddad and Saad39,Reference Nampoothiri, Sudhir and Joseph41–Reference Yu, Liu and Qu45,Reference Zhou48 with exception of seven studies Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference van den Bergh, Messina and Goff27,Reference Du, Li and Wang30,Reference Kerr, Sefah and Essah35,Reference Messina, van den Bergh and Goff40,Reference Zhang, Qian and Zhang46,Reference Zhang, Li and He47 which did not specify the antimicrobials involved. The predominant interventions in the studies were education and training, mainly of medical teams and clinical pharmacists (11 studies) Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29,Reference Du, Li and Wang30,Reference Kerr, Sefah and Essah35,Reference El Masri, Haddad and Saad39–Reference Xu, Huang and Yu44,Reference Zhou48 , audit and feedback (11 studies) Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference van den Bergh, Messina and Goff27,Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29–Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38–Reference Nampoothiri, Sudhir and Joseph41,Reference Wang, Zhou, Liu, Liu, Liu and Li43,Reference Yu, Liu and Qu45 , development of protocols addressing common infections like community acquired pneumonia, targeted de-escalation of carbapenems and treatment guides (10 studies) Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25–Reference van den Bergh, Messina and Goff27,Reference Díaz-Madriz, Cordero-García and Chaverri-Fernández29,Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Kerr, Sefah and Essah35,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38,Reference El Masri, Haddad and Saad39,Reference Nampoothiri, Sudhir and Joseph41,Reference Zhang, Li and He47 , and daily rounds (6 studies) Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Cheng, Dang and Li28,Reference Du, Li and Wang30,Reference Kuruvilla, Madhan and Puttaswamy36,Reference Li, Cheng and Zhang37,Reference Xu, Huang and Yu44,Reference Yu, Liu and Qu45,Reference Zhou48 . Most of the studies employed more than one intervention (Table 2).
Table 2. Description of the interventions carried out
Most of the outcomes reported involved reduction in antimicrobial consumption (12 studies) Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25–Reference Du, Li and Wang30,Reference Li, Cheng and Zhang37,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38,Reference Nampoothiri, Sudhir and Joseph41–Reference Xu, Huang and Yu44 , cost reduction (6 studies) Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38,Reference Wang, Zhou, Liu, Liu, Liu and Li43,Reference Xu, Huang and Yu44,Reference Zhou48 , reduction in the spectrum of action of antibiotics (6 studies) Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Holguín, Amariles, Ospina, Pinzón and Granados33,Reference Li, Cheng and Zhang37,Reference El Masri, Haddad and Saad39,Reference Yu, Liu and Qu45,Reference Zhou48 , increased adherence to protocols (3 studies) Reference Jantarathaneewat, Camins and Apisarnthanarak34,Reference Kerr, Sefah and Essah35,Reference Messina, van den Bergh and Goff40 , and reduction in hospitalization days (3 studies) Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Mahmoudi, Sepasian, Firouzabadi and Akbari38 . Of the twenty-four studies included, only seven of them evaluated mortality as an outcome from which two observed a significant reduction. Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Li, Cheng and Zhang37 Out of all the studies, only three studies identified reduction in AMR Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Li, Cheng and Zhang37,Reference Yu, Liu and Qu45 .
Regarding the quality assessment scores of the 24 included studies, five of them were considered good (19–23 of 28 possible points) per the Downs and Black quality tool Reference van den Bergh, Messina and Goff27,Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Xu, Huang and Yu44,Reference Zhang, Qian and Zhang46,Reference Zhang, Li and He47 . More than half of the studies (17 studies) were considered fair (14–18 points) Reference Apisarnthanarak, Lapcharoen, Vanichkul, Srisaeng-Ngoen and Mundy25,Reference Abubakar, Syed Sulaiman and Adesiyun26,Reference Cheng, Dang and Li28–Reference Gebretekle, Haile Mariam and Abebe Taye31,Reference Holguín, Amariles, Ospina, Pinzón and Granados33,Reference Jantarathaneewat, Camins and Apisarnthanarak34,Reference Kuruvilla, Madhan and Puttaswamy36–Reference Messina, van den Bergh and Goff40,Reference Wang, Wang and Yu42,Reference Wang, Zhou, Liu, Liu, Liu and Li43,Reference Yu, Liu and Qu45,Reference Zhou48 , and two studies were considered poor quality (≤13 points) Reference Kerr, Sefah and Essah35,Reference Nampoothiri, Sudhir and Joseph41 .
Discussion
In our systematic review, we found that ASPs are associated with decreased antibiotic utilization in multiple studies conducted in LMICs. Between a variety of different ASPs tasks and interventions, the main ones were audit and feedback, education, training, implementation of protocols, and daily rounds, which were mostly led by non-ID-trained pharmacists. Previous systematic literature reviews have also shown that pharmacist-led ASP improved the correct antibiotic prescription and adherence to guidelines in countries with high rates of inappropriate prescription. Reference Otieno, Campbell, Maley, Obinju Arunga and Otieno Okumu49–Reference Kooda, Canterbury and Bellolio51 However, none of them focused only on LMICs. In our systematic review, we observed that clinical pharmacists are key professionals well prepared to manage these programs and are successfully involved in implementing ASPs interventions in LMIC inpatient settings. They play an important role by following protocols and contribute positively to the outcomes of these programs.
In most studies, the interventions were conducted by clinical pharmacists who were not specialized in IDs, and the outcomes obtained in their interventions were favorable, contributing to the reduction in the use of antimicrobials. For example, one study demonstrated the positive impact of a multifaceted clinical pharmacist-led program on antimicrobial stewardship through segmented regression analysis. A reduction in the quantity of antibiotic use after the ASPs intervention was seen, additionally, there was a significant decrease in the proportion of patients receiving more than one antibiotics and the average length of hospital stay. Reference Du, Li and Wang30 Through the review, we identified that more than half of programs did not have ID specialization, this can be justified by the lack of specific specializations degrees available in these countries. There is no formalization and obligation from hospitals to have an ID specialized pharmacist to manage the ASPs. In addition to this, other factors that prevents the progress of ASPs in some regions are the lack of adequate resources, technology and laboratory support, workforce, and support from leadership. Reference Howard, Pulcini and H17,Reference Wang, Zhou, Liu, Liu, Liu and Li43,Reference Xu, Huang and Yu44
In developing countries, such as Brazil, clinical pharmacists are trained in a wide range of activities and there are no postgraduate options in ID as in the United States. In most of the included studies on LMICs, management programs were led by pharmacists without specialization in ID. The ID training consists of training focused on antimicrobial therapy combining optimization strategies and broad knowledge in pharmacokinetics, pharmacodynamics, microbiology, and IDs. However, most of these pharmacists received basic training in antimicrobial management by the hospital infection control service’s medical team, qualifying them to lead different interventions. Despite the limited availability of ID-trained pharmacists as well as ID physicians in some of these settings, clinical pharmacists already obtain successful outcomes in reducing AMR, encouraging further investments in education and training.
In the United States, ASPs allows multiple pharmacist-led activities, led by both clinical pharmacists and ID pharmacists. This review shows that antimicrobial stewardship interventions led by clinical pharmacists are an alternative solution found by LMICs. Several studies evaluated the effectiveness of clinical pharmacists’ interventions, described the development of ASPs, and elucidated the role of pharmacists in these programs. The outcomes and benefits achieved by these practices were determined and, consequently, multiple pharmacist-led interventions and protocols were identified. Reference Bessesen, Ma and Clegg18,Reference Mahmood, Gillani, Alzaabi and Gulam50–Reference Turner, Wrenn and Sarubbi55
We found an increase in publications on ASPs in recent years from LMICs, mainly in countries classified as upper middle income, such as China, South Africa, and some countries in Latin America, demonstrating the importance of the contribution of pharmacists in the rational use of antimicrobials and in the fight against bacterial resistance. There were differences observed between the studies. In China, ASPs are more widespread as it is a country with more resources, receiving greater support from local authorities, so much so that the majority of the reviewed studies are from this country. They mainly involved auditing actions, feedback, and daily rounds. In Africa, we noticed that there is an incentive through partnerships with developed countries, helping to disseminate the importance of ASPs by providing support through training and implementation of protocols. Reference Kerr, Sefah and Essah35 In Latin America, we noticed mainly educational activities with two Colombian studies having the evaluation of antimicrobial prescriptions carried out by ID specialized pharmacists Reference Holguín, Amariles, Ospina, Pinzón and Granados32,Reference Holguín, Amariles, Ospina, Pinzón and Granados33 , demonstrating that they receive support from the management team of these institutions.
In most of the studies there was either an overall reduction in the use of antimicrobials or in the use of a specific type of antibiotic, depending on the pharmaceutical intervention carried out. Moreover, as a consequence of the several interventions carried out by pharmacists, such as shorter treatment duration, discontinuation of unnecessary antibiotics, de-escalation from broad-spectrum to lower-spectrum drugs, switch therapy, and reduced duration of hospitalization, there was an overall cost reduction. Other outcomes, such as the decrease in mortality rate and in the microbial resistance rate were also observed in some of the studies but not all were able to achieve statistical significance. Although it is suggested that ASPs can prevent the increase in bacterial resistance rates, studies with longer intervention periods are needed to demonstrate the real impact.
Regardless of the country and the regional differences observed, different interventions were performed simultaneously to achieve positive results. It is possible to verify that hospitals with well-structured ASPs in LMICs have a variety of interventions and protocols that can be implemented and successfully managed by pharmacists. Reference Li, Cheng and Zhang37,Reference Wang, Zhou, Liu, Liu, Liu and Li43,Reference Zhou48 It is worth mentioning that in recent years, some countries have increased the participation of clinical pharmacists in ASPs Reference Gebretekle, Haile Mariam and Abebe12 and this increase is likely underestimated. Education, training, and protocol development are, in short, the components for initial implementation Reference Otieno, Campbell, Maley, Obinju Arunga and Otieno Okumu49,Reference Gebretekle GB, Mariam DH, Mac S56,Reference Almeida, Baptista and Lima57 These interventions combined with auditing and feedback are present in most of the studies analyzed in this review. The latter being necessary to compose strategies and lead to better outcomes in the ASPs.
Our study had some limitations. First, the search was carried out only in one database. Second, no meta-analysis was carried out. Third, most studies were classified as having poor quality by the Downs and Black quality tool. Lastly, there was great heterogeneity between studies, such as different populations, great variation between study periods, and no description of which antimicrobials were evaluated, and how interventions were carried out. Despite the importance of the pharmacist and the positive impact on patient outcomes, there is still a lack of studies with strong evidence to corroborate this practice. Therefore, our review suggests that further studies should be conducted in this regard, mainly in LMICs.
In LMICs, we would expect that most pharmacists would not have expertise in ID. Nonetheless, 66% of studies within our review had pharmacists with comparable skill, which is a positive and promising number. It shows that, in recent years, hospital management has supported this cause by implementing ASPs and focusing on training and education of pharmacist in ID; therefore, contributing to the rational use and reduction of AMR. In our perception, ASPs must have at least one ID pharmacist who disseminates their knowledge by conducting trainings focused on strategic interventions led by clinical pharmacists.
Our findings reinforce the importance of clinical pharmacists´ participation in this context and the importance of a global need to increase ID training to continue to expand the impact of these healthcare professionals on ASPs, especially in LMICs.
Financial support
This study was not funded.
Competing interests
All authors report no conflict of interest relevant to this article, except MH who is contributor of Up To Date, consultant for the World Health Organization and member of SHEA Antimicrobial Stewardship Committee.
Open access
