The American Law Institute (ALI), after a debate spanning multiple annual meetings, voted on May 20, 2024, to recognize, as part of the Restatement (Third) of Torts, “medical monitoring,” either as a separate tort or as a possible remedy for an existing cause of action. Medical monitoring has been adopted in a number of states as a way of addressing, absent present injury, a significant increase in the risk of developing a disease or ailment following tortious exposure to a toxic substance, radiation, or other impact. The tortfeasor is obligated to pay for the monitoring in the form of periodic medical examinations and tests intended to diagnose those specific medical conditions associated with the improper release or impact. The goal is to detect those conditions earlier than would otherwise be the case thereby improving a patient’s prognosis.
The ALI Reporters have done a masterful job of synthesizing a divergent body of case law into an elegant cause of action that balances opposing considerations. Thus, a tortious act that causes a significant increase in risk of developing an ailment is not sufficient to satisfy the elements of the New Tort; the risk, after having been increased, must also be so substantial (1) that it warrants medical monitoring, (2) that the monitoring would be beneficial to those who would go on to develop the ailment, and (3) that monitoring or screening would not have been medically indicated absent the increased risk. Nonetheless, the New Tort is controversial and, as this Article points out, there are issues that remain to be addressed and resolved.
To now, the controversy has been largely conceptual—can a court properly label behavior as tortious where there has been no present or imminent physical injury, only an increased possibility of developing one in the future? Because most jurisdictions that have adopted medical monitoring do not require actual or imminent physical injury, monitoring claims are viewed, by its proponents, as “non-traditional torts.”
There is also vibrant disagreement over the number of states that have adopted the tort. The ALI Reporters note that the nation (including the District of Columbia) is divided almost equally: seventeen states have adopted or appear to have adopted the tort; nineteen have declined or appear to have declined to recognize the tort; and fifteen have yet to consider the issue or it is unclear where they stand. Others claim that only fourteen states have adopted the tort and twenty-eight states have rejected it. Many courts have adopted the tort in one form or another as a way of enabling those without health insurance or financial means to participate in a medical monitoring program if the risk of developing a particular ailment has been significantly increased to a substantial level by the defendant’s tortious conduct. Other courts have rejected the tort primarily on conceptual grounds, namely that “in a negligence action, an increased risk of harm is not an injury.” Another reason given is that “dispensing with the physical injury requirement could permit ‘tens of millions’ of potential plaintiffs to recover monitoring costs, effectively flooding the courts while concomitantly depleting the purported tortfeasor's resources for those who have actually sustained damage.”
This Article takes no position with respect to the conceptual propriety of the New Tort or which of the competing categorizations of state law is more accurate other than noting that the dispute over the “state count” is a distraction. Instead, this Article focuses on four different issues raised by the New Tort, as adopted by ALI.
First, this Article examines whether the New Tort is consistent with current scientific thought. The New Tort’s criteria for assessing whether monitoring is appropriate appear to focus exclusively on the benefits to those who go on to develop the ailment of interest (e.g., Element (4)—“a medical monitoring regimen exists that makes expedited detection and treatment of the particular serious future bodily harm both possible and beneficial”). The New Tort does not expressly consider, as it should, the risks of monitoring to those who never develop the targeted ailment nor does it require the monitoring to be based on sound scientific data (e.g., Element (6)—monitoring reasonably necessary “according to generally accepted contemporary medical practices”).
These two elements are especially relevant since ALI Members were being asked to consider the first tort that could, in and of itself, cause injury to an otherwise healthy plaintiff: medical monitoring usually is not a risk-free endeavor. According to the National Institutes of Health (NIH), the potential harms from medical screening, including follow-up testing, “run the gamut of physical, psychological, emotional, and financial effects,” and are due to “false positives” which tend to outnumber true positives by an order of magnitude. A physician-researcher at Oxford University summarized the cervical cancer screening program that she was overseeing by quoting a fellow researcher: “By offering screening to 250[,]000 we have helped a few, harmed thousands, disappointed many, used £1.5m each year, and kept a few lawyers in work.” As such, scientists and regulators have long recognized that in evaluating a screening or monitoring proposal, one must take into account the risks to the entire cohort, including the larger number who will never develop the ailment of interest. The New Tort, as adopted by ALI, although remarkably sophisticated legally, does not appear to do that.
Second, if the New Tort were revised to require a court to assess the risks to the entire cohort being monitored, it is questionable whether a lay court or jury would have the necessary data or competence to assess and apply it in a given setting. This concern is not theoretical but is based on the fact that the Restatement sought to accurately reflect the tort as formulated in many of the states recognizing it. However, those court formulations largely ignore the risks of false positives and therefore do not appear to be in tune with contemporary science. If courts are unable to formulate a test for monitoring that is scientifically valid, it is questionable whether they would be able to accomplish the more challenging feat of applying the correct and far more complicated test in any given case.
Many may be inclined to brush these concerns aside, noting that judges and juries are constantly called upon to wade into scientific thickets, especially in products liability cases involving pharmaceutical products, in medical malpractice cases, and in patent cases. There should be no reason to treat monitoring any differently. However, when judges and juries are being asked to right a past wrong requiring resolution of scientific issues, we are willing to accept a court’s lack of scientific expertise because an incorrect decision has limited impact on the prospective health of the citizenry. Usually, all that is at stake is money. In contrast, when judges and juries are asked to assess and bless a prospective medical monitoring policy that will likely, adversely affect the physical and mental well-being of otherwise healthy individuals, we have an obligation to consider their competence to entertain those cases in terms of both science and policymaking.
Nor is this a case where the only remedy-related issue is the amount that the defendant should be ordered to pay for monitoring. Elements (4), (5), and (6) of the New Tort require courts and juries to get into the scientific weeds by making various assessments about a specific monitoring program, one likely proposed by plaintiffs, or between competing monitoring programs proposed by the opposing parties. Three U.S. Public Health Service agencies, with relevant expertise, have been authorized by Congress to develop and assess monitoring programs in certain settings. ALI should have considered the extent to which courts should defer to or request assistance from these agencies when asked to assess a monitoring program, assuming the program falls within the jurisdiction of one of these agencies.
Third, it is unclear whether the policy objectives underlying medical monitoring have been overtaken by events. If the primary objective of the tort were to ensure cost-free access to needed diagnostic testing, as stated in the New Tort, then the Affordable Care Act largely meets that objective without judicial intervention. If an objective is to promote health, also as stated in the New Tort, then using, where appropriate, one of the three U.S. Public Health Service agencies and their administrative processes rather than the courts is a safer and more rational approach with significantly lower transaction costs. If a purpose of the New Tort is to promote deterrence, then the “secondary payer laws” and a health care plan’s subrogation and reimbursement, eschewed in the New Tort, would accomplish that purpose potentially without judicial intervention.
Fourth, the New Tort suggests that “courts may choose to offset defendants’ liability by payments from collateral sources” as a way of “conserv[ing] the defendant’s financial resources.” If adopted by a state, this suggestion is likely inconsistent with both over-arching federal law and with one of the goals of tort law, namely deterrence, as noted in above.
This Article recommends that courts revise the wording of the New Tort to reflect contemporary science and then consider whether they have the necessary data and competence to apply the revised tort taking into account that there are three expert federal agencies which have been charged, in specific circumstances, with reviewing and recommending medical monitoring. ALI should have considered whether it would be more prudent for courts to defer to those agencies either in toto or in part, by requesting an assessment from the agency with jurisdiction before adopting a given course of monitoring. Despite these issues, the New Tort is sophisticated and well done. As such, it warrants careful consideration by courts in states that have not definitively adopted monitoring as either a remedy or tort.
I. The New Tort Does Not Appear to Consider the Risks and Benefits Across the Entire Group to be Monitored
Screening and medical monitoring are designed to lead to the detection of a disease or an ailment before it becomes symptomatic with the expectation that early detection will, in turn, lead to better treatment outcomes. The New Tort focuses exclusively on the benefits of monitoring to a plaintiff who ultimately develops the ailment or condition at issue. Current scientific practice though is to assess the risks to the entire cohort being screened or monitored and balance those risks against the benefits to those few who may develop the ailment or condition and who may benefit from early detection. As written, the New Tort appears to ignore the risks of monitoring to those who may never develop the ailment of interest.
According to the National Cancer Institute (NCI) within NIH,
The conversation about cancer screening is changing within the medical community. Overall, the recent trends have been towards recommending less routine screening, not more. These recommendations are based on an evolving—if counterintuitive—understanding that more screening does not necessarily translate into fewer cancer deaths and that some screening may actually do more harm than good.
As the Institute for Quality and Efficiency in Health Care (IQWiG) states, “[I]t is important that both the benefits and harms of a screening test are assessed in studies before introducing it at a large scale. That way, researchers can find out whether the benefits of the screening test outweigh the risks.”
The risks of harm caused by medical screening or monitoring can be significant and the benefits may prove difficult to measure properly. The potential harms from monitoring “run the gamut of physical, psychological, emotional, and financial effects. Plus, those harms can come not just from the screening tests themselves, but also from follow-up tests and treatments.” This is so because, “[w]ith few exceptions, screening tests do not diagnose the illness. Rather subjects who test positive typically require further evaluation with subsequent diagnostic tests or procedures.”
Injuries to otherwise healthy individuals can result from the monitoring itself, such as a punctured bowel during a colonoscopy. Even diagnostic tests which themselves appear to be virtually risk-free (e.g., a blood draw), can lead to injury through false positives triggering unnecessary follow-up tests that pose risk. This is known as the “cascade effect,” which “refers to a chain of events initiated by an unnecessary test, an unexpected result, or patient or physician anxiety, which results in ill-advised tests or treatments that may cause avoidable adverse effects and/or morbidity.” For example, a high prostate specific antigen (PSA) test involving only a relatively risk-free blood-draw may lead to a biopsy of the prostate tissue, which if performed through the rectum, usually with a local anesthetic, can lead to sepsis. If performed transperineally, the risk of sepsis is greatly reduced, but the procedure is usually done under a general anesthetic, which carries its own risks and also carries the risk of temporary erectile dysfunction (less than 5%). The accuracy of the monitoring is a pivotal factor in determining whether the risks outweigh the benefits.
A. Measuring Accuracy and Assessing Potential Risks
The potential risks and the net benefit (i.e., risks and benefits) of a diagnostic test turn in large measure on its “accuracy,” which is related to its “sensitivity” and “specificity.” Sensitivity is a measure of how well the test spots individuals who actually have the ailment of interest. A test with a 95 percent sensitivity means that if it were applied to 100 individuals suffering from the ailment, it would identify 95 (i.e., true positives) and fail to identify 5 (i.e., false negatives). Specificity, on the other hand, measures how well the test identifies those who are free of the disease at issue. If a test’s specificity is 95 percent, then of 100 individuals without the disease (i.e., true negatives), the test will correctly identify 95 of those individuals as likely disease free. However, the test will also incorrectly identify 5 individuals as suspected of having the disease (i.e., false positives), requiring further testing. Usually there is a trade-off between sensitivity and specificity: as sensitivity increases, specificity decreases, and vice versa, all else being equal.
For example, suppose that the prevalence of a given ailment is 0.2 percent, the sensitivity of the test is 90 percent, and its specificity is 95 percent. Given the above assumptions, in a cohort of 10,000 who undergo monitoring, there will be 18 individuals who test as true positives (i.e., 10,000 x 0.2% x 0.90) and 499 who test as false positives (i.e., (10,000 - 20) x (1 - 0.95)). Thus, for every true positive, there will be about 28 false positives. As it happens, this cohort size and prevalence closely parallel the cohort size and prevalence of cancer in the Uranium monitoring settlement.
A specificity of 95 percent sounds impressive, but it may not be. If the monitoring is conducted annually, the cumulative number of false positives will climb each year. For example, NCI noted that
according to one analysis, screening 10,000 women for breast cancer every year for 10 years starting at age 60 will prevent 43 deaths from breast cancer (88 women will still die from breast cancer despite getting screened). It will also result in nearly 5,000 false positives that lead to nearly 1,000 unnecessary biopsies.
But 5,000 false positives out of 100,000 screenings (i.e., 10,000 x 10 years) equates to a specificity of 95 percent.
False positives along with their risks and harms are real, significant, and cannot be ignored. Indeed, the Agency for Toxic Substances and Disease Registry (ATSDR), an arm of the U.S. Public Health Service, in implementing medical monitoring under the Comprehensive Environmental Response, Compensation, and Liability Act of 1980 (CERCLA), emphasized the need to assess the rate of false positives before adopting a monitoring program because “[f]alse positive results can result in follow-up testing that is uncomfortable, expensive and potentially harmful.”
In certain instances, the risks associated with false positives may be mitigated by repeating a positive monitoring test. However, repetition is not a panacea. We do not know whether an initial positive test result is true or false. Therefore, if the test is repeated yielding a negative result, we have no way of knowing whether that second result is a true negative or a false negative. A physician’s recommendation to a patient after a positive test repeated by a negative test result will turn on the predictive value of the test, the consequences of a missed diagnosis, and the risk associated with going forward with the next test in the cascade. In many cases, merely repeating a test which was positive will not resolve the issue short of a more invasive procedure.
Nor is it likely that these risks can be mitigated through “informed consent.” Researchers have been attempting to implement this practice for more than a quarter of century without success for a number of reasons. First, patient comprehension is low. In a review of the literature, one physician noted that
many patients do not understand the fundamental information regarding their treatment plans. Even though a patient may sign a consent form stating that he or she understands what has been communicated, studies show that even after agreeing to or receiving care, 18% to 45% of patients are unable to recall the major risks associated with their surgeries . . . .
Patient comprehension may not be legally necessary. In Canterbury v. Spence, the seminal informed consent case, the court held that “the physician discharges the duty when he makes a reasonable effort to convey sufficient information although the patient, without fault of the physician, may not fully grasp it.” However, studies and commentary suggest that patient comprehension of the risks associated with a given course of treatments or diagnostics can be improved through various communication strategies. Moreover, physicians, fulfilling their responsibilities as learned intermediaries, can and should suggest alternatives, assuming they are available, to reduce the risk of a procedure or test to a given patient. Second, “tension may exist between the aims of promoting” the monitoring program and promoting patient choice. Indeed, that tension is likely even greater where the patient is also a plaintiff who sued to receive the benefits of the monitoring program. And third, while common sense suggests that better informed consent and better communication between doctor and patient should reduce the percentage of (1) false positives or (2) those who opt to proceed to the next diagnostic step after a positive initial test result, the author is unaware of any study supporting that notion.
Shortly before the May 2023 ALI meeting, researchers at the University of Michigan and University of Minnesota set out the probability of patients receiving at least one false positive for various screening tests over the course of their lives. The researchers found that for breast cancer, assuming biennial mammography, the likelihood of a false positive was 47.7 percent; for cervical cancer, assuming annual PAP smear tests, the likelihood was 53.5 percent; for colorectal cancer, assuming a colonoscopy once every ten years, the likelihood was 38.2 percent; for prostate cancer using PSA once every two years, the likelihood of a false positive was 57.9 percent; and for lung cancer using low-dose CT scan yearly for smokers, the likelihood was 20.7 percent. The authors “estimated [the] probability of receiving at least one false positive in a lifetime is 85.5% (±0.9%).” Thus, long-term monitoring is likely to generate a discernibly significant number of false positives.
B. Measuring Benefits
The New Tort assumes that early detection will lead to better outcomes. However, according to the NCI, there is significant confusion about how to measure better outcomes. As the NCI notes, “An improvement in survival—how long a person lives after a cancer diagnosis—among people who have undergone a cancer screening test is often taken to imply that the test saves lives. But survival cannot be used accurately for this purpose because of several sources of bias.” The two major sources of bias are (1) lead-time bias and (2) length bias. Lead-time bias occurs when a researcher uses five-year survival as a measure of the efficacy of screening. NCI provides the following illustration of lead-time bias:
[A] man experiencing a persistent cough and weight loss is diagnosed with lung cancer at age 67, and he dies of his cancer at age 70 . . . . Five-year survival for a group of patients like this man is 0%.
If this man is screened and his cancer detected earlier, say at 60, but he still dies at age 70 . . . , his life has not been extended, but the measure of 5-year survival for a group of patients like this is 100%. . . .
"In this example, the man does not live even a second longer. This distortion represents lead-time bias.”
Lead-time bias is inherent in any comparison of survival. It makes survival time after screen detection—and, by extension, earlier cancer diagnosis—an inherently inaccurate measure of whether screening saves lives.
Unfortunately, the perception of longer life after detection can be very powerful for doctors, noted Donald Berry, Ph.D., professor of biostatistics at the University of Texas MD Anderson Cancer Center.
“Length bias” occurs because “screening is more likely to pick up slower-growing, less aggressive cancers, which can exist in the body longer than fast-growing cancers before symptoms develop.” Suppose that 1,000 individuals with an aggressive cancer are screened and five years later, 400 are alive. It would appear that screening prolonged the lives of 40 percent of those screened. Instead, suppose 3,000 individuals were screened, 2,000 had a nonaggressive cancer and 1,000 had an aggressive cancer. In five years, all those with a nonaggressive cancer would still be alive, while only 400 of those with the aggressive cancer would be alive. Screening, here, appears to have prolonged the lives of 2,400 or 80 percent. This result is a function of sampling that is biased based on the aggressiveness of the cancer.
Given these biases, according to NCI, “the only reliable way to know if a cancer screening test reduces deaths from cancer is through a randomized trial that shows a reduction in cancer deaths in people assigned to screening compared with people assigned to a control (usual care) group.” Randomized trials of this sort take years to conduct. With respect to a significant risk caused by an environmental insult, it is unlikely that there will be a randomized trial in the peer-reviewed literature. And even if one existed, the results, at least with respect to cancer, may be so modest as to raise serious questions about the efficacy of monitoring. For example, in a low-dose computed tomography (LDCT) trial of smokers sponsored by NCI, there was a 20 percent reduction in the relative risk of death from lung cancer for those screened with LDCT as opposed to those screened with ordinary X-ray. This may sound impressive, but a 20 percent reduction in the relative risk of death translates into only a 0.4 percent reduction in deaths (“1.7% in the chest x-ray group to 1.3% in the CT scan group”). The LDCT trials are discussed in greater detail in the following section.
C. The New Tort Appears to Ignore Both the Risks Associated with Monitoring and the Difficulty of Assessing Its Benefits
The New Tort appears to focus exclusively on the benefits from early detection for those who eventually develop the ailment or condition. The dangers associated with monitoring, especially false positives, are not incorporated into the New Tort and were not discussed in the accompanying Comment section. Nor did the Comment section or Illustrations note that false positives usually occur at least ten times more frequently than a true positive, making the dangers of medical monitoring palpable by encouraging a significant disease-free cohort to submit to monitoring. The New Tort requires that the monitoring make “expedited detection and treatment of the particular serious future bodily harm both possible and beneficial.” The word beneficial “additionally demand[s] that the plaintiff show that monitoring has the potential to alter the plaintiff’s prognosis, the course of the plaintiff’s illness, or the ultimate disability or impairment.” The benefit therefore is limited to the individual who actually develops the ailment of interest. As noted above, monitoring, though, cannot be evaluated through a lens that focuses only on those who eventually develop the ailment, but instead, must examine the much larger cohort of healthy individuals—most of whom will never develop the ailment, but who will risk injury through unnecessary diagnostic testing resulting from the New Tort. The incompatibility between the New Tort and scientific norms could raise unique challenges for any putative expert under Fed. R. Evid. 702 and Daubert. This also raises an ethical question: should ALI have adopted a tort that departs from scientific teachings or should it have modified the tort to comport with those teachings?
Reliance exclusively on gross benefits to the subpopulation that eventually develops the ailment is not surprising because the “potential harms of cancer screening are more complex and harder to measure.” Nonetheless, balancing the risks and benefits is the only scientifically valid way to assess whether to adopt a proposed monitoring test. Reliance on “contemporary medical practices” in Element (6) does not provide adequate safeguards or even obliquely suggest that the “practice” must be the result of weighing the harms and benefits to the entire population to be monitored, especially because the phrase is mushy and not necessarily evidence based. There are significant limitations built into the language of Element (6)—the Element’s wording, like the wording of Element (4), focuses exclusively on those who will or would develop the ailment in future, that is, “prevent or mitigate the harm.” It uses the word prevent, suggesting that individuals who would otherwise suffer from the ailment somehow escape as a result of monitoring. Monitoring rarely prevents disease. According to IQWiG,
[s]creening tests are often advertised with slogans like “prevention is better than cure.” But most of them cannot influence whether someone will get ill. Many people wrongly mistake screening for “prevention,” and some think that having regular screening tests can protect them from a disease. But that is not the case: screening tests usually cannot prevent diseases. Sometimes doctors also tend to be too optimistic when it comes to the benefits of screening.
Further, reliance on “contemporary medical practices,” may be misplaced. First, many practice guidelines, such as those for screening and monitoring, are developed by professional medical societies or ad hoc groups. According to the National Academy of Medicine, these societies and groups may have financial conflicts of interest and often lack “procedures for identifying and managing conflicts of interest.”
Second, NCI researchers “evaluated cancer screening guidelines from more than 10 medical organizations, including the U.S. Preventive Services Task Force (USPSTF), the American Cancer Society, and National Comprehensive Cancer Network (NCCN). They found that none of the guidelines had complete information on the potential harms of screening.” To further complicate the problem, “as many screening experts stress, most harms tend to occur during or soon after a person gets screened, while the benefits don’t show up until many years later.” If the actual diagnostic tests used in a traditional screening are comparable to those used in a proposed medical monitoring, the harms may be predictable even in the absence of complete information. However, the benefits in the monitored cohort may be more elusive and data may be wanting. We can expect significantly less information about medical monitoring, especially its benefits, following haphazardly occurring environmental insults. In many cases, there will be few if any studies on point, which may adversely affect the quality of the decision-making. In the absence of robust studies, any litigation will likely devolve into a battle between hired experts, leaving it up to the jury to decide which experts are more credible. Developing healthcare policy in this way is potentially dangerous. Given the risks to those who will never develop the target ailment, the U.S. Public Health Service agencies that deal with monitoring, two of which only deal with monitoring as opposed to screening, require sufficient, albeit less than complete, data before recommending a monitoring program. Courts, in jurisdictions where monitoring is a viable cause of action, should require sufficient clinical data about both risks and benefits before imposing a monitoring requirement.
The incentives are also misaligned, further compromising the quality of the decision-making. In the typical tort case, the incentives for plaintiff’s attorney and the plaintiff are aligned—the greater the plaintiff’s recovery, the greater the attorney’s contingent fee. Here, though, the more costly the monitoring (e.g., the larger the population, the greater the frequency, and more in-depth the diagnostic testing), the greater the fees for plaintiff’s counsel. However, more intense or more frequent monitoring, or the more relaxed the monitoring-coverage criteria, may not benefit a plaintiff. According to NCI, it may actually harm a plaintiff, but will likely increase the magnitude of the contingency fee.
Courts should be mindful that assessing a proposed monitoring program is complex: medical and scientific groups often disagree about the net benefit of a given screening protocol. For example, while many men over the age of fifty routinely have their PSA measured during an annual physical, the USPSTF “recommends against PSA-based screening for prostate cancer in men 70 years and older” (grade of “D”) and for those between fifty-five to sixty-nine years, “the decision to undergo periodic prostate-specific antigen (PSA)-based screening for prostate cancer should be an individual one,” that is, a grade of “C” for the younger cohort. Despite this recommendation, a recent study revealed that about 50 percent of the physicians surveyed routinely screen their male patients who are over seventy years of age.
Similarly, LDCT had been proposed as a screening test for lung cancer among smokers because of its potentially high sensitivity. A nationwide screening program for a population at elevated risk had been proposed and endorsed by several organizations for those with a history of smoking at a certain level, including USPSTF. The USPSTF found that LDCT’s sensitivity “ranged from 59 to 100 percent (13 studies; n=76,856) and was over 80 percent in most studies. The specificity ranged from 26.4 to 99.7 percent (13 studies; n=75,819) and was over 75 percent in most studies.” Its positive predictive value (PPV) “ranged from 3.3 to 43.5 percent,” where values under 5 percent are normally viewed as potentially too low to warrant recommending a diagnostic test. It also found that “[h]arms of screening included radiation-induced cancer (estimated 0.26 to 0.81 major cancers for every 1,000 people screened with 10 annual LDCTs),” as well as 17 unnecessary invasive procedures for every 1,000 people screened. The Medicare Evidence Development and Coverage Advisory Committee (MEDCAC) in the United States, “[b]ased largely on concerns over the high false positive rate, . . . recently recommended against covering the procedure for this patient group based on a lack of evidence to support the benefits of the screening test.” USPSTF continues to recommend LDCT with a grade of B; the Centers for Medicare & Medicaid Services ultimately adopted, with slight modification, the USPSTF recommendation overriding the recommendation of its own advisory committee—MEDCAC. The disagreement between two advisory committees within the same department illustrates the complexity of determining whether the benefits of screening outweigh its risks.
Medical monitoring, as a result of an environmental insult, may make epidemiological sense in various settings. If there were a significant uptick in the risk of developing a specific ailment, then one would expect the prevalence of that ailment in the monitored population to be higher than would otherwise be the case. The heightened prevalence increases the likelihood of a PPV that is sufficiently high to justify monitoring. To satisfy Element (5), the PPV for the nonexposed population, though, would have to be sufficiently low to contraindicate screening for the general population.
II. Congressionally Authorized Agencies are Likely Better Suited Than Courts to Craft or Assess Prospective Science-Based Monitoring
A. The Way Courts Have Analyzed Monitoring Programs Is Troubling
The Dutch graphic artist Maurits Escher designed some of the most interesting, elegant, and perplexing structures. They all shared one characteristic—none could be built. The medical monitoring tort is much like an Escher creation. It is an elegant remedy, but it is questionable whether it can be properly applied by a court in a way that yields meaningful, valid, and consistent results. The evidence suggests that courts have not risen to the challenge of scientifically evaluating proposed medical monitoring by weighing its harms and benefits.
Medical monitoring has been a remedy in some jurisdictions for about forty years. One would have thought that during that forty-year period courts would have focused on the potential harms caused by the inevitable false positives. As of August 5, 2023, there were 621 state court cases reported on casetext.com where the term medical monitoring was mentioned; none discussed false positives, let alone the harm caused by them. The federal side did not fare much better: there were more than 1,000 reported federal cases where medical monitoring was mentioned; seven mentioned false positives and four that are relevant are discussed. One, Block v. Abbott Laboratories, was ironic: the underlying tort arose out of a “false positive.”
In Gates v. Rohm and Haas Co., plaintiffs, exposed to vinyl chloride, sought medical monitoring in the form of MRI testing. The court rejected the specific remedy stating that “[p]laintiffs have submitted evidence that the false positive rate for MRIs is low and the screening of asymptomatic individuals may be beneficial. . . . However, this evidence is not persuasive, when weighed against the contrary testimony of Dr. Friedman, as well as other evidence submitted by Rohm and Haas.” While Gates suggests that courts are willing to examine the harm caused by false positives, the court’s analysis is too cursory to draw any firm conclusions. And in Batchelder v. Kerr-McGee Corp., two class members out of more than 6,000 unsuccessfully objected to the medical monitoring portion of the settlement on the grounds that it was inadequate because “the proposed screening could result in false positive readings (allegedly as a result of simply ‘having eaten a barbeque sandwich the day before’), thereby causing panic among class members who may not be sick.” As in Gates, the impact of false positives on the court’s decision-making is unclear.
In Bowling v. Pfizer, a class action settlement involving allegedly defective artificial heart valves, defendants set up a corpus to fund valve-related medical research aimed “at developing a non-invasive medical device that would diagnose fracture-prone BSCC valves.” One member of the class sought to redirect some of those funds to another program at the Hershey Medical Center. In rejecting the request, the court observed that
[m]ost troubling is that among the [Hershey] study's four clear errors was a false positive—in other words, one valve was imaged as “Probably SLS,” but, in fact, it was a perfectly good, intact valve. To be clear: This patient risked dying from inherently risky open-heart surgery for no reason. In addition, one other error also could be considered a false positive—it was imaged as “suspicious,” but, in fact, it was completely “intact.” In sum, based on the results of the original Hershey program, reinstituting the program could be dangerous, risking more false positives for class members and thus post-operative death or serious injury.
Bowling is the closest any court has come to expressly recognizing and assessing the harm of false positives associated with medical monitoring. The ALI Reporters have been faithful to the tort: Element (4) does not require one to assess the impact of false positives because it only involves the gross benefits to those who will eventually suffer from the ailment of interest. Before adopting medical monitoring, a court should carefully reflect whether it wants to adopt elements in a Restatement that appear to be inconsistent with contemporary medical science. Courts and attorneys may have ignored false positives because they add a level of complexity that may be beyond their competence.
The decision-maker must have the competence both to recognize that it must assess, and to actually assess both the risks and benefits of a medical monitoring test across the entire cohort. Additionally, the decision-maker must possess sophisticated policymaking skills to appreciate where pivotal lines should be drawn. For example, how many lives need to be saved to justify a monitoring program with its attendant risks? In addition, the decision-maker must be able to recognize when the data are insufficient to draw any relevant conclusions concerning the advisability of approving a monitoring program. These knowledge-based skills may fall well outside a court’s comfort zone.
B. The U.S. Public Health Service Is Better Qualified than a Court to Assess Whether Monitoring Is Warranted and if so, the Nature of that Monitoring
Congress has created three agencies, all within the U.S. Public Health Service, with the authority to recommend or to impose medical monitoring for those at increased risk of developing an ailment or condition: (1) the United States Preventive Services Task Force (USPSTF); (2) the Food and Drug Administration (FDA); and (3) the Agency for Toxic Substances and Disease Registry (ATSDR). USPSTF’s authority is broad and covers screening and monitoring programs that have significant public health implications. The FDA’s authority to impose monitoring is limited to drugs and biologics with significantly elevated risks. ATSDR’s authority is limited to environmental hazards at so-called Superfund sites. Most courts lack the expertise of these agencies. This expertise can be brought to bear to help determine whether monitoring is warranted and if so, what form that monitoring should take. In those instances where an agency’s jurisdiction is implicated, the need for plenary judicial intervention should be carefully considered by a court.
There are no time limits imposed on USPSTF action, and some actions have taken years to complete owing to the absence of relevant data. However, there are “soft” time limits imposed on the FDA and ATSDR. Subject to two exceptions not relevant here, if someone requests action by the FDA, usually in the form of a citizen petition, then the FDA “shall furnish a response to each [citizen] petitioner within 180 days of receipt of the petition.” The FDA meets this 180-day deadline through a tentative response in about half of the cases, but frequently takes much longer to issue a final response, especially where the issue is complicated. As the Ninth Circuit stated, “ATSDR [must] complete a ‘health assessment’ within one year of an Environmental Protection Agency (EPA) proposal to list a site on the [Superfund List].” Based on the health assessment, the ATSDR Administrator determines whether there is a significant increased risk of adverse health effects in humans from exposure to hazardous substances and if so, ATSDR must initiate a health monitoring program. This assumes that there are adequate epidemiological data available.
The upside of relying on these administrative agencies is their expertise; the downside is their speed. As one appeals court noted, some administrative agencies move at “[g]lacial . . . speed.” All three agencies would likely take a few years, depending on the available data, to issue conclusions. For USPSTF, for example, the average time from “selecting a topic for review through publishing the final recommendation statement [is] 3 years.” The speed with which these agencies act must be balanced against the speed of the alternative civil litigation process for class actions, which is not necessarily much faster. This is an area warranting further attention and assessment.
1. United States Preventive Services Task Force (USPSTF)
a. Establishment and Jurisdiction
The USPSTF was created in 1984 as an advisory committee within the U.S. Public Health Service to make recommendations to the Surgeon General about preventive services that ought to be encouraged. As part of the Healthcare Research and Quality Act of 1999, Congress formally established the USPSTF as an advisory committee reporting to the Director of the Agency for Healthcare Research and Quality, within the U.S. Public Health Service, but not subject to the Federal Advisory Committee Act. Under the Act, USPSTF was instructed to “review the scientific evidence related to the effectiveness, appropriateness, and cost-effectiveness of preventive services for the purpose of developing recommendations for the health care community, and updating previous clinical preventive recommendations.” The USPSTF usually considers only topics of “[p]ublic health importance (i.e., burden of suffering and expected effectiveness of the preventive service to reduce that burden).” The USPSTF was reestablished by the ACA § 4003(a), which provided, among other things, that the Task Force would be “independent” and further that “any recommendations made by such members, shall be independent and, to the extent practicable, not subject to political pressure.” The ACA § 1001, added a new section 2713 to the Public Health Service Act providing as follows:
(a) In General.—A group health plan and a health insurance issuer offering group or individual health insurance coverage shall, at a minimum provide coverage for and shall not impose any cost sharing requirements for—
(1) evidence-based items or services that have in effect a rating of ‘A’ or ‘B’ in the current recommendations of the United States Preventive Services Task Force.
Thus, under the ACA, a private insurer, in most cases, is obligated to cover a diagnostic test recommended by the USPSTF and graded as either A or B without cost-sharing, for example, no coinsurance, no copayment, and no deductible.
The calculus used by USPSTF is instructive. For example, USPSTF evaluated whether monitoring those at high risk for chronic obstructive pulmonary disease using spirometry was warranted. One of the relevant risk factors was “[e]xposure to occupational and environmental pollutants.” Ultimately, USPSTF recommended against screening for chronic obstructive pulmonary disease using spirometry. The agency’s recommendation was based, in part, on its determination that “[o]pportunity costs (time and effort required by both patients and the health care system), anxiety associated with false-positive results, and adverse effects from appropriately or inappropriately prescribed medications are all potential harms of screening for COPD using spirometry.”
b. Requesting USPSTF Review and Its Recommendations
In theory, the USPSTF is open to reviewing any monitoring program. According to the USPSTF, “Anyone can nominate a new topic for Task Force consideration or request an update of an existing topic at any time online.” Thus, a trial court or the parties in given litigation could ask USPSTF to opine on the propriety of monitoring in a given case or which of competing monitoring programs provides the largest net benefit to plaintiffs. However, like the Supreme Court of the United States, USPSTF is not obligated to review, assess, and grade every monitoring program presented to it. Rather it will only undertake a review of a proposed monitoring program where the number of affected individuals is significant, and the matter is not within the jurisdiction of the FDA or ATSDR. Given that USPSTF limits its jurisdiction to issues of public health importance, environmental insults that affect relatively few individuals will likely not be suitable for Task Force consideration.
There are other practical limitations even where the issues are of public health importance. Most notably, USPSTF will only recommend a course of monitoring where sufficient data exist to assess whether the program’s benefits outweigh its risks. Frequently, robust data do not exist, making it unlikely that USPSTF would recommend monitoring. The fact that USPSTF believes that the data are insufficient to recommend monitoring should be of import to a court. Indeed, it would arguably be unwise and potentially dangerous for a court to approve monitoring where experts have declined to do so for want of data.
c. Constitutionality of USPSTF Is Before the Supreme Court
As of this writing, the constitutionality of the USPSTF process is the subject of an ongoing litigation. The Fifth Circuit concluded that USPSTF members are “principal officers” within the meaning of the Appointments Clause, but none had been appointed by the President with the advice and consent of the Senate. The Secretary of Health and Human Services sought to address this concern by treating the Task Force members as subject to his control and by ratifying the Task Force’s recommendations. However, the appeals court found that ratification was not possible because the statute requires Task Force members to act independent of political influences. According to the Fifth Circuit, “the Task Force cannot be ‘independent’ and free from ‘political pressure’ on the one hand, and at the same time be supervised by the HHS Secretary, a political appointee, on the other.” The Supreme Court granted certiorari to decide whether Task Force violates the Appointments Clause and if so, whether the court of appeals should have severed “the statutory provision that it found to unduly insulate the Task Force from the HHS Secretary’s supervision.”
Upholding the USPSTF may prove challenging for the government. Even if the government were to prevail on the issue of severability, the Secretary’s ratification could be viewed as a substantive rule under the Administrative Procedure Act (APA). The recommendations of the USPSTF have the attributes of a legislative rule: they have prospective effect and general or particular applicability, and are legally binding on those outside of government—both patients and insurers; their economic impact is usually significant by any measure. The grading of a monitoring protocol by the USPSTF would also qualify as rule separately. Grading can affect the rates paid by an insured individual for a given service reducing them from, for instance twenty percent, a common cost-sharing amount, to zero percent. This is a quintessential “prescription for the future of rates,” a hallmark of a legislative rule. However, the Secretary has no general rulemaking authority under the Public Health Service Act and section 2713 grants no specific rulemaking authority to the Secretary. Further, even if the Secretary had rulemaking authority, the APA would still require public participation through notice and comment. There is nothing explicit in the ACA that would relieve him of this obligation. To the contrary, the APA provides that a “[s]ubsequent statute may not be held to supersede or modify this [Act] . . . except to the extent that it does so expressly.” Whether the Secretary’s ratification constitutes a rule for which the Secretary lacks rulemaking authority is not before the Supreme Court, but could be addressed by a lower court on remand.
Even if the Court affirms the Fifth Circuit that does not mean that USPSTF ceases operations. It merely means that health plans would no longer be required to cover certain preventive services. However, most insurers would continue to provide coverage because, by definition, the preventive services have to be cost-justified to be recommended. And further, even if the Secretary were found to lack rulemaking authority that would not preclude the Secretary from issuing Task Force recommendations as nonlegislative rules. Moreover, even if the Court rejects the government’s position, there is no reason precluding a group of individuals—who believe that they are entitled to medical monitoring due to a significant increase in risk of developing an ailment in the future or a court—from petitioning USPSTF to evaluate the propriety of a monitoring program under the Task Force’s own rules of procedure, as noted above or under the APA’s petition provision.
2. Food and Drug Administration
a. Monitoring Programs Under the FDA’s REMS Authority
Articles such as drugs and biologics are subject to premarket review and approval by the FDA. As part of that review process, the FDA uses data from clinical trials to assess the risks and benefits of a new article. In some cases, additional clinical trials are conducted after approval. Where a drug or biologic is effective, but associated with a serious adverse drug experience, the Commissioner, as a condition of approval or continued marketing, may impose a “risk evaluation and mitigation strategy,” known as REMS. This better ensures that the benefits of the drug outweigh its risks. One controversial REMS was imposed on the drug Accutane, which was approved in 1982 as an effective treatment for teenage and certain adult acne. In 2005, the FDA imposed a REMS known as iPLEDGE, “to prevent fetal exposure to isotretinoin[.]” As a condition of receiving the drug, patients who are capable of becoming pregnant are required to submit to monthly medical monitoring to ensure that they are not pregnant and that they are using at least two forms of birth control.
The FDA REMS program is not triggered by tortious conduct, but rather by a significant risk of an adverse event in the future and belief, based on scientific data, that certain conduct, including medical monitoring, can mitigate that risk or harm.
b. Petitioning for a Monitoring Program
The FDA has a robust tradition of entertaining citizen petitions, including petitions to impose, strengthen or weaken REMS. The FDA is obligated to respond to a citizen petition within 180 days of receipt. It can satisfy that requirement by issuing a tentative response, including one indicating that there are insufficient data available or that the matter is too complex to resolve within 180 days. Those who believe that they are at elevated risk as a result of taking a drug or biologic may petition the agency to impose a REMS requiring medical monitoring. It is unclear whether monitoring would apply to those who have ceased taking the drug given that its purpose is to enable patients to continue with a beneficial drug that carries significant risks. Monitoring imposed by the FDA as a condition of receiving a drug or biologic is by definition “reasonable and necessary” under most health insurance policies and therefore, should be covered by those insurers.
Patients would be responsible for any cost-sharing associated with the monitoring. However, any monitoring program developed by the FDA would be based on robust data from clinical trials permitting the agency to shape a targeted monitoring program which would likely be substantially more focused than a program fashioned by a court.
3. Agency for Toxic Substances and Disease Registry
ATSDR was created in section 104(i) of CERCLA and its role greatly expanded by the Superfund Amendments and Reauthorization Act of 1986 to “provide medical care and testing to exposed individuals, including but not limited to tissue sampling, chromosomal testing where appropriate, epidemiological studies, or any other assistance appropriate under the circumstances.” Its Administrator is the Director of the Centers for Disease Control and Prevention (CDC). As the Ninth Circuit stated, “ATSDR [must] complete a ‘health assessment’ within one year of an EPA proposal to list a site on the [Superfund List]. The court observed that
[t]he purpose of the health assessment is to help determine “whether actions . . . should be taken to reduce human exposure to hazardous substances from a facility and whether additional information on human exposure and associated health risks is needed and should be acquired by conducting epidemiological studies . . . , establishing a health surveillance program . . . , or through other means.”
The court also reasoned that “[o]nly if the ATSDR Administrator determines that ‘there is a significant increased risk of adverse health effects in humans from exposure to hazardous substances based on the results of a health assessment’ must the ATSDR initiate a health surveillance program.’” Additionally, the court found that “[o]nce initiated, the program must include both periodic medical testing to screen the exposed population for disease and a mechanism to refer for treatment anyone who needs medical attention.”
Thus, ATSDR provides a science- and evidence-based determination of whether the release of hazardous materials significantly increased the risk of adverse health effects and, if so, the types of monitoring that ought to be made available to those so affected. In such cases, the party responsible for the release must bear the cost of the health assessment including the costs of monitoring. ATSDR monitoring is not tort based and likely does not preempt or supplant state remedies. Two circuits—the Ninth and Third—have concluded that state law monitoring claims may be maintained notwithstanding CERCLA. However, for our purposes, a court should assess whether it should entertain monitoring as a remedy when the experts at ATSDR are making the same assessment and how the court should treat an ATSDR determination.
III. The Need for The Proposed Tort as a Way of Ensuring Adequate Access to Medical Diagnostics and Treatment Should Be Reevaluated In Light of the ACA
Medical monitoring, as a tort, may have been born of necessity in 2009 and earlier, especially given that its primary purpose was to “foster[] access to beneficial diagnostic testing.” That may no longer be necessary following enactment of the Affordable Care Act in 2010, and full implementation in 2014. In 2010, according to the CDC, 48.2 million persons in the United States under the age of sixty-five years or 18.2 percent of that group lacked health insurance coverage at the time of the interview, due to affordability or eligibility, and 60.3 million (19.8%) persons of all ages were uninsured for at least part of the year prior to interview.
In light of the large number of persons uninsured, a set of the pre-2010 cases adopted medical monitoring, in part, to improve access to diagnostic testing by
avoid[ing] the potential injustice of forcing an economically disadvantaged person to pay for expensive diagnostic examinations necessitated by another’s negligence. Indeed, in many cases a person will not be able to afford such tests, and refusing to allow medical monitoring damages would in effect deny him or her access to potentially life-saving treatment.
The few post-ACA cases that adopted medical monitoring without relying on binding precedent merely parroted back the notion from pre-ACA cases that the tort will improve access to health care but failed to address the impact of the ACA. Thus, in Sadler v. PacifiCare of Nev., Inc., the court adopted medical monitoring noting that “[i]f medical monitoring claims are denied, plaintiffs who cannot afford testing may, through no fault of their own, be left to wait until their symptoms become manifest, losing valuable treatment time.” Interestingly, the plaintiffs in Sadler all had health insurance and the type of testing that they sought would have been covered as an “essential health benefit” under the ACA or as a benefit under their group health plan. The issue arose before the ACA was fully implemented, but the decision was issued the year following full implementation. The ALI Reporters allude to the impact of the ACA obliquely in one sentence. Given that the lack of access to diagnostics was offered as a primary rationale for the tort and that ACA has altered that landscape, a deeper dive into whether access remains an issue is warranted.
Had that impact of the ACA been analyzed, the courts would have found that it was designed to dramatically improve access to health care in a number of ways. First, it offered states the opportunity to expand their Medicaid coverage to include those with annual incomes of less than 138 percent of the Federal Poverty Level (FPL) without regard to the applicant’s assets. Forty-one states, and the District of Columbia, took advantage of this opportunity and expanded their Medicaid programs accordingly. As a result, according to the Centers for Medicare & Medicaid Services, the Medicaid and Children’s Health Insurance Program (CHIP) populations grew from 54.6 million in 2010 to over 87.3 million by October 2023.
Second, the new law subsidized premiums by “providing tax credits to help people buy insurance and establish[ed] online marketplaces where insurers could sell plans.” For those purchasing health insurance online through the government-marketplace, the ACA subsidized premiums through tax credits for those with family incomes of between 100 percent and 400 percent of the FPL, and significantly reduced cost-sharing (e.g., copayments, deductibles, coinsurance) for those with family incomes of not greater than 250 percent of the FPL.
Third, employer-sponsored plans must offer “minimum essential coverage” and must be affordable which means that in 2023, the premium for the lowest-cost plan offered by an employer to its employees must be less than 9.12 percent of the employee’s household income.
Fourth, especially relevant to medical monitoring, the ACA triggered “zero cost-sharing” for those preventive services, which include screening and monitoring, graded A or B by the USPSTF. This applies to virtually all public and private plans except a few employer plans that pre-date the ACA.
The ACA has dramatically increased health insurance coverage especially for those with limited incomes, although ascertaining its precise numerical impact may be challenging. As noted above, the ALI Reporters inferentially recognized in passing the ACA’s impact on coverage (noting “that in 2018, 11 percent of those under 65 were uninsured, down from 17 percent in 2000”) and in 2023, according to one accounting, approximately 7.7 percent of the population lacked health insurance. Of the non-elderly uninsured cohort, approximately 26.1 percent were uninsured, because they “Do Not Need or Want” insurance. Thus, assuming the 26.1 percent figure holds across the entire population, approximately 5.7 percent of the U.S. population is uninsured because of affordability or eligibility.
Adding the number of individuals with health coverage in each insurance category yields comparable results. Specifically, there are (1) approximately 65.7 million covered by Medicare divided almost equally between those with traditional Medicare and those with Medicare Advantage, (2) 87.3 million covered by Medicaid or CHIP, (3) 133 million with Employee Retirement Income Security Act of 1974 (ERISA) health coverage, (4) 21.3 million covered by an ACA Marketplace health plan, (5) 15.2 million (not counting dependents) eligible to receive coverage through a non-ERISA plan (e.g., state, or local government employee), and (6) 2.95 million (not counting dependents) eligible to receive coverage through a non-ERISA federal government employee plan. Of the Medicaid total, 12.5 million are dual eligible and must be subtracted from either the Medicaid or Medicare count to avoid double counting. This totals approximately 312.94 million individuals in the United States out of a population in 2023 of about 334.9 million. Of those without insurance (i.e., 6.55%), 26.1 percent, as noted above, have decided not to purchase health insurance for reasons other than affordability or eligibility. This 26.1 percent figure translates to 5.73 million (i.e., 26.1% x (334.9 million - 312.94 million)). Thus, the total of those with insurance and those who have declined to purchase health insurance for reasons other than affordability or eligibility is 318.7 million for a rate of 4.8 percent who lack insurance because of affordability or eligibility. There are a number of caveats. For example, the numbers in each of the six categories were measured at different months or years; one individual may be eligible for and be enrolled in more than one health plan; and the number of dependents for the nearly 18 million insured through a federal or other governmental plans is unknown and has not been counted in the total. Even so, this closely mirrors the government’s calculations of 7.6 percent insured, in toto, or 5.61 percent if discounted by those without insurance who were eligible and could afford to purchase the insurance.
Given this widespread improvement in coverage, albeit not universal, it is unclear whether the New Tort’s rationale to ensure greater access to monitoring still holds sway. At the very least, the ALI Reporters ought to have considered the impact of the ACA on both coverage and cost-sharing to determine the likelihood that courts would adopt this nontraditional tort in the future.
IV. The Proposal Raises Secondary and Third-Party Payer Issues
One of the more interesting aspects of the New Tort is the recommendation that “courts may decide to deviate from the traditional collateral source rule to require the actor to pay medical monitoring expenses only to the extent that the cost . . . has not been, or will not be, fully borne by . . . a government fund.” Comment l. states that “courts may choose to limit liability for medical monitoring when the cost of the relevant diagnostic testing has been fully borne, or will be fully borne, by the plaintiff’s insurance, the plaintiff’s employer, a government fund, or another collateral source” This invitation to deviate from the collateral source rule by requiring a defendant to pay only for those monitoring expenses not otherwise covered by insurance, plaintiff’s employer, or a government fund is designed to limit defendant’s liability. This makes sense. Unfortunately, the recommendation, if adopted by a court, would collide with federal laws affecting a majority of health plans, including Medicare, Medicaid, and most self-insured, employer-sponsored health plans. The recommendation would also undermine the ALI’s goal of deterrence, by reducing the damages a defendant would otherwise be obligated to pay for monitoring. It would shift those expenses to others, including ratepayers, taxpayers, and the Medicare Trust Funds.
The Medicare as a Secondary Payer (MSP) provision “makes Medicare the secondary payer for medical services provided to Medicare beneficiaries whenever payment is available from another primary payer.” Medicaid has an analogous provision enforced by the states. MSP applies where payment may be made by a tortfeasor or other responsible party; MSP would apply even in those jurisdictions which have not addressed the viability of the monitoring tort, if, as part of a settlement, a monitoring program were established. As of March 2023, 67.5 million are insured through Medicare with 33.948 million on traditional Medicare parts A and B, and 31.799 million enrolled in a Medicare Advantage plan, under Medicare part C. As of October 2023, an additional 87.289 million were enrolled in Medicaid or CHIP. Thus, approximately 140.5 million, after subtracting 12.5 million who are dual-eligible, are covered by Medicare, Medicaid, CHIP, or combination thereof, and subject to secondary payer rules. This represents approximately 42 percent of the entire population.
Suppose that there is a class action for medical monitoring in a state that either recognizes the tort or has taken no position. Further suppose that the case settles, and a fund is set up to pay for monitoring class members. One of those members, Susan, is over 65 years old and is covered by traditional Medicare part A (hospital fund) and part B (supplemental fund for physicians, laboratories, etc.). Also, suppose that medical monitoring ordered by the court has not been graded by USPSTF as A or B and further the court adopted Comments b. and l. If the diagnostic tests are “reasonable and necessary” and therefore covered by Medicare, Susan will pay 20 percent of the Medicare allowable and Medicare will “conditionally” pay 80 percent. If Susan joins the class, her 20 percent would be paid out of the monitoring fund. However, the conditional Medicare payment triggers the fund’s and defendant’s MSP obligations.
Inasmuch as there is a judgment in favor of a class and Susan is a member of that class, the fund would arguably be obligated to recompense Medicare for its 80 percent conditional payment, even if the Comments b. and l. were applied by that state. CMS would likely argue that any judgment that invokes Comments b. and l. to shift medical monitoring costs to Medicare is impliedly preempted as to Medicare parts A and B and expressly preempted as to Medicare Advantage (part C—managed care) and Prescription Drug Plans (part D). In U.S. v. Rhode Island Insurers’ Insolvency Fund, the court held that a state law provision authorizing payment from the state insolvency fund only after all government payment sources had been exhausted was preempted by Medicare’s MSP provision. For traditional Medicare (i.e., parts A and B) there is no express preemption provision, so courts have relied on obstacle preemption when state laws are inconsistent with the MSP provisions. However, for Medicare Advantage plans and Prescription Drug Plans (i.e., parts C and D, respectively), there is express preemption. It would appear odd for a court to treat Medicare parts A and B differently from Medicare parts C and D. This is especially so since parts C and D are funded from the parts A and B Trust Funds.
Employer plans under ERISA, could also enjoy the same protection as Medicare and Medicaid. According to the Department of Labor, as of 2020, approximately 133 million are covered by an ERISA health plan. Since most ERISA plans, like Medicare and Medicaid, are secondary to other insurers, Comments b. and l. would conflict with those plans and if so, those Comments, if adopted by a state legislature, could be preempted by ERISA, depending on the nature of the plan. State laws regulating insurance are “saved” from ERISA preemption for plans that are insured rather than self-funded under 29 U.S.C. § 1144(b)(2). Under that savings clause, state insurance laws mandating particular benefits may be applied to insurance coverage provided through insured ERISA health plans. Thus, a state law requiring an ERISA plan to provide primary coverage (other than with respect to federal plans such a Medicare) would be saved from preemption if the plan were insured, but it would not be saved from preemption in the event that the plan were self-insured. To further complicate the landscape, Medicare and virtually all health plans contain provisions authorizing subrogation or reimbursement. Certain states preclude subrogation against an insured plan participant. If the plan is a self-insured ERISA plan, then the state law provisions are preempted by ERISA and the plan provisions govern. At least 35 million individuals are covered by self-insured ERISA plans. If, however, the plan is insured, then the state law provisions, as regulations governing insurance, would not be preempted by ERISA.
Conclusion
The Restatement is intended to “reflect the law as it presently stands or might appropriately be stated by a court.” However, where a proposed new section to the Restatement involves the intersection of three disciplines—tort law, epidemiology, and health care finance law—ALI should have considered whether its synthesis is consistent with all three. It should have avoided adopting a provision that, while an accurate reflection of the law in a subset of states, is inconsistent with scientific principles or with federal laws governing health care finance. The New Tort, as adopted by ALI, appears to have ignored an aspect of public health (e.g, the consequences of false positives) and an aspect of federal law (e.g., secondary payer provisions). This is not to say that medical monitoring absent present injury is not worth considering. If ALI sought to encourage medical monitoring, it should have done so in manner that addresses the salient scientific and federal legal issues. This is especially so where some thoughtful wordsmithing could possibly have addressed some of these issues. Consideration should also have been given to the role of the Public Health Service in helping ensure that any monitoring proposal makes scientific and medical sense.
