Care for Cognition: Science communication using different modalities of communication

(1)

Care for Cognition

Science communication using different

modalities of communication

Floor de Boer

(10764372)

Research Master Brain and Cognitive Sciences (UvA)

Internship Report

Research Project 2 (36 ECTS)

First Examiner: Prof. Dr. Iris Sommer (RUG) Second Examiner: Dr. Eline Smit (UvA)

Department: Cognitive Neuropsychiatry, UMC Groningen

Date: 25 – 02 – 2021

(2)

2

Abstract

Cognitive impairment is a prevalent symptom of many brain disorders and diseases. It is important for the empowerment of patients and their health literacy to inform them about recent research findings in a way they prefer and understand. The aim of this study is to investigate how people with brain disorders and cognitive impairment prefer to be informed about recent findings in research. Three types of materials were developed: text, image, and video. The participants were asked to select their preferred modality of communication (or a combination) and to rate their satisfaction of the texts, images, and videos. By asking for feedback about the materials, the main points of action could be established. Fifty-six people participated in the research, 27 participants were diagnosed with a brain disease or disorder and 29 participants were allocated to the control group. ‘Image’ is mostly preferred by both the diagnosed group and the control group, however the difference between ‘image’ and ‘text’ and ‘image’ and ‘video’ was greater for the diagnosed group. A lower satisfaction of the materials was correlated to a higher subjective cognitive impairment. A lower percentage of preference for a multimodal option for communication was associated with a high subjective cognitive impairment. In conclusion, when scientific findings are communicated to people with a brain disorder or disease with cognitive impairment, it is preferred to communicate using an image over a video or text. However, to accommodate as many people as possible it might be useful to invest in text as well. Both image and text are easy to update when new scientific findings are published in contrast to video.

Keywords: cognition; cognitive impairment; health communication; learning preferences; multimodal learning; science communication.

(3)

3

1. Introduction

1.1 Literature review

A well-functioning cognition is crucial and important for every human being. Problems with cognition could lead to many secondary problems, including a reduced ability to work, less social interaction, and more depressive symptoms[1,2]. Many (mental) health issues are accompanied by cognitive decline, including psychotic disorders[3], neurodegenerative diseases[2,4] and depression[5,6], as well as cognitive decline caused by postoperative complications[7]. For most disorders, cognition is not the focus of the treatment. For example, for schizophrenia the main treatment option is antipsychotic medication. However, this treatment has no effect on cognition, but only on the psychotic symptoms of the disorder[8]. Furthermore, cognition is a major determiner for quality of life[2,4,9,10] and functional capability[11].

A lot of research has been performed on cognitive rehabilitation and it continues to be a point of interest. Many different areas are identified to have an effect on cognition including physical activity, medication, computer training, and nutrition (e.g.[12–16]). Therefore, there are treatment options established and proven to have a (moderate) effect on cognition or subdomains of cognition (e.g.[12–16]). However, many of these possibilities are not fully known by patients and are not completely applied in clinical practice[16].

It is important to find ways to inform patients and their families about cognition and the problems most of these patients face. Also, to inform them about how they might be able to improve their cognition, or at least decrease further decline, as well as recent developments in research. Since various disorders and diseases all present cognitive symptoms, one platform might be sufficient for all patients. In order to reach the patients and to explain cognition and the possible cognitive rehabilitation treatments, more information will be incorporated in the website of the Dutch Brain Foundation (Hersenstichting: www.hersenstichting.nl). However, it is important to study in which way the information needs to be presented in order to accommodate the patients the best. The information could be presented in various way, including in verbal (text), visual (images and graphics), audio (voice-over), or audio-visual (videos/animations).

Health communication is important, because when patients’ knowledge about their condition is increased, their empowerment improves[17]. This is related to an increased health literacy[18], referring to the ability to understand basic health related information[19]. When health communication is altered to both the level of health literacy and the preferred learning style of the patient, the knowledge of the patient about his or her own condition improves[20,21].

(5)

5

Most people have a preferred learning modality or modality of communication, such as visual, verbal or auditory learning. However, previous research found that multimodal information presentation does inform people better compared to singular mode[22]. For example, adding icons or graphics to medical information increases the knowledge about the medical procedures and help to rate the risks of the procedures better than only numbers[23], as visual presentation of medical information is important[24]. Research has found that mode tailoring, selecting the modality of communication that the user prefers, works well to inform people and to increase the overall satisfaction of the website[25]. Research has found that mode tailoring improved recall of information for older adults (>65 years)[26]. This research showed that older participants did select all three modalities (multimodality)[26]. However, these participants often did not finish or watch the audio-visual mode (video)[26].

Most studies mentioned above are performed with healthy controls instead of patients[25,26]. The healthy controls had to imagine how they would react in a certain scenario. This is also the case for the design of many hospital websites[27]. Since, (older) adults with poorer cognitive functioning and higher cognitive decline are at risk for low health literacy[28], it is important to let patients not only check the content (understandability), but also the design of the website (usability)[29]. By this, all information is accessible for this group of patients as well. Therefore, it is important to study the preferences of these patients, since the preferences could differ between people with and without cognitive problems.

1.2 Aim

The aim of this research is two-fold. Firstly, the aim is to investigate how people with brain disorders and cognitive impairment prefer to be informed about recent findings in research. Secondly, the aim is to learn how to develop clear explanatory materials in (neuro)science, in this particular case for patients with cognitive problems and their families. To reach these aims, three research questions were asked.

1.3 Research questions and hypotheses

RQ1) Which modality of communication is most preferred and satisfactory by with a diagnosis of a brain disease or disorder to explain current research about cognition?

H1) It is expected that the preference for a modality of communication, is for a more visual modality of communication, because research has shown that adding icons and graphics increases the understandability of the information better[23]. Furthermore, it is expected that a multimodality of communication is preferred instead of a singular modality of communication[22].

(6)

6

RQ2) Does subjective cognitive impairment have an effect on the preference and satisfaction of a modality of communication?

H2) It is expected that the extend of subjective cognitive impairment does influence the preferred modality of communication. Participants with higher level of subjective cognitive impairment are expected to prefer less text, but more visual or audio-visual modalities of communication. This is expected, because research has shown that older participants, who also suffer from cognitive decline, did select more modalities than younger participants[26].

RQ3) When developing science communication materials in multiple modalities of communication, where do you need to pay attention to?

H3) It is expected that developing the science communication materials in multiple modalities of communication is based on a lot of feedback from the participants. As discussed in the study by Nguyen et al. 2017[29], a redesign based on the feedback is needed to optimize the materials in both the content as the design. Therefore, it is important to mainly focus on how people with problems with cognition would understand and use the materials. It is expected that the feedback is mainly negative, focussing on what to improve. The most feedback will be regarding the video, because a video is the most difficult to develop.

(7)

7

2. Methodology

2.1 Participants

The participants were recruited mainly via social media. The Dutch Brain Foundation shared the questionnaire on their Facebook page. Furthermore, the foundation shared the questionnaire with patient organisations and asked them to share the questionnaire with their patients. Since the questionnaire was spread via social media, both people with and without a diagnosis of brain disorders participated in the study. No exclusion criteria were established, since the assumption stands that everyone who filled out the questionnaire was able to do so individually.

2.2 Materials

The materials developed, were based on recently published meta-analyses about four potential treatment options for cognitive impairment: computer training[12], physical activity[14,30], brain stimulation[31] and anti-inflammatory agents[7,13]. Each research topic was ‘translated’ to be understandable for laymen in three different modalities of communication: text, video, and image. Text

The text was written by using the main findings of the meta-analyses. The length of the text was aimed to be around 350 words. Although, some texts were longer. No extra layout options were used for the text.

Video

The objective of the video was to explain the various research topics in a simple and clean way. For that reason, an animation was chosen. An animation, rather than a live action video, can explain what is happening inside the brain when having cognitive impairment. Also, it would be easier to keep the video simple and without any unnecessary distractions, when all the colours in the video could be chosen, which would not be the case when using live action. The colours used in the animation were chosen to be neutral, so patients would not experience oversensitivity towards the animation. A voice-over was used to explain the animations in the video. The length of the video was aimed to be around 2 minutes.

The animations were made using Microsoft Office PowerPoint. This program was chosen, because no extra payment was necessary and PowerPoint includes many icons and has standard animations, which limits the skills needed to use the program.

Image

The image was made based on the format of infographics. Since, an image cannot explain all the information and research findings as elaborately as texts or videos can do, the information had to be

(8)

8

short and only focused on the main findings and conclusions. The video was used to determine the main points of focus. Text was added to the image to explain the concepts and research findings more clearly than only using visual aids.

The image was made using Microsoft Office PowerPoint, Word, and Paint. The general layout was developed using PowerPoint, while the text was added in Word and the final image file was made using Paint. The colours used had to be neutral, but still inviting enough that people would read it.

2.3 Questionnaire

The questionnaire existed of three types of questions: the demographic, quantitative, and qualitative questions. The two latter type of questions were asked four times, for each research topic the same set of questions were repeated. First, the participants were asked to read the text, watch the video and to look at the images for the four topics. The order of the materials was different for each topic. Then, the quantitative and qualitative questions were asked.

Demographics

Firstly, in the demographic part, general questions about the participant were asked. These questions included gender, age, education level, whether the participant had been diagnosed with a brain disorder, and the extend of subjective cognitive problems.

Quantitative questions

Secondly, in the quantitative part the participants were asked to rate their satisfaction of each modality of communication separately. Furthermore, they were asked to select their preference (text, graphic or video) in each area. It was possible to tick multiple boxes, resulting in a preferred combination. To make sure that the participants were not favouring one modality of communication over another based on the order of presentation, the order was randomly presented, as stated before.

Qualitative questions

Thirdly, the qualitative questions asked the participants to give feedback on the materials and what they can be improved. In this way, the materials can be improved before they are published.

2.4 Variables

Subjective cognitive impairment

Subjective cognitive impairment was measured on a Likert scale of 1-7 of how much the participant experiences cognitive impairment. The lowest value 1 meant that no cognitive impairment was experienced, while the rate of 7 meant that extreme cognitive impairment was experienced by the patient. The variable is a subjective measurement.

(9)

9

To compare subjective cognitive impairment to the preference for a certain modality of communication, a median split has been performed. All values lower than the median, were grouped in the lower subjective cognitive impairment group. All other values, including the median and higher, were grouped into the high subjective cognitive impairment group.

Satisfaction of modality of communication

Every participant was asked to rate to what extend the information in every modality was satisfactory. The participants had to do this by rating the satisfaction on a Likert scale from 1, not satisfactory at all, to 7, very satisfactory. Every text, image, and video had an own question about the satisfaction, resulting in twelve separate questions about the satisfaction (each modality four times). To be able to draw a conclusion about each modality, the four questions of each modality were all incorporated into one variable, consequentially all participants had four inputs in each variable. This resulted into three satisfaction variables: ‘satisfaction of text’, ‘satisfaction of video’, and ‘satisfaction of image’.

Preference

The variable ‘preference’ was measured by the participant’s preference for one of the three modalities of communication or a combination. Multiple boxes could be ticked, resulting in seven possible answers. The question about one’s preference was asked four times, for each research subject once. In order to measure the total preference, all four separate preference questions were combined into one variable. Therefore, the number of answers of the preference questions is four times the number of participants. Every participant has four entries in this variable. All combinations of modalities, multimodal preferences, thus two or more ticked boxes in the questionnaire, are grouped into the category of ‘multimodality’. Furthermore, each category of preference was made into a new variable with a binominal input (e.g.: ‘preference text’: yes or no).

2.5 Statistical analysis

The analysis was performed in SPSS. Since most of the data is categorical, non-parametric tests were mainly used. For differences between groups for the ordinal variables, the Whitney Mann U test was used. Only for the continuous variable ‘age’ the independent sample T-test was used. For differences within groups for these variables, the Wilcoxon Signed Rank test was used. For the nominal variable ‘preference’, the Chi-Square test was used to determine if there was a different distribution between groups. The Chi-Square Goodness-of-Fit was performed in order to determine if there was a preference for a modality of communication or a combination. The Chi-Square test of Independence. When the expected count was less than five in over 20% of the cells, the Fisher Exact

(10)

10

test was used. For correlations, the Spearman’s Rank test was used instead of the Pearson correlation, because of the nominal and ordinal variables.

2.6 Qualitative analysis

The qualitative answers to the survey were used to review the materials in a later stage and to improve all the materials. The process of designing educational materials, is an ongoing process and it is important to incorporate the feedback of patients/users of the materials into the designing process[29]. In order to analyse the feedback, the feedback was coded in multiple categories based on which type of feedback was given. Using frequencies tables, the main point of focus to improve the materials could be determined.

(11)

11

3. Results

3.1 Demographics

In table 1, the demographics of the two groups are shown. In the control group 29 participants filled in the questionnaire. For the group with patients of brain disorder, 27 people participated. It is shown that the frequencies for gender do significantly vary between the two groups (diagnosed: 52% is female; control: 79% is female; p=0.030), as well as the distribution of education level (the diagnosed group has a lower education level; p=0.035) and subjective cognitive impairment (diagnosed: 4.00 (2.00 – 5.00); control: 2.00 (1.00 – 3.00); p=0.002). The age between the groups did not differ (diagnosed: 52.07 (14.05); control: 52.31 (15.34); p=0.847).

Table 1: Demographics for the diagnosed group and the control group

Diagnosed group (N=27) Control group (N=29) P-value Gendery_(f/m) _14/13 _23/6 _0.030* Age 52.07 (14.05) 52.31 (15.34) 0.847 Education levelx

Lower than secondary education Secondary education

Trade school

University of Applied Sciences University Bachelor University Master PhD 2.00 (2.00 – 3.00) 1 4 10 7 1 4 0 3.00 (2.00 – 5.00) 0 3 5 10 2 8 1 0.035*

Subj. cognitive impairmentx _{4.00 (2.00-5.00)} _{2.00 (1.00-3.00)} _0.002*

x_{Non-normally distributed variables for which median and interquartile range are presented; Tested with Mann Whitney U test.}

y_{Tested with the X}2_test

*Below significance level α = 0.05. f: female; m: male; subj.: subjective

3.2 Hypothesis 1: Preference and satisfaction by diagnosed group vs. control group

Preference for modality of communication

As tested for both the diagnosed groups and the control group with the Chi-Square test of goodness-of-fit, the null hypothesis that there is no preference for a certain modality of communication (the frequency for all categories is the same) is rejected (diagnosed: p=0.000; control: p=0.000). Table 2 shows the difference in preference for both groups. It is shown that the distribution of the frequencies for each category do significantly differ (p=0.002). When each category of modality of communication is compared between groups, it can be observed that only the categories ‘image and text’ (diagnosed: 1% vs. control: 13%; p=0.001) and ‘text, image, and video’ (diagnosed: 0% vs control: 4%; p=0.029) do significantly differ between groups.

(12)

12

Table 2: Preference for modality of communication

Figure 1, below, shows the preference for modality of communication for both the diagnosed group and the control group, as also shown by table 2. It is shown that image is most preferred for both the diagnosed group (40%) and control group (30%). For the diagnosed group, category ‘image’ did significantly differ from all other categories, including the category ‘text’ (p=0.041) and from the category ‘video’ (p=0.004). For the control group, no significant difference was observed between ‘image’ and the categories ‘text’ (p=0.085) and ‘video’ (p=0.378). Although, ‘text’ versus ‘image’ is close to significant. As shown in table 2, between the two groups, the three categories ‘text’ (p=0.353), ‘video’ (p=0.307), and ‘image’ (p=0.131) did not significantly differ from each other. Neither do the categories ‘text’ and ‘video’ significantly differ within the two groups.

Figure 1: Bar graph of preference for modality of communication. Diagnosed group vs. control group; preference given in percentages. *below significance level α = 0.05.

0% 5% 10% 15% 20% 25% 30% 35% 40% 45%

Text Video Image Image and video Image and text Text and video Text, image and video Multimodal Perc en ta ge o f p re fe re n ce

Preference for modality of communication

Diagnosed group Control group

Diagnosed group (N=27) Control group (N=29) P-value Preferencex _{3.00 (2.00 – 3.00)} _{3.00 (2.00 – 4.00)} _0.002* 1. Text 24% 19% 0.353 2. Video 19% 24% 0.307 3. Image 40% 30% 0.131 4. Image and video 9% 6% 0.364 5. Image and text 1% 13% 0.001* 6. Text and video 7% 3% 0.190 7. Text, image, and video 0% 4% 0.029* 8. Multimodaly _18% _27% _0.102

x_{Non-normally distributed ordinal variables for which median and interquartile range are presented; Tested with Chi-Square Test.}

Y_{Sum of all multimodal preference options (4-7).}

*Below significance level α = 0.05

* *

(13)

13

The last two bars in figure 1 show that more participants in the control group (27%) have chosen a multimodal combination compared to diagnosed participants (18%), however this difference is not significant (p=0.102). For the diagnosed group, image is significantly preferred over a multimodal combination (image: 40% vs. multimodal: 18%; p=0.002). However, for the control group, there is no significant difference between ‘multimodality’ (27%) and ‘image’ (30%; p=0.622).

Satisfaction of the materials

The satisfaction of the materials, as shown in table 3, are only for the modality ‘text’ significantly lower for participants who are diagnosed with a brain disorder than for participants in the control group (p=0.002). The median for ‘satisfaction of the text’ of the control group is 5 (out of 7), while the diagnosed group rates the satisfaction at 4 (out of 7).

Table 3: Satisfaction of the modalities of communication for both groups. Tested with the Whitney Mann U test

Furthermore, figure 2 shows the satisfaction of the different modalities of communication. For the diagnosed group, the satisfaction for text is significantly lower than for video (p=0.031) or image (p=0.001). For the control group, there are no significant differences between the three variables of satisfaction.

Figure 2: Box plot for satisfaction of the modalities of communication for both groups

Diagnosed group (N=27) Control group (N=29) P-value Satisfaction textx _{4.00 (3.00-5.00)} _{5.00 (4.00-6.00)} _0.002* Satisfaction videox _{5.00 (3.25-6.00)} _{5.00 (4.00-6.00)} _0.329 Satisfaction imagex _{5.00 (4.00-6.00)} _{6.00 (4.00-6.00)} _0.327

x_{Non-normally distributed ordinal variables for which median and interquartile range are presented; Tested with Whitney Mann U Test.}

(14)

14

3.3 Hypothesis 2: Association with subjective cognitive impairment

Satisfaction of the materials

When the rate of satisfaction is correlated with the subjective cognitive impairment score, see table 4 second column, it is shown that how higher participants rate their own cognitive impairment, how less satisfied they are with the various materials. This effect seems to be the greatest for ‘text’ (Rho= -0.317; p=0.000).

Table 4: Subjective cognitive impairment compared to satisfaction

By using the median split of subjective cognitive impairment, two groups were formed: low subjective cognitive impairment and high subjective cognitive impairment, whether a participant has a diagnosis for a brain disorder, has not been included. Table 5 shows the frequencies of the two variables ‘diagnosis’ and ‘subjective cognitive impairment’ (median split).

Table 5: Frequency table for the variables “diagnosis” and “subjective cognitive impairment” (median split)

As also shown in table 4, when comparing the two groups based on the median split of subjective cognitive impairment, the ‘satisfaction of text’ (p=0.000), ‘satisfaction of video’ (p=0.000), and ‘satisfaction of image’ (p=0.001) are all three significantly different between the two groups. For all categories, participants within the high subjective cognitive impairment do rate the satisfaction lower than the low subjective cognitive impairment group. For both groups the ‘satisfaction of text’ is rated significantly lower than the ‘satisfaction of image’ (high: p= 0.006; low: p=0.014), as also can be observed in figure 3 below.

Subjective Cognitive Impairmentx (N=56) High Subjective Cognitive Impairmenty (N=30) Low Subjective Cognitive Impairmenty (N=26) P-valuey Satisfaction text -0.317 0.000* 4.00 (3.00-5.75) 5.00 (4.00-6.00) 0.000* Satisfaction video -0.274 0.000* 5.00 (3.00-6.00) 6.00 (4.00-6.00) 0.000* Satisfaction image -0.293 0.001* 5.00 (4.00-6.00) 6.00 (5.00-6.00) 0.001*

x_{Tested with the Spearman Rank Test; Spearman’s Rho and p-value is given.}

y_{Non-normally distributed ordinal variables for which median and interquartile range are presented; Tested with}

Whitney Mann U Test.

*Below significance level α = 0.05.

High Subjective Cognitive Impairment (N=30) Low Subjective Cognitive Impairment (N=26) Diagnosed: Yes (N=27) 20 7 Diagnosed: No (N=29) 10 19

(15)

15

Figure 3: Satisfaction of the different modalities of communication by the median split of ‘subjective cognitive impairment’.

Preference for modality of communication

For both groups, the null hypothesis that for ‘preference’ all categories are equally preferred is rejected (high: p=0.000; low: p=0.000). A median split of ‘subjective cognitive impairment’ is not associated with a different distribution in preference (p=0.070), as tested with the Fisher Exact test, seen in table 6. Participants in the low subjective cognitive impairment, preferred the category ‘text and video’ significantly more than in the high subjective cognitive impairment group (p=0.009).

Table 6: Preference by median split of ‘subjective cognitive impairment’.

A significant difference for the preference for multimodalities between the low and high subjective cognitive impairment groups (p=0.012) is found. Participants within the group of low subjective

High Subjective Cognitive Impairment (N=30) Low Subjective Cognitive Impairment (N=26) P-value Preferencex _{3.00 (2.00-3.00)} _{3.00 (2.00-4.00)} _0.070 1. Text 24% 18% 0.284 2. Video 25% 17% 0.163 3. Image 35% 35% 0.952 4. Image and video 7% 9% 0.576 5. Image and text 7% 8% 0.767 6. Text and video 2% 10% 0.009* 7. Text, image, and video 1% 4% 0.129 8. Multimodaly _16% _30% _0.012*

x_{Non-normally distributed ordinal variables for which median and interquartile range are presented; Tested with Chi-Square Test.}

Y_{Sum of all multimodal preference options (4-7).}

(16)

16

cognitive impairment do in general prefer more often a multimodal option than the participants in the high subjective cognitive impairment group.

Figure 4 shows the percentages of the preference by low and high subjective cognitive impairment, as given by table 6. The preference for ‘image’ in the group ‘low subjective cognitive impairment’ is significantly higher than ‘text’ (p=0.022) and than ‘video’ (p=0.014), but not compared to ‘multimodality’ (p=0.541). The categories ‘video’ and ‘text’ versus ‘multimodality’ are close to significantly different (text: p=0.090; video: p=0.063). For the group ‘high subjective cognitive impairment’, no significant preference is observed for ‘image’ compared to ‘text’ (p=0.123) and ‘image’ compared to ‘video’ (p=0.157). Compared to ‘multimodality’, the category ‘image’ is more preferred (p=0.003).

Figure 4: Preference by the median split of ‘subjective cognitive impairment’. Given in percentages. *below significance level α = 0.05.

3.4 Hypothesis 3: Qualitative feedback

In total 336 points of feedback were given (diagnosed group: 176 (N=27); control group: 160 (N=29)). As shown in figure 5, the diagnosed group gave more negative feedback about the text than the control group did. The most negative feedback given by both groups was regarding the video. The image received the most positive feedback compared to text and video. Also, it can be observed that there was less negative feedback regarding the image than compared to the other two modalities.

0% 5% 10% 15% 20% 25% 30% 35% 40%

Text Video Image Image and video Image and text Text and video Text, image and video Multimodal

Preference by low and high subjective cognitive impairment

High Subjective Cognitive Impairment Low Subjective Cognitive Impairment

*

(17)

17

Figure 5: Percentage of positive and negative feedback by type of modality of communication.

The feedback was coded in several main points of feedback and further combined to result in 16 different categories of feedback. In figure 6, the top-8 of the categories of feedback are shown. Most feedback by the diagnosed group was related to the layout of the text, image, and video (diagnosed: 22%; control: 10%). The highest percentage for control group was ‘no feedback’ (control: 19%; diagnosed: 18%). Furthermore, the control group had more positive feedback than the diagnosed group (diagnosed: 3%; control: 14%), which can also be observed in figure 5 above.

Figure 6: Bar graph for frequency (in percentages) of feedback received

Specific feedback, which was repeated frequently, was that the voice-over of the video’s was not very engaging and sometimes speaking too fast (diagnosed: 11%; control: 13%). Also, the length of

0% 5% 10% 15% 20% 25% 30% 35% 40% General negative General positive Text negative Text positive Video negative Video positive Image negative Image positive

Feedback by type of modality of communication

Diagnosed group Control group

0% 5% 10% 15% 20% 25% No feedback Positive feedback

Layout Voice-over Too long Boring Language use Animation type

Feedback percentages for top-8 topics

(18)

18

the video and the text was often recalled as too long (diagnosed: 8%; control: 10%). Also, it was mentioned that mainly the videos were boring, the text used was not always appropriate to the target audience, and the type of animations were not very engaging.

(19)

19

4. Discussion

4.1 Summary

The aim of this study existed of two parts. The first aim was to investigate how people with brain disorders and cognitive impairment prefer to be informed about recent findings in research. Three types of materials were developed: text, image, and video. The participants were asked to select their preferred modality of communication (or a combination) and to rate their satisfaction of the texts, images, and videos. The second aim of this study was to investigate how to develop science translation materials. By asking for feedback about the materials, the main points of action could be established.

4.2 Discussion of the results

H1: Preference and satisfaction by diagnosis

The study has found that people both with and without a diagnosis of a brain disorder or disease did prefer the informative image about the research topics the most. The diagnosed group showed a difference between ‘text’ and ‘image’ and between ‘video’ and ‘image’. In both cases, image was preferred over the other two categories. For the control group, no difference between these three modalities of communication was found.

As expected, visual aids did help to explain difficult research topics to laymen[23]. However, it was expected that ‘video’ would have been preferred, which would have been in accordance with other studies[32,33]. The study by Lim et al.[32] found that video is better for information retention than text, while Meppelink et al.[33] concluded that animation with a voice-over is a good way to inform people with a low health literacy. Another study by Meppelink et al.[34] found that compared to text, images are better for information recall for people with low health literacy.

The satisfaction rate for text was lower for the diagnosed group than for the control group. For the diagnosed group, the satisfaction for ‘text’ was also lower than for the other two modalities of communication. The lower satisfaction of ‘text’ could be due to the length of the text compared to the image. Violante et al.[35] found in their study that visual features in an e-learning application do induce positive satisfaction. Satisfaction of e-learnings is found to be affected by the perceived usefulness, playfulness and multimedia content effectiveness by the users[36]. Therefore, it might be the case that for ‘text’ these factors were perceived low by both groups, resulting in a lower satisfaction for ‘text’. This value for playfulness could also be the reason why ‘text’ had a lower satisfaction rate than video did, even though ‘video’ and ‘text’ were preferred equally.

(20)

20 H2: Subjective cognitive impairment

The results showed that the higher ‘subjective cognitive impairment’ was rated on the Likert scale, the lower the satisfaction for all three modalities of communication was. In the same light, it is not a surprise that the ‘high subjective cognitive impairment’ group had a lower satisfaction rate for ‘text’ and ‘video’ than the ‘low subjective cognitive impairment’ group. This might be due to a too complex explanation for those with a higher rate of subjective cognitive impairment. The materials might contain too much information and are not clear enough in the explanation. Furthermore, the length might play a role, since the texts and videos are considerably longer to read and view than it takes to read and view the image. On the other hand, the image is quite simple and short, which makes in understandable for all.

Furthermore, it was shown that the preference for a certain modality of communication was more distributed for people with low subjective cognitive impairment than for people with high subjective cognitive impairment. As suggested before, this could be due to too complex explanations of the research topics in the modalities text and video. Only, one certain modality was, ‘image’, was perceived better than the other modalities.

Participants in the group of ‘low subjective cognitive impairment’ preferred multimodal options for communication more than the ‘high subjective cognitive impairment’ group. An explanation for this might be that the ‘high subjective cognitive impairment’ group had a difficulty with comparing the information between the modalities to see whether the information is the same, complementing or contradicting. Since, the information cannot be exactly the same in each modality of communication. This is called the “media gap”[37,38]. Since multitasking can be difficult for people with cognitive impairment and brain disorders or diseases[39], it might indicate why a single modality is preferred by the group with ‘high subjective cognitive impairment’.

H3: Qualitative feedback

When asked for qualitative feedback, various suggestions were made by the participants. In order to improve the materials, these suggestions should be implemented. Most feedback in general and the most negative feedback was concerning the video. The voice-over of the video should speak slower and more enthusiastically. A study by Ritzhaupt et al.[40] found that when a video was accelerated, the satisfaction decreased. This could indicate that a voice-over which speaks too fast could decrease the satisfaction. Many participants called the video “boring”, which could be decreased by a more enthusiastic voice, but also by decreasing the length of the video. Furthermore, participants missed a practical and real-life video. In the study by Lim et al.[32], videos consisted of photos and infographics instead of animations. In this study, the researchers did not ask for the satisfaction of

(21)

21

the video, but as mentioned before, the information retention was better than text[32]. A possible clarification could be that the motion of animations did distract the participants, which could lead to an inferior satisfaction[41]. The study by Dong & Goh[42] identified multiple tips for videos in medical education. They found that it is important to have the presenter occasionally on-screen and that cognitive overload should be avoided by having text, audio, and video all be integrated together with accompanying transcript[42]. Lastly, they gave the tip that it is important to have the videos be from professional quality in order to engage the students. This last point about professional quality is something that has returned in the feedback. Multiple participants made comments about the quality of the voice-over and the animations, suggesting that the professional quality of the videos is insufficient.

The feedback about the text mainly included that the length of the text should be shortened and that the layout should be changed. No layout options were chosen for the questionnaire to put more focus on the content instead on the layout. The suggestions given included that the colour of the text was too light (grey instead of black) and that bold and italic text should be used. By using bullet points to recap the main findings of a research article and by using bold text to emphasize important words and findings, the readability of the text can increase. Furthermore, the use of language should be improved, since some grammar mistakes were made and too difficult words were occasionally used.

The most positive feedback was regarding the images. Participants found them clear. However, quite some feedback regarded the layout of the images. The colours were sometimes perceived as too busy, while others said that the colours were too monotone. The images highlighted only the main findings of the studies explained in the materials. By this the content was presented in a simple, but not simplistic way, which regarding to Bogza et al.[43] is a successful way of explaining health information to older people. Furthermore, information layering, plain language and the use of pictograms was encouraged[43].

4.3 Implementations of the study

An implementation of this study is to improve websites and other e-health applications that are currently being developed. By implementing the results of this study, the health literacy among people with cognitive impairment might be able to improve. An important finding of this study is that the preference for ‘image’ was the highest both for diagnosed and control participant. This finding could lead to a greater focus on developing educational images instead of videos. Developing videos is much more time consuming than developing images, as well as more expensive, if the videos are be made professionally. Text could be used to complement the images to have as many

(22)

22

people as possible use their preferred modality of communication, even though multimodality is not preferred.

Furthermore, by knowing that the preference for a certain modality of communication is comparable between people diagnosed with a brain disorder or disease and controls, it is possible to assess the content of health websites (about cognition) with a varied group of participants instead of only patients. A test panel with patients might be more difficult to find, therefore controls might be able to help the website developers as well. However, it should be taken into account that people with a lower subjective cognitive impairment are easier satisfied with the information provided than people with high subjective cognitive impairment and also prefer more multimodal options.

4.4 Limitations

There are various limitations of this study. Firstly, the questionnaire was quite long to fill out (around thirty minutes). Since many participants did have (subjective) cognitive impairment and/or a diagnosis of a brain disorder or disease, it might have been too long to remain focused. Not everyone managed to watch all the videos for that reason. This could have led to less precise feedback on the materials at the end of the questionnaire. Secondly, the questionnaire took the aim of the materials out of context, since only the materials were presented instead of on a health website. Therefore, it might have been more difficult for the participants to understand the information given by the materials. Thirdly, the number of participants was not very high. For both groups (diagnosed and control) less than thirty participants took part in the study. More participants could have shown an even clearer view on the preference of the modality of communication. Lastly, the study and questionnaire were not checked by an ethical committee from an university (either the University of Amsterdam or the University of Groningen). Therefore, privacy sensitive information could not be asked, such as type of diagnosis, and length of diagnosis. This information could have been interesting to compare answers between diagnoses.

4.5 Further research

For further research, the study should be repeated when the feedback on the materials is incorporated to see whether there is a shift in the preference distribution. Since the videos received most (negative) feedback, much can be improved. Therefore, it might be possible that there will be a higher preference for the video, after implementing the feedback. Also, the feedback on the layout of the text could have an impact on the preference and satisfaction rate of the text. Furthermore, not only satisfaction and preference should be measured, but also the information retention. Not many studies measure preference and satisfaction, which makes it difficult to compare between studies. Also, when this study would be repeated, it would be interesting to investigate the type of

(23)

23

brain disorder or disease the participants have. This could give an insight in the assumption that patients with various diagnoses can all benefit from communication in the same way. Lastly, it would be interesting to study if the satisfaction of the website of the Dutch Brain Foundation will increase after the materials of the study are implemented. By this, we can study the effect within context instead of a questionnaire which is taken out of the context of the implementation.

4.6 Conclusion

In conclusion, in this study, the image is the preferred modality of communication and received the most positive feedback. The satisfaction of the text was the lowest and the videos received most negative feedback. A multimodal option was not preferred by participants with high subjective cognitive impairment. The materials should be improved by implementing the feedback, which could lead to different results, mainly regarding the video modality of communication. The images are easier to develop than a video, therefore it is advised to focus on developing educational images for the website instead of developing videos. Videos take a lot of time to develop and can be quite expensive, if done professionally. The images could be supported by text. More research should be performed to help websites to improve and to explain health information in a clearer way for all laymen and patients.

(24)

24

5. Acknowledgements

This research would not have been possible without the help of Prof. Iris Sommers from the University of Groningen. I am very thankful that she has been able to supervise me for almost a year, during this internship and during my literature thesis. Furthermore, the Dutch Brain Foundation has helped me with this research by publishing the questionnaire on their social media platforms. Thank you all very much for all the meetings and feedback during this process!

This internship report indicates the end of my student life. I am very grateful for all of those who have supported me throughout my entire university career, starting in 2014. It has been a hell of a ride, but I enjoyed it all! Thank you very much to my parents, brothers, and all my friends.

Lastly, I would like to dedicate this report to my grandfather who passed away the 20th_{of February.} He would have been very proud to hear that I finalised my studies.

(25)

25

6. References

[1] H.J. Conradi, J. Ormel, P. De Jonge, Presence of individual (residual) symptoms during depressive episodes and periods of remission: A 3-year prospective study, Psychol. Med. 41 (2011) 1165–1174. https://doi.org/10.1017/S0033291710001911.

[2] J. Campbell, W. Rashid, M. Cercignani, D. Langdon, Cognitive impairment among patients with multiple sclerosis: Associations with employment and quality of life, Postgrad. Med. J. 93 (2017) 143–147. https://doi.org/10.1136/postgradmedj-2016-134071.

[3] J.M. Sheffield, N.R. Karcher, D.M. Barch, Cognitive Deficits in Psychotic Disorders: A Lifespan Perspective, Neuropsychol. Rev. 28 (2018) 509–533. https://doi.org/10.1007/s11065-018-9388-2. [4] R.A. Lawson, A.J. Yarnall, G.W. Duncan, D.P. Breen, T.K. Khoo, C.H. Williams-Gray, R.A. Barker, D.

Collerton, J.P. Taylor, D.J. Burn, Cognitive decline and quality of life in incident Parkinson’s disease: The role of attention, Park. Relat. Disord. 27 (2016) 47–53.

https://doi.org/10.1016/j.parkreldis.2016.04.009.

[5] Y.S. Woo, J.D. Rosenblat, R. Kakar, W.-M. Bahk, R.S. McIntyre, Cognitive Deficits as a Mediator of Poor Occupational Function in Remitted Major Depressive Disorder Patients, Clin. Psychopharmacol. Neurosci. 14 (2016) 1–16. https://doi.org/http://dx.doi.org/10.9758/cpn.2016.14.1.1.

[6] B. Ravnkilde, P. Videbech, K. Clemmensen, A. Egander, N.A. Rasmussen, R. Rosenberg, Cognitive deficits in major depression, Scand. J. Psychol. 43 (2002) 239–251. https://doi.org/10.1111/1467-9450.00292.

[7] L.S. Jorna, J.M. Spikman, R.G. Schoemaker, B.L. Van Leeuwen, E. Iris, iMedPub Journals The Efficacy of Anti-Inflammatory Medication in Postoperative Cognitive Decline : A Meta-Analysis, (2020) 1–11. https://doi.org/10.36648/2171-6625.11.1.318.

[8] R.E. Nielsen, S. Levander, G. Kjaersdam Telléus, S.O.W. Jensen, T. Østergaard Christensen, S. Leucht, Second-generation antipsychotic effect on cognition in patients with schizophrenia-a meta-analysis of randomized clinical trials, Acta Psychiatr. Scand. 131 (2015) 185–196.

https://doi.org/10.1111/acps.12374.

[9] A. Jenkins, J.J. Tree, I.M. Thornton, A. Tales, Subjective Cognitive Impairment in 55-65-Year-Old Adults Is Associated with Negative Affective Symptoms, Neuroticism, and Poor Quality of Life, J. Alzheimer’s Dis. 67 (2019) 1367–1378. https://doi.org/10.3233/JAD-180810.

[10] T.B. Cumming, A. Brodtmann, D. Darby, J. Bernhardt, The importance of cognition to quality of life after stroke, J. Psychosom. Res. 77 (2014) 374–379. https://doi.org/10.1016/j.jpsychores.2014.08.009. [11] D.I. Velligan, R.K. Mahurin, P.L. Diamond, B.C. Hazleton, S.L. Eckert, A.L. Miller, The functional

significance of symptomatology and cognitive function in schizophrenia, Schizophr. Res. 25 (1997) 21– 31. https://doi.org/10.1016/S0920-9964(97)00010-8.

[12] M. Prikken, M.J. Konings, W.U. Lei, M.J.H. Begemann, I.E.C. Sommer, The efficacy of computerized cognitive drill and practice training for patients with a schizophrenia-spectrum disorder: A meta-analysis, Schizophr. Res. 204 (2019) 368–374. https://doi.org/10.1016/j.schres.2018.07.034.

[13] N. Çakici, N.J.M. Van Beveren, G. Judge-Hundal, M.M. Koola, I.E.C. Sommer, An update on the efficacy of anti-inflammatory agents for patients with schizophrenia: a meta-analysis, Psychol. Med. 49 (2019) 2307–2319. https://doi.org/10.1017/S0033291719001995.

[14] M. Dauwan, M.J.H. Begemann, M.I.E. Slot, E.H.M. Lee, P. Scheltens, I.E.C. Sommer, Physical exercise improves quality of life, depressive symptoms, and cognition across chronic brain disorders: a

transdiagnostic systematic review and meta-analysis of randomized controlled trials, J. Neurol. (2019). https://doi.org/10.1007/s00415-019-09493-9.

[15] J. De Boer, M. Prikken, W.U. Lei, M. Begemann, I. Sommer, The effect of raloxifene augmentation in men and women with a schizophrenia spectrum disorder: A systematic review and meta-analysis, Npj

(26)

26

Schizophr. 4 (2018). https://doi.org/10.1038/s41537-017-0043-3.

[16] J.O. Nuninga, T.F.I. Claessens, M. Somers, R. Mandl, W. Nieuwdorp, M.P. Boks, S. Bakker, M.J.H. Begemann, S. Heringa, I.E.C. Sommer, Immediate and long-term effects of bilateral electroconvulsive therapy on cognitive functioning in patients with a depressive disorder, J. Affect. Disord. 238 (2018) 659–665. https://doi.org/10.1016/j.jad.2018.06.040.

[17] E. Tenison, A. Smink, S. Redwood, S. Darweesh, H. Cottle, A. Van Halteren, P. Van Den Haak, R. Hamlin, J. Ypinga, B.R. Bloem, Y. Ben-Shlomo, M. Munneke, E. Henderson, M.C. Portillo, Proactive and

Integrated Management and Empowerment in Parkinson’s Disease: Designing a New Model of Care, Parkinsons. Dis. 2020 (2020). https://doi.org/10.1155/2020/8673087.

[18] A.F. Jorm, Mental health literacy; empowering the community to take action for better mental health, Am. Psychol. 67 (2012) 231–243. https://doi.org/10.1037/a0025957.

[19] B. Weiss, Health Literacy, ELDER CARE A Resour. Interprofessional Provid. (2016).

[20] T.Y. Koonce, N.B. Giuse, S. V. Kusnoor, S. Hurley, F. Ye, A personalized approach to deliver health care information to diabetic patients in community care clinics, J. Med. Libr. Assoc. 103 (2015) 123–130. https://doi.org/10.3163/1536-5050.103.3.004.

[21] N.B. Giuse, T.Y. Koonce, A.B. Storrow, S. V Kusnoor, F. Ye, Using health literacy and learning style preferences to optimize the delivery of health information, in: J. Health Commun., 2012: pp. 122–140. https://doi.org/10.1080/10810730.2012.712610.

[22] R.E. Mayer, Multimedia learning, second edition, 2009. https://doi.org/10.1017/CBO9780511811678. [23] L. Fraenkel, W. Benjamin Nowell, C.E. Stake, S. Venkatachalam, R. Eyler, G. Michel, E. Peters, Impact of

Information Presentation Format on Preference for Total Knee Replacement Surgery, Arthritis Care Res. 71 (2019) 379–384. https://doi.org/10.1002/acr.23605.

[24] K. Alamantariotou, D. Zisi, Consumer health informatics and interactive visual learning tools for health, Int. J. Electron. Healthc. 5 (2010) 414–424. https://doi.org/10.1504/IJEH.2010.036211.

[25] M.H. Nguyen, E.M.A. Smets, N. Bol, E.F. Loos, J.C.M. Van Weert, How Tailoring the Mode of

Information Presentation Influences Younger and Older Adults’ Satisfaction with Health Websites, J. Health Commun. 23 (2018) 170–180. https://doi.org/10.1080/10810730.2017.1421729.

[26] M.H. Nguyen, J.C.M. van Weert, N. Bol, E.F. Loos, K.M.A.J. Tytgat, A.W.H. van de Ven, E.M.A. Smets, Tailoring the Mode of Information Presentation: Effects on Younger and Older Adults’ Attention and Recall of Online Information, Hum. Commun. Res. 43 (2017) 102–126.

https://doi.org/10.1111/hcre.12097.

[27] S. Bolle, G. Romijn, E.M.A. Smets, E.F. Loos, M. Kunneman, J.C.M. Van Weert, Older cancer patients’user experiences with web-based health information tools: A think-aloud study, J. Med. Internet Res. 18 (2016). https://doi.org/10.2196/jmir.5618.

[28] B. Geboers, E. Uiters, S.A. Reijneveld, C.J.M. Jansen, J. Almansa, A.C.J. Nooyens, W.M.M. Verschuren, A.F. De Winter, H.S.J. Picavet, Health literacy among older adults is associated with their 10-years’ cognitive functioning and decline - The Doetinchem Cohort Study, BMC Geriatr. 18 (2018). https://doi.org/10.1186/s12877-018-0766-7.

[29] M.H. Nguyen, N. Bol, J.C.M. van Weert, E.F. Loos, K.M.A.J. Tytgat, D. Geijsen, E. Drenth, M. Janse, E.M.A. Smets, Optimising eHealth tools for older patients: Collaborative redesign of a hospital website, Eur. J. Cancer Care (Engl). 28 (2019) e12882. https://doi.org/10.1111/ecc.12882.

[30] M. Dauwan, M.J.H. Begemann, S.M. Heringa, I.E. Sommer, Exercise improves clinical symptoms, quality of life, global functioning, and depression in schizophrenia: A systematic review and meta-analysis, Schizophr. Bull. 42 (2016) 588–599. https://doi.org/10.1093/schbul/sbv164.

[31] M.J. Begemann, B.A. Brand, B. Ćurčić-Blake, A. Aleman, I.E. Sommer, Psychological Medicine Efficacy of non-invasive brain stimulation on cognitive functioning in brain disorders: a meta-analysis, (n.d.).

(27)

27

https://doi.org/10.1017/S0033291720003670.

[32] S.L. Lim, J.C. Yang, J. Ehrisman, L.J. Havrilesky, S.D. Reed, Are Videos or Text Better for Describing Attributes in Stated-Preference Surveys?, Patient. 13 (2020) 401–408. https://doi.org/10.1007/s40271-020-00416-9.

[33] C.S. Meppelink, J.C.M. Van Weert, C.J. Haven, E.G. Smit, The effectiveness of health animations in audiences with different health literacy levels: An experimental study, J. Med. Internet Res. 17 (2015) e11. https://doi.org/10.2196/jmir.3979.

[34] C.S. Meppelink, N. Bol, Exploring the role of health literacy on attention to and recall of text-illustrated health information: An eye-tracking study, Comput. Human Behav. 48 (2015) 87–93.

https://doi.org/10.1016/j.chb.2015.01.027.

[35] M.G. Violante, E. Vezzetti, Virtual interactive E-learning application: An evaluation of the student satisfaction, Comput. Appl. Eng. Educ. 23 (2015) 72–91. https://doi.org/10.1002/cae.21580. [36] L. Calli, C. Balcikanli, F. Calli, H.I. Cebeci, O.F. Seymen, Identifying factors that contribute to the

satisfaction of students in e-learning, Turkish Online J. Distance Educ. 14 (2013) 85–101. https://doi.org/10.17718/tojde.36824.

[37] Y. Peng, X. Huang, Y. Zhao, An Overview of Cross-Media Retrieval: Concepts, Methodologies, Benchmarks, and Challenges, IEEE Trans. Circuits Syst. Video Technol. 28 (2018) 2372–2385. https://doi.org/10.1109/TCSVT.2017.2705068.

[38] Z. Xie, L. Li, X. Zhong, Y. He, L. Zhong, Enhancing multimodal deep representation learning by fixed model reuse, Multimed. Tools Appl. 78 (2019) 30769–30791. https://doi.org/10.1007/s11042-018-6556-6.

[39] C. Beste, T. Ziemssen, Why Cognitive–Cognitive Dual-Task Testing Assessment Should Be Implemented in Studies on Multiple Sclerosis and in Regular Clinical Practice, Front. Neurol. 11 (2020) 905.

https://doi.org/10.3389/fneur.2020.00905.

[40] A.D. Ritzhaupt, R. Pastore, R. Davis, Effects of captions and time-compressed video on learner performance and satisfaction, Comput. Human Behav. 45 (2015) 222–227.

https://doi.org/10.1016/j.chb.2014.12.020.

[41] K.A. Austin, Multimedia learning: Cognitive individual differences and display design techniques predict transfer learning with multimedia learning modules, Comput. Educ. 53 (2009) 1339–1354.

https://doi.org/10.1016/j.compedu.2009.06.017.

[42] C. Dong, P.S. Goh, Twelve tips for the effective use of videos in medical education, Med. Teach. 37 (2015) 140–145. https://doi.org/10.3109/0142159X.2014.943709.

[43] L.M. Bogza, C. Patry-Lebeau, E. Farmanova, H.O. Witteman, J. Elliott, P. Stolee, C. Hudon, A.M.C. Giguere, User-centered design and evaluation of a web-based decision aid for older adults living with mild cognitive impairment and their health care providers: Mixed methods study, J. Med. Internet Res. 22 (2020). https://doi.org/10.2196/17406.

(28)

28

7. Appendix

(29)

(30)

(31)

(32)

(33)

(34)

(35)

(36)

(37)

(38)

(39)

(40)

(41)

(42)

(43)

Care for Cognition: Science communication using different modalities of communication