Are your lab costs high? Here’s how to cut them

Try these suggestions for tighter controls

(Editor’s note: This is the first part of a two-part series on ways to improve the cost-effectiveness of your practice’s laboratory services. Part two will run in next month’s Physician’s Managed Care Report.)

After a brief conversation with your lab manager, you might remember why you haven’t looked at lab costs too closely lately. The technical nature of laboratory management can quickly make the prospect of analyzing costs seem daunting at best. Yet, as practices rely increasingly on fixed monthly payments via capitation, keeping all costs in line is critical.

Also, it may be tempting to farm out your lab services, but an on-site lab may be a major reason patients stick with your practice, so they can avoid the inconvenience of making yet another stop for their lab work.

That’s the view of Ruth A. Lander, practice administrator for Columbus (OH) Oncology Associates. Even though the practice is small (six physicians), the group is dedicated to keeping its lab on site and keeping those costs in line.

"It’s for the sake of our patients," Lander says. Sending patients to another site would take too much time and it would drastically damage their relationships to patients, she says. Some years ago CIGNA pressured the group to farm out its lab work, but the practice stood firm and agreed to eat the costs if it had to. As it turns out, lab work is not an unprofitable part of the practice, she says.

The good news is that some rather simple changes can make a big difference in your lab costs and help your practice continue to provide optimal care so it’s convenient and cost-effective for patients and providers, says Alan H.B. Wu, PhD. Wu is director of the department of pathology and laboratory medicine at Hartford (CT) Hospital and professor of laboratory medicine at the University of Connecticut in Farmington.

Getting rid of obsolete tests and cutting back unnecessary tests are two ways to start making your lab more cost-efficient, Wu tells Physician’s Managed Care Report. He recently presented specifics on how to make cost-effective changes in light of managed care to the Connecticut Primary Care Association Conference in Westbrook, and he published them in the Connecticut Medical Association’s monthly journal.1

Particularly in primary care, there are ways physicians can reduce ordering of tests and still obtain the data needed to make effective therapeutic decisions, says Wu. This month, Wu summarizes four of his eight recommendations (see next month’s issue for the next four recommendations):

1. Eliminate panels of tests, such as the general chemistry profile. Multiple testing became prevalent with the development of automated "continuous flow" chemistry analyzers. For many of these instruments, such as the Technicon sequential multiple analyzer-continuous flow the cost of running multiple tests was the same as for one test. This is because regardless of the number of tests ordered, reagents were continuously fed through the analyzer, Wu says.

Given that kind of production system, the higher the volume, the lower the costs, particularly if tests are paid on a fee-for-service basis. Modern clinical instrumentation, however, operates under a "random access" method, which means any combination of tests can be assayed on a given specimen. For that reason, technically there is an additional although minor lab cost for testing and reporting results of the extra unnecessary tests ordered on a profile. As a result, Medicare stopped reimbursing general multiple test panels that are not organ-specific as of March 1, 1996.

Watch out for false positives

Regardless of those slight costs, however, major costs can mount up quickly when you start running a wide range of tests. You have to strike a balance between screening for new, previously unknown cases found (true positives), and the number of false indications of disease, or false positives.

The more tests are ordered, the higher the statistical frequency that one of the tests will be abnormal in a subject who has no disease, says Wu. For a 25-test profile, there is only a 28% probability that a healthy person will have all tests within the normal range.

Each positive result requires a follow-up exam and confirmation tests that can greatly increase costs and cause stress and inconvenience to the patient, Wu says. Many studies show that the majority of abnormal chemistry results are caused by hyperlipidemia, diabetes mellitus, and liver disease — many of which are identified before the test in question. As practices become more proficient at identifying high-risk patients, these are the populations most critical for broader testing, he suggests. Also, intra-individual ranges or trends may need to be monitored and recorded; not all the ranges are appropriate for all individuals.

In Lander’s practice, most panel tests are limited to a set of 12 tests that the doctors deem as critical, she says. "We dropped down to 12 after the Medicare policy, but we are sticking with the 12 we think are very important."

2. Remove antiquated tests or those that provide redundant information. (See chart above.) The lab diagnostic industry continues to develop new tests, which sometimes make older ones obsolete. The chart above offers a list of tests Wu views as old, along with the newer, preferred alternative and the disease indication for each test. Checking on the status of these new vs. old tests could significantly improve your lab’s efficiency, he says.

In some cases, the newer tests can be more expensive up front (such as ferritin for microcytic anemia), but less costly in the long run when the misdiagnosis or inappropriate use of follow-up tests that can occur with older tests (such as serum iron) are taken into account.

Lander maintains that antiquated or redundant testing would be more prevalent in large practices than in small ones.

3. Use reflex testing algorithms. The term "reflex testing" refers to ordering or canceling one test based on the result of another. The use of established reflex patterns, agreed upon by the physician and the lab staff, can make testing more effective and cost-efficient, says Wu. In effect, rather than testing for everything — or large profiles — you test based on a chain of test events.

For example, a negative result for hepatitis B core antibody makes it unnecessary to order the IgM core antibody test, as the table illustrates. A Western blot confirms a positive ELISA screen for HIV. Immunofixation electrophoresis is only necessary if there is a monoclonal band on a serum protein electrophoresis gel.

In some cases, you’ll need a more complex algorithm than the chart offers, he says. For example, the National Cholesterol Education Program has established an algorithm based on total cholesterol. The algorithm guides physicians to the appropriate follow-up tests, such as the lipid profile. Use of new lipid markers may eventually become a part of this algorithm. Also, a new algorithm for thyroid disease is now under consideration.

4. Review your use of tests for illicit drugs/ drug abuse and be sure you’re using them judiciously. Rather than ordering a broad panel that includes all drug classes, a specific panel of either stimulants or depressants should be ordered to match the clinical presentation.

A knowledge of prevalent illicit drugs in use in your locality can narrow down your testing ranges. For example, some communities may be well-known for crack use, while others are known for designer drugs. Phencyclidine, methaqualone, and amphetamines are drugs that are rarely abused in Connecticut. Unless there are specific clinical indications that suggest otherwise, testing for them would not be cost-effective. Hospital medical records and public hearing surveillance data can help with tapping into drug use patterns.

Marijuana is widely used across the country, but it rarely produces acute clinical problems, says Wu. But cocaine, opiates, benzodiazepines, barbiturates, and tricyclic antidepressants are key drug classes and should be tested for when patients show symptoms of their use, Wu says.


1. Wu AH. Reducing the inappropriate utilization of clinical laboratory tests. Conn Med 1997; 61:15-21.