https://orcid.org/0000-0002-8374-1589
Figures
Abstract
Background
Health literacy has a direct impact on the health of populations. It is related to education, capacity for self-care, and management of health resources. The Health Literacy Survey Questionnaire HLS-Q12 is one of the reference instruments but has not yet been adapted to Spanish. The aims of the study were to cross-culturally adapt and evaluate the psychometric properties of the Spanish version of the HLS-Q12.
Methods
Data was collected from June 2020 to March 2022. The sample consisted of 60 patients who initiated cancer treatment for the first time within a clinical trial. Double direct translation, back-translation, cognitive debriefing with a 10-patient sample, and an expert committee were used for cross-cultural adaptation. For validation of the HLS-Q12, a psychometric analysis was performed to assess feasibility, reliability, sensitivity to change and construct validity with other measures such as health-related quality of life, empowerment, and health needs.
Results
The HLS-Q12 is equivalent at the semantic, conceptual, and content level to the original version and its psychometric properties demonstrated good internal consistency with a Cronbach’s alpha of 0.88 and a McDonald´s omega of 0.91, a high degree of fit for the confirmatory factor analysis, and a statistically significant sensitivity to change (p = 0.025).
Conclusions
Based on robust psychometric values, the Spanish version of HLS-Q12 was found to be a good cross-culturally adapted tool for collecting correct information on health literacy in cancer patients regardless of tumour type or stage. Although more studies are needed, this version of HLS-Q12 could be used in research for collecting data on the health literacy needs of Spanish-speaking patients.
Citation: Muñoz-Villaverde S, Serrano-Oviedo L, Martínez-García M, Pardo Y, Tares-Montserrat L, Gómez-Romero FJ, et al. (2024) Spanish version of the short European Health Literacy Survey Questionnaire HLS-Q12: Transcultural adaptation and psychometric properties. PLoS ONE 19(2): e0299736. https://doi.org/10.1371/journal.pone.0299736
Editor: I Gede Nyoman Mindra Jaya, Padjadjaran University: Universitas Padjadjaran, INDONESIA
Received: December 27, 2022; Accepted: February 14, 2024; Published: February 29, 2024
Copyright: © 2024 Muñoz-Villaverde et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the manuscript and its Supporting information files.
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Introduction
In recent years health literacy (HL) has received increasing attention from the scientific community. It allows individuals to manage their own health by enhancing self-care capacity, decision-making skills, and active participation in both individualised care and health programmes and therefore has a broad impact on health care systems and public health [1–3]. From this perspective, the World Health Organization (WHO) defines HL as "the social and cognitive skills that determine a person’s level of motivation and ability to access, understand and use information in ways that enable them to promote and maintain good health" [4].
In 2015 the European Health Literacy Survey Project (HLS-EU) estimated that 12.4% of the European population had inadequate literacy levels and 35.2% had marginal levels [5]. By contrast, the same study showed that 7.5% of Spanish citizens had inadequate HL levels and 50.8% were marginal [6]. The HLS-EU Project created the HLS-EU-Q86 survey for the measurement of HL. This questionnaire assesses four dimensions of HL (access, understanding, evaluation, and application of health information) that are divided into three different domains (health promotion, disease prevention, and disease care). Later versions such as the HLS-EU-Q47 and others such as the HLS-Q16, HL-SF12, and HLS-Q12 followed [6]. However, the HLS-Q12 does not have an adapted and validated Spanish version [7].
Very few health questionnaires measure HL. The availability of tools that measure HL through easy-to-use and quick-to-administer questionnaires will facilitate the construction of formative education strategies with the aim of increasing patient independence. This is especially true for those with chronic diseases whose personal involvement in the care process is essential to increase the likelihood of adherence to pharmacological therapy and successful disease management [8]. The purpose of this study was to conduct a cross-cultural adaptation and validation of the Spanish shortened version of the HLS-Q12.
Methods
Study design and participants
A quantitative psychometric study with a descriptive, longitudinal design and a cross-cultural adaptation was conducted. The study population consisted of patients initiating cancer treatment at the oncology unit of a tertiary hospital in the city of Barcelona. Inclusion criteria were: a) agreement to written informed consent and a follow-up call via telenursing; b) patients >18 years of age; c) patients with solid tumours at any stage of the disease; and d) patients who had not previously received oncological treatment as part of a clinical trial. Exclusion criteria were: a) lack of remote connection devices; b) inability to participate in the study due to clinical situation (i.e., scores of 2–3 on the Eastern Cooperative Oncology Group (ECOG) performance scale); c) patients with cognitive impairment; d) a level of schooling that did not allow completion of the questionnaires; and e) a language barrier that made it difficult to complete the questionnaires.
For sample size, we considered the number of items that form part of the scale, accepting that the number of subjects for the sample should be between 2 and 10 times the number of items. Based on this, 60 subjects were sufficient for the study [9–14]. Loss-to-follow-up rate of 25% was estimated. Data was collected from June 2020 to March 2022.
Measurements
Sociodemographic variables included age, gender, marital status, level of education, and whether there were cohabitants at home. Clinical variables included trial phase, tumour type and stage, presence and number of comorbidities and degree of independence. Management of oncological symptomatology was quantified by recording and evaluating expected outcomes and interventions according to the hospital’s individualized care plan for cancer patients. CTCAE v5.0 (Common Terminology Criteria for Adverse Events) was used to measure the different degrees of adverse events (G1-G5) related or not to treatment produced by chemotherapeutic, radiotherapeutic, or immunotherapy agents. Among the events collected we highlighted anorexia, nausea, xerostomia, mucositis, diarrhoea, constipation, pain, and fatigue. Pain and fatigue symptoms were measured using the categorical pain scale (no pain-unbearable) and the numerical fatigue scale (0-absence of fatigue, 10-worst possible fatigue), respectively.
HL was measured using the HLS-Q12, a 12-item scale, which measures HL on a Likert scale from 1 (very difficult) to 4 (very easy). The theoretical range is from 12 to 48 points, the higher the score, the higher the HL [7]. The full Spanish HLS-Q12 instrument can be found in S1 Appendix.
Two questionnaires were used to assess the health-related quality of life (HRQoL) of patients: one by the European Organization for Research and Treatment of Cancer (EORTC QLQ-C30) [15,16] and the scale designed by the American Eastern Cooperative Oncology Group (ECOG) [17] and validated by WHO. Empowerment was assessed using the Patient Empowerment in Long-Term Conditions questionnaire (PELC) [9,18] and the Holistic Needs Assessment (HNA) [19,20] to identify individual patient needs.
The 30-item EORTC QLQ-C30 scale incorporates 5 functional dimensions (physical functioning, activities of daily living, emotional functioning, cognitive functioning, and social relationships), 3 symptom scales (fatigue, nausea and vomiting, pain), a global health status scale, and several individual items, to assess additional symptoms commonly reported by cancer patients. All measures range in score from 0 to 100. A high score for the functional scale represents a high/healthy level of functioning, a high score for the global health status represents a high HRQoL, but a high score for a symptom scale represents a high level of symptomatology/problems [15,16].
The ECOG is a hetero-administered scale that assesses the evolution of the patient’s abilities in daily life while maintaining maximum autonomy, and its results help to guide therapeutic decisions and the prognosis of the disease. The ECOG is scored from 0 to 5 (normal to death, respectively) [17].
PELC is a self-administered questionnaire that measures empowerment in chronically ill patient and contains 47 items that are scored on a Likert scale from 1 (strongly disagree) to 5 (strongly agree). The scale ranges from 47 to 235, with higher scores indicating higher levels of empowerment [9,18].
The HNA Tool comprises a self-assessment of health needs of patients living with cancer through a simple questionnaire. It measures the physical, practical, emotional, spiritual, social, socio-economic, and environmental need of individuals. A higher number of marked needs indicates a higher number of concerns about the disease process [19,20].
Data collection
Five visits (V) were made: the first before treatment initiation (V1); 24 hours (V2) and 10 days (V3) after treatment initiation; completion of the educational intervention but before starting a new treatment cycle (V4); and 3 months after enrolment (V5). All patients received an educational intervention with a nurse via synchronous teleconsultation which consisted of information about the clinical trial, resolving doubts about disease process, and providing health education on the warning signs and symptoms of adverse effects of the trial treatment. Both patients and their relatives were informed with the intention of empowering the patient.
Data on sociodemographic and clinical variables were collected at V1. The PELC, EORTC QLQ-C30, HNA, and HLS-Q12 instruments were administered to assess validity and internal consistency of the HLS-Q12. Clinical variables were collected in V2 and V3 and questions were answered based on HNA results. In V4, clinical variables and the PELC, EORTC QLQ-C30, and HNA questionnaires were collected. In V5, information was collected, queries were resolved according to patients’ needs, and the treatment status was documented. The self-administered questionnaires were sent by e-mail via Microsoft forms with a mandatory response design. Sociodemographic and clinical data were collected using Research Electronic Data Capture (REDCap). Anonymized study data can be found in S2 Appendix.
Cross-cultural adaptation process
Permission was received from the author of the original English language version of the HLS-Q12. The aim of the adaptation process is to ensure that the instrument is semantically, conceptually, and content-wise equivalent to the original version. Semantic equivalence seeks to obtain same meaning for each of the items, conceptual equivalence ensures the questionnaire measures same theoretical construct in both cultures, and content equivalence proves each item has relevance for both cultures [18,21]. The adaptation was carried out taking into account the methodology recommended by the International Society for Pharmacoeconomics and Outcomes [22]: 1) forward translation; 2) reconciliation and synthesis of the translations; 3) back translation into English; 4) comparison and harmonization of the back translations with the original; 5) cognitive debriefing; 6) review of the cognitive interviews by the committee of experts; 7) reading test, spelling and grammar check; and 8) drafting of the process report.
Psychometric properties and data analysis
Sociodemographic and clinical data were collected and expressed as absolute and relative frequencies with means and standard deviations (SDs). The statistical package R software version 4.1.0 was used, and statistically significant values were set to p<0.05.
For psychometric analysis, feasibility, reliability, construct validity, and sensitivity to change were considered [23]. Feasibility was measured by recording the time needed to complete the questionnaire, as well as the descriptive characteristics of the questionnaire items (ceiling and floor effects). To assess relevance of each item, an item-total correlation analysis was performed. It recommended that values ≤0.2 be discarded or reformulated as they represent an insufficient level of homogeneity, items should be at least ≥0.3, and levels >0.4 are considered very good [24].
Reliability was analysed by internal consistency and assessed with Cronbach’s alpha and McDonald’s omega, both have values between 0–1 with higher values being more reliable [25–28] Cronbach’s alpha was also considered if the item was removed; this index shows how reliability improves when the item is removed from the scale, allowing identification of items affecting internal consistency.
For construct validity [29], instrument structure was studied by performing a confirmatory factor analysis (CFA). CFA parameter estimation was carried out using structural equation modelling and the weighted least squares means and variance adjusted (WLSMV) model for calculating estimators as it provided the best option for working with categorical or ordered data [30]. The estimates of latent variables and variances are assumed to take values from -1 to 1 [31]. Absolute fit indicators were considered with chi-square (χ2), SRMR (standardised root mean residual), and RMSEA (root mean square error of approximation) tests. These indicators are considered adequate for values >0.05 for χ2, <0.05 for SRMR, and <0.06 for RMSEA [32–34]. Incremental fit indicators utilised were AGFI (Adjusted Goodness of Fit Index), BBNFI (Bentler Bonnet Normed Fit Index), BBNNFI (Bentler Bonnet Non-Normed Fit Index), CFI (Comparative Fit Index), and TLI (Tucker-Lewis Index). These indicators range from 0 (null fit of the model to the data) to 1 (perfect fit); values close to 0.90 are considered adequate [34–36]. Normalised χ2 was used for parsimony fit indicators which establishes the ratio between χ2 and the number of degrees of freedom (χ2/gl). Values <2 are considered acceptable [32–34].
Convergent-divergent validity of the constructs was by performed with a Pearson correlation matrix between HLS-Q12, the rest of the V1 questionnaires, and different clinical variables. The correlation (r) was interpreted as low (≤0.29), moderate (0.30–0.49), and high (>0.50) [37]. For convergent validity, we hypothesised that HLS-Q12 is related to PELC, EORTC-QLQC30 (functioning, symptomatology, and global health), ECOG before intervention, and the number of comorbidities [38–40]. Acceptable convergence is considered when significant correlations p<0.05 were present [26,41]. Regarding divergent validity, we hypothesised that HLS-Q12 has different constructs with HNA, and the clinical variable related to the management of symptomatology in oncological processes [38–40]. Divergence is accepted when a non-significant value p>0.05 is expressed [26,41].
Sensitivity to change was assessed by comparing means of pre-post HLS-Q12 scores, defined by an educational intervention through teleconsultation consisting of five fixed visits plus a series of on-demand consultations (V1 and V4 in the total sample and in two subsamples that we differentiated in the study; stable patients ECOG 0 or experiencing an improvement in ECOG, from 1 to 0, and non-stable patients, who did not show clinical improvement). Cohen’s effect size was also calculated. The effect size (ES) values used were considered as either large (>0.8), moderate (0.5), or small (0.2) changes [29,34,42].
Ethics approval and consent to participate
The project has been approved by the Clinical Research Ethics Committee of the Parc de Salut Mar (No. 2020/9408/I). An information sheet and written informed consent were provided so that participants could be aware of the objectives, the purpose of the study, and could be informed of how their data was managed. At the time of inclusion, they were assigned an identification code consisting of the initials of the recruitment centre and a correlative number.
Results
Cross-cultural adaptation of the HLS-Q12
The Spanish HLS-Q12 is presented in Table 1.
A double direct translation of the English questionnaire was completed independently by two bilingual translators whose mother tongue was Spanish and who came from different cultural contexts within the Spanish territory. Both translations were reconciled and synthesised into a single document by a committee of experts yielding a first Spanish version. An expert translator in the healthcare field whose mother tongue was English was contracted to back-translate the document into English. Both English versions were compared, and the synthesised document was sent to the original author who suggested modifications to some of the items, and these opinions were considered by the authors of the Spanish translation who harmonised a second Spanish version.
A cognitive debriefing was conducted via semi-structured and individual interviews with a heterogeneous sample of 10 patients (5 men and 5 women) undergoing cancer treatment who gave their opinion on how the questionnaire was constructed, its ease of use, the type of format, and the brevity and clarity of the questions. 40% of the patients had lung cancers, 40% had genitourinary cancers, 10% had breast cancer, and 10% had digestive cancer. Mean age of the participants was 63.9±10.5 years. A 10% rate of incomprehension was obtained, and participants made suggestions for improvement. Of the 12 questions, two were identified (items 2 and 12) for which participants expressed difficulty in understanding and answering. They were asked for alternative wording and provided ideas to improve their understanding. Of the 10 participants, 30% had marginal, 60% had intermediate, and 10% had advanced levels of HL. The same expert committee reviewed the results of the cognitive interviews, a reading test, and a spelling and grammar check, and then a consensus was reached on the final version of the Spanish questionnaire.
Psychometric analyses of the HLS-Q12 questionnaire
Characteristics of the participants.
Table 2 describes the characteristics of the total sample consisting of 60 patients and the two subsamples that were used for the analysis of sensitivity to change, 47 stable patients (78.3%) which ECOG was 0 or experienced an improvement in ECOG, from 1 to 0, and 13 non-stable (21.7%) or worsening patients, who did not show clinical improvement.
Feasibility.
The average time needed to complete the HLS-Q12 was 4.1±3.9 min (range 1–19 min). Since the questionnaires had an online design with compulsory response, all items were completed for HLS-Q12, QLQC30, PELC, and HNA. In Table 3, it can be observed that scores were distributed over the theoretical range, that the floor effect for the symptomatology dimension of QLQC30 (13.33%), and Dimension (D) 1 (15%), D2 (26.67%), D3 (20%), D4 (45%), and D5 (65%) of HNA obtained the highest percentages and that the questionnaires that obtained the highest ceiling effect percentages were the functioning dimension of the QLQC30 (8.33%), and D3 (5%), D4 (10%), D5 (8.33%), and D6 (5%) of HNA.
Descriptive analysis of the HLS-12 items.
The frequency of responses for each item, the values of central tendency and dispersion, and the item-total correlation of the HLS-Q12 are shown in Table 4. 75% of items showed a response rate >50% in some of their categories (items 2, 3, 4, 6, 7, 9, 10, 11, 12). All items obtained a correlation >0.51 indicating a high degree of homogeneity and strong association between the item and the total score.
Reliability.
Cronbach’s alpha scores indicated a good internal consistency in HLS-Q12 showing values of 0.88. McDonald´s omega showed values of 0.91.
In addition, Table 5, shows that when the alpha values of some of the items were removed it did not fall below 0.86 indicating a good relationship of the item with the total scale.
Construct validity.
Confirmatory factor analysis. CFA results concluded that HLS-Q12 is a one-dimensional model (Fig 1). Table 6 shows the weight of each of the items in the WLSMV model tested in the study. Most of the items have values >0.4 indicating that all items had a relevant contribution to the model.
f1, single factor model.
Goodness of fit of the model. The χ2 value was significant (p = 0.017), indicating that the hypothesis of a perfect model should be rejected. RMSEA showed a reasonable fit, and although the SRMR did not reach the cut-off value, it was very close. The χ2/gl ratio and the incremental measure indices (AGFI, BBNFI, BBNNFI, CFI, and TLI) gave acceptable values. The model offers a high degree of fit. A summary can be found in Table 7.
Convergent-divergent validity. Pearson’s correlation matrix, showed in Table 8, for HLS-Q12 with PELC, EORTC-QLQC30, and HNA and clinical variables indicated that some of the convergent and divergent hypotheses were confirmed.
PELC and the global health dimension of the EORTC-QLQC30 expressed convergence with HLS-Q12. PELC obtained a high correlation with a significant p-value, while the global health dimension of the EORTC-QLQC30 obtained a low correlation, although with a significant p-value. However, functioning and symptom dimensions of EORTC-QLQC30, ECOG before intervention, and number of comorbidities obtained low correlations and non-significant p-values, not expressing convergence with HLS-Q12.
In the divergent hypotheses, low correlations have been established with symptomatology management and HNA, however, the p-value has been significant for HNA and has not expressed significant values with symptomatology management, expressing divergence between symptomatology management and HLS-Q12.
Sensitivity to change.
Score differences, in Table 9, for HLS-Q12 between the first and fourth visits were statistically significant for the total sample and the subsample of stable or improving patients, but not for the subsample of worsening patients. The ES had a small coefficient for the total sample and the subsample of stable or improving patients, but a large effect for the subsample of worsening patients.
Discussion
This study describes the process of cross-cultural adaptation for HLS-Q12 and its psychometric properties in terms of feasibility, reliability, construct validity, and sensitivity to change. The results indicate that the Spanish translation is equivalent at the semantic, conceptual, and content levels to the original version with good internal consistency, and a small-moderate ES, but statistically significant sensitivity to change. The construct validity showed mixed results as factor analysis verified the structure of the original model with a high degree of fit, but not all convergent-divergent validity hypotheses were confirmed.
According to the HLS-EU project, it is necessary to have reduced instruments adapted and validated in different languages that are customised to particularities and contexts where they are applied. HLS-EU-Q86 was followed by reduced versions including HLS-EU-Q47, HLS-Q16, HL-SF12, and HLS-Q12 [6]. For example, HLS-Q47 in an Asian sample that showed good construct validity, satisfactory goodness-of-fit with a 3-domain model, high internal consistency, adequate convergent validity, and no significant ceiling/floor effects [43]. For HLS-Q12, validation results showed good construct validity, adequate goodness-of-fit with a 3-domain model, and high internal consistency [7,44].
In our translation of HLS-Q12, feasibility showed a good response rate due to the online, compulsory response design, together with a reduced number of questions. The item-total correlation of the HLS-Q12 obtained values >0.50 suggesting a strong homogeneity and that each item measures the same construct [24]. Values >0.7 were obtained for reliability scores both for the total sum of the instrument (alfa and omega) and for the individual items, which are considered acceptable to indicate that an instrument has good internal consistency [25–28]. These results agree with those obtained by previous versions, with values of 0.87 for HL-SF12, 0.85 for HLS-Q12 in its English version, 0.98 for HLS-Q16 in its Spanish version, and values of >0.90 in its original version HLS-EU-Q47 [43–45].
For construct validity, the original instrument is described as a unidimensional model, which also corroborates results of this study [7]. However, another study using a different methodology (Rasch model) described HLS-Q12 as having a three-subdimensional model [44]. In our study CFA showed values for X2/gl, SRMR, RMSEA, CFI, TLI, BBNNFI, AGFI, and BBNFI, that are in general better than those of the original HLS-EU-Q47 [43].
Regarding convergent-divergent validity, based on findings in the literature we were challenged to define hypotheses of interrelationships between literacy and other concepts as well as with other health questionnaires and different clinical variables [38–40]. For convergent validity, the hypotheses we proposed regarding the relationship between HL with PELC and the global health dimension of the EORTC-QLQC30 were fulfilled, indicating that the higher the literacy level, the greater the empowerment and the more positive the perception of health status, which is in agreement with the literature [38–40]. We observed a positive correlation between literacy level and how it influences better patient self-care and a better perception of global health. Conversely, we could not confirm the convergence hypotheses with EORTC-QLQC30 constructs of functioning and symptomatology, ECOG before intervention, and the number of comorbidities. Some previous studies have shown that the higher HL, the better the quality of life and the lower the reported symptomatology, and that the number of previous comorbidities negatively affects the HL of individuals [38–40]. We could not establish these hypotheses, which aligns with previous findings that show that correlations do not occur in all cases [38–40]. That some of the relationships were not significant may result from our sample size, but even so the directions of the correlations are in the direction we hypothesised.
Findings on confirmation of the divergent validity of HLS-Q12 establishes that HL alone does not have a positive relationship on the correct management of cancer symptomatology [38]. This finding could be due to the complex aetiology of cancer where symptomatology can affect each person differently depending on the stage or severity of the cancer disease, their quality of life, and their psychological state [38–40,46]. The HNA tool indicated that, for HLS-Q12, higher level of HL correlated with higher number of health needs expressed [38]. Our results showed that HL equated to lower number of expressed needs, therefore we reject our divergent hypothesis. The literature is ambiguous in this area as data are found where, despite good HL, patients continue to express doubts and changing needs [38,39]. As with convergence, sample size could have an influence.
For sensitivity to change, both overall and the stable or improvement subsamples showed a small ES but with a statistically significant p-values. This may indicate that, although the changes in scores were small, there was an increase in patient literacy. In comparison, the subsample of worsening patients had a large ES and a non-significant p-value. This could indicate that in the subsample of patients where the ECOG was worse the educational intervention achieved a substantial change in HL level since their previous literacy levels were lower, however, the sample of patients is too small to obtain a significant p-value. None of the studies concerning the HLS questionnaires have considered sensitivity to change, which limits result comparisons [43–45].
Limitations
The main limitation of the validation study is the sample size. Although the number of participants should be 2 to 10 times the number of questions contained in the questionnaire [13], others have suggested the minimum should be 5 participants per item [14]. In our case, for a 12-question survey with 60 participants, a participant/question ratio of 5 could be considered adequate but a minimum of 200 participants would be optimal for high communalities and well-determined factors and for low communalities and poorly determined factors this number could be as high as 500 [47,48]. The most penalizing aspect of not having a larger sample was in the formation of the subsamples of sensitivity to change. However, ECOG is widely used in oncology to describe patient’s quality of life and self-care and degree of independence, which is a good indicator of patient stability [17].
For reliability, we did not consider calculating the Intraclass Correlation Coefficient (ICC), which measures the temporal stability of the responses to the questionnaire in the same patient in the same conditions [26,29], given that we proposed an educational intervention from the first visit.
Follow-up times for educational interventions were variable and dependent upon the clinical trial which could impact HL measurement. Even so, there is a lack of consensus when it comes to defining exact times that correctly identify sensitivity to change [42].
Strengths and future studies
To the best of our knowledge, this is the first study to validate the HLS-Q12 in the Spanish context and in a heterogeneous oncologic population. The original questionnaires were developed in large samples of patients coming from different health backgrounds, however, there are very few instruments that address the oncological setting [49–51]. The fact that this adaptation and validation of the HLS-Q12 in Spanish has exclusively been conducted in an oncology population may be considered a strength in the absence of previous studies of the HLS project in cancer patients in Spain.
This is one of the first studies to perform convergent-divergent validity with several health questionnaires and different clinical variables [43,45], although studies with larger samples would still be necessary to improve these results and for the comparison of data in terms of sensitivity to change.
This tool could serve as an instrument for conducting multicentre studies that would provide information on the HL of patient populations in Spain, which could facilitate the development of effective and efficient strategies aimed at improving self-care, control, and critical and contrasted decision-making by patients about their own disease [8].
Conclusions
The results of our study suggest that HLS-Q12 is a robust cross-culturally adapted tool for collecting accurate HL information in cancer patients regardless of tumour or stage. More studies are needed to confirm and expand these findings, but our results support that HLS-Q12 could be used to enable interventions aimed at reducing gaps in health services and even influence the discussion of policy strategies to help improve equality and equity in health, especially in the field of oncology.
Supporting information
S1 Appendix. Health Literacy Survey-Questionnaire 12 (Spanish version).
https://doi.org/10.1371/journal.pone.0299736.s001
(PDF)
Acknowledgments
The authors would like to thank Hanne Søberg Finbråten, Guillermo Pedreira Robles, Juan Riesgo Martín, and Ágata Aguirre Martínez for their contributions.
Editorial assistance, in the form of language editing and correction, was provided by XpertScientific Editing and Consulting Services.
References
- 1. Bas-Sarmiento P, Fernández-Gutiérrez M, Poza-Méndez M, Pelicano-Piris N. Propuestas de evaluación de la alfabetización en salud [Health literacy evaluation proposals]. Psychologia Latina. 2015;6(1):1–11.
- 2. Sørensen K, Van den Broucke S, Fullam J, Doyle G, Pelikan J, Slonska Z, et al. Health literacy and public health: a systematic review and integration of definitions and models. BMC Public Health. 2012;12:80. pmid:22276600
- 3.
Oliva Ramírez C. Alfabetización en salud en pacientes oncológicos [Health literacy in oncology patients] [Master’s thesis on the internet]. San Cristóbal de La Laguna: Universidad de la Laguna. 2021. https://riull.ull.es/xmlui/bitstream/handle/915/25222/ALFABETIZACION%20EN%20SALUD%20EN%20PACIENTES%20ONCOLOGICOS.pdf
- 4.
Organización Mundial de la Salud. Promoción de la salud: Glosario [Health promotion: glossary]. Ginebra: Ministerio de Sanid y Consumo; 1998. https://www.sanidad.gob.es/areas/promocionPrevencion/promoSaludEquidad/equidadYDesigualdad/glosario/docs/glosario.pdf
- 5. Sørensen K, Pelikan JM, Röthlin F, Ganahl K, Slonska Z, Doyle G, et al. Health literacy in Europe: Comparative results of the European health literacy survey (HLS-EU). Eur J Public Health. 2015;25(6):1053–8. pmid:25843827
- 6.
Departamento de sanidad pública y enfermedad infecciosa. Adult ognitive Decline Conscientiousness Project health literacy in Europe. Roma: Sapienza (Universidad de Roma); 2017. https://www.acdcproject.eu/documents/health_literacy_in_europe_en.pdf
- 7. Finbråten HS, Wilde-Larsson B, Nordström G, Pettersen KS, Trollvik A, Guttersrud Ø. Establishing the HLS-Q12 short version of the European Health Literacy Survey Questionnaire: Latent trait analyses applying Rasch modelling and confirmatory factor analysis. BMC Health Serv Res. 2018;18(1):1–17. pmid:29954382
- 8. Fernandez-Gonzalez L, Bravo-Valenzuela P. Effective interventions to improve the health literacy of cancer patients. Ecancermedicalscience. 2019;13:1–14. pmid:31921337
- 9. Garcimartín P, Comín-Colet J, Pardo-Cladellas Y, Badosa N, Linas A, Rosenfeld L, et al. Validation of the Spanish version of the questionnaire on Patient Empowerment in Long-term Conditions. PLoS One. 2020;15(6):e0233338. pmid:32530951
- 10. Doostfatemeh M, Taghi Ayatollah SM, Jafari P. Power and sample size calculations in clinical trials with patient-reported outcomes under equal and unequal group sizes based on graded response model: A simulation study. Value Heal. 2016;19(5):639–47. pmid:27565281
- 11. Anthoine E, Moret L, Regnault A, Sbille V, Hardouin JB. Sample size used to validate a scale: A review of publications on newly-developed patient reported outcomes measures. Health Qual Life Outcomes. 2014;12(1):1–10. pmid:25492701
- 12.
Morales-Vallejo P. El Análisis Factorial en la construcción e interpretación de tests, escalas y cuestionarios [Factor analysis in the construction and interpretation of tests, scales and questionnaires]. Madrid: Universidad Pontificia Comillas; 2011. https://pdfslide.tips/documents/xxx-el-analisis-factorial-en-la-construccion-de-test.html?page=1
- 13.
Argimon Pallás JM, Jiménez Villa J. Métodos de investigación clínica y epidemiológica [Clinical and epidemiological research methods]. 4th ed. Madrid: Elsevie; 2013.
- 14.
Tornimbeni S, Pérez E, Olaz F. Introducción a la psicometría [Introduction to psychometrics]. 1st ed. Buenos Aires: Paidós; 2008.
- 15. Husson O, Rooij BH, Kieffer J, Oerlemans S, Mols F, Aaronson NK, et al. The EORTC QLQ‐C30 summary score as prognostic factor for survival of patients with cancer in the “real‐world”: Results from the population‐based PROFILES registry. Oncologist. 2020;25(4):722–32. pmid:32297435
- 16. Aaronson NK, Ahmedzai S, Bergman B, Bullinger M, Cull A, Duez NJ, et al. The European organization for research and treatment of cancer QLQ-C30: A quality-of-life instrument for use in international clinical trials in oncology. J Natl Cancer Inst. 1993;85(5):365–76. pmid:8433390
- 17. Pérez-Cruz PE, Acevedo F. Clasificaciones en Gastroenterología Escalas de estado funcional (o performance status) en cáncer [Classifications in Gastroenterology Scales of functional status (or performance status) in cancer]. Gastroenterol latinoam. 2014;25:219–26.
- 18. Garcimartin P, Pardo-Cladellas Y, Verdú-Rotellar JM, Delgado-Hito P, Astals-Vizcaino M, Comin-Colet J. Transcultural adaptation into Spanish of the Patient Empowerment in Long-term Conditions questionnaire. Aten Primaria. 2019;51(1):24–31. pmid:29277376
- 19. Snowden A, Young J, White C, Murray E, Richard C, Lussier MT, et al. Evaluating holistic needs assessment in outpatient cancer care—A randomised controlled trial: The study protocol. BMJ Open. 2015;5(5):e006840. pmid:25967990
- 20. Young J, Cund A, Renshaw M, Quigley A, Snowden A. Improving the care of cancer patients: Holistic needs assessment. Br J Nurs. 2015;24(4):S17–20. pmid:25723367
- 21. Muñiz J, Elosua P, Hambleton RK. Directrices para la traducción y adaptación de los tests: Segunda edición [Guidelines for translation and adaptation of tests: Second edition]. Psicothema. 2013;25(2):151–7. pmid:23628527
- 22. Wild D, Grove A, Martin M, Eremenco S, McElroy S, Verjee-Lorenz A, et al. Principles of good practice for the translation and cultural adaptation process for patient-reported outcomes (PRO) measures: Report of the ISPOR Task Force for Translation and Cultural Adaptation. Value Heal. 2005;8(2):94–104. pmid:15804318
- 23.
Mokkink LB, Prinsen CA, Patrick D, Alonso J, Bouter LM, de Vet HC, et al. COSMIN Study Design checklist for patient-reported outcome measurement instruments. Amsterdam: COSMIN; 2019. https://www.cosmin.nl/wp-content/uploads/COSMIN-study-designing-checklist_final.pdf
- 24.
Frías-navarro D. Apuntes de consistencia interna de las puntuaciones de un Instrumento de medida [Notes on internal consistency of the scores of a measurement instrument]. Universidad de Valencia; 2022. https://www.uv.es/friasnav/AlfaCronbach.pdf
- 25. Batista-Foguet JM, Coenders G, Alonso J. Análisis factorial confirmatorio. Su utilidad en la validación de cuestionarios relacionados con la salud [Confirmatory factor analysis. Its role on the validation of health related questionnaires]. Med Clin (Barc). 2004;122(1):21–7. pmid:14980156
- 26. Lohr KN. Assessing health status and quality-of-life instruments: Attributes and review criteria. Qual Life Res. 2002;11(3):193–205. pmid:12074258
- 27. Hair JF Jr, Sarstedt M, Hopkins L, Kuppelwieser VG. Partial least squares structural equation modeling (PLS-SEM): An emerging tool in bussiness research. Eur Bus Rev. 2014; 26 (2):106–121.
- 28. Cheung GW, Cooper-Thomas HD, Lau RS, Wang LC. Reporting reliability, convergent and discrminant validity with structural equation modeling: A review and best-practice recommendations. Asia Pac J Manag. 2023.
- 29. Carvajal A, Centeno C, Watson R, Martínez M, Rubiales AS. ¿Cómo validar un instrumento de medida de la salud? [How is an instrument for measuring health to be validated?]. An Sist Sanit Navar. 2011;34(1):63–72.
- 30. Sass DA, Schmitt TA, Marsh HW. Evaluating model fit with ordered categorical data within a measurement invariance framework: A comparison of estimators. Struct Equ Model. 2014;21(2):167–80.
- 31.
Romero García M. Diseño y validación de un cuestionario de satisfacción con los cuidados enfermeros desde la perspectiva del paciente crítico [Design and validation of a nursing care satisfaction questionnaire from the perspective of the critically ill patient] [Doctoral Thesis]. Barcelona; 2016. http://hdl.handle.net/10803/383031
- 32. Ruiz M, Pardo A, San Martín R. Modelos de ecuaciones estructurales [Structural equation modeling]. Papeles del Psicólogo. 2010;31(1):34–45.
- 33. Escobedo Portillo MT, Hernández Gómez JA, Estebané Ortega V, Martínez Moreno G. Modelos de ecuaciones estructurales: Características, fases, construcción, aplicación y resultados [Structural equation models: Characteristics, phases, construction, application and results]. Cienc Trab. 2016;18(55):16–22.
- 34.
Garcimartín P. Adaptación cultural y validación al español del cuestionario patient Empowerment in Long-term Conditions [Cultural adaptation and validation to Spanish of the Patient Empowerment in Long-term Conditions questionnaire] [Doctoral Thesis]. Barcelona; 2018. https://www.tdx.cat/handle/10803/664423
- 35.
Ramírez R, Guevara F, D´Armas M, Pena R, Farias R, Bravo F, et al. Análisis multivariante teoría y práctica de las principales técnicas [Multivariate analysis theory and practice of the main techniques]. Ecuador: Universidad Estatal de Milagro; 2017. https://repositorio.unemi.edu.ec/bitstream/123456789/3855/1/MULTIVARIANTE%20watermark.pdf
- 36.
Hair J, Anderson R, Tatham R, Black W. Análisis multivariante [Multivariate analysis]. 5th ed. Madrid: Prentice Hall; 1999.
- 37.
Cohen J. Statistical power analysis for the behavioral sciences. 2nd ed. New York: Lawrence Erlbaum Associates; 1988.
- 38. Holden CE, Wheelwright S, Harle A, Wagland R. The role of health literacy in cancer care: A mixed studies systematic review. PLoS One. 2021;16(11):1–26. pmid:34767562
- 39.
Pelikan JM, Röthlin F, Ganahl K. Comparative report on health literacy in eight EU member states. Austria: Ludwig Boltzmann Institute Health Promotion Research; 2012.
- 40. Kieffer Campbell J. Health literacy in adult oncology: An integrative review. Oncol Nurs Forum. 2020;47(1):18–32. pmid:31845912
- 41.
Pallant J. SPSS survival manual. 6th ed. United Kingdom: Open University Press; 2016.
- 42. García de Yébenes Prous MJ, Rodríguez Salvanés F, Carmona Ortells L. Sensibilidad al cambio de las medidas de desenlace [Sensitivity to change of outcome measures]. Reumatol Clínica. 2008;4(6):240–7. pmid:21794541
- 43. Duong TV, Aringazina A, Baisunova G, Nurjanah , Pham TV, Pham KM, et al. Measuring health literacy in Asia: Validation of the HLS-EU-Q47 survey tool in six Asian countries. J Epidemiol. 2017;27(2):80–6. pmid:28142016
- 44. Maie A, Kanekuni S, Yonekura Y, Nakayama K, Sakai R. Evaluating short versions of the European Health Literacy Survey Questionnaire (HLS-EU-Q47) for health checkups. Heal Eval Promot. 2021;48(4):351–8.
- 45. Van Duong T, Chang PWS, Yang SH, Chen MC, Chao WT, Chen T, et al. A new comprehensive short-form health literacy survey tool for patients in general. Asian Nurs Res (Korean Soc Nurs Sci). 2017;11(1):30–5. pmid:28388977
- 46. León-González R, García-Esquinas E, Paredes-Galán E, Ferrero-Martínez AI, González-Guerrero JL, Hornillos-Calvo M, et al. Alfabetización en salud y resultados de salud en pacientes muy ancianos con insuficiencia cardiaca [Health literacy and health outcomes in very elderly patients with heart failure]. Rev Española Cardiol. 2018;71(3):178–84.
- 47. Ferrando P, Anguiano-Carrasco C. El análisis factorial como técnica de investigación en psicología [Factor analysis as a research technique in psychology]. Papeles del Psicológo. 2010;31(1):18–33.
- 48. Lloret-Segura S, Ferreres-Traver A, Hernández-Baeza A, Tomás-Marco I. El análisis factorial exploratorio de los ítems: Una guía práctica, revisada y actualizada [The exploratory factor analysis of items: A practical guide, revised and updated]. An Psicol. 2014;30(3):1151–69.
- 49. Huang YJ, Lin GH, Lu WS, Tam KW, Chen C, Hou WH, et al. Validation of the European Health Literacy Survey Questionnaire in women with breast cancer. Cancer Nurs. 2018;41(2):E40–8. pmid:28221213
- 50. Dai Minh L, Quang BV, Ngoc Le Mai D, Le Quyen L, Gia NH, Hang NT, et al. Health literacy of newly-admitted cancer patients in Vietnam: Difficulties understanding treatment options and processing health-related information. Heal Serv Insights. 2022;15:1–10. pmid:35035220
- 51. Kobayashi R, Ishizaki M. Relationship between health literacy and social support and the quality of life in patients with cancer: Questionnaire study. J Particip Med. 2020;12(1): e17163. pmid:33064103