Article 1.

[In header:]Sharek PJ, et. al.. Agreement among measures of asthma status: a prospective study of low-income children with moderate to severe asthma. Pediatrics, 2002 Oct; 110 (4): 797-804

There is no single variable used in heath research that best represents asthma. Due to this lack of any “gold standard” for measuring asthma outcomes, several disease status measures are used as proxy for asthma when evaluating interventions. Some of these measures may be redundant and unnecessary while some may be complementary and useful. It is not necessarily better to use several variables for asthma in a study. In fact, use of multiple outcomes may lead to ambiguous results, increased type I error rates, and be an inefficient use of resources.
Sharek, et. al., examine correlations between measures of asthma commonly used in medical research. They hoped to find relationships between proxy asthma variables that would help to create parsimonious models using only complementary (correlated) variables.
Their results suggest that asthma status and change in asthma status over time are best characterized by parent-reported symptoms, parent-reported utilization, and functional health status measures. Asthma diaries and pulmonary function tests did not seem to provide additional benefit. Their findings suggest that parsimonious models would only include collection of key data elements regarding symptoms, utilization, and functional health status.
[In footer:]Reviewed by Andrew Wilson