The Success of Data Analytics for Carolina Health Systems
With the development of Dickson
Advanced Analytics (DA2) group, the Carolinas Healthcare
System (CHS) was able to strategically unify disparate analytics departments to
guide the organization to its goals. The
analytics driven foundation significantly helped to improve CHS in terms of
better investment strategy and better patient outcomes. The impact of a data
driven organization eventually helped CHS move towards a more evidence-based
focus and reduce costs significantly not only for CHS but for the patients as
well (Quelch & Rodriguez, 2015). In terms of success, the strategy CHS used
to leverage analytics was a great move. Notably, the four pilot programs discussed
in the article all have their own significant impact and, in my opinion, not
one is more impactful then the other.
Mapping
Underserved Communities
This pilot project was created to
evaluate and identify populations with poor access to care. Through the process
of leveraging specific data markers that included median household income,
insurance type, and emergency department incidence data etc. CHS was able to
create a single measure for need of primary care offices in a given area. The
outcome essentially led to improved access to primary care and a method for
evaluating medical need in underserved communities (Quelch & Rodriguez,
2015).
Reducing
Readmissions
The Readmission predictive risk
model was a piloted project to reduce readmissions, which took a risk score
that identified patients in a high-risk category. The data was composed of
readmission rates, disease type, and an additional 38 categories that are
thought to be highly predictive. The outcome essentially led to better
facilitation of care services amongst high-risk patients (Quelch &
Rodriguez, 2015).
Advanced
Illness
The advanced illness management
pilot project was created to help patients with complex medical conditions. The
focus was to identify patients with complex conditions who had a higher risk of
readmission due to poor intervention. The outcome of a first cohort which led
to better quality of care and reduced costs for patients, eventually led to the
creation of a second and third cohort (Quelch & Rodriguez, 2015).
Patient
Segmentation
This pilot program placed patients
in seven categories based on medical need. The data derived from this program
allowed CHS to better tailor care to a patient’s specific needs. The long-term
impact aided in identifying patients who could benefit from physician
intervention. Moreover, this also served as an aid in influencing bids on new
payer contracts (Quelch & Rodriguez, 2015).
Success
of DA & Strengths of Pilot Programs
The shift to value-based care has
spearheaded the movement to a more evidence-based health system. This change
has come with challenges. However, in the case of CHS their grouping of
disparate analytics teams seemed to be a strategy that worked. This increased
their analytics capabilities and allowed CHS to combine multiple groups for one
common goal. This model of engaging the right people essentially led to organization
wide buy-in and the growth of the program. Kakad, Rozenblum, and Bates (2017)
acknowledge the necessity of having the right people. In leveraging the right
people for building a predictive model you not only improve frontline buy-in
but will also have the means of developing a tool that can further engage those
at the C-suite level. With regards to
each pilot program, each has the potential to significantly improve healthcare
outcomes of different high-risk categories. The potential impact this has not
only for patients but on the reduction of costs for wasteful use will
significantly impact the organizations bottom line. Overall CHS has done a
phenomenal job both with implementation of DA and in leveraging this team to
pilot programs for success.
References:
Quelch,
J. A., & Rodriguez, M. L. (2015, April 14). Carolinas HealthCare System:
Consumer Analytics. Retrieved from https://hbsp.harvard.edu/download?url=%2Fcourses%2F614971%2Fitems%2F515060-PDF-ENG%2Fcontent&metadata=e30%3D
Kakad,
M., & Bates, M. R. (2017, June 23). Getting Buy-In for Predictive Analytics
in Health Care. Retrieved from
https://hbr.org/2017/06/getting-buy-in-for-predictive-analytics-in-health-care
