Impact of obesity and income on the likelihood of kidney transplantation: A retrospective cohort study

Abstract

BACKGROUND

This study evaluated the joint and modifying effects of obesity and income on access to kidney transplantation, stratified by race/ethnicity.

AIM

To investigate whether income modifies the association between body mass index (BMI) and the likelihood of deceased donor kidney transplantation (DDKT) and living donor kidney transplantation (LDKT).

METHODS

We conducted a retrospective cohort study of 54891 adult kidney-only candidates added to the United States waiting list in 2020 and 2022, using data from the Organ Procurement and Transplantation Network and LexisNexis.

RESULTS

Overall, 47% of candidates were obese (BMI ≥ 30 kg/m²). In multivariable models, BMI 30- < 35 was associated with a lower likelihood of DDKT for non-Hispanic White and non-Hispanic Black candidates [sub-distribution hazard ratio (sHR): 0.93 for both]. BMI ≥ 35 was associated with an 18% lower likelihood to DDKT for non-Hispanic White candidates only. Low income was associated with higher likelihood of DDKT among non-Hispanic White (sHR: 1.34) and Hispanic/Latino (sHR: 1.20) candidates but was not associated with DDKT among non-Hispanic Black candidates. In interaction models, income-related differences in DDKT were most consistently observed among nonobese nonHispanic White and Hispanic/Latino candidates. For LDKT, both BMI ≥ 35 and low income were associated with lower likelihood of transplantation across all racial and ethnic groups; low income was also associated with reduced likelihood of LDKT among non-obese candidates in interaction models.

CONCLUSION

Obesity and income are differentially associated with access to transplantation by racial/ethnicity and transplant modality, highlighting the importance of jointly considering clinical and social risk factors in transplantation.

Key Words: Kidney transplantation; Obesity; Body mass index; Income; Social determinants of health; Organ Procurement and Transplantation Network

Core Tip: This study examines how obesity, measured by body mass index (BMI), and individual income jointly influence access to deceased donor kidney transplantation and living donor kidney transplantation among adults listed in the United States. By stratifying analyses by race and ethnicity, we identify distinct patterns in how BMI and income affect transplant likelihood. These findings highlight the need to consider both clinical and social risk factors when designing strategies to improve equity in kidney transplant access.

INTRODUCTION

The prevalence of obesity in the United States has risen substantially in recent decades, with current estimates indicating that 42.4% of United States adults meet criteria for obesity, defined as a body mass index (BMI) ≥ 30 kg/m²[1]. Obesity is not only a growing public health concern in the general population, but is also disproportionately common among patients with end-stage kidney disease (ESKD), in part because obesity contributes to the development and progression of chronic kidney disease[2,3].

While obesity has been associated with inferior post-transplant outcomes compared with normal weight, studies have consistently demonstrated that both obese and non-obese patients derive a substantial survival benefit from kidney transplantation relative to remaining on dialysis[4]. Moreover, improvements in surgical techniques and perioperative care, including the use of robotic-assisted kidney transplantation, have contributed to better post-transplant outcomes for obese recipients in recent years[5,6]. Despite these advances, however, obese transplant candidates continue to face considerable barriers to transplantation[7].

In addition to clinical factors, such as obesity, social determinants of health (SDOH) may play a role in access to transplant. Adverse SDOH may be more prevalent among obese individuals[8-10], raising concerns that obesity may compound the effects of socioeconomic disadvantage in limiting access to transplantation. Importantly, both obesity and adverse SDOH are more prevalent among racial and ethnic minority groups, who also face additional systemic barriers to transplantation. These intersecting factors may result in compounded disadvantages for minority candidates. Given these disparities, the impact of obesity and SDOH on transplant access may differ by race and ethnicity.

Despite growing recognition of the importance of both clinical and social risk factors in transplantation, limited research has explored the interplay between obesity and SDOH – particularly income – in determining likelihood of transplantation. This study aims to examine the interplay between obesity and income on the likelihood of transplantation, with particular attention to racial and ethnic differences. Specifically, we investigate whether income modifies the relationship between obesity and likelihood of deceased donor kidney transplantation (DDKT) and living donor kidney transplantation (LDKT) and determine the extent to which obesity may explain the relationship between income and likelihood of transplantation. A better understanding of these relationships may help inform strategies to reduce disparities and improve equity in access to kidney transplantation.

MATERIALS AND METHODS

Study design, participants, and data sources

We conducted a retrospective cohort study of adult (≥ 18 years at transplantation) kidney-only candidates who were added to the United States waiting list in 2020 and 2022. This study used data from the Organ Procurement and Transplantation Network (OPTN). The OPTN data system includes data on all donor, wait-listed candidates, and transplant recipients in the United States, submitted by the members of the OPTN. The Health Resources and Services Administration, United States Department of Health and Human Services provides oversight to the activities of the OPTN contractor. Estimated individual annual income was obtained from LexisNexis, a commercial data source that uses a proprietary algorithm to combine data from a combination of sources, including real estate/tax assessor records, mortgage records, motor vehicle registrations, driver’s license records, court filings (including bankruptcy federal and state tax liens), criminal history records and voter registrations[11]. As previously described[12,13], OPTN data, including patient name, social security number, date of birth, age, sex, and home residence zip code were securely transmitted to LexisNexis and used to merge OPTN data with data from LexisNexis. The social vulnerability index was obtained from the Centers for Disease Control and Prevention and Agency for Toxic Substances and Disease Registry[14] and linked at the census tract level. Kidney transplant candidates were included if they were United States citizens or United States residents and had a valid social security number.

Study variables

The primary outcomes were the cumulative incidence of DDKT and LDKT censored at 3 years. The primary exposure variables of interest were BMI and estimated individual annual income. BMI was categorized using the United States Centers for Disease Control and Prevention (CDC) classification: 18.5-30 kg/m² (non-obese), ≥ 30- < 35 kg/m² (class 1 obesity), ≥ 35 kg/m² (class 2 or 3 obesity)[15]. Estimated individual annual income (rounded to the nearest $1000) was arbitrarily categorized into tertiles: Low (≤ $43000), middle (> $43000-$76000), and high (> $76000). Estimated individual annual income was categorized into tertiles based on the distribution within the analytic cohort to facilitate interpretability and relative comparisons across income strata, rather than to represent externally generalizable income thresholds. Patient characteristics, including age, birth sex, race/ethnicity, blood type, calculated panel reactive antibody at listing, primary cause of ESKD (diabetes, glomerular disease, hypertensive nephrosclerosis, and other), and highest level of education, were obtained from OPTN data. Race/ethnicity was categorized as non-Hispanic White, non-Hispanic Black, and Hispanic/Latino. Non-Hispanic Other racial groups, including non-Hispanic Asian/Asian-American, non-Hispanic American Indian/Alaska Native, non-Hispanic Multiracial, and non-Hispanic Native Hawaiian/Other Pacific Islander, were excluded from the analysis, due to small samples size (Supplementary Figure 1).

Statistical analysis

Comparisons between groups were performed using the χ² test and Fisher’s exact test for categorical variables and the Kruskal-Wallis test for non-normally distributed continuous data. To account for competing events (DDKT, LDKT, death, and removal for other reasons), we used competing risks regression to evaluate the association of obesity and income with the likelihood of DDKT and LDKT. We used the Gray’s test to compare the unadjusted 3-year cumulative incidence of DDKT and LDKT. Multivariable Fine-Gray subdistribution hazard models were constructed separately for DDKT and LDKT. For the DDKT models, LDKT, death, and removal for other reasons were treated as competing risks; conversely, for the LDKT models, DDKT, death, and removal for other reasons were considered competing events. In a sensitivity analysis designed to explore the impact of censoring LDKT on DDKT estimates, we estimated the likelihood of DDKT by censoring LDKT while continuing to treat death and removal for other reasons as competing events. Covariates included in multivariable models were selected a priori based on their known or plausible associations with access to transplantation. To determine the extent to which BMI may explain the relationship between income and likelihood of transplantation, we performed multivariable competing risks models with and without adjustment for BMI and compared the total and direct effects of income on the likelihood of transplantation. To examine whether income modified the effect of obesity on the likelihood of DDKT and LDKT, we examined the interaction between income and obesity subgroups using a likelihood ratio test that compared the multivariable model with vs without the interaction terms. All analyses were stratified by race/ethnicity. We tested for inter-center variation in likelihood of transplant by including transplant center as a random effect in multivariable models. We used a casewise deletion strategy for individuals missing any covariates included in the analysis. Details on missing data are provided in Supplementary Figure 1. All analyses were conducted using R version 4.4.0.

RESULTS

Baseline characteristics

A total of 25805 (47%) of candidates were obese [n = 15364 (28%) BMI ≥ 30- < 35; n = 10441 (19%) BMI ≥ 35; Table 1]. Candidates with BMI ≥ 35 were more likely to be younger than candidates with lower BMI. The proportion of male candidates varied with increasing BMI (< 30, ≥ 30, ≥ 35: 62%, 64%, 59%). The proportion of non-Hispanic Black candidates increased with increasing BMI (33%, 37%, 40%). The median estimated individual annual income decreased with increasing BMI [$66000, interquartile range (IQR): $40000-$86000, $63000, IQR: $39000-$82000, and $61000 IQR: $38000-$78000].

Table 1 Baseline characteristics of 54891 kidney transplant candidates listed in 2020 and 2022, by body mass index subgroup, n (%).

	Body mass index (kg/m2)
Candidate characteristics, median (IQR)	18.5- < 30	30- < 35	≥ 35
	(n = 29086)	(n = 15364)	(n = 10441)
Age at registration (years)	56 (44, 65)	57 (47, 64)	53 (44, 62)
Birth sex
Female	10916 (38)	5501 (36)	4302 (41)
Male	18170 (62)	9863 (64)	6139 (59)
Race/ethnicity
Non-Hispanic White	14547 (50)	7182 (47)	4746 (45)
Non-Hispanic Black	9667 (33)	5624 (37)	4225 (40)
Hispanic/Latino	4872 (17)	2558 (17)	1470 (14)
Blood type
A	9605 (33)	5114 (33)	3511 (34)
B	4014 (14)	2151 (14)	1461 (14)
AB	1082 (4)	568 (4)	403 (4)
O	14385 (49)	7531 (49)	5066 (49)
CPRA at listing
0	24711 (85)	13089 (85)	8993 (86)
1-80	3091 (11)	1739 (11)	1112 (11)
81-94	534 (2)	216 (1)	148 (1)
95-100	750 (3)	320 (2)	188 (2)
Kidney diagnosis at listing
Diabetes	8283 (28)	6699 (44)	4599 (44)
Glomerular disease	5470 (19)	2201 (14)	1582 (15)
Hypertensive nephrosclerosis	5722 (20)	2962 (19)	2081 (20)
Other	9611 (33)	3502 (23)	2179 (21)
Peripheral vascular disease
No	25142 (86)	12843 (84)	8759 (84)
Yes	3717 (13)	2420 (16)	1611 (15)
Unknown	227 (1)	101 (1)	71 (1)
History of malignancy
No	26899 (92)	14357 (93)	9939 (95)
Yes	2187 (8)	1007 (7)	502 (5)
Prior kidney transplant
No	26999 (93)	14709 (96)	10118 (97)
Yes	2087 (7)	655 (4)	323 (3)
Distance from listing center to residence (miles)	29 (11, 86)	33 (12, 91)	33 (11, 87)
Education
High school or less	11081 (38)	6069 (40)	3963 (38)
College/technical school	7905 (27)	4381 (29)	3095 (30)
College or more	9470 (33)	4587 (30)	3146 (30)
Unknown	630 (2)	327 (2)	237 (2)
Estimated individual annual income	66000 (40000, 86000)	63000 (39000, 82000)	61000 (38000, 78000)
Estimated individual annual income
Low (≤ $45000)	9828 (34)	5610 (37)	4195 (40)
Medium (> $45000-$77000)	9270 (32)	5169 (34)	3595 (34)
High (> $77000)	9988 (34)	4585 (30)	2651 (25)
Social vulnerability index	54 (27, 78)	57 (30, 80)	56 (30, 79)
Year of listing
2020	13745 (47)	7246 (47)	4686 (45)
2022	15341 (53)	8118 (53)	5755 (55)

IQR: Interquartile range.

Association of obesity and income with the likelihood of DDKT by race/ethnicity

The unadjusted 3-year cumulative incidence of DDKT was highest for non-Hispanic Black candidates across all BMI categories (< 30, ≥ 30, ≥ 35: 46.6%, 45.1%, 44.7%) and decreased with increasing BMI for non-Hispanic White (36.4%, 34.5%, 31.3%) and Hispanic/Latino candidates (35.8%, 31.7%, 29.7%; Figure 1A). The unadjusted 3-year cumulative incidence of DDKT was highest for non-Hispanic Black candidates across all income groups (low, middle, high income: 46.1%, 45.6%, 45.2%) and decreased with increasing income for non-Hispanic White (40.2%, 34.9%, 31.3%) and Hispanic/Latino candidates (36.4%, 32.6%, 31.6%; Figure 1B).

Figure 1 Unadjusted cumulative incidence probability of receiving a deceased donor kidney transplant. A and B: Race/ethnicity and body mass index subgroup (A) and estimated individual annual income (B), with 95% confidence interval. BMI: Body mass index.

In multivariable analysis, BMI 30- < 35 was associated with a 7% lower likelihood of DDKT for non-Hispanic White [sub-distribution hazard ratio (sHR): 0.93; 95% confidence interval (CI): 0.90-0.97] and non-Hispanic Black (sHR: 0.93; 95%CI: 0.89-0.96) candidates (Table 2). BMI ≥ 35 was associated with an 18% (sHR: 0.82; 95%CI: 0.78-0.86) lower likelihood of DDKT for non-Hispanic White candidates (Table 2). Low income was associated with a 34% and 20% higher likelihood of DDKT for non-Hispanic White (sHR: 1.34; 95%CI: 1.23-1.45) and Hispanic/Latino (sHR: 1.20; 95%CI: 1.06-1.36) candidates, respectively, in models adjusted for BMI, representing the direct effect of income not mediated by BMI (Table 2). However, income was not associated with likelihood of DDKT among non-Hispanic Black candidates. These findings were consistent with results from models excluding BMI, which represent the total effect of income (Table 3).

Table 2 Multivariable Fine-Gray proportional sub-distribution hazards model for the association of income and body mass index for time from listing to receiving a deceased or living donor kidney transplant.

	Non-Hispanic White candidates		Non-Hispanic Black candidates		Hispanic/Latino candidates
Characteristic	sHR (95%CI)	P value	sHR (95%CI)	P value	sHR (95%CI)	P value
Sub-distribution hazard ratios for time from listing to receiving a deceased donor kidney transplant1
BMI
< 30	1 (Reference)		1 (Reference)		1 (Reference)
30 < 35	0.93 (0.90, 0.97)	< 0.001	0.93 (0.89, 0.96)	< 0.001	0.97 (0.91, 1.02)	0.25
≥ 35	0.82 (0.78, 0.86)	< 0.001	0.97 (0.93, 1.02)	0.21	0.94 (0.87, 1.02)	0.12
Estimated individual annual income
High income	1 (Reference)		1 (Reference)		1 (Reference)
Middle income	1.13 (1.07, 1.20)	< 0.001	1.03 (0.97, 1.10)	0.31	1.07 (0.97, 1.17)	0.19
Low income	1.34 (1.23, 1.45)	< 0.001	1.05 (0.97, 1.14)	0.25	1.20 (1.06, 1.36)	< 0.001
Sub-distribution hazard ratios for time from listing to receiving a living donor kidney transplant1
BMI
< 30	1 (Reference)		1 (Reference)		1 (Reference)
30- < 35	0.94 (0.90, 0.99)	0.01	0.95 (0.86, 1.05)	0.33	0.90 (0.81, 0.98)	0.02
≥ 35	0.83 (0.79, 0.88)	< 0.001	0.80 (0.69, 0.92)	< 0.001	0.79 (0.71, 0.89)	< 0.001
Estimated individual annual income
High income	1 (Reference)		1 (Reference)		1 (Reference)
Middle income	0.74 (0.70, 0.78)	< 0.001	0.61 (0.54, 0.69)	< 0.001	0.99 (0.89, 1.11)	0.87
Low income	0.51 (0.47, 0.56)	< 0.001	0.42 (0.37, 0.48)	< 0.001	0.80 (0.68, 0.95)	0.01

¹Models adjusted for candidate age, birth sex, blood type, calculated panel reactive antibody at listing, time on dialysis, diagnosis, history of peripheral vascular disease, prior transplant, distance from transplant center, education, social vulnerability index, and year of listing. BMI: Body mass index; sHR: Sub-distribution hazard ratio; CI: Confidence interval.

Table 3 Multivariable Fine-Gray proportional sub-distribution hazards model for the association between income and time from listing to receiving a deceased or living donor kidney transplant.

	Non-Hispanic White candidates		Non-Hispanic Black candidates		Hispanic/Latino candidates
Estimated individual annual income	sHR (95%CI)	P value	sHR (95%CI)	P value	sHR (95%CI)	P value
Sub-distribution hazard ratios for time from listing to receiving a deceased donor kidney transplant1
High income	1 (Reference)		1 (Reference)		1 (Reference)
Medium	1.12 (1.06, 1.19)	< 0.001	1.03 (0.97, 1.10)	0.33	1.06 (0.97, 1.17)	0.20
Low	1.32 (1.21, 1.43)	< 0.001	1.05 (0.96, 1.14)	0.27	1.20 (1.06, 1.36)	0.01
Sub-distribution hazard ratios for time from listing to receiving a living donor kidney transplant1
High income	1 (Reference)		1 (Reference)		1 (Reference)
Medium	0.73 (0.69, 0.78)	< 0.001	0.60 (0.53, 0.68)	< 0.001	0.98 (0.88, 1.10)	0.74
Low	0.51 (0.47, 0.55)	< 0.001	0.41 (0.36, 0.47)	< 0.001	0.80 (0.67, 0.94)	0.01

¹Models adjusted for candidate age, birth sex, blood type, calculated panel reactive antibody at listing, time on dialysis, diagnosis, history of peripheral vascular disease, prior transplant, distance from transplant center, education, social vulnerability index, and year of listing. Model excludes body mass index. sHR: Sub-distribution hazard ratio; CI: Confidence interval.

In sensitivity analyses censoring LDKT (Supplementary Table 1), the total effect of income on the likelihood of DDKT was attenuated. However, low income remained associated with a higher likelihood of DDKT for non-Hispanic White and Hispanic/Latino candidates.

In models testing the interaction between income and BMI, low income was associated with a higher likelihood of DDKT for non-obese non-Hispanic White (sHR: 1.34; 95%CI: 1.23-1.46) and Hispanic/Latino (sHR: 1.16; 95%CI: 1.02-1.32) candidates (Table 4). Compared with non-obese candidates with high income, candidates with BMI ≥ 35 and high income had a lower likelihood of DDKT across all racial and ethnic groups. However, compared with non-obese candidates with high income non-Hispanic White and non-Hispanic Black candidates with BMI ≥ 35 and low income had a higher likelihood of DDKT.

Table 4 Multivariable Fine-Gray proportional sub-distribution hazards model assessing the interaction between income and body mass index for time from listing to receiving a deceased or living donor kidney transplant.

	Non-Hispanic White candidates		Non-Hispanic Black candidates		Hispanic/Latino candidates
Characteristic	sHR (95%CI)	P value	sHR (95%CI)	P value	sHR (95%CI)	P value
Sub-distribution hazard ratios for time from listing to receiving a deceased donor kidney transplant1
BMI: < 30 & high income	1 (Reference)		1 (Reference)		1 (Reference)
BMI: < 30 & middle income	1.13 (1.06, 1.20)	< 0.001	0.98 (0.91, 1.06)	0.66	0.93 (0.83, 1.03)	0.16
BMI: < 30 & low income	1.34 (1.23, 1.46)	< 0.001	1.03 (0.94, 1.13)	0.49	1.16 (1.02, 1.32)	0.02
BMI: 30- < 35 & high income	0.97 (0.91, 1.03)	0.31	0.92 (0.85, 0.99)	0.04	0.83 (0.74, 0.93)	< 0.001
BMI: 30- < 35 & middle income	0.98 (0.90, 1.06)	0.58	1.05 (0.95, 1.17)	0.36	1.43 (1.23, 1.66)	< 0.001
BMI: 30- < 35 & low income	0.91 (0.83, 1.00)	0.05	0.98 (0.89, 1.08)	0.68	1.11 (0.95, 1.30)	0.18
BMI: ≥ 35 & high income	0.75 (0.69, 0.82)	< 0.001	0.86 (0.78, 0.95)	< 0.001	0.84 (0.72, 0.98)	0.03
BMI: ≥ 35 & middle income	1.09 (0.97, 1.21)	0.14	1.21 (1.06, 1.37)	< 0.001	1.31 (1.07, 1.59)	0.01
BMI: ≥ 35 & low income	1.16 (1.04, 1.30)	0.01	1.13 (1.01, 1.26)	0.03	1.07 (0.88, 1.30)	0.50
Sub-distribution hazard ratios for time from listing to receiving a living donor kidney transplant1
BMI: < 30 & high income	1 (Reference)		1 (Reference)		1 (Reference)
BMI: < 30 & middle income	0.73 (0.68, 0.78)	< 0.001	0.66 (0.57, 0.77)	< 0.001	1.11 (0.97, 1.26)	0.14
BMI: < 30 & low income	0.53 (0.49, 0.58)	< 0.001	0.41 (0.34, 0.48)	< 0.001	0.81 (0.66, 0.97)	0.03
BMI: 30- < 35 & high income	0.94 (0.88, 1.00)	0.06	0.98 (0.82, 1.16)	0.78	1.01 (0.88, 1.17)	0.84
BMI: 30- < 35 & middle income	1.04 (0.94, 1.15)	0.49	0.84 (0.65, 1.09)	0.20	0.79 (0.64, 0.96)	0.02
BMI: 30- < 35 & low income	0.96 (0.84, 1.09)	0.52	1.08 (0.83, 1.42)	0.56	0.84 (0.68, 1.05)	0.14
BMI: ≥ 35 & high income	0.86 (0.80, 0.93)	< 0.001	0.85 (0.67, 1.07)	0.17	0.81 (0.67, 0.98)	0.03
BMI: ≥ 35 & middle income	0.99 (0.88, 1.11)	0.82	0.80 (0.58, 1.11)	0.18	0.73 (0.56, 0.97)	0.03
BMI: ≥ 35 & low income	0.87 (0.75, 1.01)	0.07	1.02 (0.74, 1.41)	0.90	1.29 (0.97, 1.72)	0.08

¹Models adjusted for candidate age, birth sex, blood type, calculated panel reactive antibody at listing, time on dialysis, diagnosis, history of peripheral vascular disease, prior transplant, distance from transplant center, education, social vulnerability index, and year of listing. BMI: Body mass index; sHR: Sub-distribution hazard ratio; CI: Confidence interval.

Association of obesity and income with likelihood of the LDKT by race/ethnicity

The unadjusted 3-year cumulative incidence of LDKT was highest for non-Hispanic White candidates across all BMI categories (< 30, ≥ 30, ≥ 35: 23.4%, 19.5%, 18.3%), followed by Hispanic/Latino candidates (14.2%, 13.1%, 12.1%), and non-Hispanic Black candidates (5.8%, 5%, 4.4%; Figure 2A). The unadjusted 3-year cumulative incidence of LDKT increased with increasing income and was highest for non-Hispanic White (low, middle, high: 13.5%, 19.8%, 28.5%), followed by Hispanic/Latino (10.8%, 13.8%, 16.2%), and non-Hispanic Black candidates (3.3%, 5.2%, 9.8%; Figure 2B).

Figure 2 Unadjusted cumulative incidence probability of receiving a living donor kidney transplant. A and B: Race/ethnicity and body mass index subgroup (A) and estimated individual annual income (B), with 95% confidence interval. BMI: Body mass index.

In multivariable analysis, BMI 30- < 35 was associated with lower likelihood of LDKT for non-Hispanic White and Hispanic/Latino candidates. BMI ≥ 35 was associated with lower likelihood of LDKT for all racial and ethnic groups (sHR: 0.83; 95%CI: 0.79-0.88 for non-Hispanic White, sHR: 0.80; 95%CI: 0.69-0.92 for non-Hispanic Black and sHR: 0.79; 95%CI: 0.71-0.89 for Hispanic/Latino candidates; Table 2). Middle income was associated with lower likelihood of LDKT for non-Hispanic White and non-Hispanic Black candidates; low income was associated with lower likelihood of LDKT for all racial and ethnic groups in models that adjusted for BMI (Table 2), which was consistent with findings from models that excluded BMI (Table 3).

In models examining the interaction between income and BMI, low income was associated with lower likelihood of LDKT for non-obese candidates across all racial and ethnic groups (Table 4). Compared with non-obese candidates with high income, non-Hispanic White and Hispanic/Latino candidates with BMI ≥ 35 and high income had a lower likelihood of LDKT.

DISCUSSION

We found that the associations between BMI, income, and access to kidney transplantation differed by race/ethnicity and type of transplant. Higher BMI was associated with lower likelihood of DDKT among non-Hispanic White candidates and, to a lesser extent, non-Hispanic Black candidates, whereas no association was observed among Hispanic/Latino candidates. Lower income was associated with a higher likelihood of DDKT among non-Hispanic White and Hispanic/Latino candidates but not among non-Hispanic Black candidates. Low income and BMI ≥ 35 were universally associated with reduced access to LDKT. We found little evidence that BMI mediated the relationship between income and transplant access. However, income and BMI interacted in ways that varied by race/ethnicity and type of transplant, indicating that the combined effects of social and clinical risk differ across subgroups.

We observed racial/ethnic differences in the relationship between BMI and likelihood of DDKT: A linear association for non-Hispanic White candidates, a non-linear association for non-Hispanic Black candidates, and no significant association for Hispanic/Latino candidates. Prior studies have reported mixed relationships between obesity and transplant access across settings[7,16,17]. Consistent with our findings among non-Hispanic White candidates, some studies have demonstrated an overall lower likelihood of DDKT among obese candidates in the United States and France[7,16]. However, this is the first study to examine these associations stratified by race/ethnicity. A prior United States-based study by Gill et al[18] identified sex differences in BMI thresholds linked with reduced access to transplant, and our findings extend this work by demonstrating that BMI thresholds associated with access to DDKT also vary by race and ethnicity within the United States transplant system.

Decreasing income demonstrated a dose-response relationship with higher likelihood of DDKT among non-Hispanic White candidates, while only low income was associated with higher likelihood of DDKT among Hispanic/Latino candidates and income was not associated with DDKT among non-Hispanic Black candidates. In prior work, we observed that lower income was associated with higher likelihood of DDKT overall, although results were not stratified by race/ethnicity[12]. This pattern should not be interpreted as indicating improved access among lower-income non-Hispanic White and Hispanic/Latino candidates but rather reflects greater reliance on DDKT among these racial/ethnic groups when access to LDKT is limited.

In contrast, income was not significantly associated with likelihood of DDKT among non-Hispanic Black candidates, who had the highest likelihood of DDKT and the lowest likelihood of LDKT across all income levels, such that even high-income non-Hispanic Black candidates were less likely to receive LDKT than low-income non-Hispanic White candidates (9.8% vs 13.5%). This pattern suggests that LDKT may not represent a viable pathway for many non-Hispanic Black candidates regardless of socioeconomic position, limiting opportunities to shift away from DDKT. In the United States, lower access to LDKT combined with more frequent dialysis initiation prior to waitlisting results in greater reliance on DDKT. Once listed, longer dialysis exposure may increase priority for DDKT, independent of income. Hence, the lower likelihood of LDKT, and longer dialysis exposure, suggest that higher likelihood of DDKT does not reflect equitable access to transplantation overall.

We also observed significant interactions between income and BMI and likelihood of DDKT that varied by race/ethnicity. Among non-obese candidates, lower income was associated with a higher likelihood of DDKT for non-Hispanic White and Hispanic/Latino candidates; however, this association was attenuated at higher BMI. In contrast, among candidates with BMI ≥ 35, middle and low income were sometimes associated with higher likelihood of DDKT relative to non-obese candidates with high income. These patterns suggest differential reliance on DDKT among candidates with combined clinical and social risk, particularly when access to LDKT is limited.

For LDKT, overall, low income and BMI ≥ 35 were consistently associated with reduced likelihood of transplantation across all racial and ethnic groups, although the effects of BMI 30- < 35 and middle income varied by race/ethnicity. Interactions between income and BMI reinforced these main effects: Lower income was consistently associated with substantially reduced likelihood of LDKT among non-obese candidates across racial and ethnic groups, while higher BMI was associated with lower likelihood of LDKT even among candidates with high income. Together, these findings indicate that income and BMI jointly influence access to living donor transplantation and identify candidates with the combined burden of higher BMI and lower income as a particularly high-risk group for reduced access to LDKT.

Although obesity is not a contraindication for transplant and no formal BMI cutoff exists in clinical guidelines, many centers require or encourage weight loss through counseling or metabolic surgery prior to transplant. Prior studies have shown that weight loss among obese candidates is associated with increased likelihood of transplantation[16]. Barriers to LDKT among candidates with higher BMI may arise earlier in the evaluation process, including donor–recipient matching, surgical planning considerations, and center-level risk assessment, rather than reflecting differences in transplantation once listed. These findings highlight opportunities for targeted intervention, including structured weight-management pathways, enhanced navigation and financial support for potential living donors, and more consistent evaluation practices across transplant centers, to mitigate the compounded effects of clinical and social risk factors on access to LDKT.

This study has several limitations. First, BMI is an imperfect proxy for obesity. Measures such as waist circumference or waist-to-hip ratio may better reflect fat distribution and surgical suitability[17]; however, these measures are not collected by the OPTN. We categorized BMI based on the CDC guidelines, often used by clinicians and transplant centers; however, examining BMI as a continuous variable could provide additional insight[16,19]. Moreover, we did not assess changes in BMI over time, which has been shown to be associated with likelihood of transplantation[16]. Second, income data from LexisNexis have not been validated for research purposes and their proprietary nature limits transparency and reproducibility. These data approximate individual income at the time of listing and may be subject to misclassification and measurement error. In addition, they reflect personal, rather than household-level income, which may not fully capture an individual’s financial status. To facilitate interpretability and mitigate the influence of measurement error, income was categorized into cohortspecific tertiles and used as a relative indicator of socioeconomic position rather than an absolute, externally generalizable threshold.

Third, although all multivariable models included transplant center as a random effect to partially account for unmeasured centerlevel variation, our analyses could not directly assess differences in transplant center practices. Variability in obesity eligibility thresholds, referral and activation patterns, and the availability of resources to support candidates with lower income differ across transplant programs and may influence access to transplantation beyond patientlevel characteristics, contributing to observed differences across subgroups. Fourth, our analytic cohort was limited to individuals already added to the kidney transplant waiting list. Obesity and income likely influence earlier steps in the transplant pathway, including referral, evaluation, and listing. As a result, our findings should be interpreted as reflecting disparities in access to transplantation among waitlisted candidates and may underestimate inequities that arise prior to listing. Additionally, although we extended the analysis to include candidates listed in 2022 and adjusted for calendar year, residual effects of the coronavirus disease 2019 pandemic cannot be fully excluded. The pandemic coincided with disruptions in donor supply, transplant program capacity, and clinical workflows, which may have differentially affected access to transplantation across centers and patient subgroups, particularly in earlier years of the cohort. Lastly, nonHispanic individuals from other racial groups were excluded due to small sample sizes that limited our ability to generate stable racespecific estimates. While this exclusion does not affect inference within the included racial and ethnic groups, it limits the generalizability of our findings to other racial populations.

CONCLUSION

In conclusion, we demonstrate that obesity and income are differentially associated with access to kidney transplantation across racial and ethnic groups and by type of transplant, underscoring the importance of distinguishing between DDKT and LDKT when evaluating socioeconomic disparities. Our findings identify highrisk subgroups, including lowerincome candidates with lower likelihood of LDKT and candidates with higher BMI whose access varies by transplant type. These results highlight the need for targeted interventions that promote more equitable and standardized evaluation and management of obesity across transplant centers, and account for how SDOH impact access to DDKT and LDKT. Addressing both clinical and social risk factors may help reduce DDKT- and LDKTspecific disparities and improve equity in access to kidney transplantation.

ACKNOWLEDGEMENTS

The data reported here have been supplied by UNOS as the contractor for the Organ Procurement and Transplantation Network (OPTN). The interpretation and reporting of these data are the responsibility of the author(s) and in no way should be seen as an official policy of or interpretation by the OPTN or the United States Government.