This is an open-access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in the Journal of Medical Internet Research, is properly cited. The complete bibliographic information, a link to the original publication on http://www.jmir.org/, as well as this copyright and license information must be included.
Low health literacy is associated with poor health-related knowledge, illness self-management, health service use, health, and survival, and thus addressing issues related to low health literacy has been highlighted as a pressing international priority.
To explore views of a digital health promotion intervention designed to be accessible to people with lower levels of health literacy, in particular examining reactions to the interactive and audiovisual elements of the intervention.
Qualitative think-aloud interviews were carried out with 65 adults with type 2 diabetes in the UK, Ireland, USA, Germany, and Austria, with purposive sampling to ensure representation of people with lower levels of health literacy. Inductive thematic analysis was used to identify common themes. We then systematically compared views in subgroups based on country, health literacy level, age, gender, and time since diagnosis.
Most participants from the chosen countries expressed positive views of most elements and features of the intervention. Some interactive and audiovisual elements required modification to increase their usability and perceived credibility and relevance. There were some differences in views based on age and gender, but very few differences relating to health literacy level or time since diagnosis.
In general, participants found the intervention content and format accessible, appropriate, engaging, and motivating. Digital interventions can and should be designed to be accessible and engaging for people with a wide range of health literacy levels.
Addressing the problem of low health literacy has been highlighted as a pressing international priority, as it is associated with poor health-related knowledge, illness self-management, health service use, health, and survival [
Many people have difficulty accessing face-to-face diabetes self-management education, due to barriers such as work, caring responsibilities, disability, cost, and lack of transport [
Interventions to reduce the literacy burden and improve health literacy have included using simple language; information presented in audio, audiovisual, or pictorial formats; (when Internet delivered) employing tailoring of content to individual’s needs; and other forms of interactivity. Reviews of the effectiveness of such interventions for general public and mixed patient populations [
As part of the Diabetes Literacy project [
The intervention was a website designed to motivate people with diabetes to increase their levels of physical activity. The interactive features of the website comprised tailoring of images and advice based on user responses to questions (eg, about the user’s age, concerns about physical activity, current activity levels), a quiz, and a physical activity planner. The audiovisual features included positive images and audiovisual sequences illustrating lifestyle physical activities. We followed established good practice for designing written medical information and making it accessible to people of all literacy levels [
The intervention was developed in consultation with an expert panel of patient representatives, clinicians, and behavioral scientists. We employed the person-based approach to intervention development [
Interactive elements of the intervention
Delivery of information about the health benefits of physical activity for people with diabetes (and health risks of inactivity) in the form of an interactive “fun” quiz.
Tailored advice in response to questions about current physical activity level and concerns about physical activity (eg, barriers such as cost, health problems).
Tailoring of images based on users’ reported age.
A physical activity planner that enabled users to create personal plans for increasing physical activity, building on their current activities.
Audiovisual elements of the intervention
Positive visual images throughout the intervention.
Audiovisual sequences modeling people undertaking a range of physical activities: these comprised a narrative illustrated by a sequence of photographs with a voice-over, and had an informal tone intended to suggest real-life scenarios.
Option to access details of study aims and procedures in audio format.
Good practice design features of the intervention
To maximize accessibility to those with low levels of literacy, text as short as possible, suitable for reading age of 12.
To maximize accessibility to those with low levels of computer literacy, simple page layout and navigation.
Advice appropriate for those with low current levels of physical activity, limited time, motivation and resources, and health problems (ie, promoted gradual increase in preferred lifestyle-compatible activities).
Screenshots illustrating elements of the Healthy Living with Diabetes website. (A) Example of positive visual image on welcome page. (B) Example of quiz feedback with positive visual image. (C) Example of audiovisual sequence. (D) Example of interactive physical activity planner.
Qualitative interviews were carried out with adults with type 2 diabetes in the UK, Ireland, USA, Germany, and Austria. Full details of all study procedures, participants, and analyses are published online [
Qualitative think-aloud methods [
Audio recordings of interviews were transcribed verbatim, translated into English where necessary, and checked for accuracy by the researcher who carried out the interview. Few substantial differences in views between countries were identified (see [
The relatively large sample size (for qualitative research) allowed us to carry out a second stage of analysis to explore whether any differences could be identified between specific subgroups of participants. We did this by creating tables to systematically compare the occurrence and content of themes across subgroups (see ). First, country comparisons were carried out to examine whether there were differences in the data between countries that might preclude pooling the data across countries for analysis. Only the data from the German sample appeared systematically different, largely in terms of less views being expressed on most topics. This finding may have been related to deviations from the interview protocol evident in the transcripts (eg, a less open-ended interviewing style than in other countries, omission of substantial parts of the interview schedule). In view of this difference, the 6 German interviews were excluded from the subsequent subgroup analyses. Further subgroup analyses on the remaining pooled data comprised comparisons based on level of health literacy (low, intermediate, or high), age (under or over 60 years), gender, and time since diagnosis (less or more than 5 years).The cutoff points were chosen so as to create roughly equal groups (to maximize the sample size in each) that were likely to differ in terms of experiences of diabetes (based on the research team’s clinical experience of diabetes patients). Because of the small size of the subgroups and large number of comparisons, to avoid overinterpretation of minor differences we adopted a criterion of only reporting comparisons where the differences observed between the groups compared were based on at least five people (between country comparisons) or 30% (n=19) of participants (pooled data), or suggested a consistent trend from high to low health literacy (even if not quite reaching our criterion within the specific subgroups).
In total, 65 participants were interviewed for this study, comprising 35 from the UK, 8 from USA, Ireland and Austria, and 6 from Germany (
Participant characteristics.
Participant characteristics | Country | ||||||
|
|
UK | USA | Ireland | Austria | Germany | Overall |
|
|
|
|
|
|
|
|
|
Mean | 58 | 57 | 66 | 60 | 67 | 62 |
|
Range | 44-75 | 49-64 | 37-77 | 42-79 | 48-77 | 37-79 |
|
≥60 years | 18 | 3 | 6 | 3 | 4 | 34 |
|
<60 years | 17 | 5 | 2 | 5 | 2 | 31 |
|
|
|
|
|
|
|
|
|
<5 years | 14 | 4 | 4 | 2 | 1 | 25 |
|
≥5 years | 21 | 4 | 4 | 6 | 5 | 40 |
|
|
|
|
|
|
|
|
|
Male | 18 | 3 | 6 | 6 | 4 | 37 |
|
Female | 17 | 5 | 2 | 2 | 2 | 28 |
|
|
|
|
|
|
|
|
|
High | 16 | 4 | 6 | 7 | 5 | 38 |
|
Intermediate | 13 | 2 | 2 | 0 | 1 | 18 |
|
Low | 6 | 1 | 0 | 1 | 0 | 8 |
|
Not known | 0 | 1 | 0 | 0 | 0 | 1 |
|
|
|
|
|
|
|
|
|
White/Caucasian | 32 | 2 | 7 | 8 | 6 | 55 |
|
Black/African/Caribbean | 2 | 2 | 0 | 0 | 0 | 4 |
|
Asian | 1 | 0 | 1 | 0 | 0 | 2 |
|
Other (mixed) | 0 | 4 | 0 | 0 | 0 | 4 |
Total sample |
|
35 | 8 | 8 | 8 | 6 | 65 |
Thematic analysis of the data pooled across countries generated 40 codes, which we organized into 3 main themes, namely, (1) general reactions to website content and format; (2) reactions to interactive features; and (3) reactions to audiovisual features.
The majority of users described the level of information in the intervention as appropriate and easy to understand, without being patronizing (
I like the whole thing. It didn’t blind you with science; it didn’t treat you as a total idiot.
It’s written in a way that’s not—because I mean I’m not good at reading and stuff like that—but it’s written in a way that I can understand. You know, sometimes you look at, whether it’s books or websites or whatever, and sometimes you’re reading and you think “What are they talking about?”
The thing is that all the knowledge I received during hours of lectures [at “wellness clinics”] is briefly put together here, so really, really well done for this!
No, it’s very straight forward and compact. What you need to know is there.
Participants frequently spoke appreciatively of acquiring new information through the website and the positive framing of health information was described as encouraging.
Yeah, I know they say physical, mental activity can, can help you stay healthy [reads to self], “twice,” woah, “twice as likely!” Get out, really? I don’t remember ever hearing that one.
To find out that Alzheimer’s thing, yes that was quite shocking, a sit up and take notice moment.
A number of people discussed their intention to increase their physical activity as a direct result of what they had learned or seen in the website.
Interesting, I did not know that—so it's almost like, if I learn nothing else from this survey, I need to start my physical activity regime.
Specific intervention features described as motivating included the planner and audiovisual sequences as well as learning new health information through the quiz. Details of how participants viewed these features are given below.
Reactions to website content and format.
Subtheme | Content |
Information novelty | Almost all participants mentioned learning new, often surprising information (particularly about the benefits of physical activity for preventing Alzheimer’s disease and for healthy liver function) |
Level of website advice | Most participants felt that the advice was delivered at the right level—easy to understand but not patronizing |
Views of website advice | The vast majority of participants endorsed the advice given by the website. |
Many were surprised and some skeptical about the information that physical activity is more important than controlling blood sugar levels for preventing complications from diabetes. | |
The humor in the website was mainly appreciated by the minority who commented on it. | |
Views on website appearance | Most participants commented that they found the website clear and easy to navigate. |
Some expressed a desire for greater simplicity in presentation and less text. | |
Effects of website on motivation | Most participants (though not all) found the website generally motivating. |
All elements of website content were described by some participants as motivating them to engage in greater physical activity. |
Many participants commented that the interactive features of the website were engaging and motivating (
I’m getting really curious now to see the answers [both laugh], it’s lovely, it’s not boring, the whole Web thing is very good. It’s very interesting.
I liked the quizzes, I liked that—you know—it’s nice to have something you can use, interact with and join in with.
If I was doing this on my own I would skip this bit [quiz section], I’m bored with that now, because it's treating me like a child. I want the information but I don't like the way it's given to you.
Participants described valuing the instant tailored feedback, and often commented on encouragement provided by the positive framing of this advice.
Woo hoo—“doing enough physical activity to keep healthy”—now says star pupil actually, that’s a bonus.
Well done, you got it right, you got something right. See, they love me. Finally! [laughs] They love me.
However, participants had more mixed responses to the activity planner, some finding it cumbersome and difficult to complete, particularly (but not exclusively) before modifications were made to simplify it. For example, the original planner required participants to specify the activity type and the amount of time spent on the activity on the same page. Many participants found this difficult to navigate, and the planner was subsequently separated onto 2 pages.
Reactions to the interactive features.
Subtheme | Content |
Views of the interactive quiz | Most comments on the quiz were that it was enjoyable and informative. |
Some found it irritating, and disliked the humorous “trick” questions (ie, that physical activity would not improve hearing or alcohol consumption). | |
Views of tailored feedback | Most comments on the feedback were that participants appreciated getting immediate, positive feedback. |
Participants noted that actively engaging with the website kept their attention. | |
Views on the interactive activity planner | Some participants found using the planner difficult, due mainly to uncertainty about (1) how to estimate their activity level and (2) how to enter data into the planner (before simplification of the planner) |
The vast majority of participants found the images acceptable or liked them overall (
These pictures are quite nice as well, you know, they are happy pictures—photographs of elderly people, probably my age, dancing with umbrellas.
A few participants mentioned disliking certain images, although the images that people disliked varied. For example, some younger participants could not relate to images of older people; consequently, the intervention was modified to tailor all images by age.
Most participants also enjoyed the audiovisual sequences, appreciating the informal style and relating positively to the stories and example activities these sequences narrated.
The videos were really, really good...informative.
It was good, straight to the point, because it’s like, not everybody likes going to the gym and that. It’s too expensive, too hot to try to stay and that, and the person that was speaking was really clear.
Some participants commented positively on the use of humor in the main audiovisual sequence (which showed pictures of snowboarding and deep sea diving as examples of impractical forms of exercise before suggesting walking as more feasible) but some found it confusing or inappropriate. In addition, some people disliked particular scenarios that they found unconvincing or irrelevant:
I don’t like it at all. It doesn’t fit the pictures, there’s an old man who says he’s picking up his children. He says his day is hectic yet he can go shopping before work and during lunch. It doesn’t fit.
Some of the scenarios therefore required modification for different cultures or age groups; for example, a scenario that suggested playing football was changed to baseball for US participants.
Reactions to the audiovisual features.
Subtheme | Content |
Views of visual images in the website | Overall, use of images in website appreciated as positive, attractive. |
Images of walking and family activities generally appreciated, medical illustrations mainly well received. | |
Reactions to some images mixed (prior to modification), for example, images of older people, wheelchair activity, alcohol. | |
Views of audiovisual sequences in the website | Most participants were very positive about the audiovisual sequences, most (though not all) liking the informal style, and finding them engaging, funny. |
Most people enjoyed relating to stories that they saw as realistic, helpful examples of lifestyle activity. | |
Negative comments were often based on seeing specific content (prior to modification) as unrealistic or irrelevant to the participant’s own situation (eg, due to cultural or lifestyle differences or activity preferences). |
We found few differences in how people with high, intermediate, or low levels of health literacy viewed the website. The only differences that emerged were that people with higher health literacy were most likely to comment that they found the website content easy to understand, mention features of the website they found motivating, discuss how the interactivity maintained their attention, and express appreciation of the visual images. There were also few systematic differences in views linked to country, although there were some culture-specific preferences (as noted earlier); for example, the Austrian participants tended to dislike the audiovisual sequences, commenting negatively on the use of speakers with German accents.
With regard to other subgroup comparisons, women were more likely than men to voice their intention to be more physically active as a result of viewing the website. Women tended to express much more positive views of the quiz than men did; some men found it tedious or irrelevant. Women were also more positive about the audiovisual sequences, describing them as engaging and discussing relating to the characters and stories. Women, however, were less technically confident in completing the interactive planner. There were age differences in the activities preferred; those aged over 60 mainly intended to do more walking, whereas younger participants also spoke about cycling and swimming. Participants over 60 years of age and those who had been diagnosed for longer were more positive about the audiovisual sequences (particularly the walking stories) and the idea of exercising at home. This age group was also more likely to mention that they found the level of information in the website straightforward, helpful, or pitched at the correct level.
The main aim of this study was to investigate whether it is possible to design a digital intervention that is acceptable and engaging for people with varying levels of health literacy. Our findings are encouraging; most participants from most countries expressed positive views about most elements of our digital intervention. Very few people found the accessible format patronizing or the information provided inadequate or inappropriate. Despite concerns frequently expressed about whether it is possible to appeal to a wide and diverse target population, it is not unusual to find that materials designed to be accessible to those with lower levels of literacy and health literacy are also liked by those with higher literacy levels [
Although most participants had mainly positive reactions to the interactive and audiovisual presentation of advice, it was necessary to iteratively modify these based on participant feedback to optimize acceptability and feasibility. There were also differences in reactions to these elements of the intervention relating to age, gender, and cultural context. We were only able to evaluate views of a very limited set of digital materials, and it is quite possible that the views expressed were specific to these particular resources. For example, reactions to professionally produced videos of patients or actors might be entirely different from reactions to our amateur-style audiovisual sequences. Nevertheless, some previous research has also noted that it can be difficult to produce audiovisual narratives that precisely match the psychosocial context of all users [
The method we used for comparing views across subgroups is unusual; it can be regarded as an extension of the constant comparison technique employed in grounded theory, but with more systematic and explicit assessment of the frequency with which views were expressed in different subgroups. Because these subgroups were too small to permit reliable quantitative evaluation, it is not possible to interpret the trends observed in our data as definitive evidence for the presence or absence of group differences. Nevertheless, the consistent absence of any variations in views clearly associated with health literacy and time since diagnosis, despite clear differences based on age and gender, suggests the latter may have had a more important influence on reactions to the intervention. However, we were unable to recruit many participants with the lowest levels of health literacy, and the sample size in each country was small. Consequently, this study was only able to identify substantial differences in views due to country or very low health literacy (ie, views that would be expressed by most people in these subgroups). Most participants were also white, and these findings may therefore not be generalizable to people from other ethnic backgrounds. A further limitation in our ability to fully investigate the perspective of users with lower levels of health literacy is that these participants provided fewer comments about the website and our analysis of the articulated views of users was unable to capture nonverbal indications of accessibility or engagement barriers, such as pauses and silences. Further research into the views of people with very low health literacy is required.
It is important to be aware that, despite our best efforts to encourage participants to freely express negative views, some participants may have been reluctant to do so. For example, it is possible that women did not find the intervention more engaging than men, but were less willing to express negative views of it. It is also important to remember that for the purpose of health promotion it is necessary but not sufficient for an intervention to be acceptable and engaging—it must also be effective in achieving the intended behavioral outcomes. For this reason, we are now undertaking a large trial to test the effectiveness of our digital intervention for promoting health literacy improvement (in knowledge, understanding, and self-efficacy) and behavioral intentions with regard to increased physical activity.
Most participants found the intervention generally acceptable and engaging. Surprisingly, reactions were similar—and equally positive—among those with higher and lower levels of health literacy. This finding has importance for the reach and cost effectiveness of digital health care, because it suggests that it may not be necessary to develop multiple versions of interventions for people with differing levels of health literacy. However, our sample did not include people with the very lowest levels of health literacy, who may well require different online or offline interventions.
Some marked variations in preferences for how advice was presented did emerge; for example, many people particularly appreciated the interactive quizzes and audiovisual sequences, whereas a minority strongly disliked them. These variations in preferences were partly linked to age, gender, and culture but were not closely mapped onto demographic and psychosocial characteristics, making it difficult to prescribe exactly what format should be used for which population subgroup. This suggests that perhaps a major benefit of Internet delivery of health promotion is that it is possible to offer recipients a choice of formats, allowing them to self-select those that they find most accessible, attractive, and useful. This approach permits users to engage in what has been termed “self-tailoring” [
These findings have clear implications for those who develop health-related websites and digital interventions; these can and should be designed to be accessible and engaging for people with a wide range of levels of health literacy, to help to overcome the “digital divide” and reduce health inequalities. This can be achieved by following the established good design principles we drew on and then using findings from iterative qualitative research to maximize the perceived relevance, credibility, and feasibility of the intervention for different members of the target population. Incorporating interactive and audiovisual elements to increase interest and engagement may also be useful but requires careful development to ensure that they are appropriate for people of both genders, different ages, and from different cultures. Finally, with regard to international dissemination of digital interventions, our findings indicate that the acceptability of interventions is likely to be similar across different countries but can also be improved by making modifications on the basis of feedback from interviews to increase the perceived relevance to the specific cultural context.
This research was part of the Diabetes Literacy project supported by a grant of the European Commission (Grant No FP7-Health-2012-Innovation-1/306186). The authors would like to thank the experts and members of the Diabetes Literacy Study Group who contributed by commenting on the intervention content, design, and implementation, in particular Robyn Dixon, Paul Roderick, Deborah Morrison, Richard Holt, Wendy Lowe, Pal Kingshuk, Scott Lloyd, Brian Williams, and Kath Barnard. We would especially like to thank our two lay advisors, Andy Hamson and Wendy Heath, and all of the people who contributed their time and efforts to the audiovisual content of the website. We are also grateful to those who helped translate and transcribe the data. CDB is supported in part by the Southampton National Institute for Health Research Biomedical Research Centre.
None declared.