Individual-specific assessment of the risk of visual impairment in diabetes
The overall aim is to verify the outcome of a calculation algorithm (RetinaRisk ™), developed in Iceland and tested in several European countries, on a very large Swedish material, available in The Swedish National Diabetes Register (NDR).
Diabetes is globally a growing public health problem. The number of patients will increase by 42% in industrialized countries between 1995 and 2025. Approx. 400 000 persons in Sweden have diabetes and the health care costs is approx. 21 000 SEK per person year. On average, diabetics have 9 more days of sick leave per year than non-diabetics. In addition to direct costs, indirect costs, such as lost productivity due to sickness, absences, disability, premature retirement etc. also forms a burden to society. Retinopathy (DRP) is the most common complication in diabetes, 75% of all who have had diabetes for more than 20 years have some degree of DRP, a large proportion of them sight-threatening. Impaired vision due to DRP is a major cause of reduced work capacity in working-age subjects. Blindness due to diabetes can be dramatically reduced to about 0.5% in the diabetic population by systematic screening using retinal photography, and preventive laser treatment. Such screening is cost-effective in terms of sight years preserved compared with no screening and many countries, including Sweden, have large-scale programs for early detection and treatment of DRP. The screening intervals are based on statistical means and the recommendations are the same for large groups of patients, regardless of their individual characteristics. Published studies have shown that person-specific medical data as an assessment basis leads to extended screening intervals for many patients. It frees up resources for patients at highest risk, while low-risk patients save time and money.
The aim of the study is to verify the outcome of a calculation algorithm (RetinaRisk ™), developed in Iceland and tested in several European countries, on a very large Swedish material, available in the National Diabetes register (NDR). The very large amount of data in NDR is also expected to add new knowledge about the relationship between the algorithmic risk assessment and the actual outcome for individual patients.
Research questions: How is the distribution of the calculated risk of developing vision-threatening DRP among diabetic patients in Sweden? How does the level of risk vary over time? How well does RetinaRisk's prognosis correspond to the actual onset of vision-threatening retinopathy.? Can RetinaRisk's algorithm forecasts be improved?
Knowledge gains and significance: Experience from international studies indicates that the new screening method means direct savings in screening operations without increasing the risk of visual impairment in patients. At the same time, the analysis of the very large amount of data in NDR can provide new knowledge, which can lead to further improvement of the underlying algorithm. The introduction of the algorithms in the form of computer support, useful at the point of care, can mean practical improvements for patients, reduced costs for healthcare and, in the long run, better quality of life for diabetic patients.