There is no universally accepted and validated measure of asthma sever
ity. For community research, clinical tests are too costly, and epidem
iological assessments provide inadequate data on severity. Symptom mea
sures may offer a practical alternative. This study assessed psychomet
ric properties of symptom ratings of 91 asthmatic children. Reliabilit
y and validity of scales created from these items were examined. A sum
scale of symptom ratings was internally consistent, reliable across t
ime, and associated with concurrent health indices. This scale may be
a practical measure of severity for use in community-based research.