Optimizing carbapenem use through a national quality improvement programme

Optimizing carbapenem use through a national quality improvement programme Abstract Background Concern about increasing carbapenem and piperacillin/tazobactam use led the Scottish Antimicrobial Prescribing Group (SAPG) to develop national guidance on optimal use of these agents, and to implement a quality improvement programme to assess the impact of guidance on practice. Objectives To evaluate how SAPG guidance had been implemented by health boards, assess how this translated into clinical practice, and investigate clinicians’ views and behaviours about prescribing carbapenems and alternative agents. Methods Local implementation of SAPG guidance was assessed using an online survey. A bespoke point prevalence survey was used to evaluate prescribing. Clinicians’ experience of using carbapenems and alternatives was examined through semi-structured interviews. National prescribing data were analysed to assess the impact of the programme. Results There were greater local restrictions for carbapenems than for piperacillin/tazobactam. Laboratory result suppression was inconsistent between boards and carbapenem-sparing antibiotics were not widely available. Compliance with local guidelines was good for meropenem but lower for piperacillin/tazobactam. Indication for use was well documented but review/stop dates were poorly documented for both antibiotics. Decisions to prescribe a carbapenem were influenced by local guidelines and specialist advice. Many clinicians lacked confidence to de-escalate treatment. Use of both antibiotics decreased during the course of the programme. Conclusions A multifaceted quality improvement programme was used to gather intelligence, promote behaviour change, and focus interventions on the use of carbapenems and piperacillin/tazobactam. Use of these antimicrobials decreased during the programme—a trend not seen elsewhere in Europe. The programme could be generalized to other antimicrobials. Introduction MDR Gram-negative bacteria (MDRGNB) are an escalating global problem,1 and in Europe, increases in carbapenem use2 have been associated with increases in MDRGNB.3 In 2015 no European country showed a significant decrease in carbapenem use and use of piperacillin/tazobactam increased compared with 2014 data.4 Globally, carbapenem use is also increasing5 as is the incidence of carbapenem-resistant GNB.6,7 Carbapenems and piperacillin/tazobactam have been designated as critically important antibiotics by the WHO since 2005,8 and in 2013, the Department of Health in England recommended protecting carbapenems and antipseudomonal agents to preserve their efficacy.9 In Scotland, the reported incidence of resistant Gram-negative organisms, including bacteria producing ESBLs, was stable between 2009 and 2012,10 although the small numbers of carbapenemase-producing organisms (CPOs) were increasing year on year. Piperacillin/tazobactam and carbapenem use was relatively low in Scottish hospitals in 2012: 1.9% and 1.3% of total antibiotic use, respectively (DDD/100 admissions), but use of both antibiotics had increased between 2009 and 2014 (by 51.1% and 23.1%, respectively).10 In Scotland, the NHS comprises 14 regional health boards providing hospital and community services, plus one national hospital. The national antimicrobial stewardship programme is led by the Scottish Antimicrobial Prescribing Group (SAPG), an NHS organization hosted by Healthcare Improvement Scotland, and delivered by health board Antimicrobial Management Teams (AMTs). With the increasing threat from MDRGNB and CPOs and increased use of carbapenems and piperacillin/tazobactam in Scotland, in October 2013 SAPG produced guidance related to MDRGNB infections, and disseminated this to AMTs (Figure S1, available as Supplementary data at JAC Online). The guidance emphasized optimizing the use of carbapenems and piperacillin/tazobactam and considering the use of carbapenem-sparing antibiotics (CSAs), e.g. aztreonam, temocillin, fosfomycin and pivmecillinam. The intention was for AMTs to integrate this national guidance within local policies and education programmes. This project aimed to evaluate local implementation of the national guidance and to investigate its impact on clinical practice. Materials and methods Study design The programme was overseen by a multiprofessional steering group. There were three elements: a national implementation survey of health boards’ prescribing guidance and laboratory reporting practice; a bespoke point prevalence survey (PPS) of carbapenems and piperacillin/tazobactam to assess their use in clinical practice; and qualitative interviews in selected boards to explore clinicians’ attitudes, strategies and barriers to the use of these antibiotics and CSAs. Study outputs were regularly shared with SAPG members and AMTs. An interrupted time-series (ITS) analysis of antibiotic use was used to determine the impact of data sharing and clinician awareness of the programme. Survey A Survey Monkey© online tool (Figure S2) consisting of 49 questions was developed to seek feedback on: adoption of the SAPG MDRGNB guidance; implementation strategies; education; current local recommendations for use of carbapenems, piperacillin/tazobactam and CSAs; and local microbiology laboratory policy and practice for Gram-negative isolates. In May 2015, a link to the survey was sent to AMTs (n = 15) asking them to submit one response per board. Responses were compared to assess variation in clinical use and diagnostic microbiology laboratory practice across boards. National PPS A bespoke PPS focusing on meropenem (the predominant carbapenem used by NHS Scotland) and piperacillin/tazobactam was undertaken in all acute Scottish hospitals (n = 32) using the National Antimicrobial Stewardship PPS database and paper data collection forms for ward information and patient information (Figure S3). PPS data coding was based on the European Surveillance of Antimicrobial Consumption dataset (included in Figure S3), and staff were trained through online webinar sessions. The PPS was conducted during a 4 week period in September–October 2015. Information was collected on every prescription of a carbapenem or piperacillin/tazobactam for treatment of infection on the day of the survey. Prescriptions for antibiotic prophylaxis administered in the 24 h prior to the survey were also included although neither antibiotic is recommended for use as prophylaxis. Following completion of data entry, boards could analyse their own data and results were extracted by SAPG to produce summary reports for each board and a national report. Semi-structured interviews A semi-structured interview was developed to explore factors influencing the prescribing of meropenem and CSAs. The interview (Figure S4) consisted of five questions about prescribing, monitoring, reviewing and de-escalating meropenem; five about factors encouraging or limiting the prescription of CSAs; and an opportunity to make any other comments. Four health boards were selected based on either their good practice in use of carbapenems or use of CSAs as identified through the survey and PPS. AMTs within each board identified a representative sample of clinicians from various specialities and grades (Table S1) and each clinician was sent an invitation letter and study information. Twenty nine one-to-one interviews were conducted by author A. M. between June and November 2016. Interviews were audio recorded, transcribed verbatim and anonymized. A thematic analysis was conducted in NVivo 11 by author A. M. and was validated by author S. E. R., followed by the two researchers reaching a consensus on thematic coding. Table 1. Thematic analysis of clinician interviews about meropenem- and carbapenem-sparing agents (CSAs) (n = 21) Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Table 1. Thematic analysis of clinician interviews about meropenem- and carbapenem-sparing agents (CSAs) (n = 21) Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Sharing of project data Summary reports on each phase of the programme were shared via SAPG meetings, and with AMTs via e-mail and presentations at SAPG national network events. ITS analysis Data on carbapenem and piperacillin/tazobactam use between January 2012 and December 2016, as DDDs, were obtained from the Hospital Medicines Utilisation Database—a national database of medicines supply. Population estimates were obtained from National Records of Scotland and data were reported in DDDs per 100 000 population. The time-series was split into three segments to estimate the level and trend changes in the two segments that followed each intervention compared with the preceding segment (Figure 4). Segment 1 was 21 months (January 2012–September 2013) followed by the introduction of the SAPG Guidance in October 2013 (intervention 1). Segment 2 was 19 months (October 2013–April 2015). Intervention 2 was the quality improvement phase which included the AMT survey in May 2015, the bespoke PPS in October 2015, the sharing of reports with boards in January 2016 and the AMT event in March 2016. Segment 3 was 23 months (May 2015–December 2016). A segmented regression analysis of ITS data was used to examine intervention effects,11 using lag terms to adjust models for autocorrelation present in the residual terms and using heteroskedastic robust standard errors when residual terms were not homoskedastic. Intervention effect sizes are the estimated absolute and relative changes, with 95% CIs.12 The absolute change is the difference between the modelled estimate at the specified post-intervention point and the modelled estimate assuming the pre-intervention trend continued. The relative change is the absolute change as a percentage of the modelled estimate at the specified post-intervention timepoint. Absolute and relative effects are calculated at 1 month, 6 months and 18 months after each intervention. All analyses were carried out in SAS (Statistical Analysis Software).13 Ethics Caldicott Guardian approval for use of prescribing information was obtained locally within each health board. Clinicians involved in the interviews gave written informed consent. Formal ethical review and approval were not required because the project was a service evaluation. The project was conducted in accordance with the Declaration of Helsinki and national and institutional standards. Results National survey All 15 health boards responded to the survey and the key results are reported below. Meropenem was reported to be subject to prescribing restrictions in 13 boards (87%), but piperacillin/tazobactam was only restricted in 7 boards (47%) (Figure 1). The most common mechanism for authorization was through an infection specialist (microbiologist or infectious diseases physician) following a restricted antibiotics policy. These policies are not effectively monitored in many boards; however, one small board uses a highly effective coding system that also controls access to stock. Access to meropenem is mostly limited by having a 24 h supply available via an emergency cupboard or located on specific wards. Meropenem susceptibility reporting was automatically suppressed by laboratories in 9/15 boards (60%) but piperacillin/tazobactam only in 5 boards (33%) (Figure 1). Figure 1. View largeDownload slide NHS board responses to survey questions on meropenem and piperacillin/tazobactam use. FY1–2, Junior doctors; ST/CT, Specialty or Core Trainee doctors; BMS, Biomedical Scientist. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide NHS board responses to survey questions on meropenem and piperacillin/tazobactam use. FY1–2, Junior doctors; ST/CT, Specialty or Core Trainee doctors; BMS, Biomedical Scientist. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The four most commonly reported approved indications for meropenem were as second-line treatment of febrile neutropenia (80% of boards), severe sepsis unresponsive to piperacillin/tazobactam (53%), infections with Pseudomonas spp. or resistant Gram-negative organism colonization in cystic fibrosis patients (40%), and exacerbation of bronchiectasis (33%). The following CSAs were formulary approved for use on specialist advice: fosfomycin oral (87% of boards), pivmecillinam (73%), temocillin (67%), fosfomycin intravenous (iv) (60%), aztreonam (53%). Health boards either updated local guidelines based on the SAPG MDRGNB guidance recommendations or reviewed their local guidelines and found them to be inline with the SAPG guidance. Many boards also informed clinicians about the guidance during medical education sessions or electronically. Training on prescribing of carbapenems and piperacillin/tazobactam is integrated into routine training in most boards, mainly targeted to junior and middle-grade medical staff and pharmacists. National PPS PPS data were submitted by all 15 health boards but data from 2 small island health boards were excluded from the analysis due to delays in receiving the data. A total of 12 478 patients were sampled in 32 hospitals; all patients prescribed the study antibiotics on the day of the survey were included. Data were not collected on the total number of antibiotics prescribed or on whether the study antibiotics were prescribed as monotherapy or in combination with other antibiotics. In total, 466 prescriptions were included (129 of meropenem and 337 of piperacillin/tazobactam), and patient demographics are shown in Figure 2(a). The majority of prescriptions were for patients >50 years old (70% of meropenem and 84% of piperacillin/tazobactam) and ∼60% of prescriptions were for ≥4 days. Figure 2(b) shows the number of prescriptions by specialty. The most common diagnoses for meropenem use were pneumonia, intra-abdominal sepsis, febrile neutropenia or clinical sepsis, which accounted for 66% of all prescriptions. For piperacillin/tazobactam, 70% of prescriptions were for pneumonia, intra-abdominal sepsis, febrile neutropenia or bacteraemia. The source of infection was most often community acquired (CAI) defined as present or starting within 48 h of admission (58% of meropenem and 53% of piperacillin/tazobactam prescriptions). The prevalence of CAIs was similar to that observed in the national PPS of hospital-acquired infection and antimicrobial prescribing in 2016.14 Figure 2. View largeDownload slide Summary of data from point prevalence survey of meropenem and piperacillin/tazobactam use. BAC, bacteraemia; HAI, healthcare-associated infection. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide Summary of data from point prevalence survey of meropenem and piperacillin/tazobactam use. BAC, bacteraemia; HAI, healthcare-associated infection. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The reason for the antibiotic prescription was documented in 97% of meropenem prescriptions and 88% of piperacillin/tazobactam prescriptions. Compliance with local policy was 88% for meropenem and 70% for piperacillin/tazobactam. Documentation of a review or stop date for antibiotic prescriptions was 31% for both drugs (Figure 3). Figure 3. View largeDownload slide PPS data for quality measures of meropenem and piperacillin/tazobactam use. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 3. View largeDownload slide PPS data for quality measures of meropenem and piperacillin/tazobactam use. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. To confirm that use of meropenem and piperacillin/tazobactam on the day of the PPS was typical, data were compared with the previous year’s annual use of the drugs in each health board, measured in DDDs (Figure S5). Semi-structured interviews The main themes arising from the thematic analysis of interview data were grouped into three topic areas: initiation of a prescription, continuation of a prescription and areas for improvement. Key findings included the following: clinicians rely on specialists’ (microbiologist/infectious disease) advice regarding initiation (which would be expected given their restricted status) but also relied on specialist advice regarding continuation/de-escalation, which may indicate a lack of confidence amongst clinical teams; acknowledgement of overuse of very broad-spectrum agents; a need for tools to facilitate review, de-escalation and iv to oral switch therapy to support clinicians; lack of awareness and confidence amongst clinicians in using CSAs unless within local guidelines or on microbiology reports or recommendation (Table 1). ITS Monthly carbapenem and piperacillin/tazobactam DDDs per 100 000 population were plotted over the entire study period (Figure 4). Before intervention 1, carbapenems were increasing by 1 DDD per 100 000 population each month (P = 0.006) from a baseline of 128.7 DDDs per 100 000 population. Intervention 1 was associated with an immediate decrease of 21.3 DDDs per 100 000 population (P = 0.001) and a change in trend of 0.58 DDDs per 100 000 population (P = 0.28). Intervention 2 was associated with an immediate reduction of 12.3 DDDs per 100 000 population (P = 0.05) and a change in trend of 2.3 DDDs per 100 000 population (P < 0.001). Before intervention 1 piperacillin/tazobactam was increasing by 1.4 DDDs per 100 000 population each month (P < 0.001) from a baseline of 188.8 DDDs per 100 000 population. Intervention 1 was associated with an immediate increase of 14.9 DDDs per 100 000 population (P = 0.02) and a change in trend of −1.5 DDDs per 100 000 population (P = 0.002). Intervention 2 was associated with an immediate decrease of 17.6 DDDs per 100 000 population and a change in trend of −1.6 DDDs per 100 000 population (P = 0.002). Figure 4. View largeDownload slide NHS Scotland: carbapenem and piperacillin/tazobactam use (DDDs) from January 2012 to March 2017. Intervention 1: SAPG guidance on MDR Gram-negative bacteria (October 2013); intervention 2: quality improvement [AMT Survey (May 2015), bespoke point prevalence survey (October 2015), reports shared with boards (January 2016) and AMT event (March 2016)]. Figure 4. View largeDownload slide NHS Scotland: carbapenem and piperacillin/tazobactam use (DDDs) from January 2012 to March 2017. Intervention 1: SAPG guidance on MDR Gram-negative bacteria (October 2013); intervention 2: quality improvement [AMT Survey (May 2015), bespoke point prevalence survey (October 2015), reports shared with boards (January 2016) and AMT event (March 2016)]. Segmented regression analysis showed that 6 months following the release of SAPG Guidance in October 2013 there was an 11.4% decrease (95% CI 19.0%–3.9%) in carbapenems and a 2.5% increase (95% CI −3.2% to +8.2%) in piperacillin/tazobactam. By April 2015 the intervention effect was diminishing for carbapenem use with a smaller reduction of 6.5% (95% CI −18.4% to +5.5%), whereas piperacillin/tazobactam use showed a decrease of 5.2% (95% CI −12.9% to +2.4%). Six months after the start of the quality improvement work (intervention 2) there was a reduction in carbapenem use of 15.5% (95% CI 8.3%–22.6%), which further decreased to a 28.5% reduction (95% CI 19.3%–37.7%) by November 2016. Piperacillin/tazobactam use continued to decrease after intervention 2, so that by November 2016 there was a 20.4% decrease (95% CI 12.7%–28.1%). Discussion The survey showed that the SAPG MDRGNB guidance was implemented in most boards. Meropenem is more often subject to prescribing restrictions than piperacillin/tazobactam and authorization for use is typically through an infection specialist. There is inconsistency in the approach of microbiology laboratories towards antimicrobial stewardship nationally and the suppression and release of antimicrobial susceptibility testing results occurs via a variety of mechanisms. There is scope and an appetite amongst laboratory clinicians and scientists for standardization, which is being progressed via collaboration between SAPG and the Scottish Microbiology and Virology Network. Most boards only use CSAs for specific indications on specialist advice and only two boards have embraced their use through inclusion in local antibiotic guidance. Barriers to use of CSAs are additional costs compared with generic meropenem and issues with stock shortages. Older CSAs have a limited evidence base and further studies are required to demonstrate their efficacy in the current resistance landscape.15 However, new agents are coming to market (e.g. ceftolozane/tazobactam) and may offer alternatives to carbapenems. SAPG utilizes periodic online surveys of AMTs to obtain feedback on implementation of national stewardship initiatives, barriers to implementation and suggestions for future improvements. This provides an essential evaluation element to the stewardship programme and also informs future planning. The survey on the use of carbapenems and piperacillin/tazobactam was the fourth AMT survey and focused on the implementation of national guidance, which was subsequently reviewed and updated in 201616 to reflect the findings of this work and additional evidence from the literature. A multipronged approach to hospital stewardship is highlighted in the recent Cochrane review,17 and it is therefore encouraging that our survey confirmed that implementation of local guidance was supported by education for key clinical staff. Extension of stewardship training beyond junior and middle-grade doctors to include consultants may be helpful to ensure leadership for stewardship and to drive behaviour change. Antimicrobial pharmacists are also a key source of specialist advice for clinical teams in Scotland, and training for nursing staff is also important owing to their evolving role in stewardship.18 In addition to the reported results, the survey confirmed that most boards monitor consumption of carbapenems and piperacillin/tazobactam quarterly as recommended in national surveillance guidance.19 Consumption reports are shared at AMT meetings and, in many boards, with Infection Prevention and Control Committees, supporting an integrated approach to stewardship. Awareness of consumption trends is crucial to improving prescribing practice and to assessing the impact of interventions.20 The survey described the local processes to support appropriate use of carbapenems and piperacillin/tazobactam, but from a stewardship perspective it is important to understand how this translates into prescribing practice, which was the key aim of the PPS. National PPSs are used throughout Europe21 to evaluate the prevalence of healthcare-associated infections and antimicrobial prescribing, and have provided SAPG with quantitative and qualitative data to inform on areas for improvement.14 In the bespoke PPS, the lack of good documentation for piperacillin/tazobactam use may reflect its place as the ‘go to’ antibiotic for severe infection. The recent worldwide shortage of piperacillin/tazobactam has gone some way to changing this, with national agreement via SAPG in May 2017 to reserve piperacillin/tazobactam for treatment of suspected neutropenic sepsis and as directed by infection specialists for other specific infections. Further analysis of the PPS data showed that carbapenem use was <2% of all antibiotics in all boards and <1% in many. Piperacillin/tazobactam use varied from 1% to >6%, possibly reflecting different controls over use rather than clinical justification. Another key finding from the PPS was that over half of patients had received antibiotics for >72 h and about one-third of these patients had no documented review or stop date recorded in their medical notes. These findings are informing SAPG’s work on antibiotic review to support clinical teams through education and quality improvement tools to optimize prescribing practice. The interviews with clinicians suggest that many prescribers are not confident in reviewing iv antimicrobial therapy in patients with severe infection where oral switch options may be unclear, and there is a perceived need for additional input from infection specialists. Although carbapenems, and to some extent piperacillin/tazobactam, are often prescribed following advice from microbiology, there is a perception that there is a relative lack of follow-up discussion between the clinical team and microbiology. In addition, variance in the suppression or release of full microbiology reports can lead to patients remaining on the original treatment despite clinical improvement and lack of positive microbiology. This can be addressed through antimicrobial ward rounds,22 but these are unlikely to capture all patients prescribed these agents in a timely manner. Therefore, there appears to be a learning need to upskill prescribers, as well as to develop systems to more easily identify prescription of these antibiotics to facilitate review. Evidence from the interviews clearly identified that there was a need for a whole-system approach that includes the organizational systems and local policies (the environment), improved communication within the multidisciplinary team (the clinicians), and better availability and use of CSAs (the medicines). We acknowledge that selection bias is a limitation of this phase of the programme because we involved clinicians in only 4/15 health boards selected based on local good practice. However, they represented boards of varying size, a mix of teaching hospitals and district generals, and urban and rural populations. During the course of this 2 year improvement programme, national use of carbapenems and piperacillin/tazobactam has decreased although there is some variation between boards in terms of reduced consumption. Some of this change can be attributed to the various elements of the programme as illustrated by the ITS analysis. The impact on consumption may be a Hawthorn effect, but measurement and in-depth study of organizational systems coupled with continuous feedback of findings through multiple forums appears to be supportive in reducing use. During the last 2 years, use of CSAs has increased in some health boards, particularly aztreonam and temocillin, and reassuringly there has been no upward trend in the use of third-generation cephalosporins or fluoroquinolones in Scottish hospitals (data not shown). SAPG had previously completed a quality improvement programme for gentamicin and vancomycin,23 and this work on carbapenems used a similar approach. Such programmes utilize several methods to gain intelligence about clinical practice and target areas for improvement. SAPG has an extremely well-engaged network of local AMTs that support our work, facilitating a resource-light approach. The study findings are continuing to shape the direction of SAPG quality improvement initiatives, including: Highlighting the need to feature CSAs in local guidelines and ensure availability of stock. Working with microbiology colleagues to develop a standardized approach to antimicrobial susceptibility testing and reporting. Encouraging boards to develop local systems to identify initiation of a carbapenem to enable a formal review process by the attending clinical team and/or infection specialists. Developing a national standard and supporting toolkit for review of iv antibiotic therapy. This work demonstrates how a multifaceted quality improvement programme can be used to gather intelligence, promote behaviour change and focus interventions to optimize use of very broad-spectrum antibiotics. Recent trends in Scotland in the use of these antibiotics continue to show a downward trend, and rates are significantly lower than in other UK nations.24 Comparison with other European countries4 suggests that Scotland is ‘bucking the trend’ of stable or increasing rates of carbapenem and piperacillin/tazobactam use. We consider this three-part improvement project will be of interest to stewardship colleagues as it can be applied to other antimicrobials to investigate and inform safe and effective clinical practice. Acknowledgements We thank the Carbapenems Steering Group members not already cited as authors: Ms Ysobel Gourlay, Dr David Griffith, Mrs Heather Kennedy, Mrs Susan Roberts, and Mr Alan Timmins. We also thank Dr Becky Wilson of the Scottish Microbiology and Virology Network for analysing survey results relevant to microbiologists, and Dr Emma Corcoran for her assistance in the thematic analysis of interview data.  Interim data from the AMT survey and PPS have been submitted as abstracts to the Federation of Infection Societies (FIS) conferences in 2015 (Poster no. 0069) and 2016 (Poster no. 4744) and to ECCMID 2016 (Poster no. #P0899), accepted for presentation as posters. Funding This work was carried out as part of the work programme of the Scottish Antimicrobial Prescribing Group which is funded by Scottish Government. The work was supported by additional funding from the Scottish Government for an Antimicrobial Pharmacist (Project Manager: A. C.) from June 2015 to March 2016. The work was also supported by a postgraduate pharmacist (A. M.) undertaking a PhD funded by the government of Saudi Arabia represented by the Saudi Arabian Cultural Bureau in London. Transparency declarations None to declare. Supplementary data Figures S1–S5 and Table S1 are available as Supplementary data at JAC Online. References 1 O’Neill J. Tackling Drug-Resistant Infections Globally: Final Report and Recommendations, May 2016 . https://amr-review.org/. 2 European Food Safety Agency . Joint Interagency Antimicrobial Consumption and Resistance Analysis (JIACRA) Report, January 2015 . https://www.efsa.europa.eu/en/efsajournal/pub/4006. 3 European Centre for Disease Prevention and Control . Antimicrobial Resistance Surveillance in Europe 2015. https://ecdc.europa.eu/en/publications-data/antimicrobial-resistance-surveillance-europe-2015?ID=1637&List=4f55ad51-4aed-4d32-b960-af70113dbb90. 4 European Centre for Disease Prevention and Control. Summary of the Latest Data on Antibiotic Consumption in EU: 2016. https://ecdc.europa.eu/en/publications-data/summary-latest-data-antibiotic-consumption-eu-2016. 5 Van Boeckel TP , Gandra S , Ashok A et al. Global antibiotic consumption 2000 to 2010: an analysis of national pharmaceutical sales data . Lancet Infect Dis 2014 ; 14 : 742 – 50 . Google Scholar CrossRef Search ADS PubMed 6 Canton R , Akova M , Carmeli Y et al. Rapid evolution and spread of carbapenemases among Enterobacteriaceae in Europe . Clin Microbiol Infect 2012 ; 18 : 413 – 31 . Google Scholar CrossRef Search ADS PubMed 7 Levy Hara G , Gould I , Endimiani A et al. Detection, treatment, and prevention of carbapenemase-producing Enterobacteriaceae: recommendations from an International Working Group . J Chemother 2013 ; 25 : 129 – 40 . Google Scholar CrossRef Search ADS PubMed 8 World Health Organization . Critically Important Antimicrobials for Human Medicine, 5th Revision 2016: Ranking of Antimicrobial Agents for Risk Management of Antimicrobial Resistance Due to Non-Human Use. Geneva, Switzerland, 2016 . http://apps.who.int/iris/bitstream/10665/255027/1/9789241512220-eng.pdf?ua=1. 9 Department of Health. Antimicrobial Resistance and Healthcare Associated Infection (ARHAI) ACoARaHAI. 4th Annual Report, February 2012–March 2013. 2013. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/256479/ARHAI_Annual_Report_2012-2013.pdf. 10 National Services Scotland. Report on Antimicrobial Use and Resistance in Humans in 2012. 2014 . http://www.isdscotland.org/Health-Topics/Prescribing-and-Medicines/Publications/2014-01-28/2014-01-28-SAPG-2012-Report.pdf. 11 Wagner AK , Soumerai SB , Zhang F et al. Segmented regression analysis of interrupted time series studies in medication use research . J Clin Pharm Ther 2002 ; 27 : 299 – 309 . Google Scholar CrossRef Search ADS PubMed 12 Zhang F , Wagner AK , Soumerai SB et al. Regression based quasi-experimental approach when randomisation is not an option: interrupted time series analysis . J Clin Epid 2009 ; 62 : 143 – 8 . Google Scholar CrossRef Search ADS 13 Statistical Analysis Software SAS Institute Inc. SAS OnlineDoc, Version 9.4. Cary, NC, USA, 2005 . 14 Health Protection Scotland . National Point Prevalence Survey of Healthcare Associated Infection and Antimicrobial Prescribing 2016. 2017 . http://www.hps.scot.nhs.uk/haiic/amr/resourcedetail.aspx?id=3236. 15 Wilson APR. Sparing carbapenem usage . J Antimicrob Chemother 2017 ; 72 : 2410 – 7 . Google Scholar CrossRef Search ADS PubMed 16 Scottish Antimicrobial Prescribing Group . Position Paper on Optimising Antimicrobial Prescribing in Possible or Suspected Infections Due to Multi-Drug Resistant Gram Negative Bacteria, April 2016 . http://www.scottishmedicines.org.uk/files/sapg1/Position_paper_to_Optimise_Antimicrobial_Prescribing_in_MDRGNB.pdf. 17 Davey P , Marwick CA , Scott CL et al. Interventions to improve antibiotic prescribing practices for hospital inpatients . Cochrane Database Syst Rev 2017 ; issue 2: CD003543 . 18 Edwards R , Drumright LN , Kiernan M et al. Covering more territory to fight resistance: considering nurses’ role in antimicrobial stewardship . J Infect Prev 2011 ; 12 : 6 – 10 . Google Scholar CrossRef Search ADS PubMed 19 Scottish Antimicrobial Prescribing Group. Local Surveillance of Antimicrobial Use. 2015 . http://www.scottishmedicines.org.uk/files/sapg1/Local_surveillance_framework_for_antimicrobial_use_2015.pdf. 20 De With K , Allerberger F , Amann S et al. Strategies to enhance rational use of antibiotics in hospital: a guideline by the German Society for Infectious Diseases . Infection 2016 ; 44 : 395 – 439 . Google Scholar CrossRef Search ADS PubMed 21 European Centre for Disease Prevention and Control. Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Use in European Acute Care Hospitals—Protocol Version 5.3, 2016 . https://ecdc.europa.eu/en/publications-data/point-prevalence-survey-healthcare-associated-infections-and-antimicrobial-use-3. 22 National Institute for Health and Care Excellence Guideline (NG15), Antimicrobial Stewardship: Systems and Processes for Effective Antimicrobial Medicine Use. 2015 . https://www.nice.org.uk/guidance/ng15/resources/antimicrobial-stewardship-systems-and-processes-for-effective-antimicrobial-medicine-use-pdf-1837273110469. 23 Newham R , Thomson AH , Semple Y et al. Barriers to the safe and effective use of intravenous gentamicin and vancomycin in Scottish hospitals and strategies for quality improvement . Eur J Hosp Pharm 2015 ; 22 : 32 – 7 . Google Scholar CrossRef Search ADS 24 Health Protection Scotland . Scottish One Health Antimicrobial Use and Antimicrobial Resistance Report 2016. 2017 . http://www.hps.scot.nhs.uk/haiic/amr/resourcedetail.aspx?id=3378. © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices) http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Antimicrobial Chemotherapy Oxford University Press

Loading next page...
 
/lp/ou_press/optimizing-carbapenem-use-through-a-national-quality-improvement-0ypDpmnunp
Publisher
Oxford University Press
Copyright
© The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com.
ISSN
0305-7453
eISSN
1460-2091
D.O.I.
10.1093/jac/dky171
Publisher site
See Article on Publisher Site

Abstract

Abstract Background Concern about increasing carbapenem and piperacillin/tazobactam use led the Scottish Antimicrobial Prescribing Group (SAPG) to develop national guidance on optimal use of these agents, and to implement a quality improvement programme to assess the impact of guidance on practice. Objectives To evaluate how SAPG guidance had been implemented by health boards, assess how this translated into clinical practice, and investigate clinicians’ views and behaviours about prescribing carbapenems and alternative agents. Methods Local implementation of SAPG guidance was assessed using an online survey. A bespoke point prevalence survey was used to evaluate prescribing. Clinicians’ experience of using carbapenems and alternatives was examined through semi-structured interviews. National prescribing data were analysed to assess the impact of the programme. Results There were greater local restrictions for carbapenems than for piperacillin/tazobactam. Laboratory result suppression was inconsistent between boards and carbapenem-sparing antibiotics were not widely available. Compliance with local guidelines was good for meropenem but lower for piperacillin/tazobactam. Indication for use was well documented but review/stop dates were poorly documented for both antibiotics. Decisions to prescribe a carbapenem were influenced by local guidelines and specialist advice. Many clinicians lacked confidence to de-escalate treatment. Use of both antibiotics decreased during the course of the programme. Conclusions A multifaceted quality improvement programme was used to gather intelligence, promote behaviour change, and focus interventions on the use of carbapenems and piperacillin/tazobactam. Use of these antimicrobials decreased during the programme—a trend not seen elsewhere in Europe. The programme could be generalized to other antimicrobials. Introduction MDR Gram-negative bacteria (MDRGNB) are an escalating global problem,1 and in Europe, increases in carbapenem use2 have been associated with increases in MDRGNB.3 In 2015 no European country showed a significant decrease in carbapenem use and use of piperacillin/tazobactam increased compared with 2014 data.4 Globally, carbapenem use is also increasing5 as is the incidence of carbapenem-resistant GNB.6,7 Carbapenems and piperacillin/tazobactam have been designated as critically important antibiotics by the WHO since 2005,8 and in 2013, the Department of Health in England recommended protecting carbapenems and antipseudomonal agents to preserve their efficacy.9 In Scotland, the reported incidence of resistant Gram-negative organisms, including bacteria producing ESBLs, was stable between 2009 and 2012,10 although the small numbers of carbapenemase-producing organisms (CPOs) were increasing year on year. Piperacillin/tazobactam and carbapenem use was relatively low in Scottish hospitals in 2012: 1.9% and 1.3% of total antibiotic use, respectively (DDD/100 admissions), but use of both antibiotics had increased between 2009 and 2014 (by 51.1% and 23.1%, respectively).10 In Scotland, the NHS comprises 14 regional health boards providing hospital and community services, plus one national hospital. The national antimicrobial stewardship programme is led by the Scottish Antimicrobial Prescribing Group (SAPG), an NHS organization hosted by Healthcare Improvement Scotland, and delivered by health board Antimicrobial Management Teams (AMTs). With the increasing threat from MDRGNB and CPOs and increased use of carbapenems and piperacillin/tazobactam in Scotland, in October 2013 SAPG produced guidance related to MDRGNB infections, and disseminated this to AMTs (Figure S1, available as Supplementary data at JAC Online). The guidance emphasized optimizing the use of carbapenems and piperacillin/tazobactam and considering the use of carbapenem-sparing antibiotics (CSAs), e.g. aztreonam, temocillin, fosfomycin and pivmecillinam. The intention was for AMTs to integrate this national guidance within local policies and education programmes. This project aimed to evaluate local implementation of the national guidance and to investigate its impact on clinical practice. Materials and methods Study design The programme was overseen by a multiprofessional steering group. There were three elements: a national implementation survey of health boards’ prescribing guidance and laboratory reporting practice; a bespoke point prevalence survey (PPS) of carbapenems and piperacillin/tazobactam to assess their use in clinical practice; and qualitative interviews in selected boards to explore clinicians’ attitudes, strategies and barriers to the use of these antibiotics and CSAs. Study outputs were regularly shared with SAPG members and AMTs. An interrupted time-series (ITS) analysis of antibiotic use was used to determine the impact of data sharing and clinician awareness of the programme. Survey A Survey Monkey© online tool (Figure S2) consisting of 49 questions was developed to seek feedback on: adoption of the SAPG MDRGNB guidance; implementation strategies; education; current local recommendations for use of carbapenems, piperacillin/tazobactam and CSAs; and local microbiology laboratory policy and practice for Gram-negative isolates. In May 2015, a link to the survey was sent to AMTs (n = 15) asking them to submit one response per board. Responses were compared to assess variation in clinical use and diagnostic microbiology laboratory practice across boards. National PPS A bespoke PPS focusing on meropenem (the predominant carbapenem used by NHS Scotland) and piperacillin/tazobactam was undertaken in all acute Scottish hospitals (n = 32) using the National Antimicrobial Stewardship PPS database and paper data collection forms for ward information and patient information (Figure S3). PPS data coding was based on the European Surveillance of Antimicrobial Consumption dataset (included in Figure S3), and staff were trained through online webinar sessions. The PPS was conducted during a 4 week period in September–October 2015. Information was collected on every prescription of a carbapenem or piperacillin/tazobactam for treatment of infection on the day of the survey. Prescriptions for antibiotic prophylaxis administered in the 24 h prior to the survey were also included although neither antibiotic is recommended for use as prophylaxis. Following completion of data entry, boards could analyse their own data and results were extracted by SAPG to produce summary reports for each board and a national report. Semi-structured interviews A semi-structured interview was developed to explore factors influencing the prescribing of meropenem and CSAs. The interview (Figure S4) consisted of five questions about prescribing, monitoring, reviewing and de-escalating meropenem; five about factors encouraging or limiting the prescription of CSAs; and an opportunity to make any other comments. Four health boards were selected based on either their good practice in use of carbapenems or use of CSAs as identified through the survey and PPS. AMTs within each board identified a representative sample of clinicians from various specialities and grades (Table S1) and each clinician was sent an invitation letter and study information. Twenty nine one-to-one interviews were conducted by author A. M. between June and November 2016. Interviews were audio recorded, transcribed verbatim and anonymized. A thematic analysis was conducted in NVivo 11 by author A. M. and was validated by author S. E. R., followed by the two researchers reaching a consensus on thematic coding. Table 1. Thematic analysis of clinician interviews about meropenem- and carbapenem-sparing agents (CSAs) (n = 21) Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Table 1. Thematic analysis of clinician interviews about meropenem- and carbapenem-sparing agents (CSAs) (n = 21) Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Topic Themes Initiation phase Factors influencing prescribing of meropenem and CSAs: Local guidelines and policies Prescribers seeking advice or laboratory results Patient-related factors Carbapenem-sparing agent prescribing levers Continuation phase Factors influencing review of meropenem and CSA prescriptions: Formal review policy and guidance Duration documentation De-escalation guide Microbiology evidence and reports Areas for improvement Factors to target identified by clinicians: Better communication with specialists and within clinical teams Review prescribing practice in high usage wards Piperacillin/tazobactam overuse Audit and feedback to prescribers on their use Sharing of project data Summary reports on each phase of the programme were shared via SAPG meetings, and with AMTs via e-mail and presentations at SAPG national network events. ITS analysis Data on carbapenem and piperacillin/tazobactam use between January 2012 and December 2016, as DDDs, were obtained from the Hospital Medicines Utilisation Database—a national database of medicines supply. Population estimates were obtained from National Records of Scotland and data were reported in DDDs per 100 000 population. The time-series was split into three segments to estimate the level and trend changes in the two segments that followed each intervention compared with the preceding segment (Figure 4). Segment 1 was 21 months (January 2012–September 2013) followed by the introduction of the SAPG Guidance in October 2013 (intervention 1). Segment 2 was 19 months (October 2013–April 2015). Intervention 2 was the quality improvement phase which included the AMT survey in May 2015, the bespoke PPS in October 2015, the sharing of reports with boards in January 2016 and the AMT event in March 2016. Segment 3 was 23 months (May 2015–December 2016). A segmented regression analysis of ITS data was used to examine intervention effects,11 using lag terms to adjust models for autocorrelation present in the residual terms and using heteroskedastic robust standard errors when residual terms were not homoskedastic. Intervention effect sizes are the estimated absolute and relative changes, with 95% CIs.12 The absolute change is the difference between the modelled estimate at the specified post-intervention point and the modelled estimate assuming the pre-intervention trend continued. The relative change is the absolute change as a percentage of the modelled estimate at the specified post-intervention timepoint. Absolute and relative effects are calculated at 1 month, 6 months and 18 months after each intervention. All analyses were carried out in SAS (Statistical Analysis Software).13 Ethics Caldicott Guardian approval for use of prescribing information was obtained locally within each health board. Clinicians involved in the interviews gave written informed consent. Formal ethical review and approval were not required because the project was a service evaluation. The project was conducted in accordance with the Declaration of Helsinki and national and institutional standards. Results National survey All 15 health boards responded to the survey and the key results are reported below. Meropenem was reported to be subject to prescribing restrictions in 13 boards (87%), but piperacillin/tazobactam was only restricted in 7 boards (47%) (Figure 1). The most common mechanism for authorization was through an infection specialist (microbiologist or infectious diseases physician) following a restricted antibiotics policy. These policies are not effectively monitored in many boards; however, one small board uses a highly effective coding system that also controls access to stock. Access to meropenem is mostly limited by having a 24 h supply available via an emergency cupboard or located on specific wards. Meropenem susceptibility reporting was automatically suppressed by laboratories in 9/15 boards (60%) but piperacillin/tazobactam only in 5 boards (33%) (Figure 1). Figure 1. View largeDownload slide NHS board responses to survey questions on meropenem and piperacillin/tazobactam use. FY1–2, Junior doctors; ST/CT, Specialty or Core Trainee doctors; BMS, Biomedical Scientist. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 1. View largeDownload slide NHS board responses to survey questions on meropenem and piperacillin/tazobactam use. FY1–2, Junior doctors; ST/CT, Specialty or Core Trainee doctors; BMS, Biomedical Scientist. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The four most commonly reported approved indications for meropenem were as second-line treatment of febrile neutropenia (80% of boards), severe sepsis unresponsive to piperacillin/tazobactam (53%), infections with Pseudomonas spp. or resistant Gram-negative organism colonization in cystic fibrosis patients (40%), and exacerbation of bronchiectasis (33%). The following CSAs were formulary approved for use on specialist advice: fosfomycin oral (87% of boards), pivmecillinam (73%), temocillin (67%), fosfomycin intravenous (iv) (60%), aztreonam (53%). Health boards either updated local guidelines based on the SAPG MDRGNB guidance recommendations or reviewed their local guidelines and found them to be inline with the SAPG guidance. Many boards also informed clinicians about the guidance during medical education sessions or electronically. Training on prescribing of carbapenems and piperacillin/tazobactam is integrated into routine training in most boards, mainly targeted to junior and middle-grade medical staff and pharmacists. National PPS PPS data were submitted by all 15 health boards but data from 2 small island health boards were excluded from the analysis due to delays in receiving the data. A total of 12 478 patients were sampled in 32 hospitals; all patients prescribed the study antibiotics on the day of the survey were included. Data were not collected on the total number of antibiotics prescribed or on whether the study antibiotics were prescribed as monotherapy or in combination with other antibiotics. In total, 466 prescriptions were included (129 of meropenem and 337 of piperacillin/tazobactam), and patient demographics are shown in Figure 2(a). The majority of prescriptions were for patients >50 years old (70% of meropenem and 84% of piperacillin/tazobactam) and ∼60% of prescriptions were for ≥4 days. Figure 2(b) shows the number of prescriptions by specialty. The most common diagnoses for meropenem use were pneumonia, intra-abdominal sepsis, febrile neutropenia or clinical sepsis, which accounted for 66% of all prescriptions. For piperacillin/tazobactam, 70% of prescriptions were for pneumonia, intra-abdominal sepsis, febrile neutropenia or bacteraemia. The source of infection was most often community acquired (CAI) defined as present or starting within 48 h of admission (58% of meropenem and 53% of piperacillin/tazobactam prescriptions). The prevalence of CAIs was similar to that observed in the national PPS of hospital-acquired infection and antimicrobial prescribing in 2016.14 Figure 2. View largeDownload slide Summary of data from point prevalence survey of meropenem and piperacillin/tazobactam use. BAC, bacteraemia; HAI, healthcare-associated infection. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 2. View largeDownload slide Summary of data from point prevalence survey of meropenem and piperacillin/tazobactam use. BAC, bacteraemia; HAI, healthcare-associated infection. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. The reason for the antibiotic prescription was documented in 97% of meropenem prescriptions and 88% of piperacillin/tazobactam prescriptions. Compliance with local policy was 88% for meropenem and 70% for piperacillin/tazobactam. Documentation of a review or stop date for antibiotic prescriptions was 31% for both drugs (Figure 3). Figure 3. View largeDownload slide PPS data for quality measures of meropenem and piperacillin/tazobactam use. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. Figure 3. View largeDownload slide PPS data for quality measures of meropenem and piperacillin/tazobactam use. This figure appears in colour in the online version of JAC and in black and white in the print version of JAC. To confirm that use of meropenem and piperacillin/tazobactam on the day of the PPS was typical, data were compared with the previous year’s annual use of the drugs in each health board, measured in DDDs (Figure S5). Semi-structured interviews The main themes arising from the thematic analysis of interview data were grouped into three topic areas: initiation of a prescription, continuation of a prescription and areas for improvement. Key findings included the following: clinicians rely on specialists’ (microbiologist/infectious disease) advice regarding initiation (which would be expected given their restricted status) but also relied on specialist advice regarding continuation/de-escalation, which may indicate a lack of confidence amongst clinical teams; acknowledgement of overuse of very broad-spectrum agents; a need for tools to facilitate review, de-escalation and iv to oral switch therapy to support clinicians; lack of awareness and confidence amongst clinicians in using CSAs unless within local guidelines or on microbiology reports or recommendation (Table 1). ITS Monthly carbapenem and piperacillin/tazobactam DDDs per 100 000 population were plotted over the entire study period (Figure 4). Before intervention 1, carbapenems were increasing by 1 DDD per 100 000 population each month (P = 0.006) from a baseline of 128.7 DDDs per 100 000 population. Intervention 1 was associated with an immediate decrease of 21.3 DDDs per 100 000 population (P = 0.001) and a change in trend of 0.58 DDDs per 100 000 population (P = 0.28). Intervention 2 was associated with an immediate reduction of 12.3 DDDs per 100 000 population (P = 0.05) and a change in trend of 2.3 DDDs per 100 000 population (P < 0.001). Before intervention 1 piperacillin/tazobactam was increasing by 1.4 DDDs per 100 000 population each month (P < 0.001) from a baseline of 188.8 DDDs per 100 000 population. Intervention 1 was associated with an immediate increase of 14.9 DDDs per 100 000 population (P = 0.02) and a change in trend of −1.5 DDDs per 100 000 population (P = 0.002). Intervention 2 was associated with an immediate decrease of 17.6 DDDs per 100 000 population and a change in trend of −1.6 DDDs per 100 000 population (P = 0.002). Figure 4. View largeDownload slide NHS Scotland: carbapenem and piperacillin/tazobactam use (DDDs) from January 2012 to March 2017. Intervention 1: SAPG guidance on MDR Gram-negative bacteria (October 2013); intervention 2: quality improvement [AMT Survey (May 2015), bespoke point prevalence survey (October 2015), reports shared with boards (January 2016) and AMT event (March 2016)]. Figure 4. View largeDownload slide NHS Scotland: carbapenem and piperacillin/tazobactam use (DDDs) from January 2012 to March 2017. Intervention 1: SAPG guidance on MDR Gram-negative bacteria (October 2013); intervention 2: quality improvement [AMT Survey (May 2015), bespoke point prevalence survey (October 2015), reports shared with boards (January 2016) and AMT event (March 2016)]. Segmented regression analysis showed that 6 months following the release of SAPG Guidance in October 2013 there was an 11.4% decrease (95% CI 19.0%–3.9%) in carbapenems and a 2.5% increase (95% CI −3.2% to +8.2%) in piperacillin/tazobactam. By April 2015 the intervention effect was diminishing for carbapenem use with a smaller reduction of 6.5% (95% CI −18.4% to +5.5%), whereas piperacillin/tazobactam use showed a decrease of 5.2% (95% CI −12.9% to +2.4%). Six months after the start of the quality improvement work (intervention 2) there was a reduction in carbapenem use of 15.5% (95% CI 8.3%–22.6%), which further decreased to a 28.5% reduction (95% CI 19.3%–37.7%) by November 2016. Piperacillin/tazobactam use continued to decrease after intervention 2, so that by November 2016 there was a 20.4% decrease (95% CI 12.7%–28.1%). Discussion The survey showed that the SAPG MDRGNB guidance was implemented in most boards. Meropenem is more often subject to prescribing restrictions than piperacillin/tazobactam and authorization for use is typically through an infection specialist. There is inconsistency in the approach of microbiology laboratories towards antimicrobial stewardship nationally and the suppression and release of antimicrobial susceptibility testing results occurs via a variety of mechanisms. There is scope and an appetite amongst laboratory clinicians and scientists for standardization, which is being progressed via collaboration between SAPG and the Scottish Microbiology and Virology Network. Most boards only use CSAs for specific indications on specialist advice and only two boards have embraced their use through inclusion in local antibiotic guidance. Barriers to use of CSAs are additional costs compared with generic meropenem and issues with stock shortages. Older CSAs have a limited evidence base and further studies are required to demonstrate their efficacy in the current resistance landscape.15 However, new agents are coming to market (e.g. ceftolozane/tazobactam) and may offer alternatives to carbapenems. SAPG utilizes periodic online surveys of AMTs to obtain feedback on implementation of national stewardship initiatives, barriers to implementation and suggestions for future improvements. This provides an essential evaluation element to the stewardship programme and also informs future planning. The survey on the use of carbapenems and piperacillin/tazobactam was the fourth AMT survey and focused on the implementation of national guidance, which was subsequently reviewed and updated in 201616 to reflect the findings of this work and additional evidence from the literature. A multipronged approach to hospital stewardship is highlighted in the recent Cochrane review,17 and it is therefore encouraging that our survey confirmed that implementation of local guidance was supported by education for key clinical staff. Extension of stewardship training beyond junior and middle-grade doctors to include consultants may be helpful to ensure leadership for stewardship and to drive behaviour change. Antimicrobial pharmacists are also a key source of specialist advice for clinical teams in Scotland, and training for nursing staff is also important owing to their evolving role in stewardship.18 In addition to the reported results, the survey confirmed that most boards monitor consumption of carbapenems and piperacillin/tazobactam quarterly as recommended in national surveillance guidance.19 Consumption reports are shared at AMT meetings and, in many boards, with Infection Prevention and Control Committees, supporting an integrated approach to stewardship. Awareness of consumption trends is crucial to improving prescribing practice and to assessing the impact of interventions.20 The survey described the local processes to support appropriate use of carbapenems and piperacillin/tazobactam, but from a stewardship perspective it is important to understand how this translates into prescribing practice, which was the key aim of the PPS. National PPSs are used throughout Europe21 to evaluate the prevalence of healthcare-associated infections and antimicrobial prescribing, and have provided SAPG with quantitative and qualitative data to inform on areas for improvement.14 In the bespoke PPS, the lack of good documentation for piperacillin/tazobactam use may reflect its place as the ‘go to’ antibiotic for severe infection. The recent worldwide shortage of piperacillin/tazobactam has gone some way to changing this, with national agreement via SAPG in May 2017 to reserve piperacillin/tazobactam for treatment of suspected neutropenic sepsis and as directed by infection specialists for other specific infections. Further analysis of the PPS data showed that carbapenem use was <2% of all antibiotics in all boards and <1% in many. Piperacillin/tazobactam use varied from 1% to >6%, possibly reflecting different controls over use rather than clinical justification. Another key finding from the PPS was that over half of patients had received antibiotics for >72 h and about one-third of these patients had no documented review or stop date recorded in their medical notes. These findings are informing SAPG’s work on antibiotic review to support clinical teams through education and quality improvement tools to optimize prescribing practice. The interviews with clinicians suggest that many prescribers are not confident in reviewing iv antimicrobial therapy in patients with severe infection where oral switch options may be unclear, and there is a perceived need for additional input from infection specialists. Although carbapenems, and to some extent piperacillin/tazobactam, are often prescribed following advice from microbiology, there is a perception that there is a relative lack of follow-up discussion between the clinical team and microbiology. In addition, variance in the suppression or release of full microbiology reports can lead to patients remaining on the original treatment despite clinical improvement and lack of positive microbiology. This can be addressed through antimicrobial ward rounds,22 but these are unlikely to capture all patients prescribed these agents in a timely manner. Therefore, there appears to be a learning need to upskill prescribers, as well as to develop systems to more easily identify prescription of these antibiotics to facilitate review. Evidence from the interviews clearly identified that there was a need for a whole-system approach that includes the organizational systems and local policies (the environment), improved communication within the multidisciplinary team (the clinicians), and better availability and use of CSAs (the medicines). We acknowledge that selection bias is a limitation of this phase of the programme because we involved clinicians in only 4/15 health boards selected based on local good practice. However, they represented boards of varying size, a mix of teaching hospitals and district generals, and urban and rural populations. During the course of this 2 year improvement programme, national use of carbapenems and piperacillin/tazobactam has decreased although there is some variation between boards in terms of reduced consumption. Some of this change can be attributed to the various elements of the programme as illustrated by the ITS analysis. The impact on consumption may be a Hawthorn effect, but measurement and in-depth study of organizational systems coupled with continuous feedback of findings through multiple forums appears to be supportive in reducing use. During the last 2 years, use of CSAs has increased in some health boards, particularly aztreonam and temocillin, and reassuringly there has been no upward trend in the use of third-generation cephalosporins or fluoroquinolones in Scottish hospitals (data not shown). SAPG had previously completed a quality improvement programme for gentamicin and vancomycin,23 and this work on carbapenems used a similar approach. Such programmes utilize several methods to gain intelligence about clinical practice and target areas for improvement. SAPG has an extremely well-engaged network of local AMTs that support our work, facilitating a resource-light approach. The study findings are continuing to shape the direction of SAPG quality improvement initiatives, including: Highlighting the need to feature CSAs in local guidelines and ensure availability of stock. Working with microbiology colleagues to develop a standardized approach to antimicrobial susceptibility testing and reporting. Encouraging boards to develop local systems to identify initiation of a carbapenem to enable a formal review process by the attending clinical team and/or infection specialists. Developing a national standard and supporting toolkit for review of iv antibiotic therapy. This work demonstrates how a multifaceted quality improvement programme can be used to gather intelligence, promote behaviour change and focus interventions to optimize use of very broad-spectrum antibiotics. Recent trends in Scotland in the use of these antibiotics continue to show a downward trend, and rates are significantly lower than in other UK nations.24 Comparison with other European countries4 suggests that Scotland is ‘bucking the trend’ of stable or increasing rates of carbapenem and piperacillin/tazobactam use. We consider this three-part improvement project will be of interest to stewardship colleagues as it can be applied to other antimicrobials to investigate and inform safe and effective clinical practice. Acknowledgements We thank the Carbapenems Steering Group members not already cited as authors: Ms Ysobel Gourlay, Dr David Griffith, Mrs Heather Kennedy, Mrs Susan Roberts, and Mr Alan Timmins. We also thank Dr Becky Wilson of the Scottish Microbiology and Virology Network for analysing survey results relevant to microbiologists, and Dr Emma Corcoran for her assistance in the thematic analysis of interview data.  Interim data from the AMT survey and PPS have been submitted as abstracts to the Federation of Infection Societies (FIS) conferences in 2015 (Poster no. 0069) and 2016 (Poster no. 4744) and to ECCMID 2016 (Poster no. #P0899), accepted for presentation as posters. Funding This work was carried out as part of the work programme of the Scottish Antimicrobial Prescribing Group which is funded by Scottish Government. The work was supported by additional funding from the Scottish Government for an Antimicrobial Pharmacist (Project Manager: A. C.) from June 2015 to March 2016. The work was also supported by a postgraduate pharmacist (A. M.) undertaking a PhD funded by the government of Saudi Arabia represented by the Saudi Arabian Cultural Bureau in London. Transparency declarations None to declare. Supplementary data Figures S1–S5 and Table S1 are available as Supplementary data at JAC Online. References 1 O’Neill J. Tackling Drug-Resistant Infections Globally: Final Report and Recommendations, May 2016 . https://amr-review.org/. 2 European Food Safety Agency . Joint Interagency Antimicrobial Consumption and Resistance Analysis (JIACRA) Report, January 2015 . https://www.efsa.europa.eu/en/efsajournal/pub/4006. 3 European Centre for Disease Prevention and Control . Antimicrobial Resistance Surveillance in Europe 2015. https://ecdc.europa.eu/en/publications-data/antimicrobial-resistance-surveillance-europe-2015?ID=1637&List=4f55ad51-4aed-4d32-b960-af70113dbb90. 4 European Centre for Disease Prevention and Control. Summary of the Latest Data on Antibiotic Consumption in EU: 2016. https://ecdc.europa.eu/en/publications-data/summary-latest-data-antibiotic-consumption-eu-2016. 5 Van Boeckel TP , Gandra S , Ashok A et al. Global antibiotic consumption 2000 to 2010: an analysis of national pharmaceutical sales data . Lancet Infect Dis 2014 ; 14 : 742 – 50 . Google Scholar CrossRef Search ADS PubMed 6 Canton R , Akova M , Carmeli Y et al. Rapid evolution and spread of carbapenemases among Enterobacteriaceae in Europe . Clin Microbiol Infect 2012 ; 18 : 413 – 31 . Google Scholar CrossRef Search ADS PubMed 7 Levy Hara G , Gould I , Endimiani A et al. Detection, treatment, and prevention of carbapenemase-producing Enterobacteriaceae: recommendations from an International Working Group . J Chemother 2013 ; 25 : 129 – 40 . Google Scholar CrossRef Search ADS PubMed 8 World Health Organization . Critically Important Antimicrobials for Human Medicine, 5th Revision 2016: Ranking of Antimicrobial Agents for Risk Management of Antimicrobial Resistance Due to Non-Human Use. Geneva, Switzerland, 2016 . http://apps.who.int/iris/bitstream/10665/255027/1/9789241512220-eng.pdf?ua=1. 9 Department of Health. Antimicrobial Resistance and Healthcare Associated Infection (ARHAI) ACoARaHAI. 4th Annual Report, February 2012–March 2013. 2013. https://assets.publishing.service.gov.uk/government/uploads/system/uploads/attachment_data/file/256479/ARHAI_Annual_Report_2012-2013.pdf. 10 National Services Scotland. Report on Antimicrobial Use and Resistance in Humans in 2012. 2014 . http://www.isdscotland.org/Health-Topics/Prescribing-and-Medicines/Publications/2014-01-28/2014-01-28-SAPG-2012-Report.pdf. 11 Wagner AK , Soumerai SB , Zhang F et al. Segmented regression analysis of interrupted time series studies in medication use research . J Clin Pharm Ther 2002 ; 27 : 299 – 309 . Google Scholar CrossRef Search ADS PubMed 12 Zhang F , Wagner AK , Soumerai SB et al. Regression based quasi-experimental approach when randomisation is not an option: interrupted time series analysis . J Clin Epid 2009 ; 62 : 143 – 8 . Google Scholar CrossRef Search ADS 13 Statistical Analysis Software SAS Institute Inc. SAS OnlineDoc, Version 9.4. Cary, NC, USA, 2005 . 14 Health Protection Scotland . National Point Prevalence Survey of Healthcare Associated Infection and Antimicrobial Prescribing 2016. 2017 . http://www.hps.scot.nhs.uk/haiic/amr/resourcedetail.aspx?id=3236. 15 Wilson APR. Sparing carbapenem usage . J Antimicrob Chemother 2017 ; 72 : 2410 – 7 . Google Scholar CrossRef Search ADS PubMed 16 Scottish Antimicrobial Prescribing Group . Position Paper on Optimising Antimicrobial Prescribing in Possible or Suspected Infections Due to Multi-Drug Resistant Gram Negative Bacteria, April 2016 . http://www.scottishmedicines.org.uk/files/sapg1/Position_paper_to_Optimise_Antimicrobial_Prescribing_in_MDRGNB.pdf. 17 Davey P , Marwick CA , Scott CL et al. Interventions to improve antibiotic prescribing practices for hospital inpatients . Cochrane Database Syst Rev 2017 ; issue 2: CD003543 . 18 Edwards R , Drumright LN , Kiernan M et al. Covering more territory to fight resistance: considering nurses’ role in antimicrobial stewardship . J Infect Prev 2011 ; 12 : 6 – 10 . Google Scholar CrossRef Search ADS PubMed 19 Scottish Antimicrobial Prescribing Group. Local Surveillance of Antimicrobial Use. 2015 . http://www.scottishmedicines.org.uk/files/sapg1/Local_surveillance_framework_for_antimicrobial_use_2015.pdf. 20 De With K , Allerberger F , Amann S et al. Strategies to enhance rational use of antibiotics in hospital: a guideline by the German Society for Infectious Diseases . Infection 2016 ; 44 : 395 – 439 . Google Scholar CrossRef Search ADS PubMed 21 European Centre for Disease Prevention and Control. Point Prevalence Survey of Healthcare-Associated Infections and Antimicrobial Use in European Acute Care Hospitals—Protocol Version 5.3, 2016 . https://ecdc.europa.eu/en/publications-data/point-prevalence-survey-healthcare-associated-infections-and-antimicrobial-use-3. 22 National Institute for Health and Care Excellence Guideline (NG15), Antimicrobial Stewardship: Systems and Processes for Effective Antimicrobial Medicine Use. 2015 . https://www.nice.org.uk/guidance/ng15/resources/antimicrobial-stewardship-systems-and-processes-for-effective-antimicrobial-medicine-use-pdf-1837273110469. 23 Newham R , Thomson AH , Semple Y et al. Barriers to the safe and effective use of intravenous gentamicin and vancomycin in Scottish hospitals and strategies for quality improvement . Eur J Hosp Pharm 2015 ; 22 : 32 – 7 . Google Scholar CrossRef Search ADS 24 Health Protection Scotland . Scottish One Health Antimicrobial Use and Antimicrobial Resistance Report 2016. 2017 . http://www.hps.scot.nhs.uk/haiic/amr/resourcedetail.aspx?id=3378. © The Author(s) 2018. Published by Oxford University Press on behalf of the British Society for Antimicrobial Chemotherapy. All rights reserved. For permissions, please email: journals.permissions@oup.com. This article is published and distributed under the terms of the Oxford University Press, Standard Journals Publication Model (https://academic.oup.com/journals/pages/about_us/legal/notices)

Journal

Journal of Antimicrobial ChemotherapyOxford University Press

Published: May 24, 2018

There are no references for this article.

You’re reading a free preview. Subscribe to read the entire article.


DeepDyve is your
personal research library

It’s your single place to instantly
discover and read the research
that matters to you.

Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.

All for just $49/month

Explore the DeepDyve Library

Search

Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly

Organize

Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.

Access

Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.

Your journals are on DeepDyve

Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.

All the latest content is available, no embargo periods.

See the journals in your area

DeepDyve

Freelancer

DeepDyve

Pro

Price

FREE

$49/month
$360/year

Save searches from
Google Scholar,
PubMed

Create lists to
organize your research

Export lists, citations

Read DeepDyve articles

Abstract access only

Unlimited access to over
18 million full-text articles

Print

20 pages / month

PDF Discount

20% off