An Analytical Hierarchy Processing Approach: Developing and Testing a Basic Health Literacy Measure Among Karen Refugees

Background and Purpose: Health literacy is an important component of assessing an individual’s general health and health care utilization (Institute of Medicine, 2004). Currently existing health literacy measures, however, focus on the person’s perception of health literacy, rather than actual level of health literacy. The purpose of this study is to describe the process of developing a basic health literacy measure, using a multiple-criteria methodological approach, called analytic hierarchy processing (AHP). We assess multiple dimensions of health literacy among Karen refugees residing in the U.S.

Method: We conducted a series of focus groups (4 women and 2 men) among Karen refugees to explore health knowledge that is lacking among community members. Based on focus group data, we developed a basic health literacy scale with scores ranging from 0 to 9. Questions include basic knowledge on healthcare use (e.g., You have a blood test scheduled at 8 am tomorrow, when should you stop eating?) or systems knowledge (e.g., You need to recertify Medicaid every __ ). Data were collected from 201 Karen refugees from Burma between ages 18 to 83.

The responses to each item were recorded (0=incorrect; 1=correct). Then we used two scoring systems to create scores. First, we created the original health literacy (HL) scale, representing the number of correct answers to 9 items. Next, the AHP approach was used to create a novel version of the basic health literacy scale. Each item is provided a weight in its overall importance of predicting BHL among this population. First, we established four subscales of HL suggested by the Office of Disease Prevention and Health Promotion including management of chronic conditions, healthcare knowledge, systems knowledge, self-efficacy, and the total score. Next, scores were assigned to each item of the BHL scale according to pairwise comparisons based on a group consensus between five independent coders. Regression analysis was used to assess the effect of each criterion on a set of outcome variables selected based on their predictive ability. Analyses were conducted using Stata v12.

Results: Management of health conditions was associated with a greater likelihood of making an appointment for routine care with a provider (beta=.34, p<.05).

The healthcare knowledge scale was associated with a higher level of reading medication labels (beta=.39, p<.10) and lower cigarettes smoking (beta=-.29, p<.05).

The systems knowledge scale was associated with a greater likelihood of PCP visits (beta=1.63, p<.05) and routine appointments with a provider (beta=.60, p<.001).

The self-efficacy scale was associated with a lower likelihood of questioning a doctor’s advice (beta=-.98, p<.001).

Conclusion and Implication: Results point to the importance of improving the management of health conditions and systems knowledge to increase PCP visits and routine appointments among Karen refugees. Additionally, improvements in healthcare knowledge can increase medication adherence and assist with smoking cessation efforts. Finally, self-efficacy plays an important role in treatment adherence among Karen populations. The four newly created criterion scales can be compared to the original basic health literacy scale to test for their effectiveness in measuring healthcare use among Karen refugees.