Preventing the onset of acute myocardial infarction (AMI) and its recurrence, and reducing the morbidity and mortality associated with AMI, remain of significant public health and clinical concern. Monitoring contemporary trends in AMI incidence, treatment, and in-hospital and long-term outcomes is of considerable importance given periodic national updates of treatment guidelines, emphasis on reducing hospital readmissions, and revised definitions and classifications of AMI. Continuously supported by the NHLBI, we have conducted more than 35 years of population-based surveillance of AMI incidence and attack rates, hospital management practices, and the in-hospital and long-term prognosis associated with AMI among residents of central MA hospitalized at all central MA medical centers. We have a highly experienced team of cardiologists, epidemiologists, clinical informatics, and health services researchers who will build on multi- decade long trends (1975-2011) in our principal study endpoints examined previously in this study to the two new study years of patients hospitalized with AMI at all central MA medical centers in 2014 and 2017. To sustain our efforts into the era of electronic medical records (EMRs), and after implementation of the ICD-10 system in 2015, we will develop a new automated AMI surveillance system that efficiently utilizes EMRs by taking advantage of state-of-art natural language processing (NLP) methods that will be compatible with ICD-10 (Aim 1). We will use the new NLP method to streamline traditional chart review-based collection of socio-demographic, clinical, treatment, and hospital and post-discharge outcomes data in patients hospitalized with AMI at all 11 central MA medical centers in 2014 and 2017. The data extracted from NLP-streamlined chart reviews will be used to validate and refine the NLP system. Issues related to changes from ICD-9 to ICD- 10 will be carefully addressed. The new NLP-enriched EMR-based surveillance system will eventually be implemented in all participating central MA hospitals. Using the NLP-enriched and EMR-based surveillance data, we will monitor the contemporary clinical epidemiology of AMI, and out-of-hospital deaths due to coronary disease, and changing landscape, over a more than 40 year period (1975-2017) (Aim 2). The new EMR-based and NLP-enriched system will enhance the population-based surveillance of acute coronary disease. This new system will be cost-effective, more efficient and near-real time, have greater accuracy and precision, and can be readily updated to accommodate changes in information technologies and broadly applicable to other hospital systems. It will support our continued efforts to provide unique community- based observational data on several populations that are often excluded from clinical trials, and that are increasing in numbers, namely the elderly and patients with multiple morbidities. Furthermore, it will generate critical data to inform more national clinical guidelines on the enhanced prevention and management of AMI. If successful, the system can serve as a model and be implemented statewide in MA and elsewhere in the US.
The results of the proposed community-based study will provide data about 40 year trends with regards to the changing magnitude of, and outcomes associated with, heart attacks in residents of a large central New England community. The results of this investigation will also provide contemporary insights on how patients who experience heart attacks in the community are treated by physicians.
Showing the most recent 10 out of 12 publications
