May 03 2017 | 0 Comments | 175 reads Average Rating: 3
Casting a Wide Data Net
Healthcare professionals often turn to clinical and claims data to gather insight about patients and members, but they might do well to look beyond these usual suspects. Also with the movement of how providers are being paid it is now critical for the doctor to know if the patient will comply with their orders and if not what are the barriers. For example, if a patient has an income of $2,000 or $3,000 per month picking up a drug that costs $100 per month may not be possible as most of their net pay will need to go toward everyday needs such as rent or food/utilities. It’s a topic that I recently addressed during a Webinar entitled “Is Your Data and Risk Strategy Enabling You to Take on Risk Contracts?”
For example, a study conducted by the Robert Wood Johnson Foundation found that zip code data – while primarily thought of as something that gets letters from point A to point B – can actually uncover significant insight into the health status of individuals and populations, and could perhaps be even more telling than genetics. Indeed, RWJF together with Virginia Commonwealth University published an analysis comparing the life expectancy of residents of two cities in New Jersey – affluent Princeton and economically strapped Trenton. The average life expectancy for a person in Princeton was 87 years, while in Trenton, the average life expectancy was just 73 years.
Because such insights can be garnered from seemingly unexpected places, healthcare organizations should look to a broad array of data sources to gather insight on patients and members. Organizations can create data marts by looking to the following type of sources:
- Medical Claims
- Prescription Drug
- Household income
- Electronic Medical Record
- Government available data
- Gross Domestic Index
- Flu rates
Figure 1: Layers of Data fuels outcomes, © SCIOinspire, Corp., 2017
By creating robust data marts through these inputs, healthcare organizations can drill down into the data to get needed answers that might be difficult to access when simply relying on commonly used data sources. For example, a payer organization might want to assess the level of financial risk that new members bring with them when joining their health plan through the exchanges or as an employer where there is no prior data or even as a member who has no prior healthcare services. Claims information is typically not available for new members, however, so health plans have difficulty assessing risk.
That being said, if plans develop robust data marts populated with various types of data, they can use the various data sets to create personas that can be linked to the level of financial and clinical risk associated with the new members combined with what preferred mode of communication works best for them such as email, phone calls or texts. Median risk scores can then be developed and applied to various persona categories such as “healthy & affluent,” “quality driven” or “chronic older adults.” These personas can project how these individuals will use health care services.
This is just one of the ways that healthcare organizations can use data to assess risk. Can you think of others? Perhaps even more important than leveraging data to assess risk, organizations might want to leverage data to implement improvements for their care management or consumer engagement departments. To that end, my co-presenter during the Webinar, Teresa Rueckert, RN, MSN, Clinical Solution Consultant at SCIO Health Analytics, has addressed how data can be used to determine “impactability” and affect change in her blog.
1. Robert Wood Johnson Foundation. Tackling Life Expectancy Gaps in New Jersey. http://www.rwjf.org/en/culture-of-health/2016/11/tackling_life_expect.html