advertisement

Editors Selection IGR 10-4

Medical Treatment: General management

Comment by Ravi Thomas on:

22276 Predictors of and barriers associated with poor follow-up in patients with glaucoma in South India, Lee BW; Sathyan P; John RK et al., Archives of Ophthalmology, 2008; 126: 1448-1454

Find related abstracts

Several questions arise in the analysis, interpretation and extrapolation of this case-control study by Lee et al. (1592) A case-control design is probably inappropriate for the study question and very prone to biases. The few non-paying patients enrolled despite a large non-paying section suggests that the sample is unlikely to be representative of the hospital's or typical south Indian patients. The fact that selection of patients depended on the availability the social worker and the lack of masking are of concern. The exact questions asked and whether the social worker was specially trained in the 'neutral' administration of questions are also of relevance. Testing large numbers of un-hypothesized predictors can lead to implausible results, such as changing three buses to get to the hospital being predictive of good follow up. It is not clear if power calculations were performed for all the factors investigated or just the three postulated apriori; considering the problems, the lack of association for the latter too needs cautious interpretation. As a baseline, it would be important to know the counseling process in the study hospital. With the usual 'cataract fixated' approach to large volumes of patients in the subcontinent, education, especially for 'non-surgical' glaucoma is not the norm, with counseling usually restricted to 'use these drops and return in six months'. Dissatisfaction with the usual short 'chair times' available for follow up, restricted perhaps to IOP measurement may also be a factor, as is advising unnecessarily frequent follow ups for 'early glaucoma' without field defects. The short time span of the study is also a limitation as far as 'completeness of follow up' is concerned. The solutions to improve follow up have been nicely discussed, but extrapolation of the study is limited, and limited to paying patients in similar hospitals who actually do return for follow up.

• ← Previous
• Next →