Arthritis Care & Research Vol. 66, No. 10, October 2014, pp 1472–1481 DOI 10.1002/acr.22319 © 2014, American College of Rheumatology

ORIGINAL ARTICLE

Proof-of-Concept Study of a Web-Based Methotrexate Decision Aid for Patients With Rheumatoid Arthritis LINDA C. LI,1 PAUL M. ADAM,2 CATHERINE L. BACKMAN,1 SYDNEY LINEKER,3 C. ALLYSON JONES,4 DIANE LACAILLE,1 ANNE F. TOWNSEND,1 ELAINE YACYSHYN,4 CHARLENE YOUSEFI,5 PETER TUGWELL,6 JENNY LEESE,5 AND DAWN STACEY7 Objective. To assess the extent to which an online patient decision aid reduced decisional conflict and improved self-management knowledge/skills in patients who were considering methotrexate for rheumatoid arthritis (RA). Methods. We used a mixed-methods pre-post study design. Eligible participants had a diagnosis of RA, had been prescribed methotrexate but were unsure about starting it, and had access to the internet. Outcome included the Decisional Conflict Scale, the Methotrexate in RA Knowledge Test, and the Effective Consumer Scale. Paired t-tests were used to assess changes before and after the intervention. Randomly selected participants were interviewed at the end of the study about their experiences with the decision aid. Results. Of 30 participants, 23 were women. Mean ⴞ SD age was 54.9 ⴞ 14.9 years and the median disease duration was 1 year (interquartile range 0.3–5.0 years). Mean ⴞ SD decisional conflict changed from 49.50 ⴞ 23.17 preintervention to 21.83 ⴞ 24.12 postintervention (change ⴚ27.67 [95% confidence interval ⴚ39.89, ⴚ15.44]; P < 0.001). Knowledge of methotrexate improved (mean ⴞ SD 30.62 ⴞ 9.26 preintervention and 41.67 ⴞ 6.81 postintervention; P < 0.001), but there was no change in effective consumer attributes (mean ⴞ SD 68.24 ⴞ 12.46 preintervention and 72.94 ⴞ 12.74 postintervention; P ⴝ 0.15). Three themes emerged from interviews of 11 participants: seeking confirmation of one’s own knowledge of methotrexate, amplifying reluctance when they encountered information contradicting their own experiences, and clarifying thoughts about the next step during the process. Conclusion. Patients’ decisional conflict and knowledge improved after using the patient decision aid. Interview findings further highlighted the power of patients’ prior knowledge and experiences with RA on how they approach the information presented in a decision aid.

INTRODUCTION For people with rheumatoid arthritis (RA), there is a window of opportunity early in the disease to successfully Supported by a Canadian Institutes of Health Research operating grant (funding reference KAL-94482) and a Canadian Rheumatology Association Canadian Initiative for Outcomes in Rheumatology Care grant. Dr. Li’s work was supported by the Michael Smith Foundation for Health Research, the Canadian Institutes of Health Research New Investigator Award, The Arthritis Society research chair award, and the Canada Research Chair program. Dr. Jones’s work was supported by a Canadian Institutes of Health Research New Investigator Award. Dr. Lacaille’s work was supported by The Arthritis Society research chair award. Dr. Tugwell’s work was supported by the Canada Research Chair program. 1 Linda C. Li, BSc (PT), MSc, PhD, Catherine L. Backman, PhD, Diane Lacaille, MDCM, FRCPC, MHSc, Anne F. Townsend, BA, MA, PhD: University of British Columbia and Arthritis Research Centre of Canada, Vancouver, British Columbia, Canada; 2Paul M. Adam, MSW: Mary Pack

1472

control inflammation and prevent joint damage. According to the 2012 American College of Rheumatology recommendations on treatment for RA, disease-modifying antirheuArthritis Program and Vancouver Coastal Health, Vancouver, British Columbia, Canada; 3Sydney Lineker, PhD: The Arthritis Society, Toronto, Ontario, Canada; 4C. Allyson Jones, PT, PhD, Elaine Yacyshyn, MD, FRCPC: University of Alberta, Edmonton, Alberta, Canada; 5Charlene Yousefi, MPP, Jenny Leese, MA: Arthritis Research Centre of Canada, Vancouver, British Columbia, Canada; 6Peter Tugwell, MD, PhD: University of Ottawa, Ottawa, Ontario, Canada; 7 Dawn Stacey, RN, PhD: University of Ottawa and Ottawa Hospital Research Institute, Ottawa, Ontario, Canada. Dr. Stacey has received consulting fees, speaking fees, and/or honoraria (less than $10,000) from Blue Shield of California. Address correspondence to Linda C. Li, BSc (PT), MSc, PhD, Arthritis Research Centre of Canada, 5591 No. 3 Road, Richmond, British Columbia V6X 2C7, Canada. E-mail: [email protected]. Submitted for publication August 7, 2013; accepted in revised form February 25, 2014.

MTX Decision Aid for Patients With RA

Significance & Innovations ●

Patient decision aids are evidence-based tools designed to support treatment decision making. Little is known if and how these tools help patients with arthritis.



We developed an online patient decision aid with animated videos for rheumatoid arthritis (RA) patients who were prescribed methotrexate, but were unsure about starting it.



Patients’ decision comfort and knowledge improved after using the patient decision aid, but their perception of the information was largely shaped by their prior knowledge and experiences with RA.

matic drugs (DMARDs) such as methotrexate should be used within 6 months of diagnosis to minimize irreversible joint damage and improve long-term physical function (1–7). Delays as brief as 3 months in starting DMARDs have been associated with poorer long-term outcomes, including greater physical disability (8 –10) and joint damage (11–13), and less chance of remission (14). Despite the benefits, the evidence consistently shows that less than half of patients are receiving DMARDs after a recent diagnosis (15–17). From the patient perspective, reasons for not taking DMARDs included a lack of physician advice, fear of side effects, and preference to avoid medications (18). They also perceived a lack of information presented in ways that they can fully understand and apply (19). A major dilemma about medication use is the tradeoff between benefits and risks. For DMARDs, such as methotrexate, side effects/adverse events may include nausea, diarrhea, headaches, increased risk of infections, rash, increased liver enzymes, and bone marrow suppression, although some of these events may also happen in the absence of DMARDs (2,4). For example, in randomized controlled trials, approximately 10% of patients stopped methotrexate because of side effects and adverse events, while 9% also stopped the placebo due to the same reasons (4). Some adverse events, such as liver toxicity, can be fatal, but these occurrences are rare. Making a treatment decision is a challenge for some people with chronic conditions. Several qualitative studies involving patients with chronic diseases have alluded to the profound feeling of ambivalence toward medications (19 –22). This is marked, on one hand, by an aversion to drugs because of anticipated side effects, and on the other hand, by a fear of the potentially disabling effects of an uncontrolled disease without medication. It appeared that patients who stayed on medications tended to have a good rapport and were more engaged in communicating and making treatment decisions with their doctors (19 – 22). To assist patients in making treatment decisions, patient decision aids have been developed to help individuals to choose between 2 or more options (23–26). Patient decision aids help people personalize the information about treatment effectiveness, outcomes, and the inherent

1473 uncertainties of potential benefits versus potential harms. These tools are particularly useful when a patient feels that one treatment option is not unambiguously better than other options (27,28). A few patient decision aids are available for patients with RA to use before a clinic visit (29,30). A common limitation of existing decision aids is the lack of user friendliness, which may have contributed to poor uptake of this type of tools in patient care (31). To address this issue, we developed a patient decision aid called the Animated, Self-serve, Web-based Research tool (ANSWER) for patients considering methotrexate for RA (http://answer. arccanada.org/). The ANSWER aimed to provide unbiased information on benefits and risks of methotrexate for RA and to guide users through thinking if this is the “right” treatment for them based on the information and their personal preferences. The purpose of this mixed-methods study was to assess the extent to which the ANSWER improved decisional quality, methotrexate knowledge, and selfmanagement skills. We hypothesized that a user-friendly online decision aid reduces patients’ decisional conflict, improves their knowledge about methotrexate, and improves their skills of being effective health care consumers. Furthermore, we examined patients’ experiences with this new tool.

PATIENTS AND METHODS Patient decision aid intervention. Development of the ANSWER was guided by the International Patient Decision Aids Standards (23,32), our qualitative study on the helpseeking experience of patients with early RA (22), and input from patient/consumer collaborators. The target audience was patients who had been prescribed methotrexate for RA, but were feeling uncertain about starting it. The ANSWER was an online interactive program designed to be used after individuals were recommended methotrexate for RA. They might use the program either at a rheumatology clinic or at home. In addition to providing information, the ANSWER guided users to think about concerns and questions about using methotrexate. It focused on 2 options: to take methotrexate as prescribed or to refuse methotrexate and talk to the doctor about other treatment options. It consisted of an information module and a password-protected value clarification module. In a usability test, patients with RA took a mean ⫾ SD of 56.1 ⫾ 34.8 minutes to review and reflect on all the contents within one session (33); however, participants in the proof-of-concept study were allowed to complete the program in multiple sessions and were instructed to use it based on their knowledge needs. Individuals who had good knowledge of RA, for example, might choose to spend less time on the content about RA. The information module was based on a Cochrane systematic review on methotrexate for RA (2) and the evidence-based recommendations from the 3E (Evidence, Experts, Exchange) Initiative (34). The module addressed 6 topics: 1) about RA, 2) about methotrexate, 3) side effects of methotrexate, 4) planning a family and methotrexate, 5) alcohol use and methotrexate, and 6) other medication

1474

Li et al

Figure 1. Animated, Self-serve, Web-based Research tool (ANSWER) summary for patient.

options and adjunctive treatments (e.g., exercise and joint protection techniques). To cater to an audience with different learning styles, each topic was presented in text, voice narration, and a vignette. The latter was shown in 6 animated videos; each was approximately 4 –7 minutes in length. The value clarification module could be accessed by a username and password set up by the patient. In this module, patients were asked to rate on a 5-point scale the importance of 1) improving joint pain, 2) preventing joint damage, 3) improving physical function, 4) avoiding side effects, 5) becoming pregnant/starting a family, and 6) consuming alcohol. They were also asked to list their questions/concerns about using methotrexate and to indicate their preferred choice of the 2 options, or to declare that they remained uncertain. All information was stored in a secured server at the Arthritis Research Centre of Canada. Once completed, the ANSWER produced a 1-page

summary (Figure 1), which could be printed by the patient for their next rheumatologist visit or sent electronically to the rheumatology clinic. This summary was intended to facilitate a discussion between the patient and the rheumatologist about using methotrexate (i.e., decision coaching). According to the decision aid literature, decision coaching could be done by physicians, nurses, or other health professionals. This component is beyond the scope of this study (35,36). The ANSWER was made available for public access after this proof-of-concept study at http:// answer.arccanada.org/. Study design. A prospective mixed-methods pre-post design was used in this proof-of-concept study. Eligible individuals were patients who 1) were diagnosed with RA; 2) were candidates for methotrexate as indicated by their rheumatologists, but self-identified as unsure about initi-

MTX Decision Aid for Patients With RA ating treatment; and 3) had internet access. We excluded individuals who had already started methotrexate. Recruitment flyers were posted in rheumatologist offices in British Columbia, Alberta, and Ontario, Canada, and distributed by patient/consumer groups, including the Canadian Arthritis Patient Alliance and Arthritis Consumer Experts. In addition, we posted the recruitment information on web sites and Facebook sites of the Arthritis Research Centre of Canada and The Arthritis Society. Interested individuals were invited to contact the research coordinator, who provided details about the study, screened for eligibility, and obtained informed consent. Eligible participants were granted access to the ANSWER within 2 days after enrollment. They could use the tool at their own pace and were able to sign off and return later. They were also asked to complete a series of outcome measures before and within 2 days after using the ANSWER. At the time of enrollment, participants were informed that they had a 50% chance of being selected for a 60 –90minute telephone interview about their experience with the ANSWER within a month after using the tool. Consent to participate in the interview did not affect their eligibility to participate in the study. The semistructured interviews focused on 2 broad topics: participants’ experiences with the ANSWER and their use of the internet for health information. All interviews were audio recorded and transcribed verbatim. Transcripts were checked for accuracy against the recordings and all identifying information was removed. Participants will only be identified by their selected pseudonyms in transcripts and publications. Outcome measures. The Decisional Conflict Scale (DCS), the primary outcome measure, evaluates personal perceptions of uncertainty in choosing options, factors contributing to uncertainty, and effective decision making (37). The low literacy version has 10 questions and 3 response categories (yes, no, and unsure). DCS scores range from 0 –100 (where 0 ⫽ no decisional conflict and 100 ⫽ extremely high decisional conflict), with scores ⬍25 associated with following through with decisions and scores ⬎37.5 associated with decision delay or feeling uncertain about implementation (37). The DCS consists of 4 subscales (uncertainty, informed, values clarity, and support), each of which can be standardized to range from 0 –100 points. Both internal consistency and test–retest reliability of the DCS exceed 0.78. In studies of decision aids for a variety of screening and treatment decisions, the effect sizes range from 0.4 –1.2 (27,37,38). The Methotrexate in RA Knowledge Test (MiRAK) is a unidimensional scale consisting of 60 items on knowledge about methotrexate, with 3 response categories (true, false, and don’t know) (39). The total score is 60; each correct response gets 1 point. The questionnaire was tested in 131 patients with RA, of whom ⬎50% had a high school education or lower and 11% had English as a second language (40). The results have demonstrated internal consistency (Cronbach’s ␣ ⫽ 0.84) and test–retest reliability (intraclass correlation coefficient 0.89) (40).

1475 The Effective Consumer Scale (EC-17) was developed within the Outcomes Measures in Rheumatology process with researchers, health professionals, and people with arthritis (41,42). This unidimensional scale was designed to evaluate patients’ perceptions of their ability to effectively manage and participate in their health care. Development of the EC-17 was based on a comprehensive literature review and semistructured interviews with people living with arthritis and their caregivers. The initial scale had 48 items, which was later reduced to 17 after item reduction using classic and item response theory (43). Sample size and statistical analysis. Sample size calculation was based on previous before and after studies using the DCS (27,37). Applying a difference of 0.4, 0.35, and 0.3, with an SD of 0.6 (alpha level 0.05, 90% power), a sample size of 24, 31, and 43, respectively, would be needed. We decided a priori to continue recruitment until the upper bound of the sample size estimation was reached or at the end of an 18-month recruitment period. For the telephone interviews, we estimated approximately 10 participants would be required. Consistent with qualitative research methods, the final sample size for the interviews was determined when the point of information redundancy (i.e., data saturation) was reached (44). If by the tenth interview, saturation of the data had not occurred, we would continue interviewing until saturation was reached. Exploratory analysis was conducted using paired t-tests to assess changes in the 3 outcome measures. Effect size was calculated for each outcome measure by dividing the mean difference before and after the intervention by the SD at baseline. Descriptive statistics were used to summarize participant characteristics and health status. For the participant interviews, a single investigator (LCL) conducted thematic analysis using an iterative process, whereby codes were identified and then revised as more interviews were analyzed. Initial open coding, i.e., assigning conceptual labels to the content, was followed by clustering the labels into thematic categories. Quotations representative of the thematic categories were identified. The research protocol was approved by the University of British Columbia Behavioural Research Ethics Board (application H09-00898).

RESULTS Between November 2010 and April 2012, 43 patients expressed interest in participation (Figure 2). Ten were deemed ineligible and 1 declined to participate after screening. Of the 32 consented participants, 2 subsequently dropped out, and 30 completed the study (Table 1). The majority of participants were women (n ⫽ 23 [76.7%]), the mean ⫾ SD age was 54.90 ⫾ 14.91 years, and 73.3% (n ⫽ 22) attended/graduated from university. The median disease duration was 1 year (interquartile range 0.3–5.0 years) and the mean ⫾ SD Health Assessment Questionnaire score was 1.16 ⫾ 0.68. The mean ⫾ SD DCS score was 49.50 ⫾ 23.17 preintervention and 21.83 ⫾ 24.12 postintervention (change

1476

Li et al In-depth interviews. Eight women and 3 men ages 31– 67 years were interviewed (Table 3). Of the 11 participants, 7 declared methotrexate as their preferred choice and 4 remained unsure. Regardless of their decision, participants valued the opportunity to review the information about methotrexate and consider the options at their own time and pace, as compared to during a medical visit. This was particularly appreciated by patients who were recently diagnosed: “. . .it’s a format. . .that you can absorb at your own pace, that helps to put some clarity, like when you go for an appointment, it’s not just your time, it’s the health professional’s time too and you’re aware that they have a limit to the amount of time that they have available. Whereas in a format like this [ANSWER], it’s only your time so you can take as much time as you want it” (Amy, diagnosed less than 1 week before using the ANSWER).

Figure 2. Study design.

⫺27.67 of 100 [95% confidence interval (95% CI) ⫺39.89, ⫺15.44]; P ⬍ 0.001, effect size 1.19). Significant improvement was also found in all subscales (Table 2). At baseline, 10% of participants scored ⬍25 on the DCS (i.e., a high likelihood of implementing a decision). This improved to 66.7% after the intervention (P ⬍ 0.001). Similar results were observed in the MiRAK, with a change of 11.03 of 60 (95% CI 6.73, 15.34; P ⬍ 0.001) after the intervention, but not in the EC-17 (change 4.71 of 100 [95% CI ⫺1.81, 11.22]; P ⫽ 0.15). After using the ANSWER, 20 participants (66.7%) were able to make a decision (14 chose to take methotrexate and 6 declined methotrexate and would talk to their doctors about other treatment options). Ten participants (33.3%) remained unsure about their preferred choice.

Although uncertain about using methotrexate, all interview participants had some knowledge about RA and its treatment at the time of enrollment. This prior knowledge appeared to shape how they perceived the information presented in the ANSWER. Our analysis identified 3 themes related to their experiences: seeking confirmation of one’s own knowledge of RA and methotrexate, amplifying reluctance when individuals encountered information that was different from their own experiences with the disease, and clarifying thoughts about the next step by reviewing the information and completing the value clarification module. Seeking confirmation. Participants indicated that their desire to learn more about methotrexate was the main impetus for participating in this study (George) (Table 4). While using the ANSWER, they constantly compared the new information with what they had known and experienced. In addition to seeking confirmation on their own

Table 1. Participant characteristics*

Age, mean ⫾ SD years Women, no. (%) Disease duration, median (IQR) years Education, no. (%) University (attended/graduated) High school graduate Had not completed high school Employment, no. (%) Employed Retired/homemaker Disability leave Married, no. (%) Annual family income, no. (%)† ⬍$40,000 $40,000–80,000 ⬎$80,000 No answer HAQ score, mean ⫾ SD

All (n ⴝ 30)

Interview participants (n ⴝ 11)

54.9 ⫾ 14.9 23 (76.7) 1.0 (0.3–5.0)

53.9 ⫾ 11.4 8 (72.7) 0.5 (0.05–3.5)

22 (73.3) 6 (20.0) 2 (6.7)

7 (63.6) 3 (27.3) 1 (9.1)

13 (43.3) 13 (43.3) 4 (13.3) 22 (73.3)

5 (45.5) 4 (36.4) 2 (18.1) 9 (81.8)

11 (36.7) 3 (10.0) 12 (40.0) 4 (13.3) 1.2 ⫾ 0.7

5 (45.5) 0 (0) 5 (45.5) 1 (9.1) 1.2 ⫾ 0.6

* IQR ⫽ interquartile range; HAQ ⫽ Health Assessment Questionnaire. † Canadian dollars.

MTX Decision Aid for Patients With RA

1477

Table 2. Pre-post study results (n ⴝ 30)*

DCS (range 0–100, lower ⫽ less conflicted) Uncertainty subscale Informed subscale Values clarity subscale Support subscale Methotrexate in RA Knowledge Test (range 0–60, higher ⫽ better) EC-17 (range 0–100, higher ⫽ better) DCS score, no. (%) ⬍25 25–37.5 ⬎37.5

Before

After

Difference (95% CI)

P

49.50 ⫾ 23.17 78.33 ⫾ 31.98 38.33 ⫾ 34.51 42.50 ⫾ 34.83 44.44 ⫾ 30.43 30.62 ⫾ 9.62

21.83 ⫾ 24.12 37.50 ⫾ 43.43 18.89 ⫾ 35.49 11.67 ⫾ 22.49 21.11 ⫾ 25.12 41.67 ⫾ 6.81

⫺27.67 (⫺39.89, ⫺15.44) ⫺40.83 (⫺60.54, ⫺21.12) ⫺19.44 (⫺37.53, ⫺1.36) ⫺30.83 (⫺45.99, ⫺15.68) ⫺23.33 (⫺37.75, ⫺8.91) 11.03 (6.73, 15.34)

⬍ 0.001 ⬍ 0.001 0.036 ⬍ 0.001 0.002 ⬍ 0.001

68.24 ⫾ 12.46

72.94 ⫾ 12.74

4.71 (⫺1.81, 11.22) –

0.15

3 (10.0) 6 (20.0) 21 (70.0)

20 (66.7) 3 (10.0) 7 (23.3)

⬍ 0.001

* Values are the mean ⫾ SD unless indicated otherwise. 95% CI ⫽ 95% confidence interval; DCS ⫽ Decisional Conflict Scale; RA ⫽ rheumatoid arthritis; EC-17 ⫽ Effective Consumer Scale.

knowledge, they also sought validation of their experiences as someone living with RA by viewing the ANSWER animated videos (Lucy) (Table 4). In general, participants found it to be a positive experience when they saw that the ANSWER content matched what they had learned from their health care providers (Arty) (Table 4). Amplifying reluctance. For participants who perceived a mismatch between the decision aid content and their own experience with RA, the use of the ANSWER invoked frustration, especially when they were faced with the uncertainty of side effects. This amplified the sense of reluctance about using methotrexate in some participants. For example, Oscar believed that her joint symptoms were well managed by nonsteroidal antiinflammatory drugs and was unsure about starting methotrexate. She wanted to know the chances of different side effects, but was dissatisfied by the level of detail she received from the ANSWER (Table 4). Oscar was frustrated by the experience and continued to feel uncertain about using methotrexate after the intervention. Conversely, Hannah firmly believed that lifestyle changes should be tried before taking methotrexate, which contradicted the information presented in the ANSWER (Table 4). In the end, she did not find the patient

decision aid helpful and was unable to decide whether she should start methotrexate or discuss with her rheumatologist about other treatment options. Clarifying thoughts about the next step. Participants felt that the process of reviewing and thinking through the pros and cons of methotrexate had brought clarity about the next step after using the ANSWER (Lucy, DE2R, Cher, Jeremy, and Amy) (Table 4). Several found the value clarification module helpful for illuminating their preferences, and this subsequently helped them reach a preferred choice (DE2R, Cher, and Jeremy) (Table 4). Interestingly, 1 participant (Amy) (Table 4) who remained unsure after using the ANSWER came up with new questions about methotrexate during the process. She discussed these questions with a rheumatology nurse 2 days later, and subsequently started methotrexate the next day. For some participants, the practice of asking questions during medical visits carried on after the initial methotrexate decision: “. . . I have some different side effects [from methotrexate] . . . , there’s things that are happening to me that didn’t happen to me before and I know now that I need to really be open with her

Table 3. Interview participant characteristics* Participant pseudonym

Age, years

Sex

Time since diagnosis

DCS before

DCS after

Preferred choice after using ANSWER

Cher George Arty Lucy DE2R Marilyn Jeremy Kneedler Hannah Oscar Amy

49 59 61 43 51 31 64 67 49 65 40s†

F F M F M F M F F F F

3 months 1 month 2 years 1 year 6 months 5 years ⬍1 week 16 years 8 years 1 week ⬍1 week

0 85.0 50.0 35.0 50.0 20.0 45.0 75.0 35.0 30.0 85.0

0 5.0 55.0 5.0 45.0 5.0 0 5.0 0 25.0 30.0

Methotrexate Methotrexate Methotrexate Methotrexate Methotrexate Methotrexate Methotrexate Not sure Not sure Not sure Not sure

* DCS ⫽ Decisional Conflict Scale (range 0 –100, lower ⫽ less decisional conflict); ANSWER ⫽ Animated, Self-serve, Web-based Research tool. † This participant did not provide her exact age.

1478

Li et al

Table 4. Themes demonstrating experiences of participants using the ANSWER (Animated, Self-serve, Web-based Research tool) Quotations from participants Seeking confirmation

Amplifying reluctance

Clarifying thoughts

It [methotrexate] was prescribed for me and I filled it—and didn’t take it . . . I knew I was kind of in trouble because I had carried this pill bottle for six months and hadn’t taken anything yet. . . . And oddly enough, as I was sitting in the waiting room [of the doctor’s office] is when I saw the sign on the wall about this program. (George) Just seeing some of the things [on the ANSWER videos] . . . it gives it some validation that your feelings or when you feel silly that you have to tell someone that you need help with something or that there’s something that you can’t do because of your medication or your condition, and just to see that acted out or role-played just gives identity, it makes it feel okay perhaps. (Lucy) I had a very thorough discussion with the pharmacist . . . she explained to me all the different side effects and the possibility of side effects. But in the [ANSWER] program itself, I mean it asked me those questions. . . . Are you aware of this? Are you aware of this? Well, in the end, it allowed me to understand that first of all, the pharmacist was very thorough with me and I think my doctor has been very upfront with me about it. . . . (Arty) I didn’t expect that my arthritis was ready for a drug like methotrexate because I had been taking, all along . . . , Celebrex, naproxen. . . . Now she gave me the prescription for methotrexate. . . . It’s the side effects that bother me mostly and that’s probably why I haven’t filled it yet. (Oscar) . . . It seemed anywhere I looked [on ANSWER] the side effects were the same, but it didn’t tell me how common, I didn’t seem to find any like percentages of how often do these side effects happen. (Oscar) Well, it [ANSWER] wasn’t particularly helpful for me because I was still, I still had the mindset that . . . these things [quality of life] are important to me, but it doesn’t tell me that I need to use methotrexate. And everything else that I was discovering . . . and that the doctors were discovering at the time told me, or in a sense almost give me . . . a reinforcement for my view that why change something that’s working for me? . . . I had been prescribed a low-dose steroid and everything was calming down. . . . I knew by the time I got to the decision-making stage with the ANSWER program, I knew that I didn’t have significant RA damage yet, that what I had was osteo damage. (Kneedler) It [the ANSWER videos] made me realize you know what’s important and what isn’t . . . but it didn’t make me, at the end, I still didn’t feel like I should be going on it right away. I mean if in time I have no choice ’cause I can’t live the way you know I am forever, I mean I’m not bedridden and I’m not in severe pain, but you know I walk with a bit of a limp and you know my hands in the morning are stiff and stuff. So I know I have to do something and so my first step is to see if quitting smoking will help and if it doesn’t then I would like to try sulfasalazine, and then if it doesn’t work then I guess I have no choice but to go on methotrexate. (Hannah) I thought the whole process was both interesting and informative. I had done a lot of research about methotrexate on my own beforehand and I was really quite reluctant to use it. And the whole process helped me to, it didn’t convince me to, I wouldn’t say it convinced me, but what I would say is it helped me accept it was the best choice. (Lucy) What the ANSWER program did for me was that it just, it assisted me in recognizing certain things, identifying with certain things. (DE2R) . . . from everything I had read, so answering the questions and stuff just kind of helped to know that I was making the right decision. (Cher) Once I’d done the ANSWER questionnaire and seen the videos and read the literature, I decided. . . . It’s obviously the best route to go initially. (Jeremy) . . . even though at the end I was kind of disappointed that I still hadn’t decided based on the information that was presented, and I thought . . . I should have decided based on that, this was good information. But it did help me to identify two questions which then I took to the clinical resource nurse on Wednesday and I started the drug on Thursday. (Amy)

[my doctor] and find out if these are in fact something that comes from taking the drugs. And so I think the program opened my eyes to say yeah, you can ask that question” (Arty, diagnosed 2 years before using the ANSWER).

DISCUSSION This study demonstrated that a novel online patient decision aid could reduce decisional conflict and improve knowledge of patients who self-identified as feeling uncertain about starting methotrexate. The mean reduction of

the DCS was 27.67 of 100 points, with an effect size of 1.19. Our finding is consistent with previous pre-post patient decision aid studies that measured DCS, with effect sizes ranging from 0.4 –1.2 (37). Another important finding is that two-thirds of participants went from feeling unsure at baseline to making a decision after using the ANSWER. In addition, their methotrexate knowledge significantly improved after the intervention. The large effect size may be related to the fact that only participants who were most in need of decisional support (i.e., those who felt uncertain

MTX Decision Aid for Patients With RA about the treatment options and had not initiated methotrexate) were enrolled in this study. In contrast, previous studies often included patients who had already made a treatment decision (37,45). However, this proof-of-concept study should be interpreted with caution due to the lack of a randomization or a control group. Our results lend support to the hypotheses that a userfriendly patient decision aid that provides unbiased information reduces decisional conflict and improves knowledge. It failed, however, to find a significant difference in participants’ perception of their ability to effectively manage and participate in their health care. One reason could be insufficient statistical power to detect a change in the EC-17, since the sample size was estimated based on the DCS. Another explanation could be the timing of the measurement. Participants completed the outcome measures before and after using the ANSWER. Unlike knowledge and decision quality, for which an immediate effect is expected, it likely takes longer for individuals to develop effective consumer skills, such as using health information, setting priorities, and communicating with health professionals. Furthermore, the EC-17 might measure broader concepts about disease self-management beyond the scope of a patient decision aid targeting one specific treatment decision. Nevertheless, findings from the indepth interviews suggested that some participants did develop some effective consumer skills, such as being prepared to ask questions during medical visits, after the brief intervention period. Therefore, it might be useful to explore the change of EC-17 in a 3– 6-month followup period in future studies. Our qualitative analysis of a subsample of participants revealed that their understanding of the information and the subsequent decision making was largely shaped by their prior experiences and knowledge about RA and its treatment. As such, the ANSWER served either as a credible tool that confirmed and clarified their treatment preferences or as a source that amplified frustration and reluctance in the decision-making process. Since an individual’s prior knowledge might come from various sources and include a mix of fact and myth, user experience might be further improved by explicitly addressing myths about RA and its treatment. This hypothesis is testable in future studies. Also, given the issues raised about the animated video, further refinement in the storyline and presentation would further improve decision aid users’ experiences. Development of the ANSWER was driven by 3 issues in RA care: 1) the low use of effective DMARDs (15–17), 2) patients’ feeling of uncertainty and ambivalence toward medication (19,22,46), and 3) patients’ desire for information that can be accessed at a time and place of their choice (47). Our approach is in line with knowledge-to-action process by Graham et al (48), wherein we identified a care gap, examined the challenges experienced by patients in using an effective treatment, and designed a user-friendly decision support tool that could be accessed anywhere with an internet connection. The next step would be to further refine this tool and examine the effectiveness of this tool for improving shared decision making in rheumatology practices. Furthermore, given that the uptake of patient decision aids has been modest across clinical spe-

1479 cialties (31), future implementation research examining barriers to patient decision aid use and strategies to improve uptake is warranted. This study has several limitations. First, although the selected outcome measures captured changes from the patient’s perspective, we were not able to determine if the ANSWER affects the rheumatologist’s participation in shared decision making or if it improves patient– physician communication during the clinician visit. Second, this study employed a pre-post study design, which is inherently limited for studying the effect of an intervention. Nonetheless, our work has added to the large body of literature synthesized by the recent Cochrane review (36), which found that patient decision aids improve knowledge, reduce decisional conflict, increase patient engagement in decision making, and reduce the feeling of uncertainty. Third, although the ANSWER could be viewed in any order, this tool was not designed explicitly to show information in the order of importance deemed by the individual. Since people may read and recall information presented first (i.e., the primacy effect) (49,50), future computerized decision aids may benefit from starting with the value elicitation module to obtain the level of importance of different treatment attributes to the patient, and then use this information to individualize the presentation of information. Fourth, the time spent by each participant in using the patient decision aid was not recorded. Therefore, we did not know how much time they spent on each section compared to those in the usability testing, where they reviewed and commented on all content in detail. Finally, more than 70% of the participants attended or graduated from university; therefore, generalizability of the results is likely limited to those with high literacy or socioeconomic status. A strength of this study is the inclusion of patients at the time when an actual treatment decision was needed. Our recruitment, however, was challenged by the stringent eligibility criteria, which limited our sampling pool. Despite a broad recruitment strategy, we only enrolled 30 participants within 18 months. In conclusion, we have developed an online patient decision aid to assist people making decisions about taking methotrexate for RA. This mixed-methods study showed that our decision aid improved patients’ decisional quality and knowledge. Furthermore, our findings suggest that user experience can be further improved by explicitly addressing myths about the disease and treatment options in patient decision aids.

ACKNOWLEDGMENTS The authors are grateful for the support of patient/ consumer collaborators, including Otto Kamensek (Arthritis Research Centre Consumer Advisory Board), Cheryl Koehn (Arthritis Consumer Experts), Colleen Maloney (Canadian Arthritis Patient Alliance), health education consultant Gwen Ellert, and information scientist Jessie McGowan. We also thank our digital media collaborators Jeannette Kopak and George Johnson (Centre for Digital Media) for organizing and supervising 2 Master of Digital Media student teams to develop the ANSWER. The Design

1480

Li et al

and Development Team consisted of Conrad Chan, Fouad Hafiz, Felwa Abukhodair, Liam Kelly, Karin Schmidlin, Yamin Li, and Shao Yingyun. The Production Team included Shahrzad Aghasharifianesfahani, Erez Barzilay, Jason Ho, Milim Kim, Clark Kim, Natalia Mitrofanova, and Al Sinoy. The ANSWER programming was led by Matt Jenkins. Original music in the videos was composed by Ben Euerby.

12. 13.

14.

AUTHOR CONTRIBUTIONS All authors were involved in drafting the article or revising it critically for important intellectual content, and all authors approved the final version to be published. Dr. Li had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study conception and design. Li, Adam, Backman, Jones, Lacaille, Townsend, Yacyshyn, Tugwell, Stacey. Acquisition of data. Li, Lineker, Jones, Yacyshyn, Yousefi, Leese. Analysis and interpretation of data. Li, Townsend, Stacey.

15.

16. 17.

REFERENCES 1. Singh JA, Furst DE, Bharat A, Curtis JR, Kavanaugh AF, Kremer JM, et al. 2012 update of the 2008 American College of Rheumatology recommendations for the use of diseasemodifying antirheumatic drugs and biologic agents in the treatment of rheumatoid arthritis. Arthritis Care Res (Hoboken) 2012;64:625–39. 2. Suarez-Almazor ME, Belseck E, Shea BJ, Tugwell P, Wells G. Methotrexate for treating rheumatoid arthritis. Cochrane Database Syst Rev 2008;3:CD002047. 3. Osiri M, Shea B, Robinson V, Suarez-Almazor M, Strand V, Tugwell P, et al. Leflunomide for treating rheumatoid arthritis. Cochrane Database Syst Rev 2008;3:CD000957. 4. Suarez-Almazor M, Osiri M, Emery P, and the Ottawa Methods Group. Rheumatoid arthritis. In: Tugwell P, Shea B, Boers M, Brooks P, Simon LS, Strand V, et al, editors. Evidencebased rheumatology. London: BMJ Books; 2003. p. 287–94. 5. Sokka T, Mottonen T, Hannonen P. Disease-modifying antirheumatic drug use according to the ‘sawtooth’ treatment strategy improves the functional outcome in rheumatoid arthritis: results of a long-term follow-up study with review of the literature. Rheumatology (Oxford) 2000;39:34 – 42. 6. Fries JF, Williams CA, Morfeld D, Singh G, Sibley J. Reduction in long-term disability in patients with rheumatoid arthritis by disease-modifying antirheumatic drug– based treatment strategies. Arthritis Rheum 1996;39:616 –22. 7. Abu-Shakra M, Toker R, Flusser D, Flusser G, Friger M, Sukenik S, et al. Clinical and radiographic outcomes of rheumatoid arthritis patients not treated with disease-modifyingdrugs. Arthritis Rheum 1998;41:1190 –5. 8. Munro R, Hampson R, McEntegart A, Thomson EA, Madhok R, Capell HA. Improved functional outcome in patients with early rheumatoid arthritis treated with intramuscular gold: results of a five year prospective study. Ann Rheum Dis 1998; 57:88 –93. 9. Tsakonas E, Fitzgerald AA, Fitzcharles MA, Cividino A, Thorne JC, M’Seffar A, et al. Consequences of delayed therapy with second-line agents in rheumatoid arthritis: a 3 year followup on the Hydroxychloroquine in Early Rheumatoid Arthritis (HERA) Study. J Rheumatol 2000;27:623–9. 10. Van der Heide A, Jacobs JW, Bijlsma JW, Hcurkens AH, Booma-Frankfort C, van der Veen MJ, et al. The effectiveness of early treatment with second-line antirheumatic drugs: a randomized, controlled trial. Ann Intern Med 1996;124:699 – 707. 11. Lard LR, Visser H, Speyer I, van der Horst-Bruinsma IE, Zwinderman AH, Breedveld FC, et al. Early versus delayed treatment in patients with recent-onset rheumatoid arthritis:

18. 19. 20. 21. 22.

23. 24.

25. 26. 27.

28.

29. 30.

31.

comparison of two cohorts who received different treatment strategies. Am J Med 2001;111:446 –51. Buckland-Wright J. Quantitative microfocal radiography detects changes in erosion area in patients with early RA treated with myochrisine. J Rheumatol 1993;20:243–7. Egsmose C, Lund B, Borg G, Pettersson H, Berg E, Brodin U, et al. Patients with rheumatoid arthritis benefit from early 2nd line therapy: 5 year followup of a prospective double blind placebo controlled study. J Rheumatol 1995;22:2208 –13. Mottonen T, Hannonen P, Korpela M, Nissila M, Kautiainen H, Honen J, et al, for the FIN-RACo Trial Group. Delay to institution of therapy and induction of remission using singledrug or combination– disease-modifying antirheumatic drug therapy in early rheumatoid arthritis. Arthritis Rheum 2002; 46:894 – 8. Shipton D, Glazier RH, Guan J, Badley EM, Shipton D, Glazier RH, et al. Effects of use of specialty services on diseasemodifying antirheumatic drug use in the treatment of rheumatoid arthritis in an insured elderly population. Med Care 2004;42:907–13. Lacaille D, Anis AH, Guh DP, Esdaile JM. Gaps in care for rheumatoid arthritis: a population study. Arthritis Rheum 2005;53:241– 8. Widdifield J, Bernatsky S, Paterson JM, Thorne JC, Cividino A, Pope J, et al. Quality care in seniors with new-onset rheumatoid arthritis: a Canadian perspective. Arthritis Care Res (Hoboken) 2011;63:53–7. Lacaille D, Rogers P. Why are people with rheumatoid arthritis not using DMARDs? Understanding gaps in care [abstract]. Arthritis Rheum 2007;56 Suppl:S86. Townsend A, Hunt K, Wyke S. Managing multiple morbidity in mid-life: a qualitative study of attitudes to drug use. BMJ 2003;327:837. Kay EA, Punchak SS. Patient understanding of the causes and medical treatment of rheumatoid arthritis. Rheumatology (Oxford) 1988;27:396 – 8. Carder PC, Vuckovic N, Green CA. Negotiating medications: patient perceptions of long-term medication use. J Clin Pharm Ther 2003;28:409 –17. Townsend AF, Backman CL, Adam P, Li LC. A qualitative interview study: patient accounts of medication use in early rheumatoid arthritis from symptom onset to early post diagnosis. BMJ Open 2013;3:e002164. Ottawa Health Research Institute. International Patient Decision Aid Standards (IPDAS) checklist. 2008. URL: http:// decisionaid.ohri.ca/methods.html#checklist. Elwyn G, O’Connor A, Stacey D, Volk R, Edwards A, Coulter A, et al. Developing a quality criteria framework for patient decision aids: online international Delphi consensus process. BMJ 2006;333:417. O’Connor AM, Graham ID, Visser A. Implementing shared decision making in diverse health care systems: the role of patient decision aids. Patient Educ Couns 2005;57:247–9. Legare F, Stacey D, Forest PG. Shared decision-making in Canada: update, challenges and where next! Z Arztl Fortbild Qualitatssich 2007;101:213–21. O’Connor AM, Tugwell P, Wells GA, Elmslie T, Jolly E, Hollingworth G, et al. A decision aid for women considering hormone therapy after menopause: decision support framework and evaluation. Patient Educ Couns 1998;33:267–79. O’Connor AM, Stacey D, Entwistle V, Llewellyn-Thomas H, Rovner D, Holmes-Rovner M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2003;2:CD001431. Cochrane Musculoskeletal Group. Decision aids. 2011. URL: http://musculoskeletal.cochrane.org/decision-aids. Fraenkel L, Peters E, Charpentier PA, Olsen B, Errante L, Schoen RT, et al. Decision tool to improve the quality of care in rheumatoid arthritis. Arthritis Care Res (Hoboken) 2012; 64:977– 85. Legare F, Ratte S, Gravel K, Graham ID. Barriers and facilitators to implementing shared decision-making in clinical practice: update of a systematic review of health professionals’ perceptions. Patient Educ Couns 2008;73:526 –35.

MTX Decision Aid for Patients With RA 32. Elwyn G, O’Connor AM, Stacey D, Volk R, Edwards A, Coulter A, et al. Developing a quality criteria framework for patient decision aids: online international Delphi consensus process. BMJ 2006;333:417. 33. Li LC, Adam PM, Townsend AF, Lacaille D, Yousefi C, Stacey D, et al. Usability testing of ANSWER: a web-based methotrexate decision aid for patients with rheumatoid arthritis. BMC Med Inform Decis Mak 2013;13:131. 34. Visser K, Katchamart W, Loza E, Martinez-Lopez JA, Salliot C, Trudeau J, et al. Multinational evidence-based recommendations for the use of methotrexate in rheumatic disorders with a focus on rheumatoid arthritis: integrating systematic literature research and expert opinion of a broad international panel of rheumatologists in the 3E Initiative. Ann Rheum Dis 2009;68:1086 –93. 35. Stacey D, Murray MA, Legare F, Sandy D, Menard P, O’Connor A. Decision coaching to support shared decision making: a framework, evidence, and implications for nursing practice, education, and policy. Worldviews Evid Based Nurs 2008;5:25–35. 36. Stacey D, Bennett CL, Barry MJ, Col NF, Eden KB, HolmesRovner M, et al. Decision aids for people facing health treatment or screening decisions. Cochrane Database Syst Rev 2011;10:CD001431. 37. O’Connor A. User manual: Decisional Conflict Scale. Ottawa: Ottawa Health Research Institute; 2005. URL: http://decision aid.ohri.ca/docs/develop/User_Manuals/UM_Decisional_ Conflict.pdf. 38. Drake ER, Engler-Todd L, O’Connor AM, Surh LC, Alasdair H. Development and evaluation of a decision aid about prenatal testing for women of advanced maternal age. J Genet Couns 1999;8:217–33. 39. Ciciriello S, Wicks IP, Osborne RH, Buchbinder R. Development and validation of the methotrexate in rheumatoid arthritis knowledge test (MiRAKTM). Intern Med J 2010;40 Suppl:9.

1481 40. Ciciriello S, Buchbinder R, Osborne RH, Wicks IP. Improving treatment with methotrexate in rheumatoid arthritis: development of a multimedia patient education program and the MiRAK, a new instrument to evaluate methotrexate-related knowledge. Semin Arthritis Rheum 2014;43:437– 46. 41. Tugwell PS, Wilson AJ, Brooks PM, Driedger SM, Gallois C, O’Connor AM, et al. Attributes and skills of an effective musculoskeletal consumer. J Rheumatol 2005;32:2257– 61. 42. Tugwell PS, Santesso NA, O’Connor AM, Wilson AJ, and the Effective Consumer Investigative Group. Knowledge translation for effective consumers. Phys Ther 2007;87:1728 –38. 43. Kristjansson E, Tugwell PS, Wilson AJ, Brooks PM, Driedger SM, Gallois C, et al. Development of the effective musculoskeletal consumer scale. J Rheumatol 2007;34:1392– 400. 44. Lincoln Y, Guba E. Naturalistic inquiry. Beverly Hills: Sage; 1985. 45. Frankel L, Bogardus S, Concato J, Felson D. Preference for disclosure of information among patients with rheumatoid arthritis. Arthritis Care Res 2001;45:136 –9. 46. Townsend A, Wyke S, Hunt K. Self-managing and managing self: practical and moral dilemmas in accounts of living with chronic illness. Chronic Illn 2006;2:185–94. 47. Li LC, Townsend AF, Badley EM. Self-management interventions in the digital age: new approaches to support people with rheumatologic conditions. Best Pract Res Clin Rheumatol 2012;26:321–33. 48. Graham ID, Logan J, Harrison MB, Straus SE, Tetroe J, Caswell W, et al. Lost in knowledge translation: time for a map? J Contin Educ Health Prof 2006;26:13–24. 49. Schkade DA, Kleinmuntz DN. Information displays and choice processes: differential effects of organization, form, and sequence. Organ Behav Hum Decis Process 1994;57:319 – 37. 50. Lohse GL. Consumer eye movement patterns on yellow pages advertising. J Advert 1997;26:61–73.

Proof-of-concept study of a Web-based methotrexate decision aid for patients with rheumatoid arthritis.

To assess the extent to which an online patient decision aid reduced decisional conflict and improved self-management knowledge/skills in patients who...
332KB Sizes 3 Downloads 3 Views