From the period of April 2014–March 2015, the database showed 115,606 patients who were screened. 1880 patients aged 80 years and 1105 patients aged 85 years making a total of 2985 patients were included who met the criteria of the study being 80 and 85 years respectively at the time of the study. The average age of the first ever screen for both cohorts was 73 years. The demographics of the patients for the 80 cohort were 60% Caucasian, 21% Asian and 11% Black and remaining not known. 961 (51%) were male and 919 (49%) were female in this cohort. The demographics of the patients for the 85 cohort were 55% Caucasian, 7 % Asian, 5% Black and remaining not known. 531(48%) were male and 574 (52%) were females in this cohort.

Figure 1 shows detailed screening outcomes for both cohorts. The low rates of referable DR can be noted in both groups and unassessable images (inadequates) on standard photography were referred to slit lamp clinics. The unassessables due to dense cataract on slit lamp were referred to cataract clinics in HES and no referable DR was picked up in this group from the available follow up data in HES.

Bar charts showing data for 80 -year cohort (A) and 85- year cohort (B).

Table 1 shows the data for rates of referable retinopathy, referral rates to eye clinics and treatment rates at baseline screening and subsequent years. The 80-year cohort showed referable DR rates between 3.2 and 3.9% but actual referral to HES ranged between 0.7 and 1.4% with a low treatment rate of 0.6%. The same pattern was reflected in the 85-year cohort where referable DR rates ranged between 2 and 3.9%. The actual referral to HES was noted to be 0.1% to 1.3% here with an even lower treatment rate of 0.4%. Patients with referable DR not referred to HES were the ones with suspect maculopathy who were seen in the Digital surveillance (DS) clinics. These patients were monitored in DS clinics in 6 months if stable, back to annual recall if improved and referred to HES if worsened. Having OCT in our DS clinics helped to identify true maculopathy and diabetic macular oedema for referral to HES. False positives based on vision were sent back to annual recall and minimal dry maculopathy monitored in DS until they resolved or referred to HES if worsened.

As depicted by Fig. 2, in the 80-year cohort, a total of 76 (4%) patients were referred to HES over the five years and 11 (0.6) patients received treatment with macular focal laser and injections for maculopathy. R2 referrals were all stable or improved and none progressed to R3. The patients who were referred for R3 were all found to be either previously treated stable R3 or false positives who were downgraded to R2/ R1 in HES. None of these patients required any treatment and were monitored in HES or discharged back to the screening programme. In the 85-year cohort, a total of 27 (2.4%) patients were referred to HES over the five years and 4 (0.4%) patients received treatment with macular focal laser and injections for maculopathy. All R2 referrals were deemed stable or improved to R1 when seen in HES. Of the four R3 referrals, three were stable treated ones and one was a false positive who was downgraded and all were discharged back to the screening programme. It was only maculopathy that required treatment in both these cohorts and nobody in this age group worsened to proliferative retinopathy over the follow up period and none required pan retinal photocoagulation.

(A) showing data for the 80-year cohort and (B) depicting the 85-year cohort.

Figure 3 shows the non-DR referrals to HES over the 5 years in both these cohorts. In the 80-year cohort, 396 (21%) patients were referred to HES during the follow up period for other eye conditions and similarly in the 85-year cohort, a total of 140 (13%) patients were referred for other eye conditions. The detailed break up for this can be noted in the Fig. 3 which shows the various non-DR eye conditions for which these patients were referred to HES over the follow up period. Unassessable images due to cataract accounted for the highest proportion of referrals in both the cohorts.

(A) Showing data for the 80-year cohort and (B) depicting the 85-year cohort.

It was also noted that, 403 (21%) patients in the 80-year cohort and 541 (49%) patients in the 85-year cohort died over the course of follow up. This was reflected in the progressive reduction in the total numbers screened in each subsequent year of follow up for both cohorts and can be noted in Fig. 1.