Aims To determine short-term repeatability of the effect of intraocular pressure (IOP) reducing medication and the number of repeated measurements necessary to estimate therapeutic effect with a given degree of precision.

Methods IOP was measured at 8:00, 11:00 and 16:00 h at each of three weekly visits in untreated patients with primary open-angle glaucoma or ocular hypertension. After starting travaprost (0.004%) to both eyes, the measurements were repeated for a further three weekly visits. Repeatability of the change in IOP 1, 2 and 3 weeks after starting treatment was reported as the coefficient of repeatability (CR) and coefficient of variability (CV).

Results Mean reduction in IOP was 7.5 mm Hg (29.9%). CR and CV between visits were 7.8 mm Hg and 37.2%, respectively. Repeated estimates of the effectiveness of treatment in the same subject at the same time of day would, therefore, be expected to lie within a range of 7.8 mm Hg and within ±73.2% of the mean effect with 95% confidence. A reduction in IOP less than 7.8 mm Hg over a single pair of measurements would be indistinguishable from measurement error. Precision improves from ±73.2% for a single pair of readings to ±25.9% for eight pairs.

Conclusions Day-to-day repeatability in the estimation of medication effectiveness is poor even for time-of-day standardised measurements and the number of IOP measurements required to achieve precision adequate for use in clinical practice is large.