Even subsequent studies purporting to demonstrate a clear benefit for incorporation of Feno into clinical management had methodological design problems, and several meta-analyses failed to demonstrate a cumulative benefit. Such trial design problems confounded the interpretation of whether Feno yielded an incremental, beneficial role in various subpopulations of people with asthma beyond that offered from a routine history, physical examination, and peak flow monitoring. For example, Powell et al identified an important, hard-to-study clinical problem: poorly controlled, undertreated asthma during pregnancy. Unfortunately, in their trial, most patients were not receiving inhaled corticosteroids (ICSs) at study entry, and with only a single clinic visit 2 weeks after study entry, little time existed to optimize asthma care by standard therapy. The clinical care algorithm relied on the Asthma Control Questionnaire (ACQ) as the major determinant of whether to increase or decrease ICS dosages; and for patients in the ACQ category of “partial loss of control,” the trial algorithm did not call for a change in ICS or other therapy, potentially leaving asthma undertreated. Unsurprisingly, at the end of the study, patients monitored by Feno measurement used more ICSs and experienced fewer exacerbations than patients in the clinical algorithm arm. Feno advocates cite the study as evidence that Feno testing beneficially guides effective asthma therapy. In contrast, Feno critics cite the study as uninterpretable, essentially comparing an inadequate treatment regimen in the clinical algorithm group with an unproven Feno intervention, since ideally patients in both arms should have been receiving higher doses of ICS at both study entry and study end. These differences in perspective highlight the continuing uncertainty over whether the cost and benefit (eg, value) warrants widespread or frequent use of Feno testing in clinical practice.