Using a Virtual World to Teach Joint Protection to People Living with Rheumatoid Arthritis: A Pilot Randomised Controlled Trial
MetadataShow full item record
Abstract Background: Rheumatoid arthritis (RA) is a systemic autoimmune disease affecting an estimated 1% of the global population. Joint protection is one intervention with some quality evidence of efficacy for RA self-management. However, joint protection education is often provided only in urban centres during Arthritis Self-Management Programs (ASMPs) in classroom sessions at designated times. These programs, therefore, may not be available to all who need them. Providing and testing more accessible methods of delivering joint protection education to people living with RA may improve accessibility. Aims: (i) To develop a virtual world (VW) intervention available via the Internet in Second Life®, that aims to improve the knowledge of joint protection among people with RA and (ii) to undertake a pilot randomised controlled trial (RCT) to assess the feasibility of conducting a subsequent large scale RCT. Methods: First, qualitative interviews with occupational therapists and clients living with RA who had previous experiences teaching or taking arthritis self-management programmes were undertaken and thematically analysed. This analysis informed the design of the VW joint protection education intervention. Second, the intervention was constructed and tested with these same participants. Their feedback helped refine the VW intervention and select assessment tools for the pilot RCT. Third, in a pilot RCT, three primary methods of advertising and invitation were used to recruit subjects: (i) poster invitations with take-home paper copies from clinical settings; (ii) direct messages to Twitter® users living with RA; and (iii) online discussion forums. Participants were recruited after contacting the principal investigator, reading an invitation letter and giving written informed consent. Participants were randomised to intervention or (30-day) waiting list control group, and completed a series of measures. These were completed after 30 days of program access for the treatment group and on enrolment in the study for the control group. Survey completion was online and included piloted knowledge-based questions about joint protection, validated during the second phase of the study with occupational therapists who were experts in joint protection education. A higher score was indicative of better joint protection knowledge. Standardized measures used on the survey included the Arthritis Impact Measurement Scale, Short Form, version II (AIMS2SF) and Pain Self-Efficacy Questionnaire (PSEQ). Results: It was possible to develop a VW education program focused on RA and joint protection based on the content identified by participants in the first part of the study and test with the tools selected. The program developed included input from client users, following the theorectical basis of occupational therapy as a client-centred practice. Additionally, the program developed applied principles of adult-learning and the recommendations of existing programs regarding chronic disease management. Recruitment of 50 participants for the pilot RCT was challenging, taking 6 months with low response rates for all three methods. The poorest response rates were to poster and paper invitations in clinical settings. The most effective means of recruitment was via electronic bulletin boards, such as blogs. All subjects, once randomised to the control or intervention group completed the online questionnaire. However, adherence to the intervention was poor; only 15 out of 25 randomised reported using the program. On the other hand, all 15 who used the program indicated that this medium was acceptable to learn about joint protection, despite 5/15 of these subjects reporting some difficulty accessing the program. All participants completed the three questionnaires (knowledge, impact, pain self-efficacy) and these may be useful in a definitive RCT. Although the main purpose of using Intention to Treat Analysis in pilot studies is to practice and check that analysis is feasible, there was a positive statistically significant difference between the treatment (x̄=52.8%) and control (x̄=24%) group scores on a test of joint protection knowledge using an independent samples t-test (F value, 20.8 p < 0.05) comparing joint protection knowledge scores after the treatment group had access to the program for 30 days. A higher score was indicative of better joint protection knowledge. The difference between the two groups was considerable, with the intervention group score mean being more than double that of the control group. Given the magnitude of this difference between groups, a smaller difference between groups would also be worth finding. The difference between groups for the AIMS2SF and PSEQ were not statistically significant using an independent samples t-test (F values, 0.5 and 0.2) but there was some suggestion that the intervention group scored more favourably on some of the subscales more relevant to joint protection on both the AIMS2SF and PSEQ, particularly noteworthy was a higher score pertaining to ability to carry out work on both measures. In a definitive trial a sample size of 1250 participants would give 80% power to find a difference of 28.8% on joint protection knowledge, weighted score of 1.8 on the AIMS2SF and overall score of 1.8 on the PSEQ at 5% level of significance. Smaller samples would be required if the PSEQ was dropped as a measure in a future study. Sample sizes of 14 and 558 would be required for the joint protection knowledge and AIMS2SF respectively at the same level of power and significance. Conclusion: A VW intervention to improve joint protection knowledge has been developed and is worth testing further. The intellectual contribution of the creation of this program using this methodology is that an occupational therapy based study using client input and priniciples of adult learning to create the intervention has been conducted, applying client-centred practice in research, which is, in reality, present in a minority of studies at this time. A full RCT would be feasible, though very challenging, given the numbers of subjects required for recruitment, most likely recruiting via the Internet on relevant RA focus sites, such as RA bloggers, and using the same outcome measures as in this study. A sample size of 1250 could feasibly be recruited in 36 months if a full time study were undertaken with suggestions discussed to assist with future study recruitment. However, given the number of study dropouts at enrolment seen in this study, close to double this number would be needed, entailing a recruitment period of up to 72 months, or 6 years, making a full RCT less practical. A future study may need to consider either a longer enrolment period, different outcome measures as well as address the limitations of this study, including the limited time of enrolment in this pilot RCT. However, longer enrolment duration would increase the amount of time required for a future full RCT, reducing the feasibility of a future study. Findings from this study indicate that the program developed would likely to be useful to people who are not able to access the urban centred classroom based program. On the other hand, those participants who used the program incurred no costs, appeared to have no risks or detrimental impact with possible improvement in knowledge and self-efficacy. Now the intervention has been developed, refinement, maintenance, and use is low cost for service providers, so it could be used routinely now for those who prefer it to ASMPs with an ongoing preference trial.