An Unexpected Lesson from the Emergency Room

As actuaries, we are constantly scrutinizing unit costs, utilization patterns, reimbursements, and more in an attempt to blunt the impact of distressing healthcare cost trend. I have a confession to make – in my personal life I recently violated many of these efforts during an emergency room visit. In the end, it was a valuable learning experience about what can easily be forgotten when analyzing claims data at any sort of aggregated or depersonalized level.

My Emergency Room Experience

As a carrier of a genetic blood clotting condition, I was on high alert for any symptoms of pulmonary embolisms following a recent surgery, which left me mostly bedridden for several days. When I began experiencing shortness of breath and chest pressure five days into recovery, and the symptoms did not resolve, I decided to make a trip to the emergency room at 10 pm that night. After explaining everything to the doctor, including my family history, where two immediate family members nearly died from pulmonary embolisms, he explained that the next step would be to perform a D-dimer blood test. A CT scan would be performed if levels were elevated.

I knew my next words were going to ruffle some feathers with both the doctor and my actuarial peers, but I proceeded anyway to ask whether I could get a CT scan regardless of the D-dimer results. Given my family history and current risk factors, I wanted peace of mind. The doctor, who knew my profession, conceded to ordering the test for reasons I laid out but quickly made fun of the irony that an actuary was asking for additional tests. I ate my serving of humble pie, but it was worth whatever extra amount I’ll be billed in 4 months in order to sleep easily (not that night; it was already 11 pm by this point). Fortunately, even though the D-dimer levels were slightly elevated, the CT scan was all clear, and I returned home well past midnight.

As we grapple with untenable healthcare inflation, we must keep the patient experience at the forefront.

The Importance of Patient Satisfaction

ER visits rarely have any entertainment value, but this one did. I left that night with two takeaways to remember when analyzing population-level data. First, we cannot neglect the value of patient satisfaction. In nearly every case, the doctor would have been right that a low D-dimer meant I could’ve gone home and gotten two extra hours of sleep. Most patients would probably have been slightly less informed than me (for better or worse!) and be satisfied with that level of care. But what is the value of peace of mind? For an insurer, what degree of leniency can be permitted in order to retain a happy and loyal member? As we grapple with untenable healthcare inflation, we must keep the patient experience at the forefront.

As value-based reimbursement continues to gain momentum with no slowing in sight, perhaps patient satisfaction can be incorporated into quality metrics as well. Most value-based programs reward providers for stellar HEDIS scores. This is crucial to incentivize care management rather than care suppression. Incorporating a patient satisfaction component is critical to both member retention and customer happiness. Providers are on the frontlines dealing with patients who need to be assured they’re getting all the care they need. In doing so, providers should be financially incentivized to create this experience for insurers. Formulating an effective way to capture satisfaction is beyond this article’s scope, but member surveys seem to be a good place to start.

Clinical and Actuarial Integration

The second lesson learned was the importance of clinical expertise working hand-in-hand with actuaries. After discussing various studies with my ER doctor and reading numerous studies, it is true that a normal D-dimer means no pulmonary embolisms for the general population. But other studies have shown that a normal D-dimer level for patients with multiple clinical risk factors is not as reliable[1]. One study found that 9.3% of high-risk patients with normal d-dimer levels had pulmonary embolisms, compared to 1.1% in the low-risk group[2].

Actuaries must understand these nuances if our data analysis is going to impact the healthcare of so many people. This is just one case where clinical considerations have such a clear impact on appropriate care delivery. CT scans are expensive and unnecessary utilization should be avoided, but the stakes are life-or-death. Studies have clearly demonstrated the relevance of clinical risk factors that would probably be missed by an actuary analyzing claims data.

A “ground-up” approach, which includes coordination and communication with physicians, should reach similar conclusions as a top-down data-driven approach. Our work has real consequences, and sometimes, the stakes are life or death. Gaining physician trust and buy-in is important so that they can focus on doing their best work for their patients. This is one of the reasons AHP has always employed physicians and delivered the best of both actuarial and clinical expertise to customers.

Endnotes

[1] https://www.sciencedirect.com/science/article/pii/S1538783622032512

[2] https://www.aafp.org/pubs/afp/issues/2009/1001/p742.html