View Article PDF

Antimicrobial resistance (AMR) is a growing issue that threatens human health globally. To help combat this, both the New Zealand1 and Global2 AMR Action Plans include effective antimicrobial stewardship (AMS) as a core control measure. AMS involves coordinated strategies to optimise antimicrobial use for the prevention and treatment of infections, while minimising possible adverse sequelae of antimicrobial use such as Clostridioides difficile infections, adverse reactions and AMR. The total quantity of inpatient antibacterial use is similar in five of New Zealand’s larger district health boards (DHBs) (700–800 defined daily doses per 1,000 inpatient bed days), and lower than in Australian and English hospitals (mean of ~900 and ~1,300 defined daily doses per 1,000 inpatient bed days, respectively) but higher than in countries like Switzerland and Sweden (~500 and ~300 defined daily doses per 1,000 inpatient bed days, respectively).3 Collectively, these data offer some reassurance regarding the total volume of hospital antimicrobial use in New Zealand DHBs. However, AMS programmes need to assess the quality of antimicrobial prescribing as well as the quantity of antimicrobials used.

Point prevalence surveys can provide a practical and repeatable ‘snapshot’ of antimicrobial prescribing patterns to inform AMS programmes, and monitor their impact.4,5 In this work, we present findings from the first point prevalence surveys of antimicrobial prescribing in adult inpatients at Canterbury DHB (CDHB) hospitals. Our aims were to investigate the:

1. prevalence and nature of antimicrobial use,

2. compliance with local antimicrobial prescribing guidelines,

3. appropriateness of antimicrobial prescribing based on defined criteria, and

4. compliance with funding criteria, per Pharmaceutical Management Agency (PHARMAC) Section H (Hospital Medicines List) prescribing restrictions.

Methods

Setting

CDHB provides publicly funded healthcare for a population of ~570,000 people across 14 certified facilities (~1,500 beds).6,7 We sought to understand antimicrobial prescribing in adult inpatients at the three largest CDHB facilities, excluding Hillmorton Hospital, which exclusively cares for people with psychiatric conditions. Together these three facilities comprise ~1,100 beds and ~75% of all CDHB beds:

Christchurch Hospital campus (~830 beds) includes the geographically co-located Christchurch Women’s Hospital and provides a full range of emergency, acute, elective and outpatient services including bone marrow transplant and intensive care units,

Burwood Hospital (~230 beds) provides rehabilitation, elective orthopaedic surgery, spinal injury, geriatric and psychogeriatric services,

Ashburton Hospital (~50 beds) offers secondary-level acute medical and surgical services and is situated in Ashburton ~90 km south of Christchurch.

The remaining CDHB facilities were excluded from this as they provide psychiatric care (eg, Hillmorton Hospital) or are small rural hospitals focused mainly on rehabilitative, geriatric and/or maternity care (eg, Oxford Hospital).

Audit development

We modelled our overall audit process on the Australian National Antimicrobial Prescribing Survey (‘NAPS’; www.naps.org.au), using their recommended multidisciplinary and hospital-wide approach.8 Data collection forms were similar to those used in the NAPS and were completed for each patient on antimicrobial therapy. These forms included general demographic information, details of the antimicrobial regimen(s) and specifics regarding antimicrobial allergies, surgical procedures and microbiology results where relevant. The start date, route of administration, dose, frequency, review/stop date (where documented) and indication (documented by prescriber or if this was absent, presumed indication if auditors were able to determine this from the clinical record) were recorded for each antimicrobial prescribed. Key points of difference between the CDHB and NAPS processes were that we assessed antimicrobial prescribing against CDHB’s guidelines (published in The Pink Book9 and in HealthPathways10) rather than the Australian guidelines (Therapeutic Guidelines: Antibiotic),11 and evaluated compliance with PHARMAC Section H, which provides the criteria for funded medicine use in DHB hospitals.12 Prior to the audit days, we undertook sequential pilot studies in general surgical and medical wards to ensure the data collection form was fit for purpose and to estimate resourcing for the audit days.

Audit process

The audits were conducted on three days over a one-year period: 30 November 2017 (Christchurch and Ashburton Hospitals), 13 June 2018 (Christchurch Women’s Hospital) and 22 November 2018 (Burwood Hospital). On the first audit day, 11 multidisciplinary teams (around one team per 50 patients) consisted of an infectious diseases or clinical microbiology consultant or registrar (advanced trainee) paired with a pharmacist, clinical pharmacologist or registered nurse. On the subsequent audit days, multidisciplinary teams (physician or registrar plus pharmacist) included at least one member from the first audit. Prior to the audit days, teams were given verbal and written education on how to collect data and perform the assessments, to assist with standardisation.

On each audit day, a list of all adult inpatients at 08:00am was generated from the patient management system (PMS) and divided by clinical area among audit teams. Auditors manually reviewed the electronic and hardcopy prescription records for all these patients to identify those who were prescribed at least one antimicrobial agent (antibacterial, antiviral, antifungal or antiparasitic administered via any route) at 08:00am on the study day and/or administered a one-off (‘stat’) antimicrobial dose or surgical prophylaxis in the preceding 24 hours. One data collection form was completed for each patient who fulfilled these criteria. The auditing teams assessed compliance with CDHB antimicrobial prescribing guidelines and PHARMAC Section H restrictions. Guideline compliance was categorised as compliant, non-compliant, directed therapy (according to culture susceptibilities), recommended by Infectious Diseases/Clinical Microbiology, no guideline available, or not assessable (such as when the indication was not known). Further, prescribing was classified as ‘appropriate’ (optimal or adequate) or ‘inappropriate’ (suboptimal or inadequate) based on the NAPS criteria.8 These criteria included consideration of choice of antimicrobial agent, dose, route and duration of treatment, patient allergies and microbiology results. An infectious diseases physician was available to resolve any disputes within multidisciplinary teams. Finally, documentation of indication and review/stop date, and cessation of surgical antimicrobial prophylaxis within 24 hours post-surgery, were examined. These are considered key indicator assessments with target compliance >95%, recommended in Australia.8

Details from the data collection forms were entered into Microsoft Excel™ spreadsheets for analysis. Approximately 10% of entries were checked for accuracy by a second member of the research team as a quality assurance measure. Results are presented as median, interquartile range and range, or as number and percentage, as appropriate. Results are both presented separately and combined for the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital.

Support for the point prevalence audits was given by Te Komiti Whakarite at CDHB. The audit was regarded as out of scope by the Health and Disability Ethics Committee (Ministry of Health, Wellington, New Zealand).

Results

Number of antimicrobials prescribed

The number of inpatients at the three sites on the audit days together with the prevalence of antimicrobial use and patient demographics are shown in Table 1. The overall prevalence of antimicrobial prescribing was 42.4% (322/760 patients), with the Christchurch Hospital campus the highest of the three facilities. There was a total of 480 antimicrobial prescriptions for the 322 patients on antimicrobial therapy; most (79.8%) usage was at the Christchurch Hospital campus (383/480 prescriptions), which comprised 69.3% of all the inpatients (527/760 patients). The majority of patients were prescribed one [208/322 patients (64.6%)], two [79/322 patients (24.5%)] or three [29/322 patients (9.0%)] agents each, while a very small proportion [6/322 patients (1.9%)] had four to six different agents.

Table 1: Demographic details of the adult inpatients on antimicrobial therapy at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aSelf-reported. Patients may identify with more than one ethnicity. Number of ethnicities identified were: Christchurch Hospital campus = 284, Burwood Hospital = 72, Ashburton Hospital = 11 (total 367).

Antimicrobial indication

The indications for antimicrobial use at the three sites are shown in Table 2. Overall, most prescriptions [377/480 prescriptions (78.5%)] were for treatment of infection, with around one-fifth being for surgical or medical prophylaxis [93/480 prescriptions (19.4%]. Six indications accounted for half of all antimicrobial use [249/480 prescriptions (51.9%)]: surgical prophylaxis (13.8%), community-acquired pneumonia (11.7%), intra-abdominal infection (8.3%), urinary tract infection (7.5%), medical prophylaxis (5.6%) and sepsis (5.0%); most of these prescriptions were at the Christchurch Hospital campus [211/249 prescriptions (84.7%)].

Table 2: Indications for antimicrobial use at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined), N(%).

aListed in order from highest to lowest frequency, based on the Christchurch Hospital campus, where CDHB AMS resources are focused. Indications defined as per the Australian National Antimicrobial Prescribing Survey, with some indications combined as indicated by superscript letters.
bPeritonitis, appendicitis, cholangitis, cholecystitis, diverticulitis or intra-abdominal abscess.
cCystitis, pyelonephritis or catheter-associated urinary tract infection.
dCellulitis/erysipelas, abscess/boils/folliculitis.
eProsthetic or native joint.
fSurgical or non-surgical.
gNon-oral mucosa, eg, vaginal candidiasis.
hExacerbation or chronic.

Antimicrobial agents used and route of administration

The top 20 antimicrobial agents used at the three sites and the associated routes of administration are shown in Table 3. These 20 antimicrobial agents accounted for 85.0% (408/480 prescriptions) of all antimicrobial usage, and was similar at the three facilities. Six systemic agents—amoxicillin+clavulanic acid, cefazolin, metronidazole, cefuroxime, flucloxacillin and gentamicin—accounted for nearly half (47.5%) of all antimicrobial use (228/480 prescriptions).

Table 3: Top 20 antimicrobial agents and routes at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aListed in order from highest to lowest frequency, based on the Christchurch Hospital campus where CDHB AMS resources are focused.

Amoxicillin+clavulanic acid use exceeded that of each of the five next most commonly used agents by two- to three-fold. The IV route was used for half of all prescriptions [232/480 prescriptions (48.3%)], and was the predominant route at the Christchurch Hospital campus (54.6%).

Compliance with antimicrobial prescribing guidelines

The overall antimicrobial guidelines compliance at the three sites is shown in Table 4; Appendix Figures 1 and 2 further stratify these by indication (as listed in Table 2) and antimicrobial agent (as listed in Table 3), respectively.

Table 4: Antimicrobial guidelines compliance, appropriateness of prescribing, PHARMAC compliance and key indicators at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aShaded area indicates overall guidelines compliance assessment, which excludes therapy that is directed, recommended by Infectious Diseases/Clinical Microbiology, has no guideline available or is not assessable.
bShaded area indicates overall appropriateness assessment, which excludes non assessable prescriptions.
cPrescriptions may have more than one reason for being inappropriate: Christchurch Hospital campus = 121 reasons across 57 prescriptions, Burwood Hospital = 39 reasons across 14 prescriptions, Ashburton Hospital = 7 reasons across 5 prescriptions.

Of the 278 prescriptions able to be assessed for guidelines compliance, 73.7% (205/278) were compliant. Pelvic inflammatory disease, intra-abdominal infections, skin and soft tissue infections, medical prophylaxis, sepsis and urinary tract infections accounted for 20.6% (99/480) of all prescriptions and were all ≥80% compliant. Surgical prophylaxis and community-acquired pneumonia accounted for 25.4% (122/480) of all prescriptions and had lower guidelines compliance at 67.9% (38/56 assessable prescriptions) and 58.7% (27/46 assessable prescriptions), respectively. Prescribing for infective exacerbation of chronic obstructive pulmonary disease had the lowest guidelines compliance at 36.4% (4/11 assessable prescriptions) (Table 2; Appendix Figure 1). Compliance with the six most commonly used antimicrobial agents, which accounted for around half of all antimicrobial prescriptions, ranged from 51.0% to 92.9%; the lowest was for

amoxicillin+clavulanic acid at 51.0% (25/49 assessable prescriptions) and highest for gentamicin at 92.9% (13/14 assessable prescriptions) (Appendix Figure 2). One-fifth (20.8%) of all antimicrobial prescriptions (100/480 prescriptions) did not have a local guideline available to support prescribing; half of these were for oral or cutaneous/mucosal candidiasis [29/100 assessable prescriptions (29.0%)], or medical or surgical prophylaxis [20/100 assessable prescriptions (20.0%)] (Table 4).

Appropriateness of antimicrobial prescribing

The evaluation of appropriateness of antimicrobial prescribing at the three sites is shown in Table 4. Appendix Figures 3 and 4 further stratify these by indication (as listed in Table 2) and antimicrobial agent (as listed in Table 3).

Of the 451 prescriptions that could be assessed for appropriateness (ie, excluding prescriptions that could not be assessed), 83.1% (375/451) were considered appropriate (Table 4). Prescriptions for most indications were appropriate in at least 80.0% of cases; the exceptions were community-acquired pneumonia [44/56 (78.6%)], chronic obstructive pulmonary disease [8/15 (53.3%)], pelvic inflammatory disease [3/6 (50.0%)] and empyema [1/3 (33.3%)] (Appendix Figure 3). Prescriptions for seven antimicrobial agents were appropriate in less than 80.0% of cases: piperacillin+tazobactam [11/14 (78.6%)], azithromycin [12/16 (75.0%)], doxycycline [12/16 (75.0%)], cefalexin (8/11 (72.7%)], chloramphenicol (topical) [4/6 (66.7%)], ceftriaxone [5/9 (55.6%)] and ciprofloxacin [6/12 (50.0%)]. The most common reason for an inappropriateness assessment was that an agent with too broad a spectrum was used [49/76 (64.5%)]. This was particularly noticeable for community-acquired pneumonia and chronic obstructive pulmonary disease with 38.5% (20/52) inappropriate cases.

Compliance with Pharmaceutical Schedule Section H

Table 4 shows the compliance with PHARMAC Section H, which was 97.7% (469/480 prescriptions) overall and similar across the three sites. Non-compliance in 11 prescriptions was due to lack of documented speciality endorsement (Infectious Diseases or Clinical Microbiology, or Respiratory Medicine as relevant) in the clinical record for use of an agent (ciprofloxacin, piperacillin+tazobactam, cefepime, meropenem or vancomycin) outside of a DHB guideline.

Key indicators—indication, review/stop date, surgical prophylaxis

Table 4 shows results for key indicator assessments at the three sites. Overall, 73.5% (353/480 prescriptions) had the indication for antimicrobial use documented by the prescriber in the clinical notes, 30.2% (145/480 prescriptions) had the review or stop date for antimicrobial therapy documented and 80.3% (53/66 prescriptions) for surgical prophylaxis ceased within 24 hours of the procedure.

Discussion

This is the first antimicrobial prescribing survey across CDHB hospitals, and the only one that we are aware of to be published from a New Zealand DHB. In the absence of a New Zealand national programme, we elected to adopt the Australian method (‘NAPS’) that was first implemented online in 2013. CDHB’s overall survey results were similar to the findings reported in the most recent Australian survey (2018) of participating Australian public hospitals for guidelines compliance (74% vs 76%), appropriateness (83% vs 80%), indication documented (74% vs 84%), review/stop date documented (30% vs 42%) and surgical prophylaxis stopping within 24 hours (80% vs 72%), but indicate improvement is needed for all five quality markers.8 We are unable to directly compare our results with other New Zealand DHBs given the absence of a centralised national programme with transparent reporting.

Antimicrobial prescribing surveys are a useful tool to identify and monitor AMS quality improvement initiatives within DHB hospitals and other healthcare facilities. Our survey results have enabled us to direct our limited AMS resourcing to areas that will likely have the greatest impact on patient safety and quality of care. For example, in addition to noting areas for improvement against the Australian quality markers, we found that surgical prophylaxis, community-acquired pneumonia and chronic obstructive pulmonary disease accounted for quarter of all our antimicrobial prescriptions (29%), and that these were frequently non-compliant with guidelines and inappropriate (39% and 23% of assessable prescriptions, respectively). The main issues were use of unnecessarily broad spectrum agents, an incorrect dosing regimen or an excessive duration as has been identified in the Australian report.8

Since these surveys were performed, we have commenced and/or completed a number of quality improvement initiatives to address some of the issues identified. These have included work to:

• understand the potential prescribing impact and staff resource requirements of a post-prescription AMS ward round.13

• increase utilisation of our electronic prescribing and administration system to facilitate appropriate antimicrobial prescribing, including identifying patients on restricted antimicrobial agents (eg, ciprofloxacin) that could benefit from AMS review and ‘within system’ techniques to facilitate guidelines compliance (eg, prescribing via a guideline rather than entering the prescription manually).

• establish an AMS intranet page for our clinical staff that provides education on AMR and AMS along with access to key resources to support appropriate antimicrobial prescribing including policies and guidelines, antibiograms and contact details for AMS personnel.

• improve understanding of antimicrobial prophylaxis of general surgery and gynaecological procedures (longitudinal auditing), in collaboration with these surgical specialities.

• improve the appropriateness of inpatient quinolone use including changes to prescribing guidelines for urinary tract and intra-abdominal infections, cessation of quinolone susceptibility reporting on urinary isolates, and educational bulletins for staff.

• improve documentation of the indication for antimicrobial use within the electronic prescription.

We plan to conduct a second point prevalence survey at the Christchurch Hospital campus in November 2020 to evaluate our progress since this baseline survey and to inform our future AMS initiatives. Repeat point prevalence surveys have been associated with improvements in most quality improvement markers in Australia,8 likely reflecting between-survey work to address issues identified and greater focus of healthcare facilities on the need for responsible antimicrobial prescribing.14 Our focus on the Christchurch Hospital campus reflects the need to use our limited AMS resourcing judiciously. However, Ashburton Hospital is also planning to undertake their own longitudinal audit to better understand guidelines compliance and appropriateness.

We have shared our survey findings with hospital staff via verbal presentations (education sessions, including a grand round) and through direct engagement with certain specialities (eg, general surgery, gynaecology, respiratory, pharmacy) to assist with quality improvement initiatives. The quality indicator findings of guidelines compliance (one in four antimicrobial prescriptions were not compliant) and appropriateness (nearly one in five prescriptions were not appropriate) will be highlighted during Antibiotic Awareness Week in November 2020.

We attempted to minimise the limitations of this type of survey work by following the recommended NAPS process. As such, the subjectivity of guidelines and appropriateness assessments was mitigated through use of multidisciplinary auditing teams, formal education sessions on the procedures prior to the audit days, and a standardised approach to assessments. The timing of the three surveys (spread out over 12 months) was for practical reasons, and we thought reasonable given that our intent was not to compare the sites which have different case mixes. Repeat surveys at the two main sites (Christchurch Hospital and Burwood) will ideally be done at the same time of year as their respective baseline surveys. NAPS recommends that large facilities like the Christchurch Hospital campus (~830 beds) and Burwood Hospital (~250 beds) are ideally surveyed all at once (hospital-wide) when resourcing permits. This was achieved with the exception of obstetrics and gynaecology, which was studied separately and concurrently with the neonatal and paediatric services (not reported as part of this work). Smaller hospitals (<100 beds) like Ashburton Hospital (~54 beds, with only 32 inpatients on the audit day) should be surveyed repeatedly rather than relying on a ‘snapshot’ approach, which limits the conclusions that can be drawn because of the small number of patients studied. As recent survey work indicates that rural hospitals may not have their AMS needs met by simple transference of a tertiary hospital AMS model,15 it is helpful that more longitudinal auditing is planned at this site.

At least half of New Zealand DHBs now use the Australian NAPS method (eg, standardised approach to assessments, template for data collection), but only a few elect to submit their own data into the NAPS online portal (for pre-set reports and/or data export and self-analysis) and aggregate New Zealand data are not currently provided to enable between-facility comparisons. The success of the Health Quality and Safety Commission’s Surgical Site Infection Improvement Programme16 demonstrates the value of assessing DHB practices against national guidelines and quality indicators. We urge the Ministry of Health to develop or adopt a centralised approach to monitoring the quantity and quality of antimicrobial usage. We note that “national benchmarks…on reducing antimicrobial consumption and increasing appropriate prescribing” is included within the New Zealand AMR Action Plan,1 and that the Health and Disability Standards offer a mechanism to make AMS a priority for all healthcare service providers.

In conclusion, we have completed our first CDHB antimicrobial point prevalence survey covering around 75% of all beds. This provides us with the foundation with which to strategically plan and prioritise AMS work, and monitor for improvement. A co-ordinated and standardised national approach to antimicrobial prescribing surveys, including assessment against our own national guidelines would facilitate AMS within DHBs.

Appendix

Appendix Figure 1: Antimicrobial prescribing guidelines compliance for the most common indications (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least compliant with CDHB guidelines. Number of prescriptions given in brackets after indication. Percentage of prescriptions for each indication specified only when ≥10%.
*Non-oral mucosa, eg, vaginal candidiasis.

Appendix Figure 2: Antimicrobial prescribing guidelines compliance for most commonly prescribed antimicrobial agents (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least compliant with CDHB guidelines. Number of prescriptions given in brackets after indication. Percentage of prescriptions for each indication specified only when ≥10%.

Appendix Figure 3: Appropriateness for the most common indications (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least appropriate. Number of prescriptions given in brackets after the indication. Percentage of prescriptions for each indication only specified where ≥10%.
*Non-oral mucosa, eg, vaginal candidiasis.

Appendix Figure 4: Appropriateness for the most commonly prescribed antimicrobial agents (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least appropriate. Number of prescriptions given in brackets after the indication. Percentage of prescriptions for each indication only specified where ≥10%.

Summary

Abstract

Aim

To determine the nature and appropriateness of antimicrobial prescribing in adult inpatients at Canterbury District Health Board (CDHB).

Method

Multidisciplinary teams collected clinical details for all adult inpatients on antimicrobial therapy at three CDHB facilities (~1,100 beds) and made standardised assessments based on the Australian National Antimicrobial Prescribing Survey (http://naps.org.au) against local guidelines and national funding criteria.

Results

Antimicrobial therapy was prescribed to 42% of inpatients (322/760), usually to treat infections [377/480 prescriptions (79%)], with amoxicillin+clavulanic acid the agent most commonly prescribed [72/480 prescriptions (15%)]. Of assessable prescriptions, 74% (205/278) were guideline compliant, 98% (469/480) were funding criteria compliant, and 83% (375/451) were appropriate clinically. Prescriptions for the most common indications—surgical prophylaxis [66/480 (14%)] and community-acquired pneumonia [56/480 (12%)]—were often non-compliant with guidelines (32% and 41%, respectively) and inappropriate (18% and 21%, respectively). Overall, the indication was documented in 353/480 (74%) prescriptions, the review/stop date documented in 145/480 (30%) prescriptions, and surgical prophylaxis stopped within 24 hours in 53/66 (80%) prescriptions.

Conclusion

Most antimicrobial prescriptions were appropriate and complied with guidelines. Compliance with key quality indicators (indication documented, review/stop date documented, and surgical prophylaxis ceased within 24 hours) were well below target (> 95%) and needs improvement.

Author Information

Sharon J Gardiner, Antimicrobial Stewardship Pharmacist, Infectious Diseases, Clinical Pharmacology and Pharmacy Departments, Christchurch Hospital, Canterbury District Health Board, Christchurch; Ari B Basevi, Medical Student, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Niall L Hamilton, Clinical Pharmacology Registrar, Clinical Pharmacology Department, Christchurch Hospital, Canterbury District Health Board, Christchurch; Sarah CL Metcalf, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Stephen T Chambers, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Pathology Department, University of Otago, Christchurch; Stephen G Withington, Rural Hospital Medicine Physician, Ashburton Hospital, Canterbury District Health Board, Ashburton; Paul K Chin, Clinical Pharmacologist, Clinical Pharmacology Department, Christchurch Hospital, Canterbury District Health Board, Christchurch; Joshua T Freeman, Clinical Microbiologist, Microbiology, Canterbury Health Laboratories, Christchurch; Simon C Dalton, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch.

Acknowledgements

The authors thank their fellow CDHB auditors for their enthusiasm and expertise on the study day: Mark Birch, Vicki Campbell, Kevin Chen, Liane Dixon, Nigel Dean, Matthew Doogue, Nicholas Douglas, Allan Edwards, Bevan Harden, Julia Howard, Heather Isenman, Aaron Keene, Melissa Kerdemelidis, Ashleigh Kortegast, Georgi Lynch, Liz Malcolm, Cate McCall, Michaela Smith, Tim Vincent and Mary Young. Thank you also to Andrew Villazon for technical support.

Correspondence

Sharon J Gardiner, Department of Infectious Diseases, Christchurch Hospital, PO Box 4710, Christchurch.

Correspondence Email

sharon.gardiner@cdhb.health.nz

Competing Interests

Nil.

1. Ministry of Health and Ministry for Primary Industries. New Zealand antimicrobial resistance action plan. Wellington: New Zealand; 2017. Available at: http://www.health.govt.nz/publication/new-zealand-antimicrobial-resistance-action-plan Accessed 04 March 2020.

2. World Health Organisation. Global Action Plan on Antimicrobial Resistance. Geneva: Switzerland; 2015. Available at: http://www.who.int/antimicrobial-resistance/publications/global-action-plan/en/ Accessed 04 March 2020.

3. Duffy E, Gardiner S, du Plessis T, et al. A snapshot of antimicrobial use in New Zealand hospitals – a comparison to Australian and English data. NZ Med J 2015; 128:1421.

4. Australian Commission on Safety and Quality in HealthCare. Antimicrobial Stewardship in Australian Health Care 2018. Sydney: Australia; 2018. Available at: http://www.safetyandquality.gov.au/publications-and-resources/resource-library/antimicrobial-stewardship-australian-health-care Accessed 07 October 2020.

5. British Society for Antimicrobial Chemotherapy. Antimicrobial Stewardship from principles to practice. Birmingham: United Kingdom; 2018. Available at: http://bsac.org.uk/antimicrobial-stewardship-from-principles-to-practice-e-book/ Accessed 06 April 2020.

6. Ministry of Health. Canterbury DHB. Available at: http://www.health.govt.nz/new-zealand-health-system/my-dhb/canterbury-dhb (updated 25 February 2019). Accessed 04 March 2020.

7. Ministry of Health. Canterbury DHB. Wellington: New Zealand; 2020. Available at: http://www.health.govt.nz/your-health/certified-providers/public-hospital. Accessed 27 October 2020.

8. National Centre for Antimicrobial Stewardship. National Antimicrobial Prescribing Survey. Melbourne: Australia; 2018. http://www.naps.org.au. Accessed 27 October 2020.

9. Antimicrobial Stewardship Committee. Canterbury District Health Board. Antimicrobial Guidelines (published in ‘The Pink Book’). Christchurch: New Zealand; 2020. Available at: http://www.pinkbook.org.nz/LoginFiles/Landing.aspx?from=192f5ef9b32e4c0d8ab1f59e9c3e2a83&page=90640.htm Accessed 04 March 2020.

10. Ardagh M, McGeoch G (clinical leaders). Canterbury Hospital HealthPathways. Christchurch: New Zealand; 2020. Available at (password protected): http://canterbury.hospitalhealthpathways.org/ Accessed 04 March 2020.

11. Antibiotic Expert Groups. Therapeutic Guidelines: Antibiotic. Version 16. Melbourne: Australia; 2019. Available at (password protected): http://www.tg.org.au/.  Accessed 27 October 2020.

12. Pharmaceutical Management Agency (PHARMAC). Pharmaceutical Schedule (Section H). Available at: http://www.pharmac.govt.nz/tools-resources/pharmaceutical-schedule/section-h/ Accessed 04 March 2020.

13. Hamilton NH, Gardiner SJ, Chuah Q, et al. Feasibility of a tertiary hospital antimicrobial stewardship ward round using an electronic prescribing system – a pilot study. Appl Clin Inform Open 2020; 4:119–125.

14. Arnoldo L, Smaniotto C, Celotto D, et al. Monitoring healthcare-associated infections and antimicrobial use at regional level through repeated point prevalence surveys: what can be learnt? J Hosp Infect 2019; 101:447–54.

15. Green JK, Gardiner SJ, Clark SL, et al. Antimicrobial stewardship practice in New Zealand’s rural hospitals. NZ Med J 2018; 131:16–26.

16. Health Quality and Safety Commission. Surgical Site Infection Improvement Programme. Available at: http://www.hqsc.govt.nz/our-programmes/infection-prevention-and-control/projects/surgical-site-infection-improvement/ Accessed 05 April 2020.

Contact diana@nzma.org.nz
for the PDF of this article

View Article PDF

Antimicrobial resistance (AMR) is a growing issue that threatens human health globally. To help combat this, both the New Zealand1 and Global2 AMR Action Plans include effective antimicrobial stewardship (AMS) as a core control measure. AMS involves coordinated strategies to optimise antimicrobial use for the prevention and treatment of infections, while minimising possible adverse sequelae of antimicrobial use such as Clostridioides difficile infections, adverse reactions and AMR. The total quantity of inpatient antibacterial use is similar in five of New Zealand’s larger district health boards (DHBs) (700–800 defined daily doses per 1,000 inpatient bed days), and lower than in Australian and English hospitals (mean of ~900 and ~1,300 defined daily doses per 1,000 inpatient bed days, respectively) but higher than in countries like Switzerland and Sweden (~500 and ~300 defined daily doses per 1,000 inpatient bed days, respectively).3 Collectively, these data offer some reassurance regarding the total volume of hospital antimicrobial use in New Zealand DHBs. However, AMS programmes need to assess the quality of antimicrobial prescribing as well as the quantity of antimicrobials used.

Point prevalence surveys can provide a practical and repeatable ‘snapshot’ of antimicrobial prescribing patterns to inform AMS programmes, and monitor their impact.4,5 In this work, we present findings from the first point prevalence surveys of antimicrobial prescribing in adult inpatients at Canterbury DHB (CDHB) hospitals. Our aims were to investigate the:

1. prevalence and nature of antimicrobial use,

2. compliance with local antimicrobial prescribing guidelines,

3. appropriateness of antimicrobial prescribing based on defined criteria, and

4. compliance with funding criteria, per Pharmaceutical Management Agency (PHARMAC) Section H (Hospital Medicines List) prescribing restrictions.

Methods

Setting

CDHB provides publicly funded healthcare for a population of ~570,000 people across 14 certified facilities (~1,500 beds).6,7 We sought to understand antimicrobial prescribing in adult inpatients at the three largest CDHB facilities, excluding Hillmorton Hospital, which exclusively cares for people with psychiatric conditions. Together these three facilities comprise ~1,100 beds and ~75% of all CDHB beds:

Christchurch Hospital campus (~830 beds) includes the geographically co-located Christchurch Women’s Hospital and provides a full range of emergency, acute, elective and outpatient services including bone marrow transplant and intensive care units,

Burwood Hospital (~230 beds) provides rehabilitation, elective orthopaedic surgery, spinal injury, geriatric and psychogeriatric services,

Ashburton Hospital (~50 beds) offers secondary-level acute medical and surgical services and is situated in Ashburton ~90 km south of Christchurch.

The remaining CDHB facilities were excluded from this as they provide psychiatric care (eg, Hillmorton Hospital) or are small rural hospitals focused mainly on rehabilitative, geriatric and/or maternity care (eg, Oxford Hospital).

Audit development

We modelled our overall audit process on the Australian National Antimicrobial Prescribing Survey (‘NAPS’; www.naps.org.au), using their recommended multidisciplinary and hospital-wide approach.8 Data collection forms were similar to those used in the NAPS and were completed for each patient on antimicrobial therapy. These forms included general demographic information, details of the antimicrobial regimen(s) and specifics regarding antimicrobial allergies, surgical procedures and microbiology results where relevant. The start date, route of administration, dose, frequency, review/stop date (where documented) and indication (documented by prescriber or if this was absent, presumed indication if auditors were able to determine this from the clinical record) were recorded for each antimicrobial prescribed. Key points of difference between the CDHB and NAPS processes were that we assessed antimicrobial prescribing against CDHB’s guidelines (published in The Pink Book9 and in HealthPathways10) rather than the Australian guidelines (Therapeutic Guidelines: Antibiotic),11 and evaluated compliance with PHARMAC Section H, which provides the criteria for funded medicine use in DHB hospitals.12 Prior to the audit days, we undertook sequential pilot studies in general surgical and medical wards to ensure the data collection form was fit for purpose and to estimate resourcing for the audit days.

Audit process

The audits were conducted on three days over a one-year period: 30 November 2017 (Christchurch and Ashburton Hospitals), 13 June 2018 (Christchurch Women’s Hospital) and 22 November 2018 (Burwood Hospital). On the first audit day, 11 multidisciplinary teams (around one team per 50 patients) consisted of an infectious diseases or clinical microbiology consultant or registrar (advanced trainee) paired with a pharmacist, clinical pharmacologist or registered nurse. On the subsequent audit days, multidisciplinary teams (physician or registrar plus pharmacist) included at least one member from the first audit. Prior to the audit days, teams were given verbal and written education on how to collect data and perform the assessments, to assist with standardisation.

On each audit day, a list of all adult inpatients at 08:00am was generated from the patient management system (PMS) and divided by clinical area among audit teams. Auditors manually reviewed the electronic and hardcopy prescription records for all these patients to identify those who were prescribed at least one antimicrobial agent (antibacterial, antiviral, antifungal or antiparasitic administered via any route) at 08:00am on the study day and/or administered a one-off (‘stat’) antimicrobial dose or surgical prophylaxis in the preceding 24 hours. One data collection form was completed for each patient who fulfilled these criteria. The auditing teams assessed compliance with CDHB antimicrobial prescribing guidelines and PHARMAC Section H restrictions. Guideline compliance was categorised as compliant, non-compliant, directed therapy (according to culture susceptibilities), recommended by Infectious Diseases/Clinical Microbiology, no guideline available, or not assessable (such as when the indication was not known). Further, prescribing was classified as ‘appropriate’ (optimal or adequate) or ‘inappropriate’ (suboptimal or inadequate) based on the NAPS criteria.8 These criteria included consideration of choice of antimicrobial agent, dose, route and duration of treatment, patient allergies and microbiology results. An infectious diseases physician was available to resolve any disputes within multidisciplinary teams. Finally, documentation of indication and review/stop date, and cessation of surgical antimicrobial prophylaxis within 24 hours post-surgery, were examined. These are considered key indicator assessments with target compliance >95%, recommended in Australia.8

Details from the data collection forms were entered into Microsoft Excel™ spreadsheets for analysis. Approximately 10% of entries were checked for accuracy by a second member of the research team as a quality assurance measure. Results are presented as median, interquartile range and range, or as number and percentage, as appropriate. Results are both presented separately and combined for the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital.

Support for the point prevalence audits was given by Te Komiti Whakarite at CDHB. The audit was regarded as out of scope by the Health and Disability Ethics Committee (Ministry of Health, Wellington, New Zealand).

Results

Number of antimicrobials prescribed

The number of inpatients at the three sites on the audit days together with the prevalence of antimicrobial use and patient demographics are shown in Table 1. The overall prevalence of antimicrobial prescribing was 42.4% (322/760 patients), with the Christchurch Hospital campus the highest of the three facilities. There was a total of 480 antimicrobial prescriptions for the 322 patients on antimicrobial therapy; most (79.8%) usage was at the Christchurch Hospital campus (383/480 prescriptions), which comprised 69.3% of all the inpatients (527/760 patients). The majority of patients were prescribed one [208/322 patients (64.6%)], two [79/322 patients (24.5%)] or three [29/322 patients (9.0%)] agents each, while a very small proportion [6/322 patients (1.9%)] had four to six different agents.

Table 1: Demographic details of the adult inpatients on antimicrobial therapy at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aSelf-reported. Patients may identify with more than one ethnicity. Number of ethnicities identified were: Christchurch Hospital campus = 284, Burwood Hospital = 72, Ashburton Hospital = 11 (total 367).

Antimicrobial indication

The indications for antimicrobial use at the three sites are shown in Table 2. Overall, most prescriptions [377/480 prescriptions (78.5%)] were for treatment of infection, with around one-fifth being for surgical or medical prophylaxis [93/480 prescriptions (19.4%]. Six indications accounted for half of all antimicrobial use [249/480 prescriptions (51.9%)]: surgical prophylaxis (13.8%), community-acquired pneumonia (11.7%), intra-abdominal infection (8.3%), urinary tract infection (7.5%), medical prophylaxis (5.6%) and sepsis (5.0%); most of these prescriptions were at the Christchurch Hospital campus [211/249 prescriptions (84.7%)].

Table 2: Indications for antimicrobial use at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined), N(%).

aListed in order from highest to lowest frequency, based on the Christchurch Hospital campus, where CDHB AMS resources are focused. Indications defined as per the Australian National Antimicrobial Prescribing Survey, with some indications combined as indicated by superscript letters.
bPeritonitis, appendicitis, cholangitis, cholecystitis, diverticulitis or intra-abdominal abscess.
cCystitis, pyelonephritis or catheter-associated urinary tract infection.
dCellulitis/erysipelas, abscess/boils/folliculitis.
eProsthetic or native joint.
fSurgical or non-surgical.
gNon-oral mucosa, eg, vaginal candidiasis.
hExacerbation or chronic.

Antimicrobial agents used and route of administration

The top 20 antimicrobial agents used at the three sites and the associated routes of administration are shown in Table 3. These 20 antimicrobial agents accounted for 85.0% (408/480 prescriptions) of all antimicrobial usage, and was similar at the three facilities. Six systemic agents—amoxicillin+clavulanic acid, cefazolin, metronidazole, cefuroxime, flucloxacillin and gentamicin—accounted for nearly half (47.5%) of all antimicrobial use (228/480 prescriptions).

Table 3: Top 20 antimicrobial agents and routes at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aListed in order from highest to lowest frequency, based on the Christchurch Hospital campus where CDHB AMS resources are focused.

Amoxicillin+clavulanic acid use exceeded that of each of the five next most commonly used agents by two- to three-fold. The IV route was used for half of all prescriptions [232/480 prescriptions (48.3%)], and was the predominant route at the Christchurch Hospital campus (54.6%).

Compliance with antimicrobial prescribing guidelines

The overall antimicrobial guidelines compliance at the three sites is shown in Table 4; Appendix Figures 1 and 2 further stratify these by indication (as listed in Table 2) and antimicrobial agent (as listed in Table 3), respectively.

Table 4: Antimicrobial guidelines compliance, appropriateness of prescribing, PHARMAC compliance and key indicators at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aShaded area indicates overall guidelines compliance assessment, which excludes therapy that is directed, recommended by Infectious Diseases/Clinical Microbiology, has no guideline available or is not assessable.
bShaded area indicates overall appropriateness assessment, which excludes non assessable prescriptions.
cPrescriptions may have more than one reason for being inappropriate: Christchurch Hospital campus = 121 reasons across 57 prescriptions, Burwood Hospital = 39 reasons across 14 prescriptions, Ashburton Hospital = 7 reasons across 5 prescriptions.

Of the 278 prescriptions able to be assessed for guidelines compliance, 73.7% (205/278) were compliant. Pelvic inflammatory disease, intra-abdominal infections, skin and soft tissue infections, medical prophylaxis, sepsis and urinary tract infections accounted for 20.6% (99/480) of all prescriptions and were all ≥80% compliant. Surgical prophylaxis and community-acquired pneumonia accounted for 25.4% (122/480) of all prescriptions and had lower guidelines compliance at 67.9% (38/56 assessable prescriptions) and 58.7% (27/46 assessable prescriptions), respectively. Prescribing for infective exacerbation of chronic obstructive pulmonary disease had the lowest guidelines compliance at 36.4% (4/11 assessable prescriptions) (Table 2; Appendix Figure 1). Compliance with the six most commonly used antimicrobial agents, which accounted for around half of all antimicrobial prescriptions, ranged from 51.0% to 92.9%; the lowest was for

amoxicillin+clavulanic acid at 51.0% (25/49 assessable prescriptions) and highest for gentamicin at 92.9% (13/14 assessable prescriptions) (Appendix Figure 2). One-fifth (20.8%) of all antimicrobial prescriptions (100/480 prescriptions) did not have a local guideline available to support prescribing; half of these were for oral or cutaneous/mucosal candidiasis [29/100 assessable prescriptions (29.0%)], or medical or surgical prophylaxis [20/100 assessable prescriptions (20.0%)] (Table 4).

Appropriateness of antimicrobial prescribing

The evaluation of appropriateness of antimicrobial prescribing at the three sites is shown in Table 4. Appendix Figures 3 and 4 further stratify these by indication (as listed in Table 2) and antimicrobial agent (as listed in Table 3).

Of the 451 prescriptions that could be assessed for appropriateness (ie, excluding prescriptions that could not be assessed), 83.1% (375/451) were considered appropriate (Table 4). Prescriptions for most indications were appropriate in at least 80.0% of cases; the exceptions were community-acquired pneumonia [44/56 (78.6%)], chronic obstructive pulmonary disease [8/15 (53.3%)], pelvic inflammatory disease [3/6 (50.0%)] and empyema [1/3 (33.3%)] (Appendix Figure 3). Prescriptions for seven antimicrobial agents were appropriate in less than 80.0% of cases: piperacillin+tazobactam [11/14 (78.6%)], azithromycin [12/16 (75.0%)], doxycycline [12/16 (75.0%)], cefalexin (8/11 (72.7%)], chloramphenicol (topical) [4/6 (66.7%)], ceftriaxone [5/9 (55.6%)] and ciprofloxacin [6/12 (50.0%)]. The most common reason for an inappropriateness assessment was that an agent with too broad a spectrum was used [49/76 (64.5%)]. This was particularly noticeable for community-acquired pneumonia and chronic obstructive pulmonary disease with 38.5% (20/52) inappropriate cases.

Compliance with Pharmaceutical Schedule Section H

Table 4 shows the compliance with PHARMAC Section H, which was 97.7% (469/480 prescriptions) overall and similar across the three sites. Non-compliance in 11 prescriptions was due to lack of documented speciality endorsement (Infectious Diseases or Clinical Microbiology, or Respiratory Medicine as relevant) in the clinical record for use of an agent (ciprofloxacin, piperacillin+tazobactam, cefepime, meropenem or vancomycin) outside of a DHB guideline.

Key indicators—indication, review/stop date, surgical prophylaxis

Table 4 shows results for key indicator assessments at the three sites. Overall, 73.5% (353/480 prescriptions) had the indication for antimicrobial use documented by the prescriber in the clinical notes, 30.2% (145/480 prescriptions) had the review or stop date for antimicrobial therapy documented and 80.3% (53/66 prescriptions) for surgical prophylaxis ceased within 24 hours of the procedure.

Discussion

This is the first antimicrobial prescribing survey across CDHB hospitals, and the only one that we are aware of to be published from a New Zealand DHB. In the absence of a New Zealand national programme, we elected to adopt the Australian method (‘NAPS’) that was first implemented online in 2013. CDHB’s overall survey results were similar to the findings reported in the most recent Australian survey (2018) of participating Australian public hospitals for guidelines compliance (74% vs 76%), appropriateness (83% vs 80%), indication documented (74% vs 84%), review/stop date documented (30% vs 42%) and surgical prophylaxis stopping within 24 hours (80% vs 72%), but indicate improvement is needed for all five quality markers.8 We are unable to directly compare our results with other New Zealand DHBs given the absence of a centralised national programme with transparent reporting.

Antimicrobial prescribing surveys are a useful tool to identify and monitor AMS quality improvement initiatives within DHB hospitals and other healthcare facilities. Our survey results have enabled us to direct our limited AMS resourcing to areas that will likely have the greatest impact on patient safety and quality of care. For example, in addition to noting areas for improvement against the Australian quality markers, we found that surgical prophylaxis, community-acquired pneumonia and chronic obstructive pulmonary disease accounted for quarter of all our antimicrobial prescriptions (29%), and that these were frequently non-compliant with guidelines and inappropriate (39% and 23% of assessable prescriptions, respectively). The main issues were use of unnecessarily broad spectrum agents, an incorrect dosing regimen or an excessive duration as has been identified in the Australian report.8

Since these surveys were performed, we have commenced and/or completed a number of quality improvement initiatives to address some of the issues identified. These have included work to:

• understand the potential prescribing impact and staff resource requirements of a post-prescription AMS ward round.13

• increase utilisation of our electronic prescribing and administration system to facilitate appropriate antimicrobial prescribing, including identifying patients on restricted antimicrobial agents (eg, ciprofloxacin) that could benefit from AMS review and ‘within system’ techniques to facilitate guidelines compliance (eg, prescribing via a guideline rather than entering the prescription manually).

• establish an AMS intranet page for our clinical staff that provides education on AMR and AMS along with access to key resources to support appropriate antimicrobial prescribing including policies and guidelines, antibiograms and contact details for AMS personnel.

• improve understanding of antimicrobial prophylaxis of general surgery and gynaecological procedures (longitudinal auditing), in collaboration with these surgical specialities.

• improve the appropriateness of inpatient quinolone use including changes to prescribing guidelines for urinary tract and intra-abdominal infections, cessation of quinolone susceptibility reporting on urinary isolates, and educational bulletins for staff.

• improve documentation of the indication for antimicrobial use within the electronic prescription.

We plan to conduct a second point prevalence survey at the Christchurch Hospital campus in November 2020 to evaluate our progress since this baseline survey and to inform our future AMS initiatives. Repeat point prevalence surveys have been associated with improvements in most quality improvement markers in Australia,8 likely reflecting between-survey work to address issues identified and greater focus of healthcare facilities on the need for responsible antimicrobial prescribing.14 Our focus on the Christchurch Hospital campus reflects the need to use our limited AMS resourcing judiciously. However, Ashburton Hospital is also planning to undertake their own longitudinal audit to better understand guidelines compliance and appropriateness.

We have shared our survey findings with hospital staff via verbal presentations (education sessions, including a grand round) and through direct engagement with certain specialities (eg, general surgery, gynaecology, respiratory, pharmacy) to assist with quality improvement initiatives. The quality indicator findings of guidelines compliance (one in four antimicrobial prescriptions were not compliant) and appropriateness (nearly one in five prescriptions were not appropriate) will be highlighted during Antibiotic Awareness Week in November 2020.

We attempted to minimise the limitations of this type of survey work by following the recommended NAPS process. As such, the subjectivity of guidelines and appropriateness assessments was mitigated through use of multidisciplinary auditing teams, formal education sessions on the procedures prior to the audit days, and a standardised approach to assessments. The timing of the three surveys (spread out over 12 months) was for practical reasons, and we thought reasonable given that our intent was not to compare the sites which have different case mixes. Repeat surveys at the two main sites (Christchurch Hospital and Burwood) will ideally be done at the same time of year as their respective baseline surveys. NAPS recommends that large facilities like the Christchurch Hospital campus (~830 beds) and Burwood Hospital (~250 beds) are ideally surveyed all at once (hospital-wide) when resourcing permits. This was achieved with the exception of obstetrics and gynaecology, which was studied separately and concurrently with the neonatal and paediatric services (not reported as part of this work). Smaller hospitals (<100 beds) like Ashburton Hospital (~54 beds, with only 32 inpatients on the audit day) should be surveyed repeatedly rather than relying on a ‘snapshot’ approach, which limits the conclusions that can be drawn because of the small number of patients studied. As recent survey work indicates that rural hospitals may not have their AMS needs met by simple transference of a tertiary hospital AMS model,15 it is helpful that more longitudinal auditing is planned at this site.

At least half of New Zealand DHBs now use the Australian NAPS method (eg, standardised approach to assessments, template for data collection), but only a few elect to submit their own data into the NAPS online portal (for pre-set reports and/or data export and self-analysis) and aggregate New Zealand data are not currently provided to enable between-facility comparisons. The success of the Health Quality and Safety Commission’s Surgical Site Infection Improvement Programme16 demonstrates the value of assessing DHB practices against national guidelines and quality indicators. We urge the Ministry of Health to develop or adopt a centralised approach to monitoring the quantity and quality of antimicrobial usage. We note that “national benchmarks…on reducing antimicrobial consumption and increasing appropriate prescribing” is included within the New Zealand AMR Action Plan,1 and that the Health and Disability Standards offer a mechanism to make AMS a priority for all healthcare service providers.

In conclusion, we have completed our first CDHB antimicrobial point prevalence survey covering around 75% of all beds. This provides us with the foundation with which to strategically plan and prioritise AMS work, and monitor for improvement. A co-ordinated and standardised national approach to antimicrobial prescribing surveys, including assessment against our own national guidelines would facilitate AMS within DHBs.

Appendix

Appendix Figure 1: Antimicrobial prescribing guidelines compliance for the most common indications (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least compliant with CDHB guidelines. Number of prescriptions given in brackets after indication. Percentage of prescriptions for each indication specified only when ≥10%.
*Non-oral mucosa, eg, vaginal candidiasis.

Appendix Figure 2: Antimicrobial prescribing guidelines compliance for most commonly prescribed antimicrobial agents (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least compliant with CDHB guidelines. Number of prescriptions given in brackets after indication. Percentage of prescriptions for each indication specified only when ≥10%.

Appendix Figure 3: Appropriateness for the most common indications (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least appropriate. Number of prescriptions given in brackets after the indication. Percentage of prescriptions for each indication only specified where ≥10%.
*Non-oral mucosa, eg, vaginal candidiasis.

Appendix Figure 4: Appropriateness for the most commonly prescribed antimicrobial agents (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least appropriate. Number of prescriptions given in brackets after the indication. Percentage of prescriptions for each indication only specified where ≥10%.

Summary

Abstract

Aim

To determine the nature and appropriateness of antimicrobial prescribing in adult inpatients at Canterbury District Health Board (CDHB).

Method

Multidisciplinary teams collected clinical details for all adult inpatients on antimicrobial therapy at three CDHB facilities (~1,100 beds) and made standardised assessments based on the Australian National Antimicrobial Prescribing Survey (http://naps.org.au) against local guidelines and national funding criteria.

Results

Antimicrobial therapy was prescribed to 42% of inpatients (322/760), usually to treat infections [377/480 prescriptions (79%)], with amoxicillin+clavulanic acid the agent most commonly prescribed [72/480 prescriptions (15%)]. Of assessable prescriptions, 74% (205/278) were guideline compliant, 98% (469/480) were funding criteria compliant, and 83% (375/451) were appropriate clinically. Prescriptions for the most common indications—surgical prophylaxis [66/480 (14%)] and community-acquired pneumonia [56/480 (12%)]—were often non-compliant with guidelines (32% and 41%, respectively) and inappropriate (18% and 21%, respectively). Overall, the indication was documented in 353/480 (74%) prescriptions, the review/stop date documented in 145/480 (30%) prescriptions, and surgical prophylaxis stopped within 24 hours in 53/66 (80%) prescriptions.

Conclusion

Most antimicrobial prescriptions were appropriate and complied with guidelines. Compliance with key quality indicators (indication documented, review/stop date documented, and surgical prophylaxis ceased within 24 hours) were well below target (> 95%) and needs improvement.

Author Information

Sharon J Gardiner, Antimicrobial Stewardship Pharmacist, Infectious Diseases, Clinical Pharmacology and Pharmacy Departments, Christchurch Hospital, Canterbury District Health Board, Christchurch; Ari B Basevi, Medical Student, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Niall L Hamilton, Clinical Pharmacology Registrar, Clinical Pharmacology Department, Christchurch Hospital, Canterbury District Health Board, Christchurch; Sarah CL Metcalf, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Stephen T Chambers, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Pathology Department, University of Otago, Christchurch; Stephen G Withington, Rural Hospital Medicine Physician, Ashburton Hospital, Canterbury District Health Board, Ashburton; Paul K Chin, Clinical Pharmacologist, Clinical Pharmacology Department, Christchurch Hospital, Canterbury District Health Board, Christchurch; Joshua T Freeman, Clinical Microbiologist, Microbiology, Canterbury Health Laboratories, Christchurch; Simon C Dalton, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch.

Acknowledgements

The authors thank their fellow CDHB auditors for their enthusiasm and expertise on the study day: Mark Birch, Vicki Campbell, Kevin Chen, Liane Dixon, Nigel Dean, Matthew Doogue, Nicholas Douglas, Allan Edwards, Bevan Harden, Julia Howard, Heather Isenman, Aaron Keene, Melissa Kerdemelidis, Ashleigh Kortegast, Georgi Lynch, Liz Malcolm, Cate McCall, Michaela Smith, Tim Vincent and Mary Young. Thank you also to Andrew Villazon for technical support.

Correspondence

Sharon J Gardiner, Department of Infectious Diseases, Christchurch Hospital, PO Box 4710, Christchurch.

Correspondence Email

sharon.gardiner@cdhb.health.nz

Competing Interests

Nil.

1. Ministry of Health and Ministry for Primary Industries. New Zealand antimicrobial resistance action plan. Wellington: New Zealand; 2017. Available at: http://www.health.govt.nz/publication/new-zealand-antimicrobial-resistance-action-plan Accessed 04 March 2020.

2. World Health Organisation. Global Action Plan on Antimicrobial Resistance. Geneva: Switzerland; 2015. Available at: http://www.who.int/antimicrobial-resistance/publications/global-action-plan/en/ Accessed 04 March 2020.

3. Duffy E, Gardiner S, du Plessis T, et al. A snapshot of antimicrobial use in New Zealand hospitals – a comparison to Australian and English data. NZ Med J 2015; 128:1421.

4. Australian Commission on Safety and Quality in HealthCare. Antimicrobial Stewardship in Australian Health Care 2018. Sydney: Australia; 2018. Available at: http://www.safetyandquality.gov.au/publications-and-resources/resource-library/antimicrobial-stewardship-australian-health-care Accessed 07 October 2020.

5. British Society for Antimicrobial Chemotherapy. Antimicrobial Stewardship from principles to practice. Birmingham: United Kingdom; 2018. Available at: http://bsac.org.uk/antimicrobial-stewardship-from-principles-to-practice-e-book/ Accessed 06 April 2020.

6. Ministry of Health. Canterbury DHB. Available at: http://www.health.govt.nz/new-zealand-health-system/my-dhb/canterbury-dhb (updated 25 February 2019). Accessed 04 March 2020.

7. Ministry of Health. Canterbury DHB. Wellington: New Zealand; 2020. Available at: http://www.health.govt.nz/your-health/certified-providers/public-hospital. Accessed 27 October 2020.

8. National Centre for Antimicrobial Stewardship. National Antimicrobial Prescribing Survey. Melbourne: Australia; 2018. http://www.naps.org.au. Accessed 27 October 2020.

9. Antimicrobial Stewardship Committee. Canterbury District Health Board. Antimicrobial Guidelines (published in ‘The Pink Book’). Christchurch: New Zealand; 2020. Available at: http://www.pinkbook.org.nz/LoginFiles/Landing.aspx?from=192f5ef9b32e4c0d8ab1f59e9c3e2a83&page=90640.htm Accessed 04 March 2020.

10. Ardagh M, McGeoch G (clinical leaders). Canterbury Hospital HealthPathways. Christchurch: New Zealand; 2020. Available at (password protected): http://canterbury.hospitalhealthpathways.org/ Accessed 04 March 2020.

11. Antibiotic Expert Groups. Therapeutic Guidelines: Antibiotic. Version 16. Melbourne: Australia; 2019. Available at (password protected): http://www.tg.org.au/.  Accessed 27 October 2020.

12. Pharmaceutical Management Agency (PHARMAC). Pharmaceutical Schedule (Section H). Available at: http://www.pharmac.govt.nz/tools-resources/pharmaceutical-schedule/section-h/ Accessed 04 March 2020.

13. Hamilton NH, Gardiner SJ, Chuah Q, et al. Feasibility of a tertiary hospital antimicrobial stewardship ward round using an electronic prescribing system – a pilot study. Appl Clin Inform Open 2020; 4:119–125.

14. Arnoldo L, Smaniotto C, Celotto D, et al. Monitoring healthcare-associated infections and antimicrobial use at regional level through repeated point prevalence surveys: what can be learnt? J Hosp Infect 2019; 101:447–54.

15. Green JK, Gardiner SJ, Clark SL, et al. Antimicrobial stewardship practice in New Zealand’s rural hospitals. NZ Med J 2018; 131:16–26.

16. Health Quality and Safety Commission. Surgical Site Infection Improvement Programme. Available at: http://www.hqsc.govt.nz/our-programmes/infection-prevention-and-control/projects/surgical-site-infection-improvement/ Accessed 05 April 2020.

Contact diana@nzma.org.nz
for the PDF of this article

View Article PDF

Antimicrobial resistance (AMR) is a growing issue that threatens human health globally. To help combat this, both the New Zealand1 and Global2 AMR Action Plans include effective antimicrobial stewardship (AMS) as a core control measure. AMS involves coordinated strategies to optimise antimicrobial use for the prevention and treatment of infections, while minimising possible adverse sequelae of antimicrobial use such as Clostridioides difficile infections, adverse reactions and AMR. The total quantity of inpatient antibacterial use is similar in five of New Zealand’s larger district health boards (DHBs) (700–800 defined daily doses per 1,000 inpatient bed days), and lower than in Australian and English hospitals (mean of ~900 and ~1,300 defined daily doses per 1,000 inpatient bed days, respectively) but higher than in countries like Switzerland and Sweden (~500 and ~300 defined daily doses per 1,000 inpatient bed days, respectively).3 Collectively, these data offer some reassurance regarding the total volume of hospital antimicrobial use in New Zealand DHBs. However, AMS programmes need to assess the quality of antimicrobial prescribing as well as the quantity of antimicrobials used.

Point prevalence surveys can provide a practical and repeatable ‘snapshot’ of antimicrobial prescribing patterns to inform AMS programmes, and monitor their impact.4,5 In this work, we present findings from the first point prevalence surveys of antimicrobial prescribing in adult inpatients at Canterbury DHB (CDHB) hospitals. Our aims were to investigate the:

1. prevalence and nature of antimicrobial use,

2. compliance with local antimicrobial prescribing guidelines,

3. appropriateness of antimicrobial prescribing based on defined criteria, and

4. compliance with funding criteria, per Pharmaceutical Management Agency (PHARMAC) Section H (Hospital Medicines List) prescribing restrictions.

Methods

Setting

CDHB provides publicly funded healthcare for a population of ~570,000 people across 14 certified facilities (~1,500 beds).6,7 We sought to understand antimicrobial prescribing in adult inpatients at the three largest CDHB facilities, excluding Hillmorton Hospital, which exclusively cares for people with psychiatric conditions. Together these three facilities comprise ~1,100 beds and ~75% of all CDHB beds:

Christchurch Hospital campus (~830 beds) includes the geographically co-located Christchurch Women’s Hospital and provides a full range of emergency, acute, elective and outpatient services including bone marrow transplant and intensive care units,

Burwood Hospital (~230 beds) provides rehabilitation, elective orthopaedic surgery, spinal injury, geriatric and psychogeriatric services,

Ashburton Hospital (~50 beds) offers secondary-level acute medical and surgical services and is situated in Ashburton ~90 km south of Christchurch.

The remaining CDHB facilities were excluded from this as they provide psychiatric care (eg, Hillmorton Hospital) or are small rural hospitals focused mainly on rehabilitative, geriatric and/or maternity care (eg, Oxford Hospital).

Audit development

We modelled our overall audit process on the Australian National Antimicrobial Prescribing Survey (‘NAPS’; www.naps.org.au), using their recommended multidisciplinary and hospital-wide approach.8 Data collection forms were similar to those used in the NAPS and were completed for each patient on antimicrobial therapy. These forms included general demographic information, details of the antimicrobial regimen(s) and specifics regarding antimicrobial allergies, surgical procedures and microbiology results where relevant. The start date, route of administration, dose, frequency, review/stop date (where documented) and indication (documented by prescriber or if this was absent, presumed indication if auditors were able to determine this from the clinical record) were recorded for each antimicrobial prescribed. Key points of difference between the CDHB and NAPS processes were that we assessed antimicrobial prescribing against CDHB’s guidelines (published in The Pink Book9 and in HealthPathways10) rather than the Australian guidelines (Therapeutic Guidelines: Antibiotic),11 and evaluated compliance with PHARMAC Section H, which provides the criteria for funded medicine use in DHB hospitals.12 Prior to the audit days, we undertook sequential pilot studies in general surgical and medical wards to ensure the data collection form was fit for purpose and to estimate resourcing for the audit days.

Audit process

The audits were conducted on three days over a one-year period: 30 November 2017 (Christchurch and Ashburton Hospitals), 13 June 2018 (Christchurch Women’s Hospital) and 22 November 2018 (Burwood Hospital). On the first audit day, 11 multidisciplinary teams (around one team per 50 patients) consisted of an infectious diseases or clinical microbiology consultant or registrar (advanced trainee) paired with a pharmacist, clinical pharmacologist or registered nurse. On the subsequent audit days, multidisciplinary teams (physician or registrar plus pharmacist) included at least one member from the first audit. Prior to the audit days, teams were given verbal and written education on how to collect data and perform the assessments, to assist with standardisation.

On each audit day, a list of all adult inpatients at 08:00am was generated from the patient management system (PMS) and divided by clinical area among audit teams. Auditors manually reviewed the electronic and hardcopy prescription records for all these patients to identify those who were prescribed at least one antimicrobial agent (antibacterial, antiviral, antifungal or antiparasitic administered via any route) at 08:00am on the study day and/or administered a one-off (‘stat’) antimicrobial dose or surgical prophylaxis in the preceding 24 hours. One data collection form was completed for each patient who fulfilled these criteria. The auditing teams assessed compliance with CDHB antimicrobial prescribing guidelines and PHARMAC Section H restrictions. Guideline compliance was categorised as compliant, non-compliant, directed therapy (according to culture susceptibilities), recommended by Infectious Diseases/Clinical Microbiology, no guideline available, or not assessable (such as when the indication was not known). Further, prescribing was classified as ‘appropriate’ (optimal or adequate) or ‘inappropriate’ (suboptimal or inadequate) based on the NAPS criteria.8 These criteria included consideration of choice of antimicrobial agent, dose, route and duration of treatment, patient allergies and microbiology results. An infectious diseases physician was available to resolve any disputes within multidisciplinary teams. Finally, documentation of indication and review/stop date, and cessation of surgical antimicrobial prophylaxis within 24 hours post-surgery, were examined. These are considered key indicator assessments with target compliance >95%, recommended in Australia.8

Details from the data collection forms were entered into Microsoft Excel™ spreadsheets for analysis. Approximately 10% of entries were checked for accuracy by a second member of the research team as a quality assurance measure. Results are presented as median, interquartile range and range, or as number and percentage, as appropriate. Results are both presented separately and combined for the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital.

Support for the point prevalence audits was given by Te Komiti Whakarite at CDHB. The audit was regarded as out of scope by the Health and Disability Ethics Committee (Ministry of Health, Wellington, New Zealand).

Results

Number of antimicrobials prescribed

The number of inpatients at the three sites on the audit days together with the prevalence of antimicrobial use and patient demographics are shown in Table 1. The overall prevalence of antimicrobial prescribing was 42.4% (322/760 patients), with the Christchurch Hospital campus the highest of the three facilities. There was a total of 480 antimicrobial prescriptions for the 322 patients on antimicrobial therapy; most (79.8%) usage was at the Christchurch Hospital campus (383/480 prescriptions), which comprised 69.3% of all the inpatients (527/760 patients). The majority of patients were prescribed one [208/322 patients (64.6%)], two [79/322 patients (24.5%)] or three [29/322 patients (9.0%)] agents each, while a very small proportion [6/322 patients (1.9%)] had four to six different agents.

Table 1: Demographic details of the adult inpatients on antimicrobial therapy at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aSelf-reported. Patients may identify with more than one ethnicity. Number of ethnicities identified were: Christchurch Hospital campus = 284, Burwood Hospital = 72, Ashburton Hospital = 11 (total 367).

Antimicrobial indication

The indications for antimicrobial use at the three sites are shown in Table 2. Overall, most prescriptions [377/480 prescriptions (78.5%)] were for treatment of infection, with around one-fifth being for surgical or medical prophylaxis [93/480 prescriptions (19.4%]. Six indications accounted for half of all antimicrobial use [249/480 prescriptions (51.9%)]: surgical prophylaxis (13.8%), community-acquired pneumonia (11.7%), intra-abdominal infection (8.3%), urinary tract infection (7.5%), medical prophylaxis (5.6%) and sepsis (5.0%); most of these prescriptions were at the Christchurch Hospital campus [211/249 prescriptions (84.7%)].

Table 2: Indications for antimicrobial use at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined), N(%).

aListed in order from highest to lowest frequency, based on the Christchurch Hospital campus, where CDHB AMS resources are focused. Indications defined as per the Australian National Antimicrobial Prescribing Survey, with some indications combined as indicated by superscript letters.
bPeritonitis, appendicitis, cholangitis, cholecystitis, diverticulitis or intra-abdominal abscess.
cCystitis, pyelonephritis or catheter-associated urinary tract infection.
dCellulitis/erysipelas, abscess/boils/folliculitis.
eProsthetic or native joint.
fSurgical or non-surgical.
gNon-oral mucosa, eg, vaginal candidiasis.
hExacerbation or chronic.

Antimicrobial agents used and route of administration

The top 20 antimicrobial agents used at the three sites and the associated routes of administration are shown in Table 3. These 20 antimicrobial agents accounted for 85.0% (408/480 prescriptions) of all antimicrobial usage, and was similar at the three facilities. Six systemic agents—amoxicillin+clavulanic acid, cefazolin, metronidazole, cefuroxime, flucloxacillin and gentamicin—accounted for nearly half (47.5%) of all antimicrobial use (228/480 prescriptions).

Table 3: Top 20 antimicrobial agents and routes at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aListed in order from highest to lowest frequency, based on the Christchurch Hospital campus where CDHB AMS resources are focused.

Amoxicillin+clavulanic acid use exceeded that of each of the five next most commonly used agents by two- to three-fold. The IV route was used for half of all prescriptions [232/480 prescriptions (48.3%)], and was the predominant route at the Christchurch Hospital campus (54.6%).

Compliance with antimicrobial prescribing guidelines

The overall antimicrobial guidelines compliance at the three sites is shown in Table 4; Appendix Figures 1 and 2 further stratify these by indication (as listed in Table 2) and antimicrobial agent (as listed in Table 3), respectively.

Table 4: Antimicrobial guidelines compliance, appropriateness of prescribing, PHARMAC compliance and key indicators at the Christchurch Hospital campus, Burwood Hospital and Ashburton Hospital (separately and combined).

aShaded area indicates overall guidelines compliance assessment, which excludes therapy that is directed, recommended by Infectious Diseases/Clinical Microbiology, has no guideline available or is not assessable.
bShaded area indicates overall appropriateness assessment, which excludes non assessable prescriptions.
cPrescriptions may have more than one reason for being inappropriate: Christchurch Hospital campus = 121 reasons across 57 prescriptions, Burwood Hospital = 39 reasons across 14 prescriptions, Ashburton Hospital = 7 reasons across 5 prescriptions.

Of the 278 prescriptions able to be assessed for guidelines compliance, 73.7% (205/278) were compliant. Pelvic inflammatory disease, intra-abdominal infections, skin and soft tissue infections, medical prophylaxis, sepsis and urinary tract infections accounted for 20.6% (99/480) of all prescriptions and were all ≥80% compliant. Surgical prophylaxis and community-acquired pneumonia accounted for 25.4% (122/480) of all prescriptions and had lower guidelines compliance at 67.9% (38/56 assessable prescriptions) and 58.7% (27/46 assessable prescriptions), respectively. Prescribing for infective exacerbation of chronic obstructive pulmonary disease had the lowest guidelines compliance at 36.4% (4/11 assessable prescriptions) (Table 2; Appendix Figure 1). Compliance with the six most commonly used antimicrobial agents, which accounted for around half of all antimicrobial prescriptions, ranged from 51.0% to 92.9%; the lowest was for

amoxicillin+clavulanic acid at 51.0% (25/49 assessable prescriptions) and highest for gentamicin at 92.9% (13/14 assessable prescriptions) (Appendix Figure 2). One-fifth (20.8%) of all antimicrobial prescriptions (100/480 prescriptions) did not have a local guideline available to support prescribing; half of these were for oral or cutaneous/mucosal candidiasis [29/100 assessable prescriptions (29.0%)], or medical or surgical prophylaxis [20/100 assessable prescriptions (20.0%)] (Table 4).

Appropriateness of antimicrobial prescribing

The evaluation of appropriateness of antimicrobial prescribing at the three sites is shown in Table 4. Appendix Figures 3 and 4 further stratify these by indication (as listed in Table 2) and antimicrobial agent (as listed in Table 3).

Of the 451 prescriptions that could be assessed for appropriateness (ie, excluding prescriptions that could not be assessed), 83.1% (375/451) were considered appropriate (Table 4). Prescriptions for most indications were appropriate in at least 80.0% of cases; the exceptions were community-acquired pneumonia [44/56 (78.6%)], chronic obstructive pulmonary disease [8/15 (53.3%)], pelvic inflammatory disease [3/6 (50.0%)] and empyema [1/3 (33.3%)] (Appendix Figure 3). Prescriptions for seven antimicrobial agents were appropriate in less than 80.0% of cases: piperacillin+tazobactam [11/14 (78.6%)], azithromycin [12/16 (75.0%)], doxycycline [12/16 (75.0%)], cefalexin (8/11 (72.7%)], chloramphenicol (topical) [4/6 (66.7%)], ceftriaxone [5/9 (55.6%)] and ciprofloxacin [6/12 (50.0%)]. The most common reason for an inappropriateness assessment was that an agent with too broad a spectrum was used [49/76 (64.5%)]. This was particularly noticeable for community-acquired pneumonia and chronic obstructive pulmonary disease with 38.5% (20/52) inappropriate cases.

Compliance with Pharmaceutical Schedule Section H

Table 4 shows the compliance with PHARMAC Section H, which was 97.7% (469/480 prescriptions) overall and similar across the three sites. Non-compliance in 11 prescriptions was due to lack of documented speciality endorsement (Infectious Diseases or Clinical Microbiology, or Respiratory Medicine as relevant) in the clinical record for use of an agent (ciprofloxacin, piperacillin+tazobactam, cefepime, meropenem or vancomycin) outside of a DHB guideline.

Key indicators—indication, review/stop date, surgical prophylaxis

Table 4 shows results for key indicator assessments at the three sites. Overall, 73.5% (353/480 prescriptions) had the indication for antimicrobial use documented by the prescriber in the clinical notes, 30.2% (145/480 prescriptions) had the review or stop date for antimicrobial therapy documented and 80.3% (53/66 prescriptions) for surgical prophylaxis ceased within 24 hours of the procedure.

Discussion

This is the first antimicrobial prescribing survey across CDHB hospitals, and the only one that we are aware of to be published from a New Zealand DHB. In the absence of a New Zealand national programme, we elected to adopt the Australian method (‘NAPS’) that was first implemented online in 2013. CDHB’s overall survey results were similar to the findings reported in the most recent Australian survey (2018) of participating Australian public hospitals for guidelines compliance (74% vs 76%), appropriateness (83% vs 80%), indication documented (74% vs 84%), review/stop date documented (30% vs 42%) and surgical prophylaxis stopping within 24 hours (80% vs 72%), but indicate improvement is needed for all five quality markers.8 We are unable to directly compare our results with other New Zealand DHBs given the absence of a centralised national programme with transparent reporting.

Antimicrobial prescribing surveys are a useful tool to identify and monitor AMS quality improvement initiatives within DHB hospitals and other healthcare facilities. Our survey results have enabled us to direct our limited AMS resourcing to areas that will likely have the greatest impact on patient safety and quality of care. For example, in addition to noting areas for improvement against the Australian quality markers, we found that surgical prophylaxis, community-acquired pneumonia and chronic obstructive pulmonary disease accounted for quarter of all our antimicrobial prescriptions (29%), and that these were frequently non-compliant with guidelines and inappropriate (39% and 23% of assessable prescriptions, respectively). The main issues were use of unnecessarily broad spectrum agents, an incorrect dosing regimen or an excessive duration as has been identified in the Australian report.8

Since these surveys were performed, we have commenced and/or completed a number of quality improvement initiatives to address some of the issues identified. These have included work to:

• understand the potential prescribing impact and staff resource requirements of a post-prescription AMS ward round.13

• increase utilisation of our electronic prescribing and administration system to facilitate appropriate antimicrobial prescribing, including identifying patients on restricted antimicrobial agents (eg, ciprofloxacin) that could benefit from AMS review and ‘within system’ techniques to facilitate guidelines compliance (eg, prescribing via a guideline rather than entering the prescription manually).

• establish an AMS intranet page for our clinical staff that provides education on AMR and AMS along with access to key resources to support appropriate antimicrobial prescribing including policies and guidelines, antibiograms and contact details for AMS personnel.

• improve understanding of antimicrobial prophylaxis of general surgery and gynaecological procedures (longitudinal auditing), in collaboration with these surgical specialities.

• improve the appropriateness of inpatient quinolone use including changes to prescribing guidelines for urinary tract and intra-abdominal infections, cessation of quinolone susceptibility reporting on urinary isolates, and educational bulletins for staff.

• improve documentation of the indication for antimicrobial use within the electronic prescription.

We plan to conduct a second point prevalence survey at the Christchurch Hospital campus in November 2020 to evaluate our progress since this baseline survey and to inform our future AMS initiatives. Repeat point prevalence surveys have been associated with improvements in most quality improvement markers in Australia,8 likely reflecting between-survey work to address issues identified and greater focus of healthcare facilities on the need for responsible antimicrobial prescribing.14 Our focus on the Christchurch Hospital campus reflects the need to use our limited AMS resourcing judiciously. However, Ashburton Hospital is also planning to undertake their own longitudinal audit to better understand guidelines compliance and appropriateness.

We have shared our survey findings with hospital staff via verbal presentations (education sessions, including a grand round) and through direct engagement with certain specialities (eg, general surgery, gynaecology, respiratory, pharmacy) to assist with quality improvement initiatives. The quality indicator findings of guidelines compliance (one in four antimicrobial prescriptions were not compliant) and appropriateness (nearly one in five prescriptions were not appropriate) will be highlighted during Antibiotic Awareness Week in November 2020.

We attempted to minimise the limitations of this type of survey work by following the recommended NAPS process. As such, the subjectivity of guidelines and appropriateness assessments was mitigated through use of multidisciplinary auditing teams, formal education sessions on the procedures prior to the audit days, and a standardised approach to assessments. The timing of the three surveys (spread out over 12 months) was for practical reasons, and we thought reasonable given that our intent was not to compare the sites which have different case mixes. Repeat surveys at the two main sites (Christchurch Hospital and Burwood) will ideally be done at the same time of year as their respective baseline surveys. NAPS recommends that large facilities like the Christchurch Hospital campus (~830 beds) and Burwood Hospital (~250 beds) are ideally surveyed all at once (hospital-wide) when resourcing permits. This was achieved with the exception of obstetrics and gynaecology, which was studied separately and concurrently with the neonatal and paediatric services (not reported as part of this work). Smaller hospitals (<100 beds) like Ashburton Hospital (~54 beds, with only 32 inpatients on the audit day) should be surveyed repeatedly rather than relying on a ‘snapshot’ approach, which limits the conclusions that can be drawn because of the small number of patients studied. As recent survey work indicates that rural hospitals may not have their AMS needs met by simple transference of a tertiary hospital AMS model,15 it is helpful that more longitudinal auditing is planned at this site.

At least half of New Zealand DHBs now use the Australian NAPS method (eg, standardised approach to assessments, template for data collection), but only a few elect to submit their own data into the NAPS online portal (for pre-set reports and/or data export and self-analysis) and aggregate New Zealand data are not currently provided to enable between-facility comparisons. The success of the Health Quality and Safety Commission’s Surgical Site Infection Improvement Programme16 demonstrates the value of assessing DHB practices against national guidelines and quality indicators. We urge the Ministry of Health to develop or adopt a centralised approach to monitoring the quantity and quality of antimicrobial usage. We note that “national benchmarks…on reducing antimicrobial consumption and increasing appropriate prescribing” is included within the New Zealand AMR Action Plan,1 and that the Health and Disability Standards offer a mechanism to make AMS a priority for all healthcare service providers.

In conclusion, we have completed our first CDHB antimicrobial point prevalence survey covering around 75% of all beds. This provides us with the foundation with which to strategically plan and prioritise AMS work, and monitor for improvement. A co-ordinated and standardised national approach to antimicrobial prescribing surveys, including assessment against our own national guidelines would facilitate AMS within DHBs.

Appendix

Appendix Figure 1: Antimicrobial prescribing guidelines compliance for the most common indications (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least compliant with CDHB guidelines. Number of prescriptions given in brackets after indication. Percentage of prescriptions for each indication specified only when ≥10%.
*Non-oral mucosa, eg, vaginal candidiasis.

Appendix Figure 2: Antimicrobial prescribing guidelines compliance for most commonly prescribed antimicrobial agents (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least compliant with CDHB guidelines. Number of prescriptions given in brackets after indication. Percentage of prescriptions for each indication specified only when ≥10%.

Appendix Figure 3: Appropriateness for the most common indications (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least appropriate. Number of prescriptions given in brackets after the indication. Percentage of prescriptions for each indication only specified where ≥10%.
*Non-oral mucosa, eg, vaginal candidiasis.

Appendix Figure 4: Appropriateness for the most commonly prescribed antimicrobial agents (as listed in Table 2) for the three CDHB sites combined.a

aIndications listed in order from most to least appropriate. Number of prescriptions given in brackets after the indication. Percentage of prescriptions for each indication only specified where ≥10%.

Summary

Abstract

Aim

To determine the nature and appropriateness of antimicrobial prescribing in adult inpatients at Canterbury District Health Board (CDHB).

Method

Multidisciplinary teams collected clinical details for all adult inpatients on antimicrobial therapy at three CDHB facilities (~1,100 beds) and made standardised assessments based on the Australian National Antimicrobial Prescribing Survey (http://naps.org.au) against local guidelines and national funding criteria.

Results

Antimicrobial therapy was prescribed to 42% of inpatients (322/760), usually to treat infections [377/480 prescriptions (79%)], with amoxicillin+clavulanic acid the agent most commonly prescribed [72/480 prescriptions (15%)]. Of assessable prescriptions, 74% (205/278) were guideline compliant, 98% (469/480) were funding criteria compliant, and 83% (375/451) were appropriate clinically. Prescriptions for the most common indications—surgical prophylaxis [66/480 (14%)] and community-acquired pneumonia [56/480 (12%)]—were often non-compliant with guidelines (32% and 41%, respectively) and inappropriate (18% and 21%, respectively). Overall, the indication was documented in 353/480 (74%) prescriptions, the review/stop date documented in 145/480 (30%) prescriptions, and surgical prophylaxis stopped within 24 hours in 53/66 (80%) prescriptions.

Conclusion

Most antimicrobial prescriptions were appropriate and complied with guidelines. Compliance with key quality indicators (indication documented, review/stop date documented, and surgical prophylaxis ceased within 24 hours) were well below target (> 95%) and needs improvement.

Author Information

Sharon J Gardiner, Antimicrobial Stewardship Pharmacist, Infectious Diseases, Clinical Pharmacology and Pharmacy Departments, Christchurch Hospital, Canterbury District Health Board, Christchurch; Ari B Basevi, Medical Student, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Niall L Hamilton, Clinical Pharmacology Registrar, Clinical Pharmacology Department, Christchurch Hospital, Canterbury District Health Board, Christchurch; Sarah CL Metcalf, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Stephen T Chambers, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch; Pathology Department, University of Otago, Christchurch; Stephen G Withington, Rural Hospital Medicine Physician, Ashburton Hospital, Canterbury District Health Board, Ashburton; Paul K Chin, Clinical Pharmacologist, Clinical Pharmacology Department, Christchurch Hospital, Canterbury District Health Board, Christchurch; Joshua T Freeman, Clinical Microbiologist, Microbiology, Canterbury Health Laboratories, Christchurch; Simon C Dalton, Infectious Diseases Physician, Infectious Diseases, Christchurch Hospital, Canterbury District Health Board, Christchurch.

Acknowledgements

The authors thank their fellow CDHB auditors for their enthusiasm and expertise on the study day: Mark Birch, Vicki Campbell, Kevin Chen, Liane Dixon, Nigel Dean, Matthew Doogue, Nicholas Douglas, Allan Edwards, Bevan Harden, Julia Howard, Heather Isenman, Aaron Keene, Melissa Kerdemelidis, Ashleigh Kortegast, Georgi Lynch, Liz Malcolm, Cate McCall, Michaela Smith, Tim Vincent and Mary Young. Thank you also to Andrew Villazon for technical support.

Correspondence

Sharon J Gardiner, Department of Infectious Diseases, Christchurch Hospital, PO Box 4710, Christchurch.

Correspondence Email

sharon.gardiner@cdhb.health.nz

Competing Interests

Nil.

1. Ministry of Health and Ministry for Primary Industries. New Zealand antimicrobial resistance action plan. Wellington: New Zealand; 2017. Available at: http://www.health.govt.nz/publication/new-zealand-antimicrobial-resistance-action-plan Accessed 04 March 2020.

2. World Health Organisation. Global Action Plan on Antimicrobial Resistance. Geneva: Switzerland; 2015. Available at: http://www.who.int/antimicrobial-resistance/publications/global-action-plan/en/ Accessed 04 March 2020.

3. Duffy E, Gardiner S, du Plessis T, et al. A snapshot of antimicrobial use in New Zealand hospitals – a comparison to Australian and English data. NZ Med J 2015; 128:1421.

4. Australian Commission on Safety and Quality in HealthCare. Antimicrobial Stewardship in Australian Health Care 2018. Sydney: Australia; 2018. Available at: http://www.safetyandquality.gov.au/publications-and-resources/resource-library/antimicrobial-stewardship-australian-health-care Accessed 07 October 2020.

5. British Society for Antimicrobial Chemotherapy. Antimicrobial Stewardship from principles to practice. Birmingham: United Kingdom; 2018. Available at: http://bsac.org.uk/antimicrobial-stewardship-from-principles-to-practice-e-book/ Accessed 06 April 2020.

6. Ministry of Health. Canterbury DHB. Available at: http://www.health.govt.nz/new-zealand-health-system/my-dhb/canterbury-dhb (updated 25 February 2019). Accessed 04 March 2020.

7. Ministry of Health. Canterbury DHB. Wellington: New Zealand; 2020. Available at: http://www.health.govt.nz/your-health/certified-providers/public-hospital. Accessed 27 October 2020.

8. National Centre for Antimicrobial Stewardship. National Antimicrobial Prescribing Survey. Melbourne: Australia; 2018. http://www.naps.org.au. Accessed 27 October 2020.

9. Antimicrobial Stewardship Committee. Canterbury District Health Board. Antimicrobial Guidelines (published in ‘The Pink Book’). Christchurch: New Zealand; 2020. Available at: http://www.pinkbook.org.nz/LoginFiles/Landing.aspx?from=192f5ef9b32e4c0d8ab1f59e9c3e2a83&page=90640.htm Accessed 04 March 2020.

10. Ardagh M, McGeoch G (clinical leaders). Canterbury Hospital HealthPathways. Christchurch: New Zealand; 2020. Available at (password protected): http://canterbury.hospitalhealthpathways.org/ Accessed 04 March 2020.

11. Antibiotic Expert Groups. Therapeutic Guidelines: Antibiotic. Version 16. Melbourne: Australia; 2019. Available at (password protected): http://www.tg.org.au/.  Accessed 27 October 2020.

12. Pharmaceutical Management Agency (PHARMAC). Pharmaceutical Schedule (Section H). Available at: http://www.pharmac.govt.nz/tools-resources/pharmaceutical-schedule/section-h/ Accessed 04 March 2020.

13. Hamilton NH, Gardiner SJ, Chuah Q, et al. Feasibility of a tertiary hospital antimicrobial stewardship ward round using an electronic prescribing system – a pilot study. Appl Clin Inform Open 2020; 4:119–125.

14. Arnoldo L, Smaniotto C, Celotto D, et al. Monitoring healthcare-associated infections and antimicrobial use at regional level through repeated point prevalence surveys: what can be learnt? J Hosp Infect 2019; 101:447–54.

15. Green JK, Gardiner SJ, Clark SL, et al. Antimicrobial stewardship practice in New Zealand’s rural hospitals. NZ Med J 2018; 131:16–26.

16. Health Quality and Safety Commission. Surgical Site Infection Improvement Programme. Available at: http://www.hqsc.govt.nz/our-programmes/infection-prevention-and-control/projects/surgical-site-infection-improvement/ Accessed 05 April 2020.

Contact diana@nzma.org.nz
for the PDF of this article

Subscriber Content

The full contents of this pages only available to subscribers.

LOGINSUBSCRIBE