SP723VALIDATION OF A 3-ITEM HEALTH LITERACY SCREENER IN A MULTIETHNIC NEW ZEALAND DIALYSIS POPULATION

Abstract

Introduction and Aims: Poor treatment adherence is an important barrier to successful chronic disease management. A recent systematic review showed a statistically significant relationship between health literacy and treatment adherence (Zhang et al. Annals of Pharmacotherapy 2014, Vol. 48(6) 741-751). However, traditional health literacy screening methods (e.g. the Rapid Estimate of Adult Literacy in Medicine, REALM and the Short Test of Functional Health Literacy in Adults, S-TOFHLA) are slow to administer (REALM 66 items, S-TOFHLA 36 items), can only be administered face-to-face, and can be associated with patient embarrassment. Brief, quickly administered, and flexible instruments are required to allow clinicians to efficiently assess the health literacy level of their patients, and tailor their health services delivery accordingly. Methods:We evaluated 3 single-item health literacy screeners in 100 randomly selected patients from a large multi-ethnic New Zealand dialysis cohort, using items that have previously been validated against gold standard in general populations or to a limited degree in dialysis populations: Q1. How confident are you at filling out medical forms by yourself? Q2 How often do you have someone (like a family member, friend, hospital/clinic worker or caregiver) help you read hospital materials? Q3 How often do you have problems learning about your medical condition because of difficulty understanding written information? Each item is scored on a Likert scale (1-5), with 3 or above on each scale representing poor health literacy.We surveyed patients in person using translators if necessary, either in dialysis facilities or their homes.We performed factor analysis to determine the contribution of each item to the measurement of the health literacy construct, and scale reliability analysis to determine internal consistency and whether any item should be dropped. Results: 62 participants were males, 38 females; mean age 55.4 (range 17-87) years; 67 on HD (mean Kt/V 1.4), 33 on PD (mean Kt/V 1.82); mean age-adjusted Charlson Co-morbidity Index 5.32 (range 2-10). 35 participants self-classified as NZ Maori, 35 as Pacific People, and 30 were either Indian, other Asian or NZ European. 50 participants did not have secondary education. 39 participants had trade certificates, postgraduate and professional type qualifications. Factor analysis showed loadings (before rotation) of 0.71, 0.77 and 0.74 for Q1, Q2, Q3, respectively, suggesting each of the 3 items are contributing equally strongly to the measurement of the health literacy construct. Overall, the Cronbach's alpha score for the combined 3 single-item scale was 0.78, providing evidence of strong internal consistency. The item-to-total statistics show that all 3 items contribute to Cronbach's Alpha, and that dropping any one of them would reduce the overall score (Q1 - 0.703, Q2 - 0.722, Q3 - 0.686). Conclusions: Our results provide strong support for a combined 3 single-item instrument for detection of poor health literacy, using an average of the 3 items. Of note, 43% of the participants in our sample were classified as having poor health literacy using the combined instrument, with a mean score 2.63. In conclusion, poor health literacy is common, and best detected through a combined 3 single-item instrument.