Coordinated Health Care Interventions for Childhood Asthma Gaps in Outcomes (CHICAGO) plan

Abstract

Background: Evidence-based strategies to improve outcomes in minority children with uncontrolled asthma discharged from the emergency department (ED) are needed.

Objectives: This multicenter pragmatic clinical trial was designed to compare an ED-only intervention (decision support tool), an ED-only intervention and home visits by community health workers for 6 months (ED-plus-home), and enhanced usual care (UC).

Methods: Children aged 5 to 11 years with uncontrolled asthma were enrolled. The change over 6 months in the Patient-Reported Outcomes Measurement Information System Asthma Impact Scale score in children and Satisfaction with Participation in Social Roles score in caregivers were the primary outcomes. The secondary outcomes included guideline-recommended ED discharge care and self-management.

Results: Recruitment was significantly lower than expected (373 vs 640 expected). Of the 373 children (64% Black and 31% Latino children), only 63% completed the 6-month follow-up visit. In multivariable analyses that accounted for missing data, the adjusted odds ratios and 98% CIs for differences in Asthma Impact Scores or caregivers' Satisfaction with Participation in Social Roles scores were not significant. However, guideline-recommended ED discharge care was significantly improved in the intervention groups versus in the UC group, and self-management behaviors were significantly improved in the ED-plus-home group versus in the ED-only and UC groups.

Conclusions: The ED-based interventions did not significantly improve the primary clinical outcomes, although the study was likely underpowered. Although guideline-recommended ED discharge care and self-management did improve, their effect on clinical outcomes needs further study.

Keywords: Pragmatic clinical trial; asthma; community health worker; emergency department.

Fig 1

CHICAGO plan study schematic. Children aged 5 to 11 years who presented to the ED with uncontrolled asthma were randomly allocated to 1 of 3 groups: an enhanced UC versus ED-only intervention (an ED-based intervention using the CAPE, a decision support and communication tool) versus an ED-plus-home intervention (the same ED-only intervention plus CHW-led home visits at 2 or 3 days, 2 weeks, 1 month, 3 months, and 6 months after ED discharge to help implement the CAPE and reduce environmental triggers). Outcomes were assessed at baseline (in person before ED discharge), 1 month (via phone), 3 months (via phone), and 6 months (in home or via phone). The prespecified primary time point for assessing the primary outcome was 6 months after ED discharge. A 12-month follow-up assessment via phone was discontinued owing to low retention rates. See the Supplementary Methods for the details.

Fig 2

Participant flow diagram: screening, randomization, follow-up, and analysis. Of 528 children approached to assess eligibility, 373 were eligible and randomized to the ED-plus-home, ED-only, or enhanced UC group. We defined complete data for any primary outcome as evaluable data for at least 1 of the 2 primary outcomes (PROMIS Asthma Impact Scale score or PROMIS Satisfaction with Participation in Social Roles score) at each time point. For the PROMIS Asthma Impact Scale score (the primary outcome for children), complete data for the change since enrollment at 1, 3, and 6 months were collected for 95 (79%), 83 (69%), and 76 (63%) children, respectively, in ED-plus-home group; for 103 (82%), 97 (77%), and 82 (65%) children, respectively, in the ED-only group; and for 98 (78%), 88 (70%), and 78 (62%) children, respectively, in the enhanced UC group. Similarly, for the PROMIS Satisfaction with Participation in Social Roles, the corresponding complete data were collected for 92 (76%), 82 (68%), and 75 (62%) children, respectively, in the ED-plus-home group; for 103 (82%), 95 (75%), and 82 (65%) children, respectively, in the ED-only group; and for 99 (79%), 86 (68%), and 76 (60%) children, respectively in the enhanced UC group. Lost to follow-up refers to the 6-month follow-up visit. Percentages may not add up to 100% because of rounding. Analyses for primary outcomes were by intention to treat and used multiple imputation for missing data. ∗Because of slower than expected enrollment, some exclusion criteria were modified after the start of the study; see the Methods for details.

Fig 3

Primary outcomes in adjusted analyses. The 6-month change (value at 6 months minus value at baseline) in PROMIS Asthma Impact Scale (children [A]) and Satisfaction with Participation in Social Roles (caregivers [B]) T-scores were the primary outcomes. The results of analyses using ordinal logistic regression models fitted with generalized estimating equations to estimate the adjusted odds ratios (ORs) and 98% CIs for higher quartile of change in T-scores for each outcome in the first group versus in the second group after accounting for prespecified baseline characteristics (race, ethnicity, sex, health insurance, site, and all-cause acute care use in the 12 months before enrollment) are presented. Hollow circles and squares indicate results of complete-case analyses; solid circles and squares indicate analyses after multiple imputations for missing data. The 98% CIs were used to correspond to a Bonferroni-adjusted P < .0167 (.05 of 3). Higher T-scores on the Asthma Impact Scale indicate worse asthma (A); thus, an OR greater than 1.0 indicates a smaller improvement in asthma impact in the first group than in the second group. Higher T-scores on the Satisfaction with Participation in Social Roles scale indicate better satisfaction (B); thus, an OR greater than 1.0 indicates a greater improvement in satisfaction in the first group than in the second group. Results of complete-case analyses (and after multiple imputation for missing data) indicate no significant differences (Bonferroni-adjusted 98% CIs all cross OR = 1.0) for either primary outcome between the 3 study groups over 6 months.

Fig 4

Discharge instructions provided to children and caregivers in the ED. Documented discharge instructions to child and caregiver were significantly more likely to be consistent with the guidelines in the ED-plus-home and ED-only groups than in the enhanced UC group, as follows: (1) use of systemic corticosteroids, P < .0001 and P < .0001, respectively; (2) use of inhaled corticosteroid or other controller, P < .0001 and P = .0004, respectively; (3) use of inhaled rescue medications, P = .0005 and P < .0001, respectively; and (4) scheduling of a follow-up appointment within 4 weeks after ED discharge, P < .0001 and P < .0001, respectively. Differences between the ED-plus-home and ED-only groups were not significant (P > .2 for all). Between-group differences were considered significant if the P was less than .004 (0.05 of 12 hypothesis tests).

Fig 5

Self-management after ED discharge. Pharmacy dispensation data were available for 82% of participants (80%, 86%, and 79% in the ED-plus-home, ED-only, and enhanced UC group, respectively). The evaluable data suggest that slightly more than 80% of those in all 3 study groups filled prescriptions for systemic corticosteroids within 7 days of discharge. Those in the ED-plus-home group were significantly more likely to fill prescriptions for inhaled corticosteroids or some other controller than were those in the ED-only group (P = .0003) and the enhanced UC group (P < .0001). Attendance at an outpatient visit with the patient-identified asthma clinician within 4 weeks of ED discharge was significantly more common in the ED-plus-home group than in the group with enhanced UC (P = .005). Other pairwise comparisons were not significant (P > .2 for all). Between-group differences were considered significant if P was less than .006 (0.05 of 9 hypothesis tests).