Why doctors order unnecessary tests

I’m happy that this study is getting some play in the media. Essentially, many of the “routine” tests done on a physical are not recommended (the whole concept of a “routine physical” is controversial – but that’s for another discussion). Merenstein concludes:

# 37 percent of checkups included a urinalysis.

# 9 percent of checkups included an electrocardiogram.

# 8 percent of checkups included an X-ray.

# 43 percent of checkups included at least one of these three tests.

None of these studies are recommended screening tests by the USPSTF. I have written previously about what routine tests should be done.

So, why are they being ordered? There are several reasons.

1) Blame the media

The mainstream media continues to push a “more testing equals better medicine” philosophy. High-profile malpractice cases often highlight missed diagnoses due to a failure to test.

Publications like this article from Forbes continues to push evidence-bereft screening tests:

Among the battery of screening tests you could talk to your doctor about getting: a lipid profile that includes a C-reactive protein measurement, a colonoscopy if you’re over 50 (or earlier if you’re at high risk for colorectal cancer), a diabetes risk test (fasting plasma glucose test or the oral glucose tolerance test), a stress test and a skin cancer exam.

There is no evidence to support a screening CRP, stress test or even skin cancer screening.

2) “Defensive medicine”

This blog’s favorite buzzword. It seems that Merenstein has taken up a crusade against unnecessary testing after being burned as a resident. Good for him.

I will not start another discussion about this topic and have the contrarian folks come back with how little data there is that defensive medicine exists/decreases lawsuits/lowers malpractice premiums etc. Instead, I will speak only for myself. If there were no-fault malpractice, health courts, or even caps (which I think is the least-effective solution), I would order less diagnostic tests. Period.

3) Reimbursement

Reimbursement in the US is essentially fee-for-service. Physicians get paid a set amount from an insurance company or Medicare/Medicaid for every service that they do. With reimbursements declining and overhead increasing, ordering more tests generates more revenue for a practice.

Of course, this is a ridiculous system for which there is no easy solution. Capitation was tried and failed miserably. Pay-for-performance is an untested answer. Increasing reimbursement may relieve some of the financial pressures physicians have to keep a practice afloat – subsequently decreasing the need for unnecessary testing.

As long as there is a fee for every service performed, there will continue to be an underlying financial incentive to order tests. That’s the simple reality – interpret that as you wish.

4) “Consumer-driven healthcare”

Another buzzword. There is a push for patients, I mean “healthcare consumers”, to shoulder a bigger financial responsibility for their health. With the media publicizing non-evidence-based testing (see point 1), consumers will have a greater say in what tests are ordered. It is their right after all – since they are paying for the tests. Besides, who wouldn’t like an executive physical? Expect this trend to increase.

There you have it. But before asking for that routine stress echo, chest x-ray, CA-125 test, or urinalysis, consider what you’re getting yourself into:

“The patient has no symptoms and doesn’t smoke, but he gets a routine chest X-ray. If there is a small shadow, doctors are obligated to look further.”

“That X-ray becomes a CT scan. That may show a small little nodule. The next thing you know, the patient ends up with a cardiothoracic surgeon who wants a needle biopsy, or even an open ,” he says. “In a lot of these cases, he comes up with nothing, a benign nodule or something.”

Aside from the costs in time and the potential for unnecessary suffering, these procedures add up to big money. Merenstein’s modest estimate of the cost of just these three simple tests is $47 million to $194 million a year. And that doesn’t include the cost of follow-up tests.

View 30 Comments >