The Change in Living Kidney Donation in Women and Men in the United States (2005–2015): A Population-Based Analysis

Discussion

This study represents the first population-level analysis of living kidney donation in women and men in the United States in the current era, when overall living donation rates have declined and during an economically strained time in the United States, with increased rates of unemployment and decreased job security.¹⁴ We found that women have a 44% higher incidence of living kidney donation compared with men, after adjustment for factors that have been previously hypothesized to explain this disparity, including sex-specific differences in ESRD. We also found that living donation in women has remained stable over the last decade, whereas there has been a marked decrease in living donation in men, resulting in a widening gap in donation between the sexes. We found that living donation rates varied by income in both women and men, but the decline in living donation was most marked in men from lower income groups. Finally, we found that living related donations declined in both women and men, in all income groups.

The last national population-based analysis examining sex differences in living donation was on the basis of transplantations performed during the years 1991–1993.¹⁵ Our analysis is timely because it examines sex differences in donation during a period when living donation is declining. Although it is difficult to compare the results from the 1990s to our findings, the fact that we demonstrate a widening of the discrepancy in donation between women and men over the study period should renew interest in understanding the basis for this long-standing discrepancy. Importantly, the previous work did not include consideration of sex related differences in income or ESRD.¹⁵ Our finding that sex-related differences in living donation persisted after adjustment for differences in ESRD between women and men suggests the importance of other factors including differences in the financial implications of donation, attitudes toward donation, or societal expectations of donation between women and men contribute to the sex-related difference in living donation.^5,15

Our findings also draw further attention to financial barriers to living donation (see recommendations from the American Society of Transplantation Living Donor Community of Practice consensus conference¹⁶ and other recent publications^17,18 for a comprehensive overview of this issue). The finding that income had a greater effect on donation from men should be interpreted with the knowledge that living donors and recipients are usually from the same socioeconomic background and, in many cases, the same household.¹⁹ It is notable that the largest difference (IRR, 2.48) in donation between women and men was observed for spousal donations. Because a larger proportion of men are the primary household income earner in the United States²⁰ and are likely to earn higher wages compared with women,⁷ it follows that the potential for lost wages may disproportionately affect donations from men. With the exception of selected private charities and paired exchange registries, most current expense reimbursement programs for living donors do not cover lost wages. The National Living Donor Assistance Center does not provide support for lost wages, but it is currently engaged in a randomized trial in which potential transplant recipients will be randomized to be offered or not offered wage reimbursement for their donor.²¹

It is important to note that financial barriers to donation are not limited to out of pocket expenses and lost wages. Concerns about financial, employment, or insurance consequences of donation may have been particularly relevant in our analysis, as it was conducted during the period of economic downturn in the United States. The potential for an interruption in employment or loss of employment may pose a greater concern for loss of work-related benefits, including health insurance, in men. Ensuring adequate medical follow-up for living donors postdonation is increasingly recognized a priority for transplantation programs,^22,23 and therefore lack of health insurance may be a significant barrier to donation. Because a larger proportion of women in the United States have their health insurance benefits covered as a dependent,²⁴ the threat of job loss may not necessarily affect their ability to access health care in the future. Conversely, the prospect of a male donor losing his job may more frequently result in loss of health insurance for him and his dependents. The proposed Living Donor Protection Act²⁵ would entitle a covered employee to medical leave in the event of living kidney donation. The bill would also prohibit discrimination on the basis of an individual’s status as a living organ donor in “the offering, issuance, cancellation, coverage, price, or any other condition of a life insurance policy, disability insurance policy, or long-term care insurance policy.” Our findings suggest that the passage of this bill into law may indirectly help increase living donation in the future, particularly from men.

Although our study emphasizes the importance of income as a barrier to donation in men, it is important to highlight that income was associated with lower donation rates in both women and men. We previously reported that the retraction in living donation since 2005 was restricted to lower income populations.¹³ Together, these findings highlight the need for broad based policies to ensure donation is financially neutral for living donors.^26,27 This includes reimbursement for out of pocket expenses at the time of donation, and for long-term health insurance coverage for donation related complications, as well as for health care required to preserve postdonation kidney function.²⁸

Our findings also highlight the role of additional factors beyond financial considerations in the decline in living donation over the past decade. We found that living related donation declined by 42% in women and 47% in men between 2005 and 2015. Of note, this decline did not vary by median household income. Parent to child living donations have declined, in part, because of changes in deceased donor organ allocation that prioritize pediatric recipients,²⁹ but this only constitutes a minority of living related donations. Studies about the long-term health outcomes of living donors were published in 2013³⁰ and 2014,³¹ and may have had some limited effect on the use of living related donations in the latter years of our analysis, but were unlikely to contribute to the decline in donation that preceded these studies. It is important to understand that other than the potential role of APOL-1 gene variants,^32–36 the basis for an increased risk of kidney failure in living related donors remains uncertain. Importantly, lifestyle factors that are shared among family members may lead to conditions such as diabetes and hypertension, which appear to be important antecedents of CKD in previous living donors.^37,38 The importance of these recent observations is that much of the apparent familial risk in related living donors may be modifiable with appropriate postdonation care.

The strengths of this analysis are that it includes all living donors in the United States with a valid zip code and directly examined prevalent and longitudinal changes in living donor transplantation in women and men, after adjustment for differences in age, race, geography, and ESRD. Our population-level analysis may not be applicable to individual donors. The use of zip codes to determine median household income is frequent in the medical literature and assumes the same income for individuals living in a given zip code. This assumption may be incorrect, especially in metropolitan areas. Median household income is only one indicator of socioeconomic status and may not directly relate to the financial status of an individual. However, when we examined other metrics at the zip code level, such as education, we found that these were colinear with median household income. Because we examined median household income as a categorical variable rather than a continuous variable, we were unable to account for the variability in income within each income quintile and how this relates to living donation. Data from the 2010 United States Census were used to assign median household income and determine population figures. It is important to note that the lack of individual level data limit our ability to account for changes in median household income over time, particularly during periods of economic stress. We adjusted for the age and race standardized prevalence of ESRD, but are not able to directly adjust for the incidence of other health conditions (i.e., diabetes, heart disease) that may affect living donation. The effect of such conditions would only be accounted for to the extent that they are correlated with the prevalence of ESRD in our study. We have previously reported on the equivalent rate of living donation in black and white Americans in the United States, and therefore we did not specifically examine three-way interactions of ancestry, income, and sex on living donation in this study as these analyses are difficult to interpret and we have shown that black ancestry is not an independent determinant of living donation after accounting for differences in income and ESRD. Specific to black people, recent survey work has shown that United States transplant centers do not routinely currently use APOL-1 testing in their evaluation of living donors,³² and therefore we do not believe the evolving information regarding APOL-1 risk alleles and the risk of CKD affected the use of living related donation in black people in this study. The trends in this analysis may not be applicable for small selected subgroups of donors, such as nondirected anonymous donors. Kumar et al.³⁹ recently reported on nondirected donation in the United States and showed that only 44% of these donors were men. Although we did not separately examine trends in living donation by sex and income among nondirected anonymous donors in this analysis because of the small numbers of such donors, it is unlikely that these donors were key drivers of the overall trends in donation in women and men, given their small numbers.

In summary, we found that changes in living donation in the past decade varied by sex, resulting in a widened gap in donation between women and men and that income may be an important factor contributing to this discrepancy. These findings suggest that strategies to remove financial barriers to living donation as may be important to maintain living donation in the future. The reasons for the decline in living related donation require further study.