Estimation of the FEV.

The procedure recommended by the Medical Research Council for estimating a subject's forced expiratory volume in one second (FEV1) is to require five separate attempts, discard the first two results, and average the last three. The most popular alternatives are to use the largest of the last three or the largest of a smaller number of results. Nine different indices derived from some or all of five attempts were compared in two studies. In one 40 normal subjects were studied. In the other 335 men exposed to industrial dust, whose forced expiratory volume declined with their degree of radiological pneumoconiosis as well as with age, were studied. There were small but consistent differences between indices. The index which emerged as the best overall in both studies was the mean of the largest three results from five attempts. It was better than the recommended index for all the comparisons made, but at the same time it gave a very similar mean value for the FEV1. Excluding the lowest two results rather than the first two from five blows is a rational procedure, and it should be formally recognised as providing the best index available.