A Scoping Review of Health Literacy Measurement Tools in the Context of Cardiovascular Health

Background. Despite the strong link between health literacy and cardiovascular health outcomes, health literacy measurements remain flawed and fragmented. There exists a gap in the knowledge when formulating a valid measurement to capture the broad concept of health literacy. The existence of various tools for health literacy measurement also hampers the availability of health literacy data. Additionally, little research is available on a valid measurement tool for cardiovascular health literacy. Objective. This study aims to provide an overview of the health literacy measurement tools used in the context of cardiovascular health. Method. A scoping review was conducted. Two electronic databases, Medline and Embase, were searched to identify studies that described a tool for the measurement of health literacy in the context of cardiovascular health. Results. After reviewing the available studies, 53 studies met the inclusion criteria. A total of 26 health literacy measurement tools were identified in the studies. Among the 26 tools, 16 used an objective measurement approach, 9 adopted a subjective approach, and 1 employed a mixed approach. Additionally, 28 studies used tools to measure print literacy, 15 studies measured print literacy and numeracy, and 5 studies measured print literacy, oral literacy, and numeracy. Conclusions. STOFHLA, TOFHLA, and REALM were the mostly commonly used tools in the selected studies. The majority of tools were based heavily on reading skills and word recognition. Researchers should focus on the development of more comprehensive and reliable health literacy measurement tool(s) specific to cardiovascular health to assist health care providers to more efficiently and accurately identify people with cardiovascular problems who have inadequate health literacy.

abilities are assessed by having them solve tasks dealing with print literacy, oral literacy, and/or numeracy, whereas a subjective approach is characterized by the self-report of respondents' own perceived abilities in multiple domains. The Rapid Estimate of Adult Literacy in Medicine (REALM), the Medical Term Recognition Test (METER), the Newest Vital Sign (NVS), and the Test of Functional Health Literacy in Adults (TOFHLA) are commonly used tools for the measurement of HL in the literature.
Regarding the domains that the different tools measure, early HL measurements focused exclusively on reading capacity, seeking to establish links between reading skills and health outcomes (Dickson-Swift, Kenny, Farmer, Gussy, & Larkins, 2014). Contemporary measurements extend beyond simply the capacity to read. Many researchers suggest that HL tools should focus on writing, numeracy, speaking, listening capacity, and understanding of the health care system (Nutbeam, 2000;Parker & Kreps, 2005;Sorensen et al., 2012). It is believed that these concepts are central for promoting self-management and individual responsibility for health care (Berkman, Sheridan, Donahue, Halpern, & Crotty, 2011;Rothman, Montori, Cherrington, & Pignone, 2008).

Health Literacy and Cardiovascular Health
Cardiovascular disease (CVD) is a term used to describe all diseases of the heart and blood vessels, and an estimated 17.5 million people died of CVDs in 2012, representing 31% of all global deaths (World Health Organization, 2012). Cardiovascular health (CVH) emphasizes a more integrative definition of health to include protective biological factors and behaviors, but it has not been investigated in relation to individuals' perspectives of their own health (Manczuk et al., 2017). Cardiovascular HL can therefore refer to the degree to which individuals have the capacity to obtain, process, and understand basic health information and services needed to make appropriate health decisions to manage their CVH.
In particular, a growing number of scientific reports have shown that HL plays a significant role in the management of CVD (Safeer, Cooke, & Keenan, 2006). Limited HL is a major barrier that blocks many people from achieving good CVH or benefitting from effective treatment for myocardial infarction, heart failure, stroke, and other CVDs (Manczuk et al., 2017). Most CVDs can be prevented by addressing behavioral risk factors such as tobacco use, unhealthy diet and obesity, physical inactivity, and harmful use of alcohol through populationwide strategies. At the same time, evidence has accumulated that HL is correlated with healthy behavior and better health outcomes. Improvement of cardiovascular HL is significant to the management of CVH. The identification of HL levels among cardiovascular patients and the general populations would also enable the provision of tailored-made support in prevention and disease management.
Thus, it is important to understand which HL measurement tools have been used among people with CVDs in different settings. This study identifies current cardiovascular HL measurement tools in the literature and provides some information on the advantages and limitations of the identified tools. To the best of our knowledge, there has not yet been a scoping review that specifically addresses this question in the context of cardiovascular health.

Objective of the Study
A scoping review of the literature was conducted using the framework approach (Arksey & O'Malley, 2005) to identify the tools that currently exist for the measurement of cardiovascular HL. Scoping reviews allow researchers to examine the extent, range, and nature of research activities on a specific topic, to summarize and disseminate research findings, and to identify research gaps in the existing literature (Arksey & O'Malley, 2005;Berkman et al., 2010).

Search Strategy
A comprehensive literature search of the Medline and Embase databases was conducted for all available years to March 31, 2017 in English and Chinese. Only original research articles focusing on cardiovascular health literacy measurements published in peer reviewed journals were included. Two researchers and two health sciences librarians at the University of Oslo Medical Library were included in the study.

Search Terms
In order to capture all studies to contribute to a thorough review, the final search terms included were [(health AND literacy) OR (cardiovascular AND literacy)] AND (cardiovascular OR heart OR vascular OR arrhythmias OR cardiac OR myocardial OR aneurysm OR aortic OR cerebrovascular OR stroke) AND (instrument* OR tool* OR questionnaire* OR survey* OR interview*).

Study Selection and Charting the Data
After removal of duplicates, articles were screened in order by titles, abstracts, and then full text. The screening process was performed independently by two authors, and any disagreement was resolved through discussion until consensus was achieved. Data were extracted by the first author and cross-checked by the second author.
Our systematic search of the electronic databases yielded 740 potential articles. After deletion of duplicates, 555 articles remained for the analysis. Irrelevant studies that were not original research, did not use a tool to measure HL, or had no cardiovascular focus were screened out (see Figure 1). At the full text screening stage, 31 articles were excluded because they were disease-specific systematic review studies, studies of HL level among caregivers without a specific HL tool, measurements of health professionals' knowledge of health literacy without any HL tool, educational intervention programs for families without any HL tool, or ehealth and health literacy studies without HL tools.
As a result of the full text review, 53 articles were included in this study. Using Microsoft Excel 2010 (Microsoft Corporation, Redmond, WA) spreadsheet, the studies were charted and summaries developed that included author, journal, publication year, research question or aim, setting, sample, and instrument used (see Table 1).
The study analyzed the tools according to the domain they measured. The findings of the review showed that the studies measured print literacy only (n = 28), print literacy and numeracy (n = 15), print literacy, oral literacy, and numeracy (n = 5), print literacy and oral literacy (n = 4), or oral literacy only (n = 1).

Discussion
This scoping review aimed to provide a better understanding of current cardiovascular HL measurement tools in the literature. The discussion of results is handled under four categories: approaches used in the HL tools, common tools, measurement mode, and cultural settings.

Objective, Subjective, or Mixed Approach
The findings of the research show that the majority of the studies used an objective approach for the measurement of HL in the CVH context. An advantage of using an objective approach is that it avoids the responses being subjected to a social desirability bias, which decreases the validity of the measurement (Cho et al., 2008;McNaughton et al., 2013;Reading et al., 2017).
However, the research shows that tools with a subjective measurement approach often address wider domains of HL such as patient-provider encounters; interaction with the health care system; rights and responsibilities; health information-seeking; understanding, processing, and using health care information; and communication with health care professionals (Begoray & Kwan, 2012).
A mixed-measurement approach was only detected in one tool from the screened studies . The mixed-measurement approach may portray the advantages of both objective and subjective approaches (Wu et al., 2010). The instruments used in objective and subjective approaches are different in their administration, methodology, and construct meaning; thus, researchers should expect inconsistencies in the ways in which they measure the varied constructs of health literacy (Peterson et al., 2011;Battersby et al., 1989;(Parker et al., 1995).

Implications from Commonly Used Tools
The results of the study reported that the most commonly used tool in the measurement of HL in the CVH context was STOFHLA developed by Parker, Baker, Williams, and Nurss (1995).
The original version, the TOFHLA, also developed by Parker et al. (1995), was also widely used. The TOFHLA is composed of two major sections, namely reading comprehension and numeracy. TOFHLA reading comprehension requires respondents to read three health-related passages and fill in missing words as they read along in 12 minutes. The numeracy session takes 10 minutes to complete and involves respondents reading numerical information and answering 17 questions using simple math skills related to prescription labels, appointment cards, simple directions, and financial information. The STOFHLA, which is an abbreviated version of the TOFHLA reading comprehension section, consists of only two passages and 36 fill-in-theblanks and takes only 7 minutes (Parker et al., 1995).
Both TOFHLA and STOFHLA are validated in English and Spanish and have been used in a wide variety of clinical settings and cardiac populations (Peterson et al., 2011). However, according to Hickey, Sciacca, Gonzalez, Castillo, andFrulla (2015), they are not easily administered in the clinical arena. TOFHLA and STOFHLA may be too long or tedious for patients to complete in a busy clinical setting. Taking into consideration the suitability of the tool for use in everyday clinical settings is essential so that screening patients for low HL in clinical practices can be feasible, low cost, and clinically advantageous. By identifying patients with low HL, doctors can give them additional time and resources to support their understanding of health information, best treatment options, and adherence to treatment.
The study found that the second widely used tool for the measurement of HL among patients with cardiovascular problems was REALM. REALM was developed by Davis et al.
(1993) involving respondents reading out loud a list of 66 medical words arranged in increasing order of difficulty. The REALM score is calculated by awarding one point for each correctly pronounced words and nothing for each mispronounced or skipped word (Huang et al., 2015).
It takes 2-3 minutes to complete. The relative speed and ease of administering REALM compared to TOFHLA and STOFHLA means that it has the potential to serve as a screening tool in everyday clinical practice. However, it is not self-administered. REALM requires patients to read a list of words aloud, and a practitioner must be present to score pronunciation accuracy.
Another concern with REALM administration is the potential embarrassment of lowliteracy patients from struggling to read the words in front of another person, as patients with low literacy have reported a feeling of shame concerning their reading problems (Parikh, Parker, Nurss, Baker, & Williams, 1996). In addition, ambiguities in scoring (e.g., mispronunciations due to speaker's accent or other conditions like throat infections) might be another practical concern.
The findings of this study show that the most commonly used HL measurement tools in the literature are TOFHLA, STOFHLA, and REALM (Parker, 2000). Despite the comparability provided by these commonly used HL measurement tools, the administration methods for these tools have some practical limitations for use in clinical settings that should be taken into account in the development of a new tool in the future. For example, the STOFHLA is the most frequently used measure of HL, but it only measures reading fluency, leaving out key domains in HL, and is often not feasible for use in clinical settings due to limited time and resources for administering and scoring the measure.

Measurement Mode
The findings of this scoping review show that the majority of the tools focus on the measurement of print literacy and reading comprehension. Less attention is paid to the measurement of numeracy or oral literacy. Research has pointed out a lack of recognition of communication skills (oral literacy) in the HL measurement construct in many tools (Pleasant, McKinney, & Rikard, 2011). The incorporation of CVD HL can expand HL measurement to a broader and more comprehensive understanding considering the communicative and interactional requirements of health care, thus increasing the validity of the tool. Furthermore, 24 of the studies in the review were presented as multi-dimensional measurements. Given the fact that HL is a multi-dimensional construct that encompasses print literacy, oral literacy, numeracy, and so on, the use of one-dimensional measurement modes prevalent in HL measurement should be subjected to further evaluation. In addition, despite 24 of the tools being presented as multi-dimensional measurements, the majority of tools principally assessed print literacy. REALM, for instance, only tests recognition and pronunciation of medical words and, thus, has limited ability to measure the full concept of HL.
As discussed earlier, tools with an objective approach generally tend to use a one-dimensional measurement mode.
Another finding is that despite the attempt to measure HL in the context of CVH, in many cases the tools used were general HL measurement tools, such as TOFHLA, STOFHLA, REALM, and some other tools, like NVS and METER. Although these are widely used tools for the measurement of HL, they assess HL with a broad and extensive scope rather than disease-or context-specific HL (Battersby et al., 1989).
More recent developments have attempted to measure HL specific to the context of CVD. Such tools, however, tended to measure a specific type of CVD. For example, the SAQ was developed with the aim of examining the knowledge of stroke risk factors and warning signs in the adult populations (Hickey, Holly, McGee, Conroy, & Shelley, 2012). Additional studies could be beneficial to develop more comprehensive tool/s to capture HL within the specific context of CVH. It is important to develop tools consisting of very short questions or few questions because application of the instruments will take a long time to complete due to the intensive work pace in clinics and the busy schedules of the health professionals working in these clinics.

Limitations of the Study
First, a systematic review of the available literature on cardiovascular HL measures was not conducted, so there is no information available on the quality of the studies selected. Second, the search was only conducted in Medline and EMBASE, which means that studies that were not registered on either database were not included. Third, only English and Chinese articles were included, and thus important studies in other languages may have been missed. Fourth, the results of the study are limited to the key search terms used in the research.

Conclusion
The results of this scoping review show that STOFHLA, TOFHLA, and REALM are widely used cardiovascular HL measurement tools. There is no gold standard instrument currently available to adequately assess the more global concept of CVD HL, including the interactions among different abilities. Commonly used tools are based heavily on reading skills, word recognition, and numeracy, and some tools measure general HL rather than CVH literacy specifically, while others narrowly measure knowledge of one CVD.
As a final word, CVH literacy is seen as central to lifelong engagement with health, building cognitive and social skills, and the motivation necessary to navigate health care systems, disease prevention, and health promotion throughout one's life. The current measurement of cardiovascular HL is fragmented; researchers should focus on the development of more comprehensive and reliable HL measurement tool/s specific to CVH to assist health care providers to more effectively and accurately identify people with CVDs who have inadequate HL in busy clinical settings. The tools that are developed should be sensitive enough to measure changes resulting from interventions. Assessing health literacy of hypertensive patients in a primary care setting using a