Leigh Anne Bakel1, Troy Richardson2, Heidi Gruhler De Souza2, Sunitha V Kaiser3, Sanjay Mahant4, Jennifer D Treasure5, Ilana Y Waynik6, Jeffrey C Winer7, Lalit Bajaj1. 1. Section of Hospital Medicine, Department of Pediatrics, Clinical Effectiveness Team, Children's Hospital Colorado, University of Colorado School of Medicine, Aurora, Colorado, USA. 2. Children's Hospital Association, Lenexa, Kansas, USA. 3. Department of Pediatrics, University of California at San Francisco, San Francisco, California, USA. 4. Department of Paediatrics, The Hospital for Sick Children, University of Toronto, Toronto, Canada. 5. Division of Hospital Medicine, Cincinnati Children's Hospital Medical Center, Cincinnati, Ohio, USA. 6. Department of Pediatrics, Connecticut Children's Medical Center, University of Connecticut, Mansfield, Connecticut, USA. 7. Department of Pediatrics, Le Bonheur Children's Hospital, University of Tennessee Health Science Center, Memphis, Tennessee, USA.
Abstract
BACKGROUND: Previously few means existed to broadly examine variability across conditions/practices within or between hospitals for common pediatric conditions. OBJECTIVE: Our objective was to develop a novel empiric measure of variation in care and test its association with patient-centered outcomes. DESIGNS: We conducted a retrospective cohort study of children hospitalized from January 2016 to December 2018 using the Pediatric Hospital Information Systems database. SETTINGS AND PARTICIPANTS: We included children ages 0-18 years hospitalized with asthma, bronchiolitis, or gastroenteritis. INTERVENTION: We developed a hospital-specific measure of variation in care, the hospital's observed specific standard practice (HOSSP), the most common combination of laboratory studies, imaging, and medications used at each hospital. MAIN OUTCOME AND MEASURES: The outcomes were standardized costs, length of stay (LOS), and 7-day all-cause readmissions. RESULTS: Among 133,392 hospitalizations from 41 hospitals (asthma = 50,382, bronchiolitis = 54,745, and gastroenteritis = 28,265), there was significant variation in overall HOSSP adherence across hospitals for these conditions (asthma: 3.5%-47.4% [p < .001], bronchiolitis: 2.5%-19.8% [p < .001], gastroenteritis: 1.6%-11.6% [p < .001]). The majority of HOSSP variation was driven by differences in medication prescribing for asthma and bronchiolitis and laboratory ordering for gastroenteritis. For all three conditions, greater HOSSP adherence was associated with significantly lower hospital costs (asthma: p = .04, bronchiolitis: p < .001, acute gastroenteritis: p = .01), without increases in LOS or 7-day all cause readmissions. CONCLUSION: We found substantial variation in the components and adherence to HOSSP. Hospitals with greater HOSSP adherence had lower costs for these conditions. This suggests hospitals can use data around laboratory, imaging, and medication prescribing practices to drive standardization of care, reduce unnecessary testing and treatment, determine best practices, and reduce costs.
BACKGROUND: Previously few means existed to broadly examine variability across conditions/practices within or between hospitals for common pediatric conditions. OBJECTIVE: Our objective was to develop a novel empiric measure of variation in care and test its association with patient-centered outcomes. DESIGNS: We conducted a retrospective cohort study of children hospitalized from January 2016 to December 2018 using the Pediatric Hospital Information Systems database. SETTINGS AND PARTICIPANTS: We included children ages 0-18 years hospitalized with asthma, bronchiolitis, or gastroenteritis. INTERVENTION: We developed a hospital-specific measure of variation in care, the hospital's observed specific standard practice (HOSSP), the most common combination of laboratory studies, imaging, and medications used at each hospital. MAIN OUTCOME AND MEASURES: The outcomes were standardized costs, length of stay (LOS), and 7-day all-cause readmissions. RESULTS: Among 133,392 hospitalizations from 41 hospitals (asthma = 50,382, bronchiolitis = 54,745, and gastroenteritis = 28,265), there was significant variation in overall HOSSP adherence across hospitals for these conditions (asthma: 3.5%-47.4% [p < .001], bronchiolitis: 2.5%-19.8% [p < .001], gastroenteritis: 1.6%-11.6% [p < .001]). The majority of HOSSP variation was driven by differences in medication prescribing for asthma and bronchiolitis and laboratory ordering for gastroenteritis. For all three conditions, greater HOSSP adherence was associated with significantly lower hospital costs (asthma: p = .04, bronchiolitis: p < .001, acute gastroenteritis: p = .01), without increases in LOS or 7-day all cause readmissions. CONCLUSION: We found substantial variation in the components and adherence to HOSSP. Hospitals with greater HOSSP adherence had lower costs for these conditions. This suggests hospitals can use data around laboratory, imaging, and medication prescribing practices to drive standardization of care, reduce unnecessary testing and treatment, determine best practices, and reduce costs.