6533b82dfe1ef96bd12910ce

RESEARCH PRODUCT

Use of hospital and amubutory data in assessment of screnning, follow-up and complications of diabetes mellitus

subject

description

Diabetes is a priority health target worldwide. Access to conclusive data is equally a priority in order to establish clear objectives and to implement coordinated action. Medico-administrative databases are being used more and more frequently for this purpose. We set out to illustrate and assess the interest of using a French healthcare consumption database to evaluate screening, surveillance and the risk of acute or chronic complications of diabetes in certain at-risk populations. We concluded that early screening for type-2 diabetes in women with a history of gestational diabetes was inadequate. Although a slight improvement occurred following the recommendations published in 2010, one woman in two has no screening test in the first year following the pregnancy.Recommended screening for type-2 diabetes following myocardial infarction in the general population is seriously insufficient. At the time of cardiac rehabilitation, 97% of patients were screening for type-2 diabetes screening and 40% of these are diagnosed with diabetes or prediabetes. After stroke, surveillance of diabetes remains poor. In the least severe cases of stroke, it seems that the control of risk factors for recurrence is an integral part of the management of the cerebrovascular disease. We showed that women with gestational diabetes have a risk of early cardiovascular complications and that young persons with type-1 diabetes and schizophrenia have a risk of rehospitalisation for acute complications. Prevention policies for diabetes could be improved if they were based on opportunities to create strong partnerships involving actors from different sectors. The use of medico-administrative databases and observational studies has extended our knowledge of diabetes. However, before being used, data quality and the pertinence of selection algorithms must be evaluated, and it must be shown that the methods used for the statistical analysis are appropriate.