CLINICIAN’S CAPSULE
What is known about the topic?
Patients who die from poisoning are potential organ or tissue donors but may represent an under-referred population.
What did this study ask?
Do Canadian emergency physicians under-refer patients who die from poisoning for organ or tissue donation, and what characteristics influence referral rates?
What did this study find?
In a simulated context, poisoned patients were under-referred as potential organ or tissue donors, and both environmental and clinician characteristics influenced rates.
Why does this study matter to clinicians?
Understanding factors that affect donor referral may decrease the morbidity and mortality for patients waiting for organ or tissue donation.
INTRODUCTION
Organ and tissue donation is a critical part of treatment for a variety of both acute and chronic illnesses. Unfortunately, the supply of available tissue and organs is far outweighed by the demand.Reference Krishnamoorthy and Souter1, Reference Traino, Alolod, Shafer and Siminoff2 In 2015, 4,631 individuals were on a waiting list for organ transplant in Canada, and 262 died prior to receiving a transplant. In addition to a limited number of eligible donors, the supply of organs and tissue for donation is further exacerbated by missed potential donors.Reference Kutsogiannis, Asthana and Townsend3 The Canadian donation program receives approximately 500 donors per year.Reference Transplantation4 Considering the number of Canadians who die while waiting, there is a very real morbidity and mortality associated with every missed donation. The effect of missed donors also includes significant economic impact because only three additional donors can save the healthcare system more than $1 M,Reference Whiting, Kiberd and Kalo5 intangible psychological costs to families who were not able to facilitate an individual’s desire to become a donor,Reference Rodrigue, Cornell and Howard6-Reference Radecki and Jaccard8 and a failure to improve the quality of life for patients who would have received the donation. As such, any missed donor is a sentinel event, and necessary steps should be taken to avoid it.
Although organ and tissue donations have traditionally come from individuals who die from trauma, sudden cardiac death, or intracerebral bleeding, advances in medical and safety technology have reduced rates from these causes while the demand for donors has increased.Reference Wood, Dargan and Jones9, Reference Wood, Chan and Dargan10 In response, the medical community has explored other potential donors. Individuals who die from poisoning have served as one such source, and, as of 2013, 1% of donors in Canada came from this demographic.Reference Miller, Gardiner and Gubler11 Although there are limited reports regarding the use of organ and tissue donation from poisoned patients, success has been described following exposure to a wide range of toxins.Reference Duque, Duque and Henao12-Reference Chari, Hemming and Cattral17 In addition, Canadian organ donation guidelines do not specifically exclude poisoned patients as potential donors. For tissue donation, many regions such as Ontario and British Columbia require that all in-hospital deaths be referred,18, 19 whereas others such as Nova Scotia, Alberta, and Quebec do have contraindications for referral, but poisonings are not one of the contraindications.20, 21 As such, poisoning should not be a deterrent when considering a potential organ or tissue donor. Despite this, there is some indication that poisoned patients may not be considered as often when compared with other causes of death, but little evidence exists outside expert opinion, and no Canadian studies have been done to explore this issue.Reference Wood, Dargan and Jones22, Reference Hantson and Mahieu23
The limited referral rate for organ or tissue donors is multifactorial and complicated by several factors in the case of poisoning. The determination of brain death is difficult in this setting, and the decision to refer is heavily influenced by practitioner comfort and the substance involved in the poisoning. The greatest influence, though, is insufficient knowledge regarding inclusion criteria and the donation process.Reference de Tourtchaninoff and Hantson24-Reference Sorensen and Copeland28 In Canada, front-line clinicians such as emergency physicians are often responsible for initial screening of potential donors and subsequent activation of the donation network. Any failure at this stage to recognize a patient as a potential donor effectively excludes that individual. To determine whether poisoned patients are appropriately recognized as potential donors, this study explores decisions to refer this population in a simulated context and compares referral decisions to physician characteristics. This will provide information to determine whether poisoned patients are a potentially under-referred population by emergency physicians and to evaluate sources of bias surrounding the eligibility of this demographic.
METHODS
Study design and sample
In this cross-sectional unidirectional survey study, all active, affiliate, resident, and pediatrician members of the Canadian Association of Emergency Physicians (CAEP) were invited to participate (n=1,471). An invitation was sent via their registered email account asking to complete an online survey. The survey was delivered using an online tool, OpinioTM, and consisted of three parts: organ donation scenarios, follow-up questions, and demographics. This study was reviewed and approved by the Nova Scotia Health Authority Research Ethics Board (Study No. 1013656).
Survey tool
The survey tool consisted of 10 cases that were drafted by the research team, each presenting two scenarios with a decision to refer or not to refer for organ or tissue donation. The cases were pilot tested for content, structure, and validity with the research team and a small group of local physicians and residents.
Participants were unaware of the objective to explore organ or tissue donation in the context of poisonings. As such, of the 20 individual decisions to refer or not refer, 10 related directly to organ or tissue donation in the context of poisonings and the other 10 scenarios dealt with other circumstances surrounding organ or tissue donation. One scenario provided an ideal situation for organ or tissue donation that was used as a control. The survey and scenarios can be found in the Supplementary Material.
Follow-up questions and demographics
The survey included a range of follow-up questions and demographics to explore factors influencing the decision to refer or not refer for organ or tissue donation. Four follow-up questions included perceived barriers to organ or tissue donation, formal organ or tissue donation training, a history of referring for organ or tissue donation, and the presence of organ or tissue donation support at the participant’s hospital. Six demographic questions included primary practice setting, certification, province, years of practice, gender, and whether the participants had indicated on government identification their intention to be an organ or tissue donor.
Recruitment
Participation was voluntary. An invitation email was sent to all eligible physicians by the CAEP administrator, which included a weblink to the survey. The survey was open for 39 days with a reminder email sent at 14 days. As an incentive, participants were given the opportunity to provide their email address to be entered in a draw for one of four $25.00 gift cards.
Data collection and analysis
Survey responses were downloaded from OpinioTM into Microsoft ExcelTM. The data were analysed descriptively for demographics, and comparisons were made between control referral rates and key demographics. To calculate comparisons between the control and poisoning scenarios, as well as characteristics of physician groups, total referrals were summed for scenarios involving a poisoning (scenarios 2, 4a, 5a, 6a, 7, 7a, 9, 9a, 10, 10a), and comparisons were made using odds ratios and 95% confidence intervals (CI). Descriptive statistics, odds ratios, p-values, and confidence intervals were calculated using IBM SPSS, and results were reviewed by the Department of Mathematics and Statistics at Dalhousie University.
RESULTS
Response rate and demographics
Overall response rate for the donation scenarios was 208/1471 (14.1%) and 175/1471 (11.9%) for the demographics survey; 83 (47.4%) respondents were male; 118 (67.4%) indicated that they worked in a large, urban hospital; 32 (18.3%) indicated a membership with the Canadian College of Family Physicians (CCFP), 63 (36.0%) with CCFP with a certification in emergency medicine (EM), and 52 (30.0%) with the Royal College of Physicians and Surgeons of Canada (RCPSC); 104 (59.4%) participants indicated <10 years of service; 43 (24.6%) had previous organ donation training; 116 (66.2%) had referred for donation in the past; 111 (63.4%) had donation support available; and 143 (81.7%) indicated on government identification their own intent to be a donor. A summary can be found in Table 1.
Demographic information (N=175)
ICU=intensive care unit, PEI=Prince Edward Island.
Of the participants who indicated having donation support, 77 (69%) indicated having an organ donation coordinator, 45 (41%) a screening tool or guideline, and 8 (7%) a human resources coordinator. Other less commonly indicated resources included phone consults or designated nurses.
Organ or tissue donation referral in poisoned patients
The survey included 10 scenarios that explored organ and tissue donation in the context of poisonings. Excluding incomplete scenario responses, this resulted in a total of 1,871 decisions to refer or not refer, and referral was chosen in 1,404 or 75.0% (95% CI [73.0%, 77.0%]) of cases. This compared with a control scenario where 170 or 92.4% (95% CI [91.1%, 93.7%]) referrals were made. Using an odds ratio, this indicates that Canadian emergency physicians were 4.0 (95% CI [2.3, 7.0]) times more likely to make referrals in the control scenario as compared with scenarios involving a poisoning.
Organ or tissue donation referral by a specific poison
In scenarios presenting an overdose involving a prescription agent, 117 (63.6%) physicians referred for organ or tissue donation. Referral rates for other common agents included acetaminophen 143 (77.3%), carbon monoxide 171 (91.9%), non-prescription or recreational drugs 129 (64.5%), resuscitation medications 171 (89.5%), unknown chemical exposure 133 (73.5%), and organophosphate exposure 94 (52%). A full summary of referral rates by scenario can be found in Table 2.
Referral rates by scenario
Organ or tissue donation by demographic and geographic
Participants who worked in an urban centre were 3.8 (95% CI [2.8-5.1]) times more likely to refer for potential donation in the setting of poisoning when compared with non-urban centres. Likewise, those with the RCPSC or CCFP (EM) certification were 3.6 (95% CI [2.8-4.7]) times more likely to refer when compared to CCFP members. Male participants were 2.2 (95% CI [1.7-3.0]) times more likely to refer when compared with female participants. Participants with more than 10 years of practice were 2.1 (95% CI [1.6-2.7]) times more likely to refer when compared with those with less than 10 years of practice. Participants who had referred for organ or tissue donation in the past were 4.3 (95% CI [3.2-5.8]) times more likely to refer. Physicians working in an environment with organ or tissue donation support were 3.9 (95% CI [2.9-5.2]) times more likely to refer. Those who had attended organ or tissue donation training in the past were 2.6 (95% CI [2.0-3.5]) times more likely to refer; and, finally, those participants who had indicated their own desire to be a donor on government identification were 5.8 (95% CI [4.2-8.0]) times more likely to refer. A summary can be found in Table 3.
Odds ratio of organ or tissue donation referral in poisoned patients by physician characteristic
CI=confidence interval
Practice location was significantly correlated with other demographics. Physicians working in a large, urban hospital were 12.1 (95% CI [5.7-25.6]) times more likely to have available donation support, 4.3 (95% CI [2.2-8.4]) times more likely to have referred in the past, 43 (95% CI [15.6-123.5]) times more likely to have EM certification, and 3.4 (95% CI [1.5-7.5]) times more likely to have signed government identification indicating their own intention to be a donor.
Provinces with enough participation to calculate an odds ratio with statistical significance included British Columbia and Ontario. In British Columbia, physicians were 4.7 (95% CI [1.4-15.9]) times more likely to refer in the control scenario when compared with poisoned patients. In Ontario, physicians were 3.3 (95% CI [1.2-9.3]) times more likely to refer in the control scenario as compared with poisoned patients. A summary for each region can be found in Table 4.
Odds ratio for referral of non-poisoned patients by region
CI=confidence interval
Perceived barriers to organ or tissue donation
Participants had the opportunity to select from a list of potential barriers to organ or tissue donation, and some barriers were selected more often than others (p<0.0001). The most common barrier was clinician familiarity with the referral process (n=113). Other notable barriers included that the deceased patient’s wishes were unknown (n=87), the physician’s failure to identify or refer potential donors (n=78), and lack of time to discuss donation with the family (n=76). Barriers uncommonly selected included negative attitudes towards organ or tissue donation among clinicians (n=6), the perception that clinicians will provide suboptimal care to potential donors (n=16), and racial, ethnic, or religious perspectives on organ donation (n=31). A summary of these findings can be found in Table 5.
Perceived barriers to organ or tissue donation
DISCUSSION
By comparing the demographics of physicians deciding to refer or not refer poisoned patients for organ or tissue donation, we can explore factors that may affect referral rates. This is important because bias has been shown to significantly impact clinical decision-making and can lead to costly errors such as missed organ donation.Reference Croskerry27 For example, poisoned patients may be associated with characteristics such as chronic viral illnesses or other high risk behaviors that predispose clinicians to exclude them as potential organ or tissue donors.Reference Sorensen and Copeland28 Because the presence of chronic viral illness or other active disease processes may be a contraindication for tissue donation in some Canadian provinces,20, 21 clinicians should be careful that an inappropriate association between these characteristics does not influence a referral decision in an isolated poisoning context.
Physicians who indicated working in large hospitals having EM certification or male gender were associated with higher referral rates of poisoned patients. Being aware of relatively non-modifiable factors such as these allows physicians to identify biases or trends and is a first step in mitigating their effect.Reference Green, Carney and Pallin29 There are many strategies that have been developed to reduce the effect of bias, ranging from decision tools to cognitive strategies. Examples include focusing on patient characteristics independently of other attributes or attempting to view a situation from a different perspective.Reference Chapman, Kaatz and Carnes30 These strategies have yielded some success in reducing biased decision-making, and so exploring one’s own attitudes, values, and risk factors for under-referral may have a significant effect on donation rates.
Non-modifiable characteristics do not exist in isolation of other, modifiable, characteristics. Although working in a large, urban hospital was associated with a higher referral rate, physicians in this context were also more likely to indicate having access to valuable donation support, having referred in the past and to have signed government identification indicating their own intention to be a donor. Therefore, when attempting to mitigate the effects of non-modifiable factors, other related characteristics can be targeted.
Physicians indicating on government identification their own intention to be a donor, when compared with those who did not, were 5.8 times more likely to refer potential donors in the poisoned context. This contrasts with only 2.8% of participants indicating that they thought that negative attitudes towards organ or tissue donation among clinicians presented a significant barrier to donation. This supports the theory that clinicians’ values can have a significant impact on their decision-making, that they are often unaware of these biases, and that becoming aware of these biases is an important first step to improving decision-making in clinical contexts.
Level of training, both in organ and tissue donation as well as the RCPSC or CCFP (EM) certification, significantly affected one’s decision to refer. In addition, clinician familiarity with the referral process was identified as the most common barrier to successful organ or tissue donation. This illustrates that knowledge surrounding the donation process is critical for improving overall referral rates for organ donation, supporting the lack of familiarity with donation policy and process has been shown to decrease successful organ referrals by 25% to 50%.Reference Sorensen and Copeland28, Reference Caramiciu, Adams and McKown31 Although most healthcare providers recognize the urgent need for donation, significant misconceptions in eligibility exist, and only a small minority have engaged in dedicated donation training.Reference Traino, Alolod, Shafer and Siminoff2, Reference Morris, Wilcox, Noreuil and Frist32 In our study, 34.5% of participants indicated no prior training, and Canadian medical students have been found to have only a limited knowledge of organ donation.Reference Bardell, Hunter, Kent and Jain33 In addition, it is important to stress that refresher training is necessary to prevent decay of medical knowledge and skills, and that the effect of education is highly dependent on the time since training.Reference Weaver, Newman-Toker and Rosen34 This offers institutions an opportunity to implement new or reinforcing training programs, which may improve donation rates even if primary training exists.
Finally, our study also illustrates the importance of available donation support for clinicians. Physicians with available support were 3.9 times more likely to refer poisoned donors. The reasons for this are likely multifactorial. Emergency departments are often busy and unpredictable environments are not conducive to prolonged or sensitive discussions such as decisions regarding organ or tissue donation. Correspondingly, the lack of time to discuss donation was selected in our survey as one of the most common barriers encountered in the emergency department. As such, in addition to providing information and support, donation professionals may play an important role in the experience of families considering donation, because less than half of families approached by clinicians provide consent.Reference Jacoby, Breitkopf and Pease35 This stresses the importance of trained individuals serving in multiple roles in the donation process, including being available to support emergency physicians in the department.Reference DeJong, Franz and Wolfe36
LIMITATIONS
Although this was a national survey, it was limited to emergency physicians registered with CAEP. Participants varied significantly in clinical setting, stage, and type of training, and operated under a wide range of health authorities with different policies and procedures. Although our response rate was significantly powered for our calculations, it may not represent Canadian emergency physicians as a whole due to response or selection biases as evidenced by the high organ or tissue donation referral rate in our survey. In addition, although we were able to look at multiple demographics individually, some of these demographics are interrelated (i.e., larger centres also have a greater likelihood of organ and tissue donation support). Finally, this study was looking at physician decisions to refer in a survey context, and it is unclear how well this correlates with actual clinical decision-making.
CONCLUSION
This study demonstrates that, in a simulated setting, Canadian emergency physicians under-refer patients who die from poisoning as possible organ or tissue donors and identifies potential methods to improve these rates. By examining characteristics that put clinicians at risk for under-referral of organ or tissue donors, becoming aware of potential biases, improving the knowledge base of physicians, and implementing programs that support the organ and tissue donation process, we may have the opportunity to improve these rates and reduce morbidity and mortality for Canadians requiring organ or tissue donation.
Acknowledgements
We thank Melissa MacDougall for her contribution in editing this project, the physicians who assisted us in piloting the online survey, the CAEP administration who assisted in distributing the invitation emails, and the physicians for their continued support of research in emergency medicine. This research was supported by a grant received from the Dalhousie University Faculty of Medicine, Mattar fund.
Competing interests
None declared.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit https://doi.org/10.1017/cem.2018.43
