Dataset Information

Criticality: A New Concept of Severity of Illness for Hospitalized Children.

ABSTRACT:

Objectives

To validate the conceptual framework of "criticality," a new pediatric inpatient severity measure based on physiology, therapy, and therapeutic intensity calibrated to care intensity, operationalized as ICU care.

Design

Deep neural network analysis of a pediatric cohort from the Health Facts (Cerner Corporation, Kansas City, MO) national database.

Setting

Hospitals with pediatric routine inpatient and ICU care.

Patients

Children cared for in the ICU (n = 20,014) and in routine care units without an ICU admission (n = 20,130) from 2009 to 2016. All patients had laboratory, vital sign, and medication data.

Interventions

None.

Measurements and main results

A calibrated, deep neural network used physiology (laboratory tests and vital signs), therapy (medications), and therapeutic intensity (number of physiology tests and medications) to model care intensity, operationalized as ICU (versus routine) care every 6 hours of a patient's hospital course. The probability of ICU care is termed the Criticality Index. First, the model demonstrated excellent separation of criticality distributions from a severity hierarchy of five patient groups: routine care, routine care for those who also received ICU care, transition from routine to ICU care, ICU care, and high-intensity ICU care. Second, model performance assessed with statistical metrics was excellent with an area under the curve for the receiver operating characteristic of 0.95 for 327,189 6-hour time periods, excellent calibration, sensitivity of 0.817, specificity of 0.892, accuracy of 0.866, and precision of 0.799. Third, the performance in individual patients with greater than one care designation indicated as 88.03% (95% CI, 87.72-88.34) of the Criticality Indices in the more intensive locations was higher than the less intense locations.

Conclusions

The Criticality Index is a quantification of severity of illness for hospitalized children using physiology, therapy, and care intensity. This new conceptual model is applicable to clinical investigations and predicting future care needs.

SUBMITTER: Rivera EAT

PROVIDER: S-EPMC7790867 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:IntroductionPediatric mortality remains unacceptably high in many low-resource settings, with inpatient deaths often associated with delayed recognition of clinical deterioration. The Family-Assisted Severe Febrile Illness ThERapy (FASTER) tool has been developed for caregivers to assist in monitoring their hospitalized children and alert clinicians. This study evaluates feasibility of implementation by caregivers and clinicians.MethodsRandomized controlled feasibility study at Kenyatta National Hospital, Kenya. Children hospitalized with acute febrile illness with caregivers at the bedside for 24 h were enrolled. Caregivers were trained using the FASTER tool. The primary outcome was the frequency of clinician reassessments between intervention (FASTER) and standard care arms. Poisson regression with random intercept for grouping by patient was used, adjusting for admission pediatric early warning score, age, gender. Secondary outcomes included survey assessments of clinician and caregiver experiences with FASTER.ResultsOne hundred and fifty patient/caregiver pairs were enrolled, 139 included in the analysis, 74 in the intervention, 65 in the control arm. Patients' median age was 0.9 (range 0.2-10) and 1.1 years (range 0.2-12) in intervention vs. control arms. The most common diagnoses were pneumonia (80[58%]), meningitis (58[38%]) and malaria (34 [24%]). 134 (96%) caregivers were patients' mothers. Clinician visits/hour increased with patients' illness severity in both arms, but without difference in frequency between arms (point estimate for difference -0.9%, p = 0.97). Of the 16 deaths, 8 (four/arm) occurred within 2 days of enrollment. Forty clinicians were surveyed, 33 (82%) reporting that FASTER could improve outcomes of very sick children in low-resource settings; 26 (65%) rating caregivers as able to adequately capture patients' severity of illness. Of 70 caregivers surveyed, 63 (90%) reported that FASTER training was easy to understand; all (100%) agreed that the intervention would improve care of hospitalized children and help identify sick children in their community.DiscussionWe observed no difference in recorded frequency of clinician visits with FASTER monitoring. However, the tool was rated positively by caregivers and clinicians., Implementation appears feasible but requires optimization. These feasibility data may inform a larger trial powered to measure morbidity and mortality outcomes to determine the utility of FASTER in detecting and responding to clinical deterioration in low-resource settings.Clinical trial registrationClinicalTrials.gov, identifier: NCT03513861.

Project description:ImportanceThe measured severity of illness of hospitalized Medicare beneficiaries has increased. Whether this change is associated with payment reforms, concentrated among hospitalizations with principal diagnoses targeted by payment reform, and reflective of true increases in severity of illness is unknown.ObjectivesTo assess whether the expansion of secondary diagnosis codes in January 2011 and the incentive payments for health information technology under the US Health Information Technology for Economic and Clinical Health Act were associated with changes in measured severity of illness and whether those changes are reflective of true increases in underlying patient severity.Design, setting, and participantsThis cohort study of Medicare fee-for-service beneficiary discharges (N = 47 951 443) between January 1, 2008, and August 31, 2015, used a regression-discontinuity design to evaluate changes in measured severity of illness after the expansion of secondary diagnoses. Discharge-level linear regression model with hospital fixed effects was used to evaluate changes in measured severity of illness after hospitals' receipt of incentives for health information technology. The change in predictive accuracy of measured severity of illness on 30-day readmissions after the implementation of both policies was evaluated. Data analysis was performed from November 1, 2018, to March 5, 2019.Main outcomes and measuresThe primary outcome was patients' measured severity of illness determined by the number of condition categories from secondary discharge diagnosis codes. Measured severity of illness for diagnoses commonly targeted by Medicare policies and untargeted diagnoses was assessed.ResultsIn total, 47 951 443 discharges at 2850 hospitals were included. In 2008, these beneficiaries included 3 882 672 women (58.5%) with a mean (SD) age of 78.5 (8.4) years. In 2014, the discharges included 3 377 137 women (57.8%) with the mean (SD) age of 78.4 (8.7) years. The Centers for Medicare & Medicaid Services expansion of secondary diagnoses was associated with a 0.348 (95% CI, 0.328-0.367; P < .001) change in condition categories for all diagnoses, 0.445 (95% CI, 0.419-0.470; P < .001) for targeted diagnoses, and 0.321 (95% CI, 0.302-0.341; P < .001) for untargeted diagnoses. Health information technology incentives were associated with a 0.013 (95% CI, 0.004-0.022; P = .005) change in condition categories for all diagnoses, 0.195 (95% CI, 0.184-0.207; P < .001) for targeted diagnoses, and -0.016 (95% CI, -0.025 to -0.007; P < .001) for untargeted diagnoses. Minimal improvements in predictive accuracy were observed.Conclusions and relevanceChanges in Centers for Medicare & Medicaid Services policies appear to be associated with increases in measured severity of illness; these increases do not appear to reflect substantive changes in true patient severity.

Project description:ObjectivesTo assess severity of illness trajectories described by the Criticality Index for survivors and deaths in five patient groups defined by the sequence of patient care in ICU and routine patient care locations.DesignThe Criticality Index developed using a calibrated, deep neural network, measures severity of illness using physiology, therapies, and therapeutic intensity. Criticality Index values in sequential 6-hour time periods described severity trajectories.SettingHospitals with pediatric inpatient and ICU care.PatientsPediatric patients never cared for in an ICU (n = 20,091), patients only cared for in the ICU (n = 2,096) and patients cared for in both ICU and non-ICU care locations (n = 17,023) from 2009 to 2016 Health Facts database (Cerner Corporation, Kansas City, MO).InterventionsNone.Measurements and main resultsCriticality Index values were consistent with clinical experience. The median (25-75th percentile) ICU Criticality Index values (0.878 [0.696-0.966]) were more than 80-fold higher than the non-ICU values (0.010 [0.002-0.099]). Non-ICU Criticality Index values for patients transferred to the ICU were 40-fold higher than those never transferred to the ICU (0.164 vs 0.004). The median for ICU deaths was higher than ICU survivors (0.983 vs 0.875) (p < 0.001). The severity trajectories for the five groups met expectations based on clinical experience. Survivors had increasing Criticality Index values in non-ICU locations prior to ICU admission, decreasing Criticality Index values in the ICU, and decreasing Criticality Index values until hospital discharge. Deaths had higher Criticality Index values than survivors, steeper increases prior to the ICU, and worsening values in the ICU. Deaths had a variable course, especially those who died in non-ICU care locations, consistent with deaths associated with both active therapies and withdrawals/limitations of care.ConclusionsSeverity trajectories measured by the Criticality Index showed strong validity, reflecting the expected clinical course for five diverse patient groups.