Andrew C. G. Breeze’s salary was funded in part by Cambridge Fetal Care, Cambridge, United Kingdom.
Perinatal Postmortems: What Is Important to Parents and How Do They Decide?
Article first published online: 9 JAN 2012
© 2012, Copyright the Authors. Journal compilation © 2012, Wiley Periodicals, Inc.
Volume 39, Issue 1, pages 57–64, March 2012
How to Cite
Breeze, A. C. G., Statham, H., Hackett, G. A., Jessop, F. A. and Lees, C. C. (2012), Perinatal Postmortems: What Is Important to Parents and How Do They Decide?. Birth, 39: 57–64. doi: 10.1111/j.1523-536X.2011.00513.x
- Issue published online: 28 FEB 2012
- Article first published online: 9 JAN 2012
- Accepted June 2, 2011
- perinatal bereavement;
- postmortem imaging;
- pregnancy loss
Abstract: Background: Falling consent rates for postmortems, regardless of age of death, have been widely reported in recent years. The aim of this study was to explore parental attitudes to, and decision-making about, a perinatal postmortem after termination for fetal abnormality, late miscarriage, or stillbirth.
Methods: A prospective self-completion questionnaire was given to 35 women and their partners. The participants had experienced second or third trimester pregnancy loss in a single fetal medicine and delivery unit in the United Kingdom and were making decisions about having a postmortem. They were asked to complete a questionnaire about their attitudes to, and expectations of, a perinatal postmortem.
Results: Thirty-one questionnaires were received from parents of 17 babies (49% of those asked; 16 from mothers, 15 from fathers). Parents of nine babies (53%) said they would agree to a full postmortem, of three babies to a limited postmortem, and of four babies to an external examination only; one couple were undecided. The most important issues for the parents in this study that related to their decisions about a postmortem centered on the need for information, both for future planning and about what had happened. Moderately important issues related to altruism, which is, improving medical knowledge and helping other parents experiencing similar bereavement. Among the lowest scoring issues were potential barriers, such as concerns about cultural or religious acceptability of a postmortem, funeral delays, and what would happen to the baby’s body.
Conclusions: Bereaved parents who participated in this study, where postmortem consent rates were relatively high, thought that their need for knowledge eclipsed assumed barriers when deciding whether or not to have a postmortem for their baby. (BIRTH 39:1 March 2012)
Postmortem findings may explain the cause of death or pregnancy loss following late miscarriage, stillbirth, or termination for fetal abnormality (1), and provide information that will improve parental and medical decision-making in the future (2,3).
Falling consent rates for postmortems, regardless of age of death, have been widely reported in recent years (4–6). In the United Kingdom, rates for perinatal autopsies have decreased since publicized organ retention events of the 1990s at Bristol and Alder Hey hospitals, and similarly in Australia (7), although they had been declining well before this (8). This decrease may be a consequence of changes in clinicians’ perception of the value of the investigation (9,10), centralization of pathology services (11), or improvements in diagnostic imaging (12–14). In 1988, a joint working party of the Royal College of Obstetricians and Gynaecologists and the Royal College of Pathologists had considered a perinatal postmortem rate of less than 75 percent as “unacceptable” (15), but subsequently agreed that a target rate was no longer appropriate (16). A recent series showed that the postmortem rate in cases of termination for fetal abnormality fell from 84 to 67 percent over the period 1991 to 2000 in the United Kingdom (2) and from 67 to 50 percent in an Australian study (7). In England and Wales, postmortem uptake fell from 58 percent of all deaths reported to the Confidential Enquiry into Stillbirth and Deaths in Infancy (CESDI) in 1993 to 45 percent for stillbirths and 21.1 percent for neonatal deaths in 2007 (17).
Although in approximately 50 percent of cases, parents decline consent for a postmortem (18), clinicians may sometimes not even discuss the investigation with parents (19). A significant proportion of neonatologists in the United Kingdom do not routinely offer a postmortem to bereaved parents because of recent cases of organ retention, the new consent process (a more explicit consent form was introduced nationwide in 2003), concern that it might upset parents further, and being uncomfortable with the postmortem examination process themselves (10). Furthermore, consent may be inappropriately delegated to inexperienced or junior clinicians with little insight into the importance of postmortem (20).
Although obstetric and sociodemographic predictors of parental decisions about postmortem have been explored (21), the direct exploration of parental decision-making has been addressed in only a few studies and always retrospectively. Reticence about probing during this sensitive time when parents are making their decision is not surprising given their distress in the aftermath of perinatal loss (19,22) and the attendant difficulties with obtaining ethical approval for prospective research (23). The most extensive retrospective study explored reasons for having and not having a postmortem in a group of parents attending bereavement counseling (24). The findings were not easily generalizable because those who had agreed to a postmortem were asked different questions from those for whom the examination had not been carried out.
The aim of this study was to explore whether research about decision-making with respect to a postmortem can be undertaken prospectively, and to improve understanding of parents’ attitudes to a postmortem and of the factors they report as important in making the decision as to whether or not to have the investigation. A secondary objective was to examine parental interest in less invasive forms of postmortem examination, such as postmortem magnetic resonance imaging (25–31).
Recruitment of Study Participants
Between April 2005 and September 2006, parents who were making decisions about postmortems were invited to take part in a self-completion questionnaire study. Parents were recruited from the fetal medicine unit and delivery unit of a teaching hospital in the United Kingdom with approximately 5,000 deliveries per year (2006). The study received a positive ethical opinion from the local National Health Service Research Ethics committee (Cambridge 3 Research Ethics Committee, Ref. 04/Q0108/185).
After late miscarriage, stillbirth, or the decision for pregnancy termination for fetal abnormality, women and their partners were asked by the obstetric and midwifery staff if they wished to discuss a perinatal postmortem with the fetal medicine research fellow. Those agreeing to such a discussion (irrespective of whether or not they gave consent for any form of postmortem) met with him. After the clinical discussion he asked if they were willing to be given information about a research study. Those who agreed were given an information leaflet about the study, a consent form, and a self-completion questionnaire. Consent forms and completed questionnaires were returned to the researcher (H. Statham), who was not involved in the provision of clinical care.
Questionnaires were designed specifically for this study, addressing issues previously reported as important in postmortem decision-making (24,32,33) and those which the investigators considered might be important in the light of recent organ retention events and their own clinical experience. Some questionnaire items varied, depending on the nature of the pregnancy loss. Participants were asked questions within the following categories: intentions about a full or partial postmortem; issues that were important to them as they made their decision; and information they believed a postmortem could provide. Participants were also asked if participating in the research had added in any way to their distress, and about their attitudes to the questionnaire, including the impact on their postmortem decision-making. Findings in relation to these aspects are reported elsewhere (23).
All questionnaire data were coded and analyzed using the Statistical Package for the Social Sciences (34). Descriptive statistics that were used included means and medians, and age comparisons carried out using the t test. The numbers of cases in analyses and tables vary according to how many respondents answered questions: percentages are provided based on the number of valid responses. Given the small study population and distributions of dependent variables, nonparametric statistical tests were used (Spearman correlation coefficient, Mann–Whitney U test for comparisons of continuous variables).
Response Rate for Participation in Postmortem Research
Of 35 women and/or couples approached, one or both parents of 17 babies (49%) returned 31 questionnaires (16 from mothers and 15 from fathers). Fifteen questionnaires were from parents of 8 babies for whom the pregnancies ended in termination and 16 from parents of 9 babies who died in utero. The median age of study women was 32.5 years (range: 21–38). For comparison, the median age of all women delivering at this unit in 2006 was 31 years (range: 15–47). In general, male respondents were older than women (t = −3.2, 29 df, p = 0.003). Among mothers and fathers separately there were no differences in ages related to type of loss.
The educational background of respondents was high: 11/16 women and 11/15 men had at least an undergraduate degree or equivalent. The study population was predominantly white European, with one father and one mother’s father being Asian. This ethnicity largely reflects that of women delivering at this unit in 2005–2006, of whom 89 percent described themselves as white (unit data, 2005–2006).
Intentions about Postmortem
Parents of nine babies (53%) said that they would agree to a full postmortem, and parents of three babies to a limited postmortem (where particular part(s) of the baby’s body is/are examined internally); parents of four babies said they would have only an external examination. Two parents wished to find out more before deciding. When both parents participated, there was complete agreement concerning intentions about postmortem.
All six respondents (two couples, one mother, and one father) who were not having either full or limited postmortem agreed to chromosomal studies, external examination, and placental histology. Five of six parents would agree to x-rays or other imaging, and three agreed to organ needle biopsies being performed (35).
All who agreed to full or limited postmortems stated that they would choose a full postmortem over a less invasive procedure (if offered) that, they were told in the questionnaire, might provide less information. The parents who were uncertain about which to choose stated that they would opt for the less invasive postmortem, if available.
What Is Important in Deciding Whether or Not to Have a Postmortem?
Participants were asked to rate the importance to their decision on whether to have a postmortem for each of 19 items. Ratings were from 0 for unimportant issues to 5 for issues deemed very important. The items fell into two broad categories: reasons for having a postmortem and reasons for not having one. Figure 1 illustrates the items in descending order of importance for the whole study population (i.e., men and women, after termination and stillbirth).
The items rated as most important by the study participants overall centered around the need for information. Information was important for future planning: to plan subsequent pregnancies and to establish the recurrence risk. It was also important for the present, that is, to obtain an explanation of what had happened, to find out about the abnormalities or the cause of death, and to confirm whether postnatal findings corresponded with prenatal investigations.
Moderately important items related to altruism, such as improving medical knowledge and helping prevent similar things from happening to others, despite prior counseling that this item was not a primary purpose of the postmortem. Among the lowest scoring issues for study participants were the cultural or religious acceptability of postmortems, and the potential barriers to postmortem, including funeral delays or concerns about what would happen to the baby’s body.
In seeking to understand the decision-making of individual women and to inform further research, some preliminary exploration of these data were made using nonparametric tests. Small but significant correlations were shown between age of respondent and some items identified as potential barriers to having a postmortem. In contrast, differences between those who were and were not having the examination did not (in this study) relate to “barriers” but, rather, to parents’ need for knowledge. Younger respondents rated as more important four barriers: a postmortem may delay the funeral arrangements (Spearman R = −0.424, p = 0.025); my partner feels differently than I do (R = −0.513, p = 0.05); worry about what will happen to the baby’s tissues and organs after the postmortem (R = −0.400, p = 0.048); and not wanting to put my baby through anything else (R = −0.405, p = 0.0365). The “information” items for which scores differed between those having and not having a postmortem were “I want to try and find out the cause of my baby’s abnormalities (TFA)/why my baby died (SB/M)” and “I want to check if abnormalities are there that couldn’t be seen on scan,” where median scores were higher for those having the investigation; and “I already know what is wrong with my baby (TFA)/why my baby died (SB/M),” where median scores were lower. Median scores, ranges, and Mann–Whitney U statistics for the distributions of scores are shown in Table 1 for these items.
|We Know that there Are Particular Aspects about a Postmortem that Are Important for Parents as They Make Their Decision about Whether or Not to Have One. We Would Like to Know how You Feel about Things||Having Postmortem Median (No.) (Range)||Not Having Postmortem Median (No.) (Range)||Mann–Whitney U||p|
|I want to try and find out … the cause of my baby’s abnormalities (TFA)/why my baby died (SB/M)||5 (23) (1–5)||3 (6) (0–5)||28.0||0.026|
|I want to check if abnormalities are there that couldn’t be seen on scan||5 (23) (2–5)||3 (5) (0–4)||9.5||0.002|
|I already know … what is wrong with my baby (TFA)/why my baby died (SB/M)||0 (13) (0–5)||5 (5) (4–5)||3.5||0.002|
What Do You Think a Postmortem Will Show?
The second part of the questionnaire examined parents’ expectations of what a postmortem might show. The items were selected taking into account topics discussed in the clinical consultation about a postmortem, relating mostly to the fetus or baby and the circumstances of the death, but others related to a wider context including the investigation’s effect on others and on care in future pregnancies. Table 2 shows responses to this question: “How likely do you think it is that a postmortem will …”, with a score of “0” for not at all likely, and “5” for very likely, with detail of the range of responses for those having and not having a postmortem.
|How Likely Do You Think it is that a Postmortem Will ... (Items Were Scored from 0—Not at All Likely to 5—Very Likely)||Range||Mean (Median)||Having Postmortem Median (No.)||Not Having Postmortem Median (No.)|
|Confirm the findings of your scans during pregnancy (TFA only)||1–5||4.5 (5)||5 (10)||5 (5)|
|Confirm that your baby was normal (SB/M only)||2–5||3.8 (3.5)||3 (13)||5 (3)*|
|Help you plan for future pregnancies||0–5||3.7 (4)||4 (23)||4 (8)|
|Tell you the chance of something like this happening again||0–5||3.6 (3)||4 (23)||3 (8)|
|Improve medical knowledge and research||0–5||3.6 (4)||4 (23)||4 (6)|
|Find abnormalities that were not seen on scan||1–5||3.6 (4)||4 (23)||3 (6)|
|Tell you the cause of your baby’s death (SB/M)/Tell you the cause of your baby’s abnormalities (TFA)||0–5||3.5 (4)||3 (23)||5 (8)|
|Reassure you that you’ve made the right decision about the pregnancy (TFA only)||0–5||3.3 (5)||5 (10)||0 (5)*|
|Reassure you that you were not to blame||0–5||3.2 (3)||3 (23)||4.5 (8)|
|Help prevent others losing a baby like this||0–5||2.9 (3)||3 (23)||1.5 (6)|
|Help with grieving process||0–5||2.9 (3)||3 (23)||1 (6)*|
Expectations were high, with mean and median scores above the midpoint for all items. The highest scoring item was that the postmortem would “confirm findings of your scans during pregnancy,” with 13 of 15 who had terminated pregnancies scoring 4 or 5. Although the second highest scoring item overall, only 8/16 of those whose baby had died or been miscarried, scored 4 or 5 for the corresponding item “confirm that your baby was normal.” Those participants who decided not to have a postmortem tended to score this item more highly, that is, believing it to be the more likely outcome. Median scores in each group were 3 (having n = 13) and 5 (not having one, n = 3); the distributions in the two groups differed significantly (Mann–Whitney U = 4.5, p = 0.039, two-tailed.)
Expectations of the postmortem providing new knowledge were also high, particularly with respect to influencing care in subsequent pregnancies, providing risks of recurrence, and confirmation of both what was known before and reaffirmation that the “right decision was made.” Relatively high expectations were also seen that some other good would come of the postmortem, either for other people, in preventing other pregnancy losses, or for improving medical science. The lowest expectation overall was that the postmortem would help with grieving, but even that scored a mean of 2.9 (median: 3) with 65 percent of respondents scoring 3 or more. Distributions of scores differed between those who decided against a postmortem and those who went on to have the investigation. All six respondents who declined a postmortem scored 3 or less, whereas 18 (of 23, 78%) of those who went on to have the investigation scored 3 or more (Mann–Whitney U = 17.5, p = 0.003, two-tailed).
Some comments made at various points throughout the questionnaire related to the clinician-researcher who had discussed postmortem options, consent, and research participation with these parents. It seemed particularly important that the opportunity of time was given for discussion and questions, and parents welcomed discussion specific to their situation, such as being told that a postmortem might not be useful in their circumstances:
After a discussion with a doctor, many reasons I thought a postmortem may be useful to our circumstances proved wrong. This discussion did not change my attitude to postmortems as such, but did help to inform my decision on how to proceed. (Father—termination)
This study demonstrates that it is possible to investigate factors influencing parental decisions in relation to postmortem consent, expectations of postmortem outcomes, and attitudes to less invasive alternatives, and to undertake this investigation prospectively, as parents are making their decisions. We described previously the positive attitudes of these bereaved parents to participation in this research (23); a finding from that study was that parents feel it is an important area of research and that the research, which involved thinking in great detail about reasons to have or not to have a postmortem, helped them to make their decision about how to proceed, to feel better about the choice they made, or both. Participation made little difference to the decision made: of 22 participants who answered a specific question, “Has filling in this questionnaire made you think differently about having or not having a postmortem?” one participant reported that they were less likely and another more likely to have a postmortem, whereas 16 reported that they felt better about their decision.
The most important consideration underlying the decisions made by the particular group of parents in this study was the extent to which they needed knowledge. Knowledge could give information that was relevant immediately by explaining the cause of pregnancy loss or fetal abnormalities and for future reproductive decision-making and obstetric care.
Altruism, as illustrated by hopes that other parents might avoid similar experiences, or that postmortem findings might improve medical knowledge, was less important than the need for knowledge, but was higher than expected. The clinician-researcher had emphasized during the clinical counseling that a postmortem was conducted to try and find answers for them (and their health care professionals), not for anyone else. Previous research has shown that parents expect the advancement of medical knowledge to be an important benefit from postmortem (36).
Less influential in decision-making were barriers to a postmortem. Their relative lack of importance must be considered in the context of this small group of parents from one hospital, most of whom chose to consent to postmortem and were of religious backgrounds where funeral rites are less proscribed than might be the case in other populations (5). Even in the light of publicity surrounding the issues of organ retention at children’s hospitals in the United Kingdom following the events at Bristol and Alder Hey hospitals (37,38), the least important issues for the parents in this study were “worry about what has happened elsewhere”; what will happen to tissues, organs, and the baby’s body after the postmortem; and these issues delaying the funeral. In fact, these and other “barriers” to postmortem consent were of greater concern to younger women, who perhaps were less likely to be aware of the earlier adverse publicity around organ retention than older participants. These findings contrasted with a recent retrospective study where barriers were reported as important factors in decision-making (24), which may reflect different times and study populations, but could also reflect different methodologies. In the present study, all participants were asked prospectively about the importance of issues that can be viewed as reasons to have and reasons not to have a postmortem. In the earlier retrospective study (24), different questions were asked of parents according to their decision; those who did not have the procedure were asked to indicate why they did not consent. The explicit identification of issues in this prospective questionnaire could potentially have a far greater impact on participants’ decisions than we found (23), especially in units where postmortem discussions were held with less-interested or less-experienced clinicians.
Although this study is small and the number of parents who declined any form of a postmortem was low, it would suggest some implications for practice with respect to postmortem consent discussions. Namely, the particular perspectives of some women and their partners may make it more or less likely that a postmortem would be chosen. Thus, parents who had less need for knowledge declined a postmortem. Such an attitude may suggest a group of parents for whom the autopsy consent discussion or counseling is relatively unimportant in influencing their decision on postmortem. A larger study with different populations of parents may well reveal additional characteristics of parents that predispose toward declining a postmortem. If clinicians believe that such parents would benefit from the information to be gained from an autopsy, counseling needs to be more specifically targeted to address underlying attitudes. The apparent fears of the younger women reported here about the process of the postmortem also can be addressed directly. No suggestion was made that participants found it distressing to consider in detail these aspects of a postmortem, but rather, that being encouraged to think helped them to make their decision, feel better about the choice they made, or both (23).
It is of concern, however, that even with specialized preconsent counseling, parents with pregnancies that had miscarried or ended in stillbirth ascribed great importance to the items indicating that postmortems would provide particular knowledge in terms of reassurance that it was not their fault or showing them the cause of their baby’s death. Such high hopes for answers may contribute to this study’s relatively high consent rate for a postmortem, with hopes frustrated if or when parents did not get the information that they expected. Only a small number of postmortems provide information that changes the clinical classification of the cause of stillbirth (39), and perhaps more explicit discussion of this should be included in the consent process.
We were reassured that many parents would want a full conventional postmortem rather than a less invasive alternative, as current data suggest that the former is more likely to provide valuable information for counseling (23,30). As relatively few declined a full or limited postmortem, we are unable to answer with confidence the question as to whether less invasive forms of postmortem are of interest to those who either decline or who request only an external examination. The findings of this study suggest that parents who agreed to full or limited postmortems were a distinct group and would always choose these options over a “less invasive” postmortem (perhaps involving magnetic resonance imaging and organ biopsies) (27,40). However, those declining a full or limited postmortem stated that they might agree to investigations such as placental histology, external examination, and chromosome studies, x-rays (and other imaging), and organ biopsies. In a recent study from the Netherlands, 59 percent of parents agreed to postmortem magnetic resonance imaging, whereas 45 percent agreed to a conventional postmortem (25). Given the uncertainties about the comparability of postmortem magnetic resonance imaging to conventional postmortem, it is important that parents are counseled carefully before less invasive methods are offered. Future developments in postmortem magnetic resonance imaging may mean it could provide equivalent or even superior information to a conventional postmortem (41,42), particularly in the assessment of central nervous system anomalies.
The findings of this study, of course, must be viewed in the context of its small number of participants and having been conducted at a single recruitment site with specialist clinicians who had a particular interest in postmortems. The unit has a marginally higher consent rate for a postmortem (55%) than for the east of England in 2006 (43.7% for stillbirths) and England, Wales, and Northern Ireland (43%) (43). Furthermore, the ethnic mix of our population (predominantly white European) may mean that few parents had religious beliefs that conflicted with the request for a postmortem. Apart from religious beliefs, however, it remains unclear what other parental and obstetric characteristics determine the likelihood of consent to a perinatal postmortem (3,21). We suggest that characteristics of parents which may be important include their age, previous reproductive experiences, and need for knowledge, but these factors should be explored in other hospitals with different populations and postmortem practices. Couples may have discussed the content of the questionnaire, or their responses may have reflected previous conversations about the topics raised, because it was rare for the mother and father (when both responded) to have different views about a postmortem. This factor is significant in the context of law in the United Kingdom, which is interpreted as requiring the mother’s consent before autopsy is performed.
This study was undertaken at a time when consent rates for perinatal postmortems had been in decline in many countries, in part because many clinicians find it difficult to discuss a postmortem with parents for fear of increasing distress. This study suggests that prospective research into parental attitudes to postmortems and decision-making about a postmortem can be performed at the time of perinatal loss and that in-depth discussion of the issues can be helpful. The differences between those parents who do or do not have a postmortem are small. Parents who consent to a postmortem appear to have a greater need for information than those who decline, but altruism is also important for many parents.
Although previous retrospective data had suggested that parents who decline a postmortem did so because of “barriers,” such as concerns about what happens to the baby during the examination; these perceived barriers were not pertinent for parents in declining consent in this prospective study. Parents who agreed to a full postmortem indicated little enthusiasm for less invasive forms of the procedure. Parents who did not agree to a postmortem were generally agreeable to postmortem imaging (magnetic resonance imaging) as an alternative method of examination.
The authors are indebted to the parents who agreed to take part in this study, and to the obstetric and midwifery staff of the Rosie Hospital, Cambridge, United Kingdom. In particular the authors wish to thank Pat Ingham (midwife-counselor), Lisa McKechie, Susan Woolley, and Emma Te Braake, for their support to this work.
- 15Royal College of Obstetricians and Gynaecologists, Royal College of Pathologists. Joint Working Party Report on Fetal and Perinatal Pathology. London: Author, 1988.
- 16Royal College of Obstetricians and Gynaecologists, Royal College of Pathologists. Fetal and Perinatal Pathology: Report of a Joint Working Party. London: Author, 2001.
- 17Confidential Enquiry into Maternal and Child Heath (CEMACH). Perinatal Mortality 2007. London: Author, 2009.
- 18Centre for Maternal and Child Enquiries (CMACE). Perinatal Mortality 2008. London: Author, 2010.
- 34SPSS. Statistical Package for Social Sciences, Version 11.5 for Windows. Chicago, IL: Author, 2003.
- 37Bristol Royal Infirmary Inquiry, Department of Health. Learning from Bristol: The Report of the Public Inquiry into Children’s Heart Surgery at the Bristol Royal Infirmary 1984–1995. London: Stationery Office, 2001.
- 38Royal Liverpool Children’s Inquiry, Report. London: HMSO, 2001., , .
- 43Confidential Enquiry into Maternal and Child Heath (CEMACH). Perinatal Mortality 2006: England, Wales and Northern Ireland. London: Author, 2008.