Is it worth analysing antibiotic use?

Abstract

Rational antibiotic therapy requires analyses that enable the monitoring of the use of anti-infective medicines in hospitals, compliance with procedures and the achievement of set targets. One of the most popular methods is quantitative retrospective analysis based on Defined Daily Doses (DDD). This article outlines the key issues relating to how to carry out such an analysis, as well as the purpose for which it is carried out.

Keywords

DDD, monitoring of antibiotic use, retrospective quantitative monitoring, analysis of antibiotic therapy

According to the European Centre for Disease Prevention and Control, antibiotics are one of the most important therapeutic discoveries in the history of medicine. They have revolutionised the treatment of infections, significantly reducing mortality rates. One might be tempted to believe that the use of antibiotics has only benefits. Unfortunately, their overuse leads to serious complications – primarily the development of antibiotic resistance, the occurrence of adverse effects and unnecessary costs [1].

Antibiotics are the only group of medicines whose effectiveness diminishes over time – due to the growing resistance of bacteria. In principle, this is inevitable, but inappropriate use accelerates this process. If we consider the following facts:

• Around half of all prescriptions for antibiotics may be inappropriate; as many as 10 per cent of hospitalised patients who are prescribed antibiotics have no medical indication for their use
• In general hospitals, infections are found in 15–20 per cent of patients (10–15 per cent of community-acquired infections, 5 per cent of hospital-acquired infections)
• Expenditure on antibiotics often accounts for 20–30 per cent of the total cost of all medicines used in a hospital
• It seems obvious that measures need to be taken to reduce the inappropriate use of antibiotics [2].

However, in order to draw conclusions, take action and (most importantly) achieve results, it is first necessary to obtain data on the use of antibiotics at a given facility. The aim of monitoring the use of anti-infective medicines is to develop or review internal guidelines on the use of antibiotics, to provide training for hospital staff, or to evaluate the outcomes of the measures implemented. Various analyses are proposed:

1. Prospective monitoring

• routine – a practice in wards with very high antibiotic consumption, or specific practices relevant to the rational use of antibiotics, e.g.:
  • monitoring decisions regarding changes to or continuation of treatment on the third day of treatment (particularly in the ICU),
  • a review of all relevant antibiotics 2–3 times a week (so-called ‘antibiotic rounds’) – particularly in the ICU,
  • changing an ineffective course of antibiotic treatment, or adding another antibiotic
  • the use of certain antibiotics (particularly those in the ‘reserved’ category as defined by the WHO),
  • selected groups of infections (e.g. MSSA bacteraemia) or patients (e.g. those with renal failure);

• spot checks – one-day audits of all courses of antibiotic therapy (usually in a single ward) to assess:
  • the appropriateness / determination of indications for the use of antibiotics,
  • compliance of treatment with current standards,
  • accurate laboratory diagnosis,
  • adjustments to treatment following receipt of the results of microbiological tests,
  • the correct dosage.

The aim of this is:

• the identification of antibiotic misuse,
• an assessment of the duration of treatment (particularly in the context of peri-operative prophylaxis),
• monitoring the use of sequential therapy,
• an assessment of antibiotic combinations.

1. Retrospective monitoring

• qualitative – selected groups of infections (e.g. MSSA bacteraemia) or patients (e.g. those with renal failure), or an analysis of the time from diagnosis of the infection to the administration of antibiotics, and others;
• Quantitative – analysis of antibiotic usage. This type of report is the simplest to prepare, but it does not allow for an assessment of the quality of infection treatment. However, such a report does allow trends within the hospital to be observed and is valuable for drawing overall conclusions when combined with other types of monitoring (e.g. alongside microbiological mapping) [3].

Retrospective quantitative monitoring is most commonly carried out using defined daily doses (DDD). The DDD is a conventional value which does not necessarily correspond to the amount of antibiotic a patient receives per day, nor should it be regarded as a recommendation regarding the drug’s dosage. The sole purpose of the DDD is to enable the comparison of the consumption of one drug with that of another. The DDD value for a given medicine is determined by the Norwegian Institute of Public Health as part of the WHO Collaborating Centre for Drug Statistics Methodology. A medicine search tool (including combinations in compound medicines) is available at https://atcddd.fhi.no/atc_ddd_index/, along with a description of the ATC and DDD methodologies for which the Institute is responsible. In many cases, this value differs from the actual daily dose used at a given facility. This may occur in paediatric hospitals (where the DDD is higher than the doses actually administered), or in the case of many medicines, e.g.: where ampicillin with sulbactam (DDD = 6 when converted to ampicillin, i.e. 6 g of ampicillin per day) is used to treat infections caused by Acinetobacter baumannii (in which case the daily dose is often 12 or 18 g of ampicillin per day). In such cases, the PDD (prescribed daily dose) may be used. However, the use of the DDD allows for a comparison of antibiotic consumption across different centres with a similar patient profile. It is important to note that the same drug has different DDDs depending on the route of administration.

AVAILABILITY	DESCRIPTION	EXAMPLE
Access (share)	First-line treatment for most mild to moderate infections. Relatively low risk of misuse. Unless tests indicate incorrect use, they do not require special monitoring	Amoxicillin with clavulanate, metronidazole, gentamicin
Watch (follow)	Used in severe infections, or when medicines in the Access group are unsuitable. It is recommended that medicines in this group be dispensed on the basis of personalised prescriptions. Periodic retrospective analyses of the use of this group of medicines are recommended.	Ceftriaxone, meropenem, clarithromycin
Reserve (book)	Used when other medicines are unsuitable, or in the most severe forms of infection. Prospective monitoring is recommended, with an assessment of the appropriateness of treatment before each dispensing of the medicine	Intravenous fosfomycin, cefiderocol, linezolid

Example 1.

Over the course of a year, the hospital used 900 vials of ampicillin with sulbactam, each containing 2+1g. As mentioned, the DDD for this medicine is 6, calculated on the basis of ampicillin. Therefore, 1 vial (2 g of ampicillin) is 1/3 of a DDD, and 900 vials is 300 DDDs.

It is important to note that most pharmacy software programmes (e.g. AMMS) allow you to generate a report on medicine usage in DDDs, so there is no need to do this manually. However, the ‘DDD’ and ‘number of DDDs per pack’ fields must be completed in the medicine record. DDD values may change, so you should check the institute’s website for any information regarding changes to DDDs – fortunately, a list of updates is available at: https://atcddd.fhi.no/atc_ddd_alterations__cu/.

It is worth ensuring that all equivalents have the same international name so that the programme can link them correctly (if one has a Latin name, another an English name and a third is in capital letters, they will all appear separately on the report). If you are generating separate reports for oral and parenteral formulations, it is also worth standardising the ‘formulation’ in the medicines record. You can also opt for a report organised by ATC groups (Anatomical Therapeutic Chemical classification); in this case, the report will not list specific active substances, but rather the consumption of individual subgroups of antibiotics (e.g. tetracyclines). It is also worth preparing a separate report based on availability groups (according to the WHO AWaRe system – Access, Watch, Reserve) [4].

In this way, it is possible to obtain a report on consumption in DDDs per unit of time (usually one calendar year). However, it is difficult to use such a report for any practical purpose. Higher medicine consumption may be due to a larger number of patients. To address this, data should be presented in DDDs per 100 or 1,000 patient-days. However, such a report can only be used for comparison with other hospitals with a similar patient profile. To be able to observe trends, it is necessary to compare the data with previous years.

Example 2.

The use of intravenous cloxacillin over the years, as evidence of improvements in the treatment of MSSA bacteraemia and the narrowing of the spectrum of antibiotic therapy.

YEAR	DDD/1,000 PERSON-DAYS
2018	7,212
2019	10,173
2020	12,421
2021	15,775
2022	24,855
2023	39,476

If the objective was to reduce the consumption of a particular group of medicines, such a report makes it possible to show how well that objective is being met.

Example 3.

The consumption of intravenous ciprofloxacin over the years, as evidence that the target – to reduce consumption of the drug – has been achieved. 

YEAR	DDD/1,000 PERSON-DAYS
2018	34,570
2019	35,933
2020*	38,420
2021	26,195
2022	24,338
2023	19,668

As can be seen from the attached examples, retrospective quantitative monitoring is not an end in itself, but should be carried out to verify objectives, observe trends or draw general conclusions. Example:

• Microbiological monitoring shows an increase in the number of bacterial isolates with acquired carbapenem resistance in the ICU. The year-on-year DDD/1000 report shows an increase in carbapenem consumption in this ward. Specific measures should therefore be taken; examples include:
• to run a training session in the department to demonstrate the specific impact of the increase in carbapenem use
• expand the analysis to identify the causes of the situation in question; for example: has the ward drawn up a proposal for empirical treatment of the most common infections?
• Does the department have a procedure in place for switching to targeted therapy once the results of microbiological tests are available?
• Are the current guidelines or procedures being followed?
• Should further action be taken to remedy the situation?

Quantitative retrospective monitoring is therefore often the starting point for further analysis and action, as well as a relatively straightforward and objective method of verifying whether agreed targets have been met.

However, such a report does have certain ‘pitfalls’:

• If a given centre has been using subtherapeutic doses of antibiotics, such as aminoglycosides, then as a result of the measures introduced, the DDD per 1,000 patient-days for this group of medicines will increase, and this should not be viewed negatively;
• The use of higher doses and sequential therapy (switching to targeted oral antibiotic treatment) may improve treatment efficacy and shorten the length of hospital stay. In that case, the same number of patients will ‘use up’ more DDDs per 1,000 patient-days and free up hospital beds for subsequent patients, who will also be taking antibiotics. Consequently, the overall DDD per 1,000 patient-days may increase, despite an overall improvement in the quality of treatment;
• If, as part of the use of rational antibiotic therapy, the duration of treatment is extended when required, despite an improvement in the patient’s condition (minimum duration of antibiotic therapy for Staphylococcus aureus bacteraemia or Clostridioides difficile infection). The consumption of antibiotics (DDD/1,000 person-days) used to treat these infections will increase;
• It is worth analysing the total consumption of antibiotics in a given unit in the context of:
• the microbiological situation (if there is an increase in the number of infections caused by Enterobacterales with acquired ESBL-type resistance, the use of carbapenems will rise)
• other quality indicators, such as: mortality rate, average length of stay, etc.
• The objectives set;
• The report must contain clearly stated conclusions and comments.

It is worth noting that retrospective quantitative monitoring need not be limited to a particular hospital but may also cover the whole country. In accordance with the European Council’s Recommendations on stepping up action to combat antimicrobial resistance within the framework of the ‘One Health’ approach (2023/C220/01), the Council of the EU recommends the introduction of appropriate measures aimed at ensuring that, by 2030, total human consumption of antibiotics (defined as defined daily doses (DDD) per 1,000 inhabitants per day), in both outpatient and inpatient care combined, is reduced by 20 per cent in the European Union compared with 2019. A further target is to achieve a share of at least 65 per cent of antibiotics belonging to the ‘Access’ group according to the WHO classification [5]. Based on the report of the National Health Programme – neither of these targets has been met at present – consumption remains at the 2019 level (23.6 DDD per 1,000 inhabitants per day), whilst the share of ‘Access’ group consumption stands at 60.4 per cent [6].

1. https://www.ecdc.europa.eu/en/antimicrobial-resistance/facts (as at 27 August 2024)
2. Hryniewicz, W., Ozorowski, T. Hospital antibiotic policy. A proposal for hospitals. National Medicines Institute, Warsaw 2011
3. Ozorowski T., Woroń J., Misiewska-Kaczur A. Antibiotic policy. Programme for hospital antibiotic policy. Centre for Quality Monitoring in Healthcare. Kraków 2015
4. Web Annex. Infographics. In: The WHO AWaRe (Access, Watch, Reserve) antibiotic book. Geneva: World Health Organisation; 2022 (WHO/MHP/HPS/EML/2022.02) as at (27 August 2024)
5. Council Recommendations on stepping up action to combat antimicrobial resistance within the framework of the ‘One Health’ approach (2023/C 220/01) https://eurlex.europa.eu/legal-content/PL/ TXT/PDF/?uri=CELEX:32023H0622(01) (as at 27 August 2024)
6. Pawlik K., Bysiek J., Hryniewicz W., Skoczyńska A. Report: Monitoring of Antibiotic Use in Inpatient Care for 2022. Warsaw, 28 December 2023