Watch this space: a systematic review of the use of video-based media as a patient education tool in ophthalmology

Authors: Farwana R, Sheriff A, Manzar H, Farwana M, Yusuf A, Sheriff I.

Geographical coverage: The US and Ghana 

Sector: Biomedical

Sub-sector: Treatment

Equity focus: None specified

Study population: Ophthalmology patients who received video-based media as an education tool.

Review type: Effectiveness review

Quantitative synthesis method: Systematic review

Qualitative synthesis method: Not applicable

Background:

Effective clinician-patient communication is particularly important in ophthalmology where long-term adherence to treatment is often required. However, in the context of increasingly pressurised clinics, there is a tendency to resort to written information leaflets not suited to patients with visual impairment, non-English speakers or those with low levels of literacy. Video-based media could be harnessed to enhance clinician-patient communication.

Objectives:

To assess the efficacy of using video-based media for patient education in ophthalmology. Secondary objectives included to identify the extent to which video-based media influences health behaviour; whether video-based patient education has an impact on overall clinical outcome in the ophthalmic population; and to describe currently reported methods of video-based patient education in ophthalmology.

Main findings:

Though more evidence is needed, overall findings note the use of video-based media appears to be effective in improving patient understanding and in certain cases may ameliorate overall outcome.

The search strategy yielded 481 studies. Following screening, 12 studies were deemed eligible to proceed to data extraction. Overall, 58.3% of studies were randomised controlled trials. The remainder were prospective non-randomised controlled trials (25%) and pre- and post-intervention studies (16.6%). A total of 1,650 participants were included (range 13-244). The majority of included studies were from researchers based in the USA. Eleven out of 12 studies were single centre studies. Four out of 12 studies assessed changes to health behaviour, of which 50% yielded statistically significant studies which specifically measured improvements in eye drop proficiency. Furthermore, seven out of 12 studies (58.3%) measured improvements in patients’ knowledge of a condition or procedure, of which five studies (71.4%) yielded statistically significant results for the intervention. Specifically, this assessed comprehension of cataract surgery, glaucoma, intravenous fluorescein angiography process and parental understanding of retinopathy of prematurity.

The majority of studies (58.3%) reported outcomes on patient comprehension with 5/7 (71%) showing statistically significant improvement after video intervention. Four studies (33.3%) reported on patient performance in a task (for example, drop application method) or overall health-related outcome with 2/4 (50%) showing statistically significant improvement after intervention.

Authors observed paucity of well-designed studies and future research is required to fully examine the role of video-based media in patient education. According to review authors future work requires national and international focus on developing standardised, high-quality video interventions and validated scoring systems to evaluate their impact. Access for patients to these interventions will also require some thought; in particular, those who are elderly or from low income households, where electronic devices and internet access may not always be readily available. Finally, any change must be sustainable. Pilot studies trialling video-based media as an educational tool should start as early as possible, so that results can guide regular developments and updates as the field of medical innovation and technology grows and lessens the time and cost needed to generate video content.

Methodology:

Inclusion criteria included peer-reviewed published studies, including primary and secondary research. Studies were included in this systematic review if they involved the use of video-based media as a sole intervention for patient education in ophthalmology with endpoints focused on its impact on patient knowledge, health behaviour or overall clinical outcome. Publications that did not fulfil the inclusion criteria and duplicates were excluded from the search.

Authors searched EMBASE, PubMed, Medline and PsycInfo databases from inception to July 2019.

Initial screening was conducted by two independent authors. Article titles and abstracts were read and assessed for suitability against the inclusion criteria and any duplicate studies removed. Studies which passed the initial screening criteria were put forward to full-text screening to yield the final articles that were included in the data extraction stage. Any discrepancies in study selection were discussed to reach a consensus and if this was unsuccessful, a third author was contacted to review and decide on whether to include the study or exclude it.

Following screening, data from selected studies was extracted independently by two authors. All yielded studies were collated onto a spreadsheet using Microsoft Excel. No further details were provided on how synthesis was undertaken.

Applicability/external validity: The authors identified several limitations in assessing the impact of video-based patient education in ophthalmology due to scarce literature and validated tools. The studies primarily used changes in knowledge scores or pre- and post-test questionnaires as outcome measures, which can be biased and limit objective outcome measurement. Most studies were small, single-center trials conducted in the USA and the western world. The limited number of studies and the diversity of interventions and outcomes reported make it difficult to conclusively validate the effectiveness of video-based media as an educational tool in ophthalmology.

Geographic focus: Almost all included studies were conducted in the US. At least one LMIC (Ghana) was included.

Summary of quality assessment:

The methods used to identify, include, and synthesise studies were generally thorough, with two authors performing all key tasks. However, the search was somewhat restricted – it was language-limited, there was no effort to include unpublished material, and no references were made to the reference sections of the included studies. Furthermore, no attempt was made to evaluate or score the risk of bias in the included studies. The analysis had several limitations, with scant details provided. Notably, the risk of bias in individual studies was not considered when interpreting their results. As the authors noted, the studies exhibited a high degree of heterogeneity in terms of the interventions considered, outcomes reported, and their measurement. Due to these reasons, we have low confidence in the review’s findings.