The Medical Costs of Determining Eligibility and Waiting for a Kidney Transplantation

Abstract

Background:

Recent efforts to increase access to kidney transplant (KTx) in the United States (US) include increasing referrals to transplant programs, leading to more pretransplant services. Transplant programs reconcile the costs of these services through the Organ Acquisition Cost Center (OACC).

Objective:

To determine the costs associated with pretransplant services by applying microeconomic methods to OACC costs reported by transplant hospitals.

Research Design, Subjects, and Measures:

For all US adult kidney transplant hospitals from 2013 through 2018 (n=193), we crosslinked the total OACC costs (at the hospital-fiscal year level) to proxy measures of volumes of pretransplant services. We used a multiple-output cost function, regressing total OACC costs against proxy measures for volumes of pretransplant services and adjusting for patient characteristics, to calculate the marginal cost of each pretransplant service.

Results:

Over 1,015 adult hospital-years, median OACC costs attributable to the pretransplant services were $5 million. Marginal costs for the pretransplant services were: initial transplant evaluation, $9k per waitlist addition; waitlist management, $2k per patient-year on the waitlist; deceased donor offer management, $1k per offer; living donor evaluation, procurement and follow-up: $26k per living donor. Longer time on dialysis among patients added to the waitlist was associated with higher OACC costs at the transplant hospital.

Conclusion:

To achieve the policy goals of more access to KTx, sufficient funding is needed to support the increase in volume of pretransplant services. Future studies should assess the relative value of each service and explore ways to enhance efficiency.

1. Introduction

Kidney transplantation (KTx) is the preferred treatment for end-stage renal disease (ESRD)¹. However, most patients with ESRD in the United States (US) are not treated with KTx. In order to receive a KTx, a patient needs to be 1) evaluated by a transplant hospital; 2) deemed an acceptable candidate, possibly after some medical testing; and 3) have an available and matching donor organ, either from a living donor (also evaluated by a transplant hospital), or from a deceased donor (which requires that the patient be added to a national waitlist and receive an organ offer according to national allocation rules). Recent policy efforts to improve access to KTx have taken one of two approaches: 1) increasing organ supply and 2) expanding access to KTx evaluation. For instance, information on referral for transplant will soon be part of routine reporting from dialysis units to Center of Medicare & Medicaid Services²; the percent of prevalent patients waitlisted for transplant has become a quality metric under the End-stage Renal Disease Quality Incentive Program³. Expanding access to KTx evaluation may not increase KTx numbers directly (unless many of the candidates who gain access to evaluations have viable living donors), but may provide more equitable access for patients from socioeconomically disadvantaged backgrounds, who currently suffer from diminished access to KTx⁴.

Expanding access to the KTx evaluation entails a large increase in the quantity of pretransplant services delivered to patients with ESRD. We define pretransplant services as services that only serve the purpose of KTx and are not a part of a patient’s regular healthcare. They include: in-person and remote services from transplant physicians, nurse coordinators and social workers; diagnostic testing including histocompatibility testing; procedures including, for instance, revascularization of an asymptomatic coronary stenosis to mitigate peritransplant risks; and various behind-the-scene operations including maintaining up-to-date on the patient’s health status, processing deceased organ offers, and work-up of, donor nephrectomy for, and post-donation care of living donors. Transplant hospitals deliver most of these services. Due to the shortage of transplantable kidneys, many patients on the KTx waitlist die or are removed before a kidney becomes available⁵. These patients incur pretransplant costs, but accrue no benefit, of KTx.

Expanding access to the KTx evaluation will therefore lead to a large increase in the pretransplant costs of KTx. Transplant hospitals will incur most of these costs and pass them on to Medicare and private payers. Medicare, the largest provider of ESRD services in the US, reimburses transplant hospitals for pretransplant costs via the Organ Acquisition Cost Center (OACC) mechanism⁶. Each fiscal year, transplant hospitals report allowable costs attributable to pretransplant care and organ procurement as OACC on the Medicare Cost Report, and Medicare reimburses the transplant hospital a portion of the OACC cost based on the Medicare ratio. The national mean Medicare ratio is 0.72, meaning that Medicare shoulders about 72% of the total OACC costs in the US⁷ while the rest are shouldered by private payers and other state and federal programs, e.g. Medicaid. Median OACC costs per KTx increased from $81k in 2012 to $100k in 2017⁵. Of the median OACC costs per KTx, about 36% was attributable to the costs of deceased organs; the rest were attributable to the other pretransplant services described above⁷.

Key to forecasting the fiscal implications of expanding access to the KTx evaluation is understanding the per-unit costs of the pretransplant services delivered, in order to develop policy projections and strategies to manage costs such that policies aimed at improving KTx access and enhancing KTx equitability are financially plausible and sustainable. In this study, we used microeconomic methods to make inferences about costs using publicly available data from Medicare Cost Reports.

2. Methods

2.1. General Approach and Cost Function

We used a multiple-output cost function, as pioneered by Granneman et al.⁸ in the realm of healthcare. The dependent variable was the total cost attributable to delivering pretransplant services at the hospital-year level; independent variables were surrogate measures to estimate the quantity of different types of pretransplant services. We then generated marginal cost estimates of each pretransplant service from the outputs of the cost function using standard regression methods. This approach made no assumption about how costs are allocated within the KTx program’s accounting structure, since only the total cost was included. We previously used this approach to estimate the costs of dialysis and hospital services for Medicare beneficiaries^9,10 and the costs of procuring organs at an organ procurement organization level¹¹. Figure 1 illustrates the cost function framework as applied to the pretransplant setting.

Figure 1.

The structure of organ acquisition cost center (OACC) and types of services it covers (white box: facility; grey box: candidate; black box: donors). Our cost function will estimate the total OACC cost minus deceased donor organ procurement costs (blue bracket) as a function of the quantity of services (red bracket). The result will be an estimate on the marginal costs of these services (red bracket).

2.2. Data Source

We created our analytic dataset (2013–2018, all non-pediatric transplant hospitals in the US) using multiple data sources and linkage variables at the transplant hospital level. The sources are:

1. Hospital Cost Reports (CMS-2552) which include total OACC costs (worksheet D-4), by transplant hospital by year, obtained from the National Bureau of Economic Research; while the Medicare Cost Report reports the total OACC at each transplant hospital, there is no direct way of obtaining the cost of each pretransplant service (see Figure, Supplemental Digital Content 1, for sample cost report and line items).

2. Scientific Registry of Transplant Recipients (SRTR)¹², which enabled us to create, at the transplant hospital level, proxy measures for transplant hospital processes (e.g., number of waitlist additions at each hospital each year) and case-mix indicators (e.g., percent of incident waitlist patients with diabetes mellitus);

3. a previously described organ procurement organization (OPO) cost database which included the standard acquisition charge, or price of deceased donor kidneys at each of the 51 independent OPOs that is paid by transplant hospitals to OPOs every time the transplant hospital accepts a deceased donor kidney from the OPO¹³;

4. publicly available CMS hospital wage index, which reflected the costliness of the occupational mix in a hospital relative to the national average¹⁴.

2.3. Independent Variables and Covariates

From SRTR data (data source #2), we created independent variables acting as proxy measures for quantities of pretransplant services, including: 1) number of waitlist additions (proxy for the quantity of transplant evaluations); 2) number of patient-years on the waitlist (proxy for the quantity of surveillance on the waitlist); 3) number of deceased donor kidney offers (proxy for the quantity of management of deceased donor kidney offers); and 4) number of living donations (proxy for the quantity of living donor evaluation and procurement). Table 1 outlines and describes these services in detail.

Table 1.

Surrogates for quantity of transplant hospital pretransplant services examined in this study that are paid for by the OACC. KTx: kidney transplant. OACC: organ acquisition cost center. OPO: organ procurement organization.

Type of Pretransplant Service	Description of Pretransplant Service	Surrogate Measure	Description of Surrogate Measure
Initial evaluation	• Services rendered and medical testing to determine if a patient, referred to a transplant hospital, is a KTx candidate.  • If a candidate, the patient is added to the waitlist.  • If not a candidate, the patient is declined but the cost is absorbed into the OACC.	Number of new additions to the waitlist	The number of patients added to each transplant hospital’s KTx waitlist during each fiscal year.
Waitlist management	• Care coordination efforts by transplant hospital staff to monitor the medical status of each KTx waitlist candidate on their waitlist. •  Some medical testing required to keep the candidate ready for transplant (e.g. histocompatibility testing).	Number of waitlist patient-years	The number of patient-years on each transplant hospital’s KTx waitlist during each fiscal year.
Managing deceased donor offers	• Care coordination efforts by transplant hospital staff to maintain communication with OPOs and respond to deceased donor kidney offers in real time. •  If offer is accepted: Pre-operative medical examination if the candidate does not receive the KTx as intended*.	Number of deceased donor kidney offers	The number of deceased donor kidney offers each transplant hospital receives during each fiscal year.
Living donor workup and procurement	• Services rendered and medical testing to determine if an individual is a viable living donor candidate.  • If a candidate, the individual may proceed with donation.  • If not a candidate, the patient is declined but the cost is absorbed into the OACC. •  Donor nephrectomy and hospitalization. •  Costs of donation follow-up, including complications, are sometimes included in the OACC and sometimes not15.	Number of living donors	The number of living donations at each transplant hospital during each fiscal year.

^*If the candidate receives the KTx as intended, the costs are subsumed under the KTx episode under the Diagnosis Related Group prospective payment structure, rather than the OACC.

From SRTR’s cand_kipa file which contains all information on transplant candidates, we created covariates to indicate the case-mix from incidence waitlist patients at each hospital-year: median age, % sex, % racial minority (defined as non-White), insurance status (“underinsured” vs not, the former defined as not having insurance from private entities, Medicare, or the Department of Veterans Affairs), median dialysis duration, and % diabetes mellitus. We used the number of hospital beds to indicate the capital at each hospital-year, obtained from Hospital Cost Reports (data source #1).

2.4. Dependent Variable

The total OACC cost at each transplant hospital-year includes three components: costs attributable to pretransplant services (item #1), costs attributable to living donor kidney procurement (item #2), and costs attributable to deceased donor kidney procurement (item #3, Figure 1). We wanted to specifically focus on items #1 and #2, as the specifics of #3 are already known from prior work¹³. As a result, we estimated the total OACC costs attributable to #1 and #2 by subtracting the cost of deceased donor organ acquisition (#3) from the total OACC costs, and used the result as the dependent variable for the cost function. Costs attributable to deceased donor organ acquisition (#3) included:

1. Costs attributable to procurement of deceased donor kidneys at the transplant hospital that are reimbursed by the OPOs, indirectly estimated by the revenue for organs sold (Worksheet D-4, Line 66, Column 1). The median was $149k in our study years.

2. Costs attributable to paying OPOs for deceased donor kidneys, estimated by crosslinking the OPO providing each deceased donor kidney to our OPO cost database¹³ (which contains the cost per kidney at each OPO-year level). For instance, if a transplant hospital utilized 10 kidneys from OPO A (where cost per kidney is $35k) and 10 kidneys from OPO B (where cost per kidney is $45k) for deceased donor transplantation in 2018, then the transplant hospital’s total costs attributable to paying OPOs for deceased donor kidneys would be $800,000 in 2018.

2.5. Analysis

We used generalized linear equations. We log transformed the dependent variable (OACC minus deceased donor procurement costs) for analysis. We analyzed quantities of pretransplant services (e.g., number of waitlist additions at each hospital each year) in the log-linear and log-log forms; the results were very similar, so we opted for the log-log form for ease of interpretability. We used the following model:

$$ln{C}_{(h,l)t}=\alpha +{\sum }_{i=1}^4{\beta }_{i}ln{Y}_i+{\sum }_{j=1}^2{\delta }_{j}H{C}_j+{\sum }_{k=1}^6{\gamma }_{k}P{C}_k+{\tau }_t+r_l+{ϵ}_{ht}$$

where $C_{(h,l)t}$ is the cost of program at hth transplant hospital in year $t;Y_{i=\mathrm{1,2},\mathrm{3,4}}$ corresponds to four types of pretransplant services; $H{C}_{j=\mathrm{1,2}}$ are hospital characteristics (Table 2); and $P{C}_{k=\mathrm{1,2},\mathrm{3,4},\mathrm{5,6}}$ represent patient characteristics summarized at the hospital-year level (case-mix, see Table 2). To account for unobserved heterogeneity across time and location, we also include year fixed effect $\tau$ and county- or state- fixed effect $r$. Standard errors are clustered at the locality level, $l$ (county vs state). We used resultant coefficients from each model to calculate the marginal cost associated with each type of pretransplant service and incremental change in cost associated with patient characteristics (case-mix), simplified as $M{C}_i={\beta }_i·C$, where $M{C}_i$ represents marginal cost for service $i$ and $C$ represents the total OACC cost (see Supplemental Digital Content 2, for details).

Table 2.

Baseline characteristics at each hospital-year, 2013–2018. A total of 193 adult kidney transplant hospitals are represented.

Variable, Unit	Median (1st Quartile – 3rd Quartile)	Mean (Standard Deviation)	Minimum – Maximum
Total OACC Costs (minus deceased donor organ procurement costs), $	5,085,000 (2,866,000–8,092,000)	6,303,000 (4,843,000)	432,000–30,700,000
Surrogate Measures of Pretransplant Services
New waitlist additions, person	131 (78–241)	188 (163)	3–1044
Waitlist patient-years, person-year	349 (197–670)	550 (652)	0.5–5197
Number of deceased donor organ offers, offer	742 (430–1094)	831 (526)	44–3210
Number of deceased donor kidney transplants, transplant	48 (25–86)	86 (52)	0–319
Number of living donor nephrectomies, donor	18 (8–36)	29 (34)	1–261
Covariates: Transplant Hospital Characteristics
Hospital beds, bed	425 (324–568)	476 (249)	100–2155
Operating wage index	0.98 (0.92–1.12)	1.03 (0.18)	0.42–1.74
Covariates: Patient Characteristics
Median age in patients added to the waitlist, year	54 (52–56)	54 (3)	41–64
% of patients added to the waitlist with diabetes mellitus, %	42 (37–47)	42 (8)	0–86
Median dialysis vintage in patients added to the waitlist, year	1 (0–1)	0.7 (0.6)	0–4
% of patients added to the waitlist who are female, %	38 (35–41)	38 (5)	14–63
% of patients added to the waitlist who belong to a racial minority, %	52 (33–71)	52 (22)	0–100
% of patients added to the waitlist who are underinsured, %	7 (3–13)	10 (10)	0–79

To evaluate if economies of scale (EOS) exist, i.e. whether there are cost advantages at larger transplant hospitals relative to smaller transplant hospitals, we calculated the Ray EOS by $\mathrm{E}\mathrm{O}\mathrm{S}=\mathrm{C}\left(\mathrm{Y}\right)/{\sum }_{i}M{C}_i{Y}_i$.¹⁵, where $C$ represents the total OACC cost at the transplant hospital, $M{C}_i$ the marginal cost for service $i$, and $Y_i$ the unit count of the surrogate for service $i$. Ray EOS is an index measure of scale economies (increased efficiency) that may occur as the size of the firm (program) increases, holding the mix of products (services) constant. Computationally, it is calculated by dividing total costs by the weighted sum of the marginal costs. In the multi-service settings, EOS exist when increasing all pretransplant services proportionally leads to a decline of the average cost of providing all services. We used a gradient search long transplant hospital size to assess whether larger transplant hospitals had lower marginal costs as further evidence of the presence of EOS (see Supplemental Digital Content 2, for details).

We used bootstrapping with 10,000 iterations to obtain confidence intervals that are non-linear combinations of regression coefficients, including marginal costs and Ray EOS.

We conducted statistical analyses using SAS 9.4 (Cary, NC) and STATA 17 (College Station, TX). The data reported here have been supplied by the Minneapolis Medical Research Foundation as the contractor for the SRTR. The interpretation and reporting of these data are the responsibility of the authors and in no way should be seen as an official policy of or interpretation by the SRTR or the US government.

3. Results

Our dataset consisted of 1,015 adult transplant hospital-years in the US from 2013 through 2018, representing 193 distinct transplant hospitals. Table 2 illustrates their baseline characteristics.

Figure 2A shows the estimated marginal cost of each pretransplant service in both the county-level and the state-level fixed effects models (see Table, Supplemental Digital Content 3, for coefficients estimated from the generalized linear regression model). County-level and state-level fixed effects models generated estimates that were directionally consistent and within 25% of each other. Figure 2B shows the estimated increase in OACC expenditure by a median-sized transplant hospital, if it were to increase its service volume by 100%. Although the per-unit marginal cost is highest for living donor management (Figure 2A), the overall cost increase at the transplant hospital level is the highest for initial transplant evaluation and waitlist management (Figure 2B).

Figure 2.

Marginal cost of each kind of service in United States dollars (panel A), and the overall increase in OACC if a median-size transplant hospital were to increase their service volume by 100% (panel B). * Significant to p<0.05. ** Significant to p<0.01.

Table 3 shows the associations of patient characteristics, or case-mix, with the cost of pretransplant services. Of the comorbidities, a longer median time on dialysis among patients added to the waitlist was associated with higher cost OACC costs at the transplant hospital ($381k increase per one-year increase in median dialysis vintage) in the state fixed-effect model. An increase in the proportion of underinsured patients added to the waitlist was associated with lower OACC costs at the transplant hospital ($251k-$300k decrease per 5-percentage point increase in the percentage of underinsured patients). An increase in the proportion of non-white patients added to the waitlist was associated with an higher OACC costs at the transplant hospital ($124k increase per 5-percentage point increase in the percentage of non-white patients) in the state fixed-effect model.

Table 3.

Estimates on the effect of case-mix (i.e. patient characteristics at each transplant hospital) on total organ acquisition cost center (OACC) costs minus deceased donor organ procurement. Results in black indicate statistical significance (p<0.05). Results in grey indicate not reaching statistical significance. For reference, median total OACC at transplant hospitals is $5,000,000.

Case-Mix unit	Effect on Total OACC Costs ($) [95% confidence interval]
	County Fixed Effect Model	State Fixed Effect Model
Median age in patients added to the waitlist  per 1 year increase	−9,603 [−71,372, 52,166]	−7,571 [−106,402, 91,260]
% of patients added to the waitlist with diabetes mellitus  per 5% increase	85,270 [−29,870, 204,880]	84,015 [−101,290, 269,320]
Median dialysis vintage in patients added to the waitlist  per 1 year increase	284,584 [−48,970, 618,138]	380,869 [10,252, 751,485]
% of patients added to the waitlist who are female  per 5% increase	7,295 [−116,245, 130,835]	50,955 [−70,780, 174,490]
% of patients added to the waitlist who belong to a racial minority  per 5% increase	18,245 [−105,295, 141,780]	123,660 [120, 247,195]
% of patients added to the waitlist who are underinsured  per 5% increase	−299,875 [−301,470, -298,280]	−251,010 [−498,085, -3,930]

The Ray EOS had an index value of 1.3 (p<0.001), meaning that an 10% increase in volume was associated with a 3% increase in overall costs at the transplant hospital level, i.e. there were economies of scale. Our gradient search showed that the Ray EOS estimate remained stable, even when the largest 10–20% programs and smallest 10–20% programs (i.e. outlier programs) were alternatively excluded (Supplemental Digital Content 2).

4. Discussion

In this paper, we provide estimates on the costs of pretransplant services using data from all US adult transplant hospitals between 2013 and 2018, and provide insights into patient characteristics associated with higher or lower costs at the transplant hospital level. Payment for routine ESRD care (including provision of dialysis) does not include these services. Whenever a patient is considered for KTx these services are performed, whether or not the patient actually receives a KTx. As such, the costs of these services warrant special attention. Our findings of economies of scale suggests that the cost increases may be ameliorated somewhat (for instance, a 10% increase in evaluations will only result in a 3% increase in total costs). However, the overall magnitude of the cost increases is large enough such that opportunities to reduce inefficiencies to reduce or at least contain these costs need to be explored.

Initial transplant evaluation costs around $9k per new waitlist addition. This figure includes the Organ Procurement Transplant Network registration fee¹⁶ and medical evaluation and professional service fees incurred during the candidate evaluation. The per-item costs of waitlist management and deceased donor organ offer management are lower (Figure 2A). However, the sheer volume of waitlist patient-years and deceased donor offers requiring management at each transplant hospital (median 349 patient-years and 742 offers per fiscal year, respectively) translates into an exceptional high overall cost of maintaining a deceased donor kidney waitlist and the organ offers that arise from the deceased donor allocation system (Figure 2B). We anticipate that these evaluation and waitlist management costs will increase rapidly with an expansion of KTx waitlisting practices. Our results directly enable cost estimates of any policy measure to expand access to the KTx evaluation.

Living organ donor evaluation and procurement cost roughly $25k. This figure refers to only what is included in the OACC and underestimates of the true costs of living donation. Transplant hospitals frequently need to recoup the costs of living donor follow-up from sources other than OACC, including billing claims to recipient’s insurance or other sources of hospital funds¹⁷. Living donors also incur considerable out-of-pocket costs in the form of lost wages, travel, and dependent care ranging between $900 to $20k US dollars¹⁸. Our figure therefore serves only as a lower bound on the transplant program costs of living donation.

Of the patient characteristics, an increase in the median time on dialysis in patients added to the waitlist was associated with markedly higher total OACC costs at the transplant hospital level. Two explanations exist. Higher median time on dialysis may denote higher medical complexity among patients added to the waitlist; more complex patients require more resources to manage. Alternatively, as time on dialysis is a key factor in determining allocation priority, some transplant programs defer accepting patients to their waitlist until the patients have attained a certain dialysis vintage/allocation priority, and as such may be performing more services to prepare these patients for imminent transplantation during the evaluation phase of care. Such operational details are not publicly disclosed or mandated reporting, but they likely exert profound effects on transplant hospital finances as well as outcomes and warrant further investigation.

As referral patterns change, the mix of patients being evaluated and managed by transplant programs will also change with accompanying cost implications. Budgetary decisions should account for the increased needs of historically underrepresented patient subsets, to ensure that these patients receive equitable access to transplantation. For example, our data showed that as transplant programs expand waitlist access to more patients, including non-white patients or patients who carry a larger comorbidity burden such as longer dialysis duration, the pretransplant costs per transplant will rise. Removing barriers to transplantation for disadvantaged patients will therefore necessitate budgetary expansion at the transplant hospital level and/or practice innovations to reduce inefficiencies and reduce costs.

We found lower costs among transplant programs with a higher proportion of underinsured patients added to their waitlist. While counterintuitive, this finding makes sense in the context of transplant program financing. Transplant programs recoup the costs of pretransplant services from two main sources: 1) Medicare, which reimburses a majority of OACC retrospectively based on the Medicare ratio; and 2) non-Medicare payers, which pay a prospectively set rate for deceased donor organ acquisition and fee-per-service for evaluation services. Our definition of underinsured includes mostly non-Medicare-insured patients reliant on state Medicaid. Reimbursements for pretransplant services under state Medicaid are generally low. Transplant programs therefore likely cherry-pick only the healthiest underinsured candidates for their waitlist. Indeed, prior studies have shown that underinsured KTx recipients tend to be younger and healthier^19,20, factors associated with lower costs. The favorable posttransplant outcomes previously reported in this population, along with our finding of lower costs, suggest that, under current selection practices, expansion of KTx coverage for underinsured patients may be fiscally viable for many state Medicaid programs.

The unique financing of the OACC has implications for efforts to expand KTx evaluation access. Because Medicare shoulders the bulk of OACC costs at most transplant hospitals, and because Medicare payments are retrospective and at-cost, the downward pressure on costs is weaker. On the other hand, the same retrospective and at-cost reimbursement structure also allows for no margin, and it could be difficult to convince hospital administrators to grant more resources, personnel and capita, to KTx programs. Prior costing studies have shown that most KTx programs survive on thin margins or require subsidies from their associated health systems^21,22. In the short term, faced with many KTx referrals, KTx programs may resort to diverting resources from other areas (such as deceased donor offer management) to handle KTx evaluations. Further research is thus urgently needed to identify the services of highest value and to evaluate strategies to decrease cost and enhance efficiency.

A limitation of our study is we do not have direct quantities of services produced, only proxies with their inherent limitations. The number of waitlist additions as a proxy for evaluations does not account for the patients who were evaluated but not added to the waitlist. The number of living donations as a proxy for living donor evaluations does not account for donor candidates who were evaluated but did not progress to living donation. While residual interference from collinearity of proxies is a concern, we were able to obtain fairly consistent estimates using multiple model specifications. Furthermore, because we rely on administrative data, data fidelity and accuracy are areas of potential concern. These are partly mitigated by strong incentives for transplant hospitals to provide accurate figures of OACC for which they receive Medicare payments. Another limitation is that OACC understimates the complete cost of pretransplant work-up and that transplant hospitals finance pretransplant evaluation differently, hence the degree of underestimation varies by hospital. Consider, for instance, the case of a potential KTx recipient with no cardiac symptoms who undergoes a screening test for ischemic heart disease as a part of KTx evaluation, as is standard practice²³ and supported by society guidelines²⁴. The costs of the initial stress test may be assigned to the OACC or may be billed to the patient’s usual insurer, depending on the particular hospital. Any ensuing follow-up care, including repeat testing, coronary angiography, revascularization, and treatment of any complications that might arise, falls within the standard-of-care for a patient and are billed to the patient’s usual insurer. These are the incremental and largely hidden costs of pretransplant services that warrant consideration in any economic analyses of transplantation.

In summary, we provide estimates on the costs of pretransplant services delivered by transplant hospitals in the US and identify patient characteristics associated with higher and lower costs. Such work can inform future payment reforms and restructuring, as well as highlights the need for additional research to identify the cost-effectiveness of the evaluation components with respect to improving equitable access and transplant outcomes.

Data Sharing Statement:

The data used for this study are made available through a data use agreement with the Scientific Registry of Transplant Recipients and therefore no available for sharing.