Changes in the Dispensing of Antibiotics to Australian Children Between 2013 and 2023: Are We Heading in the Right Direction?

ABSTRACT

Background

Antimicrobial resistance (AMR) poses a critical public health issue, exacerbated by the overuse and misuse of antibiotics. Children are particularly susceptible to bacterial infections and are frequently prescribed antibiotics.

Objective

This study examined trends in antibiotic dispensing to children aged under 13 years in Australia between 2013 and 2023.

Methods

This retrospective observational study used a 10% random sample of dispensing records for nationally subsidised prescription antibiotics. The number of children dispensed an antibiotic was calculated for each year and expressed per 100 children. Trends were analysed using joinpoint regression overall and by age group, sex, the World Health Organisation's Access, Watch, Reserve (AWaRe) system of antibiotic classification and antibiotic subtype.

Results

Between 2013 and 2023, 3,406,208 antibiotic prescriptions were dispensed to 554,837 children. There was a decrease in the total number of antibiotic prescriptions dispensed, falling from 103 prescriptions dispensed for every 100 children in 2013 to 63 prescriptions in 2023 (annual percent change [APC]: −6.9, 95% CI: −9.8, −4.4). While decreases were observed for medications classified as ‘Access’ (APC: −5.8, 95% CI: −8.7, −3.1), the largest decrease was observed in ‘Watch’ medications (APC: −15.0, 95% CI: −19.4, −11.7). Decreases were observed in the proportion of children dispensed an antibiotic, declining from 45.7% in 2013 to 33.6% in 2023 (APC: −4.7%, 95% CI: −7.1%, −2.5%). Reductions in dispensing were observed overall and by sex, age groups and most antibiotic types.

Conclusions

Antibiotic dispensing in Australian children has decreased over the past decade, for all ages, sexes and antibiotic sub‐classes, likely reflecting implemented policies and efforts to curb overuse of antibiotic medicines and AMR during this period.

1. Background

Antimicrobial resistance (AMR) has been recognised as one of the most pressing global public health concerns, responsible for an estimated 1.14 million global deaths in 2021 [1]. Defined as the ability of bacteria to survive and proliferate despite the administration of antibiotics, AMR renders previously effective drugs ineffective, limiting treatment options for bacterial infections [2]. Consequently, infections that have historically been treatable can become life‐threatening, leading to longer hospital stays, increased medical costs, and higher mortality rates [3]. Increases in AMR have been largely driven by the misuse or overuse of antibiotics including the unnecessary use of antibiotics for infections and deviations from recommended dosing guidelines [4].

Children represent an important population in the context of AMR, due to high rates of antibiotic prescribing [5]. For example, in countries including South Korea, Italy, Spain, the US and Germany, children under 2 years old receive antibiotic courses at rates surpassing the equivalent of one per child annually [6]. High rates of antibiotic prescribing reflect high rates of infection in children, associated with their immature immune system, increased exposure to pathogens through exploratory oral behaviours, developing hygiene practices, and socialisation with other children (e.g., childcare and school). However, overuse and inappropriate prescribing of antibiotics is also common [7, 8]. In the US, 18.3% of children (aged 0–17 years) with a physician office visit were inappropriately prescribed an antibiotic for an infection‐related diagnosis that did not warrant antibiotics [9]. Liberal prescribing of antibiotics in children is in part driven by the high risks associated with not treating infections and difficulties in the timely diagnosis of bacterial infections [10, 11]. Overprescribing can be further compounded by parental pressures and expectations, driven by concern for the health of their child [10, 11, 12]. The overuse and inappropriate prescribing of antibiotics in children is also of importance, as children commonly experience negative side effects associated with the use of antibiotics [13], and there is growing evidence that early exposure to antibiotics may be associated with long‐term health risks such as increased susceptibility to infections, asthma, allergic diseases and obesity [14, 15]. Although it is not possible to determine clear causation with the current literature [15].

Globally, antibiotic use and AMR prevalence vary substantially between countries [16]. While Australia reports one of the lowest AMR‐related mortality rates [17], its use of antibiotics exceeds that of many high‐income countries, including the UK, Germany, Switzerland and Norway [18]. To mitigate this, Australia has introduced various national policies, guidelines and programs aimed at reducing antibiotic use. Nationally, Australia's first National Antimicrobial Resistance (AMR) Strategy 2015–2019 was released in 2015 with seven key objectives for reducing AMR [19]; this was followed up by the National AMR Strategy—2020 and Beyond [20]. Prescribing guidelines from the Therapeutic Guidelines and the Australian Commission on Safety and Quality in Health Care (ACSQHC) [21], as well as surveillance systems, such as the National Antimicrobial Prescribing Survey (NAPS) and the Antimicrobial Use and Resistance in Australia (AURA) surveillance system, have also been developed to help guide and monitor antibiotic prescribing practices and resistance patterns at national and local levels [22, 23]. Additionally, there have been public awareness campaigns to raise awareness of AMR and the safe use of antibiotics.

This study aimed to examine temporal trends in antibiotic dispensing to paediatric patients (0–12 years) in Australia between 2013 and 2023, overall and by sex, age group, antibiotic subtype, and the World Health Organisation's Access, Watch, Reserve (AWaRe) system of antibiotic classification.

2. Methods

2.1. Study Design and Population

The study utilised a retrospective observational design with national medicine dispensing data from the Pharmaceutical Benefits Scheme (PBS). The PBS provides subsidised medications to all citizens and permanent residents as part of Australia's universal healthcare system. A sample of individual level dispensing records for 10% of the eligible PBS population (known as the PBS 10% sample) was provided by Services Australia. From this data, we identified all dispensing of antibiotics to children aged 0–12 years from 2013 to 2023. PBS data and the PBS 10% sample have widely been used to research medicine use in Australia [24]. Within the PBS data, only year of birth was available, thus date of birth was estimated as the 1st of July of the birth year of each participant, which was then used to exclude individuals aged 13 years and over at the time of dispensing from the analysis. PBS data from 2013 to 2023 was used, as prior to 2013 data on medication dispensing that fell below the co‐payment threshold was not captured, which would have resulted in an underestimation of antibiotic dispensing. Data from 2023 was the latest complete year available at the time of analysis.

2.2. Outcomes

The primary outcome was the number of antibiotic prescriptions dispensed per year per 100 children, and the proportion of children aged 0–12 years who were dispensed an antibiotic per year. Data on population size, overall and by age and sex, were obtained from the Australia Bureau of Statistics [25]. The total cost and cost per 100 children were also calculated in Australian dollars (2024). Medication costs were taken from the PBS.

The outcome was examined by age, sex, AWaRe classification, antibiotic subtype, and prescriber specialty. Age was categorised by infants (< 1 year), toddlers (1–2 years), early childhood (3–5 years), middle childhood (6–9 years) and pre‐adolescents (10–12 years). The antibiotic sub‐type used was based on the ATC groupings within J01 (Antibacterials for systemic use). The AWaRe classification of antibiotics divides antibiotics into three categories: Access, Watch and Reserve [26]. Antibiotics in the ‘Access’ category tend to have a narrow spectrum of activity, lower cost, a good safety profile and generally low microbial resistance potential. As such, they are typically recommended as first‐ or second‐line treatment options for common infections [26]. ‘Watch’ antibiotics are broad‐spectrum antibiotics that are typically more costly and are recommended for patients with more severe clinical conditions or for infections with pathogens that are likely resistant to Access antibiotics (such as in upper urinary tract infections) [26]. Reserve antibiotics are last‐line treatments used for treating multidrug‐resistant infections [26]; however, none of these medications are currently listed on the PBS and therefore do not appear in the dataset used for this study. Antibiotic medication subgroups were categorised using their Anatomical Therapeutic Chemical (ATC) classification (3rd level).

2.3. Statistical Analysis

The total number of antibiotics dispensed per year and the number child of children per year dispensed an antibiotic was estimated and expressed per 100 children using population data from the Australian Bureau of Statistics (adjusted to 10% of the population). Joinpoint regression was used to examine trends over time (2013–2023) in the dispensing of antibiotics. This form of regression analysis was selected as within the study period, there was not a single pre‐defined point of interest (e.g., as would be required for an interrupted time series analysis). Joinpoint regression uses the observed data to fit a series of linear segments, identifying joins in the data where the trend changes [27]. The annual percentage change (APC) is calculated for each segment. Joinpoint regression analysis was carried out using the Joinpoint Regression Program 4.8.0.1.

Descriptive statistics, including measures of central tendency and dispersion, were calculated to summarise demographic characteristics of the study population, and the distribution of antibiotic dispensing practices over time. Data were managed and analysed using statistical software Stata 18 [28].

2.4. Missing Data

Individuals included in the study required non‐missing year of birth and dispensing item code. Sex was missing for a small number of individuals who were included (811 prescriptions, < 0.01%). Additionally, no AWaRe category was available for a small number of medications (3 medications, 189 dispensing).

2.5. Ethical Approval

Ethics approval was obtained for a wider project on the PBS data through the University of Western Australia Human Research Ethics Committee (2022/ET000372), and a project‐specific amendment was submitted for the current project. A waiver of consent was granted for the project as it met the criteria set out in the National Statement on Ethical Conduct in Human Research [29].

3. Results

3.1. Dispensing Characteristics

Between 2013 and 2023, 3,406,208 antibiotic prescriptions were dispensed to 554,837 children aged 0–12 years. The use of antibiotics was highest in children aged 3–5 years (Table 1) and male children were dispensed more antibiotics than female children. Antibiotics were commonly dispensed by general practitioners (GP)/family doctors. The most common subclasses of antibiotics prescribed were penicillins, with most included in the Access AWaRe classification (Table 1).

TABLE 1.

The characteristics of publicly subsidised PBS antibiotic dispensed to children aged 0–12 years in Australia between 2013 and 2023.

Characteristics	Number (%)
Total number of dispensings	3,406,208 (100.0)
Sex a
Female	1,655,150 (48.6)
Male	1,750,247 (51.4)
Age group
< 1 years	231,340 (6.8)
1–2 years	835,507 (24.5)
3–5 years	960,412 (28.2)
6–9 years	881,890 (25.9)
10–12 years	497,059 (15.6)
Antibiotic type
Tetracyclines	15,624 (0.5)
Penicillins	2,161,907 (63.5)
Other beta‐lactam antibiotics (excluding penicillins)	841,795 (24.7)
Sulfonamides & trimethoprim	77,403 (2.3)
Macrolides, lincosamides & streptogramins	277,639 (8.2)
Aminoglycoside antibiotics	812 (< 0.1)
Quinolone antibiotics	1321 (< 0.1)
Other antibiotics	29,707 (0.9)
AWaRe classification b
Access	2,980,716 (87.5)
Watch	425,304 (12.5)
Prescriber speciality
Medical specialties	424,218 (12.5)
Surgical specialties	11,490 (0.3)
Diagnostic & imaging	1007 (< 0.1)
Maternal & neonatal	1235 (< 0.1)
GP/family doctor	1,858,391 (54.6)
Emergency & intensive care	7754 (0.2)
Dentistry & oral	19,940 (0.6)
Other providers	4579 (0.1)
Miscellaneous/unknown	1,077,594 (31.6)

^a In 811 dispensing records sex was missing from the data.

^b AWaRe classification was not available for two preparations/188 prescriptions.

3.2. Dispensing Trends

Between 2013 and 2023, there was a decrease in the total number of antibiotics dispensed per year, from 103 antibiotics dispensed for every 100 children in 2013 to 63 in 2023 (APC: −6.9, 95% CI: −9.8, −4.4) (Table 2). Decreases were observed in the dispensing of antibiotics classified as Access or Watch. There was a 5.8% annual percentage decrease for medications classified as Access (APC: −5.8, 95% CI: −8.7, −3.1), and a 15.0% decrease for medications classified as Watch (APC: −15.0, 95% CI: −19.4, −11.7). Similar decreases in costs were observed with a reduction in average annual cost per 100 children from $471 to $274 per year (APC: −7.3, 95% CI: −10.1, −4.8).

TABLE 2.

Dispensing of publicly subsidised antibiotics to a 10% sample of children aged 0–12 years in Australia between 2013 and 2023.

Year	Total no. of antibiotics dispensed	Antibiotics per 100 children	Total no. of access antibiotic dispensed	Access antibiotics per 100 children	Total no. of watch antibiotics dispensed	Watch antibiotics per 100 children	Total cost	Annual cost per 100 children
2013	394,143	103.3	327,271	85.8	66,841	17.5	$1,798,273	$471
2014	400,726	103.4	337,471	87.1	63,235	16.3	$1,826,134	$471
2015	401,717	102.0	338,301	85.9	63,388	16.1	$1,873,699	$476
2016	362,818	90.4	311,948	77.8	50,856	12.7	$1,684,004	$420
2017	337,045	82.9	295,856	72.7	41,170	10.1	$1,561,367	$384
2018	301,858	73.5	267,592	65.2	34,248	8.3	$1,416,619	$345
2019	302,269	73.2	268,344	65.0	33,906	8.2	$1,374,306	$333
2020	178,310	43.2	162,299	39.3	16,001	3.9	$824,686	$200
2021	212,464	51.8	195,443	47.6	17,012	4.1	$967,512	$236
2022	257,118	62.6	235,017	57.2	22,093	5.4	$1,149,139	$280
2023	257,740	62.6	241,174	58.6	16,554	4.0	$1,128,764	$274

3.3. Trends in Proportion of Children Dispensed an Antibiotic

The proportion of children dispensed an antibiotic each year also decreased, with 45.7% of children dispensed an antibiotic in 2013 compared with 33.6% in 2023 (APC: −4.7%, 95% CI: −7.1%, −2.5%) (Figure 1).

FIGURE 1.

The number of children (0–12 year) per 100 dispensed an antibiotic between 2013 and 2023.

Similar trends were observed for male and female children, with APC of −4.8 and −4.7 from 2013 to 2023, respectively (Figure S1). The proportion of paediatric patients with at least one antibiotic dispensing per year decreased consistently from 2013 to 2023 across all age groups (Figure S2). The APC over this observation period ranged from −3.8% in children aged 1–2 years to −5.5% in neonates and infants (Table 3). Reductions in dispensing were also observed across all antibiotic subtypes (Figure S3). For most antibiotic subtypes, the reduction was linear with no joins, with the exception of other beta‐lactam antibiotics (excluding penicillins) and other antibiotics, both with a single join (Table 3).

TABLE 3.

Change in the proportion of children aged 0–12 years dispensed an antibiotic per year between 2013 and 2023.

	2013	2023	APC (95% CI)
Sex
Female	45.8%	33.7%	−4.7 (−7.0, −2.4)
Male	45.7%	33.4%	−4.8 (−7.2, −2.5)
Age
< 1 year	50.5%	34.0%	−5.5 (−8.0, −3.3)
1–2 years	60.7%	45.5%	−3.8 (−5.9, −1.8)
3–5 years	52.0%	40.1%	−4.0 (−6.4, −1.7)
6–9 years	40.3%	30.4%	−4.9 (−7.4, −2.5)
10–12 years	33.8%	24.2%	−5.4 (−7.8, −3.1)
Antibiotic types
Tetracyclines	0.20	0.18	−1.4 (−2.8, −0.1)
Penicillins	37.7	26.2	−5.7 (−8.6, −2.9)
Other beta‐lactam antibiotics (excluding penicillins)	17.1	12.8	2013–2021: −5.5 (−14.0, 2.7)
			2021–2023: 8.7 (−5.3, 19.5)
Sulfonamides & trimethoprim	2.0	0.8	−12.0 (−18.0, −7.6)
Macrolides, lincosamides & streptogramins	8.4	2.4	−14.3 (−19.1, −10.8)
Aminoglycoside antibiotics	0.02	0.01	−7.6 (−11.5, −4.2)
Quinolone antibiotics	0.03	0.02	−5.4 (−6.9, −4.0)
Other antibiotics	0.7	0.4	2013–2015: 7.7 (−6.6, 23.8)
			2015–2023: −12.5 (−18.2, −10.8)

4. Comment

4.1. Principal Findings

A reduction in the dispensing of PBS‐subsidised antibiotics in children (0–12 years) was observed in Australia, decreasing from 103 antibiotics dispensed for every 100 children in 2013 to 63 antibiotics in 2023. Reductions in dispensing were observed across sex, age, and most antibiotic types. This reflects clear progress in reducing paediatric antibiotic use over the past decade, aligning with national public health goals to combat antimicrobial resistance.

4.2. Strength of the Study

The use of the PBS dataset provides a comprehensive view of antibiotic dispensing trends across a large, representative sample of the Australian paediatric population. Its national scope allows for longitudinal analysis of dispensing trends, providing valuable insights into both long‐term trends and short‐term fluctuations.

4.3. Limitations of the Data

One of the limitations of this study is its reliance on only medication dispensing data, which may not directly reflect antibiotic prescription, consumption, or quality of use. This is particularly important in the study's context, where evidence shows correct prescribing to play a critical role in preventing AMR [30]. The data used in the study was a 10% sample of the whole PBS dataset. While we assume that this provides a good representation of the whole sample, there may be slight differences that may lead to slightly different results.

Additionally, the study's reliance on the PBS dataset posed some challenges. Since the dataset only includes the year of birth, the calculation of patient age relied upon all patients being assigned a birth date of July 1st of the year of birth, which introduced a margin of error in age‐specific analyses and reduced the overall precision of our findings. Furthermore, the PBS dataset does not contain clinical indications for antibiotic use, preventing assessment of whether antibiotics were prescribed based on clinical necessity or guidelines, a critical factor in antibiotic stewardship efforts [31]. This has several important implications in the interpretation of the results, as without knowing the reasons why antibiotics were prescribed, it is difficult to determine if reductions reflect improved stewardship/judicious antibiotic use or other external factors leading to fewer bacterial infections.

4.4. Interpretation

The overall decline in antibiotic dispensing is likely to be a positive indicator; however, continuous monitoring of broad‐spectrum antibiotics and dispensing trends is crucial to sustain these efforts. The findings are consistent with the general literature for high‐income countries, particularly those with antimicrobial stewardship programs [32, 33, 34, 35]. A systematic review and meta‐analysis examining trends in antibiotic consumption and their association with antimicrobial stewardship programs, found that programs were associated with a 21% reduction in antibiotic consumption in paediatric hospitals and a 28% reduction in WHO ‘Watch’ group antibiotics [32].

In line with the decrease in antibiotic dispensing, there were substantial reductions in both the total expenditure on antibiotics and average cost per child, with a total difference of $669,509 per annum between expenditure in 2013–2023. Given that our results are based on a 10% sample of PBS data, this reflects an estimated saving of approximately $6.7 million annually (equivalent to approximately $4.4 million USD) in dispensing of government‐subsidised medicines in Australia from antibiotics alone. As these medications are publicly subsidised, this reduction in cost would have been shared by parents and the government.

Decreases were observed in the dispensing of both Access and Watch antibiotics, with the greatest reduction seen in the use of Watch (broad‐spectrum) antibiotics (APC: 15.0%). This positively aligns with stewardship guidelines encouraging more targeted use of antibiotics as well as improved rapid diagnostics for accurate diagnosis of bacterial and viral infections [36]. With this substantial decrease, the dispensing of Watch antibiotics has decreased from 17.0% of dispensed antibiotics in 2013 to 6.4% in 2023. As such, in patients aged 0–12 years, Australia is far surpassing the country‐level target of at least 60% of total antibiotic consumption being Access group antibiotics set out by the WHO 13th General Programme of Work 2019–2023 [37]. However, as mentioned above, none of the medications classified as ‘Reserved’ are currently PBS listed, thus we cannot estimate the dispensing of these last‐line antibiotics used for treating multidrug‐resistant infections, although studies suggest that the use of ‘Reserve’ antibiotics in Australia is extremely low [38, 39].

Importantly, reductions in antibiotic dispensing were observed across all antibiotic types, except other beta‐lactam antibiotics, regardless of overall incidence, although the degree of reduction varied across subtypes. This decline in is likely due to several key factors such as the implementation of antibiotic stewardship programmes, increased awareness of AMR and compliance with clinical/paediatric guidelines, all of which reduce unnecessary antibiotic use [22, 23]. The use of common broad‐spectrum antibiotics such as penicillins decreased at similar levels (APC −5.7%) to more sparsely used antibiotics such as aminoglycoside (APC −7.6%) and quinolone antibacterials (APC −5.4%). Other beta‐lactam antibiotics followed a similar trend to the other antibiotic types; however, a substantial drop during the COVID‐19 pandemic results in a single join at this time. It is likely that with subsequent years of data, clear declines in the use of these antibiotics will be observable.

The low incidence of antibiotic dispensing observed in 2020 and 2021 is likely reflective of the COVID‐19 pandemic [40]. This is consistent with the Fourth Australian Report on Antimicrobial Use and Resistance in Human Health, which reported reductions in the dispensing of seven of the 10 most dispensed antibiotics during the pandemic period [41]. This, alongside additional public health measures of physical distancing, isolation when unwell, and vigilance on hand hygiene, reduced transmission of bacterial infection in the community during that time.

5. Conclusions

This study demonstrates a substantial decline in antibiotic dispensing among paediatric patients (0–12 years) in Australia between 2013 and 2023. Results were consistent across age, sex, antibiotic type and AWaEe category. While these trends are promising, continued efforts are needed to ensure that antibiotic dispensing remains appropriate and aligned with national and international stewardship goals. These findings recognise the likely effectiveness of robust national stewardship programs and support the ongoing need to promote the responsible use of antibiotics to combat the growing threat of antimicrobial resistance.

Author Contributions

Erin Kelty: conceptualisation (equal), data curation, formal analysis (equal), supervision (equal), writing – review and editing. Ella Tairy: formal analysis (equal), writing – original draft preparation. Scott Sims: conceptualisation (equal), writing – review and editing. Carol Orr: conceptualisation (equal), writing – review and editing. Amy Page: project administration, writing – review and editing. David B. Preen: resources, writing – review and editing. Frank M. Sanfilippo: resources, writing – review and editing. Christopher Etherton‐Beer: resources, writing – review and editing. Ebony Quintrell: supervision (equal), writing – review and editing.

Conflicts of Interest

Authors E.K., E.T., S.S., C.O., A.P., D.B.P., F.M.S. and E.Q. have no conflicts to declare. C.E.B. is a member of the Drug Utilisation Sub‐Committee of the Pharmaceutical Benefits Advisory Committee. The contents of this report are the responsibility of the authors alone.

Supporting information

Figure S1: The number of children (0–12 years) per 100, dispensed an antibiotic between 2013 and 2023, stratified by sex.

Figure S2: The number of children (0–12 years) per 100 dispensed an antibiotic between 2013 and 2023, stratified by age.

Figure S3: The number of children (0–12 years) per 100 dispensed an antibiotic between 2013 and 2023, stratified by antibiotic type.

Kelty E., Tairy E., Sims S., et al., “Changes in the Dispensing of Antibiotics to Australian Children Between 2013 and 2023: Are We Heading in the Right Direction?,” Paediatric and Perinatal Epidemiology 40, no. 3 (2026): 342–348, 10.1111/ppe.70068.

Data Availability Statement

The data that support the findings of this study are available from Services Australia. Restrictions apply to the availability of these data, which were used under license for this study. Data are available from the author(s) with the permission of Services Australia.