Laboratory quality is an overlooked area of clinical medicine, according to Douglas Beigel, president of COLA, a nonprofit organization that accredits medical laboratories in the U.S. and provides education and training on best practices.
Most of the lab tests done in physician offices and at the bedside, including popular, frequently performed tests like pregnancy tests and rapid strep tests, are considered “waived,” or exempt from government oversight. But that doesn't mean that they should be exempt from quality control, Mr. Beigel said.
He spoke to ACP Hospitalist about the importance of laboratory testing to clinical medicine and why physicians should take steps to improve the testing done in their offices and hospitals.
Q: What is the background on regulatory oversight of laboratory testing in the United States?
A: According to June 2013 CMS data, there are nearly 232,500 laboratories in the United States that presently fall under the scope of the Clinical Laboratory Improvement Amendments of 1988 (CLIA 88). About 35,000 laboratories are regulated, and in laboratory medicine-speak that means that they're performing what's called moderate- or high-complexity testing. But that means that the rest, more than 197,000, about 85% of laboratories in the United States, are fundamentally unregulated. Most fall under the CLIA category of waived testing, and “waived” means waived generally from federal regulations.
Q: How does laboratory medicine affect clinical care, especially in the current health care environment, with models such as the patient-centered medical home and accountable care organizations?
A: One key message that we're transmitting through our member organizations [including ACP] is that 70% of medical decisions are [made] as a result of laboratory tests. That is a key element of decision making that happens every day in medicine. The issue is that there should be some focus on quality, making sure that the test is done correctly, because that leads into diagnosis and treatment pretty quickly.
The other aspect is that what we see happening in patient-centered medical homes around laboratory medicine has already happened in several health care delivery models. What they are doing is “Do the test once; use it many times.” If you make the major assumption that laboratory tests are done well and they produce quality results, then it makes sense to send the data used by a primary care clinician to make an initial diagnosis to a specialist if there's a referral, or to the hospital if hospitalization is going to be required. That data's used to then either validate the initial impression or to build upon the initial diagnosis.
With the patient-centered medical home and the Affordable Care Act, we believe that the concept will be to do the test first on a very high-quality basis and then reduce redundant testing so that those test results then are used as the patient moves through the system.
However, most physician practices are using only waived testing. In the waived test environment, the methodology or the instrument used is very good. It's in the preanalytic and postanalytic phases—for example, human errors that occur while the test sample is being gathered, or after the test has been run—that industry-wide, no matter what the test classification is, there are issues involved with quality. Most laboratories in the United States now are using waived testing devices that have pre- analytic and postanalytic problems that could be resolved fairly easily, but basically without oversight, what we're seeing is that they're not. These sites are generating problem data that are being used in the diagnosis and treatment of patients.
For example, under a patient-centered medical home, testing information from a retail clinic will be fed into a medical record at some point and be the responsibility of a practicing physician to coordinate care. Unless there is some focus on quality in laboratory medicine, our belief here at COLA is that that will lead to unintended outcomes of either redundant testing that may not be reimbursed or possibly data in the medical record that would lead to misinterpretation, misdiagnosis and possibly inappropriate treatment.
Q: What can be done to encourage more oversight in primary care practices?
A: What we're attempting to do is communicate the importance of laboratory medicine, that 70% of the patient management decisions are based on laboratory tests. In the minds of a lot of primary care physicians, they do it every day, they just don't realize [its significance] to the practice of medicine. What we say to some of the primary care physicians that we talk to is “What would your practice be like if you didn't have a laboratory?” And that all of a sudden helps them change their mindset.
We are working with payers to introduce quality initiatives into laboratory medicine, so that physicians could be eligible for up to a 30% increase in reimbursement if they participate in some of the quality initiatives out there. Part of what we've done working with primary care physicians is to create a [training] program that should be easy to use but has a high impact making sure people who are doing the testing are competent and stay on top of preanalytic and postanalytic issues.
What we've done is try to take the pain out of the process. We've worked with ACP, we've worked with the American Academy of Family Physicians (AAFP) and the American Medical Association (AMA) to come up with a solution that we think is just pretty simple and could actually be targeted to the people doing the test, so that they're aware of what they need to do, how to read a package insert, what are those basic laboratory principles that they need to follow. (See sidebar, below, for more on available laboratory training resources.)
Q: You've said that CMS is investigating potential problems with waived testing that could affect patient care. If waived testing begins to be regulated, what could be the impact on a primary care physician's practice?
A: That's what we're concerned about. We think close-to-patient testing—including, for example, tests for pregnancy, strep, blood glucose and flu—is the best type of laboratory testing that can be done. The best case is that laboratory testing's done when you have an encounter with a physician, so that the physician can use those test results as part of their physical examination with the patient. They have the data, they see the patient at the same time, so they understand what conditions are being presented to them.
Fewer opportunities for close-to-patient testing can translate into increased opportunities for errors. What happens when you don't have close-to-patient testing is you have an encounter with the physician, they order tests, the patient goes off site somewhere to a collection station, and the sample goes off to a centralized lab. The lab reports the data back, and then it needs to be reviewed by the physician to make sure that the original diagnosis and treatment plan are consistent with the lab results. Sometimes the lab results don't match the patient encounter information that the physician has, and then they need to be rerun again.
So you're seeing this time delay and you're seeing a disconnect from what I think is a valuable patient-physician encounter. You see that dissolving away, if physicians say, “I'm not going to do any testing at my office anymore, I'm just going to send all the tests out and then I'll handle the results post-testing.” Bottom line is we're concerned that physicians would stop doing testing close to patient care. From our perspective, that would not be good for quality care initiatives.
Q: How does this issue affect hospitals and hospitalists?
A: The concept of doing the test once and using it many times is definitely going to be involved with hospitalists. Again, it's the continuity of care, it's the idea of making sure of what the initial diagnosis and treatment plan was and how's that going to be followed up if the patient is admitted to the hospital.
Not only are hospitalists getting results that are sent to them via the community or other programs or other patient-physician encounters, but they also are prescribing close-to-bedside testing.
The biggest problem with point-of-care testing in hospitals is that in some ways no one owns it. Close-to-patient testing or what's called point-of-care testing in hospitals can be done by literally anyone: nurses, doctors. One of the big issues that they've found in hospitals is that it's so available that sometimes the data doesn't end up in the medical record but is used to prescribe treatment. And it's been a problem forever. All this testing is being done, it's being charged, but it doesn't end up in the medical record.
Q: What can individual physicians do to improve the situation in their hospitals or primary care offices?
A: I would say again understanding the role of laboratory medicine … and I would say that physicians should participate in competency training on basic laboratory medicine. COLA was founded because primary care physicians are not trained in laboratory medicine. They don't receive it in medical school, and they don't receive training in residency programs. What it always comes down to is that there's some very easy, fundamental steps to take in [improving] waived testing, but people just need to literally attend to these fundamental steps. [They need] to make sure that they're going to be producing quality lab results not only for themselves, but for other people who've got to use those results.