Portal de Periódicos da CAPES

Liability and Payment Issues in the Selection of Pathology Assays

2002; American Medical Association; Volume: 126; Issue: 6 Linguagem: Inglês

10.5858/2002-126-0652-lapiit

1543-2165

Jack R. Bierig,

Medical Malpractice and Liability Issues

In an ideal world, the only considerations that a pathologist would have to take into account in selecting a diagnostic methodology would be the effectiveness of the methodology and its costs. The pathologist would review the literature on various alternative approaches to testing for a particular analyte and would determine which of these alternatives produced clinically reliable results most frequently. The pathologist would also take into account for each competing approach whether there was an efficient way to do confirmatory testing to check for false-positives or false-negatives. Finally, the pathologist would factor in the relative costs of various approaches in light of the difference in the reliability of their results. Based on this evaluation, the pathologist would choose the particular methodology that he or she deemed best for patients. No other factors would enter into the equation.But we do not live in an ideal world. In the real, nonplatonic world, a pathologist faced with a choice of methodologies for testing for a particular analyte may want to factor in several additional considerations. Among these factors are the following: This article explores how these nonscientific considerations can affect the pathologist's choice among several possible methods for testing for a particular analyte.One would hope, of course, that a pathologist who made a reasoned, good faith judgment about which procedure was best scientifically would not face any malpractice exposure at all. At a minimum, to the extent that the pathologist faced such exposure, the malpractice coverage that the pathologist had purchased would cover all costs and any liability associated therewith. Furthermore, having provided an important service in a professional manner, the pathologist would, of course, be paid for the service.In our world, however, the congruence between responsible scientific decision making and reward to the pathologist is sometimes less than perfect. A pathologist who has made a reasoned choice of analytical methodology may nonetheless face liability, insurance coverage, and payment issues. This unfortunate consequence derives from the fact that, in this country, a pathologist's selection of test methodologies is subject to 3 different systems: Regrettably, these systems do not always operate in perfect 3-part harmony. The dissonance between these systems can expose well-intentioned pathologists to disconcerting problems.Liability issues can arise from the disjunction between the federal system for approving test kits and the state system for determining medical negligence. The federal system for approving test kits is administered by the Food and Drug Administration (FDA), enforcing the Food, Drug, and Cosmetic Act. Under that Act, test kits are generally regulated as medical devices. As such, they may not lawfully be marketed in the United States unless the manufacturer has made 1 of 2 showings to the FDA. Under one approach, the manufacturer may demonstrate to the satisfaction of the FDA that a device is safe and effective for its intended use. This approach is known as the premarket approval, or PMA, process.Alternatively, the manufacturer may show that a device is substantially equivalent to an existing device already on the market. This approach is named after the provision of the Food and Drug Act in which it is described (§510[k] of the Act). It is known as the 510(k) process. So, there are 2 processes for obtaining FDA preclearance for a medical device: the PMA process and the 510(k) process.With certain exceptions, the Food and Drug Act regulates only manufacturers and distributors of foods, drugs, medical devices, and other covered products. The Act generally does not regulate physicians or other health care professionals who use these products, or those who make their own products but do not distribute them. The actions of these physicians and other professionals are subject only to state malpractice and similar law.This point has 2 important implications for purposes of this article. First, a manufacturer of a medical device approved by the FDA for a particular purpose may market that device only for the approved purpose and may make no claims that the device will be safe or effective for other uses. However, physicians may conclude, based on their review of articles in medical journals or on their own observations, that a device marketed exclusively for one purpose may be safe and effective for another purpose. Nothing in the Food and Drug Act prevents the physician from using the device for such other purposes. Use of a drug or device for a purpose not approved by FDA (and therefore not described in the labeling) is known as an "off-label" use. Off-label uses are perfectly lawful as a matter of food and drug law.The choice of an assay to measure the HER-2/neu status of breast tumors is illustrative. The FDA has approved the Dako Hercep Test and the Ventana Pathway kit as immunohistochemical assays for this purpose. It has also approved the Vysis Path Vysion fluorescence in situ hybridization (FISH) assay for determining HER-2/neu gene amplification. As of the writing of this article, these 3 assays were the only ones to have been approved by FDA for predicting the efficacy of trastuzumab (Herceptin) therapy. Nevertheless, as far as the FDA is concerned, a pathologist is free to use other assays even though the manufacturers of such assays may not lawfully promote them for that purpose.The second implication is that pathologists may, without running afoul of the Food and Drug Act, create and use monoclonal antibodies or other testing materials in their own laboratories, as long as they do not distribute these materials outside the laboratory. Thus, a pathologist can produce a home-brew assay to test for HER-2/neu overexpression in his or her own laboratory, provided that the home brew stays at home. This practice, too, is perfectly lawful under the Food and Drug Act.Significantly, however, the fact that conduct is permitted under the Food and Drug Act does not mean that such conduct is immune from challenge under state liability law. It is true that §521(a) of the Food and Drug Act provides that no state "may establish or continue in effect with respect to a device intended for human use any requirement which is different from, or in addition to, any requirement applicable under the chapter to the device." This provision may preempt a product liability action against the manufacturer of a device that has been approved pursuant to the PMA process. However, it will not protect a pathologist in a malpractice action, and it certainly will not protect the pathologist who makes an off-label use. Although off-label uses of devices are permitted under the Food and Drug Act, there is no "requirement" to make such a use. Therefore, §521(a) is not applicable.The product liability consequences of selection of a testing methodology can be seen in the context of HER-2/neu testing. Specifically, different considerations are presented depending on whether the pathologist (a) uses an FDA-approved test as directed by the manufacturer, (b) uses a commercially available assay that was approved by FDA for another purpose, (c) uses a home brew, or (d) uses an FDA-approved test, but in diluted form. Obviously, if all goes well, there can be no liability. If a false-negative is reported, however, the patient may not get the trastuzumab therapy she needs and may consequently suffer injury or death. Alternatively, if a false-positive is reported, the patient will receive trastuzumab therapy that she does not need—therapy which can have very toxic effects.If a patient is injured by a false-positive or a false-negative result arising out of use of an FDA-approved assay in accordance with the manufacturer's instructions, the choice of assay cannot be faulted. However, what if the pathologist has made an off-label use of an assay approved by the FDA for another purpose?It does not take great insight to recognize that the plaintiff's malpractice lawyer will make 2 basic claims: In response, the pathologist will point out that the off-label use was not prohibited by the Food and Drug Act. However, that response is not going to carry the day for the pathologist. The question is not whether the selection of assays violated the Food and Drug Act, but whether it was negligent under state malpractice law.A federal court in Pennsylvania summed up the situation in the context of the pedicle screw litigation, a large group of cases in which orthopedic surgeons were sued for making off-label uses in spine surgery of a device known as the pedicle screw:[A] physician is free to use a medical device for an off-label purpose, if, in the physician's medical judgment, he or she believes that use of the device will benefit the patient. Because the off-label use of a medical device is a matter of medical judgment, a physician may be subject to medical malpractice liability for the exercise of that judgment (Orthopedic Bone Screw Products Liability Litigation, 1996 WL 107556 at *5 [ED Pa 1996]).Transposing to the HER-2/neu context, a pathologist is free to use a non-FDA approved assay, but the pathologist may be subject to medical malpractice liability for choosing that assay.The malpractice issue comes down to whether the off-label use meets the standard of care. What is the standard of care? It is that practice that the court concludes a reasonable physician would have employed in the circumstances. Conduct by a physician that does not meet the standard of care is said to be malpractice.How is the standard of care determined? It is determined by the testimony of expert witnesses on what constitutes appropriate practice in a given situation. This testimony is based on customary practice in similar situations, medical textbooks, articles in peer-reviewed medical journals, practice guidelines of professional societies, and any other relevant and reliable sources of information.A pathologist who determines to make an off-label use of an FDA-approved kit or to use a home brew should be able to prevail by demonstrating that the choice of methodology was medically reasonable. Let us begin with the pathologist who makes off-label use of an assay approved by the FDA for another purpose. On this issue, the Investigational Devices Exemption Manual1 published by the FDA in June 1992 is quite relevant.That manual provides, in pertinent part, as follows:This statement has been cited by at least 2 courts to support the conclusion that a physician acted consistently with the standard of care, even though the physician made an off-label use of a medical device (see Staudt v Froedtert Memorial Lutheran Hospital, 580 NW2d 361, 363 [Wis App 1998]; Femrite v Abbott Northwestern Hospital, 568 NW2d 535, 542 [Minn App 1997]). The lesson of these cases is that pathologists who are using an unapproved assay to test for the efficacy of trastuzumab therapy should make a record as to why they chose that assay in preference to a test approved by FDA for that purpose, and what results were obtained using the unapproved assay. This record could include citations to relevant publications, reference to data generated in the laboratory, and other documentation supporting selection of the methodology.What about the pathologist who uses a home brew? On this question, there do not appear to be any cases directly on point. However, principles similar to those that underlie the statement in the FDA's Investigational Devices Exemption Manual should apply in this context as well. Specifically, a pathologist who chooses to test using a home-brew assay should be well informed about the assay and should make a contemporaneous record of the medical and scientific rationale for using that assay and of the results obtained with such use.What about the pathologist who dilutes the materials in an FDA-approved kit in order to make those materials go farther? The first case to address a somewhat analogous issue was the 1970 decision of the Supreme Court of Minnesota in Mulder v Parke Davis, 181 NW2d 882 (Minn 1970). In that case, a patient died of gastrointestinal hemorrhage due to aplastic anemia or bone marrow depression after her physician prescribed the drug chloramphenicol (Chloromycetin) in a dosage different from that recommended by the manufacturer. The Court ruled as follows: In other words, departure from dosage levels recommended by the manufacturer does not necessarily mean failure to meet the standard of care, but it does require the physician to provide a cogent explanation for the decision to depart from the recommended dosage. It will then generally be for the jury to determine whether the physician justified the departure from the manufacturer's recommendation.Three years later, in Ohligschlager v Proctor Community Hospital, 303 NE2d 392 (Ill 1973), a patient was injured after her physician administered the drug Sparine at 2 and 3 times the dose recommended by the manufacturer. The Supreme Court of Illinois held that this deviation from the manufacturer's instructions constituted prima facie evidence of negligence. Taken together, the Mulder and Ohligschlager cases suggest that if a pathologist chooses to dilute the amount of FDA-approved test materials used in a HER-2/neu assay, that pathologist should, at a minimum, make a contemporaneous record of why the dilution is consistent with good diagnostic practice.Significantly, a pathologist who decides to dilute for economic reasons may have a difficult time persuading the trier of fact that the dilution accords with sound practice. The plaintiff's lawyer will assert that the pathologist subordinated the patient's health to the pathologist's economic benefit. The pathologist may be able to refute this argument, but the risk does not seem worth the benefit.Let us assume that the pathologist has documented the reasons that the assay selection represents a reasonable medical judgment in the interests of the patient. Does the pathologist also have to obtain the informed consent of the patient before making an off-label use, or is the choice of a diagnostic methodology within the reasonable discretion of the pathologist, without having to involve the patient in the process?There appears to be no case squarely addressing the issue of informed consent with regard to selection of an off-label diagnostic methodology. However, cases in analogous contexts suggest that the answer may not be entirely uniform from state to state. The majority view seems to be that physicians need not disclose the FDA status of a device. The theory behind this view is that the informed consent process focuses on risks of a proposed procedure and that the FDA status of a device is not a risk.Applying this view, a Florida appellate court, for example, has held that an orthopedic surgeon who implanted a pedicle screw in a plaintiff's spine was under no obligation to inform the patient that the pedicle screw was not approved by the FDA for spine surgery (see Alvarez v Smith, 714 So2d 652 [Fla App 1998]). The court concluded that "FDA status is not a 'medical risk' and thus need not be disclosed in obtaining a patient's informed consent" (Id. at 655).A different approach was taken, however, by a New York trial court in Retkwa v Orentreich, 584 NYS2d 710 (Sup Ct 1992). In that case, a physician injected a woman with liquid-injectable silicone without first disclosing to the woman that such silicone had not been approved by the FDA. The court in that case viewed the informed consent process as involving more than a disclosure of medical risks. Rather, the process was said to encompass the provision of all information that a patient might reasonably want in deciding whether to have a procedure. Applying this approach to informed consent, the court held as follows: To be sure, the New York case deals with a product that was never approved by the FDA for any purpose, whereas the Florida case and similar cases dealt with off-label use of an FDA-approved product. Nevertheless, the 2 cases may represent different concepts of what must be included in the informed consent process.Thus, in general, whether a physician needs to obtain the prior consent of a patient before making an off-label use of a drug or device appears to depend on the jurisdiction in which the physician is sued. In most states, obtaining such consent does not seem to be required. However, in New York and in some other states, it may be required.To be sure, the practice of pathology is different from the practice of internal medicine or surgery in that the pathologist usually does not have direct contact with the patient. This distinction supports the position that a pathologist should not be required to get the consent of the patient before adopting a particular test methodology in preference to an FDA-approved alternative. Nevertheless, in light of decisions such as the New York case involving injectable silicone, a pathologist who does not use an FDA-approved test as prescribed by the manufacturer but who wants to be on the safe side may want to consider obtaining the informed consent of the patient before proceeding.Of course, it is somewhat impractical for a pathologist to seek the informed consent of the patient. After all, the pathologist does not see the patient. The pathologist can, however, develop a short document describing the methodology to be used, outlining the reasons for the decision to employ that methodology in preference to the FDA-approved test and inviting the patient to contact the pathologist if she has any questions. The pathologist can ask the oncologist to give this document to the patient and to be willing to discuss it with the patient. The process may be somewhat cumbersome, but it could avoid problems down the road.In summary, from a liability perspective, 3 points should be considered in the selection of an assay: These 3 laws of assay selection dynamics are hardly Newtonian in vision, but they can be important to remember.If there is anything worse than being sued for malpractice, it is being sued for malpractice and discovering that the suit is not covered by your malpractice policy. Thus, in selecting an assay, the pathologist will want to take into account the potential impact of the selection on malpractice coverage.The issue is brought into focus by a clause from a standard malpractice insurance policy from Miller's Standard Insurance Policies Annotated2:Of course, this particular clause applies only to drugs and not to medical devices, such as diagnostic assays. Thus, even if it were in a pathologist's malpractice policy, it would not rule out coverage for a claim arising out of a non–FDA-approved assay.Nevertheless, the clause brings up an important point: Insurance companies are free to disclaim coverage for whatever risk they choose not to accept. One of the risks that they may disclaim is coverage of claims arising out of use of non–FDA-approved devices, or even of off-label uses of FDA-approved devices. Indeed, after the financial beating that insurance companies took defending cases arising out of off-label uses of pedicle screws, it would not be surprising if clauses of this nature were to become more common.The moral of the story is clear: If a pathologist is going to use any assay for purposes of predicting the efficacy of trastuzumab therapy (or for any other purpose) other than a test approved by the FDA for these purposes and in accordance with the manufacturer's instructions, the pathologist should review the applicable malpractice policy to make sure that the policy will cover claims arising out of the assay that is used. It is important to read the relevant clauses carefully. A pathologist who has any questions about coverage should contact his or her insurance agent to clarify the situation.Consider the following clause:Needless to say, such a clause may not appear in the malpractice policy of a particular pathologist. However, let us consider its potential effect in the context of selecting an assay for HER-2/neu testing.Initially, a home brew assay will not have been approved by the FDA and will not have been the subject of a §510(k) proceeding. Thus, if a pathologist gets sued based on an error arising out of use of a home brew, that pathologist may not be covered under this policy unless the prior approval of the company has been obtained. Thus, the pathologist should do 1 of 3 things: The essential antecedent point is to examine the malpractice policy to see if a similar provision is included.What about off-label use of a kit approved for other purposes? The pathologist would have a good argument that this use is covered, but the pathologist may want to clarify the issue in advance. Once again, the key point is to be aware of whether there is a provision governing off-label uses.What about using an FDA-approved test but in diluted form? Under the clause in question, this approach will probably be covered since the test has been approved by FDA for the specific purpose. However, the outcome may depend on very specific verbiage. It behooves the pathologist to know what that verbiage is.In summary, pathologists who are going to use an assay in preference to one approved by FDA for the relevant purpose and in acccordance with the manufacturer's instructions should satisfy themselves that any claims arising out of the use will be covered by their malpractice insurance policy. They should read the policy carefully. If they have any doubts, clarification should be sought.Finally, what about the effect of a non–FDA-approved use on payment? Remember that 3 different systems impact the choice of methodology: the FDA clearance system, the state liability adjudication system, and individual payment systems. This part of this article addresses the intersection of the first and third systems.In the individual payment systems, every payer determines those products and procedures for which it will pay and those for which it will not pay. Thus, for example, many indemnity insurers will not pay for elective cosmetic surgeries. While each payer establishes its own coverage policies, virtually all payers refuse to pay for drugs, devices, or procedures that they deem "not medically necessary" or those denominated as "experimental."The Medicare program is illustrative. Section 1862(b)(1) of the Medicare Act, 42 USC §1395y(b)(1), provides that the Medicare program will not pay for items or services that are "not reasonable and necessary for the diagnosis or treatment of illness or injury." Every private payer has a similar provision.How is it decided whether a particular product is "medically necessary" or whether it is "experimental"? One would anticipate that a device cleared by the FDA for marketing would not be deemed experimental. Somewhat surprisingly, however, that is not the case. Each payer, including the Medicare program, makes its own independent determination regarding whether a product or procedure is experimental. The FDA clearance is a relevant fact, but it is not dispositive.An interesting example occurred on June 29, 2001, with respect to positron emission tomography (PET) scans performed with a device known as a gamma camera. The FDA had found that a PET system with gamma cameras was substantially similar to a preexisting device and therefore approved its distribution pursuant to §510(k) of the Food and Drug Act. However, the Centers for Medicare and Medicaid Services, the subagency formerly known as the Health Care Financing Administration, determined not to pay for certain uses of the gamma camera. The Centers for Medicare and Medicaid Services concluded that the FDA clearance of the camera under §510(k) "does not guarantee a standard of clinical performance that is linked to clinical decision-making on patient outcomes, and does not provide an adequate basis to determine that use of PET is medically beneficial for patients with specific malignancies." Thus, even FDA approval of a device under the §510(k) process does not guarantee payment by Medicare. And if Medicare does not pay, private insurers are not likely to pay either.Let us consider this issue in the context of HER-2/neu testing. A payer is highly unlikely to deny payment for any test that has been found by the FDA to be safe and effective in predicting the efficacy of trastuzumab therapy in a PMA proceeding. Even use of the test material in diluted form is unlikely ever to get to the attention of the payer.But what about off-label use of an assay or use of a home brew? The former has not been approved by the FDA for predicting the efficacy of trastuzumab therapy, while the latter has not been approved at all. Once again, the pathologist's choice of assay is unlikely to come to the attention of a payer. Even if it does, the payer may well be indifferent to the choice of assay. These methodologies are not nearly as costly to a payer as a gamma camera.It is possible that the payer might take the position that use of the non–FDA-approved methodology is experimental. It is also possible that a payer will refuse to pay for the actual trastuzumab therapy unless the FDA-approved test has been used to predict the efficacy of such therapy. Indeed, the Medicare carrier in some states may have taken this position.Unlike the malpractice coverage issue, this issue should not be explored in advance. After all, if a payer declines to pay on grounds that an assay is experimental, the worst that is likely to happen is that it will not pay for the most recently performed procedures and for any future procedures. In these circumstances, it does not seem worth the potential downside to raise the issue prospectively. If, however, 1 or more payers start rejecting claims for payment for tests performed making off-label use of an FDA-approved kit or a home brew, or refuse to pay for subsequent trastuzumab therapy, the pathologist may want to take the issue up with the payer.In this imperfect world, a number of considerations beyond the medical and scientific merits of various methodologies might influence the pathologist's selection of an assay. Use of an FDA-approved test in undiluted form poses the least risk. However, altruistic and cost-conscious pathologists may conclude that another approach is scientifically preferable or equally efficacious but at a lower cost.Pathologists who reach that conclusion should proceed with their methodology of choice. But they should also keep in mind this 6-part prescription:By following this recipe, the pathologist will have gone a long way toward bringing our imperfect world and the ideal platonic world of assay selection closer together.