4

Mitigating Liability Associated with Clinical Research

Chapter 3 outlines strategies to minimize harm in clinical research settings, which is an important factor in reducing potential liability in those trials and in the postmarketing clinical setting. This chapter discusses additional strategies that may mitigate liability for clinical research involving pregnant and lactating women for several key stakeholders in the research enterprise: industry sponsors, institutions, institutional review boards (IRBs), and investigators. It also discusses additional legal protections for pregnant research participants.

Pregnant or lactating research participants and their offspring who were harmed in clinical research are currently able to seek compensation from sponsors, research institutions, investigators, and IRBs through the tort system. However, the process of obtaining relief is arduous, time consuming, and inequitable. Significant uncertainty prevails, and the system neither satisfies people who have suffered injury nor stakeholders who worry about liability risks. Compensation schemes offer an alternative to the tort system. There are several different models of compensation systems used by institutions and organizations for those injured in clinical research. Each of these systems has varying degrees of coverage for injuries and has specific ways of how those injured can request compensation. Compensation schemes vary in the damages that they cover—examples include costs of medical expenses, pain and suffering, lost wages, and death. The compensation programs for research injuries discussed below include self-indemnity programs, agency-affiliated compensation programs, and certain types of federal no-fault compensation schemes, such

as the Countermeasure Injury Compensation Program (CICP). While the chapter presents some of their strengths and weaknesses, it does not make recommendations for each of these compensation strategies, but simply reviews the evidence for each.

MITIGATING LIABILITY FOR INDUSTRY SPONSORS

Drug development typically happens within industry, with more than 99 percent of products approved between 2010 and 2019 sponsored by for-profit entities and the majority of phased clinical trial costs coming from industry (Cleary et al., 2020; Zhou et al., 2023). Although the committee could not find a systematic review identifying funders of research with pregnant and lactating women and acknowledges that industry funding is the most difficult funding source to track (Moran, 2009), a review of funding disclosures in the published literature shows that the National Institutes of Health (NIH), the U.S. Agency for International Development, and the Bill and Melinda Gates Foundation make up the highest proportion of funders for maternal health research (Footman, 2014). Nonetheless, industry has an important role to play in advancing research and evidence in pregnant and lactating women. Industry sponsors confer with the U.S. Food and Drug Administration (FDA) and carry out the trial design for a product and are responsible for monitoring a trial throughout. Given their critical role in all aspects of clinical research, the committee considered potential strategies to mitigate industry sponsor liability.

FDA Regulatory Guidance

As covered in Chapter 3, FDA has issued several guidance documents relevant to conducting research in pregnant and lactating women that could be improved upon to minimize harm to these populations. One way to improve guidance documents is for FDA to provide additional detail and clarity. Stakeholders for activities at different stages along the product development pathway have identified the lack of regulatory clarity from FDA as a challenge to including pregnant and lactating women in research (NASEM, 2023). Uncertainty regarding what evidence FDA would consider adequate for studies including pregnant and lactating women and additional guidance for conducting studies on conditions specific to pregnancy or lactation is of particular concern for research sponsors. This uncertainty may stem from the lack of guidance from FDA on determining the safety and efficacy of a product in pregnant women through clinical studies. In fact, FDA’s guidance on pharmacokinetics (PK) in pregnancy reads “This guidance . . . does not address ways to assess efficacy of a drug in pregnancy or how to assess whether the drug

causes adverse pregnancy or neonatal outcomes” (FDA, 2004). It may also stem from FDA guidance indefinitely remaining in draft form without finalization, leaving industry sponsors unsure if the draft truly represents the final word. FDA has not published a time line or set public goals for finalizing FDA guidance.

Although not legally binding, published guidance from FDA nonetheless sets expectations for product sponsors, clinical investigators, IRBs, FDA reviewers, and others involved in clinical research and review, as discussed in Chapter 3. Industry values guidance because it provides consistency in expectations for themselves and their competitors (Seiguer and Smith, 2005). If a lawsuit arises, clear FDA standards and evidence that the sponsor complied with those standards may support the sponsor’s defense.¹ While compliance with federal regulations or guidance is generally not a complete defense, it can bolster the defense to be able to demonstrate compliance with detailed federal requirements or, to a lesser extent, recommendations, particularly when they leave little room for discretion on the part of the sponsor. For example, the Third Restatement of Torts states that a product’s compliance with regulation and safety standards is considered when determining whether a product is defective (American Law Institute, 2023).²

Informed Consent

As outlined in Chapter 2, in addition to supporting ethical requirements, robust informed consent may provide some liability protection. This section discusses informed consent considerations that may mitigate liability for industry sponsors. Informed consent provisions to mitigate liability for institutions, IRBs, and investigators are discussed later in the chapter. Depending on state law, sponsors may be responsible for making sure that informed consent is appropriately documented and obtained. Careful attention to the information that is provided to investigators and the informed consent process may provide the best backstop to a claim based on failure to warn. Sponsors must take care to provide all the required information in the informed consent form in the investigator’s brochure. FDA has specific guidance applicable to pregnant women that should be included. Special considerations for information on the

___________________

¹ As presented to the committee in open session by Kirke Weaver on June 16, 2023.

² The Restatement of Torts is a treatise issued by the American Law Institute that summarizes the general principles of common law in United States tort law. It is a consensus-based document and a secondary source of law that courts may adopt or cite as persuasive authority.

informed consent process provided to investigators in research with pregnant women include:

• Ensure that the investigator’s brochure and the informed consent document outlines the known and potential risks associated with the investigational product. This should include any potential risks and the chance of unknown risks to an embryo or fetus, should a participant become pregnant.³
• For studies that may involve pregnant women, the investigator’s brochure should include results of animal reproductive toxicity studies with appropriate explanation of their significance in humans.⁴
• If a participant becomes pregnant during a trial, unblinding should occur to determine exposure, and risks and benefits should be reviewed with the participant to determine whether it is in their best interest to continue treatment with the investigational drug (if they were on it). A second informed consent process appropriate for a pregnant participant should then take place. Whether or not they continue with the treatment, data should be collected (FDA, 2018).
• Consistent with FDA regulations, the consent document should be updated regularly as new safety information is obtained about the investigational product.⁵

Ensure Qualifications and Experience of Investigators and Clinical Staff

The selection of qualified investigators is a factor that can minimize harm and mitigate liability within clinical trials. Even if a product is performing as expected, research participants can suffer injuries because of negligence by the investigator. Although the injury may be caused by negligence, the sponsor may face some liability on the bases of negligent hiring, failure to properly vet investigators, or negligent training or supervision of trial sites (Medmarc, 2022). Investigators with training regarding the consent process and ongoing consent obligations as well as with training on the unique considerations and ethical issues related to research involving pregnant and lactating women may be more prepared to conduct studies with these populations. Sponsors can mitigate their liability

___________________

³ Revised Policy on Inclusion of Women of Childbearing Potential in Clinical Trials, 58 Fed Reg. 39408 (Jul. 22, 1993).

⁴ Revised Policy on Inclusion of Women of Childbearing Potential in Clinical Trials, 58 Fed Reg. 39408 (Jul. 22, 1993).

⁵ Elements of Informed Consent, 21 CFR 50.25(b)(5) (Jan. 27, 1981).

by carefully selecting experienced and well-trained or well-supervised investigators and by ensuring that all staff assisting with the conduct of the study are aware of their obligations in the safe conduct of the study.

Concerns about selecting qualified investigators is particularly critical for research involving pregnant and lactating women, given the complexity of conducting these trials. However, as discussed in Chapter 5, there is a lack of expertise among researchers, members of IRBs, institutional leaders, and other stakeholders in conducting research with pregnant and lactating women. Addressing this lack of expertise in research involving pregnant and lactating women can be an important factor in mitigating potential liability.

Clinical trial monitors may also be hired or engaged by sponsors to monitor the conduct of the trial, including data quality, informed consent procedures, safety of participants, and quality assurance (Love et al., 2022). Engaging monitors in rigorous external monitoring of clinical trials involving pregnant and lactating women may be another approach to mitigation of liability.

MITIGATING LIABILITY FOR INSTITUTIONS, IRBs, AND INVESTIGATORS

Most liability mitigation strategies proposed in previous reports have primarily focused on mitigating liability for sponsors. However, clinical research institutions of all types, IRBs, and investigators also have liability concerns that warrant consideration. Those concerns must be addressed for them to overcome their reticence to conduct or approve trials involving pregnant and lactating women. There are several liability mitigation strategies that could be considered.

Informed Consent

Although an earlier section in this chapter discusses the role of research participants’ informed consent in mitigating liability for industry sponsors, this section provides an overview of steps in the informed consent process that reduces the risk that research institutions, IRBs, and investigators will be held liable for harm to participants. Under federal regulations, the clinical investigator has the main responsibility for providing and documenting each participant’s informed consent. IRBs are responsible for determining the adequacy of the process design and the content of the consent form, and clinical research institutions are responsible for overseeing the process. Here, too, a carefully prepared plan governing the informed consent process can be a mitigating factor for liability. Unique regulatory considerations for informed consent in

research with women of childbearing potential and pregnant women include the following:

• The informed consent process must disclose potential risks and the chance of unknown risks to an embryo or fetus, should a participant be or become pregnant.⁶
• For studies that may involve pregnant women, informed consent should include a discussion in appropriate lay language about any completed animal reproductive toxicity studies and their significance for humans.⁷
• Subpart B of the U.S. Department of Health and Human Services (HHS) human research regulations, which provides additional protections to pregnant women, fetuses, and neonates in research, states that the individuals who provide consent for such research should be informed of the reasonably foreseeable effect on the fetus or neonate. Subpart B also requires that when research is conducted for the benefit of the fetus alone, informed consent must be obtained from both parents (though there are exceptions to the paternal consent requirement, as described in Chapter 3).⁸ Ethical objections have been raised to the regulations that require paternal consent, especially because this rule for unborn fetuses is more burdensome than the rule in Subpart D (which contains provisions governing pediatric research), which requires dual-parent consent only for studies that offer no prospect of direct benefit to a born child and present greater than minimal risk (Little et al., 2018). As a practical matter, getting informed consent would make it unlikely that a father could succeed in objecting after the fact to the inclusion of the fetus in research that caused an injury. Nonetheless, the evidence needed either to support or to disprove this supposition is lacking because of the paucity of such research during pregnancy and the difficulty of knowing how many possible cases were resolved without a formal cause of action being brought.
• FDA regulations provide that if an IRB regularly reviews research involving pregnant women, the IRB must consider including one or more members who are knowledgeable about and experienced in working with this population.⁹ Extending this paradigm, if

___________________

⁶ Precautions in Clinical Trials Including Women of Childbearing Potential, 58 Fed Reg. 39411 (g) (Jul. 22, 1993).

⁷ Precautions in Clinical Trials Including Women of Childbearing Potential, 58 Fed Reg. 39411 (g) (Jul. 22, 1993).

⁸ Research Involving Pregnant Women or Fetuses, 45 C.F.R. 46.204.

⁹ IRB Membership, 21 C.F.R. 56.107(a).

• other IRBs and oversight bodies—such as expert consultants, data safety monitoring boards (DSMBs), radiation safety committees, or data security committees—are charged with reviewing the study protocol and its implementation, it may also be important for them to include among their membership individuals with knowledge and expertise in pregnancy.
• Special considerations may arise when the trial is expected to involve pregnant minors. Some states consider a pregnant minor as “emancipated”—that is, able to provide consent to medical care, among other legally binding contracts—a status that the minor often loses after delivery.
• Although not unique to research with pregnant women, HHS regulations require that for research involving more than minimal risk, the researcher must provide an explanation of whether any compensation or medical treatments are available if injury occurs and, if so, what such compensation consists of or where further information may be obtained.¹⁰ It may be helpful to provide to research participants the information of the person or office that participants can contact should they experience an event that requires compensation. Investigators and their research teams must be made aware of the compensation process.

Self-Indemnity Programs

Indemnity is “a promise made by one party to another that it will cover any loss suffered by the other party” (International Society of Nephrology, 2022). One example of an institution using a self-indemnity fund for research-related injuries is the University of Washington (UW). Originally, UW purchased a commercial insurance plan to provide compensation for research-related injuries, which it had from 1972 to 1979. However, the cost in insurance premiums far surpassed the amount the university was paying in claims.¹¹ Therefore, in 1979, UW created a self-funded no-fault plan that covers up to $10,000 in out-of-pocket research-related injuries and up to $250,000 for care received at UW Medical Center (Henry et al., 2015).¹² The compensation collected in the university’s indemnity pool is intended to provide necessary medical care to subjects who sustain a bodily injury directly from participation in a research project or trial funded by UW (Henry et al., 2015). Although the committee was not able to access information on the number of claims, the Director

___________________

¹⁰ General Requirement for Informed Consent, 45 CFR 46.116 (b)(6).

¹¹ As presented to the committee in open session by Jason Malone on March 23, 2023.

¹² As presented to the committee in open session by Jason Malone on March 23, 2023.

of the Human Subjects Division at UW told the committee in open session that the university does not currently receive many claims.¹³

In a qualitative study of factors to facilitate research with pregnant women, a respondent from UW noted that “UW Medical Center has fewer research-related lawsuits and tort claims than do comparable institutions” (Mastroianni et al., 2020). This style of a compensation fund is used in other academic institutions but is typically more modest and provides limited coverage for those injured during research projects or clinical trials (Henry et al., 2015). With self-indemnity funds, the claim-filing process can be simple and provide easy and accessible coverage, but this is only possible when the fund is implemented correctly. In addition, the background of the funding of self-indemnity compensation schemes is also sometimes unknown and may lead to concerns over where the money is coming from to support the fund (e.g., donations).

In the clinical setting, evidence from the study of medical errors and medical malpractice suggests that compensation for harm may be an effective solution to mitigate liability for institutions and their investigators. A model that provided compensation, along with disclosure of a medical error, implemented within the University of Michigan Health System was associated with fewer lawsuits, shorter durations between claims and claim resolutions, and decreased institutional costs for liability payments (Kachalia et al., 2010). A similar program of compensation and disclosure at the Veterans Affairs Medical Center in Lexington, Kentucky, was found to be an effective solution to limiting costs related to liability payments (Kraman and Hamm, 1999). An analysis of compensation and disclosure programs found that an offer of compensation did not influence harmed individuals’ interest in seeking legal advice, though the study did not evaluate whether harmed individuals proceeded to file a claim (Murtagh et al., 2012).

Agency-Affiliated Compensation Systems

Programs affiliated with federal government agencies include compensation systems covered by government institutions such as the Department of Defense (DoD) Clinical Investigation Program (CIP) or the Department of Veterans Affairs (VA) Office of Research and Development. The CIP program provides compensation for research-related injuries for all DoD-sponsored research. The VA Office of Research and Development provides compensation for research injuries for research approved by a VA IRB and conducted under VA supervision (Henry et al., 2015). These programs’ injury claims are submitted to their specific

___________________

¹³ As presented to the committee in open session by Jason Malone on March 23, 2023.

research offices, and then funding is adjudicated by the specific offices. For example, the CIP accepts claims and then assists the injured person to the correct facility for care. Most, if not all, agency-affiliated programs are funded by Congress, but the amount of funding for injury compensation is determined by the agencies themselves. Although these systems limit care to specific medical facilities, it provides some mitigation of liability for investigators and institutions that are covered by these policies.

Regulatory Guidance

There is no federal regulatory requirement that pregnant or lactating women be included in clinical trials; however, there is also no blanket prohibition against their inclusion. As discussed in Chapters 2 and 3, HHS regulatory requirements related to the inclusion of pregnant women in research are found in 45 CFR 46, Subpart B, “Additional Safeguards for Pregnant Women, Human Fetuses, and Neonates Involved in Research.” Subpart B, as it is commonly known, includes directives regarding acceptable levels of research-related risk in research with pregnant women (Subpart B does not contain language on the inclusion of lactating women). In practice, however, the precise meaning and applications of the regulatory terminology around research-related risk, as well as its relationship to the prospect of research-related benefit, are subject to different interpretations by IRBs as well as investigators. The ambiguity and resulting uncertainty related to risk assessment have contributed to the exclusion of pregnant women from clinical trials (Krubiner et al., 2016; Mastroianni et al., 2017; ORWH et al., 2010; van der Zande et al., 2017).

Risk assessment is essential to the regulatory oversight and ethical conduct of clinical research, with implications for the permissibility of research as well as informed consent. Debates and variations in application of regulatory definitions of risk, particularly minimal risk, are not exclusive to research in pregnancy (e.g., Kopelman, 2004; Resnik, 2005). Minimal risk is defined in HHS regulations to mean

The regulations do not specify, for example, whether the daily life risk threshold is determined in reference to a healthy research participant,

___________________

¹⁴ Additional Protections for Pregnant Women, Human Fetuses and Neonates Involved in Research, 45 CFR 46.102(i) (2018i) (i).

a patient participant, or a healthy person. The regulations also can be interpreted to suggest that cumulative risk over many days is not relevant to decision making. Even where there is agreement on a standard, risk perception and characterization vary among individuals and stakeholders, leading to inconsistencies in application.

Those same debates and ambiguities become more complex in the context of pregnancy studies because they hinge on additional considerations of fetal risk. Specifically, Subpart B provides that a study that proposes no “prospect of direct benefit” to the pregnant woman or fetus can be performed if fetal risk is assessed to be “not greater than minimal.” A study that proposes a prospect of direct benefit to the pregnant woman, the fetus, or both can go above the minimal risk threshold.¹⁵ National-level approval by HHS is required for research that does not fit those requirements, although reportedly none has been sought (Saenz et al., 2017). Should an assessment of fetal risk take into consideration that the fetus of a pregnant woman with a medical condition may already be at elevated risk compared to the fetus of a healthy pregnant woman? Should the “daily life” risk standard account for whether a pregnant woman, and by extension her fetus, live in an unhealthy or dangerous environment, relative to other participants? Notably any such determinations would be inherently subjective.

The definitions of both risk and benefit are subject to variable interpretations, and little regulatory guidance exists, particularly for minimal risk (Blehar et al., 2013; Mastroianni et al., 2017; NVAC, 2017). In practice, this lack of clarity often results in conservative interpretations by decision makers that discourage conducting research with pregnant women (Mastroianni et al., 2017). IRBs do not have sufficiently clear guidance to evaluate appropriate study designs and safeguards for including pregnant women in clinical research that would permit their inclusion (Krubiner et al., 2016; Strong, 2011; White et al., 2021).

As more research with pregnant women is conducted, IRBs would benefit from assistance interpreting Subpart B from the HHS Office for Human Research Protections (OHRP). As discussed in Chapter 3, OHRP can help IRBs provide feedback on research protocols involving pregnant and lactating women and minimize harm to research participants by providing IRBs with clear guidance on safely conducting research in those populations. Further, guidance from OHRP on interpreting Subpart B can help to mitigate liability for regulatory-compliant institutions and their IRBs, in the same way that FDA guidance, discussed above, can mitigate liability for regulatory-compliant sponsors.

___________________

¹⁵ Additional Protections for Pregnant Women, Human Fetuses and Neonates Involved in Research, 45 CFR 46.204.

MITIGATING LIABILITY FOR SPONSORS, RESEARCH INSTITUTIONS, AND INVESTIGATORS

Although there are mitigation strategies that differentially mitigate liability for industry sponsors, investigators, and institutions, some new and existing programs mitigate liability for all these stakeholders. Some of these concepts, such as clinical trial insurance, are currently in use by many stakeholders conducting clinical research but could be expanded to provide coverage for research involving pregnant and lactating women. Other strategies, such as a no-fault system for research-related injuries for pregnant and lactating women and tort reform, do not currently exist but could be created to mitigate liability for research stakeholders.

Clinical Trial Insurance

Clinical trials insurance covers the costs of compensation and legal fees if a participant suffers an injury or harm as a direct result of taking part in a clinical trial. It can be an essential tool in helping sponsors and others involved in the conduct of clinical trials protect their reputation, assets, and future research endeavors from potential liability and losses that may result from clinical trials. Such insurance may also help gain trust and confidence from the participants, regulators, investors, and other partners involved in clinical trials. Most policy holders are manufacturer–sponsors of clinical trials, but universities and other research institutions can also purchase insurance.¹⁶ However, many large sponsors of clinical research often self-insure up to a certain amount and purchase reinsurance for large trials with potentially large liabilities. Most insurance policies extend beyond the sponsor to protecting other entities in the trial, such as IRBs, investigators (assuming protocols are followed), and contract research organizations.¹⁷

Although most payments from insurance come from a claim being filed and either settled or through the finding of fault in the tort system, most insurance policies provide some limited coverage for medical expenses if a participant is injured during a clinical trial.¹⁸ In the event that an injury or illness occurs as the result of participation in the trial, including exposure to the treatment under study, the insurance policy covers the medical expenses up to a sublimit for any subject who is treated in a medical facility, called “med pay” (Dyson, 2023). The med pay system is usually a no-fault system, but the sublimit for expenses is usually fairly low.

___________________

¹⁶ As presented to the committee in open session by Jason Malone on March 23, 2023.

¹⁷ As presented to the committee in open session by Jason Malone on March 23, 2023.

¹⁸ As presented to the committee in open session by Jason Malone on March 23, 2023.

Clinical trial insurance may mitigate liability for institutions and investigators by providing even modest compensation for research-related injuries and providing liability coverage should claims arise. However, institutions may be hesitant to purchase these plans because of their cost. If investigators were able to use federal grant funds to purchase clinical trial insurance, this would provide insurance for the investigator and their institution should any claims arise, therefore mitigating liability for the institutions and investigators. NIH currently covers clinical trial insurance as an allowable expense when clinical research sites are in countries that require clinical trial insurance (Henry et al., 2015). This policy could be extended to U.S. trial sites, allowing more investigators to purchase clinical trial insurance.

Obtaining trial insurance for research that includes pregnant women may be more difficult than for research that does not include pregnant women because of the uncertainty or potential severity of the risks involved (Manningham-Buller and Brocklehurst, 2022; Mastroianni et al., 2017). For example, the sponsors of an Ebola treatment trial were unable to get insurance coverage if they included pregnant women, despite strong recommendations from the World Health Organization that this population be included (Gomes et al., 2017). Taking steps to mitigate risk, many of which overlap with liability mitigation strategies, may help insurance underwriters feel more comfortable underwriting for research involving pregnant and lactating women and may result in better terms and conditions from the insurers. Because most trials exclude pregnant and lactating women before attempting to obtain insurance, challenges with obtaining insurance have so far not been as prominent an obstacle as other factors (NASEM, 2023).

Current federal regulations do not require researchers, institutions, or sponsors to provide medical care or compensation to those who are injured during clinical trials, and this has caused most academic institutions and government agencies to opt out of providing a compensation system (Henry et al., 2015). A study done in 2014 found “More than half of the U.S. research institutions surveyed do not offer free medical care or other compensation for research-related injuries,” with less than 5 percent offering an “unconditional compensation” for harmful effects or injuries that resulted from the experimental intervention (Henry et al., 2015). While there was an increase in the number of institutions providing coverage from 2000 to 2015, the overall drive to develop a compensation system is unlikely to change without government pressure (Resnik et al., 2014).

No-Fault Administrative Compensation Systems

No-fault compensation systems include the National Vaccine Injury Compensation Program (VICP), described in detail in Box 4-1, and the

Countermeasure Injury Compensation Program (CICP). The CICP was established in 2005 by passage of the Public Readiness and Emergency Preparedness Act.¹⁹ In the event of a public health emergency or security dangers that threaten the United States, the government will support the development of countermeasures, and these can be in the form of vaccines, medications, medical devices, diagnostic test kits, or other items used to diagnose, prevent, or treat the emergency event (HRSA, 2023b). The CICP compensates for severe injuries or deaths that occur from the administration or use of a specific countermeasure (HRSA, 2023b). Some previously covered public health threats were COVID-19, Ebola, pandemic influenza A, smallpox, and anthrax (HRSA, 2023b). One important factor with the CICP is that liability protection is only enacted when a public health emergency declaration is made, it is only provided as a last resort for those seeking compensation, and the available funds are adjusted based on the type and severity of the emergency (HRSA, 2023b). No-fault systems have been particularly useful for granting compensation to injured participants as a finding of negligence is not necessary to receive compensation (Weiler, 1993).

Allowing research participants to choose an alternative to the tort system—such as a national system of no-fault compensation for injuries related to research involving pregnant and lactating participants—could achieve several aims. First, it would increase the likelihood that pregnant and lactating participants and their offspring or surviving kin would receive some financial recompense for harms that were probably caused by their participation in the research (Mariner, 1994). Second, such a system would redirect some—but not necessarily all—claims of harm away from the courts and into the no-fault system, which, by definition, does not assign blame to, or impose liability on, the individuals or institutions that sponsored or conducted the research, which means that the harms to reputation and morale caused by a finding of fault would be absent. Third, a no-fault compensation program for research-related injuries could diminish the fear those conducting the research have of large and unpredictable jury damage awards; payments by the research sponsors to fund the no-fault system would be more predictable and probably much smaller.

However, there are limitations to what a no-fault compensation program can accomplish. First, the amount of financial compensation through a no-fault system tends to be less than what could be awarded through the tort system (Engstrom, 2011). In addition, unlike the tort system, some no-fault compensation systems lack the element of deterrence. Whereas the tort system incentivizes sponsors, research institutions, and investigators

___________________

¹⁹ Public Readiness and Emergency Preparedness Act, Public Law 109-148, 109^th Cong., (December 30, 2005).

to optimize quality and minimize harm, no-fault compensation systems, depending on their design, may weaken or eliminate that incentive.

Since the evidence outlined in Chapter 2 points to limited liability for research with pregnant women and virtually no liability for research involving lactating women and there is not a national no-fault compensation program for clinical research generally, the most compelling argument for a no fault-compensation program for research involving pregnant and lactating women is that it is a better mode for compensation for research-related injuries, rather than a way to mitigate liability. While the committee supports compensating pregnant and lactating women and their offspring who are harmed when contributing to the societal benefit of clinical research, it has struggled to find reasons to create a national no-fault compensation program solely for pregnant and lactating research participants and their offspring. As many others have advocated (HEW, 1977; Mariner, 1994; Research, 1982), the committee favors a national no-fault compensation program for all clinical research participants, including pregnant and lactating women, but it is beyond this committee’s charge to recommend such an all-encompassing system.

Although the committee is not recommending a national no-fault compensation program, some of the practical considerations for implementation of such are worth noting. There are numerous practical challenges to implementing a no-fault compensation program for research including pregnant and lactating women, particularly in regard to their offspring. A special challenge involves separating compensable research-related injuries from cases in which fetal demise or a child’s congenital condition is unrelated to research participation (Mariner, 1994).

Discerning whether a child’s injuries arose from or merely during parental participation in clinical research during pregnancy or lactation could be difficult or impossible because of the following reasons:

• Normal conception, gestation, and birth produces a wide range of congenital abnormalities or poor outcomes at baseline in offspring, including many that only become apparent later in childhood, but the occurrence of which led some parents to search for a singular cause to blame.
• The manifestation of injuries that affect prenatal development may be separated in time from exposure to the drug in the clinical trial.
• Rather than a defined set of adverse consequences (as is, for example, the case with childhood vaccinations, for which a federal compensation program exists), the wide variety of medical products in question would create a much larger list of conditions in children, some of which might be very rare, which could lead to prolonged and complex investigations into causation and in some cases a

• child who suffered an injury in utero because of an investigational product would (unfairly) not be compensated by the system. The latter problem could be addressed by lowering the standard of proof of causation in the no-fault system (Mariner, 1994).

Although issues of causation are a challenge in the tort system as well, the tort system is set up to litigate individual cases to establish causation. To establish causation for individual cases on a national scale would drastically slow down payments in a national compensation program, leading to many of the same issues seen in the tort system. In fact, as covered in Box 4-1, this is what has happened with the VICP, leading to slower payments for claims than the tort system (Engstrom, 2015).

Tort Reform

Given the problems with the tort system and the potentially large awards that may result from the system, some have suggested tort reforms as a potential solution to mitigate liability (PRGLAC Task Force, 2020). Tort reform involves any attempt by lawmakers to make it more difficult for plaintiffs to file lawsuits or limit the amount of compensation a plaintiff may recover when filing a lawsuit. However, the tort system is already a difficult-to-navigate and inequitable system for some plaintiffs who have a meritorious claim (Franklin, 1967; Lytton et al., 2011; Mello, 2023).

There is a lot of imprecision in who is awarded money through the system. Unfortunately, not everyone with a meritorious claim is awarded compensation, and inequities exist among claimants with similar injuries (Lytton et al., 2011). This is partially because the tort system can prove inaccessible to injured parties, leading to only major cases being brought because of the expense of bringing a lower severity case. There is also an equity problem with the tort system, as the lower a person’s income, the lower the economic damages awarded (Paez and Liscow, 2022). If an attorney is working on a contingent fee, then lower damages awards mean a less attractive plaintiff for attorneys (Mello, 2023). In a survey of attorneys, over half of the attorneys were not willing to accept a case unless the expected damages were at least $250,000, even if they were almost certain to win the case (Sheperd, 2014). In cases with a less certain outcome, most attorneys required a minimum of $500,000 in damages to accept the case.

Therefore, the committee considered, but does not recommend, a mitigation strategy that places a cap on liability for investigators conducting research on therapeutics used during pregnancy and lactation (PRGLAC Task Force, 2020). There is some evidence that such tort reforms

do reduce levels of malpractice litigation, and they certainly reduce the amounts of damage awards (Yu, 2017). However, there is also evidence that the decrease in litigation caused by such reforms does not reduce provider anxiety about liability. At the same time, those reforms increased the likelihood of undercompensation for research-related injuries (DeVito and Jurs, 2014) and have a disproportionate effect on awards for those with the most serious injuries, as only the most seriously injured will have noneconomic damages that meet the limits set by the caps (Hubbard, 2020).²⁰ Tort reforms are also arguably more likely to disproportionately disadvantage women, children, and the elderly, who are more likely than men to have a greater proportion of their total damages come from noneconomic loss, such as emotional distress and grief, altered sense of self, impaired relationships, and more. Wages lost and health care expenses, which are more likely to be awarded to men of working age, have largely not been included in conversations around tort reform (Finley, 2004). As noted by one law professor, “by limiting noneconomic damages relative to economic damages, states may disproportionately reduce damage payments to women” (Sheperd, 2008). Finally, there is also some evidence that there is an increase in adverse patient safety events following adoption of damage caps (Zabinski and Black, 2022), which suggests that reform may undermine one of the key goals of the tort system: to deter negligent conduct.

MITIGATING CRIMINAL AND CIVIL LIABILITY FOR PREGNANT RESEARCH PARTICIPANTS

Complicated privacy concerns have long been an issue for research involving pregnant women, often stemming from a state’s stated interest in minimizing any risk to fetuses while in utero. For example, an IRB at the University of South Dakota encountered such privacy issues when the IRB was presented with a protocol for a five-state study of fetal alcohol syndrome that involved identifying and monitoring women who drink during pregnancy. South Dakota law, however, requires officials to report behavior the state defines as abusive toward a fetus, including drinking alcohol. At that time, investigators were unable to offer research participants a certificate of confidentiality or other privacy protection because of state law. As a result, women who volunteered for the study were at risk of being reported to state officials and potentially facing legal repercussions because of their substance use while pregnant. Ultimately, the governor’s office wanted the study to proceed because its objectives

___________________

²⁰ N. Broward Hosp. Dist. V. Kalitan, 219 So. 3d 49 (Fla. 2017).

involved a positive intervention—helping pregnant women with drinking problems with educational interventions intended to help them maintain sobriety. Under the state’s decision, the women would still be reported to the state, but the state would take no action against participants of the study (IRB Advisor, 2003).

As covered in Chapter 2, the U.S. Supreme Court decision in Dobbs v. Jackson Women’s Health Organization overturned its previous rulings that the U.S. Constitution protected the right to an abortion. Post-Dobbs, the breadth of privacy issues may increase as states propose and enact new laws aimed at preventing abortion, protecting fetal life, and regulating the bodies and choices of pregnant women. The current legal environment, including its instability, underscores the importance of protecting the confidentiality of all information about trial participants’ pregnancies and use of abortion services (Appendix E).²¹

The post-Dobbs climate may affect how researchers record pregnancies among subjects and whether and how that information is protected from disclosure. In many clinical trials involving nonpregnant subjects, initial and periodic pregnancy tests are a standard part of trial protocol. These tests are deemed necessary when a trial’s protocol requires exclusion of pregnant women, yet they may also detect early pregnancies that would have otherwise gone unnoticed because of high rates of first trimester miscarriages. A positive pregnancy test during the course of a trial is typically considered a “reportable event,” so participants must be willing to report their pregnancies and feel secure doing so, particularly if they are considering an abortion. According to Aoife Brennan, chief executive officer of Synlogic:

According to a recent analysis on potential implications of Dobbs, “the simple fact that a research participant is not pregnant nor has given birth, but a test indicates that they were pregnant during research, could put them at risk of legal action” (Sugarman et al., 2023).

In most if not all cases, the information reported to states maintains the patient’s confidentiality and does not provide their name or other personally identifiable information. However, where a state has banned or severely restricted abortion, state officials may seek such identifiable

___________________

²¹ Appendix E can be viewed online at https://nap.nationalacademies.org/catalog/27595.

information in pursuit of criminal charges, which may involve participants, investigators, or the investigators’ institutions.

Overall, the risk of criminal and civil liability will likely increase post-Dobbs for participants, investigators, and their institutions involved in research with pregnant women. This will be particularly true in states with fetal personhood laws, fetal homicide laws, or where state child abuse statutes have been interpreted to encompass risky behavior by a pregnant woman that may affect her fetus (Appendix C).²² Researchers have also posited that:

The potential for criminal and civil liability depends on how far states are willing to push their antiabortion and fetal protection laws. While some states may limit their actions to research explicitly studying drugs intended to induce an abortion, others could go further, seeking to impose liability on those involved in clinical research that may harm a fetus or result in fetal death. The liability could stem from a state’s abortion laws, fetal personhood laws, child endangerment and abuse laws, or other criminal laws.

Certificates of confidentiality (CoCs) provide an important tool to protect research participants against privacy breaches. CoCs were created to provide participants with greater certainty that their privacy concerns are addressed, so participants who have such concerns would be willing to participate in research (UVA, n.d.). CoCs likely provide privacy protections in many of the contexts involving pregnant women in clinical research. However, many research institutions, IRBs, and investigators do not understand the full statutory power of CoCs and therefore discount the privacy protections that they afford. These stakeholders may also misunderstand the complexities of the privacy protections that they afford and may unwittingly undermine the privacy protections provided.

The CoC is a federal statutory device that protects identifiable, sensitive information collected during “biomedical, behavioral, clinical, or other research” from compelled disclosure (UVA, 2019). Specifically, if a law enforcement officer, prosecutor, legislator, civil litigant, or other party seeks to compel information about a research participant through a subpoena or warrant, a CoC allows the researcher to refuse the disclosure and bars the use of that information as evidence. By protecting researchers

___________________

²² Appendix C can be viewed online at https://nap.nationalacademies.org/catalog/27595.

and institutions from being compelled to disclose information that would identify research subjects, a CoC can help achieve the research objectives and promote participation in studies by assuring confidentiality and privacy to participants.

A recent analysis demonstrates that legal challenges to the privacy protections afforded by CoCs will likely fail, with the possible exception of challenges based on constitutional criminal defense rights (Ram et al., 2022). CoCs are therefore a remarkably strong tool for protecting privacy rights, and there is reason to believe that they can be relied upon to protect pregnant participants’ privacy. In research contexts, CoCs add a layer of protection that may not be available under the Health Insurance Portability and Protection Act (HIPAA). HIPAA allows a covered entity to refuse a subpoena or a warrant; HIPPA does not compel a covered entity to do so. In contrast, recent amendments to the law governing CoCs prohibit researchers from disclosing “any identifiable, sensitive information about [an] individual . . . that was created or compiled for purposes of the research.”²³ Moreover, although the statute that authorizes CoCs does permit disclosures “as required by federal, state, or local laws” (e.g., public health reporting requirements),²⁴ researchers may not do so unless they have obtained the consent of the participant.²⁵ Nonetheless, researchers in states where their institutions may be subject to political or other pressures to comply with state- or court-mandated demands for information should also recognize that their efforts to protect the private information of their participants may be hampered by institutional pressures.

Legally, the issuance of CoCs is automatic for all NIH-funded research that collects or uses identifiable, sensitive information; research funded by agencies other than NIH may be granted a CoC automatically through the funding agency if the agency issues them. If the research is funded by an agency that does not issue CoCs, investigators may apply to NIH for a CoC. Researchers not engaged in federally funded research are eligible to apply for a CoC at NIH. As a result, large volumes of research data are now covered by CoCs and therefore may be beyond the reach of state and federal law enforcement, legislative bodies, and other authorities. CoCs provide some reassurance to participants that their data are safe and protected from disclosure or use in legal proceedings. As a result, participants may feel more comfortable about participating in research.

___________________

²³ Research and investigations generally, 42 U.S.C. §241(d)(1)(D).

²⁴ Research and investigations generally, 42 U.S.C. §241(d)(1)(C)(i).

²⁵ Research and investigations generally, 42 U.S.C. §241(d)(1)(C)(iii).

CONCLUSIONS

REFERENCES

This page intentionally left blank.