Participant characteristics
The survey was administered to 43 children, young people and families (of the 64 approached) within the paediatric uveitis service.
Participant characteristics are shown in Table 1. The mean age was 10.9 years (SD 3.3, range 2–16, median 11), with 51% of participants under the age of 12 years. The average duration of disease was 3.2 years (SD 3.5, range 0–13, median 2), with 49% of patients reporting a duration of under 2 years. In terms of treatment, the majority had received steroid eye drops or steroid tablets plus at least one other immunomodulatory agent.
Self-rated knowledge
Self-rated knowledge scores were compared using the Wilcoxon Signed Ranks Test (Table 2). Answers of ‘not sure’ were classified as absence of knowledge. Self-marked understanding of uveitis improved significantly after watching the video (p = 0.001). Self-marked knowledge of all the tested terms (‘uvea’, ‘inflammation’, ‘glaucoma’, ‘cataract’, and ‘macular oedema’) significantly improved after video viewing (p < 0.001), with the most benefit seen for the words ‘uvea’ and ‘macular oedema’. Although, those with a new diagnosis of uveitis (< 2 years) had a greater degree of improvement in subjective knowledge compared to those with established disease (Z= -8.124, p < 0.001 and Z= -5.745, p < 0.001; respectively).
Objective knowledge
Objective knowledge scores were also compared using the Wilcoxon Signed Ranks Test (Table 3). Answers of ‘don’t know’ were classified as absence of knowledge, as were incorrect answers. There was a statistically significant improvement in scores across all 6 questions, with a particular increase in the correct identification as false of statements 2 (that ‘Inflammation means that part of the eye has an infection’; Z= -5.292, p < 0.001), 3 (‘Once you have uveitis it never goes away’; Z= -5.477, p < 0.001), and 5 (‘Cataract is when scar tissue crosses the front of the eye; Z= -5.477, p < 0.001).
To investigate whether age (under 12 years versus 12 years or older) or duration of disease (new, ie under 2 years, versus established disease of 2 years or longer) had an impact on improvement in objective knowledge (total possible score of 6; 1 for each question), the Mann-Whitney U Test was used. Both variables were found to have a statistically significant impact on outcomes (Z= -1.991, p = 0.047 and Z= -2.469, p = 0.014; respectively). There was a mean 3.73 net marks (SD 2.004) improvement for participants under 12 years, compared to a mean 2.52 net marks (SD 1.914) for those 12 years or older. Those with duration of disease under 2 years had a mean net improvement of 3.90 marks (SD 1.58), whilst the rest had a mean net improvement of 2.3 marks (SD 2.03).
To test the effect of treatment history on net objective scores, the Friedman Test was used. A score was assigned (out of a possible total of 6) for answers at T1 and T2. Then, the difference between both scores was calculated to determine the net improvement in objective knowledge. The biggest improvement was seen in the first treatment group (steroid drops or tablets only, mean 4.44, SD 1.424), with the second (steroids and at least one other agent, mean 2.78, SD 2.279) and third (steroids and at least one other agent and at least one surgical intervention, mean 1.89, SD 1.965) following in respective order. There was a statistically significant difference in net improvement of objective knowledge depending on which treatment group each participant belonged to, χ2(2) = 11.455, p = 0.003. Post hoc analysis with Wilcoxon signed-rank tests was conducted with a Bonferroni correction applied, resulting in a significance level set at p < 0.017. Median (IQR) net improvement in objective knowledge for treatment group 1, group 2 and group 3 were 5 (3 to 5), 3 (0 to 5), and 2 (0 to 3.5), respectively.
There were no significant differences in knowledge improvement between groups 1 and 2 (Z= -1.209, p = 0.227), or between groups 2 and 3 (Z= -0.851, p = 0.395). However, there was a statistically significant difference between groups 1 and 3 (Z= -2.694, p = 0.007). To compare the impact of various patient characteristics on outcome variables, the Wilcoxon Signed Ranks test was again used. There was a statistically significant difference noted for all demographics, with the most significant for those with disease duration under 2 years (Z= -3.975, p < 0.001) (table 4).
Change score
There was an overall increase in objective knowledge across all groups. The T1-T2 change score was greater for those aged less than 12 years compared to older children (mean rank 25.64 vs 18.19 respectively, U = 151, Z= -1.991, p = 0.047), as well as for those with less than 2 years of disease compared to longer (mean rank 29.36 vs 14.98 respectively, U = 76.50, Z= -3.917, p < 0.001) (supplementary material 3).
There was no statistically significant difference in T1-T2 change score between treatment groups 1 and 2 or groups 2 and 3. However, there was a significant improvement when comparing treatment groups 1 to 3 (mean rank 15.2 vs 8 respectively, U = 27, Z= -2.485, p = 0.13).
Satisfaction
Overall, there was a good level of satisfaction with the videos, with 65% of participants stating they learned something new (supplementary material 4). All participants rated the videos as very easy or quite easy to understand, with 77% and 70% stating that the amount of information and length of videos were just right, respectively.
A common theme encountered in the free text section at the end of the survey was the need for more information (8 of 43 families); this was across varying patient groups. There was particularly good feedback regarding the use of children’s voices, and agreement that these videos would be helpful for new families. One parent of an 11-year-old with an established diagnosis wrote “Would have liked this video when my son was younger - and was good for him to watch”, while one 14-year-old commented “Good video for new families.”