Journey Through 30 Years of Healthcare Data Analytics

First Stop: A Look at Historical Per Capita Cost Trends

I recently hit a major career milestone of 30 years in healthcare. With my youngest daughter Kendall joining me in the office as our marketing intern, we came up with the idea to create a series of articles about my experience in healthcare data analytics during this often crazy but very rewarding time within the US Healthcare system.

My healthcare career began almost as a punishment by my first employer, an IT consulting firm. I was supposed to be learning a cool new technology in manufacturing. However, due to logistical issues finishing my master’s degree, I was placed with a healthcare client using the very old programming language BASIC. It was a blessing in disguise, as I was very intrigued right away by the business of healthcare and quickly redirected my career from IT into the data analytics space. In those early years, I was developing analytics supporting capitation models for primary care, cardiologists, and other specialties. We were helping to build incentive models for physicians to keep members out of the hospital and developing full-cap deals for hospital systems to take on all healthcare cost risk (unsuccessfully, I may add – more on this in a future article).

This kickoff article looks at high-level Healthcare Cost Trends over the last 30 years. For the sake of simplicity, I will generalize this time into four decades, starting with the 1990s and continuing through today. I would summarize what was happening in healthcare care cost containment and the resulting data analytics work over these four decades as follows:

 
Let’s consider US healthcare annual cost trends per capita over these decades. The chart below uses a combination of healthcare spend data from the CDC along with the US Census data to develop per capita costs. From the CDC data, we select those areas of spend that we would consider closely tied to health insurance premiums: hospital care, physician care, and prescription drugs.

• The first amazing thing we see in this chart is that trends for the 30-years prior, from 1960-1990, were a whopping 10.3% per year. If this trend rate had continued, we would be looking at 2019 per capita spend of $30K per year vs the $7,110 actual spend!
• Seeing a big drop in the 1990s to 5% trends per year on average points to HMOs, in general, being effective at reducing medical costs and keeping patients out of the hospital. Rx is an interesting story in the 1990s with double-digit trends remaining; I suspect due to a large number of drugs that we would consider maintenance medications today being in their branded heyday.
• In the 2000s, hospital and overall trends returned, coinciding with a backlash against HMOs (recall or go check out the famous line about HMOs from the movie “As Good As It Gets”).
• In the 2010s, with the resurgence of value-based care and the introduction of the ACA (Obamacare), overall trends were actually lowest.

A final overall observation considers the resilience of hospitals in keeping their spending trends on par with overall trends over the past three decades, despite continuous efforts by payers to move care “out of the hospital setting.” Underneath these hospital trends, we recognize that there has been a major shift from the inpatient setting (reduced admissions) to the outpatient and ambulatory care settings. The use of the Emergency Room was a concern 30 years ago and still is today. Hospital care has also grown significantly in areas such as cancer care, gene therapy, etc.

One last thing to consider is that we are looking at trends across the entire population. As those of us in the industry know, this population is a combination of Medicare, Medicaid, and Commercial (Employer-sponsored and Individual), along with the Uninsured. Since the Government can and does use its leverage to limit the trends of Medicare and Medicaid, the result of this is that the Commercial product trends end up higher, sometimes much higher, than the averages seen here.

This chart serves as a launchpad for our exploration of data and analytics over the past 30 years, providing a glimpse into the fascinating journey ahead.  Some of the upcoming topics I will be looking to cover:

• The Non-Traditional Actuarial Role of Medical Economics and the ROI of Healthcare
• Payer side vs. the Provider side of healthcare cost containment
• Value-based care: A look back: What is working and what is not?
• Pharmacy trends over the decades: Does the US subsidize the rest of the world?
• And other current and relevant topics such as:
  • Hospital Rate Regulation and the Maryland Model: Is there a better model out there?
  • Why have so many smart companies failed in their healthcare ventures?
• Do the traditional economic principles of supply and demand work in healthcare?

In retrospect, I feel very fortunate to have been thrown into this world of healthcare. My career has been rewarding and the work continues to challenge me every day. I’ve had the pleasure of working together with many great and talented people. In some ways, I feel that we have accomplished a great deal, but in other ways, I feel like we have not moved the needle very much at all.

Please feel free to leave your comments below on this article, or thoughts on other topics you would like me to address. I will try to keep these articles relatively short but interesting. I welcome everybody’s input to keep the conversation going. Our constant challenge is to make the healthcare system better for everyone. Cheers!

Data sources:

National health expenditures, average annual percent change, and percent distribution, by type of expenditure: United States, selected years 1960–2019

Health, United States 2020–2021 (cdc.gov)

US Census (except 2009 US Population estimate from USAFacts.org)

https://www.census.gov/data/tables/time-series/dec/popchange-data-text.html