Repeatability of physical signs in airways obstruction

Abstract

We have examined the observer variation in the detection of physical signs in airways obstruction. In the first study, 10 relatively experienced physicians examined 11 patients for the presence of six familiar signs and seven less well-known signs. Six experienced observers then underwent a training period after which they examined another group of 21 patients. The repeatability of all the signs fell about midway between that expected by chance and the maximum possible. There was no difference between the familiar and unfamiliar signs. The training period resulted in a slight but not significant improvement in reliability.We have examined the observer error in the elicitation of some familiar and some less well-known physical signs in patients with airways obstruction. To provide additional information on the importance of skill and training, the study was undertaken in two stages. In the initial study a group of physicians were briefly instructed in the elicitation of the signs, some of which were relatively unfamiliar, and they then examined a series of patients. In the second stage, a smaller group of the more experienced members of the team were trained together for a time and they then examined another series of patients. METHODSIn the initial study, 10 observers examined 11 patients. These observers were all physicians at the Hammersmith Hospital. Seven were members of the Royal College of Physicians, one was of equivalent American standard, and two were post-registration house physicians. In the second study, six observers examined 21 patients. These observers were all members of the Royal College of Physicians and ranged in experience from consultant to registrar. The patients for both studies were selected from the wards and outpatient clinics, and their co-operation was sought after the nature of the study had been explained to them.PHYSICAL SIGNS All patients were examined while reclining on a couch with the back rest raised to 45'. The following list of physical signs is based on tPresent address:

Wheezes

“…Wheezes are usual clinical features in patients with chronic obstructive airway disease. However, they are not always present in bronchial obstruction, and GODFREY et al [4] found that only 70% of patients with severe airflow obstruction (forced expiratory volume in one second (FEV1) <1L) wheeze. Wheezes are more frequent in patients with asthma than in patients with chronic obstructive pulmonary disease (COPD) [38].…”

“…However, subjective evaluation of wheezing is not very reliable [3][4][5] and, despite the nomenclature, the intraobserver agreement on lung sound terminology has been found, a few years ago, to fall somewhere between chance and total agreement [6].…”

Interobserver variability of croup scoring in clinical practice

“…Among physician observers of infants with lower respiratory tract infection, agreement ranged from 0.2 to 0.66 for auscultatory findings, respiratory effort and colour (25). In studies of faculty and trainee clinicians assessing over 30 observation items in adults with respiratory disease, only tachypnea and percussion note had agreement more than 50% beyond chance (26)(27)(28)(29). In the context of randomized, controlled clinical trials of steroid use for croup with fewer numbers of trained observers, much better agreement for at least two scoring items has been reported.…”

Pneumonia in Children and Adolescents

“…Thus, the absence of all four findings could be used to rule out pneumonia without need for a chest film. Measurement of tachypnea has good reproducibility compared with other findings such as detection of retractions, crackles, or wheezes (Godfrey et al, 1969;Wang et al, 1992Wang et al, , 1996. The accurate measurement of respiratory rate should be an integral part of the assessment of any child with suspected pneumonia.…”