Use of Prospective Audit and Feedback to Reduce Antibiotic Exposure in a Pediatric Cardiac ICU.

OBJECTIVES: We sought to determine whether a prospective audit and feedback intervention decreased antibiotic utilization in a pediatric cardiac ICU and to describe the characteristics of prospective audit and feedback audits and recommendations. DESIGN: Before-after study. SETTING: Pediatric cardiac ICU of a freestanding children's hospital. PATIENTS: All patients admitted to the cardiac ICU. INTERVENTIONS: A prospective audit and feedback program was established in our hospital's pediatric cardiac ICU on December 7, 2015. The antimicrobial stewardship program audited IV antibiotics, communicated prospective audit and feedback recommendations to the cardiac ICU, and regularly reviewed recommendation adherence. Mean monthly antibiotic utilization 18 months before ("preprospective audit and feedback"; from June 1, 2014 to November 30, 2015) and 24 months after ("prospective audit and feedback"; from January 1, 2016 to December 31, 2017) prospective audit and feedback implementation was compared. Antibiotic audit data during the prospective audit and feedback period were reviewed to capture the characteristics of prospective audit and feedback audits, recommendations, and adherence. MEASUREMENTS AND MAIN RESULTS: Mean cardiac ICU IV antibiotic use decreased 20% (701 vs 880 days of therapy per 1,000 patient days, p = 0.001) during the prospective audit and feedback period compared with the preprospective audit and feedback period. There was no difference in mean cardiac ICU length of stay (p = 0.573), mean hospital length of stay (p = 0.722), or the rate of discharge due to death (p = 0.541). There were 988 antibiotic audits and 370 prospective audit and feedback recommendations (37% recommendation rate) during the study period. The most commonly audited antibiotic category was broad-spectrum gram-negative agents and the most common indication for use was sepsis. Broad-spectrum gram-positive agents were more likely to be associated with a recommendation. CONCLUSIONS: There was a significant reduction in antibiotic use following implementation of a prospective audit and feedback program in our pediatric cardiac ICU. Over one-third of antibiotics audited in our cardiac ICU were associated with a prospective audit and feedback recommendation, revealing important targets for future antimicrobial stewardship efforts in this population.

A Guided Online and Mobile Self-Help Program for Individuals With Eating Disorders: An Iterative Engagement and Usability Study.

BACKGROUND: Numerous digital health interventions have been developed for mental health promotion and intervention, including eating disorders. Efficacy of many interventions has been evaluated, yet knowledge about reasons for dropout and poor adherence is scarce. Most digital health intervention studies lack appropriate research design and methods to investigate individual engagement issues. User engagement and program usability are inextricably linked, making usability studies vital in understanding and improving engagement. OBJECTIVE: The aim of this study was to explore engagement and corresponding usability issues of the Healthy Body Image Program-a guided online intervention for individuals with body image concerns or eating disorders. The secondary aim was to demonstrate the value of usability research in order to investigate engagement. METHODS: We conducted an iterative usability study based on a mixed-methods approach, combining cognitive and semistructured interviews as well as questionnaires, prior to program launch. Two separate rounds of usability studies were completed, testing a total of 9 potential users. Thematic analysis and descriptive statistics were used to analyze the think-aloud tasks, interviews, and questionnaires. RESULTS: Participants were satisfied with the overall usability of the program. The average usability score was 77.5/100 for the first test round and improved to 83.1/100 after applying modifications for the second iteration. The analysis of the qualitative data revealed five central themes: layout, navigation, content, support, and engagement conditions. The first three themes highlight usability aspects of the program, while the latter two highlight engagement issues. An easy-to-use format, clear wording, the nature of guidance, and opportunity for interactivity were important issues related to usability. The coach support, time investment, and severity of users' symptoms, the program's features and effectiveness, trust, anonymity, and affordability were relevant to engagement. CONCLUSIONS: This study identified salient usability and engagement features associated with participant motivation to use the Healthy Body Image Program and ultimately helped improve the program prior to its implementation. This research demonstrates that improvements in usability and engagement can be achieved by testing and adjusting intervention design and content prior to program launch. The results are consistent with related research and reinforce the need for further research to identify usage patterns and effective means for reducing dropout. Digital health research should include usability studies prior to efficacy trials to help create more user-friendly programs that have a higher likelihood of "real-world" adoption.

Performance of a Commonly Used Pressure Injury Risk Model Under Changing Incidence.

BACKGROUND: Hospital-acquired pressure injuries (HAPIs) cause patient harm and increase health care costs. We sought to evaluate the performance of the Braden QD Scale-associated changes in HAPI incidence. METHODS: Using electronic health records data from a quaternary children's hospital, we evaluated the association between Braden QD scores and patient risk of HAPI. We analyzed how this relationship changed during a hospitalwide quality HAPI reduction initiative. RESULTS: Of 23,532 unique patients, 108 (0.46%, 95% confidence interval [CI] = 0.38%-0.55%) experienced a HAPI. Every 1-point increase in the Braden QD score was associated with a 41% increase in the patient's odds of developing a HAPI (odds ratio [OR] = 1.41, 95% CI = 1.36-1.46, p < 0.001). HAPI incidence declined significantly following implementation of a HAPI-reduction initiative (ß = -0.09, 95% CI = -0.11 - -0.07, p < 0.001), as did Braden QD positive predictive value (ß = -0.29, 95% CI = -0.44 - -0.14, p < 0.001) and specificity (ß = -0.28, 95% CI = -0.43 - -0.14, p < 0.001), while sensitivity (ß = 0.93, 95% CI = 0.30-1.75, p = 0.01) and the concordance statistic (ß = 0.18, 95% CI = 0.15-0.21, p < 0.001) increased significantly. CONCLUSION: Decreases in HAPI incidence following a quality improvement initiative were associated with (1) significant deterioration in threshold-dependent performance measures such as specificity and precision and (2) significant improvements in threshold-independent performance measures such as the concordance statistic. The performance of the Braden QD Scale is more stable as a tool that continuously measures risk than as a prediction tool.

Clinical Impact of Clostridium difficile PCR Cycle Threshold-Predicted Toxin Reporting in Pediatric Patients.

BACKGROUND: Reliance on tests that detect only the presence of toxigenic Clostridium difficile can result in overdiagnosis and overtreatment of C difficile infection (CDI). The C difficile polymerase chain reaction (PCR) cycle threshold (CT) can sensitively predict the presence of free C difficile toxins; however, the clinical application for this testing strategy remains unexplored. We evaluated the impact of dual PCR and toxin result reporting, as predicted by the CT, on CDI management and outcomes in children. METHODS: Before the intervention, results for C difficile testing at Lucile Packard Children's Hospital Stanford were reported as PCR positive (PCR+) or negative (PCR-) according to the GeneXpert C diff Epi tcdB PCR assay (Cepheid, Sunnyvale, California). Beginning October 5, 2016, the presence of free toxins, as predicted by the CT, was reported also. The CDI treatment rates 1 year before and 18 months after implementation of toxin reporting were compared. Demographic and treatment-related data were collected, and patient outcomes were followed up 8 weeks later. RESULTS: CDI treatment decreased 22% after the intervention (96% [preintervention] vs 74% [postintervention]; P < .001). During the postintervention period, there were 152 PCR+C difficile results, and 94 (62%) of them were toxin positive (toxin+) according to the CT. Of the 58 PCR+/toxin-negative (toxin-) results, 38 (66%) did not result in CDI treatment. Seven (18%) of the untreated PCR+/toxin- patients underwent repeat testing within 8 weeks, and 5 (13%) of them were subsequently PCR+/toxin+ and treated. No CDI-related complications were identified. CONCLUSIONS: Addition of the CT-predicted C difficile toxin result to PCR reporting reduces the proportion of PCR+ children treated for CDI.

Inpatient Observation After Transition From Intravenous to Oral Antibiotics.

OBJECTIVES: Children hospitalized with infections are commonly transitioned from intravenous (IV) to enteral (per os [PO]) antibiotics before discharge, after which they may be observed in the hospital to ensure tolerance of PO therapy and continued clinical improvement. We sought to describe the frequency and predictors of in-hospital observation after transition from IV to PO antibiotics in children admitted for skin and soft tissue infections (SSTIs). METHODS: We conducted a retrospective cohort study of children with SSTIs discharged between January 1, 2016, and June 30, 2018, using the Pediatric Health Information System database. Children were classified as observed if hospitalized ≥1 day after transitioning from IV to PO antibiotics. We calculated the proportion of observed patients and used logistic regression with random intercepts to identify predictors of in-hospital observation. RESULTS: Overall, 15% (558 of 3704) of hospitalizations for SSTIs included observation for ≥1 hospital day after the transition from IV to PO antibiotics. The proportion of children observed differed significantly between hospitals (range of 4%-27%; P < .001). Observation after transition to PO antibiotics was less common in older children (adjusted odds ratio [aOR] = 0.69; 95% confidence interval [CI] 0.52-0.90; P = .045). Children initially prescribed vancomycin (aOR = 1.36; 95% CI 1.03-1.79; P = .032) or with infections located on the neck (aOR = 1.72; 95% CI 1.32-2.24; P < .001) were more likely to be observed. CONCLUSIONS: Children hospitalized for SSTIs are frequently observed after transitioning from IV to PO antibiotics, and there is substantial variability in the observation rate between hospitals. Specific factors predict in-hospital observation and should be investigated as part of future studies aimed at improving the care of children hospitalized with SSTIs.

Key Drivers in Reducing Hospital-acquired Pressure Injury at a Quaternary Children's Hospital.

INTRODUCTION: Despite being a participating Solutions for Patient Safety (SPS) children's hospital and having attempted implementation of the SPS hospital-acquired pressure injuries (HAPIs) prevention bundle, our hospital remained at a HAPI rate that was 3 times the mean for SPS participating children's hospitals. This performance led to the launch of an enterprise-wide HAPI reduction initiative in our organization. The purpose of this article is to describe the improvement initiative, the key drivers, and the resulting decrease in the SPS-reportable HAPI rate. METHODS: We designed a hospital-wide HAPI reduction initiative with actions grouped into 3 key driver areas: standardization, data transparency, and accountability. We paused all individual hospital unit-based HAPI reduction initiatives. We calculated the rate of SPS-reportable HAPIs per 1,000 patient days during both the pre- and postimplementation phases and compared mean rates using a 2-sided t test assuming unequal variances. RESULTS: The mean SPS-reportable HAPI rate for the preimplementation phase was 0.3489, and the postimplementation phase was 0.0609. The difference in rates was statistically significant (P < 0.00032). This result equates to an 82.5% reduction in HAPI rate. CONCLUSIONS: Having an institutional pause and retooled initiative to reduce HAPI with key drivers in the areas of standardization, data transparency, and accountability had a statistically significant reduction in our organization's SPS-reportable HAPI rate.

Predictors of Antimicrobial Stewardship Program Recommendation Disagreement.

OBJECTIVETo identify predictors of disagreement with antimicrobial stewardship prospective audit and feedback recommendations (PAFR) at a free-standing children's hospital.DESIGNRetrospective cohort study of audits performed during the antimicrobial stewardship program (ASP) from March 30, 2015, to April 17, 2017.METHODSThe ASP included audits of antimicrobial use and communicated PAFR to the care team, with follow-up on adherence to recommendations. The primary outcome was disagreement with PAFR. Potential predictors for disagreement, including patient-level, antimicrobial, programmatic, and provider-level factors, were assessed using bivariate and multivariate logistic regression models.RESULTSIn total, 4,727 antimicrobial audits were performed during the study period; 1,323 PAFR (28%) and 187 recommendations (15%) were not followed due to disagreement. Providers were more likely to disagree with PAFR when the patient had a gastrointestinal infection (odds ratio [OR], 5.50; 95% confidence interval [CI], 1.99-15.21), febrile neutropenia (OR, 6.14; 95% CI, 2.08-18.12), skin or soft-tissue infections (OR, 6.16; 95% CI, 1.92-19.77), or had been admitted for 31-90 days at the time of the audit (OR, 2.08; 95% CI, 1.36-3.18). The longer the duration since the attending provider had been trained (ie, the more years of experience), the more likely they were to disagree with PAFR recommendations (OR, 1.02; 95% CI, 1.01-1.04).CONCLUSIONSEvaluation of our program confirmed patient-level predictors of PAFR disagreement and identified additional programmatic and provider-level factors, including years of attending experience. Stewardship interventions focused on specific diagnoses and antimicrobials are unlikely to result in programmatic success unless these factors are also addressed.Infect Control Hosp Epidemiol 2018;806-813.

Displaying radiation exposure and cost information at order entry for outpatient diagnostic imaging: a strategy to inform clinician ordering.

BACKGROUND: Displaying radiation exposure and cost information at electronic order entry may encourage clinicians to consider the value of diagnostic imaging. METHODS: An urban safety-net health system displayed radiation exposure information for CT and cost information for CT, MRI and ultrasound on an electronic referral system for outpatient ordering. We assessed whether there were differences in numbers of outpatient CT scans and MRIs per month relative to ultrasounds before and after the intervention, and evaluated primary care clinicians' responses to the intervention. RESULTS: There were 23 171 outpatient CTs, 15 052 MRIs and 43 266 ultrasounds from 2011 to 2014. The ratio of CTs to ultrasounds decreased by 15% (95% CI 9% to 21%), from 58.2 to 49.6 CTs per 100 ultrasounds; the ratio of MRIs to ultrasounds declined by 13% (95% CI 7% to 19%), from 37.5 to 32.5 per 100. Of 300 invited, 190 (63%) completed the web-based survey in 17 clinics. 154 (81%) noticed the radiation exposure information and 158 (83.2%) noticed the cost information. Clinicians believed radiation exposure information was more influential than cost information: when unsure clinically about ordering a test (radiation=69.7%; cost=46.4%), when a patient wanted a test not clinically indicated (radiation=77.5%; cost=54.8%), when they had a choice between imaging modalities (radiation=77.9%; cost=66.6%), in patient care discussions (radiation=71.9%; cost=43.2%) and in trainee discussions (radiation=56.5%; cost=53.7%). Resident physicians and nurse practitioners were more likely to report that the cost information influenced them (p<0.05). CONCLUSIONS: Displaying radiation exposure and cost information at order entry may improve clinician awareness about diagnostic imaging safety risks and costs. More clinicians reported the radiation information influenced their clinical practice.

From admission to discharge: patterns of interpreter use among resident physicians caring for hospitalized patients with limited english proficiency.

Resident physicians' use of professional interpreters drives communication with hospitalized patients with limited English proficiency (LEP). We surveyed residents from three specialties across two hospitals affiliated with one academic medical institution about their communication with their last hospitalized LEP patient. Among 149 respondents (73% response rate), 71% reported using professional interpreters for fewer than 60% of hospital encounters. Most (91%) perceived their quality of communication with hospitalized LEP patients as worse than with English-speaking patients. Professional interpreter use varied substantially by resident and by hospital encounter, with more reporting use of ad hoc interpreters, their own language skills, or not talking to the patient due to time constraints during pre-rounds (39%), team rounds (49%), or check-ins (40%) than during procedural consents (9%) or family meetings (17%). The reported variation suggests targets for quality improvement efforts and the need for clear enforceable guidelines on resident communication with hospitalized LEP patients.

Clinicians' views on displaying cost information to increase clinician cost-consciousness.

OBJECTIVES: To evaluate 1) clinician attitudes towards incorporating cost information into decision making when ordering imaging studies; and 2) clinician reactions to the display of Medicare reimbursement information for imaging studies at clinician electronic order entry. STUDY DESIGN: Focus group study with inductive thematic analysis. METHODS: We conducted focus groups of primary care clinicians and subspecialty physicians (nephrology, pulmonary, and neurology) (N = 50) who deliver outpatient care in 12 hospital-based clinics and community health centers in an urban safety net health system. We analyzed focus group transcripts using an inductive framework to identify emergent themes and illustrative quotations. RESULTS: Clinicians believed that their knowledge of healthcare costs was low and wanted access to relevant cost information for reference. However, many clinicians believed it was inappropriate and unethical to consider costs in individual patient care decisions. Among clinicians' negative reactions toward displaying costs at order entry, 4 underlying themes emerged: 1) belief that ordering is already limited to clinically necessary tests; 2) importance of prioritizing responsibility to patients above that to the healthcare system; 3) concern about worsening healthcare disparities; and 4) perceived lack of accountability for healthcare costs in the system. CONCLUSIONS: Although clinicians want relevant cost information, many voiced concerns about displaying cost information at clinician order entry in safety net health systems. Alternative approaches to increasing cost-consciousness may be more acceptable to clinicians.

Clinician perspectives on considering radiation exposure to patients when ordering imaging tests: a qualitative study.

BACKGROUND AND OBJECTIVES: Increased computer tomography (CT) scan use has contributed to a rise in medically-associated radiation exposure. The extent to which clinicians consider radiation exposure when ordering imaging tests is unknown. We examined (1) outpatient clinician attitudes towards considering radiation exposure when ordering CT scans; and (2) clinician reactions to displaying radiation exposure information for CT scans at clinician electronic order entry. METHODS: We conducted nine focus groups with primary care clinicians and subspecialty physicians (nephrology, pulmonary and neurology) (n=50) who deliver outpatient care across 12 hospital-based clinics and community health centres in an urban safety-net health system, which use a common electronic order entry system. We analysed focus group transcripts using an inductive framework to identify emergent themes and illustrative quotations. FINDINGS: Clinicians felt they had limited knowledge of the clinical implications of radiation exposure. Many believed clinically relevant information such as the increased risk of malignancy from CT scans would be useful to inform decision-making and patient-clinician discussions. Clinicians noted that patient vulnerability and long wait times for tests with less radiation exposure (such as MRI or ultrasound) often acted as barriers to minimise patient radiation exposure from CT scans. Clinicians suggested providing patients' cumulative radiation exposure or formal decision aids to improve the usefulness of the radiation exposure information. CONCLUSIONS: Displaying clinically relevant radiation exposure information at order entry may improve clinician knowledge and inform patient-clinician discussions regarding risks and benefits of imaging. However, limited access to tests with lower radiation exposure in safety-net settings may trump efforts to minimise patient radiation exposure.