While USP Chapter <797> has been official since June 2008, a proposed revision to the chapter was released for public comment in September of 2015. Since that time, the Compounding Expert Committee has been reviewing more than 8000 comments received from over 2500 stakeholders. Upon completion of this review, USP will publish a revision to the proposed chapter for a second public comment period; however, a date for that event has yet to be released. It is unlikely that the final revision will become official until well into 2019, perhaps becoming official in concert with the December 1, 2019 official date for USP Chapter <800>.
Judging from the multitude of questions we receive from hospital pharmacists about the proposed chapter, we are concerned that there has been a loss of focus on compliance with Chapter <797> in its current, official form. The 2017 USP Compounding Compliance Study, CriticalPoint’s 7th annual national survey of sterile compounding practices, explored this issue as part of its analysis of sterile compounding compliance rates. This concern is reinforced by the fact that overall compliance rates in this year’s survey vary little from the compliance rates over the past 2 years (see FIGURE 1). Given the recent sterile compounding tragedies and ongoing FDA inspections of hospital pharmacies, pharmacy is not in a position to settle for anything less than full compliance. Getting close is not good enough.
The 2017 USP Compounding Compliance Study saw 1057 locations register to complete the <797> and/or <800> gap analyses. As in previous years, CriticalPoint limited reportable respondents to locations that completed at least 90% of the required survey questions. This adjustment resulted in a final sample of 662 locations, which represents roughly 50% fewer respondents than in 2016 when 1185 locations were represented after the same adjustment. The survey results demonstrate that overall sterile compounding practices have improved; nonetheless, few facilities have achieved full compliance. We assume that the people who participate in this study are among the most engaged and motivated to improve sterile compounding practice. What about all the others? There are certainly some shining examples of excellence, but those hospitals continue to be too few and far between.
In failing to achieve full compliance with <797>’s minimum practice requirements, pharmacy leaves the door open for the FDA to insist on having more authority over pharmacy sterile compounding practice. If pharmacy is to maintain control over this area of its practice, then pharmacy must pursue full compliance with <797>, while also moving aggressively to implement documented best practice. In addition, state boards of pharmacy and boards of health must be more aggressive in holding their licensees accountable to compliance with <797> where applicable. States that currently do not require compliance with these minimal standards need to move forward. This is a practice imperative.
We ask pharmacy to renew its focus on the current chapter. Without complete compliance with the current chapter, organizations will not be able to move quickly to meet the future requirements, which are likely to be more specific and more rigorous.
Having been in effect for 9 years, and given that the current chapter establishes minimum practice requirements, rather than best practices, higher overall compliance was expected by this time. USP <797> is a comprehensive quality management system wherein each component works in concert with the other components to construct a system of quality assurance for pharmacy’s compounded sterile preparations (CSPs). Consider, for example, that the storage limit and beyond-use date (BUD) requirements were developed with the assumption that they would be used in organizations that are in compliance with all of the elements of the chapter. Unfortunately, that is not always the case.
As in previous years, the findings published herein relate to hospital practice only since hospitals represent over 80% of the respondents in the study. We are providing more data than in past years, using Tableau, a powerful analytics tool. One of the new features in this year’s report are state-specific compliance maps (see FIGURE 2); additional compliance maps are displayed elsewhere in this publication. Our hope is that these data may guide state boards of pharmacy in assessing compliance with relevant sterile compounding regulation in their states.
General Quality Management
There are many areas where compliance rates remain stagnant. For example, after the High-Risk Compounding domains, the General Quality Management domain had the lowest overall compliance rate at 66%. This number shows no meaningful change from previous years, wherein compliance rates have ranged from a low of 62% (2011) to a high of 68% (2016). TABLE 1 depicts the specific items included in this subject domain.
The practices most responsible for lowering the overall compliance score all relate to quality assurance activities. This begs the question: Are we performing monitoring simply because it is required without thoughtfully considering what the results tell us? Likewise, if your facility’s sampling never exceeds established action levels, there may be a larger problem. Should 0 CFU findings be consistently observed week after week, and month after month for surface, air, and gloved fingertip sampling, something is wrong. No operation is perfect! There is likely a problem with the type of media used, sampling technique, incubation process, or method of reading samples. Is it possible that there is not a consistent process for reacting to action levels because many do not find results that require a reaction?
Certainly, there is a dearth of specific guidance in <797> about how to investigate or remediate action levels that have been exceeded. For more specific guidance on these activities, visit the June 2017 CriticalPoint Sterile Compounding Pearls of Knowledge at www.criticalpoint.info/sterile-compounding-pearls-of-knowledge-june-2017-edition. The chapter requires evidence of a consistent process that occurs anytime a patient outcome is not achieved, an action level is exceeded, or another important operational variance occurs. There also must be appropriate follow up to verify that the actions the organization takes in response to problems prove effective at eliminating the problems.
Please consider your organization’s practice objectively and take any necessary steps to improve. Do not monitor action levels simply because the chapter says to; rather, monitor action levels to ensure that your pharmacy has achieved a state of microbial control. Furthermore, drifts happen to everyone, so it is imperative to monitor frequently enough to detect any drifts away from a state of control early enough to prevent patient harm.
High-Risk Compounding May Go Unrecognized
While the number of hospital pharmacies that perform high-risk (nonsterile to sterile) compounding is small, the magnitude of the risk involved makes this group a key subset worthy of further investigation. Over the past 7 years, our study has demonstrated that 12-15% of sterile compounders overall perform nonsterile to sterile compounding, although it is typically more common among nonhospital compounders. In previous years of this survey, 10% of hospitals reported performing high-risk compounding; this year only about 5% of hospitals are conducting high-risk compounding. This difference may be due to the smaller sample size of this year’s study or perhaps this practice is decreasing in hospitals (a good thing, we think).
Our survey results highlight some serious issues that have the potential to compromise patient safety. It is imperative that improvements occur immediately. Data from this year’s and previous years’ surveys attest to the fact that some hospitals are performing high-risk compounding but do not identify themselves as doing so, and thus, do not perform the activities required of high-risk compounders.
In this year’s survey, only 28 hospitals (5%) identify themselves as performing high-risk compounding. As a double check and to provide gating logic for later items, all respondents were also asked if their location ever uses a 0.22 micron filter to sterilize solutions during compounding processes. The question purposely includes a note to respondents that tells them to answer in the negative if they only use a non-sterilizing filter for other purposes (eg, filtering particulates) to reduce the risk of incorrect positive responses. Ultimately, 55 locations (10%) report using a 0.22 micron filter to sterilize solutions. Despite the fact that 55 locations use a filter to sterilize solutions, and thus, are performing high-risk compounding, only 28 hospitals recognize that they perform high-risk compounding. It is deeply troubling that 27 hospitals do not recognize that they perform high-risk compounding.
High-Risk Compounding Compliance
When it comes to high-risk compounding, hospital compliance rates are significantly lower than they are for non-high-risk compounding practices. Because nonsterile to sterile compounding is high-risk by definition, significant expertise, focus, and consistency of practice is required versus low- and medium-risk compounding. <797> establishes more stringent practices for high-risk compounders for training and competencies, media fill testing, and environmental monitoring. For example, ongoing gloved fingertip sampling (GFS) must be conducted at least every 6 months, viable air sampling must be performed using both general growth media and fungal-specific media, and sterilization procedures must be performed correctly. Among our survey respondents, the only sterilization method reportedly used by hospitals was sterilization via filtration, which requires that a filter integrity test be conducted for each filter used to sterilize solutions. See TABLE 2 for a summary of hospital compliance rates for a variety of high-risk compounding elements.
Bacterial Endotoxin Testing
Survey respondents were asked to indicate which of the conditions listed in FIGURE 3 exist in their pharmacies. Ninety-five (95) locations state that at least one of these conditions exist in their location.
Among the hospitals that are required to perform bacterial endotoxin testing (BET), only 30% report doing so, and only one location reports having specific, written SOPs about BET based on the USP <85> Bacterial Endotoxin Test. The compliance score for hospitals on the endotoxin testing domain has not improved through the years; it has, in fact, decreased (see FIGURE 4).
The survey results for sterility testing compliance are cause for serious concern. Among the survey respondents, 95 locations operate under conditions that require sterility testing to be performed. When these locations were asked if they performed sterility testing, just 20 of 95 (21%) answered in the affirmative (see FIGURE 5). These numbers are consistent with previous years’ surveys, indicating that this is an ongoing problem, not an anomaly.
The majority of the hospitals (94%) that should be performing sterility testing are required to do so because they are extending BUDs. This finding also has remained consistent since 2011. The chapter is clear: If BUDs are extended, regardless of risk level (low, medium, or high), then sterility testing is required for every batch made. As discussed earlier, the storage limits for each risk level are only valid for operations that are fully compliant with the chapter’s quality management system; these data show that hospitals are extending BUDs without performing sterility testing.
Among the hospitals performing sterility testing, 80% state that they are performing it in accordance with the requirements of USP <71>. USP <71> clearly states that if a drug is filterable (and the majority of compounded products are filterable), then filter-based testing is the preferred method and must be employed. Based on observations and reports from practitioners, the investigators are concerned that hospitals may not have an adequate understanding of the requirements of USP <71>. From conversations and experiences with students in the sterility test lab exercise conducted during the CriticalPoint Sterile Compounding Boot Camp class, it is clear that widespread understanding of this chapter is poor. Of further concern is the fact that commercial filtration products are available that do not meet USP <71> requirements. Moreover, direct inoculation products are sold commercially that are not appropriate for use on filterable CSPs. None of the commercial products suggest performing method suitability testing according to USP <71>. This testing ensures that the drug being tested does not interfere with the test and that the test method can identify microbial growth.
Of the 20 hospitals that report performing sterility testing according to USP <71>, 14 (70%) dispense CSPs at risk (ie, before 14 days of sterility testing have been completed). When CSPs are dispensed at risk, <797> states that sterility test samples must be read daily. Of the 14 hospitals that dispense at risk, only half (7 of 14) state that their samples are read daily. We strongly discourage any provider from dispensing at risk, but to dispense CSPs before sterility testing is complete and then not read the samples daily, puts patients at increased risk and is not responsible practice.
Of the 20 hospitals performing sterility testing (remember, far fewer than the 95 that should be performing it), only 10 report having an SOP that requires notification of the prescriber and patient in the event of a sterility failure. If 100% of the hospitals performing sterility testing were not dispensing at risk, this finding might be acceptable. As this is not the case, it is vital that prescribers and patients be notified if they received a CSP that failed a sterility test.
Finally, among these same 20 hospitals, only 9 (45%) state that their SOP specifies actions to take during an investigation should a sterility test fail. This is unacceptable. Sterility testing is required in order to safeguard patients; to have a test fail and not take consistent, disciplined action through an investigation is irresponsible. FIGURE 6 provides a comparison of hospital compliance with sterility testing by select year.
In summary, significant concerns surround the typical hospital’s approach to sterility testing:
- 80% of hospitals that should perform sterility testing do not do so
- Of the 20% that do perform sterility testing, at least one-fifth are not performing it per USP <71>
- 70% of hospitals dispense at risk (ie, before the sterility test is complete)
- Only 50% of those dispensing at risk read sterility tests daily
- Only 50% of those performing sterility testing have created an SOP to notify the doctor and patient if the sterility test fails
- Only 55% of hospitals performing sterility testing have an SOP requiring them to perform a rapid, systematic investigation to identify and correct problems in processes and methods in the event of a sterility test failure
Filter Integrity Testing
<797> requires that filters used for sterilization be tested to ensure that they maintain their integrity after use. Compounding staff error is the overwhelming cause of filter failure. Hurried staff or those unfamiliar with how to use a filter are most likely to cause a filter failure. Because many hospitals that use filters to sterilize solutions perform this type of compounding infrequently, their staff may be less familiar with proper filter use. It is important to recognize that the filters from any pharmacy that does not routinely perform nonsterile to sterile compounding may be more likely to become damaged during compounding.
Among our survey respondents, 55 locations reported using a 0.22 micron filter to sterilize solutions, and thus were asked if they always perform a filter integrity test (ie, bubble point test) after each filter use. Only 14 of the 55 locations (21%) report that they do so. This result is also consistent with previous surveys. A 21% compliance rate for conducting filter integrity testing is alarming. Following the nonsterile to sterile compounding process, correct performance of the filter integrity test is the final safety measure before a patient receives a CSP made from a potentially contaminated active pharmaceutical ingredient (API) that has not been adequately filtered.
Hospitals must carefully consider all sterile compounding practices. If you perform nonsterile to sterile procedures (ie, high-risk compounding), but do not do it often, stop this practice and develop alternatives instead. Performing nonsterile to sterile compounding requires specialized knowledge, environments, practices, and testing. It is not reasonable to expect any pharmacy that does not perform high-risk compounding routinely to have everything in place to do it correctly. Be creative in developing your alternative plan; these products can be outsourced or prescriber habits can be changed.
Please review all the charts, graphs, and figures in this edition. These data demonstrate that many hospitals have yet to fully embrace and implement all the elements of the current USP Chapter <797>. For these facilities, the thoughtful, meaningful, and effective integration of contamination control, personal and environmental sampling, as well as the overwhelmingly important quality assurance follow up, is still a future goal.
Pharmacy is responsible for meeting the minimum standard of practice every day, regardless of whether your state requires compliance with <797>. As licensed professional pharmacists, you are compelled to follow widely accepted best practices, especially where they have been validated by direct or extrapolated evidence. Unlike 9 years ago when the current version of the chapter was published, a multitude of articles, studies, and training resources have been published, and many are free and available in the public sector. If it is a challenge to meet the requirements of <797> now, consider how much more arduous it will be when the FDA visits your facility. If we do not embrace these requirements, pharmacy’s control over compounding may be taken out of our hands.
As Eric Kastango likes to say: We can do it the easy way (<797>), or we can do it the hard way (cGMPs).
Kate Douglass, MS, RN, CRNI, is the vice president of CriticalPoint, LLC, and serves as the codirector of the annual USP <797> Compliance Survey.
Eric S. Kastango, MBA, RPh, FASHP, is the president, CEO, and principal of Clinical IQ, LLC, and CriticalPoint, LLC. He also serves as the co-director for the annual USP <797> Compliance Survey.
Peter Cantor, who serves as the COO and managing partner of CriticalPoint, LLC, is the study coordinator.
Acknowledgment: CriticalPoint gratefully acknowledges the assistance of Nick Kastango whose expertise using Tableau facilitated our ability to obtain and analyze our data in a much more robust and efficient manner.
Address any questions to Kate Douglass at firstname.lastname@example.org