Evaluation of a Comprehensive Algorithm for PICU Patients With New Fever or Instability: Association of Clinical Decision Support With Testing Practices.

OBJECTIVES: Previously, we implemented a comprehensive decision support tool, a "New Fever Algorithm," to support the evaluation of PICU patients with new fever or instability. This tool was associated with a decline in culture rates without safety concerns. We assessed the impact of the algorithm on testing practices by identifying the proportion of cultures pre- vs. post-implementation that were discordant with algorithm guidance and may have been avoidable. DESIGN: Retrospective evaluation 12 months pre- vs. post-quality improvement intervention. SETTING: Single-center academic PICU and pediatric cardiac ICU. SUBJECTS: All admitted patients. INTERVENTIONS: Implementing the "New Fever Algorithm" in July 2020. MEASUREMENTS AND MAIN RESULTS: Patient medical records were reviewed to categorize indications for all blood, respiratory, and urine cultures. Among cultures obtained for new fever or new clinical instability, we assessed specific testing patterns that were discordant from the algorithm's guidance such as blood cultures obtained without documented concern for sepsis without initiation of antibiotics, respiratory cultures without respiratory symptoms, urine cultures without a urinalysis or pyuria, and pan-cultures (concurrent blood, respiratory, and urine cultures). Among 2827 cultures, 1950 (69%) were obtained for new fever or instability. The proportion of peripheral blood cultures obtained without clinical concern for sepsis declined from 18.6% to 10.4% (p < 0.0007). Respiratory cultures without respiratory symptoms declined from 41.5% to 27.4% (p = 0.01). Urine cultures without a urinalysis did not decline (from 27.6% to 25.1%). Urine cultures without pyuria declined from 83.0% to 73.7% (p = 0.04). Pan-cultures declined from 22.4% to 10.6% (p < 0.0001). Overall, algorithm-discordant testing declined from 39% to 30% (p < 0.0001). CONCLUSIONS: The majority of cultures obtained were for new fever or instability and introduction of the "New Fever Algorithm" was associated with reductions in algorithm-discordant testing practices and pan-cultures. There remain opportunities for improvement and additional strategies are warranted to optimize testing practices for in this complex patient population.

A Novel Comprehensive Algorithm for Evaluation of PICU Patients With New Fever or Instability.

OBJECTIVES: There is variation in microbiology testing among PICU patients with fever offering opportunities to reduce avoidable testing and treatment. Our objective is to describe the development and assess the impact of a novel comprehensive testing algorithm to support judicious testing practices and expanded diagnostic differentials for PICU patients with new fever or instability. DESIGN: A mixed-methods quality improvement study. SETTING: Single-center academic PICU and pediatric cardiac ICU. SUBJECTS: Admitted PICU patients and physicians. INTERVENTIONS: A multidisciplinary team developed a clinical decision-support algorithm. MEASUREMENTS AND MAIN RESULTS: We evaluated blood, endotracheal, and urine cultures, urinalyses, and broad-spectrum antibiotic use per 1,000 ICU patient-days using statistical process control charts and incident rate ratios (IRRs) and assessed clinical outcomes 24 months pre- and 18 months postimplementation. We surveyed physicians weekly for 12 months postimplementation. Blood cultures declined by 17% (IRR, 0.83; 95% CI, 0.77-0.89), endotracheal cultures by 26% (IRR, 0.74; 95% CI, 0.63-0.86), and urine cultures by 36% (IRR, 0.64; 95% CI, 0.56-0.73). There was an anticipated rise in urinalysis testing by 23% (IRR, 1.23; 95% CI, 1.14-1.33). Despite higher acuity and fewer brief hospitalizations, mortality, hospital, and PICU readmissions were stable, and PICU length of stay declined. Of the 108 physician surveys, 46 replied (43%), and 39 (85%) recently used the algorithm; 0 reported patient safety concerns, two (4%) provided constructive feedback, and 28 (61%) reported the algorithm improved patient care. CONCLUSIONS: A comprehensive fever algorithm was associated with reductions in blood, endotracheal, and urine cultures and anticipated increase in urinalyses. We detected no patient harm, and physicians reported improved patient care.

Unplanned Extubations in Children: Impact on Hospital Cost and Length of Stay.

OBJECTIVE: To determine the attributable hospital cost, both operational and departmental, and length of stay associated with unplanned extubations in children admitted to PICU and cardiac ICU. DESIGN: Retrospective, matched case-control study. SETTING: Forty-four-bed PICU and 26-bed cardiac ICU in a 303-bed tertiary care pediatric hospital. PATIENTS: Cases with an unplanned extubation were retrospectively identified from July 2011 to March 2013. Controls were PICU and cardiac ICU patients admitted over the same time period and were matched at a ratio of 2:1 for age and diagnosis. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Forty-eight unplanned extubations were analyzed. There were no differences in patient demographics between the two groups, except the control group had a higher severity of illness as illustrated by a larger Paediatric Index of Mortality II Risk of Mortality. Median total hospital costs were higher in those patients with unplanned extubations as compared with controls ($101,310 vs $64,618; p < 0.001). Patients with an unplanned extubation had longer median ICU length of stay (10 d vs 4.5 d; p < 0.001) and hospital length of stay (16.5 d vs 10 d, p < 0.001). CONCLUSION: Pediatric patients with unplanned extubations have an associated increase in hospital costs ($36,692/case) and length of stay (6.5 d/case) as compared with age and diagnosis-matched controls. Further efforts are warranted to establish data-driven benchmarks and establishment of unplanned extubations as a critical metric for ICU quality.

Unplanned admissions to a pediatric cardiac critical care unit: a review of 2 years' experience.

OBJECTIVES: Unplanned admissions to the pediatric cardiac ICU may be a large and high-risk group. Our study describes the frequency of unplanned pediatric cardiac ICU admissions, their admission data, and outcomes. DESIGN: All admissions to a pediatric cardiac ICU over 2 years were reviewed and those that were unplanned were identified for a detailed chart abstraction. Demographic, laboratory, diagnostic, and outcome data were collected. Readmission or admission for adverse event was noted. SETTING: Single, tertiary, pediatric cardiac center. PATIENTS: All patients admitted unexpectedly to the pediatric cardiac critical care unit between May 2008 and May 2010. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: There were 1,203 admissions to the cardiac ICU, and 426 (35%) were unplanned. The most common reasons for admission were new heart disease (25%), infection (19%), arrhythmia (13%), and congestive heart failure (11%). The majority of unplanned admissions (62%) occurred at night. Shock was present at admission in 18.5% of patients. Structural heart disease was present in 79%, and 39% of those were patients with single ventricle. Overall mortality among unplanned admissions was 7.3%, which is higher than that reported for elective surgical admissions. Mortality for the subset of patients readmitted within 30 days was 5.5%. Mean creatinine at admission was higher among nonsurvivors (0.7) than survivors (0.5). CONCLUSIONS: Unplanned admissions accounted for over one third of all admissions and had a high mortality rate. The majority of these occur at night, which may affect staffing models. Acute deterioration leading to unplanned admission, rather than readmission status, may be the driving factor in increased mortality. However, the risk of readmission, lower renal function, or other indices may identify patients at higher risk of an unplanned admission. Continued efforts to identify patients at risk for unplanned admissions are warranted given the outcomes in this cohort.

Effect of a Real-Time Pediatric ICU Safety Bundle Dashboard on Quality Improvement Measures.

BACKGROUND: Patient daily goal sheets have been shown to improve compliance with hospital policies but might not represent the dynamic nature of care delivery in the pediatric ICU (PICU) setting. A study was conducted at Children's National Health System (Washington, DC) to determine the effect of a visible, unitwide, real-time dashboard on timeliness of compliance with quality and safety measures. METHODS: An automated electronic health record (EHR)- querying tool was created to assess compliance with a PICU Safety Bundle. Querying of the EHR for compliance and updating of the dashboard automatically occurred every five minutes. A real-time visual display showed data on presence of consent for treatment, restraint orders, presence of urinary catheters, deep venous thrombosis (DVT) prophylaxis, Braden Q score, and medication reconciliation. Baseline compliance and duration of noncompliance was established during three time periods: the first, before activation of the dashboard; the second, at one month following activation of the dashboard; and the third, at three months after activation. RESULTS: A total of 450 patients were included in the analysis. Between the first and third time periods, the median time from PICU admission to obtaining treatment consent decreased by 49%, from 393 to 202 minutes (p=.05). The number of patients with urinary catheters in place>96 hours decreased from 16 (32%) in Period 1 to 11 (19%) for Periods 2 and 3 combined (p=.01). Completion of medication reconciliation improved from 80% in the first time period to 93% and 92%, respectively, in the subsequent two periods (p=.002). There was no difference between the three periods in presence of restraint orders, DVT prophylaxis, or development or worsening of pressure ulcers. CONCLUSIONS: A unitwide dashboard can increase awareness for potential interventions, affecting patient safety in the PICU in a dynamic manner.

Monitoring the harm associated with use of anticoagulants in pediatric populations through trigger-based automated adverse-event detection.

BACKGROUND: The safety profile of anticoagulants, which are being used with increasing frequency in pediatric populations, is not well studied. Automatic triggers built into electronic health record systems (EHR) have been shown to be an effective way to monitor for and identify medication errors. Anticoagulant-associated adverse events were examined through the use of an anticoagulant trigger panel. METHODS: In a retrospective, five-year (September 2007-September 2012) observational study, four automated triggers were used to detect anticoagulant-related adverse events: activated partial thromboplastin time (aPTT) > 100 seconds in patients on an unfractionated heparin (UFH) infusion, International Normalized Ratio (INR) > 4, anti-factor Xa (anti-FXa) >1.5U/mL for patients on enoxaparin, and the documented use of protamine. RESULTS: For the 1,664 triggers evaluated, 12 were associated with the aPTT trigger, only 1 of which was preventable. Receiver operator characteristic curve analysis indicated that increasing the aPTT trigger > 140 seconds would optimize sensitivity and specificity. The INR trigger identified four outpatients with adverse events. No adverse events were associated with the anti-FXa trigger. The protamine trigger identified 12 adverse events and was associated with more severe events. Minimal overlap was found with protamine and aPTT triggers. CONCLUSION: Laboratory- and medication-based triggers can be effective monitoring tools for anticoagulants. For patients receiving a UFH infusion, an aPTT cutoff value of > 140 seconds is more precise. We also found that protamine use as a trigger adds value to a trigger-based anticoagulant monitoring system. Continued improvement in the logic algorithms associated with the EHR-based trigger tool will allow expanded use of this tool in a clinical manner.

Patient Harm Events and Associated Cost Outcomes Reported to a Patient Safety Organization.

OBJECTIVES: The aims of the study were to describe inpatient harm events detected via an automatic electronic trigger system (ETS) and to consider their financial consequences. METHODS: Over a 27-month period, inpatient harm events were identified and documented in 1 healthcare system with 37 acute care facilities. Patients who experienced harm (all harm or preventable harm only) were compared with similar patients who did not. Clinical, financial, and demographic data were used to identify labor-adjusted direct variable costs (DVC) and potential differences in length of stay (LOS) associated with all-harm, preventable-harm, and nonharmed cohorts. Age-adjusted Charlson Comorbidity Index, case mix index, diagnosis-related groups, major diagnostic category, sex, age, location, diagnosis, adverse event category and subcategory, preventability, and harm severity were used to compare cohorts. Total harm events reported via the ETS and the health system's voluntary event reporting system were compared. RESULTS: Nearly 93,000 encounters for all-harm (n = 25,665) and nonharmed cohorts (n = 67,217) were compared by random sampling of diagnosis-related group-matched all-harm and nonharmed groups to ensure similar clinical conditions, as measured by Charlson Comorbidity Index and case mix index. Sampling (2 groups, n = 100 and n = 200) showed that increased LOS was associated with harm; yet other clinical comparators were similar across groups. the preventable-harm subcohort had longer LOS (10.7 versus 5.9 days) and higher DVC ($13,442 versus $8024) than the nonharmed cohort. Identification of harm events was nearly 6-fold higher with the ETS than with the voluntary event reporting system. CONCLUSIONS: Patients with preventable harm had increased LOS that was associated with higher DVC per preventable-harm encounter in a large US healthcare system.

Disparities in Patient Safety Voluntary Event Reporting: A Scoping Review.

Voluntary event reporting (VER) systems underestimate the incidence of safety events and often capture only serious events. A limited amount of data is collected through these systems, and they may be inadequate to characterize disparities in reported safety events. We conducted a scoping review of the literature to summarize the state of the evidence as it relates to differences in safety events and safety event reporting by age, gender, and race. Using a broad-based query, a systematic search for published, peer-reviewed literature that discusses patient safety event reporting and differences by age, gender, race, and socioeconomic status was conducted. Based on modified Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, 283 studies underwent title and abstract review, yielding 56 studies for full text review. After full text review, 23 studies were carefully reviewed individually, grouped thematically, and summarized to highlight the most pertinent findings. The studies reviewed yielded important insights, particularly with regard to race, gender, and the ways events are identified. Patients from minoritized groups may be less likely to have events reported and more likely to suffer serious events. Some studies found differences in rates of reporting safety events for female vs. male providers. The rate of VER is consistently lower than the rate of events identified through identified using automated detection. The current literature describing VER data shows disparities by race, language, age, and gender for patients and providers. Further research and systematic change are needed to specifically study these disparities to guide health care institutions on ways to mitigate bias and deliver more equitable care.

Hypoglycemia adverse events in PICUs and cardiac ICUs: differentiating preventable and nonpreventable events*.

OBJECTIVES: To describe the use of an adverse event detection system to identify, characterize, and categorize preventable versus nonpreventable hypoglycemia AEs in PICUs and cardiac ICUs. DESIGN: Retrospective observational study. SETTING: PICU and cardiac ICU of a tertiary pediatric hospital. SUBJECTS: All hypoglycemia triggers generated over a 3-year period. INTERVENTIONS: All hypoglycemia triggers generated via an electronic health record-driven surveillance system were investigated to determine if they represented a true adverse event and if that event was preventable or nonpreventable. Clinical and demographic variables were analyzed to identify characteristics of patients who developed a preventable or nonpreventable hypoglycemia adverse event. MEASUREMENTS AND MAIN RESULTS: There were 197 hypoglycemia adverse events in 90 patients. Thirty percent of the adverse events in the PICU and 36% of the adverse events in the cardiac ICU were characterized as preventable. Of the adverse events, 118 (59.9%) necessitated an intravenous dextrose bolus. No adverse events were associated with reporting of symptoms of hypoglycemia including apnea, altered mental status, or seizures. Events were more likely to be preventable (p < 0.001) if the patient was receiving only parenteral sources of nutrition (intravenous fluids or total parenteral nutrition). Controlling for weekends and holidays, adverse events associated with sole parenteral nutrition source had an increased odds ratio of 9.5 (95% confidence interval: 2.8-31.9) of being preventable. Stratifying by ICU, cardiac ICU events occurring on a weekend or holiday were more likely to be preventable (p = 0.001). Stratifying by unit and controlling for parenteral nutrition source, adverse events in the cardiac ICU occurring on weekends or holidays had an increased odds ratio of 11.6 (95% confidence interval: 2.7-50.2) of being preventable. CONCLUSIONS: Preventable hypoglycemia adverse events are associated with patients receiving sole parenteral sources of nutrition in both the PICU and cardiac ICU. In the cardiac ICU, there is an association between weekend and holiday time periods and the development of preventable hypoglycemia adverse events.

Association of the Child Opportunity Index and Inpatient Illness Severity in the United States, 2018-2019.

OBJECTIVE: Children residing in impoverished neighborhoods have reduced access to health care resources. Our objective was to identify potential associations between Child Opportunity Index (COI), a composite score of neighborhood characteristics, and inpatient severity of illness and clinical trajectory among United States (US) children. METHODS: This retrospective cohort study assessed data using the Pediatric Health Information System Registry from 2018 to 2019. Primary exposure variable was COI level (range: very low [CO1 1], low [COI 2], moderate [COI 3], high [COI 4], and very high [COI 5]). Markers of inpatient clinical severity included index mortality, Pediatric Intensive Care Unit (PICU) admission, invasive mechanical ventilation (IMV), and hospital length of stay (LOS). Subgroup analysis of COI and clinical outcome variation by United States Census Geographic Regions was conducted. Adjusted regression analysis was utilized to understand associations between COI and inpatient clinical severity outcomes. RESULTS: Of the 132,130 encounters, 44% resided in very low or low COI neighborhoods. In adjusted models, very low COI was associated with increased mortality (aOR: 1.35, 95% CI: 1.05-1.74, P = .018), PICU admission (aOR: 1.06, 95% CI: 1.02-1.11, P = 0.014), IMV (aOR: 1.12, 95% CI: 1.04-1.21, P = .002), and higher hospital LOS (P = .045). Regional variation by COI depicted the East North Central region having the highest rate of mortality (20.5%), P < .001, and PICU admissions (23%), P = .014. CONCLUSIONS: Our multicenter, retrospective study highlights the interaction between neighborhood-level deprivation and worsened health disparities, indicating a need for prospective study.

Body Mass Index Category and Adverse Events in Hospitalized Children.

OBJECTIVE: To identify associations between patient body mass index (BMI) category and adverse event (AE) rate, severity, and preventability in a cohort of children discharged from an academic children's hospital. METHODS: We identified patients 2 to 17 years old consecutively discharged between June and October 2018. Patient age, sex, height, and weight were used to categorize patients as having underweight, normal weight, overweight, or obesity. We used the Global Assessment of Pediatrics Patient Safety trigger tool to identify AEs, which were scored for harm and preventability. The primary outcome was the rate of AEs; these were compared with Poisson regression. We used multivariable logistic regression to model event preventability. RESULTS: We reviewed 834 encounters in 680 subjects; 51 (7.5%) had underweight, 367 (54.0%) had normal weight, 112 (16.5%) had overweight, and 150 (22.1%) had obesity. Our cohort experienced 270 AEs, with an overall rate of 69.7 (61.8-78.5) AEs per 1000 patient-days: 67.7 (46.4-98.7) in underweight, 70.0 (59.4-82.4) in normal weight, 58.6 (42.5-79.7) in overweight, and 80.4 (62.5-103.6) in obesity, P = .46. No associations were seen between BMI category and AE severity. Children with obesity had an increased rate of preventable AEs (P < .01), but this association did not persist in the multivariable model. CONCLUSIONS: In this single-center study, we did not find associations between BMI category and rate, severity, or preventability of AEs.

Culture Ordering for Patients with New-onset Fever: A Survey of Pediatric Intensive Care Unit Clinician Practices.

INTRODUCTION: Accurate assessment of infection in critically ill patients is vital to their care. Both indiscretion and under-utilization of diagnostic microbiology testing can contribute to inappropriate antibiotic administration or delays in diagnosis. However, indiscretion in diagnostic microbiology cultures may also lead to unnecessary tests that, if false-positive, would incur additional costs and unhelpful evaluations. This quality improvement project objective was to assess pediatric intensive care unit (PICU) clinicians' attitudes and practices around the microbiology work-up for patients with new-onset fever. METHODS: We developed and conducted a self-administered electronic survey of PICU clinicians at a single institution. The survey included 7 common clinical vignettes of PICU patients with new-onset fever and asked participants whether they would obtain central line blood cultures, peripheral blood cultures, respiratory aspirate cultures, cerebrospinal fluid cultures, urine cultures, and/or urinalyses. RESULTS: Forty-seven of 54 clinicians (87%) completed the survey. Diagnostic specimen ordering practices were notably heterogeneous. Respondents unanimously favored a decision-support algorithm to guide culture specimen ordering practices for PICU patients with fever (100%, N = 47). A majority (91.5%, N = 43) indicated that a decision-support algorithm would be a means to align PICU and consulting care teams when ordering culture specimens for patients with fever. CONCLUSION: This survey revealed variability of diagnostic specimen ordering practices for patients with new fever, supporting an opportunity to standardize practices. Clinicians favored a decision-support tool and thought that it would help align patient management between clinical team members. The results will be used to inform future diagnostic stewardship efforts.

Developing a patient safety surveillance system to identify adverse events in the intensive care unit.

Aggregation of adverse drug event data has evolved in the last decade. Several approaches are available to augment the standard voluntary incident reporting system. Most of these methods are applicable to nonmedication adverse events as well. To identify appropriately system trends as well as process failures, intensive care units should participate in various collection methods. Several different methods are available for robust adverse drug event data collection, such as target chart review, nontargeted chart review, and direct observation. As the various methods usually capture different types of events, employing more than one technique will improve the assessment of intensive care unit care. Some of these surveillance methods offer real-time or near real-time identification of adverse drug events and potentially afford the practitioner time for intervention. Continued development of adverse drug event detection will allow for further quality improvement efforts and preventive strategies to be utilized.

Racial, Ethnic, and Socioeconomic Disparities in Patient Safety Events for Hospitalized Children.

OBJECTIVES: Previous studies have revealed racial/ethnic and socioeconomic disparities in quality of care and patient safety. However, these disparities have not been examined in a pediatric inpatient environment by using a measure of clinically confirmed adverse events (AEs). In this study, we do so using the Global Assessment of Pediatric Patient Safety (GAPPS) Trigger Tool. METHODS: GAPPS was applied to medical records of randomly selected pediatric patients discharged from 16 hospitals in the Pediatric Research in Inpatient Settings Network across 4 US regions from January 2007 to December 2012. Disparities in AEs for hospitalized children were identified on the basis of patient race/ethnicity (black, Latino, white, or other; N = 17 336 patient days) and insurance status (public, private, or self-pay/no insurance; N = 19 030 patient days). RESULTS: Compared with hospitalized non-Latino white children, hospitalized Latino children experienced higher rates of all AEs (Latino: 30.1 AEs per 1000 patient days versus white: 16.9 AEs per 1000 patient days; P ≤ .001), preventable AEs (Latino: 15.9 AEs per 1000 patient days versus white: 8.9 AEs per 1000 patient days; P = .002), and high-severity AEs (Latino: 12.6 AEs per 1000 patient days versus white: 7.7 AEs per 1000 patient days; P = .02). Compared with privately insured children, publicly insured children experienced higher rates of preventable AEs (public: 12.1 AEs per 1000 patient days versus private: 8.5 AEs per 1000 patient days; P = .02). No significant differences were observed among other groups. CONCLUSIONS: The GAPPS analysis revealed racial and/or ethnic and socioeconomic disparities in rates of AEs experienced by hospitalized children across a broad range of geographic and hospital settings. Further investigation may reveal underlying mechanisms of these disparities and could help hospitals reduce harm.

An international assessment of a web-based diagnostic tool in critically ill children.

Improving diagnostic accuracy is essential. The extent of diagnostic uncertainty at patient admission is not well described in critically ill children. Therefore, we studied the extent that pediatric trainee diagnostic performance could be improved with the aid of a computerized diagnostic tool. Data regarding patient admissions to five Pediatric Intensive Care Units were collected. Information included patients' clinical details, admitting team's diagnostic workup and discharge diagnosis. An attending physician assessed each case independently and suggested additional diagnostic possibilities. Diagnostic accuracy was calculated using the discharge diagnosis as the gold standard. 206 out of 927 patients (22.2%) admitted to the PICUs did not have an established diagnosis at admission. The trainee teams considered a median of three diagnoses in their workup (IQR 3-5) and made an accurate diagnosis in 89.4% cases (95% CI 84.6%-94.2%). Diagnostic accuracy improved to 92.5% with use of the diagnostic tool alone, and to 95% with the addition of attending physicians' diagnostic suggestions. We conclude that a modest proportion of admissions to these PICUs were characterized by diagnostic uncertainty during initial assessment. Although there was a relatively high accuracy rate of initial assessment in our clinical setting, it was further improved by both the diagnostic tool and the physicians' diagnostic suggestions. It is plausible that the tool's utility would be even greater in clinical settings with less expertise in critical illness assessment, such as community hospitals, or emergency departments of non-training institutions. The role of diagnostic aids in the care of critically ill children merits further study.further study.

Using Length of Stay to Understand Patient Flow for Pediatric Inpatients.

OBJECTIVES: Develop and test a new metric to assess meaningful variability in inpatient flow. METHODS: Using the pediatric administrative dataset, Pediatric Health Information System, that quantifies the length of stay (LOS) in hours, all inpatient and observation encounters with 21 common diagnoses were included from the calendar year 2013 in 38 pediatric hospitals. Two mutually exclusive composite groups based on diagnosis and presence or absence of an ICU hospitalization termed Acute Care Composite (ACC) and ICU Composite (ICUC), respectively, were created. These composites consisted of an observed-to-expected (O/E) LOS as well as an excess LOS percentage (ie, the percent of day beyond expected). Seven-day all-cause risk-adjusted rehospitalizations was used as a balancing measure. The combination of the ACC, the ICUC, and the rehospitalization measures forms this new metric. RESULTS: The diagnosis groups in the ACC and the ICUC included 113,768 and 38,400 hospitalizations, respectively. The ACC had a median O/E LOS of 1.0, a median excess LOS percentage of 23.9% and a rehospitalization rate of 1.7%. The ICUC had a median O/E LOS of 1.1, a median excess LOS percentage of 32.3%, and rehospitalization rate of 4.9%. There was no relationship of O/E LOS and rehospitalization for either ACC or ICUC. CONCLUSIONS: This metric shows variation among hospitals and could allow a pediatric hospital to assess the performance of inpatient flow.