Introduction
Retinopathy is a frequent complication of diabetes mellitus especially when blood sugar is poorly controlled. The retina is the part of the eye that transforms light into signals transmitted to the brain through the optic nerve (NHS UK 2013). The brain interprets the electrical signals into images corresponding to what was seen. While sufficient blood supply is needed for the retina to function optimally, chronic hyperglycaemia causes ischemia and bleeding in the retinal microvasculature leading to the formation of new capillaries meant to sustain blood circulation in the retina (Vithian & Hurel 2010). However, the new capillaries are fragile and prone to rupture and leakage. The loss of blood supply is what causes damage to the retina manifesting as gradual and irreversible vision loss (NHS UK 2013).
Diabetic retinopathy is the leading cause of vision loss among diabetic adults as the likelihood of going blind is 10-20 times higher in this population (Diabetes UK 2012). Studies show that within 20 years following the diagnosis of diabetes, almost all patients with type 1 disease and nearly 60% of those with type 2 disease already suffer from some degree of diabetic retinopathy (Diabetes UK 2012). In addition, persons with diabetes are twice more prone to cataracts and glaucoma than persons without the disease. Similarly, persons with diabetes in Oman are 20 times more likely to develop visual impairment with around 9% diagnosed with glaucoma (Khandekar, Mohammed & Al-Lawati 2008). The incidence rate of diabetic retinopathy was 2.7% in 2005 while recent data on prevalence shows the complication affects 14% of diabetic Omanis (Al-Shookri et al. 2011). Hence, diabetic retinopathy is a chief public health concern in the country.
Objective, Criteria, and Standards of the Audit
Diabetic retinopathy and other eye diseases can be averted through secondary prevention. The Ministry of Health (MOH) (2003) in Oman issued guidelines stating that persons with type 2 diabetes should have yearly eye screening following a baseline screening around the time of diagnosis. Hence, the objective of this audit is to determine the level of adherence with the MOH guideline. The standard is to achieve a minimum of 95% screening attendance to optimise uptake of the service (MOH 2003). This audit will also review and critically appraise the evidence supporting the current guidelines and to generate recommendations for quality improvement.
This audit is limited to the aspect of process pertaining to annual eye screening rates for a diabetic clinic. According to the MOH (2003), the criteria is for all eligible adults with diabetes to receive eye screening services at or around the time the patient is diagnosed and for the same procedure to be performed at least yearly. A tool has been created to ascertain whether screening was done or not done (see Appendix 1). The audit will also determine the reasons for defaulting or the patient’s failure to show up for the procedure.
Data Collection and Audit Sample
The purpose of the audit was explained to the clinic charge nurse. As patient records will be accessed for the date of last eye screening, permission was obtained from the health centre authority. The confidentiality of medical information and privacy of the patients’ identities was guaranteed in accordance with the General Medical Council (2009) guidelines. Patients were identified by numbers instead of their names. The sample included 15 patients with diabetes (type 1 and type 2) due for their annual eye screening within the period of two consecutive days. Those who failed to attend their scheduled screening were followed up via telephone to ascertain why the appointment was missed. Data AnalysisThe rate of screening uptake or actual attendance refers to the number of patients in the sample who came in for their scheduled eye screening divided by the total sample size (NHS 2014). The results are shown in Appendix 2. Of the total sample of 15 patients due for their annual eye screening, 11 had the procedure done. Hence, the rate of eye screening uptake was 73.33% which is way below the minimum standard of 95% for actual patient participation. Four patients or 26.67% failed to have their screening and represents deviation from national protocol. Follow-up telephone calls revealed that of this number, two forgot their appointments while one stated not having been given an appointment for screening by the doctor. One patient had an appointment and knew about it but refused screening services. He reports not having problems with his vision for the past three years and does not see the usefulness of continuing with screening. The above reasons for failure to be screened represent important barriers that must be addressed if the rate of uptake is to be improved significantly.
DiscussionSecondary prevention can prevent or delay the development of diabetic retinopathy as demonstrated by several landmark studies. One of them is the 1995 Diabetes Control and Complications randomised controlled trial that employed a sample size of 1,441 patients aged 13-39 years all of whom had type 1 diabetes (Garg & Davis 2009). All patients either had no diabetic retinopathy or in the early stage of the complication. The sample was randomised to intensive blood sugar control with an average haemoglobin A1C of 7.2% or to conventional blood sugar control with an average HbA1C of 9.1% (Garg & Davis 2009). The results showed that intensive blood sugar control resulted in a 54-percent reduction in the risk of retinopathy progression as determined by annual eye screening. The study also showed a 47-percent decline in the development of severe diabetic retinopathy as well as a 23-percent decline in the risk of macular oedema related to diabetes (Garg & Davis 2009).
As such, diabetic retinal screening serves as an indicator of the effectiveness of interventions for blood sugar management (Stratton et al. 2013). At the same time, eye screening is a method of monitoring the progression of diabetic retinopathy that enables the provision of timely and appropriate treatment for the reduction of the risk of continued vision loss and eventual blindness. In the U.K., the Royal College of Ophthalmologists (RCO) guidelines support an eye screening frequency of yearly among adults and more frequent screenings for those already with retinopathy (Ghanchi et al. 2013). This is a Level A recommendation which means that there is universal agreement on the strength of the evidence supporting it. In a modelling study, annual screening was shown to detect 1-6% of retinopathy and 6-9% of retinopathy progressing to blindness (Bachmann & Nelson 1998). Both conditions require treatment.
Further, Vijan et al. (1997) also found that although patients with advanced retinopathy benefit from laser treatment, only 35-60% received it because of the lack of yearly screening. In another study, having individualised eye screening frequencies were found to be confusing for patients with a negative effect on compliance while at the same time posing difficulties in clinical implementation (Wilson, Baker & Thompson 1995). As such, annual screening as the universal frequency was found the most efficient for a population-based prevention programme such as that implemented by the National Health Service UK.
However, Scotland recently revised its eye screening guidelines through the initiative of the Scottish Intercollegiate Guidelines Network (SIGN) (see Appendix 3). Compared with the RCO guideline, the SIGN guideline now recommends a two-year interval in eye screening for persons with diabetes who did not show evidence of retinopathy at baseline while patients with mild or background retinopathy are still recommended to have their eye screening annually and those with advanced disease recommended for more frequent screening (SIGN 2010; Looker et al. 2013). The evidence supporting this guideline was regarded as Level B which means that the likelihood of benefit outweighed the harm although there is no universal agreement among all stakeholders.
The guideline revision was based in part on an evaluation of the Scotland Diabetic Retinopathy Screening programme which showed that those without referable retinopathy at baseline showed a low rate of progression to retinopathy at 0.2-0.3% within two years so that if they were screened biennially rather than annually, the programme would have screened 40% less individuals (Looker et al. 2013). Two other studies in the UK support this evaluation study. Cohort studies using large samples of diabetic adults from the Central Norfolk Diabetic Retinopathy Screening Service (Jones et al. 2012) and the Gloucestershire Diabetic Eye Screening Service (Stratton et al. 2013) also show that the progression to referable retinopathy was low among those with no retinopathy at initial screening and suggest the transition to a screening frequency of more than a year. In the present audit, the lone patient who refused screening because of negative results for the past 3 years and had no retinopathy at baseline validates the findings of the above studies.
In the United States, the American Diabetes Association (ADA) guideline (see Appendix 4) similarly supports a two-year interval of screening if one or more screening results showed no sign of retinopathy (ADA 2014). Among its cited evidence is a systematic review. To determine which screening interval is the most effective in reducing the incidence of vision loss, Echouffo-Tcheugui et al. (2013) performed a systematic review of 15 program evaluations, 7 cost-effectiveness studies, and 3 models of the natural course of diabetic retinopathy. In general, there is support for an interval of more than one year based on programme evaluations and modelling studies, but results are mixed in cost-effectiveness studies.
Models show the typical length of time and conditions that lead to the development of diabetic retinopathy as well as its progression through different stages (Echouffo-Tcheugui et al. 2013). Diabetic retinopathy is categorised as proliferative or non-proliferative. The former is evaluated according to severity – early, high-risk, florid, and gliotic – as well as location which can be on the disk or in other areas of the retina (The Royal College of Ophthalmologists 2013). Another criterion categorizes diabetic retinopathy as background, pre-proliferative, and proliferative. Meanwhile, non-proliferative retinopathy is graded as mild, moderate, or severe. One modelling study showed that there was no large difference in the net individual benefit or years of sight saved when annual screening or screening every two years was used at 40% and 37%, respectively (Echouffo-Tcheugui et al. 2013). The same is true in another study that measured the reduction in blindness wherein it was 54% with annual screening and 51% for biennial screening. A recent modelling study that was not included in the current review further validated the safety and cost-effectiveness of biennial screening among those without retinopathy at baseline (Chalk et al. 2012).
On the other hand, programme evaluations provide information on the incidence of diabetic retinopathy and blindness associated with the screening services provided. In the studies reviewed, all of which used the cohort design, the interval for eye screening ranged from 1 to 4 years (Echouffo-Tcheugui et al. 2013). One study supported the need for more frequent eye screening in patients already with diabetic retinopathy, i.e. annual eye screening for those with background retinopathy and every 0.3 of a year for mild proliferative retinopathy. An interval of more than one year was found to be safe and appropriate in patients with no evidence of diabetic retinopathy at baseline based on 12 of the 15 studies (Echouffo-Tcheugui et al. 2013). The probable rationale was that it would take more than 2 or 3 years for these patients to develop retinopathy limiting the usefulness of annual screening. However, many of the aforementioned programme evaluation studies were small and conducted in the hospital setting limiting the generalisability to primary care settings. More research is needed to firmly establish the above findings that would warrant a change in current protocols.
Finally, cost-effectiveness studies show mixed results. There is support for biennial screening among diabetic adults found to have no or mild retinopathy. The reason is that biennial screening had no significant impact on years of sight saved compared with annual screening and shifting to this screening frequency was associated with yearly savings of £164.13 per person (Echouffo-Tcheugui et al. 2013). However, there are studies in the review supporting annual screening as the optimum interval, but recommend biennial screening when there are resource constraints as the latter is associated with lower costs. Other studies point out that there are conditions influencing the effectiveness and cost-effectiveness of screening such as the sensitivity of screening methods, level of patient compliance, time elapsed since diagnosis, age of onset, and the level of glycaemic control (Jones & Edwards 2010; Scanlon, Aldington & Stratton 2013). As such, there are also suggestions for individualised screening intervals, risk-based screening, or targeted screening with a higher frequency recommended for patients with lower glycaemic control and lower compliance and who have had the disease for a long time.
The Oman Ministry of Health guideline is similar to the UK Royal College of Ophthalmologists guideline as both advocate annual screening regardless if the patient was not found to have evidence of retinopathy. Thus, the MOH guideline differs from that of the ADA and SIGN which recommend biennial screening for those without retinopathy at baseline. A comprehensive approach to diabetic retinopathy was adopted in 1998 and was followed by the launching of the Eye Health Care Programme with the guideline in place since 2003 (Khandekar, Mohammed & Al-Lawati 2008). Suboptimal patient compliance with diabetes management and complication prevention strategies was a driving factor that influenced the adoption of annual eye screening for all diabetics (MOH 2003). To determine the most appropriate eye screening frequency in light of the current debate, it would be useful for the MOH to conduct an evaluation study of the national eye care programme similar to what Scotland’s SIGN conducted and to consider current evidence in the literature.
Finally, patient compliance with eye screening recommendations is an important issue in regards to diabetic retinopathy prevention. Studies show a compliance rate ranging from 21-28% (Echouffo-Tcheugui et al. 2013) making the compliance rate in the present audit high at 73.33%. Nevertheless, it is important to determine and address the barriers to compliance in order to achieve the minimum standard of 95%. A qualitative study on the incentives and barriers to eye screening attendance showed that fear of blindness and awareness of the negative impact of retinopathy on visual acuity were incentives while lack of recommendation by the doctor is a major barrier (van Eijk et al. 2012). In the current audit, failure to recommend was a barrier that led to one defaulter. Forgetting appointments was also a barrier noted in studies (van Eijk et al. 2012) as it was in two patients in the audit. Another barrier was the annual frequency of screening despite the consistent lack of evidence of retinopathy. Addressing this barrier would require reviewing current evidence supporting biennial screening and determining whether this is safe for the patient.
Moreover, a study involving South Asians showed that access in relation to the appointment system, clinic location, clinic hours, multiple appointments for comorbid conditions, and transportation were barriers (John, Cooper & Serrant-Green 2014). The literature also identifies inadequate awareness of the importance of eye screening in relation to diabetic retinopathy, vision loss, and blindness as well as how the procedure is done were other barriers identified (Khandekar et al. 2010). Although inadequate access and awareness are barriers among many patients, these were not noted among the defaulters in the audit. However, they still require consideration to ensure that this does not prevent future service users from receiving eye screening.
Recommendations and the Nursing Role
Based on the results of the audit and evidence from the literature, a more effective appointment system is needed. The use of checklists can help ensure that appropriate recommendations including eye screening are communicated to the patient and that appointments are given (MacDonald et al. 2013). In addition, a system of reminding patients for upcoming screening appointments must also be in place. Text messaging, e-mail, phone calls, or physical mail may be employed depending on what patients find most effective (Kannist, Koivunen & Vallimaki 2014; Austin & Wolfe, 2011). A computerised system that displays appointments for each day will make patient follow-ups for screening attendance more efficient (Anabtawi & Mathew 2013). In Oman, moreover, a register of all diabetics who received eye screening at baseline was created. To combat defaulting or patients not coming for their appointments, those who did not show up were followed up to ascertain the reason and to reschedule the patient if necessary (Khandekar, Al-Lawati & Barakat 2011). The system must be maintained to improve compliance. Addressing the barriers identified makes a 95% screening uptake goal by the next audit a reasonable target. At the same time, the evidence generated from recent studies should be considered to determine the continuing effectiveness and cost-effectiveness of guidelines (Stevens 2013).
Nurses have an important role to play in improving the eye screening attendance rate. By exercising leadership, they can initiate structure and process improvements (Smolowitz et al. 2014) as mentioned above to address the barriers to attendance. Nurses also play a role in educating patients about the need for and value of screening even if no evidence of diabetic retinopathy is found (Gosden et al. 2009; Watkinson & Chetram, 2005). Moreover, nurses can engage in research to generate more evidence as to the optimal interval for eye screening especially for patients who have had negative results since baseline. Research can also be done to determine other barriers to screening participation so that approaches can be more comprehensive (Gosden et al. 2009).
Finally, an audit serves as a strategy for quality improvement and should be done periodically to identify areas needing further improvement, formulate action plans, and to ascertain the effectiveness of strategies employed to bring about improvement (Bowie, Bradley & Rushmer 2010). The results of the audit must be communicated to the physician, physician assistant, nurses, and other stakeholders to justify the need for change. A re-audit is planned after six months following the implementation of improvements. Dissemination of the process and results of quality improvement efforts through the audit will be done by writing a scholarly article and submitting to nursing journals and poster presentations during conferences (Stevens 2013).
Conclusion
Diabetic retinopathy is a common but preventable complication of diabetes. Current protocols include annual eye screening as a strategy of determining the presence of the complication and monitoring its progression. Current national standards recommend that all eligible patients receive retinopathy screening each year with a minimum target of 95%. An audit was done to ascertain clinic adherence to standards and protocols. At 73.3%, the audit shows that the diabetic clinic did not meet the target for attendance. There were barriers to screening uptake based on follow-ups of the patients who failed to turn up for their appointments. Part of quality improvement is to address these barriers so that the target of 95% within 6 months can be achieved. With regards to the protocol, evidence supports the usefulness of eye screening among adult patients. However, there is emerging evidence suggesting the comparable safety and greater cost-effectiveness of biennial screening although more research is needed before definitive recommendations can be made. Audits should be placed in the context of quality improvement. As such, the changes needed to address present barriers must be implemented and a re-audit planned to evaluate the success of improvement. Nurses play a role in leading change by educating patients and addressing barriers to screening attendance as well as generating evidence to support current and alternative practices. Part of improving practices in primary care is the dissemination of the clinic’s experiences which can be used in similar settings to foster quality improvement.
References
Al-Shookri, A. Khor, G.L. Chan, Y.M. Loke, S.C. Al-Maskari, M. 2011. Type 2 diabetes in the Sultanate of Oman. Malaysian Journal of Nutrition. 17(1). pp.129-41.
American Diabetes Association. 2014. Standards of medical care in diabetes – 2014. Diabetes Care. 37(Suppl. 1). pp.S14-80.
Anabtawi, A. Mathew, L.M. 2013. Improving compliance with screening of diabetic patients for microalbuminuria in primary care practice. ISRN Endocrinology. 2013(893913). pp. 1-3.
Austin, S. Wolfe, B.L. 2011. The effect of patient reminders and gas station gift cards on patient adherence to testing guidelines for diabetes. Wisconsin Medical Journal. 110(3). pp. 132-7.
Bachmann, M.O. Nelson, S.J. 1998. Impact of diabetic retinopathy screening on a British district population: Case detection and blindness prevention in an evidence-based model. Journal of Epidemiology and Community Health. 52(1). pp. 45-52.
Bowie, R. Bradley, N.A., Rushmer, R. 2010. Clinical audit and quality improvement – time for a rethink? Journal of Evaluation in Clinical Practice. 18(1). pp. 42-8.
Chalk, D. Pitt, M. Vaidya, B. Stein, K. 2012. Can the retinal screening interval be safely increased to 2 years for type 2 diabetic patients without retinopathy? Diabetes Care. 35(8). pp. 1663-8.
Diabetes UK. 2012. Diabetes in the UK 2012: Key statistics on diabetes (online). Available at http://www.diabetes.org.uk/Documents/Reports/Diabetes-in-the-UK-2012.pdf. (Accessed: 16 December 2014).
Echouffo-Tcheugui, J.B. Ali, M.K. Roglic, G. Hayward, R.A. Narayan, K.M. 2013. Screening intervals for diabetic retinopathy and incidence of visual loss: A systematic review. Diabetic Medicine. 30. pp. 1272-92.
Garg, S. Davis, R.M. 2009. Diabetic retinopathy screening update. Clinical Diabetes. 27(4). pp. 140-5.
General Medical Council. 2009. Confidentiality (online). Available at http://www.gmc-uk.org/guidance/ethical_guidance/confidentiality.asp. (Accessed: 16 December 2014).
Ghanchi, F. Bailey, C. Chakravarthy, U. Cohen, S. Dodson, P. Gibson, J. Menon, G. Muqit, M. Pilling, R. Olson, J. Prasad, S. Scanlon, P. Stanga, P. Vafidis, G. Wright, A. Wykes, W. 2013. The Royal College of Ophthalmologists’ clinical guidelines for diabetic retinopathy: A summary. Eye. 27. pp.285-7.
Gosden, C. James, J. Winocour, P. Turner, B. Walton, C. Nagi, D. Williams, R. Holt, R. 2009. Leading the way: The changing role of the diabetes. Journal of Diabetes Nursing. 13(9). pp. 330-7.
John, A. Cooper, J. Serrant-Green, L. 2014. Barriers to diabetic retinopathy screening in South Asian groups. Primary Health Care. 24(8). pp. 25-30.
Jones, S. Edwards, R.T. 2010. Diabetic retinopathy screening: A systematic review of the economic evidence. Diabetic Medicine. 27. pp. 249-56.
Jones, C.D. Greenwood, R.H. Misra, A. Bachmann, M.O. 2012. Incidence and progression of diabetic retinopathy during 17 years of a population-based screening programme in England. Diabetes Care. 35(3). pp. 592-6.
Kannisto, K.A. Koivunen, M.H. Vallimaki, M.A. 2014. Use of mobile phone text message reminders in health care services: A narrative literature review. Journal of Medical Internet Research. 16(10). pp. e222-9.
Khandekar, R. Mohammed, A.J. Al-Lawati, J.A. 2008. Care of diabetic retinopathy patients in Oman. Sultan Qaboos University Medical Journal. 8(1). pp. 5-9.
Khandekar, R. Al Harby, S. Al Harthy, H. Al Lawatti, J. 2010. Knowledge, attitude and practice regarding eye complications and care among Omani persons with diabetes – A cross sectional study. Oman Journal of Ophthalmology. 3(2). pp. 60-5.
Khandekar, R. Al-Lawati, J.A. Barakat, N. 2011. A retrieval system for patients with avoidable blindness due to diabetic retinopathy who do not present for ophthalmic assessment in Oman. Middle East African Journal of Ophthalmologists. 18(2). pp. 93-7.
Looker, H.C. Nyangoma, D.T. Cromie, J.A. Olson, G.P. Leese, S. Philip, M.W. Black, J. Doig, N. Lee, A. Briggs, E.J. Hothersall, A.D. Morris, R.S. Lindsay, J.A. McKnight, D.W. Pearson, N.A. Satter, S.H. Wild, P. McKeigue, H.M. Colhoun, H.M. 2013. Predicted impact of extending the screening interval for diabetic retinopathy: The Scottish Diabetic Retinopathy Screening Programme. Diabetologia. 56(8). pp. 1716-25.
MacDonald, L. Stubbe, M. Tester, R. Vernall, S. Dowell, R. Dew, K. Kenealy, T. Sheridan, N. Docherty, B. Gray, L. Raphael, D. 2013. Nurse-patient communication in primary care diabetes management: An exploratory study. BMC Nursing. 12(20). pp. 1-10.
Ministry of Health, Sultanate of Oman. 2003. Diabetes mellitus management guidelines for primary health care. 2nd ed. Muscat, Oman.
National Health Service UK. 2103. Diabetic retinopathy (online). Available at http://www.nhs.uk/Conditions/Diabetic-retinopathy/Pages/Introduction.aspx. (Accessed: 16 December 2014).
National Health Service UK. 2014. Interim quality standards and performance objectives for diabetic eye screening programmes pending a full standards review (online). Available at http://diabeticeye.screening.nhs.uk/standards. (Accessed: 16 December 2014).
National Institute for Health and Care Excellence. 2011. Diabetes in adults quality standard (online). Available at https://www.nice.org.uk/guidance/qs6/chapter/quality-statement-8-complications. (Accessed: 16 December 2014).
Royal College of Ophthalmologists. 2013. Diabetic retinopathy guidelines 2012 (minor update July 2013) (online). Available at http://www.rcophth.ac.uk/page.asp?section=451. (Accessed: 16 December 2014).
Scanlon, P.H. Aldinton, S.J. Stratton, M. 2013. Delay in diabetic retinopathy screening increases the rate of detection of referable diabetic retinopathy. Diabetic Medicine. 31. pp. 439-42.
Smolowitz, J. Speakman, E. Wojnar, D. Whelan, E.M. Ulrich, S. Hayes, C. Wood, L. 2014. Role of the registered nurse in primary health care: Meeting health care needs in the 21st century. Nursing Outlook. In press.
Stevens, K.R. 2013. The impact of evidence-based practice in nursing and the next big ideas. Online Journal of Issues in Nursing. 18.
Stratton, I.M. Aldington, S.J. Taylor, D.J. Adler, A.I. Scanlon, P.H. 2013. A simple risk stratification for time to development of sight-threatening diabetic retinopathy. Diabetes Care. 36. Pp. 580-5.
van Eijk, K.N. Blom, J.W. Gussekloo, J. Polak, B.C. Groenveld, Y. 2012. Diabetic retinopathy screening in patients with diabetes mellitus in primary care: Incentives and barriers to screening attendance. Diabetes Research and Clinical Practice. 96(1). pp. 10-6.
Vijan, S. Stevens, D.L. Herman, W.H. Funnell, M.M. Standford, C.J. 1997. Screening, prevention, counselling, and treatment for the complications of type II diabetes mellitus. Journal of General and Internal Medicine. 12. pp. 567-80.
Vithian, K. Hurel, S. 2010. Microvascular complications: Pathophysiology and management. Clinical Medicine. 10(5). pp. 505-9.
Watkinson, S. Chetram, N. 2005. A nurse-led approach to diabetic retinal screening. Nursing Times. 101(36). pp. 32-4.
Wilson, A. Baker, R. Thompson, J. 1995. Screening for diabetic retinopathy: Services are still patchy in Britain. British Medical Journal. 311(7014). pp.1229-1231.
Appendix 1
Eye Screening Attendance Audit Tool
Appendix 2
Eye Screening Attendance Audit Results
Appendix 3
Scottish Intercollegiate Guidelines Network (SIGN) Guideline on Eye Screening
Reference:
Looker, H.C. Nyangoma, D.T. Cromie, J.A. Olson, G.P. Leese, S. Philip, M.W. Black, J. Doig, N. Lee, A. Briggs, E.J. Hothersall, A.D. Morris, R.S. Lindsay, J.A. McKnight, D.W. Pearson, N.A. Satter, S.H. Wild, P. McKeigue, H.M. Colhoun, H.M. 2013. Predicted impact of extending the screening interval for diabetic retinopathy: The Scottish Diabetic Retinopathy Screening Programme. Diabetologia. 56(8). pp. 1716-25.
Appendix 4
American Diabetes Association (ADA) Guideline on Eye Screening
Reference:
American Diabetes Association. 2014. Standards of medical care in diabetes – 2014. Diabetes Care. 37(Suppl. 1). pp.S14-80.
