Senior researcher Patient-centered Care
Publicatie
Publication date
Health Literacy in Europe: the development and validation of health literacy prediction models
Heide, I. van der, Uiters, E., Boshuizen, H., Rademakers, J. Health Literacy in Europe: the development and validation of health literacy prediction models European Journal of Public Health: 2015, 25(spl. 3), p. 184. Abstract: 8th European Public Health Conference: "Health in Europe - from global to local policies, methods and practices". 14-17 oktober 2015 in Milan.
Introduction
Health literacy is considered an important determinant of health disparities. It is therefore important to have insight into health literacy skills of the general population within countries. Little is known on the health literacy skills of the general population in EU member states. The aim of our study was to examine whether census data can be used to provide a reliable indication of health literacy skills on population level in EU member states.
Methods
Dutch data derived from the European Health Literacy Survey (HLS-EU) and the Adult Literacy and Life Skills Survey (ALL) were used to develop prediction models to predict subjective (self-reported) and objective (tested) health literacy. The HLSEU data includes a subjective measure of health literacy and the ALL an objective measure. Stepwise linear regression analyses were applied to build one model to predict subjective health literacy and one to predict objective health literacy based on a random 2/3 of the Dutch HLS-EU data and ALL data. The models were validated based on the remaining 1/3 of both datasets and on HLS-EU data from seven other EU countries.
Results
Level of education, age, sex, income, working status and urbanization were included as predictors in the models. In the final models that included merely significant predictors (p < .05), education was an important predictor of subjective as well as objective health literacy. Age and working status significantly predicted objective health literacy. Sex and income significantly predicted subjective health literacy. The prediction models provided a reliable indication of health literacy when applied in the same population. The reliability of the models in other EU countries varied per country.
Conclusions
Census data can provide a reliable estimation of national health literacy levels in most countries. The prediction models can be used to provide an indication of health literacy when health literacy measures are absent.
Health literacy is considered an important determinant of health disparities. It is therefore important to have insight into health literacy skills of the general population within countries. Little is known on the health literacy skills of the general population in EU member states. The aim of our study was to examine whether census data can be used to provide a reliable indication of health literacy skills on population level in EU member states.
Methods
Dutch data derived from the European Health Literacy Survey (HLS-EU) and the Adult Literacy and Life Skills Survey (ALL) were used to develop prediction models to predict subjective (self-reported) and objective (tested) health literacy. The HLSEU data includes a subjective measure of health literacy and the ALL an objective measure. Stepwise linear regression analyses were applied to build one model to predict subjective health literacy and one to predict objective health literacy based on a random 2/3 of the Dutch HLS-EU data and ALL data. The models were validated based on the remaining 1/3 of both datasets and on HLS-EU data from seven other EU countries.
Results
Level of education, age, sex, income, working status and urbanization were included as predictors in the models. In the final models that included merely significant predictors (p < .05), education was an important predictor of subjective as well as objective health literacy. Age and working status significantly predicted objective health literacy. Sex and income significantly predicted subjective health literacy. The prediction models provided a reliable indication of health literacy when applied in the same population. The reliability of the models in other EU countries varied per country.
Conclusions
Census data can provide a reliable estimation of national health literacy levels in most countries. The prediction models can be used to provide an indication of health literacy when health literacy measures are absent.
Introduction
Health literacy is considered an important determinant of health disparities. It is therefore important to have insight into health literacy skills of the general population within countries. Little is known on the health literacy skills of the general population in EU member states. The aim of our study was to examine whether census data can be used to provide a reliable indication of health literacy skills on population level in EU member states.
Methods
Dutch data derived from the European Health Literacy Survey (HLS-EU) and the Adult Literacy and Life Skills Survey (ALL) were used to develop prediction models to predict subjective (self-reported) and objective (tested) health literacy. The HLSEU data includes a subjective measure of health literacy and the ALL an objective measure. Stepwise linear regression analyses were applied to build one model to predict subjective health literacy and one to predict objective health literacy based on a random 2/3 of the Dutch HLS-EU data and ALL data. The models were validated based on the remaining 1/3 of both datasets and on HLS-EU data from seven other EU countries.
Results
Level of education, age, sex, income, working status and urbanization were included as predictors in the models. In the final models that included merely significant predictors (p < .05), education was an important predictor of subjective as well as objective health literacy. Age and working status significantly predicted objective health literacy. Sex and income significantly predicted subjective health literacy. The prediction models provided a reliable indication of health literacy when applied in the same population. The reliability of the models in other EU countries varied per country.
Conclusions
Census data can provide a reliable estimation of national health literacy levels in most countries. The prediction models can be used to provide an indication of health literacy when health literacy measures are absent.
Health literacy is considered an important determinant of health disparities. It is therefore important to have insight into health literacy skills of the general population within countries. Little is known on the health literacy skills of the general population in EU member states. The aim of our study was to examine whether census data can be used to provide a reliable indication of health literacy skills on population level in EU member states.
Methods
Dutch data derived from the European Health Literacy Survey (HLS-EU) and the Adult Literacy and Life Skills Survey (ALL) were used to develop prediction models to predict subjective (self-reported) and objective (tested) health literacy. The HLSEU data includes a subjective measure of health literacy and the ALL an objective measure. Stepwise linear regression analyses were applied to build one model to predict subjective health literacy and one to predict objective health literacy based on a random 2/3 of the Dutch HLS-EU data and ALL data. The models were validated based on the remaining 1/3 of both datasets and on HLS-EU data from seven other EU countries.
Results
Level of education, age, sex, income, working status and urbanization were included as predictors in the models. In the final models that included merely significant predictors (p < .05), education was an important predictor of subjective as well as objective health literacy. Age and working status significantly predicted objective health literacy. Sex and income significantly predicted subjective health literacy. The prediction models provided a reliable indication of health literacy when applied in the same population. The reliability of the models in other EU countries varied per country.
Conclusions
Census data can provide a reliable estimation of national health literacy levels in most countries. The prediction models can be used to provide an indication of health literacy when health literacy measures are absent.