A clinician's view of laboratory utilization.
Approximately five billion laboratory tests a year are performed in the U.S., at an estimated cost of $11 billion. Since lab tests account for about 25 per cent of all inpatient charges, they are under close scrutiny as hospitals grasp for ways to cut spending under prospective payment.
Plans to curb excessive testing cannot proceed unilaterally from the laboratory or the hospital administration. They should be formulated in cooperation with the medical staff because clinicians are at the heart of the issue. I will offer a clinician's view here.
A study by Wertman et al a few years ago verified the diversity of purposes behind test requests: As Figure I shows, clinicians order tests to diagnose, monitor, and screen for diseases; to evaluate or confirm previously abnormal results; to develop prognoses; as part of house staff training to learn how useful different tests are; and for medicolegal reasons, to guard against possible malpractice action. But less apparent factors also influence test ordering, studies have shown.
Young physicians, for example, order more tests than older physicians do. That's due in part to current medical training, which places greater emphasis on laboratory investigation over the more traditional physical examination and history. Regardless of an individual physician's training, however, experience spells less testing over time. A practitioner who has seen hundreds of sore throat cases won't start thinking of a differential diagnosis and a series of cultures when presented with another such case, although an intern might.
Family practice physicians tend to request fewer tests than specialists. Specialists are geared to look for unusual diagnoses or rare diseases; this orients them toward laboratory workups.
Physicians who graduated at the top of their medical school classes don't order as many tests as those who ranked lower. These achievers may be more confident of their diagnostic skills and not need to rely on the laboratory so much.
Teaching hospitals, when controlled for case mix, register more tests per patient than nonteaching hospitals.
Studies repeatedly demonstrate that East Coast physicians out-order their colleagues on the West Coast. Perhaps the West Coast is more progressive or open to change in modes of operation. In any event, many clinicians do what their peers around them expect them to do.
In the office or clinic setting, members of small group practices tend to request fewer tests than those in large group practices, and physicians in health maintenance organizations exercise more ordering restraint than those practicing under fee-for-service arrangements. On the other hand, physicians who have laboratory instruments in their offices do not order more tests than physicians who have to send out their test work.
The point is that no single profile will encompass all clinicians who order tests excessively or prudently. Nor will a wuick, single approach eliminate overutilization of the clinical laboratory. Several factors besides clinicians' ordering habits contribute to the problem.
At times the laboratory itself is at fault. For example, continuous flow analyzers force clinicians to order whole batteries that don't always make sense for the cases in question. Then again, delays in reporting results may cause clinicians to reorder or to request additional tests. And laboratories frequently continue to use outmoded tests when newer, more effective assays are available.
That last factor cuts both ways, though. The technological explosion of recent years has spawned many new tests. Therapeutic drug monitoring is just one area that expanded rapidly. Monoclonal antibody methods are drawing great interest today. These developments and others act to increase laboratory utilization.
Patients' attitudes foster over-ordering. Many with health insurance regard medical care as a free right. To them, more is better: Studies show that patients are happier if their physicians order tests--even when it is explained that the diagnosis can be made without laboratory work.
Several years ago, an advertisement in the New York Times discussed "what every man should know about prostate cancer." It aimed at getting patients to encourage their doctors to order a commercial radioimmunoassay for prostatic cancer. Many tests were subsequently ordered--inappropriately, because as more data became available, analyses indicated that the procedure was not a good screening test for prostatic cancer. Similarly, a great popular interest in hypoglycemia has led many patients to prompt their physicians to order oral glucose tolerance tests, often without good clinical basis.
Hospitals, compelled to decrease length of stay, also exert pressure on clinicians. There's often a push to order more tests sooner and not to wait for results before asking for additional tests.
Lest this sound like an apology for clinicians, consider these additional points. Clinicians also overorder because: they don't know what laboratory tests cost; their peers, especially in training programs, will castigate them if they miss a test that might be relevant; they gain financially from markups on their office testing; they fear malpractice claims, as we have mentioned; they conduct trial and error testing out of inexperience; they misunderstand test utility, including such terms as sensitivity, specificity, and predictive value; and they simply don't care how many tests they request (Figure II).
Three kinds of theories have been advanced to explain the way physicians make decisions. The first two listed below are underpinnings for different attempts to remedy test abuse. The third theory leads to a dead end because it says in effect that test ordering patterns are preordained ta medical school and in training, before individuals go out into practice.
1. Cognitive theory. This view regards clinicians as processors of information. Their decisions are shaped by their beliefs about the effectiveness of testing and how well they understand such concepts as predictive value theory. Some can process large amounts of infomration better than others; some take better histories and perform better physicals than others, giving them more information to work with.
2. Stimulus-response theory. Here, physicians' decisions are seen as responses to specific environmental stimuli. Physicians are considered to be particularly sensitive to financial incentives, administrative orders, fear of malpractice charges, and patient reinforcement.
3. Sociological theory. Physicians belong to a social group, sharing its viewpoints and responding to its pressures. Their decisions are largely a function of when they graduated from medical school, their training site, and their medical specialty. This would mean that their basic behavior is ingrained before they start practicing.
Let's see how the first two theories translate into efforts to improve laboratory utilization.
* Educational techniques. If you believe in the cognitive theory, improving laboratory utilization is just a matter of educating clinicians through teaching conferences, individual instruction, or written materials. Offer them information on the proper use of testing, and they'll correct bad habits.
Unfortunately, relying on educational techniques alone often produces only short-lived success. Several prominent studies documented immediate, encouraging changes in ordering patterns that wore off six months to a year after instruction.
* Audit with feedback. Embracing the stimulus-response theory, this approach reviews laboratory misutilization with clinicians and compares their performance with "normal" standards. It has been tried numerous times, with some success. One program that worked was at Brigham Women's Hospital in Boston. Senior physicians conducted extensive audits of house staff test requests, and laboratory use decreased.
Elsewhere, a 1983 study concluded that audit with feedback was not economical. The expense of paying senior physicians to read charts and help modify younger physicians' ordering behavior outweighed the savings derived from reduced testing, and the program also lacked an enduring impact.
A 1973 study demonstrated that laboratory utilization could be cut by briefing clinicians on the costs of the tests they were ordering--a combination of education and audit with feedback. Many physicians have no idea what laboratory tests cost. Given two groups of clinicians, one supplied with information on test costs and the other group kept in the dark, the first group will use fewer laboratory tests per diagnosis.
* Financial incentives. Can the prospect of extra budget funds or more salary stimulate desired test ordering behavior? One hospital offered the house staff free subscriptions to medical journals if testing was reduced by a certain amount. This plan failed, as have other rewards, perhaps because they were not substantial enough. Some observers contend that the right incentives have not been tried yet.
Health maintenance organizations may enjoy more success with financial incentives because physicians can share in the extra income earned when testing is held down.
* Administrative changes. There's some strong sentiment against hospital administration efforts to change procedures for ordering or providing tests. These corrective measures have been described as "often odious, difficult to implement, ethically questionable, and costly to maintain." On the other hand, several of these overhauls have worked quite successfuly--more so, in fact, than any other approach.
The change can be as simple as a hospital's revision of the lab request form that physicians use to order tests. If you provide a form that enables clinicians to check off 10, 15, or even 20 tests, they are likely to order more tests than if the form required them to write out the name of each procedure.
Another hospital limited the number of tests that a clinician could order each day. A third created a policy whereby a senior house staff member had to approve test requests beforehand. All succeeded in paring laboratory utilization.
Why do many clinicians oppose administrative programs? One fear may be that changes will be made by the laboratory or administration without any say by clinicians. This exclusion was often the cause in the past and probably accounted for the ineffectiveness of many programs.
Administrative changes can be useful and effective, but only when the laboratory and physicians ordering the tests work closely together. In that atmosphere, changes that make sense to clinicians are more likely to result and win acceptance.
At Boston University Medical Center, other clinicians and I have worked with Dr. Bernard E. Statland, director of laboratory medicine, and the laboratory staff in developing administrative changes--most notably the test request intervener and modifier (TRIM) described by Dr. Statland in the February 1984 issue of MLO ("Prospective Payment: How Computers Can Help Us Cope").
The laboratory and the medical staff have been cooperating to formulate test ordering decision rules that make clinical sense. Computer programs will monitor test requests, flag unusual orders, and modify those that violate the decision rules.
Communication--not blame--is the key to upgrading laboratory utilization. The goals of clinicians and the laboratory are the same even if we come together from opposite directions.
|Printer friendly Cite/link Email Feedback|
|Author:||Moskowitz, Mark A.|
|Publication:||Medical Laboratory Observer|
|Date:||Jul 1, 1984|
|Previous Article:||A cooperative strategy for curbing test abuse.|
|Next Article:||Evaluating supervisory candidates by their meeting performance.|