In 1972 a British epidemiologist, Archie -Cochrane, drew attention to our great collective ignorance about the effects of health care. He pointed out the fact that those who would like to make informed decisions about health care do not have access to reliable reviews of the available evidence. It was also recognized that the quality of published reviews was poor. Cochrane emphasized the need for systematically prepared reviews of research evidence, which are kept up-to-date to take into account new evidence.

The Cochrane collaboration was developed in response to Cochrane's call for systematic, up-to-date reviews of all relevant randomized clinical trials (RCTs) of health care. In 1992 the first Cochrane Center was opened in Oxford to facilitate systematic reviews of RCTs across all areas of health care. The hope that there would be a collaborative international response to the Cochrane agenda came to fruition at the first in a series of annual Cochrane Colloquia held in October 1993; 77 people from nine countries co-founded the Cochrane Collaboration.

What are Systematic Reviews?

The results of clinical trials of specific interventions are often conflicting or lack sufficient power to reach a definite conclusion. In systematic reviews, published literature dealing with specific health interventions is collected. The studies are carefully scrutinized to assure that they meet the previously defined eligibility criteria for the review (e.g., a review may consider only RCTs where patients fulfilling certain strict diagnostic criteria are included). A dataset is assembled from the various studies which meet the study criteria (often the investigators have to be contacted for data which is not specified in the published report). The results of the eligible studies are analyzed using statistical synthesis of the data (meta-analysis), if appropriate and possible. The results of this analysis provides summary data on the results of all the trials combined which may be used in coming to a decision regarding the efficacy of the intervention.

What is a Cochrane Review?

In a Cochrane review, the reviewers use a standard methodology to assemble, appraise and synthesize data based on the work of the Cochrane methods working groups in order to improve the validity and precision of the reviews. The reviewers may also use a computer software (RevMan), prepared by the Cochrane methodologists, to help them to organize, ana- lyze and present their systematic reviews. Unlike other published reviews, the Cochrane review is constantly updated, as new information becomes available. The responsibility of updating the review lies with the reviewer:

The Cochrane Controlled Trials Register

To enable the reviewers to get a comprehensive list of articles to review, the Cochrane collaboration maintains the Cochrane Controlled Trials Register (CCTR). This is a bibliographic database of controlled trials identified by contributors to the Cochrane Collaboration through a systematic hand search of the world's health care journals. This is done with the collaboration of the National Library of Medicine in Washington D.C. (who produce Medline) and Reed Elsevier, Amsterdam (who produce Embase). This task was undertaken because it has been shown that the existing databases are inadequate for identifying all relevant studies.

The Cochrane Review Groups

The Cochrane collaboration consists of a number of collaborative review groups. The members of each group share a common interest in a particular health problem, e.g., Cochrane acute respiratory infections group, Cochrane stroke prevention group, etc. Each group defines the scope of the group and identifies specific topics for the preparation of reviews. The group takes the responsibility of preparing and maintaining the reviews related to topics in their area of interest. This is done through a group of reviewers who undertake to pre- pare and maintain (for a working lifetime) systematic reviews on different health interventions and the editorial board of the group which scrutinizes the protocol and the review prior to publication.

The Cochrane Database of Systematic Reviews

To ensure that the result of their work can be widely disseminated, the reviewers contribute their reviews to the Cochrane Database of Systematic Reviews. The re- views are constantly updated as new information becomes available and in response to criticisms from readers. The database, along with the CCTR, is distributed on disc, CD-ROM and on the Internet.

Since computers and Internet facilities

are still not widely available to practicing physicians in India, Indian Pediatrics has asked for and received the approval of the Cochrane Collaboration Acute Respiratory Infection Group to publish the abstracts of reviews prepared by this group. These abstracts will be published periodically in the journal along with a brief comment on the review from members of the Academy. The abstract of the first of these reviews on antibiotic treatment of acute otitis media is presented in this issue.

Antibiotic in Otitis Media

[Glasziou PP, Haym M, Delmar CB. Antibiotic versus placebo for otitis media in children. In: Acute Respiratory Infections Module of the Cochrane Database of Systematic Reviews (Updated 1 December; 1997). Eds. Douglas R, Bridges-Webb C, Glasziou P, Lozano J, Steinhaff M, Wang E. Available in the Cochrane Library (data- base on disk and CDROM). The Cochrane Collaboration; Issue 1. Oxford, Update Software; 1998; Updated Quarterly].

Objective: To assess the effectiveness of antibiotics in the clinical management of children with acute otitis media. Search strategy: The literature from 1958 to 1994 was systematically searched. to identify studies which looked at whether antibiotics reduce symptoms and complications of acute otitis media. Selection criteria: Studies were included if they were (a) randomized and (b) involved the use of one or more antibiotics against a control that did not receive antibiotics. Data collection and analysis: From the eight identified trials we extracted data on duration of symptpms, hearing problems and complications. Main results: Our primary interests were in (a) short-term relief of symptoms and (b) subsequent complications. The eight trials, all in developed countries, suggest no reduction in pain at 24 hours, but a 41 % relative reduction in pain at 2-7 days. Since approximately 80% of. patients will have settled spontaneously. in this time, this means an absolute reduction of about 8%, that is about 12 children must be treated with antibiotics to prevent 1 child having some pain after 2 days. There was no effect of antibiotics on complication rates, as measured by subsequent tympanometry or recurrence. However, there were few serious complications seen in these trials: only one case of mastoiditis occurred (in a penicillin treated group). Conclusion: Antibiotics provide a real but small benefit in acute otitis media in children. As most cases will resolve spontaneously, this poses a difficult decision for clinicians when weighed against the possible adverse reactions. For populations or subgroups at risk of suppartive complications such as mastoiditis, e.g., in developing countries, antibiotics would be strongly advised.

Comments

Acute otitis media is a common disease in infants and young children. Though, it has low mortality, it is associated with considerable morbidity including long term sequelae such as hearing loss. Despite, a large number of studies, there is no consensus on the use of antibiotic in the treatment of acute otitis. In the United States, antibiotics are routinely recommended for the treatment of acute otitis(1). On the other hand, antibiotic use for otitis media. in the Netherlands is only 31 %. The aim of this re- view was to determine the role of antibiotic (any versus none) in providing symptoms relief during the acute phase, occurrence of complications such as mastoiditis, and long term sequelae such as hearing loss.

The study found a significant reduction in pain at 2-7 days in the antibiotic group; pain reduction was not significant at day 1. There was no difference in the rates of complications or long term sequelae (hearing loss). Though there was a non-significant trend towards fewer recurrences in the antibiotic group, this effect was counter-balanced by the side effects of antibiotic treatment. As the authors rightly point out, the marginal benefit in pain in a small proportion of patients have to be weighed carefully against the problems associated with increased antibiotic use (primarily emergence of antibiotic resistance). The rate of mastoiditis in the trials reviewed was very low. One expects the rate of suppurative complications such as mastoiditis and chronic otitis to be much higher in India. The prevalence of hearing impairment in school children in India ranges from 11.9 to 34%; the majority of these are due to middle ear disease(2). Therefore, one must interpret the results of this meta-analysis with caution. Information from trials done in developing countries is required before we can come to firm conclusions on the need for antibiotic in acute otitis in this country. Carefully conducted trials (preferably multicentric) on common clinical conditions encountered in children in India are a challenging task but some- thing we cannot shy away from.

Thomas Cherian,
Professor,
Department of Child Health,
 Christian Medical College and Hospital,
Vellore,
Tamilnadu 632 004, India.
 

1. Dowell SF, Marcy SM, Phillips WR, Gerber MA, Schwartz B. Otitis media - Principles of judicious use of antimicro-bialagents. Pediatrics 1998; 101: 165-171.

2. Jacob A, Rupa V, Job A, Joseph A. Hearing impairment and otitis media in a rural primary school in south India. Int J Pediatr Otorhinolaryngol 1997; 39: 133- 138.