Description
This week’s assignment is designed to help you examine the potential of using “big data” analytics for community/population health. Big data analytics can be used to address a number of clinical problems as well as yield significant cost savings. Furthermore, using clinical data as it is collected may be more cost-effective than collecting new data specifically for a research study.
This group assignment involves developing a narrated PowerPoint presentation describing an aspect of a “big data” application to address a community/population health problem. Community health can be applied to a physician practice, or a specific group, region, or you may use the term more broadly to apply to a population. Big data applications can use data mining, natural language processing, text mining, machine learning, algorithms and other tools to identify patients who meet certain criteria, or data trending to a specific outcome. The goal for the assignment is for you to understand the specific methodologies used for “big data” analysis and examine the potential of big data in improving population health by exploring the feasibility, challenges and/or issues surrounding its application to another setting or scaling up for widespread use.
Guidelines for the Assignment
Select an example of a methodology that is used for a “big data” study in health care. Describe the methodology, the data requirements and the software that is necessary to conduct the analysis. What makes the study a “big data” study: volume, velocity, variety, (variability, veracity, value)? Provide an example of how the method was applied to a clinical, financial and/or administrative problem in health care. What were some of the challenges in conducting the study in your example?
Cite sources used for this assignment in APA format on the last slide.
Resources for Completing this Assignment
Search Term Help: Possible PubMed search terms include: data mining, text mining, natural language processing, machine learning, combined with electronic health record, surveillance, neural network, Bayesian classifier and/or a specific disease or clinical problem.
Here are some examples of “big data” research (available on ERES):
Chase, H. S., Mitrani, L. R., Lu, G. G., & Fulgieri, D. J., (2017). Early recognition of multiple sclerosis using natural language processing of the electronic health record. BMC Medical Informatics and Decision Making, 17(1), 24. https://doi.org/10.1186/s12911-017-0418-4
Gibbons, C., Richards, S., Valderas, J. M., & Campbell, J. (2017). Supervised machine learning algorithms can classify open-text feedback of doctor performance with human-level accuracy. Journal of Medical Internet Research, 19(3), e65. https://doi.org/10.2196/jmir.6533
Oliveira, A., Faria, B. M., Gaio, A. R., & Reis, L. P. (2017). Data mining in HIV-AIDS surveillance system: Application to Portuguese data. Journal of Medical Systems, 41(4), 51. https://doi.org/10.1007/s10916-017-0697-4
Pruinelli, L., Yadav, Hangsleben, A., Johnson, J., Dey, S., McCarty, M., Kumar, V., Delaney, C. W., Steinbach, M., Westra, B. & Simon, G. J. (2016, July 20). A data mining approach to determine sepsis guideline impact on inpatient mortality and complications. AMIA Joint Summits on Translational Science Proceedings, 194-202. eCollection 2016. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5001751/
Westra, B. L., Christie, B., Johnson, S. G., Pruinelli, L., LaFlamme, A., Sherman, S. G.,… Speedie, S. (2017). Modeling flowsheet data to support secondary use. Computers, Informatics, Nursing: CIN, 35(9), 452-458. https://doi.org/10.1097/CIN.0000000000000350