A hybrid analytic approach for understanding patient demand for mental health services
- 57 Downloads
The increase in digital/data resources available in the healthcare sector has heightened the emphasis of applying analytics to extract information to provide solutions to problems. However, the process of providing analytic-based healthcare solutions may introduce factors that require multiple analytic techniques or a hybrid approach. Data resources can involve complexities including formatting and volume issues or multiplicity of sub-tasks in achieving a full problem solution. This work extends the previous research on AI in forecasting patient demand and adds clustering methods to identify the types of ailments that need to be treated according to diagnostic codes. The hybrid approach is applied to data from a US-based psychiatry/behavioral health center and the results indicate clustering can add value to demand forecasts established by AI by identifying the type of ailments that patients require treatment for. With this information, care providers can better optimize staffing resources to meet demand in a cost-effective and efficient way by better understanding not only the amount of patient demand, but also the type of treatment that is required for select ailments.
KeywordsPsychiatric behavioral health Hybrid analytics Artificial neural networks Clustering
I would like to thank Thad Perry PhD for his assistance and guidance in this research.
Compliance with ethical standards
Conflict of interest
There are no conflicting interests and no fundings connected to this paper.
- Barr J (2017) Ensembles and regularization—analytics super heros, elder research data science and predictive analytics. https://www.elderresearch.com/company/blog/regularization-and-ensembles. Accessed 15 Dec 2017
- Catalano R (1991) The health effects of economic insecurity. Am J Public Health 18(9):1448–1552Google Scholar
- Davis D, Chawla N, Blumm N, Christakis N, Barabasi A (2008) Predicting individual disease risk based on medical history. In: Proceedings of the 17th ACM conference on information and knowledge management, California, pp 769–778Google Scholar
- Fayyad U, Shapiro G, Smyth P (1996) From data mining to knowledge discovery in databases. Artif Intell 13(3):37Google Scholar
- Ha S, Joo S (2010) A hybrid data mining method for the medical classification of chest pain. World Acad Sci Eng Technol 37:608–613Google Scholar
- Hu J, Wang F, Sun J, Sorrentino R, Shahram EA (2012) Healthcare utilization analysis framework for hot spotting and contextual anomaly detection. AMIA annual symposium proceedings, pp 360–369Google Scholar
- Kudyba S (2012) Utilizing multivariate analytics for decision support to enhance patient demand forecasting. Research and Insights. International Institute of Analytics, Portland, pp 1–7Google Scholar
- Kuo MH, Hung CM, Barnett J, Pinheiro F (2012) Assessing the feasibility of data mining techniques for early liver cancer detection. Stud Health Technol Inf 180:584–588Google Scholar
- Lurie SJ, Gawinski B, Pierce D, Rousseau SJ (2006) Seasonal affective disorder. Am Fam Physician 74(9):1521–1524Google Scholar
- Meystre S, Thibault J, Shen S, Hurdle J, South B (2010) Textractor: a hybrid system for medications and reason for their prescription extraction from clinical text documents. JAMIA 17(5):559–562Google Scholar
- Rumelhart D, McClelland J (1996) Learning internal representations by error propagation. MIT Press, LondonGoogle Scholar
- Sridhar S (2012) Improving diagnostic accuracy using agent-based distributed data mining systems. Inform Health Soc Care, September 7Google Scholar
- Vulik S, Mayer E, Darzi A (2016) Enhancing risk stratification for use in integrated care: a cluster analysis of high-risk patients in a retrospective cohort study. BMJ 6(12):e012903Google Scholar
- Wu X, Kotagiri R, Korb K (1998) Research and development in knowledge discovery and data mining. Second Pacific-Asia Conference, Melbourne AustraliaGoogle Scholar