Organ Procurement Organization Approach of ICU Families in Person Versus by Phone for Donation Authorization

Abstract

OBJECTIVES:

Due to hospital visitor restriction policies, the COVID-19 pandemic forced many conversations between ICU patients’ families and organ procurement organizations (OPOs) to be conducted by phone, as opposed to in-person. We aimed to determine if OPO phone approaches are a negative predictor of authorization for donation.

DESIGN:

Retrospective observational study.

SETTING:

Hospitals partnering with New England Donor Services (NEDS) as their OPO.

PATIENTS/SUBJECTS:

Multicenter database study of all ICU patients from 2017 to 2021 whose families had been approached by NEDS. Of note, registered organ donors who were declared brain dead were excluded due to their prior first-person authorization. However, registered organ donors who were candidates for donation after circulatory death were included in the analysis because of the family’s role in controlling the timing of ventilator withdrawal and, thus, their critical role in determining donation success.

INTERVENTIONS:

None.

MEASUREMENTS AND MAIN RESULTS:

In addition to whether OPO approaches occurred in person or by phone, we extracted covariate data on patient and surrogate demographics, OPO representative training, and extent of ICU-OPO collaboration. Univariate and multivariate analyses were conducted to determine predictors of successful authorization. Among 2240 approaches of potential organ donors, OPO donation discussions by phone constituted 221 of 1282 (17%) of authorizations, as opposed to 134 of 958 (14%) of failed attempts (p = 0.04). In a multivariate model, phone approach was no longer significantly negatively associated with authorization. However, patient race/ethnicity, registered status for donors after cardiac death, ICU-OPO collaboration, specialized OPO representative training, OPO representative being the sole participant in authorization discussions, and surrogate relationship were strong predictors.

CONCLUSIONS:

OPO approach by phone is not a barrier to organ donation. Strategies for successful donation discussions should focus on collaborative processes between ICU teams and OPOs and involvement of OPO representatives with special training for discussing authorization.

KEY POINTS

Question: Are organ procurement organization (OPO) approaches of ICU patients’ families by phone (as opposed to in person) a negative predictor of authorization?

Findings: In a multivariate analysis of a large database of ICU patients whose families had been approached by an OPO, approach by phone was not an independent predictor of authorization, but close ICU-OPO collaboration and OPO representative training were strong positive predictors.

Meaning: OPO approach by phone is not a barrier to donation success. Cultivating a collaborative process between ICU teams and OPOs and prioritizing the involvement of OPO representatives with special training for discussing authorization is critical.

The critical need for transplantable organs is a pressing problem, as demand far exceeds the utilization of available supply (1–4). Studies estimate that an average of 17 people die in the United States each day awaiting a life-saving organ transplant (5). This problem was exacerbated in the pandemic, as data suggests COVID-19 reduced patient access to solid organ transplantation (6–8). Understanding the barriers to surrogate authorization for organ donation in the ICU is a critical step in increasing organ donation rates in America.

Previous literature has found a variety of factors that are associated with increased rates of organ donation authorization for eligible patients in the ICU. Factors that have been generally established as standard practice include making an in-person request in a private setting (9) and decoupling the request, which refers to separating the timing of the request for donation from the pronouncement of death (1, 10). The quality of care provided to potential organ donors by the hospital care team during the hospitalization is also an influencing factor (1, 11), along with the empathetic delivery of information (12), emphasizing that donation can save lives (12), ensuring the family member has adequate comprehension regarding the process of organ donation (1, 11), and understanding the meaning of brain death (1, 13, 14). Personal characteristics of the surrogate decision maker also play a role in the authorization success. There is a body of literature on how minority race/ethnicity (4, 14–16), culture (13, 17), religion (17–21), and age (22, 23) may impact family authorization rates.

One factor that has not been reported is how approaching a potential surrogate decision maker in-person vs. over phone affects authorization success. This question of in-person vs. by phone approach became especially relevant in the era of COVID visitor restrictions. While before COVID-19 most organ donation discussions were commonly done in-person, this became increasingly difficult given pandemic restrictions on visitor presence in ICUs. An increased reliance on telehealth was seen during the pandemic (24), and the effects of this on organ donation rates were unknown.

Our objective was to investigate the relationship between phone vs. in-person organ donation authorization success by organ procurement organizations (OPOs) in the ICU, both before and during the peak of the COVID-19 pandemic. Our hypothesis was that a phone approach of an ICU patient’s family would be a negative predictor of success, given that historically in-person approaches have been OPO standard practice.

MATERIALS AND METHODS

Study Design and Setting

We conducted a retrospective observational study of a dataset of ICU patients from 2017 to 2021 whose families had been approached by New England Donor Services (NEDS) for organ donation. This study (No. 2000030119) was determined to be not human subjects research by the Yale Human Investigation Committee and was therefore exempt from review and approval.

Dataset

The dataset for this study came from an internal database maintained prospectively by NEDS for quality tracking purposes. Data from each case in the database was populated into a standard template with defined data fields as part of routine workflow by NEDS staff, via real-time chart review and discussions with clinical teams. One of the authors of this study (B.M.), a NEDS senior data analyst, manually organized this previously collected data for this study’s retrospective analysis. We selected 2017 as a starting year to capture pre- and post-pandemic times. Due to the COVID-19 pandemic, authorization conversations were mostly conducted by phone after March 2020, but before the pandemic, phone approaches happened only on occasion and were captured in the database. The few participants younger than 1 year old in the database were excluded from this data extraction due to a desire to keep age data expressed in whole numbers of years, as opposed to fractions. Registered organ donors who were declared brain dead were also excluded due to their prior first-person legal authorization for donation. However, registered organ donors who were candidates for donation after circulatory death were included in the analysis because of the family’s role in controlling the timing of ventilator withdrawal and, thus, their critical role in determining donation success. Of note, during the study timeframe, all possible family approaches were attempted either in-person or by phone, with no completely missed opportunities.

Variables and Data Collection

Our primary variable of interest was in-person vs. phone approach with the patient’s surrogate decision maker for donation. Patient and surrogate variables extracted included patient age, patient gender, patient race and ethnicity, patient’s hospital admission date, type of donation (mechanism of death—brain death vs. cardiac death—in conjunction with donor registration status of patient), and surrogate relationship with patient. OPO representative variables included gender of the OPO representative who approached family and whether the OPO representative had received specialized additional training in approaching families (in addition to the basic training that all OPO coordinators receive). If authorization was attempted over the phone by a specialized trained OPO representative, we included data regarding whether a basic-trained OPO team member was present in-hospital in a supportive role. Additional variables relevant to the donation process included state of hospital, name of hospital, whether a collaborative donation process between OPO and the medical team was followed, whether an unplanned mention of donation by hospital staff with surrogate occurred (i.e., without OPO input), the main hospital staff involved with OPO in authorization discussion (OPO only, physician, registered nurse, etc.), and the date and time of approach with patient’s family―either day (7 am to 7 pm) or night. Importantly, all 14 hospitals in our dataset have a transplant center and neurosurgical services. Of note, NEDS defines a collaborative donation process as a joint plan having been discussed (either in one or several meetings), made, and followed by both NEDS and the medical team for introduction of possible donation to a surrogate decision maker. This case-by-case plan may mean having an agreement before an approach that only OPO representatives be involved in any given authorization discussions. Regardless of whether a collaborative donation process was established or followed, no discussions during the study period were held without the ultimate involvement of an OPO member. Data were all de-identified by NEDS before sharing with the Yale School of Medicine.

Outcome

Our outcome of interest was whether the NEDS approach for donation resulted in authorization obtained vs. authorization not obtained.

Statistical Analysis

Univariate and multivariate analyses were conducted with the database based on the outcome of interest to identify whether phone authorizations were correlated with increased donation rates and discover other independent predictors of increased donation rates. Eleven univariate predictors with p values of less than 0.1 were included in the multivariate analysis. All analyses were conducted in R (R Core Team; R Foundation for Statistical Computing, Vienna, Austria, 2014).

RESULTS

Surrogate, Patient, and OPO Representative Variables

Two thousand two hundred forty patients were included from the NEDS database. Table 1 summarizes patient, surrogate, and OPO representative characteristics. Patients were, on average, 44.1 years old, with a sd of 16.5. Patients were mostly male (61.5%) and White/non-Hispanic (70.5%). Most patients were unregistered potential donation after brain death donors (52.3%), followed by unregistered potential donation after circulatory death donors (26.1%), and registered circulatory death donors (21.5%). In most cases, an OPO representative with specialized training in authorization approach made the approach (86.2%). Most OPO representatives were female (70.9%).

TABLE 1.

Surrogate, Patient, and Organ Procurement Organization Representative Variables and Univariate Analysis

Variables	All Potential Donors (n = 2240)	Authorization Obtained (n = 1282)	Authorization Not Obtained (n = 958)	p
Patient age, mean (sd)	44.1 (16.5)	43.3 (16.2)	45.1 (16.9)	0.008
Patient gender, female	863 (38.5)	474 (37.0)	389 (40.6)	0.09
Patient race/ethnicity				< 0.001
White/non-Hispanic	1579 (70.5)	972 (75.8)	607 (63.3)
Black/non-Hispanic	257 (11.5)	106 (8.3)	151 (15.8)
Hispanic	303 (13.5)	179 (14.0)	124 (12.9)
Other	101 (4.5)	25 (2.0)	76 (7.9)
Donor type				< 0.001
Donation after death by neurological criteria unregistered	1173 (52.3)	616 (52.5)	557 (47.5)
DCD unregistered	585 (26.1)	289 (49.4)	296 (50.6)
DCD registered	482 (21.5)	377 (78.2)	105 (21.8)
Surrogate relationship (n = 2234)				< 0.001
Parent/guardian	881 (39.4)	559/1281 (43.6)	322/953 (33.8)
Spouse/significant other	607 (27.2)	307/1281 (24.0)	300/953 (31.5)
Children	422 (18.9)	230/1281 (18.0)	192/953 (20.1)
Healthcare proxy/other	324 (14.5)	185/1281 (14.4)	139/953 (14.6)
Female OPO representative approaching (n = 2117)	1500 (70.9)	904/1267 (71.3)	596/ 850 (70.1)	0.57
Specially trained OPO representative approaching (vs. basic trained)	1931 (86.2)	1169 (91.2)	762 (79.5)	< 0.001

DCD = donation after cardiac death, OPO = organ procurement organization.

All categorical variables are reported as n (%). Percentages were calculated among completed data fields where indicated.

Variables Relevant to Donation Process

With regards to variables relevant to the donation process (Supplemental Table 1, https://links.lww.com/CCX/B519), a collaborative donation process involving the OPO and medical team was followed in most cases (80.0%). The main hospital staff most likely to be involved in the authorization approach along with the OPO were a physician (19.4%) or registered nurse (12.7%), with 61.6% of approaches conducted by the OPO only. Most surrogate decision makers were approached during the day (76.9%).

Univariate Analyses by Authorization Success

Authorization was obtained in 1282 patients (57.2%) and not obtained in 958 patients (42.8%). OPO approaches by phone constituted 221 of 1282 (17%) of authorizations, as opposed to 134 of 958 (14%) of failed attempts (p = 0.04). Table 1 and Supplemental Table 1 (https://links.lww.com/CCX/B519) summarize additional univariate comparisons of successful vs. unsuccessful authorizations. Patients for whom authorization was obtained were more likely to be younger (43.3 vs. 45.1; p = 0.008), be White/non-Hispanic (75.8 vs. 63.3; p < 0.001), be previously registered potential donation after circulatory death donors (78.2 vs. 21.8; p < 0.001), and have parents as the key family surrogate (43.6 vs. 33.8; p < 0.001). Additionally, for successful authorizations, OPO representatives were more likely to have specialized training in authorization approach (91.2 vs. 79.5; p < 0.001). More ICU-OPO collaborative donation processes (85.4 vs. 72.7; p < 0.001) and OPO-only authorization discussions (68.1 vs. 52.6; p < 0.001), as well as fewer unplanned mentions of donation by ICU staff (12.3 vs. 22.3; p < 0.001), were seen in the donation process among authorizations. Of note, an association of state location was also noted with authorizations, but not for individual hospital location.

Multiple Logistic Regression

Table 2 summarizes the results of a multiple logistic regression that included variables possibly relevant to authorization. Variables in the univariate analyses that were included in the multivariate analyses included patient age, patient gender, patient race/ethnicity, donor type, surrogate relationship, specialized trained OPO representative approaching, whether collaborative donation processes were followed, unplanned mention of donation by hospital staff with surrogate, main hospital staff involved with OPO in authorization discussion, and state. Phone approach, patient age, unplanned hospital mentions, and state no longer had significant associations with authorization, but patient race/ethnicity, registered status for donors after circulatory death, surrogate relationship, OPO representative with specialized training in the donation approach, ICU-OPO collaborative donation processes, and OPO-only authorization discussions were all strong predictors.

TABLE 2.

Multivariate Logistic Regression Analysis of Characteristics Predicting Successful Approach Outcome

Variables	OR (95% CI)	p
Patient age	0.99 (0.99–1.01)	0.78
Patient gender, female	0.84 (0.68–1.00)	0.07
Patient race/ethnicity		< 0.001
White/non-Hispanic
Black/non-Hispanic	0.46 (0.33–0.60)
Hispanic	1.01 (0.73–1.26)
Other	0.28 (0.16–0.43)
Donor type		< 0.001
Donation after death by neurological criteria unregistered
DCD unregistered	0.82 (0.70–1.04)
DCD registered	3.11 (2.49–4.27)
Surrogate relationship		0.02
Spouse/significant other
Children	1.42 (1.06–1.84)
Parent/guardian	2.06 (1.48–2.65)
Healthcare proxy/other	1.28 (1.00–1.82)
Specially trained OPO representative approaching	2.40 (1.94–3.33)	< 0.001
Approach by phone	1.09 (0.80–1.31)	0.75
Collaborative donation process was followed	2.09 (1.49–3.94)	0.006
Unplanned mention of donation by hospital staff with surrogate	1.11 (0.51–1.43)	0.71
Main hospital staff involved with OPO in authorization discussion		< 0.001
OPO only
Physician	0.61 (0.42–0.78)
Registered nurse	0.68 (0.52–0.90)
Other	0.62 (0.42–0.90)
State		0.58
Massachusetts
Connecticut	1.38 (1.10–1.73)
Maine	1.34 (0.91–1.98)
Rhode Island	1.06 (0.74–1.53)
New Hampshire	1.25 (0.86–1.81)

DCD = donation after cardiac death, OPO = organ procurement organization, OR = odds ratio.

DISCUSSION

Contrary to our hypothesis, our study found that OPO approach for organ donation authorization by phone was not less effective than approach in-person. We found other variables to be independent predictors of authorization success, including patient race/ethnicity, prior registration for circulatory death donors, surrogate relationship, ICU-OPO collaborative donation processes, having OPO-only authorization discussions, and having an OPO representative with specialized training in donation approach.

Based on the strong significance of predictors we did identify, we surmise that authorization rates may depend less on how the OPO representative approaches the surrogate (i.e., in-person vs. remote) and likely more on factors such as the ICU-OPO collaboration and expertise of the OPO representative. Due to the increased demand for telehealth during COVID-19, our finding is important in indicating that the use of telehealth does not appear to hinder authorization success. While the majority of phone approaches in our dataset occurred during the pandemic (i.e., 2020–2021), we note that the in-person vs. phone rates of authorization success (916/1603, 57%; vs. 57/95, 60%) were similar in the period before the pandemic (i.e., 2017–2019) before telehealth had become more widespread. This observation strengthens the possibility that phone approaches may not be less effective than in-person approaches, even during nonpandemic periods when in-person approaches are more available.

A strong ICU-OPO collaborative donation plan is based on mutual understanding (together with the donor hospital medical team) of family dynamics and their understanding of prognosis and next steps. Steps for collaborative donation planning include having the OPO receive a timely notification of a potential referral from the donor ICU, conduct a clinical review to meet the family at their current level of understanding, create a collaborative plan of approach with the donor ICU medical team, and ensure the family has ample time to process the death of their loved one before having a specially trained OPO representative initiate the donation discussion. This plan of approach is based on research that has found that collaboration between OPOs and the ICU medical team in approaching families for donation tends to lead to best outcomes (9, 25, 26) and having specifically trained requesters leads to higher authorization rates (27–33). Our findings suggest that a collaborative donation process is not compromised by using phone as a medium for communication with families.

In addition to having collaborative donation processes and a specially trained OPO representative, we also found personal characteristics of the patient associated with authorization success. Patients who were White/non-Hispanic and registered potential circulatory death donors were more likely to have higher authorization success rates. Of these effects, the most significant one was race/ethnicity in line with prior literature (4, 14–16, 34).

A strength of our study is a relatively large multicenter dataset, with no missed approaches during the study period. However, as a retrospective and observational study, this study does have inherent limitations. We note that the dataset was not large enough to stratify the analysis meaningfully by individual hospital, to explore if there was wide variation among hospitals regarding percentages of authorizations that were completed by phone vs. in-person. We did not have accompanying qualitative data from decision makers about their perspectives regarding the authorization discussion in-person vs. by phone to understand further from this study whether there might be specific opportunities to optimize either type of experience. We also did not have data regarding whether surrogate decision makers themselves might have individually preferred an in-person vs. phone OPO approach, if they had somehow been able to consistently express those preferences up front.

CONCLUSIONS

We found that OPO approach for organ donation authorization by phone was not less effective than in-person approach. We did identify alternative strong independent predictors of authorization, including donor hospital ICU-OPO collaborative donation processes and having an OPO representative specifically trained in donation approach.

Our findings have practical implications, as there has been a greatly increased demand for organ transplant since COVID-19 (6–8) and an increased push for telehealth due to its cost-effectiveness and accessibility of use in pandemic times (24). Our study findings suggest that we should prioritize the strategy of an OPO approach, even if the OPO is not onsite or if the family is not available for an in-person meeting, as a donation discussion can be effectively led by phone by a specially trained OPO representative. To improve future authorization rates, we should continue to promote collaboration between ICU teams and OPOs; train OPO staff specifically for authorization discussions including those that may take place over the phone; and ensure fewer unplanned mentions of donation by ICU staff.

Supplementary Material

cc9-7-e1277-s001.pdf ^{(164.6KB, pdf)}