cdiffl2qu.htm

The Evaluation and Research on Antimicrobial Stewardship's Effect on Clostridium difficile (ERASE C. difficile) Project

1. Is our organization ready for an ASP to reduce C. difficile?

An ASP for reducing C. difficile offers a potentially promising path for facilities invested in and committed to the effort. Developing and implementing a successful ASP will involve structural, process, and cultural changes in your organization. To effect the changes needed in clinical practice, organizations require multiple adjustments in roles, responsibilities, workflow, decisionmaking, and communication.

Failure to assess your organization's readiness for the change at multiple levels can lead to unanticipated implementation challenges. Bringing about organizational change of any type is difficult. You will not want to move ahead until you are confident of your organization's readiness. Even then, it will be important to balance the need to proceed thoughtfully with the need to move quickly enough to show progress and maintain momentum.

Consider the following questions as you evaluate your organization's readiness and identify action steps to prepare.

Return to Questions To Consider

1.1. Do we have the appropriate ASP foundation on which to build?

This toolkit assumes that your hospital already has an ASP or the foundation for an ASP from which to launch the ASP targeted to promote appropriate antibiotic use and potential C. difficile reduction. Implementing and maintaining an effective ASP requires a dedicated multidisciplinary team and ongoing communication and collaboration as well as ongoing monitoring of systems. Further, ongoing monitoring may necessitate adjustments and corrections as you move forward.

Before going further, you should review your facility's current ASP or, if necessary, take steps to develop a basic ASP from which to target C. difficile reduction. To develop an ASP, you can use the GNYHA ASP toolkit (PDF File; Plugin Software Help). Also remember that antimicrobial stewardship is intended to complement other antibiotic prescribing practices and efforts to promote C. difficile prevention practices in your organization. You may also want to review the other prevention measures you have in place.

Is there an active ASP in place? Who are its members and how does it operate?

While team membership will vary among organizations, the traditional core ASP team should include an infectious disease physician and a PharmD with infectious disease training or experience. The clinical microbiologist, infection preventionist, hospital epidemiologist, information technology (IT) representative, and senior administrator will act as liaisons to support and supplement the core ASP team members. Ideally, the team should be supported by an in-house lab and IT resources.

A newer policy statement from the Society for Healthcare Epidemiology of America (SHEA) suggests expanding the team and allowing members without formal infectious disease training to be part of the ASP team. They can facilitate implementation of ASPs and related activities at more health care facilities, including those that may have fewer resources. Resource 1A is the Policy Statement on Antimicrobial Stewardship by SHEA, the Infectious Diseases Society of America (IDSA), and the Pediatric Infectious Diseases Society .

Resources 1B, 1C, and 1D provide background information on ASPs that can serve as reference points for reviewing your current program's team members and activities. 1B is the GNYHA/UHF Antimicrobial Stewardship Toolkit (PDF File; Plugin Software Help). 1C is the IDSA and SHEA Guidelines for Developing an Institutional Program To Enhance Antimicrobial Stewardship (PDF File; Plugin Software Help). 1D contains examples of State and local stewardship programs.

Tool 1Ecan help you assess your current ASP staff resources and practices to determine whether you will need to strengthen the program to most effectively target C. difficile.

What other C. difficile prevention practices are in place?

A range of infection prevention and control practices have been effective in preventing C. difficile. Tool 1F highlights some of the common practices (and provides a reference for C. difficile clinical practice guidelines). You will need to consider how ASP can complement those efforts, or you may decide to strengthen some practices in conjunction with new antimicrobial targeting strategies.

Return to Questions To Consider

1.2. What do we need to do before we begin to use our ASP to target C. difficile?

Even if you have a strong stewardship foundation in place, you will need to assess your facility's stage of readiness for targeting an ASP to C. difficile reduction. Developing an ASP to work toward reducing C. difficile will require a coordinated systems approach with readiness to change on many levels. Staff support and timing are important elements of assessing your facility's readiness.

You will need to begin the process of further developing your facility's ASP by addressing six questions about timing, readiness, and support from various components of your organization. You will need to cycle back to the last five questions throughout your change process. Reviewing these questions will reinforce communication with colleagues and stakeholders and strengthen answers about the clinical case, business case, and resources needed as you gather data and experience.

Do multiple disciplines understand why a structured ASP is needed? Is there an urgency to change?

Changing clinical practices and the organizational structures and processes to support new practices is difficult. There must be a strong motivation to change in order to successfully introduce new practices. You will need to identify a current and compelling reason that makes now the opportune time to undertake this initiative. Has some recent event brought attention to the problem of C. difficile rates and precipitated the desire to change?

There are many potential reasons to implement an ASP to bring about a reduction in C. difficile. Relevant statistics, potential negative patient outcomes, and financial considerations all offer support for establishing an ASP. But local reasons or cases may prompt a more tangible and immediate increase in awareness, with a concurrent desire to move to implement an ASP to reduce C. difficile. For example:

• Your facility experienced a significant increase or spike in C. difficile rates.
• New mandated reporting requirements bring additional external scrutiny to C. difficile rates.
• Your facility is responding to changes in insurance reimbursement policies.
• Your facility had a notable adverse event that was C. difficile related.
• Your facility has been the target of a legal action related to C. difficile.
• Your staff has personal experience of a family member affected by C. difficile.

While the motivation to change may be helped along by external factors, such as Federal mandates, it is most likely to be strong and persistent if all levels of the organization understand the concerns behind the planned change.

Do senior leaders and other key stakeholders support and provide guidance to planning and implementing an ASP?

While individuals who initiate the effort to reduce C. difficile through antimicrobial stewardship may have a clear understanding of the needed changes and the reasons for them, others may not. The level of knowledge and motivation in this area may vary greatly across the organization. Others in your hospital may have different reasons or may not understand the need to change, so it is important to define the issues and state the reasons that now is the time to institute this change. Laying the groundwork in this manner will help support a C. difficile reduction initiative through ASP.

It is crucial to ensure that your organization's leadership team and key stakeholders share the urgency to support development of an ASP to reduce C. difficile and are willing and able to provide support for this change effort. Lessons learned from other antimicrobial stewardship initiatives provide evidence that support is needed from both the top-level administration and those at the bedside. Tool 1Gprovides a checklist that can help you assess the level of leadership support you have for an ASP to reduce C. difficile.

If senior leaders do not already support the effort to reduce C. difficile through strengthening ASP, you will need to build the case for change by framing the issues and your plans most effectively to gain their attention:

• For some leaders, such as your chief financial officer, it may be a business case: How much does C. difficile cost the hospital each year in terms, for example, of longer lengths of stay, additional staff time, increased readmissions, and actual medication costs?
• For senior clinical leaders, such as the clinical chiefs and nurse executive, it may be a clinical case around increased morbidity and mortality associated with C. difficile.
• For your chief executive officer, you need to consider how support for this effort fits with institutional values, current initiatives, and other commitments.

In making your case, you will need to assess the leaders' level of support and their understanding of the need for ongoing practical support in further developing your ASP. Support will involve allocation of needed resources and ongoing senior leadership oversight to ensure focus and accountability for results.

Do key stakeholders in multiple disciplines understand why a structured ASP is needed?

Your senior leadership must understand the need for both structure and staff designated to conducting and enhancing ASP activities. Beyond your senior leaders, you will need to identify key individuals of importance to this initiative by conducting a stakeholder analysis. Tool 1H can help you structure the stakeholder analysis to identify which departments and individuals are needed, what their views are about expanding the ASP, where barriers might exist, and what actions will be needed to obtain the necessary buy-in of departments and individuals.

In addition to the stakeholder analysis, you may want to survey the prescribers in your organization about their views on an ASP. Understanding current attitudes will help you determine where education is needed and where resistance may be encountered. Lack of knowledge or negative attitudes may undermine change efforts if left unaddressed.

Tool 1I is a survey instrument that was used in the ERASE C. difficile Project to assess physician and pharmacist perceptions about antimicrobial resistance. The survey includes the scope of the problem, antibiotic prescribing practices, and thoughts about ASP prior to the intervention.

While prescribers are the key target group in ASPs, C. difficile reduction is a multidisciplinary responsibility. Engaging other clinical staff in supporting ASP and coordinating it with other C. difficile control practices will be critical. Therefore, you may want to also assess their knowledge of C. difficile control best practices and their receptivity to ASP. Tool 1Ior another established survey tool may help you in that assessment.

Is there a clinical and business case for creating an ASP for reducing C. difficile? Do leadership and prescribers understand and support it?

You may decide that the most efficient and effective way to justify your wish to further develop your facility's ASP is by developing and presenting a business case to your senior leadership. In some organizations a business case may be required. In brief, this document will succinctly describe the project and its aims, the risks associated with your plan, the expected outcomes, and the benefits and cost estimates. Resource 1J, Antimicrobial Stewardship: Implementation Tools & Resources: Other Resources , includes a link to two sample program proposals in the Business Case section that you may find helpful.

While the goal of an ASP is to improve patient outcomes, potentially substantial financial benefits may be realized in cost savings. A well-conceived and presented business case that demonstrates to your facility administrators that the resources invested in implementing an ASP will be offset by benefits to the organization may help justify ongoing support. These benefits should be noted both in your original plan for establishing an ASP for C. difficile reduction and in the ASP's results after implementation.

Part of justifying certain ASP costs may require identification of areas where cost savings can be realized. Tool 1K provides a worksheet for developing a business case. Factors that you should consider in developing a business case include:

• Calculating anticipated savings. You may decide to generate anticipated savings data by calculating costs based on current standard of care practices and comparing these costs with estimated costs of proposed ASP activities. In many cases if an ASP already exists, there will probably not be many additional costs for new staff or equipment. Rather, existing staff time and resources will be redirected to highlight the topic of C. difficile in daily ASP activities.
• Calculating actual savings. You may decide to calculate actual costs for treating patients (on individual, unit, or hospitalwide basis) before introduction of ASP activities and after to show actual savings. You can perform this calculation using pilot data for a briefer timeframe or smaller scale to demonstrate savings. These pilot data can then be used to estimate savings if the ASP is implemented on a larger scale.
• Translating ASP activities into cost savings. These calculations will be based on the particular interventions at your facility, such as a shift from an expensive IV antimicrobial agent to an oral equivalent (which is likely less expensive) or a more narrow oral agent, discontinuation of the antibiotic completely, or shortening of the treatment course. Each of these scenarios has potential savings from the drug cost and administration costs, as well as potential savings from earlier discharge of the patient. In addition to the savings realized through the specific ASP activities at your facility, other costs should be accounted for (e.g., certain agents require more frequent laboratory monitoring, adding to the cost of using that particular agent). Finally, your facility may find that through smart antibiotic prescribing practices, your facility as a whole is spending less on antibiotic agents, realizing hospitalwide savings.
• Showing that improved patient outcomes bring about cost savings. Your facility may choose to estimate cost savings realized through improved patient outcomes associated with ASP activities. Such outcomes may include reduced rates of C. difficile (because of strict adherence to all infection control and environmental cleaning, along with the specific stewardship activities); shorter lengths of stay; reduction in toxicity through dose optimization of targeted agents; and reduction in rates of antibiotic resistance. It might be important not to overstate the potential from the hospital-perspective of savings associated with these types of outcomes, as they may: (1) be hard to achieve, (2) take time to see, and (3) in some instances yield mixed financial outcomes for the facility (e.g., when looking purely at the business case, shortened length of stay may or may not actually save the hospital money).

What kinds of resources are needed to develop an effective ASP?

While you may not know at the outset all the kinds of support that will be needed, it is clear that the changes are going to require new or reallocated resources, most likely both human and material.

A critical question, of course, will be who will be responsible for growing the ASP? Beginning with the current ASP team, factoring in the results of the stakeholder analysis conducted with Tool 1H, and including your plans for new ASP strategies, are additional staff members needed? Do the resources exist to augment current staffing?

Less immediately obvious, but equally important, what other resources might be needed? These may include, for example, training, education, actual supplies/equipment/technical equipment and support. Tool 1Lprovides a worksheet for assessing these other resource needs.

What barriers might we face and how might we deal with them?

No new practice or innovation is seamlessly implemented without any difficulties. For example, in the current fiscally tight climate, your facility may be considering a number of programmatic options, all of which are competing for the same limited resources. You will need to think through the barriers you are likely to encounter so that you can anticipate and address them to prevent them from delaying or stopping the ASP. To stimulate your thinking, Tool 1Mdescribes potential barriers to implementing an ASP, drawn from the hospitals participating in the ERASE C. difficile Project, and offers strategies needed to address them if an ASP is to be implemented.

Return to Questions To Consider

2. How do we determine which interventions for reducing C. difficile to implement?

This toolkit describes an individualized approach and tailoring of selected stewardship interventions based on the results of limited case-control studies and identified issues at each facility. As you begin to identify antibiotics to target, you will also need to look at prescribing practices that will be targeted for change. You will need to plan the strategies you will use with prescribers to appropriately limit the targeted antibiotics in use. The strategy will depend on a combination of known evidence-based promising practices and what will work in your organization. To address these issues, you should consider the four questions discussed in the following sections.

Return to Questions To Consider

2.1. What is the institutional risk assessment approach? How can it help us?

A targeted risk assessment offers a promising path for identifying antimicrobial stewardship interventions to reduce C. difficile, for reasons that include the following:

• Most ASPs have limited resources, time, and staff to have activities that affect all antimicrobial prescribing within their health care facility.
• Some traditional components of a stewardship program may have little impact on C. difficile rates.
• While there may be less variability in practices for infection control and environmental cleaning so that bundled approaches have controlled outbreaks and prevented healthcare-associated infections, the same is unlikely to be true of antimicrobial prescribing. Prescribing can vary widely among facilities due to factors such as patient populations, antibiotic formularies, prescriber preference, and local microbiology.
• Antibiotic type and use vary considerably at health care facilities throughout an area (formularies are large).

The institutional risk assessment approach offers a more sophisticated approach than a uniform bundle of interventions at each facility. The tools below provide a roadmap for performing an assessment and information for tallying antimicrobial use in your facility.

The major components of conducting an institutional risk assessment at your facility follow:

• Conduct a limited case-control study to identify antimicrobial risk factors for C. difficile. Specifically, which of the large and diverse array of antibiotics is associated with C. difficile at your facility and potentially among facilities in multicampus institutions? Review the magnitude of the odds ratio (OR; strength of association) and overall use (to address attributable risk, or probability of contracting a disease; e.g., one antibiotic or class may have a lower OR but be used in large volume).
• Identify and select targeted antibiotics for stewardship intervention based on the results of your case-control study.
• Study patterns of use, including, but not limited to, the target populations and infections for which the antibiotics are being prescribed; who is prescribing them; and duration of use. Identify which strategies and how many different strategies should be implemented for each specific target. Your aim is to identify an approach that will address the majority of use of that antibiotic at your facility (i.e., aiming for 80%+ of use).
• Include the following elements in the stewardship strategies and potential interventions, based on antibiotic target:
  • Restrictions and preauthorization of implicated antimicrobials.
  • Audits and feedback to providers of implicated antimicrobials.
  • Flow and algorithms for empiric and streamlined regimens for specific diagnoses/pathogens.
  • Antibiotic order form; automatic stop orders.
  • Novel approaches to use of stewardship staff or technology for stewardship (e.g., software, text paging, Pyxis pharmacy machines for tracking and promoting proper antibiotic prescribing).
  • Educational efforts for clinicians and patients upon diagnosis.

Resource 2A, Institutional Risk Assessment Approach to Selecting Stewardship Interventions, provides a roadmap for this process. Resource 2B, A Comparison of Antibiotic Data Sources, can be used by facilities when considering sources of antibiotic data, including strengths and weaknesses. Resource 2Cdescribes methods for evaluating antibiotic use.

For example, suppose an antibiotic associated with C. difficile at your facility (i.e., with a high OR or high attributable risk [high volume of use, but an OR that may not be as high]) is found to be used primarily for surgical prophylaxis. Strategies to address surgical prophylaxis choices or automatic stop orders may be appropriate to address this antibiotic. However, if the antibiotic is used for a specific syndrome or specific diagnoses, then flows and algorithms for empiric regimens for specific diagnoses, restrictions, or audit and feedback may be useful at your facility.

• The case-control results and use patterns described above would guide your interventions; the literature, resources, and current activities at your facility would also be taken into account when the strategy is designed.
• The approaches your facility develops must address how best to anticipate possible barriers and how to implement the strategy.
• Compliance with the intervention and success (including changes in antibiotic use and rates of C. difficile) would be monitored to decide whether the strategy should be modified or additional interventions added.
• A list of triggers for implementing an additional tier of interventions should be developed.

Return to Questions To Consider

2.2. How do we conduct a time-limited internal case-control study for C. difficile? What are some of the challenges?

To identify antibiotics most associated with C. difficile cases that may be targets of stewardship interventions, an internal, time-limited, focused retrospective case-control study can be performed. The limited case-control study was chosen for a variety of reasons. C. difficile is a rare event, so each facility would need to look at a large volume of patients and antibiotic use to perform a cohort study. In addition, many facilities may not have the resources or expertise to perform complex formal studies. Thus, a "limited" case control focusing primarily on antibiotic exposures seemed more feasible, even for facilities with limited resources.

There is no single perfect way to perform the case-control study. The best methods may vary by the specific characteristics of your facility, such as number of C. difficile cases, resources, personnel, availability of medical records and antibiotic use data, time, and interest. However, in selecting your method, there are common factors to consider, including the following:

• What is the timeframe? How many cases of C. difficile should we use? You can use C. difficile cases identified by the system already in place at your facility. From our experience, looking at cases during a 3- to 6-month period, at least 30 cases will probably be needed to show a statistically significant relationship between exposure to antibiotics and C. difficile. If feasible, looking at more cases or a longer time period may make it easier for you to find targets.
• How should we choose the controls? A discussion of the advantages and disadvantages of choosing controls for case-control studies in general are beyond the scope of this toolkit. (See annotated bibliography for further resources.) From our experience, the most parsimonious method is to compare controls (patients who did not have C. difficile admitted to your facility during the same period as each case) to cases at a 2:1 ratio. However, since patients who develop C. difficile may be different from the hospital population in general, some additional criteria for choosing the controls can be considered, including matching for age, gender, diagnosis, and length of stay or other severity of illness markers.
  • Matching for age (within +/- 5 years), admission date (admitted at same time to same facility as the C. difficile case within +/- 5 days), and no documented C. difficile within 3 months before or after hospital stay seemed to yield similar data as matching for the above additional factors.
  • In addition, it may be practical for stewardship staff to identify controls by choosing the controls by hand from preprinted lists obtainable from hospital administrative sources. Resource 2D, Does Choice of Control Group Affect the Association of Antibiotics With C. difficile-Associated Diarrhea?, is a slide presentation that addresses these issues.
• How should we choose and obtain the needed data? The goal of this focused case-control study is to identify the antibiotics or antibiotic classes most associated with C. difficile (not all factors associated with C. difficile) as a way to identify which antibiotics to target with your interventions. Thus, from our experience, focusing on a few exposures (mostly antibiotic and a few demographics) is optimal. Some observations and suggestions about the data collection follow:
  • The antibiotic and other related data to be aggregated in this suggested case-control study are likely already collected on all patients routinely as part of either patient care or hospital safety measures.
  • The antibiotic data to collect may include: generic antibiotic name, dose, route, interval, and start and stop dates (which gives enough elements to be able to calculate the standard antibiotic metrics).
  • The same antibiotic and related fields are needed for the controls, including all documented antibiotic exposures in the 30 days prior to admission and throughout their inpatient stay. Although not always feasible, it would be ideal to collect data on "all" antibiotic exposures. However, some may have occurred during outpatient visits or at other facilities, and you may not be able to document them completely even if you conduct a detailed retrospective chart review. Thus, including all antibiotics dispensed and documented within your facilities may be more realistic (and obtainable in many facilities in electronic form).
  • Collected data that allow description of cases and controls and ensure that these are similar populations at risk of C. difficile should also include: admission service, admission diagnosis, and top two discharge diagnoses (age and gender are likely to already be collected).
  • Retrospective antibiotic use data may be extracted from electronic patient medical and pharmacy records at your facility, but review of paper medical records may be needed.
  • Antibiotic use data can be summarized in electronic files (e.g., Microsoft® Excel®) to avoid using additional paper worksheets.

Resource 2E, Sample Tracking and Summary Forms for Case-Control Study, can be used to organize your case-control comparisons and summarize results and potential antibiotic targets.

• How do we analyze the data? What are resources for calculating odds ratios and p values?
  • Data on cases and controls can be entered or transferred into an Excel® spreadsheet and analyzed with a Chi square, Fisher's exact, or other appropriate bivariate analysis. You can use SPSS, SAS, STATA, or another statistical software program if your facility has the specific software and capability. Although more detailed analysis including regression can be used for more complex case-control studies, for most facilities it will not be feasible or necessary.

    This is a limited case-control technique and only select antibiotic exposures will be examined. Smaller facilities or those with limited computer support can make the comparison using the Epi Info statistical calculator, a free download that is widely accessible and simple to use (available at http://www.cdc.gov/epiinfo/). You can also perform the patient tally and calculations by hand using this formula:

    Odds ratio (OR) = [a/b]/[c/d] or OR= [a x d]/[b x c]).

    	Exposure to Antibiotic X	No exposure to Antibiotic X
    C. difficile	a	b
    No C. difficile	c	d

    
    

Return to Questions To Consider

2.3. What methods can we use to review the use of potential target antibiotics for intervention activities?

As shown in the Institutional Risk Assessment Roadmap in Resource 2A, multiple approaches can be taken to reviewing the use of potential target antibiotics. You will need to choose the approach that best fits your organization. For example, you can conduct a formal medication review for several weeks or longer of prescribing patterns of the targeted antibiotic. For many facilities, this timeframe is unrealistic; a review of only a few days of prescribing patterns (potentially on random days) may suffice to begin to strategize interventions.

Details to review will include but are not limited to:

• Number of patients given prescriptions/length of therapy.
• Most common prescribers.
• Most common wards or patients receiving target.
• Reason for target drug prescribing (e.g., empiric therapy, directed therapy, prophylactic regimen).
• Most common syndromes and diagnoses treated by the target drug.
• Appropriateness and potential for prescribing changes (choice, length, other options, including not treating if not indicated).

Return to Questions To Consider

2.4. What factors do we need to consider in choosing interventions?

The target antibiotic from the case-control study and the medication review will start to guide your intervention decisions. Resource 2F lists the major intervention types and their respective advantages and disadvantages, and Resource 2G reviews types of antimicrobial stewardship interventions, comparing process measures, antibiotic metrics, and other factors to track. These comparisons may help you determine what might be appropriate for your hospital. Other factors will also need to be considered, including:

• Stewardship staffing and skill set.
• Ability to affect a large enough burden of prescribing, IT, and other external resources.
• Acceptability of activities to prescribers, stewardship team, pharmacy, administration, and other key players at your facility.
• Previous and current stewardship activities (what has and has not worked in the past; what will complement current activities); and
• Literature and best practices.

Examples of intervention strategies from other hospitals as listed in Resource 2Hmay help you identify the interventions most likely to fit your organization.

Return to Questions To Consider

2.5. How do we implement the intervention?

Once you determine what the interventions will be in terms of the factors described above, you need to develop strategies tailored to your organization for implementing the interventions. With leadership support, the C. difficile ASP team will need to guide, coordinate, and support the implementation efforts during the initial phases and as the ASP practices are rolled out across the hospital to intended populations and prescribers.

Because the interventions may involve changes in the way people work, you may have difficulty incorporating them into practice. Our experience has been that many of these efforts can take longer to implement and have the full effect than initially anticipated. Consider the following questions to determine the changes that will be needed:

• Whose help is needed to implement and sustain the interventions?
  • Staff within your facility. A formal review of your intended stewardship activities (especially if these include formulary changes, restrictions, or new policies) may be needed with facility staff, including the antibiotic subcommittee and pharmacy and therapeutics committee, infectious diseases department (including practicing private infectious disease doctors), and pharmacy staff.
  • Senior leadership. Buy-in and ongoing support are needed from your senior leaders.
  • Otherdepartments. Interdepartmental cooperation and other supportive liaisons may be needed. Based on your intervention, you may need support and resources from other departments, such as pharmacy, other practicing infectious disease physicians, IT, microbiology, and infection preventionists.

• Do we need to pilot test the new practices?
  • Piloting may give your stewardship team a chance to work out the logistics of the activity, identify unexpected barriers, and develop best practices for monitoring and tracking outcomes.
  • Piloting will allow you to obtain some early data to determine feasibility to continue activities or gain additional support from administration and other stakeholders in your facility.
  • Piloting may help you identify other areas for stewardship activities or opportunities to implement an intervention.

• How do we engage staff in an ASP to reduce C. difficile? How should we work with staff at the unit level? How can we help staff learn new practices?
  • Survey prescribers on their perceptions about issues related to C. difficile, prescribing practices, willingness to comply with stewardship activities, and concerns. This may help you understand some of the barriers you may face in implementation. Tool 1I,Survey of Staff Attitudes Toward ASP and Current Practices, is one of many similar surveys that will help you understand at baseline your prescriber perceptions. This tool can be administered over time to measure changes in perceptions.
  • Plan educational and communication activities to complement your stewardship interventions. This may include efforts to disseminate baseline data on C. difficile and prescribing concerns, along with new policies and prescribing aids (e.g., one-on-one education, lectures, clinical information systems, and hospital intranet pages).
  • Identify "physician and prescribing champions" who will help educate and disseminate information and act as model prescribers from whom others can learn and emulate best practices. These may include hospitalists, chief residents, infectious disease physicians, supervising physician assistants, and other staff. In some facilities, attending physicians and physician assistants are often comfortable prescribing only a few different antibiotics and use them for many indications. Enlisting these different types of prescribing champions may help others in their prescribing.

Return to Questions To Consider

3. How do we monitor the intervention and measure outcomes?

At all points in this process you will need to monitor appropriate processes and outcomes of the stewardship efforts. The frequency that data are collected, compiled, and analyzed will vary by facility. Perhaps your facility has internal committees or boards who require regular submission of data; perhaps your State has a reporting requirement (e.g., C. difficile is reportable in New York State).

It will be important to know the requirements at your facility so that you can make the proper preparations in terms of reporting mechanisms. In deciding how to monitor your intervention and outcomes, you should consider six questions. Some of the tools and resources from earlier sections will help you in this process.

Return to Questions To Consider

3.1. How do we measure rates of C. difficile over time?

Many facilities are already tracking patient cases and rates of C. difficile. If your facility (or State/local health department) does not have a formal system, it will be easier to use or adapt from CDC's NHSN system rather than creating your own definitions and surveillance.

The following are questions to consider:

• Whom might we need help from? You will need assistance from your microbiology laboratory and infection preventionists.
• What is key information we need to be aware of? You must be aware of the methods of C. difficile testing and data collection in your facility. Optimally, your testing methodology and data collection should remain constant throughout the period of your intervention. While it is possible to change to a different type of testing, including more sensitive PCR methods, that will make meaningful comparison before and after your intervention activities more difficult. Also, it is important to know if there are any other changes that may influence C. difficile rates, such as any changes in environmental cleaning or infection control policies and procedures.
• How should we track C. difficile cases? Calculating C. difficile cases per 10,000 patient days will allow comparison both within your facility over time and against national benchmarks. Tracking can be done monthly (or at longer intervals if not feasible) and compared quarterly.
• Is it important to know whether C. difficile cases are acquired in our facility or in the community? NHSN has definitions to assess whether cases are likely acquired in the community or in facilities (see Web page definitions [PDF File; Plugin Software Help]). Because many ASPs and targeted interventions are likely to have effects mainly in the inpatient setting, you may want to compare C. difficile rates for facility-onset cases (as defined by NHSN) before and after your intervention, rather than looking at all C. difficile cases.
• Are we mandated to report C. difficile? Currently, public health laws in New York State (Public Health Law 2819) require hospitals to report several healthcare-associated infections, including C. difficile. New York State uses the NHSN MDRO/C. difficile module that includes use of NHSN platform, definitions, and formatting for submitting required data, as shown in Resource 3A. As of May 2012, five States in addition to New York (CA, IL, OR, TN, UT) mandate public reporting of facilitywide laboratory-identified (LabID) C. difficile events. As a result of the Centers for Medicare & Medicaid (CMS) Inpatient Prospective Payment System rule, starting in 2013, all hospitals will be required to report facilitywide LabID C. difficile events using NHSN. There has also been discussion that in the near future CMS is looking at the wider issue of requiring facilities to have a more formal stewardship activities/program (go to Resource 1A, Policy Statement on Antimicrobial Stewardship by the Society for Healthcare Epidemiology of America, the Infectious Diseases Society of America, and the Pediatric Infectious Diseases Society) .
• Are there Web resources to help with C. difficile surveillance? The following resources may be useful to your facility.

Return to Questions To Consider

3.2. How do we obtain, measure and analyze antibiotic data?

One of the biggest challenges in antimicrobial stewardship in general and specifically in measuring the potential effect of an intervention is obtaining needed data on antibiotic volume for more than one patient at a time. Once the data are obtained, challenges arise in cleaning, aggregating, summarizing, and comparing the data in a meaningful way. It is beyond the scope of this toolkit to address every potential issue, but the following questions may be helpful in guiding your process:

• What are possible antibiotic data sources? The strengths and weaknesses of the four main types of antibiotic data (purchasing, orders, antibiotics dispensed, and antibiotic receipt) are summarized in Resource 2B, A Comparison of Antibiotic Data Sources. If your facility is just starting ASP efforts or has little IT support, purchasing can be both a feasible way to start to identify problem drugs and a quick way to show the effects of early intervention measures.
• What are some common antibiotic data cleaning and aggregation challenges? Getting the right data from even the most sophisticated hospital or pharmacy clinical information system is often an iterative process. Even with good IT support, data may be given to you that are not ready for immediate analysis. We recommend some validation to ensure that the data make sense. Reviews of purchasing data or focused chart reviews are two means to reality check the larger volume of use data. Check the data to verify that all drugs are included. Many facilities use generic drugs; with drug shortages, products may vary throughout the year; and some drugs will need to be removed as they are no longer used for their antimicrobial properties. In multifacility medical centers, it is important to standardize drugs and dosing to aggregate and compare over campuses.
• What are some antibiotic metrics we can consider using? Resource 2C, Possible Methods for Evaluating Antibiotic Use, describes methods for summarizing and comparing antibiotic use. It includes definitions, strength, considerations regarding use, and references to learn more about each measure. Resource 2G, A Review of Antimicrobial Stewardship Interventions With Suggested Process/Monitoring, Antibiotic Use, and Additional Measures of Impact,links some of the common intervention activities and suggested potential antibiotic metrics to access and compare prior to, during, and after the interventions.
  • You will likely need IT support to obtain aggregate data. Often if antibiotic receipt data are obtained, the database can be large and difficult to manage. It also may require cleaning and combining data substantially before any metrics or comparisons can be used.
  • Often using more than one metric will give you a more complete picture of the baseline and postintervention use.
  • Depending on the extent and nature of the intervention activities, it may take time for you to see changes in antibiotic use, depending on the metric chosen for that particular intervention. For example, restrictions likely see effects sooner, while audit and feedback activities or new algorithms may take longer to take hold and change prescribing. Some interventions may be in specific units or populations, so a hospitalwide metric may not be refined enough to show differences.

Return to Questions To Consider

3.3. How can we monitor the intervention and why should we?

It is important to monitor the interventions for multiple reasons. You will need to:

• Be able to verify that the interventions are occurring.
• Look for early and tangible signs of success (antibiotic metrics may be slow to show effects).
• Identify barriers to success and ways to improve the intervention.
• Assess whether additional interventions are needed.
• Determine whether the interventions are sustainable.
• Learn whether the intervention could be an effective way to affect other outcomes, such as C. difficile and antibiotic resistance.

This last point is important because many stewardship interventions have been successful in reducing targeted antibiotic use and cost but have fallen short in reducing more concrete outcomes such as C. difficile rates and antibiotic resistance. If these outcomes are not achieved, monitoring the intervention is important to distinguish why the intervention did not demonstrate an impact. Was it because the activities were not completed; the intervention could not affect prescribing practices, antibiotic use, or outcomes; the activities were not implemented long enough to see impact; transmission was a bigger problem than anticipated; or �the mechanism of resistance presented problems? Resource 2G, A Review of Antimicrobial Stewardship Interventions, summarizes some suggested monitoring or process measures for specific antibiotic stewardship intervention activities.

Return to Questions To Consider

3.4. What other processes do we need to monitor and measure?

It is also important to ensure that your interventions do not have unintended consequences for your patients. Thus, looking at rates of reinstitution of broad antibiotic therapy, missed or delayed antibiotic doses, or readmission (especially for infection) should be part of your ongoing surveillance, along with examination of mortality data. In addition, your interventions may have other positive outcomes (e.g., reduced length of stay for patients receiving the intervention) that you will need to monitor and keep track of.

Further, your institution's infection control, isolation precautions, and environmental cleaning policies and practices should be monitored throughout your intervention as changes in these can also affect C. difficile rates.Resources 3G and 3H offer examples of tools for monitoring prevention practices.

Finally, it is important to assess how the intervention affects your stewardship team and other prescribers. Qualitative data collection (ideally pre- and postintervention) such as prescriber surveys (go to Tool 1L), informal discussions, or focus groups may enrich your understanding of the ASP and activities at your facility.

Return to Questions To Consider

3.5. How do we analyze financial data?

It will be important to conduct a financial impact analysis at your facility to estimate the costs associated with running an ASP, including staffing, software, and equipment; estimate the cost savings from reduced antimicrobial use; and understand the potential reimbursement impact of reducing infections. In addition, you may want to estimate indirect savings, including savings on isolation equipment or estimated savings from meeting external quality measures, such as those from CMS.

You may choose to examine antibiotic purchasing data if those data are available, or you may need to use industry sources to estimate costs associated with antibiotic purchasing. Data from the hospital should be available as they are routinely compiled for internal hospital purposes or required external use (e.g., billing). In the current climate of competing demands, leadership is frequently interested in the ways patient safety initiatives translate into financial savings.

Return to Questions To Consider

3.6. How do we assess the overall impact of our interventions? How do we decide which interventions have been the most successful (and which interventions were not) and why?

Assessing the overall impact of an intervention can be complex. Following are some important points gleaned from the participating facilities and interpretation of the analysis and evaluation components by our ERASE C. difficile leadership team.

Simplistically showing a decrease in the targeted antibiotic consumption might suggest success. One issue that we encountered is that depending on the specific antibiotic, dosing, and type of intervention, the commonly used antibiotic metrics of defined daily dose, days of therapy, and number of courses may not reflect the changes in prescribing exactly the same way.

In addition, a large enough proportion of the prescribing needs to be affected to be measurable in the aggregate antibiotic use measures or to potentially improve other outcomes (e.g., C. difficile infections). For example, in several of the participating facilities, piperacillin/tazobactam was a target. But this is a broad antibiotic an array of prescribers often use for empiric regimens or for various infections. Thus, an intervention that targets only a fraction of that prescribing may be successful in decreasing that specific indication for prescribing but may not be able to be measured when overall prescribing is summarized.

Since C. difficile has multifactorial causes, affecting only one drug or the prescribing of only a proportion of the prescribing of that specific drug may not affect the rates of C. difficile. This is likely also true of antibiotic resistance patterns. In addition, it may take longer to influence C. difficile and other microbiologic targets. Thus, for interventions of short duration, it will be difficult to show changes and you may not be guaranteed to see changes in these measures even in the long run.

Further, some interventions may shift antibiotic prescribing but not reduce use. If the shift is to a more narrow antibiotic, an oral antibiotic regimen, or no antibiotics, the patient may have positive outcomes but experience unintended consequences. For example, if an antibiotic is restricted and prescribers move to an alternative nonrestricted antibiotic, an unintended increase can occur in use of the second antibiotic. Patients may then experience secondary unintended consequences (e.g., have more side effects, be more difficult to dose). The drug also may have other effects on the local microbiologic flora, be a mismatch to the patient's culture results, and result in readmission or increased length of stay. Thus, a shift in prescribing can be good and bad, so reviewing the patterns of antibiotic prescribing in general and not just of single agents is important.

Other potential surrogates or outcomes for review are length of stay and cost. A decrease in the length of treatment may allow more prompt discharge of patients, which should be balanced with a review of readmissions for infections. An assessment of cost to implement the intervention and potential cost savings from fewer or different patterns of antibiotic use can be useful, especially when advocating for more formal resources from hospital administration.

Several other factors are important to consider when assessing the impact of an intervention. It is important to explore the ability to implement the intervention in terms of cost, staffing, IT support, and timeframe. You also need to assess the acceptability to the stewardship team (does not impede other needed stewardship activities) and to prescribers. Not all activities will be welcomed, but some may be less popular than others. This can make it difficult to cast the ASP team in an educational role with prescribers, making it more difficult to implement other needed changes.

All these factors can also affect the ability to sustain activities. Thus, activities have to be considered in the setting of longer range goals for the ASP.

Ultimately, your facility will probably weight a combination of these factors when you choose to implement new or supplement existing antibiotic stewardship activities. Interventions need to be tailored and must complement other factors in your facility's environment.

Return to Questions To Consider

4. How do we sustain the ASP for reducing C. difficile over time?

Early in the process of developing your ASP aimed at reducing C. difficile, you will need to think ahead about sustaining the program once it is in place. Often, sustaining changes in clinical practice introduced through a new initiative is more difficult than implementing them initially. To maintain your ASP for C. difficile reduction so that it thrives and continues to be useful, consider at least four questions; you may have others.

In considering these questions, reflect on the challenges you currently face and consider strategies for working on them in the future. Tool 4Aprovides a worksheet for doing that. The four questions are discussed below. The tools and processes introduced in earlier sections, such as the business plan Tool 1K, also can be used in developing your sustainability plan.

Will our current ASP staffing work on an ongoing basis?

How well is your stewardship team working across departments and disciplines? What is the current distribution of responsibilities? Are there changes that you would need to make?

What is the plan for ongoing measurement and feedback?

A plan is only as good as the systems in place that ensure sustainability. It is crucial to plan at the onset for ongoing monitoring, maintenance, and evaluation of the ASP for reducing C. difficile. Section 3 talked in detail about monitoring the ASP for C. difficile intervention and outcomes as you are planning and initially implementing the program. Measurement and feedback will be equally important as you transition to sustaining the program.

What ongoing organizational support will we need to keep the new ASP practices in place?

Maintaining your ASP requires organizational support on multiple levels. Ongoing organizational support for ASP to reduce C. difficile will be strongest if you can demonstrate that it is aligned with the medical center's strategic priorities and that it addresses pressing problems.

How do we keep the ASP efforts relevant and a continued focus?

As discussed in greater detail in Section 1.2, given the environment of tightening resources in most medical centers, you will likely need to keep your preliminary business case updated and current to ensure continued or expanded investment in your ASP.

Return to Questions To Consider
Return to Contents
Proceed to Next Section
Proceed to Summary of Tools and Resources