Timely treatment of acute lower respiratory tract infections in children relies upon the assessment of clinical findings, especially respiratory rate (RR) and chest indrawing. Elevated RR has been shown to be an important indicator of underlying pulmonary disease such as pneumonia, asthma, bronchiolitis, and pulmonary oedema. RR has been described as a "simple and useful, although nonspecific, pulmonary function test of thoracic and pulmonary compliance" . Elevated rates, in combination with cough, have been used, and are currently being advocated by WHO, as an entry criterion for a treatment algorithm for pneumonia among infants an children.(a)
However, the measurement of RR is typically held in "low esteem"  by health workers who find that it is often a difficult and time-consuming manoeuvre, especially in infants. The conditions of crowded, noisy primary care settings also may not be conducive to counting the RR. The lure of high technology medicine also contributes to neglect of this basic clinical indicator. The use of simple instruments such as timers with audible cues has been suggested as a means of facilitating the counting of RR because the health worker does not have to look at the chest and a wrist-watch at the same time. However, the difference in accuracy between RRs counted using timers versus watches has not been assessed.
Chest-wall indrawing, defined as an inward movement of the lower chest wall, is the key sign for referring children with severe pneumonia to hospital using the WHO Acute Respiratory Infections (ARI) treatment algorithm. However, the recognition of this sign may not be uniform among health personnel owing to varying definitions, training and experience.
In the context of instituting a national programme for the control of ARI in Egypt, ways to increase the acceptance and performance of RR counting and to assess the recognition of chest-wall indrawing were sought. The purpose of this study was (1) to briefly survey the attitudes towards and the knowledge and practices of physicians in measuring the RR, (2) to examine the effect of a training videotape on the recognition of chest indrawing, and (3) to examine the effect of using different time intervals and providing audible timers on the accuracy of RR assessment. A test videotape focused on children of different ages and breathing rates was used to conduct these assessments so that potentially modifiable aspects of counting technique alone could be addressed.
Materials and methods
Primary health care in Egypt is delivered by physicians employed by the Ministry of Health. General practitioners, pediatricians and junior doctors in primary care centres in the study area were selected randomly from a list of staff serving five governorates (Alexandria, Assuit, Cairo, Ismailia and Menoufia) targeted for the initial implementation of the ARI control programme in Egypt. The 320 physicians in this study were distributed as follows: 45% in MCH units, 30% in urban health centres, 18% in rural health units, and 7% in the district hospital or in school health. The study participants, in groups of 16 in health centres in each of the governorates, were told about the ARI programme planning and were asked to evaluate the use of audible timers versus wrist-watches in measuring the RR of infants and children.
Descriptive survey. A questionnaire was first administered to survey the physicians' attitudes and practices regarding RR measurement, their use of watches while counting, and their knowledge of the age-appropriate RR cut-offs. The study participants were anonymously surveyed about (1) how they evaluated children with ARI using two open-ended questions on history and physical examination; (2) whether or not they counted the RR, and if so, how (multiple choice); (3) what constraints they faced in counting the RR (multiple choice); and (4) what was the usual RR of children aged <2 months, 2-6 months, and 2 years (open-ended). …