ABSTRACT
OBJECTIVE: We compared single-item (SILS) and two-item (TILS) literacy screeners in predicting Short Test of Functional Health Literacy in Adults (S-TOFHLA) scores. METHODS: Adult hemodialysis patients completed SILS, which determines need for assistance when reading written medical information; TILS (last grade completed and self-reported reading ability); and S-TOFHLA. Receiver operator characteristic curves (ROC) and stratum specific likelihoods were calculated. RESULTS: Of 227 participants, median S-TOFHLA was 24 (IQR 15-34). 129 (55%) participants had adequate, 70 (30%) inadequate, and 37 (16%) marginal health literacy. SILS and TILS predicted S-TOFHLA scores equivalently. Test characteristics predicting inadequate health literacy were: SILS sensitivity for threshold >1, 54% (95%CI: 44, 64), for >2, 39% (29, 49) and specificity for >1, 73% (64, 80), for >2, 93% (87, 97), area under the ROC of 0.67 (0.60-0.74); TILS sensitivity for threshold >1, 72% (62, 80), for >2, 30% (21, 40) and specificity for >1, 54% (45, 63), for >2, 86% (79, 92), area under the ROC of 0.66 (0.59-0.73). CONCLUSION: SILS and TILS had similar test characteristics in predicting S-TOFHLA. PRACTICE IMPLICATIONS: While a positive result on either test increases the likelihood that a patient has low health literacy, the SILS is easier to administer and score.
