Evidence reviews for the effectiveness of different monitoring frequencies
Evidence review J
NICE Guideline, No. 242
Evidence review for effectiveness of different monitoring frequencies
1.1. Review question
What is the effectiveness of different monitoring frequencies for people with non-proliferative diabetic retinopathy whose care is managed under the hospital eye services but who are not having treatment?
What is the effectiveness of different monitoring frequencies for people with proliferative diabetic retinopathy or diabetic macular oedema who are receiving treatment or have had previous treatment?
1.1.1. Introduction
Diabetic retinopathy is a significant cause of vision loss in adults. The risk of the development and progression of non-proliferative retinopathy to macular oedema or vision-threatening proliferative diabetic retinopathy requires timely intervention to improve patient outcomes and reduce the risk of loss of vision.
Early detection of disease progression can play a significant role in timely treatment. Current recommendations in the Royal College of Ophthalmologists guidelines (2012) include 4-6 monthly monitoring for people with moderately severe to very severe non-proliferative retinopathy. The aim of this review was to establish the risks and benefits of different monitoring frequencies to effectively detect potentially vision-threatening changes in:
- People with moderate, severe, and very severe non-proliferative diabetic retinopathy without macular oedema, whose care is managed under hospital eye services.
- People with proliferative diabetic retinopathy or diabetic macular oedema who are receiving treatment or have had previous treatment.
The protocols for the evidence reviews are summarised in Table 1. Please see full protocols in Appendix A.
1.1.2. Summary of the protocol
Table 1
PICO for people with non-proliferative diabetic retinopathy.
Table 2
PICO for people with proliferative diabetic retinopathy or diabetic macular oedema.
1.1.3. Methods and process
This evidence review was developed using the methods and process described in Developing NICE guidelines: the manual and the methods document for the diabetic retinopathy guideline.
Declarations of interest were recorded according to NICE’s conflicts of interest policy. Methods specific to this review question are described in the review protocol in Appendix A. Additionally:
- A modified version of the NICE economics studies checklist was used to critically appraise modelling studies. Items 1.4, 1.6, 1.7, 2.6, 2.7, 2.8 and 2.9 were removed as they relate to economic aspects which are not applicable to this review.
- A modified GRADE approach was used to assess the certainty in the evidence from modelling studies. The approach to GRADE for assessing the effectiveness of interventions outlined in the methods document for the diabetic retinopathy guideline was used, with the exception that evidence from modelling studies was started with a GRADE rating of ‘high’.
1.1.4. Effectiveness evidence
1.1.4.1. Included studies
A single systematic literature search was conducted to cover both review questions. The search included randomised controlled trials (RCTs), comparative observational studies and modelling studies comparing monitoring frequencies. No date limit was applied, and the search yielded 2,686 references. These were screened on title and abstract, with 38 full-text papers ordered as potentially relevant studies.
Studies were excluded if they did not match the protocol outlined in Appendix A.
A single paper (DCCT/EDIC Research, 2017) was included after full text screening for the review question on monitoring frequencies for non-proliferative diabetic retinopathy. No studies were included for the review question on monitoring frequencies for proliferative diabetic retinopathy or diabetic macular oedema.
For the study selection process, please see PRISMA flow diagram in Appendix C
For the full evidence tables and full GRADE profiles for included studies, please see Appendix D and Appendix E.
1.1.4.2. Excluded studies
See Appendix I for a list of excluded studies with reasons for exclusion.
1.1.5. Summary of studies included in the effectiveness evidence
Table 3
Studies included in the effectiveness evidence.
See Appendix D for full evidence tables.
1.1.6. Summary of the Effectiveness evidence
Probability over a given follow-Up interval of progression from lower levels of retinopathy to higher grade of retinopathy
Table 4
Modelled risk of progression from moderate diabetic retinopathy to proliferative retinopathy or clinically significant macular oedema.
Table 5
Modelled risk of progression from severe diabetic retinopathy to proliferative retinopathy or clinically significant macular oedema.
See Appendix E for full GRADE tables.
1.1.7. Economic evidence
1.1.7.1. Included studies
A single search was performed to identify published economic evaluations of relevance to any of the questions in this guideline update (see Appendix B). This search retrieved 672 studies. Based on title and abstract screening, 670 of the studies could confidently be excluded for these questions. Two studies were excluded following the full-text review. No relevant health economic studies were included.
1.1.7.2. Excluded studies
See Appendix I for excluded studies and reasons for exclusion.
See the health economic study selection flow chart presented in Appendix F.
1.1.8. Summary of included economic evidence
No relevant health economic studies were identified to be included.
1.1.9. Economic model
Original health economic modelling was not prioritised for this review question.
1.1.10. Unit costs
Table 6
Unit cost of screening visits.
1.1.11. The committee’s discussion and interpretation of the evidence
1.1.11.1. The outcomes that matter most
The committee agreed that progression to proliferative diabetic retinopathy was an important outcome in people diagnosed with non-proliferative diabetic retinopathy because this can have very serious consequences, including retinal detachment and irreversible and severe vision loss, if not treated. Other outcomes were also considered to be important (progression of diabetic macular oedema, best corrected visual acuity, peripheral vision, adherence, and health related quality of life) but no evidence was found for these outcomes. The committee wanted the data to be separated by subgroups including pregnancy, age, and severity of disease, however the evidence available did not allow for stratification by subgroups.
1.1.11.2. The quality of the evidence
Only one study was identified for people with non-proliferative retinopathy that matched the review protocol. The quality of the evidence for the outcomes was low, with the main reasons for downgrading being a lack of information on whether the models adjusted for confounders and the data not allowing for stratification by risk factors.
The data reported combined progression of proliferative diabetic retinopathy and diabetic macular oedema as one outcome and there was no information on the relative proportions of people with the two outcomes. The committee considered this a major limitation because delaying treatment for proliferative diabetic retinopathy has more serious consequences than delaying treatment for diabetic macular oedema. Delaying treatment for proliferative diabetic retinopathy can result in retinal detachment and irreversible sight loss. Delaying treatment for diabetic macular oedema could also result in sight loss, but this is likely to be less severe and reversible in comparison to sight loss secondary to proliferative diabetic retinopathy.
The committee were also concerned that the lack of clarity on whether the model was stratified by those who received intensive glycaemic control intervention and those that received no treatment.
The evidence was downgraded for indirectness as the population in the study was limited to people with type 1 diabetes. However, the committee agreed that they did not expect a large difference in outcomes between people with type 1 and type 2 diabetes.
No evidence was identified on the effectiveness of different monitoring frequencies for people who are receiving treatment or who have previously received treatment for proliferative diabetic retinopathy or diabetic macular oedema.
1.1.11.3. Imprecision and the clinical importance of effects
The evidence identified was modelled based on a large sample, and so the 95% confidence intervals were narrow enough to allow useful comparisons between monitoring frequencies. As noted in the section above, the committee found it difficult to determine what percentage of progression between monitoring visits would be acceptable because of the composite nature of the outcome and the different clinical consequences of progression to proliferative diabetic retinopathy and diabetic macular oedema.
1.1.11.4. Benefits and harms
The committee discussed that monitoring is needed to check for disease progression that requires treatment, so that treatment can begin promptly if progression occurs. They also noted that people with diabetic retinopathy and diabetic macular oedema often attend a large number of hospital appointments to manage their diabetes care. They often have other diabetes-related complications that also require hospital visits, and so it is important to make sure that monitoring is not more frequent than necessary to reduce this burden.
People with non-proliferative diabetic retinopathy, proliferative diabetic retinopathy or diabetic macular oedema
The committee did not review any evidence that allowed them to clearly differentiate evidence for people under 18 or pregnant people. However, they agreed that the same recommendations should apply to under 18s as to adults. Although the risk of developing diabetic retinopathy is lower in under 18s, if it is identified, it should be monitored in the same way. The committee was aware of existing recommendations on monitoring diabetic retinopathy and the timing of retinal assessments in pregnancy in NICE’s guideline on diabetes in pregnancy, so they agreed to refer to this guideline.
Non-proliferative retinopathy
Based on their expertise and the modelling evidence, the committee made different recommendations, depending on severity of disease and risk of progression. The committee made a weaker ‘consider’ recommendation based on the limitations in the evidence that was identified in the quality of the evidence section. However, it should be noted that the choice of consider rather than offer in this recommendation is in relation to the frequency of monitoring, rather than the need for any monitoring at all. It was decided that the recommendations should be separated by people who have moderate non-proliferative retinopathy and those with severe to very severe non-proliferative diabetic retinopathy. People who have moderate non-proliferative diabetic retinopathy can be seen less frequently, with lower risk of progression between appointments, while those who have severe or very severe non-proliferative diabetic retinopathy will need more frequent appointments.
The committee discussed monitoring every 6–12 months for people with moderate non-proliferative retinopathy who are not being currently treated or have not been previously treated. It was noted that people under hospital eye services who are not receiving treatment occupy a lot of clinic time. It was agreed that progression of disease in this population is relatively slow and the evidence indicates that a 6–12-month window means that people have between 6.6% (6.0%-7.3%) and 12.3% (11.3%-13.5%) chance of progressing to proliferative retinopathy or clinically significant macular oedema between appointments. The committee thought that 6–12 months between appointments is therefore appropriate and should not allow for any major progression of the disease between appointments.
The committee agreed that people with severe or very severe non-proliferative diabetic retinopathy who are not being currently treated should be seen more frequently, as they are more at risk of progression than those who have moderate non-proliferative retinopathy. They noted that this group have a 14.4% (9.4%-22.0%) chance of progression if they have monitoring appointments every 3 months. It was highlighted that more frequent appointments would further reduce the risk of progression. For instance, appointments every 2 months would mean that someone only had a 10.4% (6.5%-16.0%) chance of progression. However, the committee were concerned that it may not be practical to see all of these patients more frequently than every 3 months. They also discussed how people with diabetic retinopathy have to attend a number of different appointments including these monitoring appointments for their eye disease as well as appointments for other complications associated with their diabetes. As such, very frequent appointments might be unmanageable for some people. Three months was therefore considered an appropriate follow-up time. It was also highlighted that while 3 months is ideal, some people will still not be able to attend appointments this frequently and might instead be at risk of missing appointments, which would have a greater impact on progression than less frequent monitoring. The committee therefore decided that monitoring should take place between 3 and 6 months for this group, giving a maximum risk of 23% (15.8%-32.7%) chance of progression between appointments. This time scale also reflects current practice.
Due to the limited evidence to inform recommendations on the timing of monitoring, the committee also made a research recommendation on the most effective monitoring frequencies for people with non-proliferative retinopathy who have not started treatment. The ideal study design to inform this research would be a randomised controlled trial comparing outcomes in people monitored at different frequencies. However, such a study would be difficult to carry out because it would need long follow up times and people may need to be allocated to follow up intervals that are longer than current practice. Modelling studies that report data on progression of diabetic retinopathy and diabetic macular oedema would therefore be a feasible alternative. For details of the research recommendation see Appendix J.
Proliferative retinopathy and diabetic macular oedema
No evidence was found for people who have proliferative diabetic retinopathy or diabetic macular oedema. The committee noted that monitoring during treatment would be determined by the treatment protocol and so did not make recommendations for this area. However, the committee agreed that some guidance would be useful for monitoring frequency after treatment is completed. This was based on their clinical experience that an appropriate monitoring time can vary between individuals depending on their risk factors for progression, and the need to ensure that appointments are not so frequent that there is the risk of non-attendance. After 12 months, they thought the risk of progression was lower and therefore this is an appropriate time to consider discharge back to hospital services.
The committee agreed that people who have received treatment for proliferative diabetic retinopathy or diabetic macular oedema, whose disease has regressed should be monitored under the care of hospital eye services for 12 months. They discussed how the frequency of monitoring during this time should be individualised depending on the treatment that had been given and on a person’s response to treatment. However, they agreed that some guidance on monitoring frequency after treatment completion is required to improve consistency across the country. Therefore, they agreed that disease regression in people who have received treatment for proliferative diabetic retinopathy should be assessed at 2 to 3 months after treatment has ended. This should be an appropriate time so that any progression following the end of treatment can be identified before it leads to more serious consequences. The committee also agreed that after 12 months, people who have had treatment for proliferative diabetic retinopathy or diabetic macular oedema whose disease has resolved can be discharged back to the diabetic retinopathy annual screening programme. Those that have features that would prompt immediate re-referral to hospital eye services should remain under the care of hospital eye services for monitoring. Based on their clinical knowledge and experience the committee decided that monitoring every 12 months would be appropriate in this case. For those who are eligible to be discharged to the screening programme, the committee highlighted that they should be encouraged to attend both their eye screening appointments and regular appointments with primary care optometrists. This should help identify if further treatment is needed in the future and identify other eye disease that is not covered by the eye screening programme.
Given the lack of evidence, the committee made a second research recommendation for people with proliferative diabetic retinopathy or diabetic macular oedema who have previously received treatment (see Appendix J). This should enable future guidelines to make more precise recommendations on the most effective monitoring frequencies.
1.1.11.5. Cost effectiveness and resource use
No relevant economic evaluations were identified which addressed the cost effectiveness of different monitoring frequencies for people with a diagnosis of non-proliferative diabetic retinopathy, proliferative diabetic retinopathy, or diabetic macula oedema. The committee noted that patients with a non-proliferative diabetic retinopathy diagnosis make up a large proportion of those seen within clinic, which led to the research recommendation being made for this population. The committee agreed that the recommendations made to monitor disease progression would not be expected to have a resource impact as they reflect current practice.
The committee discussed assessing disease regression in people who have received treatment for proliferative retinopathy or diabetic macular oedema would happen within a monitoring visit 2–3 months after treatment has ended. The committee discussed this currently happens within clinical practice and would not expect a resource impact other than improving consistency of practice across the country.
The committee discussed people with proliferative diabetic retinopathy or diabetic macula oedema whose disease has regressed or resolved after treatment and noted that they should still be monitored for twelve months to ensure any progression of disease is captured early. The committee also agreed that after 12 months people should be discharged back to the diabetic screening programme, however if the persons retina has features that make them ineligible for the screening programme, they should continue to be monitored under the care of hospital eye services every 12 months. This is not expected to be a change in practice, and if implemented consistently may lead to a reduction in monitoring visits as only those whose disease has not improved would continue to be monitored after 12 months. Additionally, any early signs of disease progression would likely be detected and lead to prompt treatment rather than more intensive treatment later when the persons disease has progressed further.
1.1.11.6. Other factors the committee took into account
The committee did not review any evidence that allowed them to clearly differentiate and stratify evidence for people 18 and under or pregnant women. However, the committee agreed that the same recommendations should apply to people under 18 as, although risk of developing diabetic retinopathy is lower in this group, if it is identified it should be monitored in the same way. The committee noted that there are existing recommendations on monitoring diabetic retinopathy in pregnancy in the NICE’s guideline on diabetes in pregnancy, so they agreed to refer to this guideline.
The committee were also aware of the recommendations in the Royal College of Ophthalmologists guidelines (2012) for 4-6 monthly monitoring for people with moderately severe to very severe non-proliferative retinopathy. The NICE recommendations are broadly in line with those recommendations, although they acknowledge that some people with severe to very severe non-proliferative diabetic retinopathy could benefit from monitoring as often as every 3 months. The committee also thought it was important to highlight that some of the people who are within the recommendation from the Royal College of Ophthalmologists, but who have the least severe disease, may not need to be seen as frequently as every 6 months. For this reason, they thought the additional recommendation for monitoring every 6-12 months for people who have moderate non-proliferative diabetic retinopathy was important.
1.1.12. Recommendations supported by this evidence review
This evidence review supports Recommendations 1.4.1 to 1.4.2, 1.5.11 to 1.5.15 and 1.6.13 to 1.6.14 and 1.6.16 and the research recommendations on monitoring frequencies for people with non-proliferative diabetic retinopathy who are not receiving treatment, and for people with proliferative diabetic retinopathy or diabetic macular oedema who have received treatment.
1.1.13. References – included studies
- DCCT/EDIC Research, Group, Nathan, David M, Bebu, Ionut et al (2017) Frequency of Evidence-Based Screening for Retinopathy in Type 1 Diabetes. The New England journal of medicine 376(16): 1507–1516 [PMC free article: PMC5557280] [PubMed: 28423305]
1.1.13.1. Clinical evidence
Appendices
Appendix A. Review protocols
Appendix B. Literature search strategies
Search design and peer review
NICE information specialists conducted the literature searches for the evidence review. The searches were run in May 2022. This search report is compliant with the requirements of PRISMA-S.
The MEDLINE strategy below was quality assured (QA) by a trained NICE information specialist. All translated search strategies were peer reviewed to ensure their accuracy. Both procedures were adapted from the 2016 PRESS Checklist.
The principal search strategy was developed in MEDLINE (Ovid interface) and adapted, as appropriate, for use in the other sources listed in the protocol, taking into account their size, search functionality and subject coverage.
Review Management
The search results were managed in EPPI-Reviewer v5. Duplicates were removed in EPPI-R5 using a two-step process. First, automated deduplication is performed using a high-value algorithm. Second, manual deduplication is used to assess ‘low-probability’ matches. All decisions made for the review can be accessed via the deduplication history.
Limits and restrictions
English language limits were applied in adherence to standard NICE practice and the review protocol.
Limits to exclude, conference abstract or conference paper or "conference review" were applied in adherence to standard NICE practice and the review protocol. The limit to remove animal studies in the searches was the standard NICE practice, which has been adapted from: Dickersin, K., Scherer, R., & Lefebvre, C. (1994). Systematic Reviews: Identifying relevant studies for systematic reviews. BMJ, 309(6964), 1286 [PMC free article: PMC2541778] [PubMed: 7718048].
Search filters
Download PDF (193K)
Limits and restrictions
English language limits were applied in adherence to standard NICE practice and the review protocol.
Limits to exclude, comment or letter or editorial or historical articles or conference abstract or conference paper or "conference review" or letter or case report were applied in adherence to standard NICE practice and the review protocol.
The limit to remove animal studies in the searches was the standard NICE practice, which has been adapted from: Dickersin, K., Scherer, R., & Lefebvre, C. (1994). Systematic Reviews: Identifying relevant studies for systematic reviews. BMJ, 309(6964), 1286 [PMC free article: PMC2541778] [PubMed: 7718048].
Search filters
Download PDF (193K)
Appendix C. Effectiveness evidence study selection
Download PDF (103K)
Appendix D. Effectiveness evidence
Download PDF (162K)
Appendix E. GRADE tables
Appendix F. Economic evidence study selection
Download PDF (101K)
Appendix G. Economic evidence tables
There are no included studies in this review question.
Appendix H. Health economic model
Original health economic modelling was not prioritised for this review question.
Appendix I. Excluded studies
Clinical studies
Table
Study does not contain a relevant intervention. - Does not contain a population of people with DR/DMO
Economic evidence
Appendix J. Research Recommendation
J.1.1. Research recommendation
What is the most effective monitoring frequency for non-proliferative diabetic retinopathy in people who are cared for under hospital eye services and are not receiving treatment?
J.1.2. Why this is important
While there are general guidelines for monitoring individuals with this condition, specific evidence-based recommendations are needed for people with non-proliferative diabetic retinopathy who are under the care of hospital eye services and not receiving treatment. Non-proliferative diabetic retinopathy can progress to a more severe stage, such as proliferative diabetic retinopathy, which may require treatment. Regular monitoring can help detect early signs of disease progression and enable timely intervention, reducing the risk of vision loss. Research is therefore needed to ensure that people are monitored at the most effective frequency to ensure they can receive timely treatment.
J.1.3. Rationale for research recommendation
Download PDF (129K)
J.1.4. Modified PICO table
Download PDF (159K)
J.1.5. Research recommendation
What is the most effective monitoring frequency for proliferative diabetic retinopathy or diabetic macular oedema in people who have received treatment?
J.1.6. Why this is important
The monitoring frequency for individuals with proliferative diabetic retinopathy or diabetic macular oedema who have received treatment is an important consideration for managing their condition effectively. The most effective monitoring frequency depends on various factors, including the severity of the condition, the type of treatment received, the stability of the patienťs visual function, and the presence of any additional risk factors. Frequent monitoring is essential to detect any disease progression or recurrence early and initiate timely interventions. Research is therefore needed to ensure that people are monitored at the most effective frequency to ensure they can receive timely treatment.
J.1.7. Rationale for research recommendation
Download PDF (149K)
J.1.8. Modified PICO table
Download PDF (155K)
Final
Evidence reviews underpinning recommendations 1.4.1, 1.4.2, 1.5.11 to 1.5.15, and 1.6.12 and 1.6.13, and research recommendations 12 and 13 in the NICE guideline
These evidence reviews were developed by NICE
Disclaimer: The recommendations in this guideline represent the view of NICE, arrived at after careful consideration of the evidence available. When exercising their judgement, professionals are expected to take this guideline fully into account, alongside the individual needs, preferences and values of their patients or service users. The recommendations in this guideline are not mandatory and the guideline does not override the responsibility of healthcare professionals to make decisions appropriate to the circumstances of the individual patient, in consultation with the patient and/or their carer or guardian.
Local commissioners and/or providers have a responsibility to enable the guideline to be applied when individual health professionals and their patients or service users wish to use it. They should do so in the context of local and national priorities for funding and developing services, and in light of their duties to have due regard to the need to eliminate unlawful discrimination, to advance equality of opportunity and to reduce health inequalities. Nothing in this guideline should be interpreted in a way that would be inconsistent with compliance with those duties.
NICE guidelines cover health and care in England. Decisions on how they apply in other UK countries are made by ministers in the Welsh Government, Scottish Government, and Northern Ireland Executive. All NICE guidance is subject to regular review and may be updated or withdrawn.