Background.—Forensic pathologists face difficult moral questions in their practices each day. Consistent ethical and legal guidelines for autopsy tissue use extending beyond usual clinical and legal imperatives have not been developed in this country.
Objective.—To obtain the perceptions of medical examiners regarding the ethical acceptability of autopsy tissue use for research and education.
Method.—A written, self-report questionnaire was developed and piloted by a multidisciplinary team at the University of New Mexico, Albuquerque. All individuals who attended a platform presentation at the National Association of Medical Examiners Annual Meeting in September 1997 were invited to participate.
Results.—Ninety-one individuals completed the survey (40% of all conference registrants and approximately 75% of presentation attendees). Sixty-three percent of respondents had encountered an ethical dilemma surrounding autopsy tissue use, and one third reported some professional ethics experience. Perspectives varied greatly concerning the ethical acceptability of using autopsy tissues to demonstrate or practice techniques (eg, intubation, brachial plexus dissection) and of fulfilling requests to supply varying kinds and quantities of tissues for research and education. Most respondents indicated that consent by family members was important in tissue use decisions. Respondents agreed on the importance of basic values in education and research, such as integrity, scientific or educational merit, and formal institutional approval of a project. Characteristics of the decedent did not influence decisions to release tissues, except when the individual had died from a mysterious or very rare illness. Attributes of medical examiners, with the exception of sex, also did not consistently predict responses.
Conclusion.—Significant diversity exists in beliefs among medical examiners regarding perceptions of the appropriate use of autopsy tissues for education and research. There is need for further inquiry and dialogue so that enduring policy solutions regarding human tissue use for education and research may be developed.
The autopsy has long been a focus of medical education and biomedical research. Generations of physicians have learned anatomy, pathology, and pathophysiology at the autopsy table. Furthermore, autopsy tissues have been used to help identify new diseases, to clarify the mechanisms and patterns of illness, to explore treatments that may ameliorate suffering, and to trigger and sustain advances throughout the basic sciences.1–5 Despite this long tradition of cadaveric tissue use, there has been an 80% decline in the rate of hospital autopsies during the past 50 years.3–10 Consequently, cadaveric tissues used in medical education and research increasingly derive from medicolegal autopsies performed by forensic pathologists.8
The ethical acceptability of using medicolegal autopsies for education and research is complex. This is because such autopsies are performed in compliance with legal imperatives that supersede ethical safeguards present in practice standards for hospital autopsies and other areas of clinical medical practice.8,11–14 For instance, the concept of privacy refers to the belief that there is an area of personal life, encompassing both body and mind, that should be respected and free of intrusion.12 This ideal gives rise to the physician's duty to preserve patient confidentiality (ie, to not disclose patient information without permission) to the extent allowed by law.14 When a death occurs under circumstances that mandate a medicolegal autopsy (eg, deaths due to violence or suicide), the purpose of the autopsy is to determine the condition of the body and the precise cause and manner of death. In compliance with legal statutes, this information is then documented and formally reported and may be accessible to all interested individuals.1,15–17 In the context of the medicolegal autopsy, exceptions to preserving confidentiality have been seen as necessary and defensible, if regrettable, for the immediate good of society. Similarly, informed consent is a philosophical and legal doctrine that requires that individuals truly understand and freely make choices about “intrusions” on their bodies and minds, such as their acceptance of recommended medical treatments or their personal involvement in biomedical experimentation.11,18 As with confidentiality, informed consent is an ideal19–23 that is difficult to achieve in the context of medicolegal autopsies, since they are performed to meet statutory requirements and, unlike hospital autopsies, are not voluntary. For these reasons, the usual ethical guideposts used by clinical pathologists and clinicians in other areas of medicine do not apply readily or fully to the circumstance of the medicolegal autopsy. Consequently, the safeguards protecting confidentiality and enhancing consent that allow for ethical use of cadavers and cadaveric tissues from hospital autopsies for education and research are not in place. For many, the ethical acceptability of the use of medicolegal autopsies for these purposes remains uncertain.
Thus, forensic pathologists, medical examiners, coroners, clinicians, and consultants who work with medicolegal cadaveric tissues face a number of moral questions in their practices. For example, what are the limits of personal and family wishes when an individual's death was caused by events of critical importance to science or society? Is it acceptable to perform scientifically valuable studies that are not essential to determine the condition of the body and the cause and manner of death? Under what circumstances should medicolegal autopsy specimens be used in education of medical students and residents, emergency medical technicians, and other professionals? Is it ethically justifiable to teach certain procedures, such as resuscitation or surgical techniques, on the newly and nearly dead?24,25 Who should have access to tissues of individuals who have died from rare or mysterious illnesses? In sum, when do the needs of the living properly outweigh ethics protections ordinarily accorded to the dead?1
Such questions have increasingly been raised in recent years,1,3,8,13,16,26–29 and little agreement has been reached with respect to ethical, regulatory, and legal guidelines for use of human autopsies for education and research in this country.3,4,14,16,27 An important step toward consensus is a recent statement published by the Ad Hoc Committee on Stored Tissue from the College of American Pathologists, which outlined several ethics recommendations related to genetic testing on cadaveric tissues, the need for confidentiality and consent policies and procedures in pathology departments, the role of institutional review board approval for research and teaching endeavors involving pathology specimens, and related issues.27 This document does not focus specifically on legally mandated autopsies, and insights gathered in other areas of medicine, such as the ethics of human research, the ethics of fetal tissue use in clinical and experimental endeavors, and the ethics of organ transplantation, have not been adapted or applied to the unique situation of the medicolegal autopsy.13,14,17,30–34 The federal regulations governing human research outline ethics safeguards for living human subjects and do not adequately address the special issues related to the “no longer living.”35 Furthermore, legal guidelines may conflict across regions of the country, and they offer relatively little assistance in clarifying what a health professional ought to do, as distinct from what he or she must do, in everyday practice. As a consequence, forensic pathologist, medical examiner, and coroner activities may vary widely in terms of offering or limiting access to tissue for a wide array of educational and scientific projects.29
To our knowledge, no empirical work has been performed to explore the views and practices concerning the use of medicolegal autopsy tissues for research and education. To remedy this gap, we surveyed attendees of the 1997 Annual Meeting of the National Association of Medical Examiners (NAME), the principal professional organization for practicing forensic pathologists in the United States. Four hypotheses were tested: (1) that the ethical acceptability of fulfilling requests to provide access to bodies at autopsy would differ according to the nature of the educational or research request; (2) that different quantities and kinds of tissues in such requests would be viewed as having different ethical acceptability; (3) that several issues (eg, consent of the decedent or of family; age, diagnosis, and ethnicity of the decedent; integrity of the individual researcher; and scientific merit of the project) would influence how respondents view autopsy tissue requests; and (4) that respondent characteristics, such as sex, education, practice setting, and level of training, would influence responses.
Subjects, Design, Procedure, and Survey Instrument
A 6-page self-report instrument was developed for this project by a multidisciplinary team with expertise in anatomic and forensic pathology, clinical ethics, psychiatry, psychology, statistics, and research design. The questionnaire assessed the views and experiences regarding the ethics of using human autopsy tissues for research and educational purposes. Responses to all items were ratings on 7-point scales, with each extreme value labeled from 1 (“completely acceptable”) to 7 (“completely unacceptable”). Five case vignettes were developed to reflect tissue use issues confronted by forensic pathologists. Demographic and professional characteristics were also assessed. The instrument was revised based on pilot testing with faculty and trainees in the Department of Pathology at the University of New Mexico School of Medicine.
This study used a purposive sample of participants at NAME held in September 1997 in Baltimore, Md. After a brief verbal introduction, those who attended a conference platform presentation on the topic of the ethics of tissue use for education and research were asked to complete the questionnaire. Participant voluntarism and anonymity were assured. The project was approved prospectively by the institutional review board at the University of New Mexico School of Medicine and by the executive committee of NAME.
Descriptive data tabulated and analyses performed using SPSS 9.0 for Windows36 are reported in Tables 1 through 9. Separate repeated-measures multivariate analyses of variance (MANOVA) were performed on the ratings for sets of items (organized by the tables) that were conceptually related. Between-groups variables (eg, sex of respondent, region of country in which respondent resided, grouped age, religion, ethnicity, experiences with ethical issues) were also included in these analyses, but their effects were found not to be statistically significant, and thus they are not reported. Finally, χ2 tests were performed on the proportions of respondents who would or would not supply tissues under the conditions specified in the case vignettes.
Characteristics of Survey Respondents
Ninety-one individuals completed the survey, yielding a response rate of 40% for all conference registrants and approximately 75% of platform presentation attendees. Because the current number of full-time forensic pathologists in the United States is estimated to be about 400,37 our sample represents about 17% of the national population of medical examiners. Respondents were 76% male, 92% white, and averaged 49 years of age (Table 1). Religious affiliation was 40% Protestant, 25% Catholic, 7% Jewish, 19% some other denomination, and 10% unspecified. All regions of the country were represented, with the Southeast and Midwest constituting more than half of the sample. More than three fourths of respondents worked in medical examiner or coroner's offices, and 89% held an MD degree. Eighty-five percent had completed residency training, averaging 15 years of experience after training. Most were board certified: more than three fourths in anatomic pathology, almost three fourths in forensic pathology, and half in clinical pathology. For ease of discussion, we refer to the study participants as “respondents” or “pathologists.”
Only a few pathologists had received any formal ethics training (24%), served on an institutional review board (10%), or served on a patient care ethics committee (6%) (Table 2). Nearly two thirds of respondents (63%) reported encountering ethical dilemmas related to the use of human autopsy tissues during their professional practice.
Educational Requests for Autopsy Procedures
Respondents were asked to rate the acceptability of using autopsy tissues for 4 educational procedures: demonstration of visceral anatomy; demonstration, practice, or teaching of intubation; practice of needle thoracotomy; and demonstration or practice of brachial plexus dissection. Respondents were asked to rate the acceptability of allowing different individuals to perform or be trained through these procedures (ie, faculty member, clinical team member, resident, medical student, emergency medical technician, and art student). A repeated-measures MANOVA on the 12 educational purposes revealed a main effect for item (F11,60 = 27.72, P < .0001), indicating that significant differences in perceptions of ethical acceptability exist for the varying educational purposes (Table 3). The demonstrations of visceral anatomy were, in general, perceived as most acceptable (P < .0001), particularly for medical trainees as opposed to art students. Demonstration, practice, or teaching of intubation techniques on cadavers was seen as only moderately acceptable for residents, emergency medical technicians, and medical students. Although also viewed moderately, practicing needle thoracotomy or brachial plexus dissection was less acceptable to respondents overall (P < .05).
Respondents disagreed considerably regarding acceptability of tissue use with respect to the educational procedures. To better understand the distribution of the data, the 7-point scale was collapsed into 3 categories: “highly acceptable” (ie, 1 or 2), “moderately acceptable” (ie, 3, 4, or 5), and “highly unacceptable” (ie, 6 or 7). Thus, the percentages of respondents in each of the 3 categories for each item are shown in each table, in addition to the means on which the MANOVAs were performed. In particular, except for general agreement regarding the acceptability of demonstrating visceral anatomy to residents and medical students, the procedures widely differed in the acceptability expressed by various respondents (Table 3). For example, half saw practicing intubation techniques as highly acceptable, regardless of who performed them or who would be trained. Almost another third regarded the same educational procedures as highly unacceptable, whereas one fifth felt they were only somewhat acceptable. Similar disagreement among respondents was shown for needle thoracotomy, brachial plexus dissection, and visceral anatomy with art students.
Requests to Observe an Autopsy
Respondents were asked to rate the acceptability of fulfilling requests to observe an autopsy made by individuals in 12 different roles (eg, physician whose patient died, nurse whose patient died, medical student with no relationship to patient, art student, friend of pathologist) (Table 4). A repeated-measures MANOVA showed a main effect for role (F11,60 = 24.35, P < .0001). Ratings for all but 2 of these items produced skewed unimodal distributions, generally reflecting agreement on the acceptability of autopsy observation. Requests from medical personnel and trainees whose patient had died showed very high acceptability with very little variation, as did requests from medical personnel who had no relationship to the decedent. These requests from medical personnel were all rated as highly acceptable by about 90% of respondents. Two other roles (non–medical community member of a hospital ethics committee and attorney for educational purposes not related to a specific case) showed moderate mean levels of perceived acceptability but with considerable disagreement among respondents. The last 3 roles (art student, friend of pathologist, and friend of pathology technician) showed average weak unacceptability but with much variability across the scale.
Research Requests for Autopsy Tissues
Respondents rated the acceptability of using 6 different types of human autopsy tissues (ovary or testis, serum, knee, liver, brain, and paraffin-embedded tissue) for research purposes (Table 5). With the exception of ovary or testis, the tissue types were each presented in 3 amounts that varied from small (eg, 1 g) to large (eg, 100 tubes of serum, 10 entire brains) with respect to size and/or number. Thus, both tissue type and amount of tissue were systematically varied, and survey participants rated all types of tissue and all amounts of tissue. These ratings were subjected to a tissue type (serum versus knee versus liver versus brain versus paraffin embedded) by tissue amount (small versus moderate versus large) repeated-measures MANOVA, and all effects were significant (tissue type: F4,67 = 13.27, P < .0001; tissue amount: F2,69 = 23.79, P < .0001; and type by amount: F8,63 = 5.24, P < .0001). A one-way repeated-measures MANOVA on the means was also highly significant (P < .0001) and revealed that means differing by 0.36 or more are reliably different by the Fisher least significant difference test.37
Responses to the use of serum, knee tissue, and ovary or testis tissues showed essentially identical patterns. That is, half the respondents found requests for these tissue types in all amounts to be highly acceptable, but about a third found the requests to be completely unacceptable. In contrast, questions about moderate or large amounts of liver and brain tissues show a different kind of response: about three fifths of respondents rated liver and brain tissue use for research as highly unacceptable, about a fifth as moderately acceptable, and another fifth as highly acceptable. Finally, respondents were divided but showed even less acceptance of the use of paraffin-embedded tissue for all specimen quantities.
Thus, these analyses show that perceived ethical acceptability varied by tissue type and quantity, with smaller quantities receiving greater acceptance. Finally, perceptions of ethical acceptability were variable within items, indicating considerable disagreement among medical examiners on the research use of cadaveric tissues.
Evaluating Research and Educational Requests for Autopsy Tissues
Respondents were asked to indicate the importance of 9 features they might consider when evaluating a request for autopsy tissues for research purposes and when evaluating a request for educational purposes (Table 6). The means were subjected to a sex (2) by features (9) by purpose (research versus educational) repeated-measures MANOVA. A main effect for sex (F1,170 = 7.92, P < .001) showed that women perceived these features as more important. The feature main effect was also significant (F8,78 = 21.19, P < .0001), indicating that these features affected perceptions. The purpose main effect was not significant (P > .68), indicating that responses did not differ substantively based on whether the aim was research as opposed to education. The feature by purpose interaction effect was significant (P < .0001) however, suggesting that the aim influenced the extent to which the respondent placed importance on the features offered. Simple effects analyses show that the interaction was significant primarily due to the ratings on one item—“approval of the research or educational project by the institutional review board or dean of the training institute,” which was seen as significantly more important for research than educational uses (P < .001).
The 9 features generally clustered into 2 levels of rated importance. The first group includes the following: (1) integrity of the requesting individual, (2) integrity of the requesting individual's institution, (3) the scientific or educational reputation of the requesting individual, (4) the scientific or educational reputation of the institution of the individual, and (5) the importance of the project's scientific or educational merit. These features all have means between 2.22 and 2.60, reflecting the belief that all have importance when considering the use of autopsy tissues. The second group of items includes (1) the respondent's personal interest in the project, (2) inclusion of the respondent as a coinvestigator, and (3) the institution's commitment to research or education. These 3 features were viewed as less important overall, with means between 4.24 and 5.10. Nevertheless, 5% to 25% of respondents rated these latter issues as “highly important.”
Values Bearing on Requests for Autopsy Tissues
Conference attendees were also asked to rate the importance of 7 other elements or values potentially relevant in considering a research- or education-related request for autopsy tissues (1 indicates “highly important” and 7 indicates “highly unimportant”). A repeated-measures MANOVA showed a significant main effect for values (F6,83 = 7.74, P < .0001). Overall, values were seen as highly important (means ranging between 1.74 and 2.84), but 2 were seen as significantly more important (P < .0001) than the others: (1) knowledge that the family of the person who died consents to the use for that purpose (mean, 1.74) and (2) preserving confidentiality (mean, 1.80). Other values also seen as important include the following: (1) knowledge that the person who died had consented to such use (mean, 2.25), (2) advancing education of health professionals (mean, 2.40), (3) benefiting humankind (mean, 2.47), (4) advancing science (mean, 2.53), and (5) the likelihood of benefit to patients (mean, 2.84).
Influence of Characteristics of the Person Who Died
Individuals were also asked to indicate on a 7-point scale whether they would be more (1) or less (7) likely to fulfill a request for tissues for research or educational purposes if the person who died had 1 of 15 characteristics (Table 7). A sex (2) by characteristic (15) repeated-measures MANOVA showed a main effect for sex (F1,176 = 3.22, P < .001), which indicated that women reported on average being more likely to fulfill tissue requests because the decedent had any of the listed characteristics. A main effect for characteristic (F14,75 = 4.89, P < .0001) showed that some characteristics would be more influential than others. For all but 2 characteristics, about 75% responded with a rating of 4, indicating that the specified characteristic would have no impact on their decision. The means for these 13 items ranged from 3.57 to 4.00 and did not reliably differ except for the 2 most extreme means in this group (P < .05). Thus, respondents, on average, felt that they would not be much influenced by the decedent demographic or other personal characteristics. Two characteristics stood out among the others: if the decedent had died unexpectedly from a mysterious illness or died from a known but very rare illness, respondents reported they would be more likely to fulfill the request (P < .05).
Effects on Ratings of the Characteristics of Responders
The 6 groups of ratings discussed in the previous sections were also analyzed to detect possible effects of the demographic or professional variables of respondents. Of those factors, including age, ethnicity, sex, religion, region of the country, work setting, years after training, clinical training completed, and board certification (Table 1), only sex predicted any significant effects for data in Tables 6 and 7. Women accorded greater importance to the features of the request and the characteristics of the decedent than did their male counterparts.
Respondents were asked to respond to each of 5 vignettes that depicted cases involving requests for autopsy tissues (Table 8). Table 9 summarizes the responses to 3 questions for each vignette: (1) “How ethically acceptable do you believe fulfilling this request to be?” (2) “How acceptable is it to proceed without family consent?” and (3) “Would you supply the … tissue?” A vignette (5) by question (2) by sex (2) repeated-measures MANOVA on the ratings produced 3 significant effects: a main effect for vignette (F4,62 = 11.42, P < .0001), a main effect for question (F1,65 = 23.39, P < .0001), and a marginal main effect for the vignette by question interaction (F4,62 = 2.47, P < .06).
Simple effects analysis of the first question shows that the means across vignettes reflect 3 significantly different levels of acceptability of fulfilling requests: (1) vignettes 1, 3, and 5 are on average seen as acceptable requests; (2) vignette 2 is seen as less acceptable than vignettes 1 and 3 (P < .05); and (3) vignette 4 is seen as significantly even less acceptable than the other 4 vignettes (P < .05). The vignettes show 2 similar groupings for whether it would be acceptable to proceed without family consent: (1) vignettes 1, 3, and 5 show moderate levels of acceptability, and (2) vignettes 2 and 4 show significantly less acceptability (P < .05), moving toward the unacceptable end of the dimension. Thus, medical examiners perceive different levels of acceptability, depending on case characteristics, both with and without family consent. It is clear, however, that many respondents rate acceptability much lower if the family does not provide consent (means, 2.94 versus 4.12, respectively; P < .0001; d = 0.67). An important additional finding within these data, as was found for other measures in this study, is that the responses are variable. That is, many respondents rate acceptability very high, whereas many others rate the same request as highly unacceptable.
For the third question (ie, “Would you supply the … tissue?”), many individuals responded with either an unqualified “yes” or “no” as noted in Table 9. Others provided a qualified response by indicating the circumstances under which they would supply the tissue. Those who gave qualified responses most often stipulated that they would provide tissue only under the condition that the decedent or family had consented.
Many study participants offered detailed narrative comments. Ethical dilemmas faced by participants included the following: “persistent requests” for commercial device testing; state-mandated eye harvesting even in the absence of family consent; duty conflicts in pursuing one's personal research interests; “issues of competition” among agencies that harvest or facilitate transplantation of tissues; “scandal,” false accusations, and legal liability related to supplying tissues; discomfort associated with refusing colleagues who ask for tissues; requests for tissues to perform tests “that could have been done during life”; being offered personal financial incentives for supplying tissues; and stresses associated with tissue requests in “high-profile” legal cases.
This study documents the diverse perspectives of pathologists from throughout the United States regarding the ethical acceptability of using autopsy tissues for education and research. Respondents commonly differed in their views of autopsy tissue use to demonstrate or practice techniques such as intubation, needle thoracotomy, and dissection. Respondents were divided on the topic of whether it is ethical to permit nonmedical personnel to observe autopsies for educational reasons. Requests to supply varying kinds and quantities of tissues for scientific experimentation were also often met with disagreement. Finally, decisions regarding provision of autopsy tissues in 5 ethically complex hypothetical scenarios were relatively variable. Although the respondents were relatively homogenous in terms of ethnicity and medical training, they varied in their formal training, involvement in patient care ethics committees, and service on institutional review boards. Respondents also derived from different regions of the country, religious backgrounds, varying work settings and kinds of expertise, and both sexes. Interestingly, despite this variation, sex was the only attribute that predicted differences: women in the study placed greater importance on values and on individual characteristics of decedents in making tissue-related decisions. This finding resembles those of other studies,38–40 suggesting that women clinicians are more attentive to contextual considerations in evaluating ethics questions, but the absence of an effect due to religion, for example, is unexpected.
This unanticipated array of perspectives, experiences, and backgrounds was nevertheless coupled with several areas of consensus. The pertinence of ethics issues for pathologists and medical examiners in particular was reflected in the narrative comments we received and in the finding that most respondents (63%) had encountered ethical dilemmas regarding autopsy tissue use. We found agreement about fundamental values that influence decisions related to tissue or autopsy requests, such as knowledge of the integrity of the requesting individual and respect for the merit of the project. Greater uniformity was also evident in the acceptance of autopsy observation by medical personnel and trainees and the importance accorded to family or next-of-kin consent in dealing with complex ethics issues. Some of these findings (eg, the importance of institutional review board review related to research requests) lend support to recommendations reached by the Ad Hoc Committee on Stored Tissue offered by the College of American Pathologists.27 The implications of the patterns of diversity and consensus in pathologists' ethics perceptions related to tissue use in education and research are briefly discussed in this section, as are the limitations of this study.
The Role of the Autopsy in Education
Respondents appear to consider 3 issues with respect to the role of the autopsy in education: (1) the relevance of the autopsy to the education of the individual, (2) the level of invasiveness that the educational activity requires, and (3) consent of family for the use of tissues and organs for educational purposes. The observation of an autopsy by nearly any clinician or clinical trainee was felt to be acceptable, whereas respondents were much less tolerant of such requests by others (eg, community representative on the hospital ethics committee, an attorney, an art student, or friends of the pathologist or a pathology technician). Respondents viewed the use of organs retrieved from autopsies to teach medical students about cardiac disease as important and acceptable so long as family consent had been obtained. The “need to know” within the context of health care and clinical education thus appears to be an important factor in sanctioning access to autopsies.
Demonstrating or performing more invasive procedures, such as needle thoracotomy and brachial plexus dissection, that are not routine autopsy procedures was generally viewed as less acceptable than activities that are minimally disruptive to the decedent's anatomy, such as the demonstration of visceral anatomy and practice of intubation. Even so, half of the respondents were uncomfortable with the idea of using the newly dead to teach intubation, echoing recent data on mixed perceptions of the use of the nearly dead to teach such procedures.25 This controversy has surfaced in the clinical literature41–43 and, along with other findings from this study, has significant training implications.
Use of Autopsy Tissues for Research
The medical examiners in our study were divided concerning the use of autopsy tissues for research purposes. Providing small quantities of tissue, such as 1 g of liver or brain tissue or one tube of serum for experimentation, appears to be more acceptable than providing larger amounts or volumes of fresh tissues. It is important to note, however, that roughly half of the respondents expressed some discomfort with furnishing even 1 g of tissue. Requests for paraffin blocks were commonly seen as less acceptable than requests for fresh tissue. This is curious because paraffin blocks contain relatively small amounts of tissue and are routinely saved in most autopsies regardless of whether they are to be used for research. Indeed, requests for research access to paraffin blocks may occur years to decades after an autopsy was performed, making the search for next-of-kin very difficult. When placed in the context of a specific request for paraffin blocks for anonymous genetic testing related to a potential coagulopathy (vignette 5, Table 8), most respondents indicated that fulfilling the request was highly acceptable on ethical grounds. These diverse findings corroborate the differences in opinion in the biomedical literature surrounding tissue ‘ownership.’14,26,33,44
Influence of Specific Factors on Acceptability of Tissue Use
We examined the influence of several other factors on the acceptability of cadaveric tissue use for research and education. Three elements—integrity, scientific or educational merit, and formal institutional approval—outweigh many other factors, such as personal interest, collaborative role in a project, or institutional goals, in assessing ethical acceptability of providing access to autopsies or autopsy tissues (Table 6). Interestingly, the attributes of the decedent (eg, sex, ethnicity, age, religion) did not predict perceptions of ethical acceptability of autopsy tissue use. Only with a clear clinical imperative regarding an unusual cause of death, such as a mysterious or very rare illness, did respondents acknowledge that they would be more likely to permit tissue use (Table 7). Although this may reflect a desire not to act in a discriminatory manner, there is evidence that beliefs held by some religious or ethnic groups may justify differential attitudes regarding access to cadaveric tissues.3,42,45–48 Moreover, some decedents may belong to vulnerable populations requiring special protections and/or understudied groups requiring more intensive scientific inquiry than in the past.11,18,49 Beyond attributes of the decedent, medical examiner characteristics (eg, ethnicity, age, religion), with the exception of sex and professional ethics experiences, also did not consistently determine responses. As in other clinical ethics studies,28,33,35 women physicians placed greater importance on certain ethical values and contextual considerations than their male counterparts (Tables 6 and 7).
Key Role of Family Consent in Tissue Use for Research and Education
Family consent is an important factor for medical examiners in assessing the ethical acceptability of tissue use in research and education (Table 9). Use of tissues in the absence of family consent was viewed more negatively by respondents in all the vignettes, including requests motivated by factors ranging from clinical necessity (vignette 1) to educational value (vignette 3) to more esoteric scientific inquiry (vignette 4). Regarding the actual decision of whether to supply the requested tissue, a substantial minority of respondents indicated uncertainty in each situation, and only in one case (vignette 4) was there a majority opinion against providing the tissue. These data suggest that finding an appropriate way to include the family in the process of research and educational decision making may be preferable in the views of many forensic pathologists.
Obtaining family consent for use of medicolegal autopsy tissues can be daunting and poses several clinically and ethically complex considerations. For hospital autopsies, the appropriate next-of-kin is usually identifiable, and the reasons for tissue requests are often clear and acceptable to the decedent's family. In contrast, finding family members is commonly more difficult in medical examiner cases where death may be sudden, unexpected, or unattended and may have occurred far from the decedent's home or family. Once the next-of-kin has been identified, it is unclear who should obtain the consent for educational or research uses of tissues in medical examiner cases. Should it be someone from the medical examiner's office, the researcher, or an independent third party? Medical examiners and their staff are often comfortable interacting with and are sensitive to the needs of the bereaved. However, when they are not part of the research project, they may not have time and are unlikely to have an adequate understanding of the project to solicit a truly informed consent. Because medical examiners perform autopsies that are mandated and occur without next-of-kin consent, the research and education consent may introduce confusion about what bereaved families can and cannot control and may create uncertainty about professional roles and conflicts of interest. Conversely, scientists or teachers who approach families about consent for nondiagnostic uses of autopsy tissues may be able to clearly describe their project but may not be comfortable with family members' grief and may not be aware of statutory issues. Insensitivity on the part of these professionals could be interpreted as being representative of the medical examiner's office. If several different individuals approached a family for consent, it might be viewed as highly stressful or as harassment. Therefore, an independent third party obtaining consent may be the best solution.
Although challenging, it is important to recognize that in many situations obtaining family consent may not, in fact, be a barrier to scientific and educational use of tissues. Studies performed in the United States and Sweden suggest that, when specifically asked, most individuals would permit autopsy, anatomic dissection, and organ donation on themselves and family members.10,45,46 The existing organ procurement system set up for transplantation may be best able to obtain consent for the educational or research use of tissues in medicolegal cases. These individuals are trained to handle issues of bereavement, informed consent, clinical education, and scientific protocol. They are distinct from, but work closely with, medical examiner offices. Although conflicts may occur from time to time (eg, desire for the same tissue type), they often approach families for organs and may easily be able to ‘add on’ other requests. Indeed, educators and researchers may benefit from the generosity families often demonstrate when asked about organs for transplantation. Thus, some commentators believe that obtaining next-of-kin consent for these procedures is thus feasible, reassuring, and worthwhile.43
Limitations of Study
The interpretation and generalizability of our findings are limited by several factors. First, the study's sample size was small, the response rate moderate, and the sampling strategy purposive. Second, the study used a self-report method. Respondents may inaccurately record their views and experiences, especially because of the sensitive nature of the ethics topics addressed. However, since the questionnaire was administered in a manner that preserved anonymity of respondents, we hope to have minimized this problem. The study was based on a new survey instrument that, although carefully crafted and pilot tested, may have not fully captured the concerns and experiences of respondents. Finally, the study reflects medical examiners' beliefs about the appropriate use of cadaveric tissues, but it does not examine their actual decisions and behaviors. For these reasons, we offer this study as an initial effort in a previously neglected but important area of inquiry.
We found that the ethical acceptability of fulfilling autopsy tissue requests for educational and research purposes differed according to nature and merit of the project and the kind and quantity of tissues asked for. Many factors bearing on perceptions of ethical acceptance of tissue use and related decisions were congruent with our predictions, including family consent, clinical imperatives, qualities of the individuals making requests, and institutional safeguards and missions. Unexpectedly, many elements did not influence respondents' judgments in a consistent manner, such as certain decedent characteristics and specific attributes of medical examiners. For these reasons, mixed perceptions of the ethics of cadaver use in research in education pose numerous unresolved questions.
In the absence of uniform ethics, regulatory, and legal guidelines, the medical community has an opportunity to influence the direction and shape of principles and policies regarding the use of autopsy tissues for education and research. As a preliminary to developing and advocating specific approaches in this area, however, it is important to understand the perspectives of forensic pathologists and medical examiners. Ultimately, it is medical examiners in this country who will be responsible for continuing to provide or withhold these tissues in the future.
On the basis of this empirical work, we conclude that significant diversity exists among these professionals' beliefs regarding the use of autopsy tissues for education and research. In our opinion, this lack of consensus may make it difficult for practicing medical examiners to influence future ethical, regulatory, and legal guidelines. Although initially this may seem disappointing, this appears to be an accurate reflection of the intricacy and complex history of this realm of practice. In addition, although some agreement is desirable, we see that this rich context may give rise to genuine inquiry and dialogue and to the development of enduring policy solutions regarding human tissue use. This is especially important in light of the fundamental agreement we found among respondents related to core values underlying biomedical education and research. Another major implication of this work is that families are accorded great importance in decisions related to autopsy tissue use for education and research. This finding suggests that the respectful and sensitive inclusion of families in such decisions may ultimately become a key determinant of ethically acceptable practice nationally. Finally, we conclude that the ethics of using medical examiner autopsy tissues for education and research is an area that requires further discussion and investigation. Resolution of these ethical issues is imperative if physicians and medical scientists of all disciplines are to use these tissues effectively while also providing respect and reasonable protections for the autonomy and privacy of the deceased and the deceased's family.
Dr Roberts is supported in part by grant 1K02AI01738–01 from the National Institute of Mental Health.