Comparison of two assessment tools for hospitalized subjects with asthma

Sangeeta K. Schroeder, Waheeda Samady*, Irini N. Kolaitis, Craig M. Smith, Hannah Palac, Laura Shreffler, Mary A. Nevin

*Corresponding author for this work

Research output: Contribution to journalArticlepeer-review

1 Scopus citations


BACKGROUND: Pediatric Asthma Assessment tools used to guide the weaning of inhaled therapies during inpatient hospitalization require further evaluation and validation. This study aimed to compare 2 asthma assessment tools: an asthma scale versus an asthma score. METHODS: A prospective, physician-blinded, comparison study was conducted in 2 separate 6-week phases of patients > 2 y old admitted to a tertiary care children’s hospital with status asthmaticus between July and November 2014. The asthma scale categorized 5 components (oxygen, auscultation, dyspnea, breathing frequency, and pulse oximetry) into 1 of 3 respiratory assessments: mild, moderate, or severe. The asthma score used a sum of the components, resulting in a score of 1–15. Study tool predictability was measured using a metric based on hours on continuous albu-terol, with area under the curve ≥ 0.8 indicating good predictability. Agreement between clinicians was measured using the Cohen kappa statistic. Study tool clinical correlation was measured using Spearman coefficient. Usability was evaluated using web-based surveys. RESULTS: Phase 1 included 1,971 assessments (97 unique subjects), whereas phase 2 included 607 assessments (69 unique subjects). Using the continuous albuterol metric, predictability of the asthma scale had an area under the curve of 0.62 versus the asthma score area under the curve of 0.80. Agreement early in hospitalization for the asthma scale was kappa = 0.34 (95% CI 0.18–0.5; n = 84) versus kappa = 0.55 (95% CI 0.35–0.76; n = 44) for the asthma score. Agreement late in hospitalization for the asthma scale was kappa = 0.38 (95% CI 0.17–0.59; n = 66) versus kappa = 0.41 (95% CI 0.13–0.69; n = 33) for the asthma score. Clinical correlation for the asthma scale (no. = 1,908) was r = 0.57 (P < .001) versus r = 0.80 (P < .001) for the asthma score (no. = 558). Mean asthma scale usability was 3.38 versus 3.68 for the asthma score. CONCLUSIONS: The asthma score showed better clinical predictability and clinical correlation compared to the asthma scale. Numerical scores provided more objective assessments compared to categorical scores. Validated scoring tools such as the asthma score are crucial to the success of manage-ment of inpatient asthma care.

Original languageEnglish (US)
Pages (from-to)104-112
Number of pages9
JournalRespiratory care
Issue number1
StatePublished - Jan 1 2021


  • Assessment tools
  • Asthma
  • In-patient
  • Nurses
  • Pediatric
  • Respiratory therapist

ASJC Scopus subject areas

  • Critical Care and Intensive Care Medicine
  • Pulmonary and Respiratory Medicine


Dive into the research topics of 'Comparison of two assessment tools for hospitalized subjects with asthma'. Together they form a unique fingerprint.

Cite this