Development and Pilot Testing of Multimedia Patient Education Tools for Patients with Knee Osteoarthritis, Osteoporosis, and Rheumatoid Arthritis

Abstract

Background:

We developed and tested multimedia patient education tools (video tools) for patients with knee osteoarthritis (OA), osteoporosis (OP) and rheumatoid arthritis (RA).

Methods:

We followed an “edutainment” model, incorporating educational patient story lines. The goals were designed to make the programs both didactic and entertaining, with navigation and graphical user interfaces as simple as possible. We created both English and Spanish-language versions. Once the video tool was finalized, 60 patients, 20 per disease, were shown the tool and were interviewed. Disease knowledge was our primary outcome and decision conflict, disease management, and acceptability were secondary outcomes.

Results:

We observed statistically significant differences in pre-post knowledge questionnaire scores (before and after viewing the video tool) (OA, p=0.03; OP, p=0.001; RA, p<0.0001). Most participants felt they: 1) gained “clarity” on disease duration, symptoms, and time medication takes to start acting, 2) were “encouraged to see their doctor regularly”, and 3) were more aware about taking their medications. In terms of acceptability, most patients in all disease groups found the length and amount of information presented in the video tools to be “just right”, and the presentation as ‘balanced”. In terms of comprehension, all participants provided a favorable evaluation of the video tool; all found the video easy to use, the vocabulary easy to understand, and the materials be well organized.

Conclusions:

Multimedia tools that incorporate videos may help patients better understand and manage their disease. Patient involvement in the development process is essential to ensure relevant content and usability.

Knee osteoarthritis (OA), osteoporosis (OP) and rheumatoid arthritis (RA) are musculoskeletal diseases affecting the general adult population, with significant social and economic impact (1). Studies suggest that patients’ health outcomes, knowledge about their disease, and decision making can be improved through educational programs designed to meet their informational needs (2–4).

Written materials to inform patients, consumers, healthcare providers and policymakers of updated health evidence exist for these diseases (5–10). Although electronic versions of education materials for OA, OP, and RA are readily available for use by patients, they may not reach those in most need, namely, socioeconomically and ethnically disadvantaged populations due to low or limited health literacy, or limited access. Even when accessible, the provided information may not be adequate for patients to comprehend content when reading levels are too high.

Poor health literacy has been found to negatively impact patients’ health outcomes (11–13), and affect key decision making (14). However, little importance has been given to the role of health literacy in the development and evaluation of patient educational tools. The majority of health educational tools have not been developed making design considerations for low literate populations (14).

Entertainment education is a promising strategy for providing learning content to patients with low health literacy. Factual information about options is presented to patients within the context of engaging stories, or “telenovelas” (soap operas). This type of education significantly improves decisional conflict and patient self-efficacy and has also shown to reduce pre-procedural anxiety and improving coping skills (15–18). Thus, we develop multimedia patient education tools for patients with knee OA, RA, and OP and low health literacy in English and Spanish language. We also evaluated the acceptability of the tools in the targeted populations and their effectiveness in a limited way to provide information for planning a randomized controlled trial.

Methods

Multimedia Patient Education Tools (video tools) –

Development of our video tools followed an “edutainment” model (17) to provide a guided framework for developing patient information particularly appropriate for patients with lower health literacy. The program goals were designed to make the program both didactic and entertaining, with a simple user interface and navigation. The included key messages were derived from focus groups and cognitive interviews conducted in patients with OA, OP, and RA and rheumatologists (Supplementary Table 1) (19).

The video tools were structured as a series of soap opera segments depicting a main character with OA, OP, or RA, integrated with learning modules to provide patients with factual information about their condition and treatment options, and were develop both in English and Spanish language. Total length of time for each segment ranged between 3 and 7 minutes. Each soap opera episode was linked to continuous learning modules of approximately 1 to 2 minutes, which incorporated slides or videos and a narrator, emphasizing selected key learning points. The soap opera episodes provided a real-life context for the information and tools that were included in the linked learning modules. Learning objectives were drafted for each video tools topic. The general sequence of content was: (i) overview of the health care condition, (ii) description of the treatment options, (iii) patient story (testimonial I), (iii) description of the harms/risks and benefits of various options (iv) patient story (testimonial II), (v) review of key facts, and vi) patient story (testimonial III).

The video tools used a variety of multimedia design features to enhance their utility with low health literacy patients. Both the content and the instructions are narrated, and visual cues were provided to reach patients who are poor readers or who cannot read. We worked with local producers of healthcare information products with experience developing materials for low literacy populations. The materials were reviewed by two content experts (KS, MSA), three investigators (MLO, RV, MJW), and a patient advocate (AL). Refinements were made to the program after pre-testing to ensure accuracy. We then asked 7 lay people (3 bilingual) to evaluate paper mockups and prototype versions of the video tools during production, and made additional changes to the content. Lastly, 10 cognitive interviews of patients per disease were conducted to identify potential language and imagery issues in the video tools.

Pilot Testing –

We recruited 20 participants per disease (i.e., knee OA, RA, and OP) with at least 5 Spanish-speaking participants per disease group. Participants were recruited from community clinics of the County’s Health System and The University of Texas, MD Anderson Cancer Center, in Houston, TX. Inclusion criteria for each disease group was: OA-(i) age ≥50, (ii) prior diagnosis of knee OA (unilateral or bilateral) by a physician; RA-(i) age 18 and over, (ii) compliance with American College of Rheumatology criteria for the diagnosis of RA (20) (iii) disease duration ≤10 years; OP-(i) female gender, (ii) age ≤50, (iii) at least 3 years post-menopausal, and prior diagnosis of osteoporosis or osteopenia. In addition to these disease specific inclusion criteria, we also had some common inclusion criteria as follows: (i) adequate cognitive status as determined by the research assistant (ii) not living in a nursing facility, (iii) ability to communicate in English or Spanish language without a translator, (iv) have access to a telephone. The disease duration criterion for RA was chosen because the content in the video tool is most appropriate for patients with early disease. Patients with longstanding disease are more likely to have acquired specific knowledge about their disease, and have a number of other issues, such as surgical needs and comorbidities, not covered in the current guidelines. Patients were consented, then they were asked to complete a questionnaire immediately before (pre-assessment) watching the video intervention related to their disease. Immediately after (post-assessment) watching the intervention patients completed another questionnaire. A research staff was present to assist patients in need of help.

Outcome measures –

Primary outcomes included:

1. Knowledge about disease and therapeutic options. OA, OP and RA specific knowledge instruments were used (21–23). These instruments were modified and adapted to reflect the key learner content covered in our product. The knowledge questionnaires for OA, OP and RA consisted of 16, 20 and 12 questions, respectively. Total knowledge scores were obtained by summing the scores of individual items. Scores ranged from 0–16, 0–20 and 0–12 for OA, OP and RA, respectively. Higher scores indicate greater disease knowledge. Disease knowledge was measured at both pre and post intervention. These instruments are widely used in rheumatology and have been validated in multiple languages (21–26).

Secondary outcome measures of interest were:

1. Decisional Conflict Scale (DCS). It measures the degree to which someone is conflicted or unclear about a particular health care choice. We used the low literacy version of the instrument for this study and modified it to include only 5 questions instead of 10 (items related to support and uncertainty were covered domains outside our major evaluation goals). This version of the scale has been tested in English and Spanish with alpha coefficient ranging from 0.72 to 0.86 and has proven to be responsive to change (27). The scale measures personal perceptions of a) uncertainty in choosing between options, b) modifiable factors contributing to uncertainty such as feeling uninformed, and c) effective decision making such as feeling the choice is values-based and likely to be implemented. It consists of 5 subscales: (i) informed, (ii) values clarity (iii) support, (iv) uncertainty, and (v) effective decision. For the purpose of our study we only used the “informed” and “values clarity” subscales. Scores range from 0 (feels extremely informed/clear about personal values) to 100 (feels extremely uninformed/unclear about personal values). DCS was measured after. The suggested cut points for interpretation for the total score are: ≤25 indicating decisiveness (associated with ‘implementing decisions’) and ≥37.5 indicating indecision (27).

2. Disease management-The Effective Consumer Scale (ECS) is a 17-item instrument that measures individuals’ main skills and behaviors to effectively manage their healthcare (28). The scores range from 0–100 wherein higher scores indicate greater healthcare management. ECS was measured at post intervention. The scale’s validity and responsiveness have been tested in several languages with evidence of internal consistency, test-retest reliability, construct validity and responsiveness to change (29–31).

3. Acceptability Measures-The Ottawa acceptability scale is a standard assessment tool used to obtain users’ rating of various features of the program including clarity, balance, length, and ease of use. This scale has been widely used in different conditions with robust validity, reliability and Cronbach’s alpha (32). We added additional items for qualitative assessment of the content, ease of use and format of the newly developed video tools. We also asked for participants’ suggestions for improvement of the video tools. In addition to utilizing the Ottawa acceptability instrument, we also developed a project specific measure of acceptability based on previous research (33, 34). Our instrument consisted of 18 questions (13 items) asking patients to judge the video tool based on its content, ease of use, transportability and format. These questionnaires were administered post intervention.

We also collected participants’ demographics information (age, ethnicity, education, employment status, and marital status) and health behaviors (e.g., calcium and vitamin D intake, physical activity, waiting time to contact health provider if not feeling well, regular checkups, flu shot, etc.) information. Health literacy was assessed using a single item, “How confident are you filling out medical forms by yourself?”, which was developed by Chew at al.(35, 36) to identify “inadequate health literacy level” in English and Spanish speaking participants.

Statistical Analysis –

Frequencies and percentages were used to summarize patient demographics and other categorical measures while means and standard deviations were used to summarize continuous information such as patient’s age, knowledge score, DCS score and ECS score. Demographics and baseline characteristics were compared among disease groups using Chi-square or Fisher’s Exact test of association, as well as Analysis of Variance (ANOVA) or Kruskal-Wallis test, where appropriate. Similarly, outcome measures such as DCS and ECS were compared among the 3 disease groups using ANOVA. Paired t-test was used to compare the mean knowledge scores pre and post intervention (for each disease group and, by HL group within each disease group), while independent sample t-test was used to compare knowledge improvement (post intervention – pre intervention) between the two HL groups (per disease group). Significance level was set at p<0.05. A meaningfully important difference (MID) in knowledge was defined using the distribution-based approach (37). Thus, we considered a post intervention score to be clinically important if the knowledge improvement was higher than 0.5*SD_{pre-intervention score}.

Results

Participant demographic and baseline characteristics are provided in Table 1. Mean age in the entire cohort was 61 ± 10 years; 85% were women, 54% were Hispanic, 73% had less than a high school diploma and, 70% had adequate health literacy. No significant differences were observed among the three disease groups (p > 0.05 for all characteristics) (Table 1).

Table 1.

Demographics and baseline characteristics of OA, OP and RA patients

Characteristics	OA N = 20 n (%)	OP N = 20 n (%)	RA N = 20 n (%)
Age, years Mean (SD)	61.4 (7.9)	64.1 (10.2)	56.8 (11.0)
Female	13 (65.0)	20 (100.0)	18 (90.0)
Marital status
Single/Never married	4 (20.0)	2 (10.0)	4 (20.0)
Married/living with significant other	8 (40.0)	11 (55.0)	8 (40.0)
Divorced/Separated/Widowed	8 (40.0)	7 (35.0)	8 (40.0)
Race
White	1 (5.0)	7 (35.0)	6 (31.6)
African American	6 (30.0)	4 (20.0)	1 (5.3)
Hispanic	12 (60.0)	8 (40.0)	12 (63.2)
Asian	1 (5.0)	1 (5.0)	0 (0.0)
Educational attainment
≤ High school diploma/equivalent	16 (80.0)	11 (55.0)	17 (85.0)
> High school diploma/trade school/bachelor’s degree	4 (20.0)	9 (45.0)	3 (15.0)
Health literacy level - Inadequate health literacy	7 (35)	6 (30)	5 (25)
Language
English	11 (55)	13 (65)	9 (45)
Spanish	9 (45)	7 (35)	11 (55)

Knowledge.

Post intervention, all disease groups showed statistically significant improvements in knowledge scores post intervention (Figure 1). When examining knowledge improvement by health literacy groups (Table 2), we found that patients with inadequate health literacy showed improvement in knowledge across the 3 diseases, with borderline significance for the OP group. For patients with adequate literacy, significant improvements were observed for RA, OP, and no increase was observed for OA. Clinically important improvement in knowledge was seen in both health literacy groups among RA patients, in the inadequate health literacy group among OA patients, and in the adequate health literacy among OP patients.

Figure 1.

Difference in outcome measures of knowledge, between pre and post intervention. *P values generated from Student’s T test OA=Osteoarthritis; OP, Osteoporosis; RA, Rheumatoid Arthritis

Table 2.

Knowledge improvement by health literacy groups for each disease group

Disease group	HL group	Pre intervention knowledge score	Post intervention knowledge score Mean (SD)	Knowledge improvement (Post - pre)	P value*	P value**
	Adequate HL	8.6 (1.6)	9.0 (2.2)	0.4 (1.4)	0.40	0.02
OA	Inadequate HL	7.4 (1.6)	9.6 (1.6)	2.2 (2.0)	0.03
	P value***	0.12	0.52
OP	Adequate HL	10.7 (4.8)	13.7 (5.0)	3.0 (3.7)	0.01	0.07
	Inadequate HL	4.8 (5.2)	7.3 (4.8)	2.5 (2.4)	0.05
	P value***	0.02	0.02
RA	Adequate HL	5.3 (2.2)	8.1 (1.8)	2.8 (2.3)	0.0003	0.22
	Inadequate HL	4.4 (2.2)	5.8 (2.4)	1.4 (0.9)	0.02
	P value***	0.42	0.04

OA, Osteoarthritis; OP, Osteoporosis; RA, Rheumatoid Arthritis; HL, Health literacy

P values generated using Students paired t test and independent sample t test

^*Knowledge improvement within each HL group

^**Comparison of knowledge improvement between the two HL groups

^***Difference at each time point between the two HL groups

Decisional Conflict and Disease Management.

Post intervention, low mean scores were observed in the DCS subscales indicating patients feeling informed (OA=18±32, OP=21±32, RA=18±25) and being clear about personal values (OA=15±30, OP=19±34, RA=20±31). Similarly, high mean scores indicating better disease management were observed in the OA, OP, and RA groups after watching the video tool (84±12, 88±11, 89±12). No significant differences in DCS subscale scores (informed and values clarity) and ECS scores were observed (p>0.20 for all measures), between the three disease groups. The percent of participants with total DCS score ≤25 (associated with implementing decisions) was high 70% to 80% in all groups. There was no difference in DCS between health literacy groups in all disease groups except OP (total score p=0.002; informed subscale p=0.002; and clarity subscale p=0.02).

Acceptability.

Most participants responded favorably when asked to rate each video tool section (impact of disease, medication options, evidence about medications, benefit of medications, reducing side-effects, self-care options and stories about others). The percent of patients reporting good or excellent ranged between 80 to 95 in the OA group, 90 to 100 in the OP group, and 90 to 95 in the RA group (Table 3). The majority in all three groups found the amount of information presented “just right” (OA, 70%; OP, 70%; RA, 85%). However, only 50% participants in the OP group found the presentation to be balanced compared to 75% in both, OA and RA groups. No significant differences in response options were observed, between the three disease groups for all acceptability questions (p > 0.10; all questions). Similar results were observed with respect to our project specific evaluation measure, wherein majority of the patients in all disease groups rated favorably the content, ease of use, and format (Supplementary Table 2). Results for the transportability domain (e.g., did you feel you could identify with the characters of the video, were you eager to find out how the story ended, etc.) are shown in Figure 2. Most participants among all groups would recommend the video to a friend, liked the explanation of the medical facts, were able to easily take the perspective and identify with the characters in the story. In spite of this, only half or less were emotionally affected (touched) by what they watched (OA=40%, OP=45% and RA=50%). No significant differences were observed between the three disease groups for any of the evaluation items (p>0.10 for all questions). Also, no association was observed between the acceptability questions and health literacy groups in all disease groups. Out of the 60 patients included, the preferred language for 27 (45%) was Spanish. No differences were observed in the rates of acceptability between English and Spanish respondents, but additional comments regarding what they liked about the video tools and suggestions for improvement were provided by 22 patients (11 Spanish-speakers). Patients who watched the video tool in Spanish liked all features and most did not have any suggestions for improvement except for increasing length of time of the stories (soap opera episodes) or the video tool in general. Patients who watched the video tool in English felt that it was informative and suggestions were around keeping the information updated when new treatments become available and avoiding oversimplifying medical facts.

Table 3.

Percent of responses on the Ottawa acceptability scale by each disease group.

Question	Response	OA n (%)	OP n (%)	RA n (%)
Presentation	Good/Excellent
Impact of OA/OP/RA		19 (95)	18 (90)	18 (90)
Medication Options		16 (80)	19 (95)	18 (90)
Evidence about Medications		17 (85)	18 (90)	18 (90)
Benefits of Medications		17 (85)	19 (95)	18(90)
Understanding the side effects		16 (80)	18(90)	18(90)
Self-care options		19 (95)	20 (100)	19 (95)
Presentation: Stories about others		19 (95)	19 (95)	18 (90)
	Too long	0 (0.0)	1 (5)	2 (10)
Length of the material	Too short	0 (0.0)	0 (0.0)	1 (5)
	Just right	20 (100)	19 (95)	17 (85)
	Too much	4 (20)	3 (15)	2 (10)
Amount of information	Too little	2 (10)	3 (15)	1 (5)
	Just right	14 (70)	14 (70)	17 (85)
	Slanted towards taking self-care or lifestyle options	2 (10)	5 (25)	4 (20)
Found the material presented	Slanted towards taking medical therapies	3 (15)	3 (16)	1 (5)
	Balanced	15 (75)	12 (50)	15 (75)
Have found information useful when making decision about therapy for OA/OP/RA	Yes	17 (85)	18(90)	16 (80)
Way treatments were explained	Easy to understand	18(90)	19 (95)	20 (100)
Way side effects of the treatments were explained	Easy to understand	18(90)	19 (95)	17 (85)
Enough information to help a person decide on therapy for OA/OP/RA	Yes	19 (95)	18(90)	20 (100)

Figure 2.

Transportability items from the evaluation questionnaire post intervention. This graph indicates the percent of patients responding “yes” to each transportability item in patients with knee osteoarthritis (bars), osteoporosis (diamond) and rheumatoid arthritis (line). Except for one item, 75% to 100% of the participants liked and could identify with the story and characters of the multimedia patient education tool.

Discussion

Knowledge about a condition plays an important role in successful disease management. The selected method to deliver health education to patients can impact learning and ultimately disease knowledge (38). With the new advances in technology, the use of multimedia materials (e.g., websites or webpages, DVDs, computer files, etc.) for patient education has resulted in improvement of health outcomes (39). However, multimedia health education materials vary in the type of features included (40). Moreover, evidence about the development, evaluation and quality of video tools in rheumatic care is severely limited. Commonly, video tools are developed without following a systematic process or conducting an evaluation of the quality of information and efficacy of the educational content, accessibility, respect for patient’s linguistic needs, health literacy level, and learning preferences, and ease of use (40). The objectives of our study were: (i) to systematically develop low health literacy video tools before their implementation in patients with OA, OP and RA and (ii) evaluate the acceptability and effectiveness of the newly developed tools in these patients. The process outlined in this study can serve as an example for the development and testing of future video tool for chronic conditions. The findings provided our team with information regarding the (i) feasibility of conducting a randomized controlled trial to test the efficacy of the video tools compared with other formats; (ii) acceptability of the video tools content and method of delivery; and (iii) appropriateness of outcomes measures to be evaluated.

Upon pilot testing the video tools, we found that post intervention, knowledge scores significantly increased in all three disease groups. In addition, the use of multimedia features with story lines was associated with learning health education particularly for individuals with low health literacy. Significant knowledge improvement was observed within the OA inadequate health literacy group itself, and when comparing knowledge gain between the two OA groups (inadequate vs adequate health literacy). Individual health literacy groups for OP and RA also showed significant post intervention knowledge improvement, except for the OP inadequate health literacy group, which reached near significance. Although not entirely certain, we can hypothesize that this observed result could be due to the small percentage of inadequate health literacy participants in OP group (30%). Overall, low literacy groups appear to be less knowledgeable before watching the video tool. After watching the video tools, patients do gain knowledge, but do not reach the same level as high literacy group. No differences were observed for decisional conflict (except in OP) or the acceptability scale between health literacy groups.

Few studies have reported results comparable to our observations (41, 42). Kandula et al. developed and studied the effect of a Multimedia Diabetes Education Program (MDEP), targeted for English speaking patients with low health literacy, on knowledge and examined the association between literacy and knowledge improvement (41). They divided their participants into three literacy groups; adequate (79%), marginal (13%) and inadequate (8%) using the Short Test for Functional Health Literacy in Adults (S-TOHFLA). They found a significant increase in participants’ knowledge scores after viewing the MDEP (p<0.001), across all literacy levels. However, unlike our results, they observed that participants with inadequate literacy learned significantly less after the MDEP compared to those with adequate literacy (41). Another study by the same authors, evaluated their Spanish language MDEP targeted to Spanish-speaking patients with low literacy and assessed knowledge gained after viewing the MDEP (42). Again, participants were grouped as having adequate (53%), marginal (14%), and inadequate literacy (33%). They observed significant increases in mean knowledge scores for all participants, regardless of literacy level. Moreover, they also found that knowledge gains among the Spanish speaking study participants were comparable to those seen in native English-speaking patients from their previous study (42). Due to the inclusion of a limited number of Spanish speaking individuals in our study, we did not compare differences in knowledge gain by language (English vs Spanish) and hence were unable to ascertain whether our video tool was equally effective in both language groups.

In the aforementioned studies, the S-TOHFLA was the standard measure for health literacy, whereas, we used a single question as a measure of health literacy. Considering that this question simplifies screening of health literacy and has been validated showing comparable performance to the observed with the Rapid Estimation of Adult Literacy in Medicine (REALM) and the S-TOFHLA (35), we are confident that the health literacy classification of participants in our study is approximate to those in the other studies.

We also conducted user evaluation and assessed acceptability for qualitative analysis of our video tools. We used measures routinely reported in research focused on developing and evaluating multimedia patient education tools. Most participants provided a favorable evaluation of the video tool; all found the video easy to use, the vocabulary easy to understand and the material to be well organized. In terms of acceptability, majority of patients in all disease groups (OA, OP and RA) found the length and amount of information presented in the video tools to be “just right”, and the presentation as ‘balanced”.

To our knowledge no other studies have developed and tested low health literacy educational multimedia tools specifically for OA, OP and RA patients. Cranney and colleagues developed and pilot tested an evidence based decision aid with a 45 minute audiotape for post-menopausal women with osteoporosis who are considering options to prevent factures. They used the Ottawa Decision Support Framework as a guideline (43). Compared to our results; only 61% of women in their study found the length of the presentation as ‘just right”, 28% found the amount of information “just right” and, 61% found the presentation as “balanced”(43). Another study by Volk et al. developed and field tested a brief, video-based patient decision aid about lung cancer screening (44). Similar to our results, acceptability of the decision aid was high. Results indicated highly favorable ratings for the amount of information, length, and clarity of the decision aid (44). Majority of patients (> 94%) said they would recommend the video to others, felt it held their interest, and wanted to view similar videos about health care decisions (44).

Our study has important limitations. Our educational tool by definition is not a decision aid according to the certification criteria for decision aids, the International Patient Decision Aids Standards (IPDAS) Collaboration since it provides information on all currently available treatments for OA, OP, and RA and no probabilities (absolute risks) are discussed (45). Nonetheless, our video tool includes sections about what to discuss with one’s doctor, when a modification in therapeutic management is necessary, and what factors should be taken into account when deciding a treatment strategy (including preferences in route and dose frequency, side effects, and costs). In addition, because of the nature of this pilot study our sample size was small, not allowing us to analyze differences in participant characteristics other than health literacy and only before and after results are provided. There is also an inherent probability of social desirability bias, given by the nature of this pilot study and the used of structured direct questions to evaluate acceptability of the intervention. However, we also included structure projective questions (i.e., asking respondents to answer questions from the perspective of another person) that is a common approach used to reduce the distortion that can result from direct questions (46). Because the outcome measures were administered before and immediately after the intervention, the results observed could be only a reflection of the participant’s recall of the information presented in the tools. Therefore, a longer follow-up will be needed when evaluating the efficacy of these tools in a randomized controlled trial. Finally, possible ceiling effects on health literacy could have clouded the effects of the intervention in some outcomes for those participants with less ability to improve. Nonetheless, our results consistently indicate a trend towards outcomes improvement post intervention, suggesting that the efficacy of our video tools could be evaluated in a larger setting. A randomized controlled trial is needed to evaluate the efficacy of this type of educational intervention in improving outcomes in patients with these musculoskeletal conditions.

In conclusion, multimedia tools that incorporate video modeling may help patients better understand and manage their disease. Patient involvement in the development process is essential to ensure relevant content and usability. We developed our video tools following a systematic and rigorous process which can serve as an evidence-informed framework to improve content, development, implementation, and evaluation of future video tools. Additionally, our testing results provide information to further evaluate the use of our video tools in larger controlled studies with longer follow-up and including limited health literacy populations.

Significance and Innovations.

• We developed three different video tools following an “edutainment” model, incorporating educational patient story lines.

• Entertainment education is a promising strategy for providing education to patients with low health literacy.

• The process outlined in this study can serve as an example for the development and testing of future video tools for other rheumatic conditions.