By searching thousands of electronic medical records, a team of NIA-funded researchers may have uncovered key differences in pathology and clinical features between men and women with Alzheimer’s disease, as well as between people with and without this disease. The study, published in Nature Communicationdemonstrates the potential of a “big data” approach involving advanced computer algorithms.
Electronic medical records are generated during visits to healthcare professionals and contain clinical data about patients, including details such as age, medical history, medications, and laboratory and diagnostic test results . Before these records are viewed by researchers, all patient data is anonymized – a process that removes any information that could identify a patient. Studying Alzheimer’s disease using data from these records can provide insight into the complexity of the disease and identify sex-specific associations and differences in diagnoses, medications, and test results. laboratory. The research team – from the University of California, San Francisco (UCSF); Icahn School of Medicine in Mount Sinai, New York; and the University of Minnesota School of Medicine, Minneapolis, used a big data approach known as deep integrative phenotyping to help map and analyze the electronic medical records of 44,288 people.
Alzheimer’s disease is a complex disease that can be linked to several traits, risk factors and other disorders. However, assessing the validity of these connections has been difficult. For example, although it is known that twice as many women develop Alzheimer’s disease as men, attempts to determine exactly why have produced mixed results. In this study, the researchers tapped into the wealth of information provided by electronic medical records. This allowed them to search for connections in a systematic and scientifically unbiased way.
To do this, the researchers first fed an advanced data-mapping computer algorithm with all of the diagnostic information from the UCSF electronic medical record dataset. The algorithm then displayed the patient data as clusters on a graph with similar patient data coming close together. Remarkably, data from people with Alzheimer’s disease showed a different clustering pattern than people without the disease. Similar results were seen when they used electronic medical record data from Mount Sinai. The results supported and validated the idea that this big data approach could be used to visualize and detect links between diseases.
Next, the researchers used a different algorithm to look for links between the co-occurrence of different medical conditions, prescribed medications, and lab tests. Using this approach, they found several differences in data from people with Alzheimer’s disease. For example, consistent with previous studies, people with Alzheimer’s disease were more likely than others to be diagnosed with hypertension, diabetes mellitus, anemia, vascular disease, osteoporosis, or urinary tract infections. The researchers also found differences between men and women with Alzheimer’s disease. Namely, men seemed more likely to be diagnosed with neurological, sensory and behavioral disorders. In contrast, women had more frequent diagnoses of arthritis, bone fractures, atrial fibrillation and accidents.
Overall, the findings support the idea that studying electronic medical records can not only help scientists assess patterns and validity of Alzheimer’s associations, but also uncover new ones. Ultimately, this may help researchers get a clearer picture of the forces behind this very complex disease.
This research was supported in part by NIA grants R01AG060393, R01AG057683, RF1AG068325, and RF1AG059319.
These activities relate to the NIH Alzheimer’s Disease and Related Dementias Research Implementation Milestones:
- 1.F, “Support the inclusion of measures of AD-related phenotypes and environmental exposures in non-AD cohorts to enable new discovery research and accelerate cross-validation of findings made in AD cohorts.”
- 1.H, “Enable access to, and support for, electronic health record (EHR) data integration with clinical and molecular data to create person-specific predictive models of disease and wellness and enable sub-classification of diseases.These efforts should include better electronic phenotyping of AD through the application of machine learning methods.
Reference: Tang AS, et al. Extensive Alzheimer’s disease phenotyping using electronic medical records identifies sex-specific clinical associations. Communication Nature. 2022;13(1):675. two: 10.1038/s41467-022-28273-0.