Article Text
PDF
Abstract
Objectives To investigate risk perception relating to living kidney donation, to compare the risk donors would accept with current practice and identify influential factors.
Design An observational study consisting of questionnaires completed by previous living donors and the general public. Participants selected the risk they would accept from a list of options, in various scenarios. Risk communication was investigated by randomly dividing the sample and presenting risk differently.
Setting Primary care (two centres) and secondary care (one centre), London.
Participants 175 questionnaires were sent to patients who had previously undergone living-donor nephrectomy and to members of the public consulting a general practitioner. The living-donor sample comprised 77 consecutive donors at Guy's Hospital from May 2003 to January 2005. The general-public sample was recruited from two London healthcare centres. Of the eventual 151 participants, 61 were living donors and 90 were from the general public.
Main outcome measure The amount of risk a participant would accept to donate a kidney.
Results 74% of participants were willing to accept a risk of death higher than 1/3000. The most commonly accepted risk was 1/2 (29%). Those presented with a ‘chance of survival’ accepted higher risks than those presented with a ‘risk of death’ (p<0.01). Greater risks were accepted when the recipient was closely related and, for some, when the recipient's prognosis was worse. No difference was observed between the living-donor and general-public groups.
Conclusions Kidney donors will accept a higher risk of death than is currently quoted, especially if risks are presented in terms of chance of survival.
- Applied and professional ethics
- codes of/position statements on professional ethics
- donation/procurement of organs/tissues
- informed consent
- kidneys
Statistics from Altmetric.com
- Applied and professional ethics
- codes of/position statements on professional ethics
- donation/procurement of organs/tissues
- informed consent
- kidneys
Introduction
Kidneys obtained via living-donor nephrectomy are favoured over cadaveric donor kidneys for a number of reasons, including longer graft survival and less time on the transplant waiting list. The wait for a cadaveric organ is often long and unpredictable. Organs are allocated according to tissue typing and recipient need rather than on a first-come first-served basis. An elective, live-donor procedure has many benefits, including the optimisation of the recipient before surgery. This optimisation maximises the chance of graft survival and a successful long-term outcome. Prior to donation, both the donor and the recipient undergo extensive counselling, in which living donation is discussed alongside the alternative treatment options. The risks of living donation are also discussed. The estimated risk of death from donating a kidney is 1 in 3000.1 The risk of perioperative complications is around 10%–15%.2 3
Many contributory factors may influence the level of risk someone is willing to accept in order to donate a kidney.4 These may include their relationship to the recipient, the urgency of recipient need, the donor's previous experience of major surgery and the impact of preoperative counselling. The language and methodology used to present risk may also be influential, with risk communicated in a positive way leading to higher uptake than when it is presented negatively.5 Many transplant units refuse to accept donors who are at increased risk (marginal kidney donors) and there is also controversy over the acceptability of living liver donation where the risks of mortality and complications are more substantial.6 7 Living kidney donors are usually highly motivated and are often dismissive of perioperative risks.8
The aims of this study were fourfold: (1) to determine the risk of death that kidney donors would accept; (2) to determine whether risk-taking is affected by recipient-related factors, such as clinical need and relationship to the donor; (3) to assess which donor-related factors, such as how the level of risk is presented, affected risk-taking and (4) to determine whether the general public, when faced with hypothetical donation, had similar approaches to risk as those who had already undergone the procedure for a friend or relative—that is, whether such attitudes are consistent with broad societal views rather than an individualistic response to a sick friend or relative.
Methods
One hundred and seventy-five questionnaires were sent to two sample groups of patients: those who had previously undergone living-donor nephrectomy and members of the public attending their general practitioner. The living-donor sample consisted of 77 consecutive living donors at Guy's Hospital between May 2003 and January 2005. Participants were recruited by post and those failing to reply were sent one postal reminder. Each person undergoing donation within that timeframe was invited to participate in the study. These dates were selected to ensure near-equal numbers of open and laparoscopic donors, May 2003 being when laparoscopic-assisted donor nephrectomy became established at Guy's Hospital and January 2005 being when the study began. Each questionnaire was accompanied by a covering letter which stated that participation was voluntary and that future care would not be influenced either by the answers given or a decision not to participate. Before donation, each participant in the living-donor sample received extensive counselling, meeting with a living-donor transplant co-ordinator, a nephrologist and a transplant surgeon. In addition, each donor and recipient was provided with written information and a DVD further detailing the living donation process. Questions were encouraged and were answered honestly and appropriately.
A comparative, general-public sample was recruited by a researcher (NHM) attending two different healthcare centres on two consecutive days in south east London. Ninety-eight questionnaires were distributed, one to each adult patient attending the practice on the designated days. Accompanying the questionnaire was an information sheet about the study and the practice of living donation and a consent form. It also stated that the researcher was acting independently of the general practice and that participants' answers would not be relayed to their general practitioner or influence their future care. Once consent was obtained, participants completed the questionnaire while waiting to be seen.
Participants were presented with three scenarios in which their closest relative was in need of a kidney transplant (table 1, section 1). They were asked to select the maximum risk of perioperative death from kidney donation they would accept from a list provided. In order to investigate recipient factors affecting risk acceptance, each scenario had a progressively worsening recipient prognosis. In addition, participants were also asked whether by comparison they would be more, equally or less likely to donate to a list of potential recipients with whom they would have various relationships (table 1, section 2).
Questions asked of participants
The effect of different phraseology in the communication of risk was investigated by using two questionnaires, each with the same questions but with risk presented in alternative formats (table 1, section 1). Questionnaire A presented risk as ‘risk of death’ and questionnaire B presented it as ‘chance of survival’. Participants were randomly allocated to either questionnaire.
Demographic data were collected from each participant and patient-defined postoperative complications were recorded for the living-donor sample.
Statistical method
Results were analysed using Pearson's unadjusted χ2 and Fisher's exact test where appropriate. A mortality risk of 1 in 20 was used as a threshold for analysis, since this is clearly beyond a clinically acceptable level of risk in current practice.
Results
In total, 151 responses were obtained, with a response rate of 79% (61 responses) for the living-donor sample and 92% (90 responses) for the general-practitioner sample. The mean time after surgery in the living-donor group was 298 days (SE 18). Thirty of the living donors had undergone laparoscopic surgery. For the living-donor sample there were no reasons given for not responding to the questionnaire. For the general-public sample the commonest reason for not completing the questionnaire was insufficient time.
There was a significant difference between the ages of the two groups (p=0.0004). The mean ages were 49 and 41 years, for the living-donor group and the general-public sample, respectively, and the respective numbers of women in the two groups were 33 (54%) and 51 (57%). There was no significant difference between the risks accepted by men and women (p=0.5). Demographic data collected from the general-public sample showed that risk acceptance was not affected by ethnicity (p=0.178), knowledge of those who had experienced a transplant (n=11), having previously undergone a general anaesthetic or having children.
Maximum risk of death accepted
One hundred and twelve people (74%) were willing to accept a risk of death greater than 1 in 3000 in donating a kidney (table 2), and 38 (62%) of the living-donor sample and 44 (49%) of the general-public sample were willing to accept a maximum risk of 1 in 20 or greater (difference between samples non-significant, χ2 p=0.105). A total of 44 (29%) subjects would accept a risk of death as high as 1 in 2.
Maximum risk of death accepted for living kidney donation
Recipient-related factors affecting risk
Although prognosis was an influential factor for some, worsening recipient prognosis did not significantly affect risk-taking. Table 3 shows no difference in the proportions accepting a risk of death of 1 in 20 or greater as prognosis worsened (p=1.0). A steady decline in readiness to donate was observed as the relationship between the donor and the recipient became more distant (table 4). The only exception to this trend was when the potential recipient was a child.
Subjects accepting risk of death ≥1 in 20 with worsening recipient prognosis
Proportions of respondents less likely to donate according to recipient relationship when compared with their willingness to donate to their closest relative
Donor-related factors affecting risk
Risk communication
When risk was presented positively, as ‘chance of survival’ (questionnaire B), greater risks were accepted in response to a general question. In group A (presented with a ‘risk of death’), 39% accepted a risk greater than 1 in 20, versus 71% of group B (p<0.001). However, when more specific questions were asked including details of recipient prognosis (table 3), the difference between the two groups decreased as prognosis worsened (although a trend appeared likely). Similarly there were no consistent differences between responses to questionnaires A and B as the relationship to the recipient altered (table 4).
Complications in the living-donor sample
In the living-donor sample, nine donors reported having experienced systemic complications (including raised blood pressure, postoperative nausea and vomiting, leg and foot numbness, urinary tract infection, pneumothorax, swollen testicle or back pain) and seven had had complications at the wound site (including swelling, nerve damage, scar tenderness, bleeding or delayed healing). Eleven reported having had persistent wound pain when specifically asked. Thirty-seven (61%) had had no complications at all. There was no significant difference in risk acceptance between those who reported complications and those who did not (p=0.05 for risk >1 in 20). There was no significant difference between those who had undergone open and laparoscopic procedures and the level of risk accepted (p=0.841).
Living donors versus general-public group
There was no difference between these groups in response to a general question on maximum risk of death accepted (table 2, p=0.105), but the living-donor group accepted significantly higher risk in response to specific questions including recipient prognosis (table 3). Members of this group were less likely to donate to a stranger than were those of the general-public group (table 4).
Discussion
Our findings show that kidney donors are willing to accept surprisingly high risks to their own lives, with many (29%) accepting a risk of death as high as 1 in 2. Relationship to the potential recipient and the presentation of risk both significantly influenced risk-taking, whereas recipient prognosis had little effect. These findings are of particular importance in circumstances where a donor with a higher than average risk wishes to donate.
The implications of these results may be interpreted differently when considering the living-donor and general-public groups individually. Those who had already donated were found to accept higher risks than the general-public sample. This is unsurprising, as these participants were clearly motivated towards an individual and fully understood both the implications of renal failure and the donation process. Their willingness to undergo the process again is testimony to the donation process and its benefits to both the donor and the recipient.
A significant proportion of people in the general-public sample, who were asked to hypothetically consider their ‘closest relative’ requiring a transplant, were still willing to accept the highest risks: 58% were willing to accept a risk of 1 in 20 or higher. One could argue that this sample group had relatively little understanding of the transplantation process and that they might be less willing when faced with the prospect of donation in reality. However, this appears not to be the case, considering the responses of those who had already donated. Whereas one would expect willingness to donate to decrease when the potential donor is faced with the operative complications, figures obtained from the previous living-donor sample were to the contrary, demonstrating that the willingness to donate increases when faced with the complications in reality.
It is worth considering that living donation is presented as one of many options to the patient with renal failure, and prior to agreeing to donate, potential living donors undergo extensive counselling about the risks associated with surgery. As mentioned previously, living donation has many advantages over cadaveric donation, which remains an option for those who decide not to undergo living donation. Living donation is associated with improved graft survival, both because a better tissue match is found and because the recipient's medical condition can be optimised so the transplant takes place when the patient is fit, maximising the chance of a positive long-term outcome.
Extra care and consideration are taken over difficult cases—for example, where a potential donor is considered high risk. In order to protect both the donor and the recipient, the British Transplantation Society guidelines recommend ‘independence between the clinicians responsible for the donor and the recipient … (thereby) allowing for, in effect, a donor advocate.’9 Cases are discussed impartially by surgeons, nephrologists and other transplant professionals before donation is allowed to take place. As a result, donor surgeons are often not aware of the recipient's details when considering a potential donor, although clearly the donors themselves are.
Whereas the number of people willing to donate might be expected to decrease once potential donors are faced with the associated risks, our study shows that this does not happen. By demonstrating that risk-taking was consistently high in both those who had already donated and those who did not have an individual in mind, the evidence supports that willingness to accept high risks for donation may not simply be an individualistic or personal response to be considered extraordinary, but rather a societal view that is in fact the norm.
Clearly the results of this study highlight many ethical issues, in particular the role of paternalism and autonomy in medicine. Autonomy is the capacity to think and decide, and to act on the basis of such thought and decision, freely and independently,10 and the legal requirement for consent is a means of recognising its importance. Young states, ‘the seriousness of the recognition is tested, however, whenever a patient autonomously decides on a course of action which runs counter to the judgement made by [their] healthcare advisers about what would be medically best’.11 An example of this is when a high-risk donor may wish to donate against the advice of the transplant team. Although the practice of medicine has a patient-centred approach where patient autonomy is an integral component, one cannot fail to recognise the ongoing duties of the medical profession, principally to do no harm. This is especially poignant in this context, since the living related donor is unusual12 in undergoing ‘harm’ for another's gain. One must also remember that the right to refuse must extend to both the patient and the clinical team, who, as well as having responsibilities to their patients, also have a right to autonomy. The British Transplantation Society guidelines state that ‘if a fully informed donor wishes to proceed with a course of action that involves risks of mortality or morbidity more severe than the team find acceptable, they are under no obligation to proceed. Referral for a second opinion would be appropriate in such circumstances.’9
Although the issue of donor autonomy has attracted little attention recently, it has been raised previously. In the 1980s, when living organ donor programmes were not as established as they are now, calls were being made to eliminate the ‘pseudoprotectionism’ practiced by transplant teams12 and to allow donors to accept risks over and above the recommendations of transplant professionals. A paper by Spital and Spital13 found that the ‘great majority of the public and non-transplant physicians support[ed] the use of living … donation and believes that donors should be allowed to accept added risk’. In support of this, a paper by Cotler and colleagues14 looking into living liver donation also identified the ‘substantial’ levels of risk that potential living liver donors would be willing to accept, noting that many would be prepared to donate ‘even if the transplant recipient had a relatively low chance of surviving transplantation’. This paper also noted that donors wished to remain autonomous when making decisions about living donation, being ‘allowed to have the final say regarding candidacy for living donation’.
While these studies have indeed found, as we have, that donors are willing to accept high risks, those risks have never been explored in detail nor quantified. A study published in March 200815 considered the level of risk accepted with regard to complications (such as hypertension, cardiovascular disease and kidney failure) but did not look at death in particular. Our study has further demonstrated both that there is a discrepancy between the views of donors' and of the transplant community and that hypothetical and real donors differ little in attitudes to risk.
Although this paper has demonstrated that many people would accept the highest risks in order to donate, this does not suggest that all potential donors with risks as high as these should be allowed to donate and nor are we suggesting that current standards for exclusion should be lowered. Indeed, many recipients would be unwilling to undergo such a procedure if the risks to the donor were so high. However, autonomy is clearly important when considering those people who are marginal donors and at only slightly more risk than those who are normal. While in reality it is possible that fewer people would be willing to accept the highest risk of death of 1 in 2, the majority will clearly accept a risk of death as high as 1 in 100. Although the level of acceptable risk to the transplant community has not been assessed directly in this paper, other publications15 have shown this is significantly higher than most surgeons would currently accept. In view of the recent debate about accepting higher-risk donors,16 this finding is of particular importance. Thus higher-risk donors may well accept these risks, raising difficult issues regarding patient autonomy for the transplant community. There is undoubtedly a conflict between the principles of medical ethics, mainly those supporting autonomy and non-maleficence, where one must consider which of these principles takes precedence.
A unique factor investigated as part of this study is that of risk communication. Our study showed that the method of communicating risk does affect the degree of risk accepted. Of all those selecting the highest risk of 1 in 2, 67% were from group B (where risk was phrased as a ‘chance of survival’). These results show that there is potential for patients to be strongly influenced by the language used to explain risk and highlights how subtle discrepancies in the methods used may unintentionally influence the patient. Although this finding is not new, it has not previously been discussed within the context of living donation and suggests that approach to risk is not immutable despite the highly charged emotional atmosphere surrounding donation. Given the varied appreciation and understanding of risk within the general population, the way that risk is communicated and the amount of information each person requires often varies. With regard to whether risk should be presented to patients in terms of death or survival, the best approach is clearly to do both, thereby providing the best explanation and ensuring the greatest degree of understanding.
For many donors, however, risk may not be an issue. Potential donors often become ‘impervious to or oblivious of any suggestion that they should make a more informed decision’.9 The consent itself may be very real in that it comes without coercion, but in many cases it may represent an instantaneous decision rather than one made after careful consideration of the issues. This possibility is supported by our study showing that some members of the public were willing to accept the highest risks without lengthy consideration of the issues. It is difficult to be certain whether these same people would definitely proceed if faced with such risks, but the willingness of previous donors to accept higher risks does suggest that these results have some validity. In addition, given the general public's limited experience and understanding of the donation process, this finding also suggests that the decision to donate for some people could possibly be an instantaneous one, with pre-donation education having little influence on the willingness of donors to donate. However, preoperative counselling is clearly essential and should continue.
Another important finding of the study was the influence of the prognosis for the recipient. Our results show that for most (60%), risk was unaffected by the prognosis (in terms of mortality). It is unclear whether this also reflects the instantaneous nature of the decision, with greater concern for the recipient's quality of life rather than their mortality or other factors.
Not surprisingly, our study reiterated previous research findings that donors are willing to accept higher risks for those closer to them.13 The only exception to this is when the potential recipient is a child, when some people would be willing to accept the same risk for a stranger as they would for a close family member. This clearly has important implications when considering non-directed donation.
This study has a number of recognised limitations. Although it was validated in part by comparing previous living donors with the general public, a larger sample would be desirable to validate the results further. In addition, a number of variables were discussed within a relatively small sample size and this may preclude definitive conclusions. With regard to the living-donor sample, all donors had had a positive experience of the donation process with minimal complications. With the benefit of hindsight they were possibly willing to accept higher risks than they would have done before donating. This is a source of potential bias but does not appear to have affected the results obtained. With regard to the general-public sample, it was very difficult to appreciate the degree of understanding of the topic being discussed. In addition, there is a possibility that participants from both sample groups might have been concerned that their answers would be made available to their doctor or surgeon and their care compromised as a result. The information sheets accompanying both sets of questionnaires explicitly mentioned the process of data anonymisation and (for the general-public sample) that the researcher was acting independently of the practice. Another limitation was that the only risk participants were asked to consider was the risk of death. It is possible, if not likely, that certain other risks affecting the donor's postoperative quality of life would be of more importance in this consideration of whether to become a donor, but we have not considered these. In another study performed since, we have demonstrated that donors' expectations are often not met after surgery.17 This suggests that there are specific quality-of-life issues that should be further discussed prior to donation. It would have also been beneficial to seek the opinions of the transplant community, both with regard to what risk they would be willing to accept themselves and what risk they might be prepared to operate under for their patients. A valuable extension to this study would be to question the transplant community and those currently preparing to be kidney donors.
For ethicists, this study raises some important issues that require further exploration. Can autonomy in this context be considered to ‘trump’ other ethical principles, such as non-maleficence, given that objective decision-making is hard for donors in these circumstances?18 Can beneficent paternalism have a role in the management of these patients? How can we deal with the subjectivity inherent in risk acceptance by the donor and risk presentation by the transplant team? Can recipient prognosis really have no bearing on the decision by the donor and the transplant team? Should the donor be permitted to take higher risks when the recipient prognosis is worse?
Conclusions
In conclusion, we have shown that living donors are prepared to take much higher risks than many surgeons would contemplate. Risk-taking is influenced by presentation of risk and, for some, the prognosis of the potential recipient. It is not influenced by age, sex, ethnicity, previous complications of surgery, previous experience of a general anaesthetic or knowing a transplant recipient. Similar results in members of the general public suggest that these are societal rather than individualistic behaviours.
What is already known
The risk of death from living kidney donation is 1 in 3000.
Understanding of medical risk is affected by presentation.
What this study adds
Both real and hypothetical donors are willing to accept very high surgical risk.
Risk acceptance is affected by relationship to the recipient and, for some, recipient prognosis but not other factors.
Higher risks are taken by donors when risk is presented in a positive fashion.
Acknowledgments
The authors would like to thank Lisa Burnapp (Transplant Co-ordinator, Guy's Hospital, London) for her help in acquiring data for this study and Antonia Cronin (Locum Consultant in Renal Medicine, Guy's Hospital, London) for her comments and suggestions.
References
Footnotes
Funding Financial support was received from the Guy's Hospital Renal Department research fund for administration costs in the data collection process. The research was conducted independently of the funder. The funder was not involved in data interpretation, the writing of the report or the decision to submit the article for publication.
Competing interests None.
Ethics approval This study was reviewed and approved by the Guy's and St Thomas' Research Ethics Committee. Consent was presumed for all those returning a questionnaire by post. All those completing a questionnaire in the presence of a researcher completed a consent form prior to participating.
Data All authors had full access to all of the data (including statistical reports and tables) in the study and can take responsibility for the integrity of the data and the accuracy of the data analysis.
Provenance and peer review Not commissioned; externally peer reviewed.