Purpose: This study aimed to investigate (1) the prevalence of nonadherence with eye drop treatment; (2) selected correlates of nonadherence at the patient and health-care organization level; and (3) the diagnostic value of the ophthalmologists' ratings, using patients' self-reports as standard.
Methods: This cross-sectional multicenter survey used questionnaires for ophthalmologists and their patients to assess self-reported nonadherence and its correlates. One item, using a 4-point scale [never (ie, adherent) to daily], asked the patients whether they had forgotten to administer eye drops during the past 2 weeks. Ophthalmologists rated their patients as adherent or nonadherent. Nonadherence was also determined by combined methods, whereby either could indicate nonadherence. Given the nested structure of the data, multilevel modeling was used to investigate self-reported nonadherence-correlates. Diagnostic values of ophthalmologists' report were calculated.
Results: Of 663 patients (48% female, 44% >69 years), nonadherence was indicated in 39.2% (n=260) through self-reporting, 2.1% (n=14) through ophthalmologists' ratings, and 40% (n=266) through combined measures. The multivariable, multilevel model showed following significant nonadherence-correlates: Male sex (P=0.01), younger age (P=0.027), and higher-dose frequency (P=0.001). No significant correlation with treating ophthalmologist (P=0.21) could be seen. Yet, the patients visiting their ophthalmologists at least every 3 months were less nonadherent than patients with fewer consultations (P=0.01). The ophthalmologists' report showed a sensitivity and specificity of 3% and 98.5%, respectively.
Conclusions: The prevalence of self-reported nonadherence was congruent with literature. The patients visiting their ophthalmologists at least every 3 months have a lower risk of nonadherence. Ophthalmologist report is an insensitive method for detecting nonadherence.
*Centre for Health Services and Nursing Research, Katholieke Universiteit Leuven
‡Post-doctoral researcher FWO Flanders
§I-Biostat, Katholieke Universiteit Leuven
∥Department of Ophthalmology, University Hospitals of Leuven, Belgium
†Institute of Nursing Science, University of Basel, Switzerland
Supported by an unrestricted research grant from Pfizer Belgium.
Disclosure: The authors declare no conflict of interest.
Reprints: Thierry Zeyen, MD, PhD, Department of Ophthalmology, University Hospitals of Leuven, Kapucijnenvoer 33 Blok C, 3000 Leuven, Belgium (e-mail: Thierry.firstname.lastname@example.org).
Received December 18, 2009
Accepted August 16, 2010