Sample Article Critique on Opportunities and Challenges in Developing Risk Prediction Models

The article “Opportunities and Challenges in Developing Risk Prediction Models with Electronic Health Records Data: a Systematic Review” explores the significance and challenges related to the usage of electronic health records in clinical risk prediction. Goldstein et al. (2017) argue that the use of electronic health records has significantly increased across the United States. Goldstein et al. (2017) outline that the use of electronic health records across the country increased from 12.2 percent in 2012 to 75.5 percent by 2019. The article contends that electronic health records have been largely used for clinical research purposes, and currently, they are also being utilized across various health settings for clinical risk predictions. Goldstein et al. (2017) contend that the usage of electronic health records in clinical risk predictions is associated with various advantages. The article argues that electronic health records can enable a health professional to observe more metrics, on a large population of patients, within a short period, and at a low cost compared to cohort studies. The article also argues that electronic health records provide data that can be readily implemented in clinical risk predictions, unlike traditional algorithms that need to be translated into a clinical environment before being used for medical purposes.

The article argues that although the use of electronic health records for clinical risk predictions is associated with various advantages, it is also riddled with numerous gaps. Goldstein et al. (2017) argue that one of the challenges associated with electronic health record data is the presence of missing data. Another challenge associated with the use of electronic health records is that the security of patients’ health information is not guaranteed.

Critique

Although electronic health records have proved to be of great benefit in the health sector by providing extensive health data that can be used for clinical risk predictions, I believe that there are numerous issues apart from missing data that EHRs have failed to address. One of the issues that electronic health records have failed to address is patient privacy protection. The usage of electronic health records in patient screening, tracking of patient outcomes, and clinical and health care outcomes research is associated with controversies. Many people attempt to debate whether or not EHRs fulfill or go against the national patient privacy standard. Notably, EHRs do not ensure adequate patient privacy protection. For instance, when patients present similar information in large cohorts, their data can be mixed up with others, thus rendering the data unprotected. Electronic health records also do not ensure patient privacy protection as there is a risk of patients’ data being obtained by unauthorized personnel. This is because passwords of various electronic health record systems may fall into the wrong hands, and unauthorized persons may use the opportunity to obtain patients’ data for malicious activities, such as blackmailing patients. Health settings can ensure adequate patient privacy protection by ensuring that only authorized persons have passwords to the electronic health systems and that patients’ data collected in large cohorts are stored separately in various electronic files.

Additional issues that have not been addressed as a result of the usage of electronic health records for clinical risk prediction include loss of productivity and the need to update the systems regularly. The use of electronic health records often contributes to disruption of work-flows for medical staff and providers, thus resulting in a temporary loss in productivity. The loss in productivity is often stemmed from end-user trying to learn how to use the new systems, and in the long run, this causes the quality of clinical outcomes to decline. Besides, electronic health records need to be updated regularly. Therefore, if not updated, patients’ data may end up being lost and this may prove costly to a given health setting.

 

Reference

Goldstein, B. A., Navar, A. M., Pencina, M. J., & Ioannidis, J. (2017). Opportunities and challenges in developing risk prediction models with electronic health records data: a systematic review. Journal of the American Medical Informatics Association, 24(1), 198-208. https://doi.org/10.1093/jamia/ocw042