Print large text version Print this page

Screening people with diabetes

Contents

Summary

A hypothetical intervention of an annual, mobile retinal screening service for people aged 10 years and older with registered diabetes in the ROI would result in:

  • a total screening cost of €1.6 million;
  • 1,090 cases of mild VI being detected through screening tests, including 449 cases of DR;
  • discounted five-year health care costs of €2.1 million to treat detected cases;
  • deadweight welfare losses of €1.8 million from government funding of the intervention, eye tests and health care costs; and
  • 600 DALYs being averted as a result of the intervention.

7.2.1 Description

Under a societal perspective, this intervention would be associated with a cost of €9,090 per DALY averted under a societal perspective and €6,031 under a health care perspective. Therefore, this intervention is highly cost effective using WHO thresholds.

People with diabetes mellitus are at risk of developing a number of complications, including DR which can potentially result in blindness. Timely and appropriate care for people with diabetes can significantly reduce vision loss over time, improve quality of life and reduce financial costs associated with VI. Screening people with diabetes, followed by treatment of identified DR has been found to be effective in previous studies (HSE and Irish College of Ophthalmologists, 2008).

There is currently no standardised nationally-based screening program in Ireland, with screening determined by various factors including local policy, historical activity and funding issues (Dervan et al, 2008). This may lead to many patients not receiving any screening.

Borrowing elements from a national framework for DR screening designed by the HSE and Irish College of Ophthalmologists (2008) and a previous CEA for DR screening in Australia (Access Economics, 2005), the hypothetical intervention evaluated in this section would target people in the ROI with diagnosed (registered) diabetes aged 10 years and older and deliver free eye tests via an annual, mobile retinal screening service. Although the HSE and Irish College of Ophthalmologists (2008) propose screening for people aged 12 years and above, a minimum age of 10 years has been used to fit with reported population data by age group. Under this hypothetical intervention, all people with previously diagnosed diabetes aged 10 years and older would be offered screening each year.

The screening method would be digital retinal photography, which provides reproducible and quality assured results. This method has been identified as being the optimal method of performing retinopathy screening (HSE and Irish College of Ophthalmologists, 2008).

Back to top

7.2.2 Target population and reach

Prevalence rates of type 1 and type 2 diabetes in the ROI have been estimated by the Institute of Public Health in Ireland (2006) estimated and are presented in Table 7.6. Prevalence rates for type 2 diabetes were only reported by the Institute for people aged 20 years and older. These prevalence rates were applied to the ROI population in 2010 (CSO, 2008) to estimate 161,438 people with diagnosed or undiagnosed diabetes in that year aged 10 years and over.

There are few studies in the ROI on the percentage of diabetes cases that are diagnosed (Institute of Public Health in Ireland, 2006). A study by Smith et al (2003) examined 41 general practices in the ROI and concluded that 23.5% of diabetes were undiagnosed prior to the study. In the absence of other data, this percentage was applied to diabetes prevalence estimates for each age group to estimate the prevalence of diagnosed diabetes as 123,738 people aged 10 years and over.

Prevalence rates and estimated people with diabetes are presented in Table 7.6.

Please refer to the audio files to listen to this table or to the PDF to view the table.

An Irish study (Dervan et al, 2008) estimated that 80.9% of people with diabetes had an annual eye examination to test for DR. The inverse of this percentage (19.1% having no annual eye examination) was applied to the total number of people with diagnosed diabetes aged 10 years and over to estimate a target population for screening of 23,636 people.

As part of their framework for a national DR screening program, the HSE and Irish College of Ophthalmologists (2008) specify a minimum performance standard whereby 70% of eligible people will take up an initial eye screen. This percentage was applied to the target population to estimate 16,545 screening tests resulting from the intervention.

Back to top

7.2.3 Cases of undetected mild VI diagnosed

Fundus photography involves the use of a retinal camera to photograph regions of the vitreous, retina, choroid and optic nerve. Digital retinal photography can document abnormalities related to disease processes affecting the eye, including DR, macular degeneration, glaucoma, abnormal retinal function and defects (Aetna, 2010) and cataract (Mann et al, 2008). Thus, the DR screening intervention is assumed to be able to also pick up other eye conditions.
As explained in Section 7.1, VI detected by screening is assumed to be undiagnosed mild VI. Prevalence rates of mild VI by condition and age (Section 2.4) were applied to the additional number of screening eye tests in people with diabetes, to estimate the number of diagnosed and undiagnosed mild VI cases. However, DR, by definition, is only prevalent in people with diabetes. Therefore, prevalence rates for DR were re-estimated by dividing the prevalence of DR in each age/gender group by the number of people with diabetes in each age/gender group. For example, the prevalence of mild DR was estimated to be 3.3% among males with diabetes aged 60-69 years compared with 0.3% among males aged 60-69 years in the general population.

The proportion of mild VI that is undiagnosed was presented in Section 7.1.3 and is applied in this CEA also.

Estimates of diabetic eye tests and previously undetected mild VI diagnosed through screening are presented by condition and age in Table 7.7. Please refer to the audio files to listen to this table or to the PDF to view the table. Overall, it is estimated that 1,090 cases of mild VI would be diagnosed through additional eye screening tests in people with diabetes, including 449 cases of DR, based on the population size in 2010.

Back to top

7.2.4 Cost of screening

HSE Primary Care Services Dublin North East area provides a mobile retinal screening service to patients in the Diabetes Watch Programme (HSE and Irish College of Ophthalmologists, 2008). This involves screening by digital retinal photography and three stage grading with internal and external quality assurance. Three people are involved in the screening program including an ophthalmologist, photographer and one primary grader/administrator (optometrist). The cost of this screening service was €90 per person screened in 2008.

Since this is very similar to the hypothetical screening intervention, its cost was used in the CEA. The cost per person screened is €94 in 2010 prices, applying annual changes in the health consumer price index (CPI) (CSO, 2011) .
This cost was applied to the additional number of screening tests resulting from the intervention (16,545) to estimate a total screening cost of just over €1.5 million.

7.2.5 Treatment effectiveness and compliance

Estimated parameters for treatment effectiveness and compliance were presented in Section 7.1.5 and applied in this CEA also.

7.2.6 Treatment costs of diagnosed cases

Future treatment costs of VI cases diagnosed through screening were estimated using the same approach as for the older person screening intervention (see Section 7.1.6). It was assumed that each person with diabetes who has not received an annual eye test but has undetected mild VI would seek an eye test and treatment within the next five years of their own accord as their vision loss deteriorates. Please refer to the audio files to listen to this table or to the “PDF”:/files/Cost_of_Sight_Loss_Full_Report_July_2011.pdf to view the table. Overall, the total five-year health care cost of treating mild VI diagnosed through the intervention is estimated to be nearly €2.1 million.

7.2.7 DWL

Following the methodology presented in Section 7.1.7 (MCPF 0.57 cents, 80.7% of health care costs funded by government), the DWL from treatment costs following detection of vision loss in people with diabetes was estimated to be nearly €1.8 million.

7.2.8 DALYs averted from successfully treated cases

Following the methodology presented in Section 7.1.8, it was estimated that 600 DALYs would be averted as a result of the intervention.

7.2.9 Cost effectiveness results

Societal costs of the intervention (including DWL) are estimated to total €5.5 million (societal perspective). Excluding DWL, total costs of the intervention would be €3.6 million (health care perspective). The total costs associated with the diabetes population screening intervention described above are summarised in Table 7.9.

Table 7.9: Total costs associated with intervention

Eye screening tests: €1,550,460
Discounted treatment costs: €2,065,898
DWL: €1,834,055
Total costs (health care perspective): €3,616,358
Total costs (societal perspective): €5,450,412

Source: Deloitte Access Economics estimates.

Under a societal perspective, the hypothetical screening program for people with diabetes is associated with a cost of €9,090 per DALY averted. Under a health care perspective, cost effectiveness is estimated to be €6,031 per DALY averted. The intervention is highly cost effective under both perspectives, using WHO cost effectiveness thresholds.

Back to top