247EMF Cluster Analysis of Acute Care Utilization Yields Insights for Tailored Pediatric Asthma Interventions

Abstract

Study ObjectiveTo understand patterns of pediatric asthma-related acute care use to inform interventions to reduce avoidable hospitalizations.MethodsAnalysis is based on 2010-2014 all-payer claims data from two hospitals and a freestanding emergency department (ED). Cluster analysis was performed to classify patients aged 0-17 through two variables total asthma-related ED visits and total asthma-related hospitalizations. Patient typologies were generated using a Ward’s Method with Squared Euclidean Distance hierarchical clustering procedure. After determining the optimal solution, clusters were compared based on demographics, non-asthma-related ED visits and hospitalizations, 30-day return ED visits and asthma readmissions, mental health comorbidity prevalence, use of different acute care facilities, and Medicaid status.Results3,170 children met inclusion criteria. Cluster analysis showed 5 typologies of patients with distinct asthma-related acute care use, socio-behavioral and health characteristics. In all groups more than half of the children were between the ages of 3-10 years old with a male predominance. The predominant race in all groups is Black, followed by Hispanic and White, with a large proportion of patients in group 5 being Black (65.5%). Cluster 1 (62% of patients) shows the lowest rates of acute care use. No child in cluster 1 was hospitalized for asthma. They used the ED for asthma only once (66.2%) and had moderate levels of ED use for other conditions. They were least likely to have a mental health-related discharge diagnosis, and were less likely to have visited multiple facilities, with 38.9% receiving care at only one. Cluster 2 (19% of patients) showed a pattern of low asthma ED visits and one-time hospitalizations. Almost half (47.7%) had no asthma ED visits with low rates of care received for other conditions. Cluster 3 (11% of patients) had high rates of ED visits, both for asthma and other diagnoses. About two thirds had 5 or more asthma (68.5%) and non-asthma (65.8%) ED visits. Despite high rates of ED use, hospitalization for asthma and other conditions was rare. This group had the highest rate of multiple facility use; almost 31.5% received care at all 3 facilities. Cluster 4 (7% of patients) had moderate ED use for both asthma and other illnesses, with higher rates of asthma hospitalizations compared to cluster 1 with similar levels of ED use. Nearly one-quarter (24.4%) had received care at all 3 facilities, and one in 10 had a mental health diagnosis. Cluster 5 (n=29) had the highest overall rates of acute care use. Almost half (48.3%) had been hospitalized for asthma within 30 days of a prior asthma-related hospitalization. One in 7 (13.8%) had a mental health diagnosis.Conclusion Study ObjectiveTo understand patterns of pediatric asthma-related acute care use to inform interventions to reduce avoidable hospitalizations. To understand patterns of pediatric asthma-related acute care use to inform interventions to reduce avoidable hospitalizations. MethodsAnalysis is based on 2010-2014 all-payer claims data from two hospitals and a freestanding emergency department (ED). Cluster analysis was performed to classify patients aged 0-17 through two variables total asthma-related ED visits and total asthma-related hospitalizations. Patient typologies were generated using a Ward’s Method with Squared Euclidean Distance hierarchical clustering procedure. After determining the optimal solution, clusters were compared based on demographics, non-asthma-related ED visits and hospitalizations, 30-day return ED visits and asthma readmissions, mental health comorbidity prevalence, use of different acute care facilities, and Medicaid status. Analysis is based on 2010-2014 all-payer claims data from two hospitals and a freestanding emergency department (ED). Cluster analysis was performed to classify patients aged 0-17 through two variables total asthma-related ED visits and total asthma-related hospitalizations. Patient typologies were generated using a Ward’s Method with Squared Euclidean Distance hierarchical clustering procedure. After determining the optimal solution, clusters were compared based on demographics, non-asthma-related ED visits and hospitalizations, 30-day return ED visits and asthma readmissions, mental health comorbidity prevalence, use of different acute care facilities, and Medicaid status. Results3,170 children met inclusion criteria. Cluster analysis showed 5 typologies of patients with distinct asthma-related acute care use, socio-behavioral and health characteristics. In all groups more than half of the children were between the ages of 3-10 years old with a male predominance. The predominant race in all groups is Black, followed by Hispanic and White, with a large proportion of patients in group 5 being Black (65.5%). Cluster 1 (62% of patients) shows the lowest rates of acute care use. No child in cluster 1 was hospitalized for asthma. They used the ED for asthma only once (66.2%) and had moderate levels of ED use for other conditions. They were least likely to have a mental health-related discharge diagnosis, and were less likely to have visited multiple facilities, with 38.9% receiving care at only one. Cluster 2 (19% of patients) showed a pattern of low asthma ED visits and one-time hospitalizations. Almost half (47.7%) had no asthma ED visits with low rates of care received for other conditions. Cluster 3 (11% of patients) had high rates of ED visits, both for asthma and other diagnoses. About two thirds had 5 or more asthma (68.5%) and non-asthma (65.8%) ED visits. Despite high rates of ED use, hospitalization for asthma and other conditions was rare. This group had the highest rate of multiple facility use; almost 31.5% received care at all 3 facilities. Cluster 4 (7% of patients) had moderate ED use for both asthma and other illnesses, with higher rates of asthma hospitalizations compared to cluster 1 with similar levels of ED use. Nearly one-quarter (24.4%) had received care at all 3 facilities, and one in 10 had a mental health diagnosis. Cluster 5 (n=29) had the highest overall rates of acute care use. Almost half (48.3%) had been hospitalized for asthma within 30 days of a prior asthma-related hospitalization. One in 7 (13.8%) had a mental health diagnosis. 3,170 children met inclusion criteria. Cluster analysis showed 5 typologies of patients with distinct asthma-related acute care use, socio-behavioral and health characteristics. In all groups more than half of the children were between the ages of 3-10 years old with a male predominance. The predominant race in all groups is Black, followed by Hispanic and White, with a large proportion of patients in group 5 being Black (65.5%). Cluster 1 (62% of patients) shows the lowest rates of acute care use. No child in cluster 1 was hospitalized for asthma. They used the ED for asthma only once (66.2%) and had moderate levels of ED use for other conditions. They were least likely to have a mental health-related discharge diagnosis, and were less likely to have visited multiple facilities, with 38.9% receiving care at only one. Cluster 2 (19% of patients) showed a pattern of low asthma ED visits and one-time hospitalizations. Almost half (47.7%) had no asthma ED visits with low rates of care received for other conditions. Cluster 3 (11% of patients) had high rates of ED visits, both for asthma and other diagnoses. About two thirds had 5 or more asthma (68.5%) and non-asthma (65.8%) ED visits. Despite high rates of ED use, hospitalization for asthma and other conditions was rare. This group had the highest rate of multiple facility use; almost 31.5% received care at all 3 facilities. Cluster 4 (7% of patients) had moderate ED use for both asthma and other illnesses, with higher rates of asthma hospitalizations compared to cluster 1 with similar levels of ED use. Nearly one-quarter (24.4%) had received care at all 3 facilities, and one in 10 had a mental health diagnosis. Cluster 5 (n=29) had the highest overall rates of acute care use. Almost half (48.3%) had been hospitalized for asthma within 30 days of a prior asthma-related hospitalization. One in 7 (13.8%) had a mental health diagnosis. Conclusion