There are two levels at which the efficacy of glaucoma therapy can be determined: first, the therapeutic trial and second, the experience gained when treating individual patients in the daily practice. At each level, various difficulties arise. At either level, however, the main problem is that there is no way to check on patient compliance. A trial on the efficacy of glaucoma therapy needs to be thoroughly planned and performed, correct statistical methods must be used, and the therapy must be critically evaluated a review of studies on glaucoma therapy for the years 1988 and 1993 showed the difficulties in planning and enforcing a correct study design. Most studies suffer from lack of adaptation to the characteristic features of primary open-angle glaucoma, the most important of all being its chronic course with minimal changes over many years once pressure is regulated. Thus, long-term studies are mandatory with simultaneous assessment of intraocular pressure (IOP), the visual field and the morphology of the optic nerve head (ONH). Measurement of the IOP is the parameter that can be most reliably checked on. Increased IOP represents the main cause for glaucoma damage-not the damage itself. For the early detection of glaucoma, morphometry of the ONH is the most reliable method, but it is still quite rough. The use of perimetry for early detection and follow-up of glaucoma is overestimated. Given the above-mentioned characteristic features of the disease, sufficient damage with corresponding perimetric and morphometric changes does not occur in the short time periods chosen for therapeutic trials. The future will show whether the blood supply of the ONH will be of primary interest in glaucoma detection and control of treatment efficacy. New and promising methods for measurement of the blood supply of the ONH are emerging. Last but not least, control of efficacy in glaucoma treatment should include cost-effect analyses. Such studies are almost nonexistent at present.