Psychometric Evaluation of the Persian eHealth Literacy Scale (eHEALS) Among Elder Iranians With Heart Failure

Abstract

The purpose of the present study was to examine the psychometric properties of the eHealth Literacy Scale (eHEALS) using classical test theory and modern test theory among elderly Iranian individuals with heart failure (HF). Individuals with objectively verified HF (n = 388, 234 males, mean age = 68.9 ± 3.4) completed the (i) eHEALS, (ii) Hospital Anxiety and Depression Scale, (iii) Short Form 12, (iv) 9-item European Heart Failure Self-Care Behavior Scale, and (v) 5-item Medication Adherence Report Scale. Two types of analyses were carried out to evaluate the factorial structure of the eHEALS: (i) confirmatory factor analysis (CFA) in classical test theory and (ii) Rasch analysis in modern test theory. A regression model was constructed to examine the associations between eHEALS and other instruments. CFA supported the one-factor structure of the eHEALS with significant factor loadings for all items. Rasch analysis also supported the unidimensionality of the eHEALS with item fit statistics ranging between 0.5 and 1.5. The eHEALS was significantly associated with all the external criteria. The eHEALS is suitable for health-care providers to assess eHealth literacy for individuals with HF.

Keywords

classical test theory eHealth literacy heart failure Iran modern test theory

Contemporary technology allows individuals to seek health information on the Internet via devices such as Wi-Fi-enabled smartphones, tablets, and laptops. The Pew Research Internet Project estimates that more than 85% of American adults use the Internet, and nearly three - quarters of them have searched for health information online (Pew Research Center, 2016). However, searching for health information is different from interpreting health information. More specifically, individuals may lack sufficient knowledge to interpret the health information they access and read online. Therefore, assessing eHealth literacy is deemed a prerequisite for health-care providers to promote eHealth resources to patients who may need them (Norman & Skinner, 2006).

eHealth literacy is defined as “the ability to navigate the Internet for health information” (Nguyen et al., 2016, p. 2). eHealth literacy can be challenging for patients, given the many different core skills or literacies that exist including (i) traditional literacy, (ii) health literacy, (iii) information literacy, (iv) scientific literacy, (v) media literacy, and (vi) computer literacy (Norman & Skinner, 2006). More specifically, patients should have the knowledge to access, retrieve, evaluate, and appraise the information they gain online (Norman & Skinner, 2006). Patients are likely to obtain different types and quality of information that they need to further compare and evaluate. Moreover, given the rapid change of both care routines and technology, health information is updated quickly. That is, health information yesterday may not be good practice today (Norman & Skinner, 2006).

In order to appropriately use online resources for health purposes, Norman and Skinner (2006) developed the eHealth Literacy Scale (eHEALS) for clinical decision-making. Because of the brevity and utility of the eHEALS, it has been translated into different languages including Japanese (Mitsutake, Shibata, Ishii, & Oka, 2012), Chinese (Koo, Norman, & Chang, 2012), Dutch (Van der Vaart et al., 2011), and Spanish (Aponte & Nokes, 2015, 2017) and validated in different populations. However, given that elder people may have barriers or problems to learn and use e-resources (Vaportzis, Clausen, & Gow, 2017), it is unclear whether the eHEALS is a practical and valid tool for elders. More specifically, to the best of our knowledge, the eHEALS has never been applied to individuals with heart failure (HF), who are usually elderly and need to have good self-care behaviors.

HF is often associated with several comorbidities, and fluctuations of the condition cause frequent rehospitalizations (Benjamin et al., 2017; Gheorghiade, Vaduganathan, Fonarow, & Bonow, 2013; Grady, 2008). Therefore, an individual with HF needs sufficient knowledge about, for example, how to handle signs and symptoms, self-care actions, and potential changes of pharmacological treatment to handle the situation. This knowledge could be obtained either from health-care providers or via the Internet. Furthermore, adherence to medication and self-care actions in HF have found to be poor, and a variety of factors are understood to affect self-care behaviors, of which knowledge is one (Jaarsma, Cameron, Riegel, & Stromberg, 2017; Lee et al., 2018; Sedlar et al., 2017). In addition, individuals with HF are usually elderly and are likely to have insufficient ability in using online resources. A recent study (Melholt et al., 2018) stated that individuals with HF may gain benefits from eHealth literacy. Therefore, a validation of the eHEALS in this population would increase possibilities to provide good patient education and (in the long run) prevent rehospitalizations. More specifically, after establishing the psychometric properties of the eHEALS, health-care providers can use the eHEALS to identify whether an individual with HF has sufficient competence to seek health information via the Internet or whether a face-to-face provided intervention is needed (e.g., individual patient education about self-care).

Consequently, the present study aimed to investigate the psychometric properties of the eHEALS among individuals with HF. Moreover, an attempt was made to strengthen the robustness of the psychometric findings of the eHEALS by utilizing two different psychometric theories (i.e., classical test theory and modern test theory). Classical test theory has a strong assumption that the Likert-type scale scores (a type of ordinal scale) are additive, although the nature of Likert-type scales is nonadditive. Nevertheless, classical test theory provides information that most health-care providers are familiar with (e.g., Cronbach’s α and factor loading). In contrast, modern test theory such as Rasch analysis uses the probabilities of answering a specific category in the Likert-type scale to convert all Likert-type scales into ratio scales. Using the converted ratio scales, which are additive scales in nature, psychometric properties such as item and person separation reliability can be computed in the Rasch models. Therefore, using different theories helps expand knowledge regarding the psychometric properties of the eHEALS.

Method

Design, Participants, and Procedure

The present methodological study was conducted at three university hospitals in two cities (Tehran and Qazvin) between 2017 and 2018. Patients referred to these hospitals were assessed by two physicians in terms of their eligibility for study inclusion. The inclusion criteria for the study were (i) being aged 65 years or older, (ii) having a confirmed diagnosis of HF by echocardiographic and physical examination according to the International Classification of Diseases (10th revision), (iii) having the ability to speak and write Persian, and (iv) having access to the Internet at least once a month (via smartphone, tablet, and/or computer). Patients were excluded from the study if they had dementia (determined by a Mini-Mental Status score of less than 25) or diagnosed as having Alzheimer’s dementia or Parkinson’s disease. Clinical data were obtained from medical records. Left ventricular ejection fraction was determined by echocardiography performed according to clinical routines at the current hospital and analyzed by experienced cardiologists.

Among the 458 patients approached, 47 patients were not eligible to be included in the study and 23 patients declined to participate. Consequently, 388 patients participated in the study with a response rate of 84.7%. Additionally, test–retest reliability was carried out across a 2-week interval. More specifically, 388 participants were contacted by research assistants and invited to complete the eHEALS again 2 weeks after the first administration of the scale. In the retest, 43 patients declined to complete the retest (11.1% dropout rate). Therefore, data from the retained 345 patients were used for test–retest reliability.

Translation Procedure

The validated English version of eHEALS (Norman & Skinner, 2006) was translated into Persian based on international guidelines of cross-cultural adaptation (Lin et al., 2018). The translation procedure was performed in several steps. In the first step, the English version of the eHEALS was translated into Persian by two bilingual translators whose mother tongue was Persian. In the next step, the translated versions were compared, and discrepancies were resolved in order to synthesize them into an interim Persian version. Two bilingual translators then independently translated the interim Persian version into English. Both translators who spoke English as their native language were not aware of original English version of the eHEALS. An expert committee (comprising a cardiologist, cardiovascular nurse, psychometrician, and psychologist) then reviewed all translated versions, and any discrepancies were discussed to produce a prefinal version. The prefinal version was then piloted on 21 patients with HF (9 women and 11 men with mean age of 68.6 ± 8.4 years). The patients were asked to read the questionnaire items as well as the instructions. A cognitive interview was conducted to test the feasibility and understanding of the items. All necessary changes were made, and the final Persian eHEALS was administered to 388 HF patients to assess the psychometric properties of the newly translated scale.

Measures

eHEALS

The eHEALS (sample item: Know how to find helpful health resources on the Internet. Refer to Table 1 for other item descriptions) is a self-report tool that can be administered by health-care providers with little or no training (Norman & Skinner, 2006). The eHEALS comprises 8 items rated on a 5-point Likert-type scale (score 1 as strongly disagree; score 5 as strongly agree), where a higher score indicates a higher level of confidence in the ability of finding, evaluating, and using health information to make health-related decisions. In short, a higher score represents greater perceived eHealth literacy (Paige, Krieger, Stellefson, & Alber, 2017). Also, the one-factor structure (or unidimensionality) of the eHEALS has been supported by prior confirmatory factor analysis (CFA) and Rasch analysis among patients with chronic diseases including cardiovascular diseases (Paige et al., 2017).

Table 1.

Psychometric Properties of the eHealth Literacy Scale in Item Level.

	Analyses From Classical Test Theory			Analyses From Rasch
Item #	Factor Loading^a	Item-Total Correlation	Test–Retest Reliability^b	Infit MnSq	Outfit MnSq	Difficulty	DIF Contrast Across Gender^c,d	DIF Contrast Across NYHA Classification^ce
eHEALS1: I know how to find helpful health resources on the Internet.	0.66	.72	0.80	1.06	1.05	−0.04	0.22	0.18
eHEALS2: I know how to use the Internet to answer health questions.	0.78	.82	0.79	0.84	0.79	−0.18	0.03	−0.24
eHEALS3: I know what health resources are available on the Internet.	0.78	.80	0.84	0.85	0.78	0.16	0.07	0.42
eHEALS4: I know where to find helpful health resources on the Internet.	0.64	.66	0.92	1.36	1.34	0.63	0.01	−0.23
eHEALS5: I know how to use the health information found on the Internet to help me.	0.79	.80	0.88	0.86	0.87	−0.09	−0.07	0.44
eHEALS6: I have the skills to evaluate the health resources found on the Internet.	0.70	.75	0.83	0.94	0.93	0.06	−0.12	0.35
eHEALS7: I can tell high quality from low quality health resources on the Internet.	0.60	.69	0.86	1.14	1.09	−0.13	−0.24	−0.37
eHEALS8: I feel confident in using information from the Internet to make decisions.	0.74	.78	0.90	0.97	0.92	−0.41	0.08	−0.56

Note. N = 388. Infit = information-weighted fit statistic; outfit = outlier-sensitive fit statistic; MnSq = mean square error; DIF = differential item functioning; NYHA = New York Heart Association.

^aBased on confirmatory factor analysis. ^bUsing Pearson correlation. ^cDIF contrast > 0.5 indicates substantial DIF. ^dDIF contrast across gender = difficulty for females − difficulty for males. ^eDIF contrast across NYHA classification = difficulty in class II – difficulty in classes III and IV.

Hospital Anxiety and Depression Scale (HADS)

The HADS is a frequently used self-report tool comprising 14 items (7 items on anxiety and 7 on depression) and rated on a 4-point Likert-type scale ranging from 0 to 3, where a higher score indicates a higher level of anxiety or depression. The two-factor structure of the HADS has been supported by CFA, and the unidimensionality of each factor has been found in Rasch analysis among Iranian patients with epilepsy (Lin & Pakpour, 2017).

Short Form 12 (SF-12)

The SF-12 is a self-report tool that assesses the generic quality of life of an individual. The SF-12 comprises 12 items across two domains rated on a variety of response scales, including two to six categories (i.e., 2-point to 6-point Likert-type scales). All the item scores are converted into a 0–100 scale, where a higher score indicates better quality of life. Moreover, the SF-12 can be divided into two dimensions of physical health composite score (PCS) and mental health composite score (MCS). The two-factor structure of the SF-12 has been supported by principal component analysis and CFA among Iranian individuals aged 15 years or older (Montazeri, Vahdaninia, Mousavi, & Omidvari, 2009).

European HF Self-Care Behavior Scale 9-Item Version (EHFScB-9)

The EHFScB-9 is a self-report tool and comprises 9 items rated on a 5-point Likert scale ranging from 1 (completely agree) to 5 (completely disagree), where a higher score indicates better self-care of a patient with HF (Jaarsma, Årestedt, Mårtensson, Dracup, & Strömberg, 2009). Different factorial structures have been proposed for the EHFScB-9, and the latest consensus describes a two-factor structure that includes adherence to regimen (5 items) and consulting behavior (4 items; Paige et al., 2017). More specifically, the two-factor structure of the EHFScB-9 has been supported by CFA, and the unidimensionality of each factor has been found in Rasch analysis among Iranian HF patients (Paige et al., 2017).

5-Item Medication Adherence Report Scale (MARS-5)

The MARS-5 is a self-report tool comprising 5 items rated on a 5-point Likert-type scale ranging from 1 to 5, where a higher score indicates a higher level of medication adherence. The one-factor structure of the MARS has been supported both by CFA and Rasch analysis among Iranian stroke patients. Moreover, the MARS-5 has strong relationship with the medication possession rate (r = .7; Lin, Nikoobakht, Broström, Arestedt, & Pakpour, 2018).

Ethical Considerations

The investigation conformed to the principles outlined in the Declaration of Helsinki. All participants gave their written consent to participate in the study, and the study protocol was approved by the Ethics Committee at Qazvin University of Medical Sciences in Iran.

Data Analysis

After using descriptive analyses for the participant characteristics, robust psychometric testing was applied using both classical test theory and modern test theory (i.e., Rasch analysis) to examine both item and scale properties of the eHEALS. In psychometric testing utilizing classical test theory, a number of measures were tested: (1) the ceiling and floor effects with a value <20% indicating acceptable, (2) internal consistency using Cronbach’s α with a value >0.7 indicating acceptable, (3) corrected item-total correlation with a value >.4 indicating acceptable, (4) test–retest reliability using Pearson correlation coefficient with a value >.4 indicating acceptable, (5) average variance extracted (AVE) with a value >0.5 indicating acceptable, (6) composite reliability (CR) with a value >0.6 indicating acceptable, (7) standard error of measurement with a lower value indicating better, and (8) concurrent validity using a regression model to examine the associations between eHEALS and the following external criteria: depression, anxiety, PCS, MCS, adherence to regimen, consulting behavior, and medication adherence.

Following this, CFA was performed using weighted least squares and adjusted means and variances estimation to test the one-factor structure of the eHEALS using the following fit indices to indicate acceptable data model fit: a nonsignificant χ² test, a comparative fit index (CFI), a Tucker–Lewis index (TLI) > 0.9, a root mean square error of approximation, and a standardized root mean square residual (SRMR) < 0.08 (McDonald & Ho, 2002). Furthermore, three nested models were conducted to test the measurement invariance of the eHEALS across gender and across New York Heart Association (NYHA) classifications (NYHA II vs. NYHA III and IV) using multigroup CFA (Paige et al., 2017). The three models were a configural model, a model constrained by all factor loadings equal across group, and a model constrained by all factor loadings and item intercepts equal across group. The measurement invariance (i.e., testing whether different groups share the same or similar concept in a specific instrument) was supported using the following fit indices: ΔCFI > −0.01, ΔSRMR < 0.02, and ΔRMSEA < 0.015 (Chen, 2007).

In psychometric testing under Rasch analysis, the item difficulty was reported: information-weighted fit statistic (infit) mean square (MnSq) and outlier-sensitive fit statistic (outfit) MnSq (with range between 0.5 and 1.5 indicating acceptable), item and person separation reliability (with a value >0.7 indicating acceptable), item and person separation index (with a value > 2 indicating acceptable), and differential item functioning (DIF) contrast (with a value <0.5 indicating acceptable) across gender and across NYHA classifications (NYHA II vs. NYHA III and IV; Lin et al., 2018). All statistical analyses were performed using Mplus (Version 7.4; Los Angeles, CA) and Winstep 4.1.0 software (winsteps.com, Beaverton, OR).

Results

The sample of 388 participants included 234 males (60.3%), 59 current smokers (15.2%), and 243 having NYHA classification II (62.6%). In addition, the mean age of the participants was 68.9 years (SD = 3.4) with 6.4 years of education (SD = 3.2) and a body mass index of 28.2 kg/m² (SD = 5.0). On average, participants had suffered from HF for 5.5 years (SD = 3.6) with an average left ventricular ejection fraction of 30.2% (SD = 7.4). Table 2 presents other HF characteristics of the sample.

Table 2.

Participants Characteristics.

Characteristics	Mean ± SD or n (%)
Demographic variables
Age (year)	68.9 ± 3.4
Gender (male)	234 (60.3)
Gender (female)	154 (39.7)
Years of education	6.4 ± 3.2
Body mass index (kg/m²)	28.2 ± 5.0
Current smoker	59 (15.2)
Heart failure characteristics
Duration of heart failure (years)	5.5 ± 3.6
NYHA classification (II)	243 (62.6)
NYHA classification (III)	91 (23.5)
NYHA classification (IV)	54 (13.9)
Left ventricular ejection fraction (%)	30.2 ± 7.4
Medication
Diuretic	189 (48.7)
β Blockers	242 (62.4)
Angiotensin-converting-enzyme inhibitor (ACE inhibitor)	301 (77.6)
SBP (mmHg)	143.1 ± 30.5
DBP (mm Hg)	82.3 ± 22.7

Note. N = 388. NYHA = New York Heart Association; SBP = systolic blood pressure; DBP = diastolic blood pressure.

The responses to each eHEALS item are summarized in Supplementary Table S1. In brief, most participants endorsed items in the range 3 (undecided) to 5 (strongly agree), and relatively few participants endorsed responses 1 (strongly disagree) and 2 (disagree) for all the items apart from Item 6. On eHEALS Item 6, a few participants endorsed response 5. Additionally, the mean and SD of each eHEALS item are presented in Supplementary Table S1, with the lowest mean being 3.21 and the highest being 4.30.

All 8 items of the eHEALS showed promising item properties as indicated by the strong factor loadings (0.60–0.79), satisfactory corrected item-total correlation (.66 to –.82), and high test–retest reliability (0.79–0.92) from classical test theory. Using modern test theory, Rasch analysis indicated adequate infit and outfit MnSq (0.84–1.36 and 0.78–1.34, respectively) and acceptable DIF contrast across gender (DIF contrasts between −0.24 and 0.22). The DIF contrasts were all acceptable across NYHA classification, except for Item 5 (DIF contrast = −0.56). In addition, the item difficulties ranged between −0.41 and 0.63 (Table 1).

In terms of the scale level, all the psychometric properties of the eHEALS were satisfactory using classical test theory (ceiling effect = 2.8%, floor = 1%, Cronbach’s α = 0.89, CFI = 0.987, TLI = 0.979, RMSEA = 0.053, SRMR = 0.029, AVE = 0.51, CR = 0.89, standard error of measurement = 2.09, test–retest reliability = 0.85) or using Rasch analysis (item separation reliability = 0.93, item separation index = 3.62, person separation reliability = 0.86, person separation index = 2.36; Table 3). Psychometric testing from classical test theory also showed that eHEALS was significantly correlated to different external criteria including depression (β = −0.12), anxiety (β = −0.09), PCS (β = 0.14), MCS (β = 0.17), adherence to regimen (β = 0.13), consulting behavior (β = 0.10), and medication adherence (β = 0.27; Supplementary Table S2).

Table 3.

Psychometric Properties of the eHealth Literacy Scale in Scale Level.

Psychometric Testing	Value	Suggested Cutoff
Ceiling effects (%)	2.8	<20
Floor effects (%)	1	<20
Internal consistency (Cronbach’s α)	0.89	>0.7
Confirmatory factor analysis
χ² (df)	37.75 (20)*	Nonsignificant
Comparative fit index	0.987	>0.9
Tucker–Lewis index	0.979	>0.9
Root mean square error of approximation	0.053	<0.08
Standardized root mean square residual	0.029	<0.08
Average variance extracted	0.51	>0.5
Composite reliability	0.89	>0.6
Standard error of measurement	2.09	The smaller the better
Item separation reliability from Rasch	0.93	>0.7
Item separation index from Rasch	3.62	>2
Person separation reliability from Rasch	0.86	>0.7
Person separation index from Rasch	2.36	>2
Test–retest reliability by Pearson correlation	0.85	>0.4

*p < .001.

Multigroup CFA corresponded to the DIF findings in that measurement invariance was supported across gender (nonsignificant findings between configural model and model with all factor loadings constrained, Δχ² = 2.539, df = 8, p = .96, and between model with all factor loadings constrained and model with all factor loadings and all item intercepts constrained, Δχ² = 4.295, df = 8, p = .83) and across NYHA classifications (nonsignificant findings between configural model and model with all factor loadings constrained, Δχ² = 12.728, df = 8, p = .12, and between model with all factor loadings constrained and model with all factor loadings and all item intercepts constrained, Δχ² = 4.441, df = 8, p = .82). Moreover, the ΔCFI, ΔSRMR, and ΔRMSEA all supported the measurement invariance of the eHEALS across gender and across NYHA classification (Supplementary Table S3).

Discussion

The Internet has become increasingly ubiquitous in society, and many websites have been developed for educating individuals with HF in disease management and symptom prevention (Orlowski, Oermann, & Shaw, 2013). However, this does not mean that all patient education should be provided via the Internet, especially since self-care in HF is highly complex (Sedlar et al., 2017). Jaarsma, Cameron, Riegel, and Stromberg (2017) stress that the three key concepts of self-care—self-care maintenance (e.g., adherence to medication), self-care monitoring (e.g., regular check of body weight), and self-care management (e.g., actions in response to symptoms)—are affected by (among other things) access to care and cognitive abilities. Therefore, ensuring an individual with HF has sufficient ability to use online resources is crucial. Furthermore, given the fact that the population contains mostly elderly people, sometimes with impaired cognitive function (Cannon et al., 2017) and depression (Rustad, Stern, Hebert, & Musselman, 2013), even further increases the importance of individualized patient-centered care.

Given that Normand and Skinner (2006) indicated that “eHealth literacy promotion takes place within a larger learning context” (p. 5), they further proposed that psychometric studies on eHEALS should test the relationship between eHEALS and other measures such as social functioning, health, and quality of life. In order to respond to the aforementioned recommendation, the present study used a regression model to assess the relationship between eHEALS and relevant measures on individuals with HF. The significant associations found in the regression model were as anticipated. More specifically, higher scores on the eHEALS were associated with lower levels of anxiety and depression, higher levels of quality of life and better HF self-care behaviors. Using a relatively large sample of patients with HF, the results of the present study demonstrated promising psychometric properties of the eHEALS. In other words, the use of the eHEALS was supported, and health-care providers are therefore encouraged to use the eHEALS to evaluate the eHealth literacy for individuals with HF. Through such practice, health-care providers may understand whether an individual with HF has sufficient ability to use online resources for health improvement and/or maintenance.

The present study found that both CFA and Rasch analyses supported a one-factor structure (i.e., unidimensionality) of the eHEALS, suggesting that health-care providers can use eHealth literacy as a whole in a clinical assessment. The one-factor structure is important for eHEALS because this indicates that summing the eHEALS item scores into a total score is appropriate (Chang et al., 2018). With the summated single total score of eHEALS, health-care providers can quickly and easily understand the eHealth literacy of an individual. The finding of a unidimensional construct also aligns with most previous studies using principal component analysis, exploratory factor analysis, or Rasch analysis on different populations (Aponte & Nokes, 2015, 2017; Diviani, Dima, & Schulz, 2017; Koo et al., 2012; Mitsutake et al., 2012; Nguyen et al., 2016; Norman & Skinner, 2006; Paige et al., 2017). However, the unidimensional finding contradicts the results of two recently published studies (Hyde, Boyes, Evans, Mackenzie, & Sanson-Fisher, 2018; Sudbury, FitzPatrick, & Schulz, 2017), which proposed a three-factor structure for the eHEALS. More specifically, Sudbury-Riley, FitzPatrick, and Schulz (2017) used CFA to compare one- and three-factor structures of the eHEALS and found that the three-factor structure outperformed one-factor structure among baby boomers born between 1946 and 1964 in the United States, UK, and New Zealand. Hyde, Boyes, Evans, Mackenzie, and Sanson-Fisher (2018) conducted another CFA on medical imaging outpatients and further supported the three-factor structure with minor amendments (i.e., dropping one item).

Nevertheless, using the findings of the present study on individuals with HF, it is argued that the eHEALS should be treated as having one-factor rather than three-factor structure because studies using Rasch analysis (or other analysis under modern test theory) support the one-factor structure (Diviani et al., 2017; Nguyen et al., 2016; Paige et al., 2017). Given that CFA under classical test theory has the characteristics of being sample dependent (i.e., psychometric results vary in different studied samples; Chang, Wang, Tang, Cheng, & Lin, 2014), the different factorial structures found in the previous studies (Diviani et al., 2017; Nguyen et al., 2016; Paige et al., 2017) were very likely due to the sample characteristics. In contrast, Rasch analysis with sample-independent characteristics is not influenced by the threat of sample characteristics (Chang et al., 2014). Consequently, studies using Rasch analysis (Diviani et al., 2017; Nguyen et al., 2016; Paige et al., 2017) together with the Rasch findings presented in the present study demonstrate consistent unidimensional results for the eHEALS.

Additionally, health-care providers should be cautious using eHEALS when comparing individuals with HF who have minor severity (NYHA class II) and those who have severe severity (NYHA classes III and IV) because eHEALS Item 8 displayed DIF. The DIF results indicated that those with minor severity had a tendency to answer this item higher (i.e., “I feel confident in using information from the Internet to make decisions”). A possible explanation is that those in NYHA class II follow recommendations (which is positive), but those with severe HF cannot. From a clinical perspective, this is logical since those with severe HF (NYHA III and IV) may not have the capacity to follow recommendations due to their symptoms (e.g., fatigue restricts physical activity), medication (e.g., side effects from diuretics creates difficulties to follow fluid restriction), and/or poor cardiac function (e.g., increases risk of sleep apnea creates difficulties to follow sleep recommendations). Thus, different interpretations on confidence may be made between patients with mild HF and those with severe HF. However, there is no empirical evidence to support such speculation, and future qualitative studies are warranted to investigate whether such a postulation is supported.

Limitations

There are some limitations to the present study. Firstly, given that only Iranian individuals with HF were recruited, the classical test theory results cannot be generalized to other populations regardless of their diseases or ethnicities. Secondly, although it is proposed that health-care providers can use the eHEALS to decide whether using Internet resources is appropriate for their patients with HF, the study findings did not provide any suggested cutoff for their reference. The results of the present study only provided information that eHEALS scores are robust and reliable. However, it is unclear how an individual with HF scores the eHEALS is a potential candidate to be recommended to use online resources. Future studies are warranted to determine the cutoff. More specifically, an intervention design using online resources should be conducted to observe individuals with HF and to which eHEALS scores respond the best to the intervention. Consequently, health-care providers would have good insight of using eHEALS score in clinical decision-making. Thirdly, eHEALS may not fully capture the complex concept of the eHealth literacy, and thus, eHEALS may not be a comprehensive tool for in-depth understanding of eHealth literacy. Nevertheless, the benefits of eHEALS (e.g., the strong psychometric properties, brevity, and utility) outweigh its shortcoming, and the eHEALS arguably serves as a convenient tool for health practitioners in busy clinical settings. Finally, all the instruments used in the present study, including eHEALS, were self-report in nature. Therefore, the research team was unable to control well-known biases such as social desirability and memory recall.

Conclusion

In conclusion, the eHEALS is a promising and useful tool for health-care providers to capture the eHealth literacy for individuals with HF. Health-care providers may use the eHEALS score to make further clinical decisions as to whether their patients with HF should use (or not use) online resources in health promotion and maintenance. Anecdotally, it is also worth noting that some health-care providers claim that their patients trust information they find online more than information recommended by their doctors or nurses. Therefore, it is especially important for patients to know how to evaluate online information they find and to use the information correctly to make good decisions given that health-related information found on the Internet can be wrong, exaggerated, unverified, unproved, and/or commercial.

Supplemental Material

Supplemental Material, 181024Appendix - Psychometric Evaluation of the Persian eHealth Literacy Scale (eHEALS) Among Elder Iranians With Heart Failure

Supplemental Material, 181024Appendix for Psychometric Evaluation of the Persian eHealth Literacy Scale (eHEALS) Among Elder Iranians With Heart Failure by Chung-Ying Lin, Anders Broström, Mark D. Griffiths and Amir H. Pakpour in Evaluation & the Health Professions

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iDs

Chung-Ying Lin

Amir H. Pakpour

Supplemental Material

Supplemental material for this article is available online.

References

Aponte

Nokes

K. M.

(2015). Electronic health literacy of older Hispanics with diabetes. Health Promotion International, 32, 482–489.

Aponte

Nokes

K. M.

(2017). Validating an electronic Health Literacy Scale in an older Hispanic population. Journal of Clinical Nursing, 26, 2703–2711.

Benjamin

E. J.

Blaha

M. J.

Chiuve

S. E.

Cushman

Das

S. R.

Deo

… Gillespie

(2017). Heart disease and stroke statistics—2017 update: A report from the American heart association. Circulation, 135, e146.

Cannon

J. A.

Moffitt

Perez-Moreno

A. C.

Walters

M. R.

Broomfield

N. M.

McMurray

J. J.

Quinn

T. J.

(2017). Cognitive impairment and heart failure: Systematic review and meta-analysis. Journal of Cardiac Failure, 23, 464–475.

Chang

C. C.

J. A.

Chang

K. C.

Lin

C. Y.

Koschorke

Thornicroft

(2018). Perceived stigma of caregivers: Psychometric evaluation for devaluation of consumer families scale. International Journal of Clinical and Health Psychology, 18, 170–178.

Chang

K. C.

Wang

J. D.

Tang

H. P.

Cheng

C. M.

Lin

C. Y.

(2014). Psychometric evaluation, using Rasch analysis, of the WHOQOL-BREF in heroin-dependent people undergoing methadone maintenance treatment: Further item validation. Health and Quality of Life Outcomes, 12, 148.

Chen

F. F.

(2007). Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural Equation Modeling, 14, 464–504.

Diviani

Dima

A. L.

Schulz

P. J.

(2017). A psychometric analysis of the Italian version of the eHealth literacy scale using item response and classical test theory methods. Journal of Medical Internet Research, 19, 4.

Gheorghiade

Vaduganathan

Fonarow

G. C.

Bonow

R. O.

(2013). Rehospitalization for heart failure: Problems and perspectives. Journal of the American College of Cardiology, 61, 391–403.

10.

Grady

K. L.

(2008). Self-care and quality of life outcomes in heart failure patients. Journal of Cardiovascular Nursing, 23, 285–292.

11.

Hyde

L. L.

Boyes

A. W.

Evans

T. J.

Mackenzie

L. J.

Sanson-Fisher

(2018). Three-factor structure of the eHealth literacy scale among magnetic resonance imaging and computed tomography outpatients: A confirmatory factor analysis. JMIR Human Factors, 5, e6.

12.

Jaarsma

Årestedt

K. F.

Mårtensson

Dracup

Strömberg

(2009). The European Heart Failure Self-care Behaviour Scale revised into a nine-item scale (EHFScB-9): A reliable and valid international instrument. European Journal of Heart Failure, 11, 99–105.

13.

Jaarsma

Cameron

Riegel

Stromberg

(2017). Factors related to self-care in heart failure patients according to the middle-range theory of self-care of chronic illness: A literature update. Current Heart Failure Reports, 14, 71–77.

14.

Koo

Norman

C. D.

Chang

H. M.

(2012). Psychometric evaluation of a Chinese version of the eHealth Literacy Scale (eHEALS) in school age children. International Electronic Journal of Health Education, 1, 29–36.

15.

Lee

C. S.

Bidwell

J. T.

Paturzo

Alvaro

Cocchieri

Jaarsma

… Vellone

(2018). Patterns of self-care and clinical events in a cohort of adults with heart failure: 1 year follow-up. Heart & Lung, 47, 40–46.

16.

Lin

C. Y.

Nikoobakht

Broström

Arestedt

Pakpour

H. A.

(2018). Validation of the 5-item medication adherence report scale in older stroke patients in Iran. Journal of Cardiovascular Nursing. Advance online publication. doi:10.1097/JCN.0000000000000488

17.

Lin

C. Y.

Pakpour

A. H.

(2017). Using Hospital Anxiety and Depression Scale (HADS) on patients with epilepsy: Confirmatory factor analysis and Rasch models. Seizure, 45, 42–46.

18.

Lin

C. Y.

Pakpour

A. H.

Broström

Fridlund

Årestedt

Strömberg

Mårtensson

(2018). Psychometric properties of the 9-item European Heart Failure Self-care Behavior Scale using confirmatory factor analysis and Rasch analysis among Iranian patients. Journal of Cardiovascular Nursing, 33, 281–288.

19.

McDonald

R. P.

M. H. R.

(2002). Principles and practice in reporting structural equation analyses. Psychological Methods, 7, 64–82.

20.

Melholt

Joensson

Spindler

Hansen

Andreasen

J. J.

Nielsen

… Dinesen

B. I.

(2018). Cardiac patients’ experiences with a telerehabilitation web portal: Implications for eHealth literacy. Patient Education and Counseling, 101, 854–861.

21.

Mitsutake

Shibata

Ishii

Oka

(2012). Association of eHealth literacy with colorectal cancer knowledge and screening practice among internet users in Japan. Journal of Medical Internet Research, 14, e153.

22.

Montazeri

Vahdaninia

Mousavi

S. J.

Omidvari

(2009). The Iranian version of 12-item Short Form health survey (SF-12): Factor structure, internal consistency and construct validity. BMC Public Health, 9, 341.

23.

Nguyen

Moorhouse

Curbow

Christie

Walsh-Childers

Islam

(2016). Construct validity of the eHealth literacy scale (eHEALS) among two adult populations: A Rasch analysis. JMIR Public Health and Surveillance, 2, e24.

24.

Norman

C. D.

Skinner

H. A.

(2006). eHEALS: The eHealth literacy scale. Journal of Medical Internet Research, 8, e27.

25.

Orlowski

J. L.

Oermann

M. H.

Shaw-Kokot

(2013). Evaluation of heart failure websites for patient education. Advanced Emergency Nursing Journal, 35, 240–246.

26.

Paige

S. R.

Krieger

J. L.

Stellefson

Alber

J. M.

(2017). eHealth literacy in chronic disease patients: An item response theory analysis of the eHealth literacy scale (eHEALS). Patient Education and Counseling, 100, 320–326.

27.

Pew Research Center. (2016). Smartphone ownership and internet usage continues to climb in emerging economies. Retrieved August 16, 2018, from http://www.pewglobal.org/2016/02/22/smartphone-ownership-and-internet-usage-continues-to-climb-in-emerging-economies/

28.

Rustad

J. K.

Stern

T. A.

Hebert

K. A.

Musselman

D. L.

(2013). Diagnosis and treatment of depression in patients with congestive heart failure: A review of the literature. The Primary Care Companion for CNS Disorders, 15, pii: PCC.13r01511.

29.

Sedlar

Lainscak

Mårtensson

Strömberg

Jaarsma

Farkas

(2017). Factors related to self-care behaviours in heart failure: A systematic review of European heart failure self-care behaviour scale studies. European Journal of Cardiovascular Nursing, 16, 272–282.

30.

Sudbury-Riley

FitzPatrick

Schulz

P. J.

(2017). Exploring the measurement properties of the eHealth Literacy Scale (eHEALS) among baby boomers: A multinational test of measurement invariance. Journal of Medical Internet Research, 19, e53.

31.

Van der Vaart

van Deursen

A. J.

Drossaert

C. H.

Taal

van Dijk

J. A.

van de Laar

M. A.

(2011). Does the eHealth Literacy Scale (eHEALS) measure what it intends to measure? Validation of a Dutch version of the eHEALS in two adult populations. Journal of Medical Internet Research, 13, e86.

32.

Vaportzis

Clausen

M. G.

Gow

A. J.

(2017). Older adults perceptions of technology and barriers to interacting with tablet computers: A focus group study. Frontiers in Psychology, 8, 1687.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.37 MB