Carolinas HealthCare System Consumer Analytics Case Solution

Pilot evaluation: You will evaluate the results of the first three pilots DA2 completed (i.e., Mapping Underserved Communities, Reducing Readmission, Advanced Illness Management). The following questions should guide your evaluation:

1. What was the business objective served by the pilot project?

CHS has always planned and visioned to develop a project that can help it to determine the data analytics that was integral for the future operations of CHS. In 2011 Dickson Advanced Analytics was launched (DA2) that mainly focused on the outcome of the results. As time passed, predictive risk models were developed under the supervision of DA2. The main purpose of developing such a model was to implement a change in the overall culture of CHS, meaning it should be based on data and evidence. Thus, three main pilot projects were proposed: Mapping underserved communities, Reducing readmissions, and Advanced illness Management. These pilot projects were developed to serve the business purpose to enhance the overall outcomes of the patients. In addition to this, transformative solutions could be given to the individuals so that community health issues can be addressed. 

2. What were the pilot project’s goals? How well do these goals align with strategic priorities for DA2 & CHS?

DA2 launched three pilot projects: Mapping underserved communities, Reducing readmissions, and Advanced illness management. All these projects had different goals. The objective of launching Mapping underserved communities was to make reductions in the emergency departments that were unnecessary. This was done by the identification of areas that were not served by primary care facilities. The project provides clinical data as well as descriptive data so that the best variables could be identified to provide access to primary care. This model also helped to anticipate any demand that arises in the future for the professionals the healthcare. Reducing readmissions launched in 2013 had a goal of helping CHS hospital teams to identify the patients with a high risk. This was done by creating an algorithm that served the purpose of calculating the readmission risk score for a particular patient. Internal data was used by the model that existed in the CHS facilities after which certain demographics and medical variables were shortlisted to make predictions. 

The models hold an accuracy of 79% regarding the patient risk so that they can receive extra medical focus. Advanced illness model was created to help patients that have complex medical conditions. Such patients had and higher risk of readmission to the hospital and thus were offered help from a team of experts so that they were in a position to better understand their health conditions. CHS was constantly looking to develop a project that can determine the data analytics that can be helpful for the operations of CHS’s future. Such key decisions would also make an impact on patient care as well as will be cost-effective in the long run. Therefore, DA2 was launched in 2011. The main goal of this program was to make improvements in the outcomes. CHS and DA2 both had the goal of developing an algorithm that involved consumer data. Considering this, the three pilot projects were developed that were in alignment with the goals of the projects. 

3. What did the pilot do? What were the key results of the pilot?

The pilot projects were focused on covering a variety of conditions based on geographics, medical, and functional capabilities. Mapping underserved communities collect clinical data as well as descriptive data from local primary care that helped in providing facilities to the emergency department (ED). So, the main function performed by this pilot project was to reduce the emergency department by providing access to primary care. When it comes to the primary care facilities many people are underserved and that is why ED become more necessary for them. The key results were that many locations were identified that were not initially targeted and were underserved. 

Reducing readmissions identify the high-risk patients. A model was prepared in such a way that works towards the identification of patients that have a high score in readmission risk. This provides CHS with complete data to make a certain prediction of whether admitted patients need some sort of revisits or not. For making such predictions demographics and medical variables were utilized. The results obtained from the model were segmented based on risk profit and demographic of patients that eventually helped the manager of the patients to make discharge arrangements along with timely and quality medication.