"CP Unlimited is a medical care organization that constantly strives to improve its services and support for individuals with developmental disabilities. Their mission is to help provide the most comfortable lives possible for the people who depend on them. Lately, CP Unlimited has been considering implementing a palliative care pilot program for their sites across New York. Their clients usually need more thorough medical attention, so it made sense to see if palliative care could reduce events like hospital admissions and long-term stays in the ICU. My project was to create a data analysis template that could be used to compare the reduction in hospital cost and hospitalization metrics (e.g., admissions, days in hospitalization, etc.) across different boroughs and care sites once a pilot palliative care program is implemented. Moving forward, this template can also be used for a year-over-year comparison to track performance of CPU’s palliative care program.

At CBS, I picked up various frameworks on how to report and visualize findings in an exploratory analysis. These frameworks would be useful when applied to my template. My health care class, combined with nonprofit- and philanthropy-related classes, helped provide an overall landscape to consider while building my analysis. I had some background on how a nonprofit health care organization works, which helped me understand the context of my project. Outside of class, I worked on a project that involved taking data from a school-wide survey and providing analysis and visualizations of the results to the student body. I applied the technical skills I honed from that experience to this internship, as well as my experience with delegating work and creating goals and timelines. I formally presented my template, along with proposed next steps to senior leadership. Here, my Leader’s Voice class came in quite handy! Additionally, I had various ad hoc requests for metrics to be shown for a site-wide meeting. Both of these instances were helped by my experience at CBS building decks and trying to tell a cohesive narrative with my data.

Due to COVID-19, this internship involved mostly remote work. I could not get as much facetime with my manager and senior leadership as I wanted. I came into the office two times a week and had to purposefully schedule meetings with people so I could communicate my status and findings on the project. On the technical end, I had been used to data that was already cleanly processed and provided directly to me in my class projects. The real world, of course, is a bit messier. I had to use the organizational chart and figure out which departments and contacts were relevant to helping me find the data I was looking for. Once I received the data, I found that it required standardization and sometimes imputation. The process of cleaning was something that took longer in my timeline than I anticipated, but it also made the analysis and visualization portion more efficient than I expected, because I had a streamlined, more standardized set to work with. Moving forward, I will mindfully account for inspection and repair for given data when embarking on an analytics project.

I had worked with this organization under the same leadership last year, so my goals were to deliver my work with the same quality as last year, if not more. I also wanted to extend my expertise beyond just data analysis and truly grasp the context of my project. I learned that I could best achieve this by digging deeper than what was expected. Data analysis does not have to be only limited to reporting. I could take my template and use what I knew to provide insights about what could affect hospitalization rates.

It is becoming clearer over time that understanding how to handle data in the health care sphere is helpful in delivering good quality service. Health care is a limited resource environment with scarce capacity, so each decision made needs to have a good amount of certainty backing it up. Because there is a growing availability of data in our society, we can evolve predictive and prescriptive analyses to help guide us to the right decisions. "