Journal Logo

Original Clinical Science—General

Population Health, Ethnicity, and Rate of Living Donor Kidney Transplantation

Reed, Rhiannon D.1; Sawinski, Deirdre MD2; Shelton, Brittany A.1; MacLennan, Paul A. PhD1; Hanaway, Michael MD1; Kumar, Vineeta MD3; Long, Dustin PhD4; Gaston, Robert S. MD3; Kilgore, Meredith L. PhD5; Julian, Bruce A. MD3; Lewis, Cora E. MD6; Locke, Jayme E. MD1

Author Information
doi: 10.1097/TP.0000000000002286

Although superior long-term outcomes are achieved with kidney transplantation from living donors compared to allografts from deceased donors or dialysis, living donor kidney transplantation (LDKT) has decreased in the United States since 2004.1 The development of additional regulatory policies for donors (Organ Procurement and Transplantation Network (OPTN) Policies 14 and 18.5, Living Donation, April 2017), recent publications of negative donor outcomes,2,3 and an aging transplant population with limited social networks may limit the pool of potential living donors.4,5

Another explanation may be the increasing prevalence of absolute and relative contraindications to donation, including diabetes mellitus (DM), obesity, and hypertension in the general population.6,7 Studies have suggested that medical conditions including undiagnosed hypertension or abnormal glucose tolerance are the most common reasons for nondonation.8-11 A rise in unemployment and a decrease in median household income after the economic downturn of 2008 may have amplified financial disincentives to donation, such as out-of-pocket expenses and lost wages, particularly among those with low incomes, who are already known to donate at lower rates.5,12 Transplant candidates may be reluctant to ask individuals to donate in the current economic climate, given concerns about the potential financial impact to the living donor.13

Although population health and financial statuses have been hypothesized as explanations for the decrease in living donation, these assertions remain anecdotal. To our knowledge, no study has directly examined the relationship between population health or socioeconomic status (SES) and rate of LDKT using national cross-sectional data. Identifying and quantifying factors that may influence donation rates is necessary to focus research and develop interventions to improve rates of LDKT. We performed an ecological analysis using state-specific measures of population health and SES to investigate the association with transplant center rates of LDKT.


Data Sources

The primary data source was the 2015 Behavioral Risk Factor Surveillance System (BRFSS) State Prevalence and Trends Data at the Centers for Disease Control and Prevention. Behavioral Risk Factor Surveillance System is the largest continuously conducted telephone health survey system in the world, completing more than 400 000 adult interviews in the United States every year. These data include health-related risk behaviors, chronic health conditions, and use of preventive services from all 50 states, the District of Columbia, and 3 United States territories.14

The second data source was the Scientific Registry of Transplant Recipients (SRTR). The SRTR data system includes data on all donors, waitlisted candidates, and transplant recipients in the United States; these data are submitted by the members of the OPTN. The Health Resources and Services Administration of the United States Department of Health and Human Services provides oversight to the activities of the OPTN and SRTR contractors. The study was approved by the University of Alabama at Birmingham Institutional Review Board (161212003).

Study Population

The unit of analysis was kidney transplant center. All United States kidney transplant centers that performed at least 10 transplants in 2015 were eligible for inclusion. One center performed more than 10 transplants in 2015 but did not list anyone in 2013 to 2014 and thus was excluded from the analysis, resulting in a final cohort of 213 kidney transplant centers.

Categorization of Exposures

The BRFSS prevalence measures are reported by the CDC at the state level and are weighted to account for the complex survey sampling design. To create prevalence measures that best reflected the population characteristics of a transplant center, we defined each center's “catchment area” as the list of states from which patients were added to the waiting list at that center (see Materials and Methods, SDC, Center demographic and SES indicator prevalence measures were weighted by multiplying a given state's prevalence measure by the proportion of waitlisted patients from that state (eg, if 80% of transplant candidates at center A were from state A and 20% were from state B, the prevalence of obesity in state A was multiplied by 0.8 and added to the prevalence of obesity in state B multiplied by 0.2, and the resulting prevalence was assigned to the center). This was done to make a transplant center's prevalence measures look more like the patient population of the center and to account for heterogeneity within a state that is not captured by a state-level summary measure in BRFSS. A summary of the within-state variation achieved using this weighting measure is presented in Table S1 (SDC,

The following population demographic and SES indicators hypothesized a priori to be associated with rate of LDKT were considered for analysis: prevalence of age of 65 years or older, male sex, minority race/ethnicity defined as non-white (African American [AA], Asian, Hispanic, American Indian, Native Hawaiian, other, or multiracial), less than college education, lack of health insurance (defined as report of “no healthcare coverage”), annual household income less than US $15 000, unemployment (collapsed responses for “no work for < 1 year” and “no work for > 1 year”), no internet use in past 30 days, and not married/no partner. We considered the following population health indicators for analysis, as they are absolute and relative contraindications to living kidney donation: history of cardiovascular disease (CVD), DM, hypertension, kidney disease, depression, poor self-rated health, obesity, and currently smoking.

For center-level characteristics, we examined the absolute number of living donor transplants performed in 2015 and whether the transplant center performed incompatible kidney transplants (either blood group incompatible or donor exchange programs).15

Outcome Ascertainment

Center rate of living donation was defined as the proportion of all kidney transplants performed at a center in 2015 that were from living donors.

Statistical Analyses

Using measures of central tendency and spread, we explored the distribution of center prevalence measures by OPTN region. Given that some states only have 1 active transplant center, we chose to present the rate of LDKT in a heat map at the OPTN region rather than the state level, so as not to identify unique transplant centers. Prevalence measures were also described at the region level for consistency. Spearman's correlation was used to generate the correlation coefficient between covariates to assess the potential for collinearity. We also investigated the variance inflation factor (VIF) for each covariate and obtained VIFs >10 for CVD, DM, minority prevalence, and smoking and VIFs approaching 10 for obesity, lower education, unemployment, and no internet use. As such, we chose to collapse SES and health factors into 2 indices.

To create the indices, prevalence measures were dichotomized into whether the center's weighted prevalence was greater than or equal to the national median of that factor (Table S2, SDC, The relationship between the dichotomous factor and rate of LDKT was explored. We performed principal component factor analyses using measures with P values of 0.1 or less on unadjusted analyses, to confirm the communality of the measures and obtain the factor loadings for each measure to calculate weighted factor-based scores.16,17 If a center's prevalence was greater than or equal to the national median, the factor loading was added to the total score for each index, such that a center could have a maximum score of 4.101 for the disease index and a maximum score of 3.291 for the SES index. To test internal consistency of the indices, we calculated Cronbach alpha for each index. Given that health and SES are correlated and to determine whether there was an additive effect, the indices were cut at the median and categorized as low versus high and further collapsed into a single measure, with the number of centers falling into each category presented in Table S3 (SDC,

Living donation rate was examined for normality, and model diagnostics assessed the appropriateness of the assumption of linearity, with both assumptions confirmed. Given the presence of more than 1 transplant center in some states and the potential for lack of independence of these centers, we used a linear mixed effects model with a random intercept for state accounting for within-state correlation to assess the association between population health and SES factors and center rate of living donation. The index and all demographic and center-level factors were considered for multivariable modeling, and the most parsimonious model was chosen by minimizing the Akaike Information Criterion. All analyses were conducted with SAS version 9.4 (SAS Institute, Inc., Cary, NC).

Sensitivity Analyses

To account for other factors of center performance that may influence volume of living donors at a center, we ran several sensitivity analyses. Given the concerns by Matar et al18 that proportion of all kidney transplants done that are from living donors may not accurately measure center performance of LDKT, we ran a Poisson model to estimate the rate ratio (RR) of living donor transplants per individuals on the waiting list as of January 1, 2015, as a function of population characteristics. Inferences were consistent and are reported in Table S4 (SDC, We also explored the inclusion of deceased donor organs available per waiting list population, median center waiting time, waitlist additions, and total on waitlist at the beginning of the study period as covariates, and our findings were confirmed. Additional sensitivity analyses included excluding Children's Hospitals and generating a linear model with robust standard errors (R2 = 0.37). Finally, we explored different definitions of catchment area, based on distribution of donor zip code and 200-mile radius around the transplant center. All inferences were consistent.


Center Prevalence Measures by Region

Among 213 centers, the prevalence of LDKT and population measures varied by OPTN region. The rate of LDKT and the number of centers within each region are presented in Figure 1. The region with the highest rate of LDKT was region 1 with a median rate of 48% (interquartile range [IQR], 30.1-54.8), whereas region 6 had the lowest rate of LDKT (19%; IQR, 16.8-29.4).

Heat map of rate of living kidney donor transplantation by OPTN region.

Prevalence of minority race/ethnicity (categorized as other than white, non-Hispanic) ranged from 18% in region 7 (IQR, 15.5-33.6) to 58% in region 5 (IQR, 41.4-58.1) (Table 1, Figure 2A). When examining specific groups of minority race/ethnicity, Region 3 had the highest prevalence of AA (median, 21%; IQR, 14.1-28.7), and region 4 had the highest prevalence of Hispanics (median, 35%; IQR, 33.7-35.0). The prevalence of health and SES characteristics also varied by region, with the highest rates of comorbid disease and poorest SES observed in regions 3, 4, and 11 (Table 1).

Prevalence measures by UNOS Region
Heat map of rate of minority prevalence, health score, and socioeconomic score by OPTN region.

Association of Prevalence Measures and LDKT

Of the demographic factors, only high minority prevalence was significantly associated with lower rate of LDKT. High prevalence of all SES factors except marital status was negatively associated with rate of LDKT. Health factors associated with lower rate of LDKT were high prevalence of CVD, DM, kidney disease, and poor self-rated health (Table 2).

Estimates for relationship between center prevalence measures classified as above or below national median and living donation rate (interpreted as: if prevalence of covariate X is “high” or above the national median, the rate of living donation differs by x percentage points from centers with prevalence lower than the national median)

Factor analysis revealed 1 factor for the health measures with an Eigenvalue of 3.48 (standardized Cronbach alpha = 0.92). When the health factors were collapsed into a composite measure of disease based on factor loadings, regions 3, 10, and 11 had the highest/worst median composite score, meaning that most centers in those regions had a prevalence of comorbidities that was higher/worse than the national median (Figure 2B). The factor analysis also revealed 1 factor for the SES measures with an Eigenvalue of 2.37 (standardized Cronbach alpha = 0.76). When collapsed into a composite measure of poor SES, regions 3 and 4 had the highest/worst median composite score, indicating that most centers in these regions had a prevalence of poor SES that was worse than the national median (Figure 2C).

On unadjusted analyses, we found that each 1-unit increase in the composite SES score was associated with an average decrease in the rate of LDKT by 4.13 percentage points (95% confidence interval [CI], −6.1 to −2.2; P < 0.001) (Table 3). Each 1-unit increase in the composite disease score was associated with an average decrease of 2.45 percentage points in the rate of LDKT (95% CI, −4.0 to −0.9; P = 0.003). When the SES and disease scores were categorized as low versus high and combined into 1 index, 69 (32.4%) centers were in catchment areas with low scores for both disease and SES indices, and 76 (35.7%) centers had high scores for both indices, with the remaining 68 centers having a high score for only 1 index (Table S2, SDC, Centers in areas with a high prevalence of both comorbid disease and poor SES had a rate of LDKT that was 10.35 percentage points lower than centers in areas that had low scores for both disease and poor SES (95% CI, −15.95 to −4.75; P < 0.001). Centers with a high score for only 1 factor did not differ statistically in the rate of LDKT from centers with low scores for both indices (Table 4).

Factor loadings and estimates for relationship between composite SES and disease indices with rate of living donation
Relationship between center prevalence measures and living donation rate (with collapsed composite measures for disease and SES)

In an adjusted model accounting for total center LDKT volume, high prevalence of age of 65 years or older, high prevalence of male sex, and presence of an incompatible transplant program, the significant relationship between high prevalence of minority race and LDKT persisted, with a rate of LDKT that was on average 7.1 percentage points lower than centers in areas with fewer minorities (95% CI, −11.8 to −2.3; P = 0.004). The combined disease/SES index also remained significant, with centers with higher/worse scores for both disease and SES associated with an average rate of LDKT that was 7.3 percentage points lower than that for centers with low disease and SES scores (95% CI, −12.2 to −2.3; P = 0.004). Centers with an incompatible transplant program had a rate of LDKT that was on average 5.92 percentage points higher than centers without a similar program (95% CI, 1.82-10.02; P = 0.005) (Table 4).

When we used living donor transplants performed per waiting list registrants in a Poisson model as an alternative measure of center performance of LDKT, similar results were found. The RR for centers with high disease and SES scores was 0.79 (95% CI, 0.65-0.97; P = 0.02), suggesting that centers with more comorbid disease and poorer SES have a rate of LDKT per waiting list population that is 21% lower than centers in healthier areas with higher SES. Centers in areas of high minority prevalence had a RR of 0.61 (95% CI, 0.49-0.76; P < 0.001), indicating a 39% lower rate than areas with fewer minorities. Finally, the RR for centers with an incompatible transplant program was 1.80 (95% CI, 1.54-2.11; P < 0.001), demonstrating a positive association between incompatible transplantation and rate of LDKT (Table S4, SDC,


In this ecological analysis, the first using BRFSS population health and SES measures to examine associations with center rates of LDKT, we observed a significant negative association between rate of LDKT and higher prevalence of minorities, poor SES, and comorbid disease. Centers with high scores for both disease and SES indices had LDKT rates that were on average 7.3 percentage points lower than centers serving healthier and more economically advantaged populations. Centers in areas with higher prevalence of minorities had a significantly lower rate of LDKT, 7.1 percentage points lower than the rate for centers in areas with fewer minorities. These data provide the first cross-sectional evidence that living donation is associated with population health and economic well-being.

Our study found a significant association between high minority prevalence and lower rates of LDKT, even after adjusting for health and SES factors. Racial disparities in LDKT are well-known.19,20 African Americans account for 30% to 50% of transplant waiting lists but only 10% to 15% of living donors,21 and fewer Hispanics receive LDKT than non-Hispanic whites (4% vs 10%).22 This disparity has been attributed to sociocultural barriers and lack of knowledge regarding risks and benefits of LDKT specific to minority populations.23,24 Several transplant centers have initiated culturally targeted interventions among minorities and have successfully increased transplant knowledge and rate of donor inquiry.25-27 Systematic implementation of these programs may reduce some of the racial disparity in rates of LDKT. In addition, recent studies have raised concerns about the role of Apolipoprotein L1 (APOL1) in recipients of kidney transplants from deceased AA donors.28,29 The investigation of APOL1 among living donors has been limited.30,31 However, concerns remain, particularly given population studies demonstrating higher rates of postdonation end-stage renal disease and mortality among AA living donors.3,32 These data may contribute to the persistently lower rate of living donation among AAs.

Although the presence of high/worse SES composite score alone was not significantly associated with lower rates of LDKT compared with centers with low scores for both health and SES, several SES score components yielded interesting findings. High prevalence of no health insurance was associated with a lower rate of LDKT, consistent with previous reports, as some centers will decline potential donors with no health insurance.33 The association between no internet use in the previous 30 days and a lower rate of LDKT supports the hypothesis proposed by Rodrigue about limited social networks.5 Lack of internet use may be a proxy for lower access to medical care, a factor also known to be associated with lower rates of LDKT.12,34 These findings highlight the need for programs that actively engage candidate social networks to identify potential living donors, including those that engage potential donors in their homes and communities.35,36 Additionally, the negative associations between high prevalence of lack of health insurance and high prevalence of low income with rate of LDKT further motivate discussions about testing interventions to remove or mitigate financial disincentives to living donation.37,38

Our study findings suggest that addressing socioeconomic disparities alone may not increase rates of living donation. Although there is evidence that donor selection criteria have expanded to meet the growing organ shortage,39,40 it has been suggested that some centers have become more reticent to accept donors in subgroups shown to be at increased risk of end-stage renal disease.5,41 Some donor factors may be modifiable at the time of evaluation (ie, morbid obesity), but other threats to population health, such as diabetes and hypertension, require earlier and more widespread implementation of awareness and prevention programs, particularly among children and adolescents. The Robert Wood Johnson Foundation “Roadmaps to Health” provide strategies to improve population health that have been shown to be effective through empirical evidence,42 including competitive pricing for healthy foods43 and childhood obesity prevention programs,44 which could be implemented more broadly to improve population health, thereby increasing access to living kidney donation while simultaneously decreasing the need for kidney transplantation. Additionally, community-based participatory research has been used to influence policies regarding chronic disease prevention and treatment.45 Engaging community members is essential to the success of initiatives to reduce disease prevalence and incidence.

The finding that the presence of high/worse scores for both disease and SES factors was negatively associated with rate of living donation suggests that the decline in living donation may be multifactorial and would benefit from targeted efforts in both areas. It is also an important finding, given that nearly 36% of the US transplant centers included in our analyses were located in catchment areas defined as having high disease and poor SES scores.

Our study is not without limitations. BRFSS relies on information reported directly by a participant and thus may be subject to various sources of bias. We are not able to account for transplant staff size, which may be indicative of larger-volume centers that can perform more LDKTs. The percentage of recipients listed within a state in 2013 to 2014 may have differed from the fraction of the total number of recipients from that state transplanted at a center in 2015. Some LDKTs may have been performed at centers other than those that first waitlisted the recipient. Because of the changes in the survey and sampling design in 2011, we cannot assess time trends to compare prevalence measures from 2004 (the peak of LDKT) and 2015 to determine if similar associations would have been seen in prior years. Additionally, BRFSS is population-level data, and as such, we cannot attribute our findings to be the cause of the decline in LDKT. However, this is the first study, to our knowledge, using this data source to examine the cross-sectional association of population characteristics with LDKT.

In conclusion, center-level variation in LDKT was associated with population characteristics and minority prevalence. Further examination of these factors in the context of patient and center-level barriers to LDKT is warranted.


The data reported here have been supplied by the Minneapolis Medical Research Foundation (MMRF) as the contractor for the Scientific Registry of Transplant Recipients (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 US Government. These data were presented in preliminary form at the 2017 American Transplant Congress in Chicago, IL.


1. Hart A, Smith JM, Skeans MA, et al. OPTN/SRTR 2015 Annual Data Report: Kidney. Am J Transplant. 2017;17(Suppl 1):21–116.
2. Lentine KL, Schnitzler MA, Xiao H, et al. Racial variation in medical outcomes among living kidney donors. N Engl J Med. 2010;363:724–732.
3. Muzaale AD, Massie AB, Wang MC, et al. Risk of end-stage renal disease following live kidney donation. JAMA. 2014;311:579–586.
4. Fung HH, Carstensen LL, Lang FR. Age-related patterns in social networks among European Americans and African Americans: implications for socioemotional selectivity across the life span. Int J Aging Hum Dev. 2001;52:185–206.
5. Rodrigue JR, Schold JD, Mandelbrot DA. The decline in living kidney donation in the United States: random variation or cause for concern? Transplantation. 2013;96:767–773.
6. Ogden CL, Carroll MD, Fryar CD, et al. Prevalence of Obesity Among Adults and Youth: United States, 2011-2014. NCHS Data Brief. 2015;1–8.
7. Centers for Disease Control and Prevention (CDC). Long-term Trends in Diabetes. 2016.
8. Moore DR, Feurer ID, Zaydfudim V, et al. Evaluation of living kidney donors: variables that affect donation. Prog Transplant. 2012;22:385–392.
9. Weng FL, Dhillon N, Lin Y, et al. Racial differences in outcomes of the evaluation of potential live kidney donors: a retrospective cohort study. Am J Nephrol. 2012;35:409–415.
10. Norman SP, Song PX, Hu Y, et al. Transition from donor candidates to live kidney donors: the impact of race and undiagnosed medical disease states. Clin Transplant. 2011;25:136–145.
11. Tankersley MR, Gaston RS, Curtis JJ, et al. The living donor process in kidney transplantation: influence of race and comorbidity. Transplant Proc. 1997;29:3722–3723.
12. Gill J, Dong J, Rose C, et al. The effect of race and income on living kidney donation in the United States. J Am Soc Nephrol. 2013;24:1872–1879.
13. Rodrigue JR, Cornell DL, Kaplan B, et al. Patients' willingness to talk to others about living kidney donation. Prog Transplant. 2008;18:25–31.
14. Centers for Disease Control and Prevention (CDC). Behavioral Risk Factor Surveilance System Survey Data. Atlanta, Georgia; 2015.
15. Reese PP, Feldman HI, Bloom RD, et al. Assessment of variation in live donor kidney transplantation across transplant centers in the United States. Transplantation. 2011;91:1357–1363.
16. de Vaus D. Surveys in Social Research. London, United Kingdom and New York, NY: Routledge; 2014.
17. DiStefano C, Zhu M, Mindrila D. Understanding and Using Factor Scores: Considerations for the Applied Researcher. Practical Assessment, Research & Evaluation. 2009;14 Available online:
18. Matar AJ, Files J, Burkholder R, et al. Evaluating living donor kidney transplant rates: are you reaching your potential? Clin Transplant. 2017;31: doi: 10.1111/ctr.12914.
19. Weng FL, Reese PP, Mulgaonkar S, et al. Barriers to living donor kidney transplantation among black or older transplant candidates. Clin J Am Soc Nephrol. 2010;5:2338–2347.
20. Gore JL, Danovitch GM, Litwin MS, et al. Disparities in the utilization of live donor renal transplantation. Am J Transplant. 2009;9:1124–1133.
21. OPTN. National Data. Accessed April 21, 2017.
22. Gordon EJ, Feinglass J, Carney P, et al. A culturally targeted website for Hispanics/Latinos about living kidney donation and transplantation: a randomized controlled trial of increased knowledge. Transplantation. 2016;100:1149–1160.
23. Locke JE, Qu H, Shewchuk R, et al. Identification of strategies to facilitate organ donation among African Americans using the nominal group technique. Clin J Am Soc Nephrol. 2015;10:286–293.
24. Gordon EJ, Mullee JO, Ramirez DI, et al. Hispanic/Latino concerns about living kidney donation: a focus group study. Prog Transplant. 2014;24:152–162.
25. Gordon EJ, Feinglass J, Carney P, et al. A website intervention to increase knowledge about living kidney donation and transplantation among hispanic/latino dialysis patients. Prog Transplant. 2016;26:82–91.
26. Arriola KR, Powell CL, Thompson NJ, et al. Living donor transplant education for African American patients with end-stage renal disease. Prog Transplant. 2014;24:362–370.
27. Rodrigue JR, Paek MJ, Egbuna O, et al. Making house calls increases living donor inquiries and evaluations for blacks on the kidney transplant waiting list. Transplantation. 2014;98:979–986.
28. Freedman BI, Julian BA, Pastan SO, et al. Apolipoprotein L1 gene variants in deceased organ donors are associated with renal allograft failure. Am J Transplant. 2015;15:1615–1622.
29. Freedman BI, Pastan SO, Israni AK, et al. APOL1 genotype and kidney transplantation outcomes from deceased African American donors. Transplantation. 2016;100:194–202.
30. Zwang NA, Shetty A, Sustento-Reodica N, et al. APOL1-associated end-stage renal disease in a living kidney transplant donor. Am J Transplant. 2016;16:3568–3572.
31. Kofman T, Audard V, Narjoz C, et al. APOL1 polymorphisms and development of CKD in an identical twin donor and recipient pair. Am J Kidney Dis. 2014;63:816–819.
32. Segev DL, Muzaale AD, Caffo BS, et al. Perioperative mortality and long-term survival following live kidney donation. JAMA. 2010;303:959–966.
33. Rodrigue JR, Pavlakis M, Danovitch GM, et al. Evaluating living kidney donors: relationship types, psychosocial criteria, and consent processes at US transplant programs. Am J Transplant. 2007;7:2326–2332.
34. Axelrod DA, Dzebisashvili N, Schnitzler MA, et al. The interplay of socioeconomic status, distance to center, and interdonor service area travel on kidney transplant access and outcomes. Clin J Am Soc Nephrol. 2010;5:2276–2288.
35. Rodrigue JR, Cornell DL, Lin JK, et al. Increasing live donor kidney transplantation: a randomized controlled trial of a home-based educational intervention. Am J Transplant. 2007;7:394–401.
36. Garonzik-Wang JM, Berger JC, Ros RL, et al. Live donor champion: finding live kidney donors by separating the advocate from the patient. Transplantation. 2012;93:1147–1150.
37. Tushla L, Rudow DL, Milton J, et al. Living-donor kidney transplantation: reducing financial barriers to live kidney donation—recommendations from a consensus conference. Clin J Am Soc Nephrol. 2015;10:1696–1702.
38. Hays R, Rodrigue JR, Cohen D, et al. Financial neutrality for living organ donors: reasoning, rationale, definitions, and implementation strategies. Am J Transplant. 2016;16:1973–1981.
39. Davis CL, Cooper M. The state of U.S. living kidney donors. Clin J Am Soc Nephrol. 2010;5:1873–1880.
40. Mandelbrot DA, Pavlakis M, Danovitch GM, et al. The medical evaluation of living kidney donors: a survey of US transplant centers. Am J Transplant. 2007;7:2333–2343.
41. Lentine KL, Schnitzler MA, Xiao H, et al. Consistency of racial variation in medical outcomes among publicly and privately insured living kidney donors. Transplantation. 2014;97:316–324.
42. Robert Wood Johnson Foundation. County Health Rankings & Roadmaps. 2017; Accessed April 27, 2017.
43. Mozaffarian D, Afshin A, Benowitz NL, et al. Population approaches to improve diet, physical activity, and smoking habits: a scientific statement from the American Heart Association. Circulation. 2012;126:1514–1563.
44. Waters E, de Silva-Sanigorski A, Hall BJ, et al. Interventions for preventing obesity in children. Cochrane Database Syst Rev. 2011;CD001871.
45. Wynn TA, Taylor-Jones MM, Johnson RE, et al. Using community-based participatory approaches to mobilize communities for policy change. Fam Community Health. 2011;34(Suppl 1):S102–S114.

Supplemental Digital Content

Copyright © 2018 Wolters Kluwer Health, Inc. All rights reserved.