Column: Anthem’s effort to punish patients for ‘unnecessary’ ER visits has been a bust--but still burdens patients

Indianapolis-based Anthem continued to rack up higher earnings as it put the squeeze on its own patients.
(Michael Conroy / AP)

In the last few years, Anthem Blue Cross has made a strong bid for the award for the most heartless and senseless coverage policy in the health insurance business.

Its competition entry is a policy that penalizes patients for seeking “unnecessary” treatment at an emergency room. If Anthem concludes that the reason for the visit wasn’t an emergency after all, it can deny the claim—saddling members with bills that could exceed $10,000.

Anthem’s rationale is that the ER is the costliest venue for medical treatment; therefore, weeding out patients whose medical complaints could more appropriately be managed through a doctor’s appointment or a visit to an urgent-care clinic will save money for Anthem and for its customers base. Anthem launched this program in Kentucky in 2015 and expanded it in 2017 and this year to Indiana, Georgia, Missouri and Ohio. (The company dropped plans to expand it this year into New Hampshire.)


I’m a board-certified trained doctor of emergency medicine, and I have trouble ... knowing what the patient was thinking at 3 o’clock in the morning.

— ER physician Jonathan Heidt

Now, after several years’ experience in some of those states, a few conclusions can be drawn about it. First, the program as a whole appears to be a bust. According to statistics the company provided to Sen. Claire McCaskill, D-Missouri, the vast majority of claims denials under the program have been reversed on appeal. The number of initial denials has fallen this year, too, after Anthem changed the rules to broaden the exemptions — that is, cases in which the ER claim would always be approved, no questions asked.

Perhaps most important, expecting consumers to diagnose their conditions as emergent or non-emergent before going to the ER is stupid and possibly illegal, insofar as it requires them to make judgments that ER doctors often can’t make without a professional examination. Using the ultimate diagnosis as a proxy for the urgency of the original visit to the ER is an imperfect standard to the point of being nonsensical, in medical terms.

“I’m a board-certified trained doctor of emergency medicine,” Jonathan Heidt, president of the Missouri chapter of the American College of Emergency Physicians, told me in January, “and I have trouble looking at the ER note and knowing what the patient was thinking at 3 o’clock in the morning.”

But the likelihood is that Anthem doesn’t actually want to deny members’ ER claims — what it really wants is for them not to go to the ER at all. Anthem’s policy is really just another hoop for consumers to jump through, which always translates into less usage. The drawback is that these obstacles result in less unnecessary medical care, but less necessary care too. More on that in a bit.

The statistics on claims denials and reversals come from a report McCaskill issued this summer, using Anthem data (even though the company stiff-armed her on some of her data requests). The conclusions about the wisdom of the policy come from an analysis by researchers at Yale and Harvard medical schools recently published by the Journal of the American Medical Assn. The researchers concluded that anthem’s system is so flawed it “could place many patients who reasonably seek ED [emergency department] care at risk of coverage denial.”

Anthem, the nation’s second biggest health insurer, says it’s standing by its policy, though it has no current plans to expand it to more states. The company told me by email that its “Emergency Department Review” was designed to “reduce the trend in recent years of inappropriate use of EDs for non-emergencies.” It said it found that about 5% of all claims it received for ER care were for non-emergencies, “which is in line with findings from the Centers for Disease Control and Prevention.”

The important questions, however, are what counts as a “non-emergency,” who makes the call, and when?

Anthem’s system is based on the diagnostic codes submitted by the ER with its claim — in other words, what the ER doctors ultimately judged the patient’s problem to be. In Indiana Anthem used a roster of 120 codes ranging from “abrasion” to “viral wart” and including various contusions and pain complaints; in Missouri, according to ER doctors, the list ran to more than 1,900 conditions. If the conditions appeared on the ER claim, Anthem would subject the claim to further review, with an eye to rejecting it.

Under Anthem’s original rules, the denial policy wouldn’t apply when the patient is 14 or younger, an urgent care clinic isn’t located within 15 miles, or the visit occurs on a Sunday or holiday. This year the company added several exclusions. Claims will always be paid if the patient was directed to visit the ER by a doctor; is traveling out of state; received any surgery, IV medications, or an MRI or CT scan at the ER.

The basic problem remains, however: A final diagnosis by an ER doctor isn’t very useful in judging what motivated a patient to report to the ER in the first place. That’s what the Harvard/Yale study found.

Patients aren’t diagnosticians. They don’t make decisions on whether to go to the ER based on a diagnosis, but based on their symptoms. And 90% of the symptoms that typically send a patient to the ER are common to both nonemergency conditions and potentially life-threatening emergencies.

Back or abdominal pain could be a muscle spasm — or herald a kidney stone or appendicitis; the researchers found that in their study sample of emergency cases from 2011 to 2015, abdominal pain resulted in hospital admission 16% of the time — but could result in Anthem denials in 4.3% of cases. Chest pain could be indigestion or a heart attack. Headache, vomiting, dizziness, cough and shortness of breath also could go either way.

In Anthem’s system, the researchers warned, “patients with acute illnesses are put in a difficult position of weighing the risk of delayed treatment for severe disease vs an uncovered medical bill.”

Anthem told me that “if a consumer reasonably believes that he or she is experiencing an emergency medical condition, then they should always call 911 or go to the ED.” But that’s just empty persiflage, if the consequence of guessing wrong is a bill for several thousand bucks.

The stakes are considerable. The Harvard/Yale study calculated that if Anthem’s policy were widely copied, nearly 1 in 6 ER visits by insured adults would result in a non-emergency diagnosis and be subject to denial.

McCaskill’s report traced the life cycle of Anthem ER claims denials and appeals in Kentucky, Georgia and her home state of Missouri. Her findings are eye-opening.

In July through December of last year, 5% of ER claims — 3,700 — were denied in Missouri, 4% (5,000) in Kentucky and 7% (3,500) in Georgia. Another 5% were initially denied but paid after appeals in Missouri, 7% in Kentucky and 13% in Georgia.

In fact, most denials were eventually overturned — and the rate of reversals rose almost every month into this year. In Missouri, the rate of reversals increased from 58% in July 2017 to 73% in November, a trend largely matched in the other two states.

Since January, when Anthem changed its standards, ER denials have plummeted — to zero in all three states by March 2018, McCaskill reports.

That points to the question of why Anthem’s program still exists at all. To begin with, it may well violate federal law, which requires insurers to cover ER services if a patient arrives with symptoms that a “prudent layperson” — one with an average knowledge of health and medicine — could reasonably expect to result in “serious impairment to his or her health.” Anthem says its physician reviews are aimed at matching ER diagnoses with the prudent layperson standard, which may explain why denials have plunged.

Even in 2017, when the company was still denying a sizable percentage of ER claims, the results appear to have fallen short of its expectations. McCaskill says Anthem had projected that its program would save $2.9 million a year via denials of unnecessary ER visits in Missouri alone; but in the last six months of 2017, the denials yielded only $1 million in savings through unpaid claims — not counting the reversals after appeal.

The statistics suggest that Anthem’s initiative might even have cost the company more than it saved. Thousands of ER claims had to be scrutinized by professionals before denials, examined again if they were appealed, and ultimately paid if the denials were reversed.

Some of these costs land on the shoulders of patients. Anthem customers face the tension of how to pay five-figure bills for ER visits they thought would be covered for a nominal co-pay, and weeks or months of lost work time or other inconveniences trying to challenge the decision. Emergency Departments face the uncertainty of getting reimbursed for their services.

Still, the gains Anthem may have garnered from its policy may not be obvious. If it succeeded in discouraging patients from presenting at the ER in the first place out of fear of a big bill, that wouldn’t show up in denial and appeal statistics. Anthem couldn’t be tagged for infringing the prudent layperson rule, because the decision to skip the ER would be made by its imprudent customers on their own, for their own reasons.

Perhaps it’s a bit unfair to criticize Anthem for trying to shift the costs of ER coverage to patients. After all, trying to avoid paying out on claims is what comes naturally to insurance companies. That’s what allowed Anthem to record a profit of $3.8 billion last year on revenue of $90 billion, and to pay its recently-retired chairman and CEO, Joe Swedish, nearly $50 million in 2015-2017.

Indulgent state regulators in five states have allowed Anthem to get away with this flagrantly anti-consumer practice. The blame belongs to them.

Keep up to date with Michael Hiltzik. Follow @hiltzikm on Twitter, see his Facebook page, or email

Return to Michael Hiltzik’s blog.