Medication Literacy Test for Older Adults: psychometric analysis and standardization of the new instrument.

Abstract

Low medication literacy is prevalent among older adults and is associated with adverse drug events. The Medication Literacy Test for Older Adults (TELUMI) was developed and content validated in a previously published study. To evaluate the psychometric properties and provide norms for TELUMI scores. This was a cross-sectional methodological study with older adults selected from the community and from two outpatient services. Descriptive item-analysis, exploratory factor analysis (EFA), item response theory (IRT), reliability, and validity analysis with schooling and health literacy were performed to test the psychometric properties of the TELUMI. The classification of the TELUMI scores was performed using percentile norms. A total of 344 participants, with a mean age of 68.7years (standard deviation = 6.7), were included; most were female (66.6%), black/brown (61.8%), had low schooling level (60.2%) and low income (55.2%). The EFA pointed to the one-dimensional structure of TELUMI. A three-parameter logistic model was adopted for IRT. All items had an adequate difficulty index. One item had discrimination < 0.65, and three items had an unacceptable guessing index (< 0.35) and were excluded. The 29-item version of TELUMI had excellent internal consistency (KR20 = 0.89). There was a positive and strong association between TELUMI scores and health literacy and education level. The scores were classified as inadequate medication literacy (≤ 10.0 points), medium medication literacy (11-20 points), and adequate medication literacy (≥ 21 points). The results suggest that the 29-item version of TELUMI is psychometrically adequate for measuring medication literacy in older adults.