[Clinical trials on glaucoma: differences according to the medical or surgical nature of the treatment being evaluated].

PURPOSE

To compare the quality of clinical trials on glaucoma between those evaluating the effectiveness of medical treatments and those evaluating surgical treatments.

METHOD

Clinical trials on glaucoma published in seven international journals between January 1980 and December 1999 were selected. The papers were revised by researchers with a background in epidemiology using a standard qualitative questionnaire. Proportions were compared using Fisher's exact test.

RESULTS

Sample size was pre-estimated in 19% of medical treatment trials and 2% of surgical trials (p=0.005); masking (72% vs. 9%; p<0.001) and intention-to-treat analysis (17 vs. 0 papers; p<0.001) were also more frequent in medical trials. Only 50% of the trials correctly described the patient flow.

CONCLUSIONS

Quality in clinical trials on glaucoma medical treatment was higher than in surgical trials regarding sample size pre-estimation, masking and intention-to-treat analysis. However, both medical and surgical trials should improve in these aspects and in the patient flow description