Osborne, R.H., Batterham, R.W., Elsworth, G.R., Hawkins, M., & Buchbinder, R. (2013). The grounded psychometric development and initial validation of the Health Literacy Questionnaire (HLQ). BMC Public Health, 13:658. https://bmcpublichealth.biomedcentral.com/articles/10.1186/1471-2458-13…
The Health Literacy Questionnaire (HLQ) is now one of the most widely used health literacy measures in the world. Over 800 studies are using the HLQ across diseases, settings and countries.
Health literacy is multidimensional therefore we developed a series of independent measures to measure it. It is used for surveys, trials, PREM/PROM assessments, and evaluation over time. Each scale efficiently measures one of nine aspects of health literacy (each scale with 4 to 6 items). You can use one of the scales to measure Functional Health Literacy (with a total of 5 items, akin to a screener). It also measures eight other elements to ensure you measure the full construct of health literacy. Consequently, a short version is not encouraged, many groups use 2 or more health literacy scales (aspects of health literacy), measuring there concepts on interest well and specifically.
The HLQ was developed and tested over many years using the contemporary theoretical validity testing approach. It was designed for and is used in national surveys, clinical trials, general surveys, for quality improvement, evaluation studies, and, importantly, to uncover mechanisms behind health inequalities to inform intervention development. It has been found to be useful at the patient-clinician level. Each scale is reliable and generates key information about an individual's perceived health literacy abilities, and their experiences. It was designed to be sensitive to change, identifying small differences between populations, and changes over time. These characteristics have been repeatedly demonstrated. As each scale is independent, only one or some of the 9 scales need to be used to answer your specific research questions or evaluate specific outcomes. To measure the full multidimensional concept of health literacy, all nine scales are required.
The HLQ is a generic scale. This is deliberate because these days the majority of people have more than one disease - and it is not really possible for respondents to just focus on one part of their life. We recommend the use of disease-specific scales with the HLQ. The HLQ also assists teams to take a strengths-based approach, that is identifying individuals' and groups' mix of strengths, challenges, and preferences. This means that HLQ users not only get rich information on what is really going on for respondents but also what can be done to help them in the areas of health, access, and equity.
Each year, independent validity testing studies are undertaken in a range of cultures and languages. These studies are generally highly robust and consistently demonstrate that the original 9-scale structure holds and is reproducible. This level of reproducibility is rare in the patient-reported outcomes field. See this paper for a critique of if the field: https://gh.bmj.com/content/7/9/e009623
Independent scales measuring health literacy on a continuous scale. Scores range between 1 to 4 (for first 5 scales) and 1 to 5 (for scales 6 to 9).
Items were generated through grounded processes using Concept Mapping, which systematically captures the views of the target populations. Several populations were consulted: general population, patients, healthcare providers and policymakers. Items were derived directly from consumer's statements. Items were cognitively tested, then tested with 2 diverse populations (calibration and replication samples). Extensive consultation with practitioners further provided evidence of content validity. Content has been confirmed in a wide range of settings and countries.
The nine HLQ scales capture peoples' lived experiences of trying to understand, access and engage with health information and health services and are unique aspects of health literacy. Consequently there are few direct comparisons with other scales. One of the nine scales, scale 9. "Understanding health information well enough to know what to do", is most highly correlated with common functional scales such as the TOFHLA or NVS.
In most settings, alpha > 0.8 for most scales. Due to the known weaknesses of Cronbach's alpha (ie that a high number of items results in artificially high alpha, and that its calculation is based on a normal distribution of the data and equidistance between response options), we recommend calculation of composite reliability. In most settings, the composite reliability is above 0.8 which means it is suitable for research and development purposes.
