Libraries at University of Nebraska-Lincoln


Date of this Version

March 1992


Published in British Journal of Industrial Medicine 49 (1992), pp. 869-871. Copyright 1992 BMJ Publishing. Used by permission.


Variability between readers in the evaluation of the radiographic appearances of pneumoconiosis has been investigated by several authors. To reduce inter and intrareader variability, standard reference films were introduced for epidemiological purposes. Improved films and methods of classification were published in subsequent years. The category of profusion of small opacities is determined by considering the profusion as a whole over the affected zones of the lung, and by comparing this with the standard radiographs. This requires a mental process of integrating profusion over the affected zones.

The purpose of the present study was to determine whether there would be any advantage, in terms of improved reader agreement, in reading and reporting the six lung zones separately rather than integrating all readings to produce a single classification for each film. The possible validity of evaluating the six zones separately was suggested by a previous study of the relation between radiographic appearances and underlying pathology in which a good correlation was found (kappa = 0.47, p < 0.0001).

An opportunity arose to investigate this question during an epidemiological study of hard rock miners in Ontario.