Alzheimer's disease (AD) is a devastating disease which in most cases does not have an effective cure, and therefore it is important to find ways to prevent it. Persons with type 2 diabetes are more prone to develop cognitive impairment and AD in the future, but while treatment for diabetes is available and is effective in preventing many complications, it is not clear whether it can also reduce the risk of AD. In many cases, glucose metabolism is already impaired years before diabetes is diagnosed, however it is not known whether this impairment is important in predicting an individual's risk for AD. In addition, there are many drug classes which can be used to treat diabetes, and it is not known which classes are best for AD prevention. Lastly, the effect of the treatments can be different in some persons, based on their weight, blood pressure, genetics, and other characteristics. In the proposed research we will seek to answer these questions, by combining data from 6 large studies. The large sample, with a total of 37,000 individuals who were followed-up for cognitive outcomes and 6,350 people with diabetes, will allow us to explore the association of impairment in glucose metabolism with the future risk of AD, and also with performance on cognitive tests and with measures of brain injury on MRI examination. Based on the study's findings, clinicians will be able to better assess their patients' risk to develop AD, and will consider diabetes treatment choices also based on their potential to reduce the risk for AD. Researchers will be able to utilize the findings from this study to improve the design of future studies and the selection of participants.