Outcome Measures and Risk Adjustment

Topic: Healthcare Research
Words: 579 Pages: 2

Based on the scenario, it is recommended that the data analyst should use detailed data to investigate this quality issue. Aggregate data can be considered inappropriate for the given case since they do not provide a comprehensive overview of patient information. Instead, they present information in the form of data summaries for the purposes of statistical reporting. If the data analyst used aggregate data to investigate the issue, the only thing he or she would learn is the trend in data. However, it has already been discovered that there was a spike in stroke mortality for the given quarter, and the risk of stroke mortality has a score of one to four. Thus, the investigation of aggregate data may not give insights into the nature of a rapid increase in stroke mortality. That is why the data analyst should closely examine detailed data, which could reveal some important measurable risk factors that have not been taken into account.

The data analyst is recommended to use both a retrospective data warehouse and the clinical data store. On the one hand, the retrospective data warehouse can help the data analyst examine the exposure of patients who died from stroke to suspected risk factors. On the other hand, though, the clinical data store can show consolidated data about a single patient collected from several different sources.

When thinking about the type of tools and analytical approaches that may be relevant for the use by the analyst, one may mention statistical tools, such as logistic regression and cluster analysis. Logistic regression may assist in identifying risk factors that contributed to stroke mortality, as well as their significance. Cluster analysis can help the analyst determine the differences between the patients who died from a stroke and those who did not.

Three sources of evidence discussing the challenges of utilizing data in the clinical setting have been selected. The first source is a literature review of the issues presented by the use of patient-generated data in healthcare. The second source focused on the challenges for mining big data in healthcare. The third source delineated current and potential problems stemming from the application of big data in the healthcare setting.

With the proliferation of digital health, a dramatic increase in the data generated by patients has occurred. Despite the fact that self-tracking practices may yield reliable data due to automated algorithms, there are doubts if this information can support clinicians in their decision-making. It has been found that issues associated with the use of patient-generated data relate to context, accuracy, and reliability (West, Kleek, Giordano, Weal, & Shadbolt, 2017). Particular attention should be paid to the decision context in which medical information is used.

Challenges for data mining, data storage, data sharing, and data privacy have become a major concern for healthcare providers. As the amount of unstructured medical data grows in exponential progression, the efforts of the IT staff in healthcare should be concentrated on timely and accurate data mining, maintaining safety and confidentiality of patient data in the process of storing, and improving the interoperability of data (Hong, Luo, Wang, Lu, Lu, & Lu, 2018). The authors of the third source claimed that there is little evidence about the practical benefits of the use of big data in healthcare (Lee & Yoon, 2017). Due to a number of methodological issues, such as quality and inconsistency of data, as well as analytical and legal concerns, there is a strong need to improve the data quality.

References

Hong, L., Luo, M., Wang, R., Lu, P., Lu, W., & Lu, L. (2018). Big data in health care: Applications and challenges. Data and Information Management, 2(3), 175–197.

Lee, C. H., & Yoon, H. J. (2017). Medical big data: Promise and challenges. Kidney Research and Clinical Practice, 36(1), 3–11.

West, P., Kleek, M. V., Giordano, R., Weal, M., & Shadbolt, N. (2017). Information quality challenges of patient-generated data in clinical practice. Frontiers in Public Health, 5(1), 284–291.