Data Mining Applications in Healthcare, Research Paper Example

In customer relationship management planning, data mining is at the forefront. To make predictions in customers’ patterns and behaviors, organizations can use the information gathered to make critical decisions and generate proactive initiatives. Simply collecting data is not enough anymore; the data must go through an analytic process called data mining. Data mining applications can be essential tools for the elderly, physicians, and healthcare providers, to better assess needs. This paper will discuss how data mining may be used to predict the likelihood of physicians offering vaccines to the elderly during their primary care visit. This may answer the question as to whether or not vaccination rates among the elderly is decreased due to physicians’ noncompliance with offering these vaccines.

Indicators Used to Track and Manage Data

In order to predict each physician’s behavior and patterns in relationship to immunization compliance in the elderly, it is first necessary to present a plan of action to track and manage data. Data mining, in relation to this paper’s thesis statement, can best be accomplished using the following indicators: graphing how many patients receive vaccines, how many vaccines are ordered by each physician, patient education, and provider education. Data mining models as well as physician and customer surveys, are excellent tools and techniques that can be utilized. This data can then be used to give to doctors in order for them to have a collection of data in order to design a clinical profile for them that will then be used by the physicians in order to assess their own patterns of treatment. These patterns can then be used to compare with other physicians (p. 66).

Immunization compliance in the elderly could be improved, as the data of such mining will allow physicians to recognize their failings in regard to this specific patient population.

Data Collection Process

In order to properly predict healthcare provider’s behaviors and patterns for this specific problem and population, it is necessary to find a data collection process. This can best be done through the electronic health record (EHR). The data that is gathered through stratified age population charts, clinical profiles, and pattern comparisons, can be organized into accessible, and easy to understand data sets that are given to each physician in the organization. Such data can be discovered in a myriad of ways such as paying attention to patterns. Patterns found in the patients’ encounter can be taken from the EHR and compared to other physicians to determine in what ways immunization compliance was treated. An example of a way in which data collection will be acquired can be found with the Group Health Cooperative where they stratify their patient populations by their demographic characteristics. This allows for the cooperative to find out which group is using their resources more(Koh & Tan, n.d., p. 66).

Another form of data collection may come with examining and applying data found for vaccine purchases. Also, each insurance company may have a claim’s list of vaccinations in order to find data on how many elderly patients are being vaccinated by each physician. Comparison data could be made to as to what degree the elderly are being vaccinated as compared to vaccine rates in the younger age groups. Such patterns allow forcomparisons across healthcare groups, practice patterns, resource utilization,length of stay, and costs(Koh & Tan, n.d., p. 66).

Uses of Identified Data and Impact Practice Area

Such data will add in the clinical decision support system. By having a more transparent operating system infused with data mining collection in regards to physician behavior and patterns, a better working system can begin to emerge. The data will be used to ensure that proper care is being given to all age group populations and that primary care physicians’ are being compliant with offering vaccines as indicated by the CDC.

Point-of-care reports created by the EHR for the physicians’can be used weekly to provide brazenly their own lack of effort in compliance with vaccinations. Physicians may also see their charted pattern and discover in what specific ways they have been noncompliant. Everyone including the organization and patients can benefit from this broad database of knowledge. The major impact for advanced nurse practitioners is the determination whether or not their behaviors are the direct result of decrease vaccine compliance or other factors are responsible.

Conclusion

Data mining may prove to be not only a beneficial means to come to the determination of physicians’ behaviors related to decrease in vaccine compliance, but also beneficial to the knowledge and infrastructure of other organizations onmany healthcare topics. Breaking down physicians’ patterns, patient population by age, and having that stratification grouping presents data that enforces the idea that data mining can enable support for cost-savings and decision making. This paper has shown how vaccination compliance could be improved through data mining, emphasizing on physician practice patterns, and resource utilization.

References

Aspinall, R., Giudice, G., Effros, R., Grubeck-Loebenstein, B., &Sambhara, S. (2007). Immunity and Aging 4(9). Pp. 1-9.

Koh, H. C., & Tan, G. (n.d.). Data mining applications in healthcare. Journal of Healthcare Information Management 19(2). Pp. 64-72.

MacLeod, W., Eidus, R., & Stewart, E. (2012). Clinical decision support: using technology to identify patients’ unmet needs. Family Practice Management. Retrieved from http://www.aafp.org/fpm/2012/0300/p22.html