Purchase this article with an account.
Jaya Badhwar, Hayden Bosworth, Betsy Sleath, Susanne Danus, Logan Christensen, Kelly W Muir; Evaluation of two metrics for identifying poor glaucoma medication adherence compared to objective adherence measurements. Invest. Ophthalmol. Vis. Sci. 2014;55(13):5549.
Download citation file:
© ARVO (1962-2015); The Authors (2016-present)
Poor adherence to the prescribed glaucoma medication regimen may lead to preventable blindness. Previous work suggests that providers may not accurately identify nonadherent patients. The purpose of this project was to compare 2 screening methods for identifying nonadherence on the criteria of feasibility and correlation with objective adherence measurements from an electronic monitor.
Patients from 2 eye clinics (a university practice and a VA hospital) with medically-treated glaucoma were given a survey of medication adherence including the question “how many times did you miss your drops in the last week” and a visual analog scale (VAS) on which patients mark on a line how much of the time they take their drops as prescribed. Participants received electronic monitors to record when the medication was accessed until the next clinic visit. The 2 measures were evaluated for feasibility (how many participants responded), sensitivity and specificity for detecting nonadherence.
Adherence was defined as the proportion of prescribed doses taken according to the monitor. Of 142 participants, 132 returned the monitors, average duration of use 179 days (SD 69, median=179). Mean percent of the prescribed doses taken was 83.3 (SD 23, median 94). For the question, “how many times did you miss your drops last week?”, 23 failed to answer; 100 responded that they had not missed any drops in the last week. 2 failed to complete VAS, 124 responding with <100% adherence. In regards to detecting adherence of <80%, a VAS score <1 was 90% sensitive and 13% specific. A response of >1 on the item “how many times did you miss your drops in the past week” was 26% sensitive and 88% specific for detecting nonadherence.
More participants responded to the VAS than to the question regarding missed drops. Possible explanations include that the VAS is easier to interpret or is associated with less perceived stigma than admitting to a specific number of missed doses. The VAS was more sensitive at detecting poor adherence than the question regarding missed drops, but less specific. The low specificity of the VAS suggests that although the tool may be a useful screener for nonadherence, clinicians must delve more deeply if the screen is positive to determine the nature of the patient’s difficulty with the prescribed regimen.
This PDF is available to Subscribers Only