Validity and reliability of the european health literacy survey questionnaire (HLS_EU_Q16) in the greek language

1 Introduction

Health Literacy (HL) has been identified as a priority area in primary health care strategies and policies, by both the European Commission and the World Health Organization (WHO) [1]. HL is linked with empowerment and it refers to the capacity of an individual to access and comprehend health information and hence, better handle issues related to heath. Most importantly, HL is a more powerful predictor of health than education, age, income, employment and race [2]. It is also important to note that low HL is clearly associated with to premature death [3] and is a key factor in determining health inequalities in the population [4].

According to WHO, HL is defined as “the cognitive and social skills which determine the motivation and ability of individuals to gain access, to understand and use information, in ways which promote and maintain good health” [5]. A recent definition according to the European Health Literacy Project Consortium (HLS-EU) states that “HL is linked to literacy and entails people’s knowledge, motivation and competences to access, understand, appraise, and apply health information in order to make judgments and take decisions in everyday life concerning healthcare, disease prevention and health promotion, to maintain or improve quality of life during the life course” [6].

Low levels of HL are emerging as a serious public health problem in Europe, where an estimated 30% of Europeans are facing difficulties in understanding essential health-related material and hence making sound decisions with respect to health maintenance and disease management [7]. In a recent, cross-sectional study assessing HL levels using the HLS_EU_Q47, conducted in Attica, Greece, 16.6% and 41.2% of the participants were classified as having insufficient & problematic HL levels, respectively and only 10.9% had excellent HL levels [8].

It has to be underlined that even though HL has been an important topic of research in native English-speaking countries, in parts of Europe were English is not the main language [9, 10], HL is not adequately researched and is not effectively integrated in health policy and practice. A variety of tools exist aiming at measuring different parameters of HL. The Rapid Estimate of Adult Literacy in Medicine (REALM), assessing reading ability and pronunciation [11], the Test of Functional Health Literacy in Adults (TOFHLA), assessing reading, comprehension and numeracy [12], and the Newest Vital Sign (NVS), measuring comprehension and numeracy [13], are the most frequently used instruments for assessing HL.

The European Health Literacy Questionnaire (HLS_EU_Q47) [14] was initially created to assess HL levels in eight European countries, including Greece. The tool contains 47 items in total. Three health sectors: health care, disease prevention and health promotion and four information-processing sectors: accessing, understanding, appraising and applying. It has also been translated and validated in numerous countries around the world over the past 10 years [15, 16]. The same Consortium of European Countries which developed the HLS_EU_Q47, also developed a shorter version, the HLS_EU_Q16. The HLS_EU_Q16 is a 16-item self-report questionnaire which measures the ability of the participant to access, understand, appraise and apply information in health care, disease prevention and health promotion sectors. Correlations of the HLS_EU_Q16 with the index of the long form HLS_EU_Q47 were very high in the HLS_EU study [17]. This shorter version of the questionnaire, has been the instrument of choice for measuring HL levels of the general population, in a number of recent studies around the world after appropriate validation processes, such as in Finland, Germany, Italy & Iceland [18–21]. The HLS_EU_Q16 has also been shown to be an adequate tool for assessing health literacy, in vulnerable groups [22].

The need for comprehensive, reliable HL instruments to be used for health promotion, disease prevention, disease management and in research and clinical practice, is of paramount importance. The current study is set out to test the reliability and validity of the HLS_EU_Q16 questionnaire in the Greek language, as a useful and relatively easier to complete tool, for assessing HL, in comparison to the HLS_EU_Q47.

2 Material and methods

Written approval was obtained by the Greek leader of the HLS_EU project, to use the Greek HLS_EU_Q16, for validating purposes. The HLS_EU_Q16 is a questionnaire that resulted from the analysis of the data of the HLS_EU project, were the HLS_EU_Q47 was firstly used. In the HLS_EU project, 8 European countries participated, including Greece [14]. The 16 questions of the HLS_EU_Q16 were extracted from the original HLS_EU_Q47.

3 Results

Table 1 shows samples’ sociodemographic characteristics and study’s measurements. Four hundred ninety-six participants participated in the study (81.8% women). Median (IQR) age was 36 [13] years and most of the participants were Greek (98.8%), single (52.8%) with education level over Lyceum (85.1%). The majority of the sample had either < 10.000€ or 10.000–20.000€gross annual income (43.8% and 40.5% respectively). Their perceived opinion classified their health status at the “good”, “moderate” and “very good” category (38.7%, 28.8% and 26.8% respectively). One out of two participants did not smoke, three out of four were alcohol consumers and seven out of ten were physically active. The median (IQR) for their BMI was 24.52 (6.70), for HLS_EU_Q16 score was 15 [3] and for NVS score was 5 [2]. According to HLS_EU_Q16 and NVS categories 79.4% and 76.8% were classified to the sufficient HL and adequate HL category, respectively.

One hundred and forty-nine participants completed the questionnaire twice (85.2% women), Test–retest reliability after 15 days for the HLS_EU_Q16 gave Spearman’s correlation coefficient of r = 0.628 (p < 0.0001) which show a high correlation.

Sample’s adequacy was confirmed by the Kaiser-Meyer-Olkin (KMO) measure which was 0.899 and Barlett’s Sphericity Test which was x²(120)  = 2815.18, p < 0.0001, which verified that correlation among the items was satisfactory, so as to proceed to PCA. Table 2 describes the rotated factor loadings of principal component analysis (PCA) for the 16 items of the HLS_EU_Q16. The three extracted factors had eigenvalues > 1 and in combination explained 53.312% of the variance.

Range, mean, standard deviation (SD), min, max and item-total correlation for the HLS_EU_Q16 subscales are presented in Table 3. Cronbach’s a coefficients were calculated to assess internal consistency of the tool (Table 3). Cronbach’s a for the total questionnaire was 0.884 and 0.790, 0.710 and 0.800 for the “Health Care”, “Disease Prevention” and “Health Promotion” subscales, respectively. All item-total correlations were > 0.4.

Table 4 presents correlations between the 3 subscales, where “Health Care” was positively correlated with “Disease Prevention” and “Health Promotion” (r = 0.471 and r = 0.266, p < 0.001 respectively) and “Disease Prevention” was also positively correlated with “Health Promotion” (r = 0.471, p < 0.001), as it was expected.

Table 5 presents the associations between HLS_EU-Q16 and study’s variables. Significant difference was observed between the tool and the categories of marital status and especially married participants had higher levels of HL than unmarried (p = 0.011). On the other hand, non-smokers had lower levels of HL than ex-smokers (p = 0.004). Finally, HLS_EU_Q16 was positively correlated with age, BMI and NVS score (r = 0.188, r = 0.145 and r = 0.267 respectively).

Low levels of health literacy are linked with problematic health behaviors and outcomes [28, 29]. The study was set out to develop and validate the HLS_EU_Q16 questionnaire in the Greek population, as a reliable and relatively quick tool for assessing HL.

The results show that the Greek version of the HLS_EU_Q16 is a valid and reliable instrument for assessing HL levels. Cronbach’s alpha coefficient was≥0.70 for the total scale and the sub-domains, showing that the items of the tool are consistent with each other. As for the item-total correlation coefficients, the values were≥0.40 showing that the discrimination coefficients of all items were sufficient. In two similar studies conducted recently in Iceland and Italy, it was also concluded that HLS_EU_Q16 is a valid and reliable instrument, psychometrically appropriate, with relatively clear factor structure [20, 21]. As for the concurrent validity, the results showed a significant correlation between both measuring instruments (HLS-EU-Q16 and NVS), again similar to the results obtained by others, in the past [25].

In addition, in the current study, the HLS_EU_Q16 showed that 20.6 % of the participants had inadequate and problematic HL levels, contrary to another recent study conducted in Greece using the HLS_EU_Q47, in a representative population of the Attica Greece were 57.8% of the participants were classified in the insufficient and problematic HL category [8]. This could partly be attributed to the relatively high educational level of the participants in the validation study and to the fact that the sample was not representative of the population.

The HLS_EU_Q16 is gradually emerging as the instrument of choice by numerous studies throughout the world, trying to assess HL levels in the different populations [30]. The main advantage with respect to the original HLS_EU_Q47, is that it is considerably shorter in length and relatively easier to complete, making it ideal for assessing HL in less literate or more vulnerable groups of the population. It could also be a very useful tool in assessing HL in large population-based studies, in primary care, for a quick and comprehensive assessment of HL levels.

5 Limitations, strengths, and practical implications

The main limitation of the study was the snowballing method of recruitment which resulted in a relatively large but non representative sample of the general population, with respect to women. With respect to strengths, the study provides the first evidence for the reliability and validity of the HLS_EU_Q16 in the Greek population, using appropriate statistical analysis. It is important to better communicate to primary care doctors, health workers, researchers and policy makers, that low HL is not always immediately apparent at an individual level. Even though socio-economic inequalities, in particular, lower education level and smaller annual income, were negatively associated with HL [8], anyone could potentially have low HL levels. The ability to identify individuals at risk of low HL early, using appropriate tools, is of paramount importance. Hence, we believe that the validation of the HL-16 instrument in Greece, was much needed.