Feasibility of an Alternative Pathway for Hospital Referrals From Diabetic Eye Screening Wales (DESW) for People Suspected With Sight-threatening Diabetic Eye Disease Diabetic Maculopathy (FARSight-DM)
Feasibility of an Alternative Pathway for Hospital Referrals From Diabetic Eye Screening Wales (DESW) for People Suspected With Sight-threatening Diabetic Eye Disease Diabetic Maculopathy (FARSight-DM)
To assess a pathway using opticians based in their practices and a 'virtual' review by a consultant ophthalmologist, based in the hospital to assess suspected diabetic maculopathy. We want to know if this pathway works, is it acceptable to people with diabetes and what the changes might mean in terms of the outcomes for patients and NHS resources and cost?
Screening by Diabetic Eye Service Wales currently involves taking and as-sessing 2-D (dimensional) digital photographs of the back of the eye (reti-na). Leakage from damaged blood vessels can cause swelling within the central part of the retina, known as maculopathy. This swelling cannot be seen on traditional 2-D images.
Diagnosing maculopathy requires a 3-D camera and a technique known as Optical Coherence Tomography (OCT). OCT is not part of routine screening but, is available in the Hospital Eye Service (HES) and many optician prac-tices. If any changes are seen in the macular region of the retina on 2-D images, the patient will require an OCT scan to see if there is any in-creased thickness. Many patients do not have increased thickness and not everyone with diabetic retinopathy has maculopathy.
Currently after screening, patients who have macular changes that suggest possible maculopathy are referred to the HES for OCT imaging. Many of these patients will not have increased thickness and the appointment could have been better used to see a different patient with increased thickness that requires treatment.
The new pathway that will be investigated in this study involves trained opticians in practices that have an OCT camera taking the required 3-D images and carrying out the initial examination. The images and patient record will then be reviewed by an ophthalmologist working virtually, to decide the management plan. If this pathway is safe and acceptable to people with diabetes, it would reduce the pressure on HES clinics. It will also develop new skills for opticians and allow people to be seen closer to their home, reducing the stress that referrals can create and lowering the carbon footprint of the service.
1 Background Diabetes mellitus, as estimated by the International Diabetes Federation (IDF), affects 537 mil-lion people globally which is expected to increase to 783 million by 2045. DED is amongst the commonest and most feared complications of diabetes consisting of diabetic retinopathy and diabetic macular oedema. In 2020, world-wide, 103 million people were estimated to have DED, 15 million with vision threatening DED and 29 million having diabetic macular oedema, projected to reach 161 million, 45 million and 77 million respectively by 2045.
Until recently, DED has been the leading cause of blindness (Severe Sight Impairment, SSI) in the working age population in developed countries. However, since the introduction of screening for DED in the UK, along with improvements in diabetes and ophthalmological care, DED is no longer the leading cause of SSI in the UK. In Wales, new certifications for both sight impairment (SI) and SSI have fallen by 50% since screening commenced.
Screening for DED in Wales is performed by the DESW, utilising digital photography performed by trained photographers and the images are graded by trained readers, utilising standardised protocols and quality assurance. All cases of sight-threatening diabetic retinopathy (DR) i.e., severe pre-proliferative DR, proliferative DR and diabetic maculopathy require referral to ophthalmology for review and/or treatment. However, 2-dimensional (2-D) digital photography used by DESW, whilst adequate for assessing diabetic retinopathy lesions, cannot determine the presence of diabetic macular oedema (increased retinal thickness). Current screening uses surrogate markers for suspecting the presence of maculopathy such as exudates within 1 disc diameter of the fovea or microaneurysms and haemorrhages with an unexplained visual acuity of 6/12 or worse. Diagnosing diabetic maculopathy requires a 3-D camera using OCT technology to determine whether treatment or monitoring is required. On OCT imaging, a substantial number with suspected diabetic maculopathy do not have maculopathy and are either dis-charged back to DESW or remain under monitoring for years in HES without ever having treatment.
Many HES departments experience lack of specialists and available space to cope with the ever-increasing patient numbers, which has led to significant delays in treatment and in some cases irreversible SSI. Therefore, there is a need to consider new ways of conducting ophthalmology services and to allocate certain elements into the primary care sector. Examples already exist such as the national Eye Health Examinations Wales (EHEW) service, and the national low Vision Service Wales. Increasingly, Health Boards are developing locally enhanced primary eye care pathways in partnership with the HES, such as Ophthalmic Diagnostic and Treatment Centres for glaucoma and age-related macular degeneration. In some parts of England, OCT surveillance clinics have been incorporated into diabetic eye screening, requiring additional funding. Currently, due to the increasing workload and Covid-19 related disruption, DESW cannot take on this extra work.
The Covid-19 pandemic made pre-existing pressures on the HES more evident and whilst services are returning to normal, the backlog and pressure from the numbers of new patients make it extremely difficult for the already stretched HES and DESW to continue seeing patients and to catch up.
Pressures on the NHS will not diminish, even post pandemic. Irrespective of uncertainties relating to Covid-19, as well as the possibility of other pandemics in the future, investigating new ways of working are imperative to ensure irreversible sight impairment does not become an unintentional consequence of the intense pressure on the HES aside from the backlog from Covid-19. For efficient diabetic maculopathy management related clinical decisions, it is important to consider OCT central retinal thickness measurements and 3-D imaging of the macula as well as patient's visual acuity and comorbidities. The DESW pathway does not include this clinical information currently so patients with suspected diabetic maculopathy, based on surrogate markers (e.g., exudates near the fovea) must attend the HES to allow for further imaging and clinical examination to guide their management and treatment plan.
Thus, DESW referrals with suspect maculopathy are seen in the HES in either dedicated diabetes clinics or in general eye clinics. Hospital-based data collection of retinal images/OCT scans and subsequently virtual reviews have been implemented in the UK and around the world. During those hospital appointments, clinical examination, visual acuity measurement and OCT scans are undertaken with measurements of the central retinal thickness and based on clinical findings, only a small number of patients referred with suspect maculopathy require treatment at the time of referral. As a result, many patients are discharged to DESW or often monitored in the HES for many years without the clinical indication for any specific eye related treatment. This ongoing monitoring in secondary care increases further the number of necessary appointments for review whilst only a small number of patients are suitable for discharge, The study will investigate a proposed new care pathway, created in response to the Covid-19 pandemic and the consequent backlog of patients requiring referral to the HES for suspected diabetic maculopathy and pre-proliferative diabetic retinopathy which is potentially sight-threatening. Diagnosis of diabetic maculopathy requires examination by OCT to measure the thickness of the retina. Prior to Covid-19 referral to HES for OCT examination would take upwards of 6 weeks which is often an anxious time for people with diabetes. During Covid-19 screening was stopped and cases with suspected maculopathy were not reviewed at the HES. Therefore, a backlog has built up in ophthalmology and it is expected that there will be an increase in referrals from screening once the screening service (DESW) has returned to full capacity. Moreover, rules of social distancing to prevent Covid 19 transmission in secondary care outpatients' settings have further compromised the limited capacity of the screening service and HES.
The expected outcomes of the new pathway would be:
1.1. Research Question Is the proposed primary care based optometric pathway, utilising tele-ophthalmology, for people suspected to have diabetic maculopathy, safe, feasible, acceptable, environmentally beneficial and economically viable in comparison with the current pathway where all persons with diabetes suspected of having maculopathy and pre-proliferative diabetic retinopathy are referred directly to the HES?
2. Study Outline 2.1. Aims and Objectives Primary Aim: To pilot a primary care-based optometry pathway for the assessment of diabetic maculopathy and pre-proliferative diabetic retinopathy with virtual consultant ophthalmologist review in Hywel Dda University Health Board (HDdUHB).
Secondary Aims:
Objectives:
2.2. Method
2.2.1. Study Type This study is an unblinded randomised controlled trial to assess a new pathway for the management of diabetic maculopathy.
2.2.2. Study Overview
2.2.3. Study Populations, Subject Selection, Recruitment and Study Schedule The study will recruit 150 people with diabetes referred to HDdUHB ophthalmology for diabetic maculopathy (M1) or pre-proliferative diabetic retinopathy (R2). This sample size is considered achievable based on the current waiting list size within HDdUHB for this population. HDdUHB R&D delivery unit will consent participants into the study via telephone and will then randomise participants to either usual care or the community OCT pathway.
Figure 1 shows participant flow through the study.
Participants randomised to usual care will be booked an appointment by the pathway co-ordinator into the clinics currently run outpatient department (OPD) clinics. These patients will be reviewed through the standard processes for these clinics either face to face or virtually.
Participants randomised to the community optometry pathway will be booked an appointment with a participating optometry practice by the pathway co-ordinator. This will be the practice closest to home.
At the appointment an OCT image will be taken along with clinical examination and patient d-ta collected, as per case report form v5 18.11.2024. After the appointment all participants will be provided with a link to the questionnaire v7 04.10.2024 on JISC and asked to complete. Any person who indicates difficulty with completing online questionnaires will be contacted by the research nurse who will conduct the questionnaire by telephone.
OCT images, clinical examination, and patient data for participants in the usual care arm will be recorded on consultant connect and reviewed by the consultant ophthalmologist either in person or virtually according to current processes in OPD.
OCT images, clinical examination, and patient data for participants in the community OCT arm will be recorded on consultant connect and sent to the consultant ophthalmologist for review.
In both arms, the ophthalmologist will decide the ongoing management plan i.e. discharge, continued monitoring, treatment required. Results of the appointment and management plan will be provided to the patients via the post.
Failsafe will be built into both pathways with the pathway co-ordinator booking appointments and then confirming patients have attended those appointments. If the appointments are can-celled or patients are not able to attend the pathway co-ordinator will ensure appointments are rebooked at the earliest availability.
A stratified subset of participants from each arm who have indicated they are willing to participate in interviews, will be invited to attend semi structured interviews post appointment. The interview themes are outlined in the document interview themes v1 12.12.2022. The inter-views will be conducted via a virtual platform, telephone, or face to face. The interviewer will confirm consent before conducting the interview. Following the interview transcripts will be checked and anonymised. They will be uploaded to NVIVO for thematic analysis. Figure 4 shows the selection of participants in the interviews.
Figure 1: Participant flow through the study
2.2.5. Power Calculation The sample size will have more than 80% power to detect a treatment difference at a two-sided significance level of 0.05, if the true hazard ratio is not less than 1.75.
2.2.6. Statistical Analysis Plan
Data Collection:
Date of DESW referral, date of randomisation, date when patients seen, date when patients receive treatment, types of treatment received.
Data Analysis:
Kaplan-Meier plots will be constructed to examine time from DESW referral to treatment. The log rank test will be used to assess the statistical significance of observed differences in treatment event rates between the established and alternative pathways. Semi-parametric models which make no assumption about the statistical distribution of time to treatment and baseline hazard will be used to explore factors affecting time-to-treatment.
The health economic approach will utilise a cost consequence analysis (CCA). This investigates changes in will estimate healthcare resources and related costs. The outcomes for patients (consequences in cost consequence analysis (CCA) terms) will as described above, compared with 'usual care' in HES based care. Sensitivity analyses will test any assumptions used to facilitate the estimates of resource use and related costs and/or address any uncertainty in the de-sign & inputs. Scenario & deterministic sensitivity analyses will also be undertaken. Budget impact & threshold analyses will be undertaken to estimate the minimum number of practices & optometrists required to deliver the service effectively in HDdUHB. The perspective of the economic evaluation will be both the health board, the Welsh NHS & personal & social services, and the patient. The evaluation will present several time horizons, short and medium term to inform finance cycles in the NHS and from the patient perspective, long term, that is the duration of living with DED. Costs will be obtained from a combination of published sources & local record. A discount rate of 3.5% and/or 1.5% per annum will be applied depending on the outcome of the current deliberations by the National Institute for Health & Care Excellence. The data inputs to the CCA will be taken from the study data collection, literature review, sources such as the Unit Costs of Health & Social Care and NHS reference costs. Where data gaps exist from these sources, the investigators will undertake targeted literature reviews, other NHS Sources, e.g., data drawn direct from HES, optometric practices. Estimates will be made of the level at which the primary care-based optometrists need to 'sign up' to deliver the primary care-based service equitably
2.4. Timescale Month Description Pre-start Finalise the questionnaire 1-3 Sponsor approval 3-6 Ethics & R&D approvals
Study recruitment 6-12 Participant recruitment from DESW referrals 6-12 Consent from participants 6-12 Randomisation
Study delivery 6-17 Participants are examined in each arm 6-17 Participants complete the study questionnaire following examination 10-19 Qualitative interviews 8-19 Transcription
Study close 17-19 Study data extraction 17-19 Questionnaire data extraction 17-21 Study data analysis 17-21 Questionnaire data analysis 19-23 Qualitative interview data analysis Health economics 12-16 Economic modelling literature searches to inform design, methods, relevant data sources and related best practice 16-18 Initial model development 18-20 Questionnaire data analysis to obtain modelling inputs 20-21 Analyse model results run scenario/sensitivity analysis 21-22 Write up modelling work
Management Every 4 months Quarterly and final report with financial statement Every 4 months Quarterly steering committee meetings 3-6 Dissemination of study start 22-24 Dissemination of findings
2.5. Publication Policy
The investigators would expect to produce several publications in peer reviewed journals from this project:
Besides scientific publications, the investigators will also disseminate results through:
HDdUHB and Swansea University Medical School (SUMS) will continue to work closely with primary care to include optometry practice networks, Diabetes UK Cymru to ensure widespread dissemination the findings from the proposed study. Governments, commissioners, providers, and Public Health Wales will be directly targeted. Early in the project, the investigators will have in place a publication and dissemination plan and develop an authorship agreement. This research team in SUMS has a strong record of completing research in diabetes and diabetes eye health and disseminating it through appropriate academic publications, clinical, service, policy, and user-friendly routes to achieve maximum impact.
Following the principles of co-production, our public contributors will lead dissemination to the wider public. The investigators shall also encourage them to disseminate findings and present their experiences to the INVOLVE conference and HCRW Conference. This will increase the knowledge of findings and awareness of potential applicability and thus influence the development of community-based pathways for management of eye health conditions. All the research findings and recommendations will be written in lay language and made public on the diabetes research groups website http://www.diabeteswales.org.uk/en/ These would also be presented at diabetes patient group meetings.
3. Screening, Consent and Withdrawal 3.1. Screening Duties Participants' referral letters will be screened by the pathway co-ordinator who receives all referrals for diabetic eye disease (diabetic maculopathy and diabetic retinopathy), for inclusion in the study. If they are eligible, they will be sent the consent to be contacted for research, patient information sheet (PIS), and study informed consent form and a mock completed study informed consent form in the post.
3.2. Informed Consent Procedures Patients who return the consent to contact form indicating a willingness to be contacted will be telephoned by the research nurse. For those who consent the research nurse will go through the PIS, answer any questions, and ensure the patient understands the study procedures. The research nurses will record participant consent over the phone using an electronic form.
3.3. Study Withdrawal Participants can choose to withdraw from the study at any point. If a participant chooses to withdraw from the study, no further data will be collected but data already collected and analysed will be retained.
4. Risk, Ethical Considerations and Confidentiality 4.1. Risk and Benefits The investigators do not foresee any risks to the participants or staff in this study, although the investigators acknowledge the topic of diabetic retinopathy/maculopathy can be distressing for patients. The investigators estimate appointments to take approximately 30-40 minutes with 15 minutes spent completing the questionnaire. If participants are selected to participate in the interviews, then an addition 30 minutes will be required.
4.2. Ethical Considerations This study will require HRA/HCRW, NHS Research Ethics Committee and local NHS R&D department approval. HRA/HCRW and ethical approval will be sought during the set-up phase of the study, followed by NHS permissions in accordance with the study's timeline and registered with the R&D department, HDdUHB.
It is possible that some participants may find aspects of the questionnaire completion/interview process challenging as it may raise sensitive issues, but the researchers are experienced, and additional support will be offered and provided where needed.
All potential participants will be given an approved participant information sheet to aid full understanding of the study. They will have the opportunity to ask questions and be given time to decide to consent to take part or decline. There will be no impact on individuals' clinical care or other service provision if they decline participation. Participants will be able to withdraw from the study at any time.
4.3. Patient & Public Involvement The study team is committed to actively involving patients and public members in our research and includes a lay member of the Diabetes Research groups public reference panel, who was also a co-applicant on our successful funding application. They will be involved in project implementation, strategic and practical decision-making throughout the study and dissemination of findings.
Additionally, the investigators will invite a further two members of the public reference panel to join our stakeholder group (which will includes members of another HCRW funded study involving moving secondary eye care services into the community, representation of diabetes and sight charities such as Diabetes UK Cymru, sight Cymru and Royal National Institute for the Blind (RNIB), members of optometric advisory groups and Welsh Government advisors) so that views of people with diabetes are embedded throughout the study.
4.4. Confidentiality All participant information will be stored in locked filing cabinets in limited access areas. All reports/questionnaires/interview transcripts/administrative forms will be identified by a coded identification (ID) number only, to maintain participant confidentiality. All records containing personal identifiers (i.e., informed consent forms), will be stored separately from study records, identified by a code number. All local databases will be secured with password-protected ac-cess systems. Any listings (i.e., forms, lists) linking participant ID numbers to other identifying information will be stored in a separate, locked file in a limited access area. Participants' personal information is acquired during the referral process and much of this information contains private details over which people wish to maintain control (i.e., date of birth, gender, hospital number). Only anonymised electronic data will be shared outside HDdUHB.
This study will adhere to GCP guidelines, and HDdUHB policies. Anonymised data will be retained for 5 years after publication unless a longer period is required by approvers. Consent forms will be retained for a maximum of 10 years following study closure and archived in a lockable filing cabinet, in line with HDdUHB Standard Operating Procedures.
4.5. Premature Termination or Suspension of Study The study may be suspended or terminated due to reasonable cause by the decision of the Chief or Principal Investigator (CI/PI) or sponsor/funder due to insufficient adherence to Protocol requirements, data that are not sufficiently complete/available and determination of futility, safety, or ethical concerns. Written notification, documenting the reason for study suspension or termination, will be provided by the suspending, or terminating party. If the study is prematurely terminated or suspended, the CI/PI will promptly inform all parties involved (Sponsor/funder/REC/MHRA/participants/researchers) stating the reasons.
4.6. Future Use of Stored Specimens and Other Identifiable Data N/A 4.7. Record Retention and Archiving Following study closure, study files will be retained in line with HDdUHB Standard Operating Procedure RDSOP-13: Archiving Research Records.
4.8. Study Oversight The CI will be responsible for study oversight, including monitoring safety, ensuring that the study is conducted according to the Protocol and ensuring data integrity. The CI will review the data for safety concerns and data trends at regular intervals and will promptly report to the REC and the sponsor any unanticipated problem, Protocol deviation, breach of GCP or any other significant event that arises during the conduct of the study.
4.9. Study Reporting The CI is responsible for ensuring that all required progress and end of study reports are submitted to the funder, sponsor, and REC.
4.10. Safety Reporting Any adverse events or serious adverse events occurring will be recorded and reported to sponsor, and HDdUHB R&D Quality Assurance team in line with RDSOP-05: Reporting Research-Related Adverse Events/Serious Adverse Events, RDFC-01: Flow Chart for Reporting Research-Related AEs/SAEs and RDF-01: Research-Related AE/SAE Reporting Form.
Inclusion Criteria:
Exclusion Criteria:
abigail.taylor@wales.nhs.uk01437773823