Validity of spirometry performed online.

Spirometry is essential for the diagnosis and management of common respiratory diseases. However, its use and quality are low in primary care. An important reason for this is the technical difficulty in performing conventional spirometry. If high-quality spirometry could be performed online, from the pulmonary function laboratory in hospitals, most of the technical problems could be solved. The aim of the present study was to compare spirometries performed online by remote technicians with conventional spirometry. This was a controlled, randomised crossover study of 261 patients referred from primary care centres for pulmonary consultation. They were randomised to undergo either conventional or online spirometry. The technician, located remotely, controlled the spirometer computer. Using a teleconference link, the technician guided the patient through the spirometry. The comparison between conventional and online spirometries was performed on intention to treat and per protocol bases for spirometric values and quality criteria. Agreement between the two spirometric methods was assessed with a Bland-Altman plot. A subpopulation of off-range patients was also characterised. Finally, intra- and interobserver agreement was evaluated using the intraclass correlation coefficient. No clinically significant differences were seen between the online and conventional spirometric values in both the intention to treat and per protocol analyses. The agreement in Bland-Altman analysis was poorer for intention to treat than for the per protocol analysis. The latter had a lower percentage of off-range patients and high agreement to determine abnormal spirometry in the off-range group. Conventional spirometry had a higher percentage of patients with spirometric quality criteria although the quality criteria difference was only 5.9%, when both procedures were the first to start. Very good agreement was found between intra- and interobserver reliability. Spirometry performed online from a hospital can be an adequate alternative to conventional spirometry for primary care centres.