Are drawings a reliable way of recording retraction pockets?

Abstract

Objectives. Although widely used, drawings have never been assessed for their accuracy and reproducibility when used to record retraction pockets. We set out to determine the relative reliability of drawings and endoscopic photographs to record retractions of the tympanic membrane.
Method. A series of 10 standardised ‘reference’ slides of retraction pockets were shown to 18 Otolaryngologists. Next to each slide were 10 drawings of a retraction pocket and 10 photographs of retraction pockets. Half of the photographs and half of the drawings were of the same ear as the ‘reference’ image and half were of a different ear, chosen at random. Participants were asked if each drawing and photograph was the same ear as ‘reference’ image, and to assess the relative stage of the pathology.
Results. Differences between the assessments of the photographs and drawings were dramatic. Where photographs were used, participants were able to identify an identical ear 96.4% of the time and a different ear 81.7% of the time; where drawings were used, this fell to 46.7% and 68.2% respectively (P < 0.001). Identification of the relative stage of pathology was also significantly better with a photograph of a tympanic membrane (96%) rather than a drawing (53.3%) (P < 0.001).
Conclusions. This is the first attempt in the literature to compare the relative diagnostic accuracy of drawings of the tympanic membrane with photographs. A great deal of reliance is placed on clinical drawings of varying quality by clinicians. Digital photography is considerably more accurate than drawings and its use should be encouraged.