Questioning Coverage for Coronary Artery Scans

Why we should pump the breaks on insurance coverage for CAC scores

As many readers know, I hold a highly skeptical view of CAC scans. Andrew Foy and I have made the case against CAC scans for any use. In today’s guest post, Zachary Robert Caverley argues that CAC scans should remain uncovered by medical insurance. JMM

---

The coronary artery calcium (CAC) score – an estimation of coronary atherosclerotic plaque via calcium build-up on CT – has been promoted as a powerful predictive tool for estimating the chance of future cardiovascular (CV) events. Proponents recommend CAC scores to enhance the care for patients determined to be at “intermediate” risk. The scan has minimal radiation exposure and is relatively affordable.

Despite this, CAC scoring in the US is still paid for mainly by patients and not insurance.

Typically, cardiovascular risk stratification starts with a pooled cohort equation (PCE) that uses basic clinical information. This outputs a 10-year risk of a cardiac event. Proponents of the CAC score say the imaging test may improve on the PCE.

The United States Preventive Service Task Force (USPSTF), however, has deemed the evidence for this as “insufficient.” Hence, most insurance plans do not cover the scan.

In the journal Circulation, CAC proponents have expressed dismay about the lack of coverage:

“…nearly all insurance payers, including Medicare, have continued to deny coverage for CAC testing for primary prevention risk assessment…We believe the time is overdue for this coverage to be implemented so that all patients, not just those who can pay out of pocket, have access to evidence-based, personalized preventive cardiovascular care.”

While the Circulation authors’ goal sounds admirable, their argument oversimplifies the implications of insuring the CAC score. I propose the following arguments to keep testing paid for out of pocket:

1. There is no current evidence that CAC scoring saves lives

Many of the writers advocating the test rely on the following inference: (1) CAC scoring predicts cardiovascular risk, (2) people at high risk will be treated with statins, (3) many RCTs find that statins reduce CV events, and therefore (4) CAC scoring will benefit the screened population.

While this logic appears sound, we have no firm evidence that screening with this test will save lives. Only a single randomized study, the DANCAVAS trial, has evaluated the mortality benefit for routine comprehensive screening involving the CAC score. After five years, no significant mortality benefit could be found (see footnote below). This was surprising given subjects also underwent an ECG, an Ankle Brachial Index (ABI), lifestyle education when scans were positive, and blood tests to detect diabetes and hypercholesterolemia. (See footnote).

It does not follow that insurance companies should begin covering a test without a proven mortality benefit compared to traditional methods (and this is especially true for female patients, given DANCAVAS was an all-male trial). This has unfortunately not stopped advocates from writing things like, “There is an opportunity to save lives and dollars if CAC testing is covered for appropriately selected individuals.”

2. CAC score prices are already declining

Proponents may counter that heart disease already carries a high economic burden for the US healthcare system. Coverage of the CAC score could be viewed as a “drop in the bucket” if it has the potential to save money when performed in appropriate patients.

The problem with this argument is it ignores the savings already gained by leaving testing to market competition. The same authors claiming coverage would “save lives and dollars” also noted, “although the cost of a CAC test used to be >$400, today it ranges from $100–$200.” The most likely explanation for this decline is laissez-faire: the predictable pattern of competition driving down prices as patients shop around for an affordable test.”

The Circulation authors also noted this trend, writing, “In the absence of insurance coverage, some hospitals and medical centers have markedly lowered their charges for CAC scans to the $50 to $100 range.”

Beyond the fact that the current system keeps prices reasonable, it also produces transparent costs which enable patients and clinicians to make informed decisions.

The prices described above may still exclude patients on fixed incomes. To be fair, current evidence does find an increased uptake of CAC scoring when it is free, but if we consider the lack of a mortality benefit, it is not clear that increased testing would be a good thing in an era where concerns over “low value” care are rampant. If a patient is truly in a financially precarious situation, getting a CAC score should be low on their list of priorities.

3. More coverage means more inappropriate testing

One of the main concerns of CAC scans are the risk of inappropriate testing.

Adults scanned at younger ages may not realize the test has low diagnostic utility in their age group and may fail to adequately control risk factors once their clinician has reassured them with the “power of zero” from a negative scan. The opposite issue – the potential for overtreatment – can be seen in a study showing CAC was able to successfully reclassify risk for many patients with CV events but misclassified a higher absolute number of patients without events. And although the radiation exposure is indeed mild, more coverage will inevitably lead to more exposure.

Keeping in mind lower cost sharing is associated with greater healthcare utilization, the potential for more of these cases coming through the door once scans are covered is hardly something to look forward to. Even without coverage, the real-world examples of inappropriate CAC scoring our team has encountered are troubling. Just to name a few:

• A patient diagnosed with coronary plaque via angiogram returned to clinic years later with a highly positive calcium score – a test that changed nothing but her level of anxiety.

• A 40-year-old, falsely secured by his score of zero, stopped his statin not realizing all his previous testing demonstrated he would greatly benefit from risk factor modification.

• A patient had a CAC score ordered from a clinician who failed to notice his chest CT ordered one year prior had demonstrated severe coronary calcification – a fair proxy for CAC scoring in his situation rendering a dedicated score pretty much useless.

One might counter that advocates are not asking to cover all patients, just appropriately selected ones. This may sound reasonable on its surface, but due to “scope creep” it is rare that screening recommendations remain restricted to the initial target.

The USPSTF, for instance, recently lowered the recommended mammography screening age to 40 years, but the American College of Radiology (ACR) claimed this change does “not go far enough to save more women’s lives” One could expect a similar shifting of the goal posts once CAC scores are covered.

Conclusions

The goal of expanding access for preventive care may seem desirable, but the current evidence does not support insurance coverage of CAC scoring. The market-based approach has already produced affordable pricing while maintaining transparency. Until there is clear evidence of mortality benefits, keeping CAC scoring as an out-of-pocket expense may be the most prudent approach to balance access, cost, and clinical utility.

---

Footnote 1 from the editors: The DANCAVAS trial found a 5% relative risk reduction in mortality with a comprehensive package of cardiovascular screening. Though this did not reach statistical significance, the 95% confidence intervals ranged from 0.90-1.00. The P-value calculated at 0.06. Another interpretation would be that the DANCAVAS package of screening allowed for 10% reduction in death to no reduction. The main caveats of DANCAVAS were that a) CAC scoring was only one component of the screening, b) the screening was intentionally minimalist, for instance, no overreads of the CT scans were done, c) the trial environment was Denmark, which is notable for having a national health system and low rate of downstream testing. The latter two factors would make such a program unlikely to replicate in the fee-for-service US healthcare system.

Second Postscript: I received e-messages this AM stating that I was too generous in the above footnote. Here is a previous critical appraisal of DANCAVAS by Andrew Foy that we published in 2022.