DISCUSSION

This report explores the experiences of antimicrobial stewardship clinicians as they try to optimize antimicrobial use in pediatric oncology and BMT patients. We aimed to describe current stewardship activities and to identify the goals and barriers experienced by antimicrobial stewardship clinicians working in this field and determine whether these were linked to clinician or institutional characteristics.

We found that stewardship clinicians are actively working to improve antimicrobial use in the treatment of pediatric oncology and BMT patients in North America and Australasia. Despite significant differences between practitioners and institutions, these clinicians report very similar goals and strategies; most prioritize the goals of reduced initiation or expedited cessation of unnecessary antimicrobials, and most use the commonly recommended approaches of guideline development, prospective audit with feedback, clinician education, and formulary restriction to achieve those goals.Reference Abbo and Ariza-Heredia ^{1 ,} Reference Tverdek, Rolston and Chemaly ^{2 ,} Reference Newland, Gerber and Weissman ^{7 –} Reference Mihu, Paskovaty and Seo ⁹ Barriers to effective stewardship are also strikingly homogeneous, with resource limitations and differences in risk perception or motivation between the ASP and the oncology teams appearing most frequently.

Use of audit with delayed feedback is an unexpected finding. This differs from the usual method of prospective audit by delivering feedback in batches (eg, a monthly summary of antimicrobial usage by individual clinicians or units). In contrast to prospective audit, it is not well described outside of the ambulatory setting. It may be less resource intensive and should be further explored.

Reported stewardship goals may be important for determining ways to assist ASP efforts or conducting collaborative research and quality-improvement programs. Because many goals were shared by most respondents, effective strategies addressing these might help improve stewardship at most institutions.

Only 2 goals were significantly correlated with clinician or institution characteristics. Australasian clinicians appear to view prevention of adverse effects as more central to their role than those in North America, perhaps because of local cultural expectations. Similarly, clinicians spending more time attending to stewardship apparently focus on reducing overall antibiotic use, leading to a surprising inverse association between ASP work effort and the goal of decreasing time to effective therapy in resistant infections. Future studies could explore the extent to which the proportion of work effort clinicians dedicate to stewardship and local cultural expectations influence decision-making within ASPs or affect ASP effectiveness in other settings.

The study also identified important barriers faced by stewardship clinicians in the pediatric oncology population. Reported barriers clustered tightly according to identifiable themes. The importance of understanding and addressing barriers to stewardship implementation is supported by evidence that adherence to ASP recommendations in the treatment of patients with cancer can be relatively poor and that adherence improves patient outcomes.Reference Rosa, Goldani and dos Santos ^{6 ,} Reference Yeo, Wu, Chung, Chan, Fisher and Hsu ¹⁰ Future prospective collaborative research that focuses on addressing these barriers is likely to be attractive to ASP clinicians and could identify ways to improve important institutional and patient outcomes. Previous publications have listed possible challenges to effective ASP in this population,Reference Abbo and Ariza-Heredia ^{1 ,} Reference Tverdek, Rolston and Chemaly ^{2 ,} Reference Mihu, Paskovaty and Seo ⁹ but to our knowledge this study is the first to investigate the relative importance of specific barriers to ASP clinicians.

Many clinicians reported that collaborative group treatment protocols interfere with stewardship. This could be addressed by ensuring that oncology research protocols allow flexibility for stewardship, provide evidence for antimicrobial recommendations, or include research aims to help support future recommendations. Indeed, these networks might provide opportunities for stewardship or quality-improvement projects.

The formal response rate for the survey was relatively low (38%) and was higher in North America (~50%) than Australasia (~17%). However, it should be noted that the survey link was distributed to a large group to maximize the number of eligible participants, and that the listservs used to contact potential recipients include both adult clinicians and pediatric clinicians not involved in antimicrobial stewardship. Some recipients were therefore ineligible to complete the survey and, if identifiable, would not be included in the denominator for calculation of the formal response rate. Conversely, because the survey was distributed by email, recipients could pass the link to other clinicians. The number of evaluable responses from Australasia was small (n=18) but is proportional to the size of the clinician and patient population in that region (where there are only 8 free-standing pediatric hospitals).

Additionally, the data were analyzed at the level of the individual respondent, and some institutions were represented by multiple respondents; respondent institution was not included in the regression analysis because the number of institutions was large compared with the sample size. Similarly, grouping different countries by geographic region may mask differences between those countries. Infectious diseases fellows were grouped with attending physicians because their responses to all survey components were similar, but this may not be appropriate in other studies. Lastly, the study describes the perceptions of antimicrobial stewardship clinicians, rather than objective facts about beliefs or behavior of others.

Antimicrobial stewardship clinicians are actively working to optimize antimicrobial use in the pediatric oncology patient population. Regardless of individual or institutional characteristics and geographic boundaries, ASP clinicians report similar goals and strategies. Important barriers to stewardship in this setting include personnel and data-analysis limitations, challenging relationships, and differences in priorities and risk assessment between stewardship and oncology clinicians. Further research and quality improvement efforts directed towards these shared goals and barriers may help improve antimicrobial use in this uniquely vulnerable population.