General practitioners want to know whether a treatment works and is safe in practiceBMJ 1997; 315 doi: https://doi.org/10.1136/bmj.315.7118.1310 (Published 15 November 1997) Cite this as: BMJ 1997;315:1310
- Ueli Bollag, Specialist in child healtha
Editor—Froom et al question the benefit of routine use of antimicrobials for acute otitis media.1 I agree with their arguments but believe that the inclusion solely of randomised double blind placebo controlled trials in their study is insufficient to make the point. Studies undertaken in tertiary care centres cannot dispel practitioners' doubts about the need for antibiotics in acute otitis media.
Observational studies that Bollag-Albrecht and I carried out on the natural course and management of acute otitis media in an office based general and paediatric practice were meant to fill the gap between academic research and research in practice. After a practice audit of the management of acute otitis media we developed and tested a protocol in which we defined criteria for the diagnosis and treatment of acute otitis media.2 In a two year study of 168 children aged 16 and under with acute otitis media we found that antibiotics were not needed in the first instance; treatment with analgesics, and saline nose drops plus inhalation of steam to keep the mucous membranes moist, supported the self healing process and gave relief to the child. We confirmed these results in a study of another 62 children over one year.3 We did not see any case of clinically overt mastoiditis or meningitis. We have adhered to this regimen without experiencing any failure to this date.
Last year colleagues and I reported another study on the occurrence and importance of middle ear effusion.4 With careful clinical observation and follow up, and using acoustic reflectometry as an objective measurement, we found that middle ear effusion is common. It probably reflects a pathogenetic feature of the mucous membranes in the upper respiratory tract during a catarrhal episode.
The pathological importance attributed to middle ear effusion by specialists and academic researchers needs to be reassessed. If we are to make use of the statistical rigour of randomised controlled trials in guidelines for practical use we need to conduct studies in primary care practices serving unselected populations.