Applicability of Cochrane reviews in a developing country setting- a pilot study

Article type
Authors
Mathew J, Mathew P, Singh M
Abstract
Background: Cochrane reviews are the best resource for evidence concerning clinical interventions; however most reviews are generated by and for people in developed countries.

Objectives: To assess the applicability of Cochrane reviews in a developing country setting.

Methods: Respiratory conditions necessitating hospitalization of children in India are pneumonia, tuberculosis, bronchiolitis, asthma, empyema effusion and cystic fibrosis. Cochrane reviews pertaining to these were identified using MEDLINE. MEDLINE was chosen because unlike The Cochrane Library it is accessible free of cost. Thus this method simulates the approach used by professionals in India. 'Reviewers' Conclusions' of all identified reviews were listed and shown to ten paediatricians after concealing authorship and identification details. The paediatricians were asked to categorize the applicability of each review in clinical practice as per Table 1. The number of reviews (R) multiplied by paediatricians (P) was calculated as the denominator (RP). Total responses in each of the categories A to G were divided by RP and expressed as a percentage. The sum of categories A and B was defined as 'Applicable reviews' and the sum of categories C to G 'Non-applicable reviews'.

Results: There were 242 Cochrane reviews for the clinical conditions selected. Of these, 102 were excluded for being either earlier versions of updated reviews (n=93) or pertaining exclusively to adults (9). Of 140 reviews analyzed among 10 paediatricians (RP=1400); 35.1% were 'Applicable reviews', while 64.9% were 'Non-applicable reviews'. Among the latter, the major reason (35.0%) was inconclusive review (Category G). Other reasons were: non-availability of intervention (9.9%), non-feasibility (9.4%), prohibitively expensive (5.5%) and unacceptable to patients (5.1%).

Conclusions: Only a small fraction of existing Cochrane reviews is applicable in the developing country setting. About one-third are not applicable while a further third are inconclusive and hence not applicable. The need of the hour is to identify clinical questions pertaining to interventions relevant to developing countries; thereby generating applicable Cochrane reviews.