Introduction

Health literacy is related to individuals’ knowledge, motivation, and competencies to access, understand, appraise, and apply health information, and take appropriate decisions that are relevant to health promotion, disease prevention, and self-care management [1]. Because of research and advocacy by health literacy experts, inappropriate use of health services, poor health care quality, and adverse health are no longer considered as results of patients’ limited health literacy. Rather, healthcare providers and organizations are being held responsible for creating a “health literate” system that enables patients to find, understand, and use information to inform health-related decisions for themselves and their families.

Evidence suggests that healthcare providers equipped with health literacy and communication skills contribute positively to reducing health literacy related barriers and improving healthcare quality and patients’ health outcomes [2, 3]. However, studies have shown that health providers have a tendency to overestimate patients’ health literacy and they lack adequate health literacy competency to appropriately respond to and handle patients’ low health literacy issues [4, 5]. As medical technologies continue to advance and the healthcare delivery systems become more complex, increasing healthcare providers’ health literacy competency has become more crucial than ever [6, 7].

The Institute of Medicine has recommended health-related professional schools and professional continuing education programs, including medicine, dentistry, nursing, and other professionals, to develop health literacy curricula and training programs [8]. Similarly, many medical educators and researchers have urged the need to integrate health literacy training into the medical education curriculum [9]. These recommendations have thus far led to limited progress. Few undergraduate or continuing medical education programs have successfully incorporate health literacy in their curricula [6, 10]. Few curricular standards exist that address the need for health literacy training. With few exceptions where students may be exposed to health literacy concepts and practices in independent courses or during clinical rotations [2, 3], current learning models lack in-depth health literacy contents that included essential competencies and applications of health literacy principles [2, 5].

Coleman points out that a crucial challenge in integrating health literacy into existing health professional curricula is the lack of clear and widely accepted guidelines for defining and evaluating the contents of health literacy curricula across health professional programs [11]. Another challenge is that health-related professional schools and continuing education programs have limited resources, including instructional hours, financial resources, and faculty availability, to integrate health literacy into the full curriculum even when they acknowledge the importance of health literacy training [12]. As such, an evidence-based health literacy competency guideline that enlists and prioritizes a set of measurable core health literacy knowledge, attitudes, skills, and practices may enable health professional schools and programs to effectively design health literacy training programs and set clinical practice standards.

To rationally prioritize educational competencies for health literacy training, Coleman and colleagues [13] conducted an extensive literature review of health literacy competencies and practices and employed a Delphi consensus process to develop a set of measurable knowledge, skill, attitude, and practice elements to assess health professionals’ health literacy competencies. This set of educational competencies for health professionals has been refined and validated in the US and Europe [14, 15], and has led to prioritization of important health literacy practices [16]. Thus far, no similar work has been conducted in Asia.

In Taiwan, researchers surveyed nurses regarding their health literacy knowledge and the results showed that nurses answered merely half of the questions correctly [17]. Chang et al., (2016) and Chang et al. (2017) developed a health literacy assessment tool for health professionals. Their instrument was constructed predominately by combining items from several domestic surveys [18, 19]. Although the instrument may reflect specific contexts of the Taiwanese healthcare delivery system, it is focused on health literacy knowledge and does not take into account work by international scholars who have systematically identified health literacy competencies essential for health professionals. Moreover, the instrument was not designed for guiding the development of health literacy curricula.

The purpose of this study was to develop health literacy competencies for health professionals in Taiwan based on the work of Coleman and colleagues [13] and to prioritize the importance of health literacy practices in clinical settings. Our study results contribute to promoting cross-cultural application of health literacy competencies and facilitating the development of health literacy curricula in health professional education and training programs.

Methods

We first translated the set of health literacy competencies developed by Coleman et al. [13], and then underwent a modified Delphi process.

Translation of health literacy competencies for health professionals

We followed a standardized forward translation and back translation method developed by the World Health Organization to translate health literacy competencies from English to Chinese Mandarin [20]. Two native Chinese speakers, who were proficient in English, were involved in the forward translation; one had medical background and the other was a linguist. Each of them independently translated the set of health literacy competencies for health professionals [13]. They then discussed with the research team and compared their translations to reach a consensus on a final version of the translation.

Two native English speakers proficient in Chinese enlisted in the back translation. Similarly, one had medical and the other linguistic backgrounds. Without knowledge of the original wording of those health literacy competencies, each of them independently translated the Chinese version into English. Our research team then engaged all four translators to compare the two back-translated versions. Adjustments were made and a final version was agreed upon by everyone involved.

A modified Delphi process

A three-round modified Delphi survey was applied in this study. The modified Delphi process was conducted anonymously and online.

Expert selection

Delphi experts were identified based on their expertise and leadership in health profession education in their respective fields as well as their knowledge of health literacy. Considering the short span of health literacy research in Taiwan (approximately 15 years), we employed the following eligibility criteria in the selection of Delphi experts: (1) the expert had three or more years of work experience in the field of health literacy and published at least one original article that listed them as a first author; (2) the expert had five or more years of work experience in the field of health literacy and had completed at least one systematic review article that listed them as a first author or corresponding author; or (3) the expert had been devoted to health literacy-related work for five or more years. De Villiers et al. (2005) suggested that the size of the expert panel for a Delphi study ranged from 15 to 30 [21]. In this study, 22 health literacy experts from various professional areas were invited, and 20 were finally recruited with purposive sampling to participate in 3 rounds of a Delphi process, The Delphi process was conducted in 2017.

Round 1 of Delphi

In Round 1 of the Delphi process, the expert panel was provided with 62 items of health literacy competencies to rate. Moreover, the expert panel was provided with the original and translated versions of items to review to ensure conceptual and cultural equivalence and to suggest changes [20, 22, 23].

The rating was focused on the appropriateness and importance of those items using a 4-point Likert-type scale ranging from 1 (very appropriate/important) to 4 (not appropriate/important). A priori cutoff point was set similar to the previous consensus level determined in the US and European studies, with consensus being defined as 70% or more of the expert panel agreement on the appropriateness and importance of an item [13, 14]. Items with less than 30% of agreement were removed. Experts were encouraged to recommend new items.

Round 2 of Delphi

During the second round, the experts received the quantitative results of the previous round – i.e., percentages of the panelists considering items as appropriate/important – along with anonymous suggestions. The expert panel was then asked to assess whether each item should be considered a core health literacy competency for health professionals. Items that at least 70% of the panelists considered core were retained. The panelists also discussed conceptual equivalence of the translation and the wording of each item. Items reflecting similar concepts were merged and necessary wording changes were made in accordance to the panelists’ suggestions.

During this round, the Q method was applied to prioritize the importance of health literacy practices based on their likely impact on patients. The aim of the analysis was to help healthcare providers and organizations ration their resources and time by focusing on health literacy practices that have the greatest impact. The Q method is a validated technique that has been applied to prioritize educational competencies for medical education [24]. In this study, the Q method followed the method used by Coleman et al. [13]. Each expert panel provided 32 items of health literacy practice and then placed each item in order of importance on the developed e-format Q sorting board.

Round 3 of Delphi

In the final round of consensus, the expert panel received the results of the retained competency items, the importance order of health literacy practices, and anonymous suggestions in the preceding rounds. The panelists scrutinized the final version and reached a consensus on health literacy competencies and practices. Although not a focus of this paper, the panelists also recommended additional work needed to turn health literacy competencies and practices into a measurable scale for evaluation purposes and suggested that the measurement could use a Likert-scale format.

Data analysis

SPSS Version 22.0 and Microsoft Excel 2019 were utilized to analyze the quantitative data. Descriptive statistics, including means, standard deviations, and percentages, were applied. In implementing the Q method, we calculated the weighted mean scores of health literacy practice items and arranged them into a grid (Fig. 1), from highly important to least important.

Fig. 1
figure 1

The ranking in the importance of health literacy practice. Footnote: The numbers represent mean ratings out of 9 and standard deviation in parentheses. The highest mean ratings of practice items on the left and the lowest mean ratings on the right

Full size image

Ethical approval

The Institutional Board of the National Yang Ming University reviewed the protocol and approved the study through an expedited review (YM104142E).

Results

A total of 20 experts participated in the Delphi study. Most of the experts were female (65%) and professors and researchers (65%). Their educational backgrounds included clinical medicine (25%), nursing (20%), pharmacy (10%), and public health (25%), and communication or linguistics (20%) (Table 1).

Table 1 Demographics characteristics of Delphi expert panel
Full size table

Health literacy competencies retained after the Delphi process, including 12 knowledge items, 9 attitude items, and 21 skill items, are shown in Tables 2, 3, and 4, respectively. During round 1, 57 out of 62 (91.9%) health literacy competency items were accepted. The mean of the appropriateness rating by the panelists ranged from 1.15 to 2.70 and the mean of the importance rating ranged from 1.00 to 2.25 (Appendix Table 1). Based on the group consensus criterion, 22 out of 24 knowledge items, 10 out of 11 attitude items, and 25 out of 27 skill items were considered appropriate and important. No additional health literacy knowledge, attitudes, or skills were suggested in this round. In Round 2, 44 out of 57 items considered as core competencies achieved consensus (12 knowledge items, 9 attitude items, and 23 skill items). In Round 3, based on the panel’s feedback, 3 skill items that were similar were merged into 1. Overall, 42 competency items were retained, including 12 knowledge items, 9 attitude items, and 21 skill items.

Table 2 List of health literacy knowledge
Full size table
Table 3 List of health literacy attitude
Full size table
Table 4 List of health literacy skill
Full size table

Figure 1 displays the result of the final version of Q sorts with the mean and standard deviation scores ranging from 7.6 (SD ± 1.5) to 2.5 (SD ± 1.2). The Q-sort grid showed the highest mean ratings of practice items on the left and the lowest mean ratings on the right. Additionally, we calculated the number and percentage of 20 experts who rated each of the items in Group 1 at an important level of ≥ 7 to check whether mean item ratings could be the result of an outlier. The findings showed the percentage ranging between 45 and 70%, suggesting little influence from outlier opinions. Table 5 shows the original and translated versions of health literacy practices in order of importance. “Avoiding using medical jargon” was ranked as the most important health literacy practice, followed by “speaking slowly and clearly with patients” and “using analogies and examples, avoiding idioms and metaphors, to help make oral and written information more meaningful to patients.”

Table 5 Health literacy practice ranking
Full size table

Finally, the panelists suggested a Likert scale format to measure health literacy competencies, either self-administered or observational.

Discussion

This study employed a modified Delphi process to adapt a set of evidence-based health literacy knowledge, attitudes, and skills to the contexts of Taiwan’s healthcare delivery system and to guide the development of health literacy curricula for health professionals in Taiwan. Moreover, we utilized the Q-sorting technique to assign different degrees of importance to health literacy practices so as to inform healthcare providers and organizations of practices that are most beneficial to patients in alleviating barriers due to inadequate health literacy.

In this study, 91.9% consensus was achieved regarding the appropriateness and importance of 57 out of 62 health literacy competency items among a panel of Taiwanese experts. A previous European study had an expert group census of 90% [14]. Altogether, the results suggest a high level of expert agreement geographically across the US, Europe, and Taiwan. The understanding of health literacy competencies in terms of knowledge, attitude, and skills for health professionals shows similarities across Western and Eastern countries, despite the differences in healthcare delivery systems and sociocultural contexts. We further identified a core set of health literacy knowledge, attitudes, and skills required by health professionals. In total, 42 items reached a consensus, which collectively serve as a framework for guiding the development of curricula for enhancing health professionals’ health literacy competencies.

Health literacy practices are patient-centered strategies and behaviors that minimize the negative consequences of limited health literacy. The top-rated health literacy practice item in this study is “avoiding using medical jargon,” which is consistent with a previous American study wherein “avoiding medical jargon” was the second top-rated item. Health professionals commonly overestimate patients’ health literacy and unintentionally use medical jargon [25]. Use of medical jargon creates barriers to effective communication with patients, particularly among those with low health literacy, and limits patients’ ability to fully understand the medical information provided by health professionals [26]. Experts consistently agree that avoiding medical jargon is crucial and should be routinely practiced by health professionals in clinical encounters.

In Taiwan, health professionals learn medical knowledge by reading English or translated texts and most of the terminology has no common, plain language substitutes [27]. The process of learning, while increasing the medical expertise of health professionals, may lower their sensitivity to the use of medical jargon that is unfamiliar to most patients. Clearly, there is a need for health professionals to learn how to appropriately and clearly explain medical conditions. Simulations of clinical encounters with standardized patients, for example, may be helpful. Such simulations can increase medical professionals’ awareness of their use of medical jargon and offer them opportunities to practice communication in plain language. As gamification becomes a useful pedagogical tool in medical education [28], educators may develop digital-based learning games that allow students to be aware of medical jargon and learn the corresponding plain language or interpretation that they might replace.

“Teach-back” was rated the number one health literacy practice in an American study [16]. In contrast, our panelists rated the practice as the seventh. Research suggests that teach-back is an effective and feasible method to enhance patients’ health literacy and improve patients’ health outcomes [29, 30]. Patients reported satisfaction with their interactions with health professionals when the teach-back methods are applied. Health professionals also perceived improvement in their health communication with patients. However, there is concern that teach-back may evoke patients’ adverse emotions, even stigma [29]. It is likely that this concern may be more prominent, and the teach-back may be less appropriate and effective, in a collectivist culture. This may explain the lower rating of the teach-back practice by a Taiwanese Delphi panel in comparison to the result of a U.S study.

Given the strong evidence supporting the advantages of teach-back, the challenges of health professionals’ training include educating health professionals not only on routinely practicing teach-back techniques but also on methods to avoid making patients suspect their learning ability or feel stressed as if they were taking a test. The Agency for Healthcare Research and Quality [31] unveiled a set of toolkits, such as asking for non-shaming, open-ended questions or using plain languages, that utilized teach-back more efficiently, and provided good training material resources.

It is also interesting to note that the eight top-rated health literacy practice items our panelists prioritized are not identical to the eight top-rated items identified in the previous American study. Notably, four out of eight items our panelists affirmatively rated in Group 1 were not found in the previous study, including “speaks slowly and clearly,” “using analogies and examples,” “avoiding idioms and metaphors,” “active listening,” and “ask patient’s preferred learning style in a non-shaming manner.” The differences may be due to variations in culture and health service delivery systems. Items related to avoiding idioms and metaphors in a non-shaming manner could be a cultural issue. Eastern cultures, which tend to be collective, are characterized by an indirect communication style and maintain interpersonal harmony [32]. This style is a less overtly conflictual or aggressive form of communication, and individuals are more likely to use idioms and metaphors that may hinder clear communication in clinical encounters. In terms of “speak slowly” and “active listening” items, it may reflect the global budget payment system of the National Health Insurance program in Taiwan that compels healthcare providers to have a high volume of clinical work, thus lacking the time and patience for clear communication [33]. Observing patients’ nonverbal behavior and decoding proverbs and metaphors appear even more critical in indirect communication culture. Accordingly, our panelists are especially aware of and address these practices for routine use across health professionals to increase the clarity of communication and help mitigate the negative effects of low health literacy.

Currently, shared decision-making has become a focus for promoting patients’ engagement and empowerment. It is not surprising to note that two items, “eliciting the full list of patient concerns” and “negotiating a mutual agenda,” are two of the eight top-rated items in the American study and our research. However, previous studies showed that patients might have the desire to participate in health discussions and decision-making, despite the stress they may feel due to insufficient medical knowledge. Health professionals are patients’ advocates and they should encourage and support patients to gain a full understanding of health information, options, concerns, and rights during the medical decision process. It is clear that success in adhering to health promotion and self-care management behaviors relies on clear communication and taking patients’ concerns and preferences into account. Accordingly, “eliciting the full list of patient concerns” and “negotiating a mutual agenda” are essential health literacy practices for clinical and public health professionals that should not be neglected.

This study replicated US consensus studies to refine and validate health literacy competencies for health professionals in Taiwan. More than 90% high agreement in American, European [14], and Asian research indicated that the items of Coleman et al. [13] can be applied as indicators of health literacy competencies and practices across diverse healthcare and cultural contexts. In contrast to previous studies that employed local expert opinions to develop health professionals’ health literacy competencies [18, 19], the advantage of this set of measurable health literacy knowledge, attitudes, skills, and practices for health professionals is that it is comprehensive and evidence-based and that it has achieved consensus across global expert panelists. Health profession educators should feel confident in using this set of competencies and practices as core learning objectives for developing health literacy training programs and selecting the contents to match the learning activities and expected outcomes [13]. Additionally, the translated practice items would be beneficial for health literacy training programs among Mandarin-speaking health professionals.

This study has several limitations. First, although the health literacy competencies and practices examined in this study have considerable validity as they are based on accumulated international evidence [13], the results of consensus can only represent expert opinion. Our findings do not reflect the opinions of healthcare professional students or patients. Second, although the experts on our panel had diverse disciplinary backgrounds, not all health professions and medical disciplines were represented in our study. It is possible that the consensus results may vary depending upon the disciplines and areas of expertise represented on the expert panel. Insufficient representation is a limitation of recruitment and increases the likelihood of selection bias. Third, the list of health literacy competency items in this study was based on a US study and a review of literature that may be dated. Future studies should consider bring the literature review up to date and incorporate new health literacy competencies and practices identified in more recent research.

Conclusions

This consensus study used a modified Delphi method to appraise a set of health literacy competencies and practices derived from a US study for its appropriateness, importance, and prioritization for health professionals in Taiwan. The high consensus across global health literacy experts suggests that the selected core health literacy competencies and practices could be used as guidelines and clinical assessments across healthcare professions. This set of health literacy competencies can provide a sound basis for developing health professional curricula or continuing education programs to enhance health literacy competencies and clear communication practices.