Improving Quality of Acute Asthma Care in US Hospitals: Changes Between 1999-2000 and 2012-2013

Abstract

Background: Little is known about the longitudinal change in the quality of acute asthma care for hospitalized children and adults in the United States. We investigated whether the concordance of inpatient asthma care with the national guidelines improved over time, identified hospital characteristics predictive of guideline concordance, and determined whether guideline-concordant care is associated with a shorter hospital length of stay (LOS).

Methods: This study was an analysis of data from two multicenter chart review studies of hospitalized patients aged 2 to 54 years with acute asthma during two time periods: 1999-2000 and 2012-2013. Outcomes were guideline concordance at the patient and hospital levels, and association of patient composite concordance with hospital LOS.

Results: The analytic cohort for the comparison of guideline concordance comprised 1,634 patients: 834 patients from 1999-2000 vs 800 patients from 2012-2013. Over these 15 years, inpatient asthma care became more concordant at the hospital-level, with the mean composite score increasing from 74 to 82 (P < .001). However, during 2012-2013, wide variability in guideline concordance of acute asthma care remained across hospitals, with the greatest variation in provision of individualized written action plan at discharge (SD, 36). Guideline concordance was significantly lower in Midwestern and Southern hospitals compared with Northeastern hospitals. After adjusting for severity, patients who received care perfectly concordant with the guidelines had significantly shorter hospital LOS (-14% [95% CI, -23 to -4]; P = .009).

Conclusions: Between 1999 and 2013, the guideline concordance of acute asthma care for hospitalized patients improved. However, interhospital variability remains substantial. Greater concordance with evidence-based guidelines was associated with a shorter hospital LOS.

Keywords: asthma; guideline; hospitalization; quality of care.

Figure 1

Distribution of composite guideline concordance score at the patient-level according to time period. At the patient level, the mean composite concordance score increased from 74 (SD, 14) in the 1999-2000 period to 82 (SD, 14) in the 2012-2013 period (P < .001). The superimposed curves represent the normal curve based on each sample mean and SD.

Figure 2

Distribution of composite guideline concordance score at the hospital-level according to time period. At the hospital level, the mean composite concordance score increased from 74 (SD, 6) in the 1999-2000 period to 82 (SD, 7) in the 2012-2013 period (P < .001). The superimposed curves represent the normal curve based on each sample mean and SD.