This retrospective study is part of the ELIPSY project, which aimed to evaluate the psychosocial outcomes and quality of life of a long cohort of living kidney donors across six European transplant centers (Spain, Germany, France, Portugal, Sweden, and Turkey). Due to insufficient sample sizes at some sites, the present analysis included data only from the kidney transplant units of the Hospital Clinic of Barcelona (Spain) and Charité University Medicine Berlin (Germany). At both centers, donors and recipients underwent comprehensive pre-donation medical and psychological assessments. Participants were eligible if aged 18 years or older and had adequate language proficiency of the respective country to complete the questionnaires. All individuals who donated a kidney at either center between 1998 and 2014 were contacted initially by telephone. Upon consent and home address confirmation, participants received a mailed study package including information, consent forms, a self-administered questionnaire and a prepaid envelope to return the signed consent form and completed questionnaire. If the questionnaire was not returned within two to 3 weeks, donors received a telephone follow-up reminder. Data collection took place over 2 years. A total of 1,228 potential participants were identified through hospital records. 1154 were contacted as 74 had missing contact details or were living abroad. 63 actively refused participation on the phone. Questionnaires were sent to 1,091 participants and 767 returned the questionnaire (response rate 66.46%). Subsequently, 54 participants (7%) were excluded from analysis because of missing responses resulting in a final sample of 713 participants. No significant differences were observed between included and excluded participants regarding sex, center, time since donation or anxiety rate (PHQ). However, those excluded reported 1.44 lower mean depression rate (PHQ) than those included [t (704) = −2.10, (−2.78, −0.10) p = 0.036] and those included were older than those excluded [t (765) = 2.38, (0.61, 6.39) p = 0.017] with a mean difference 3.50 (1.47). An overview of the enrollment process is shown in Figure 1.

Donors’ psychosocial outcome assessed via the above mentioned self-administered structured questionnaire, incorporated both standardized psychometric instruments and donation-specific items. Standardized tools were selected based on established validity and frequent use in international research, allowing cross-study comparison. The ad-hoc items were developed based on clinical experience and previous research, including the European Living Donation and Public Health (EULID) project [41].

Donors’ quality of life was assessed using the 36-Item Short-Form Health Survey (SF-36) [42]. This instrument comprises of eight subscales: physical functioning (PF; Cronbach’s α = 0.90), general health (GH; Cronbach’s α = 0.74), bodily pain (BP; Cronbach’s α = 0.86), role limitations due to physical health problems (RP; Cronbach’s α = 0.87), social functioning (SF; Cronbach’s α = 0.83), vitality (VT; Cronbach’s α = 0.85), role limitations due to mental health problems (RE; Cronbach’s α = 0.89), and emotional wellbeing (MH; Cronbach’s α = 0.84). A physical component summary (PCS) score was calculated by averaging the first four subscales (Cronbach’s α = 0.93), and a mental component summary (MCS) score was calculated by averaging the latter four subscales (Cronbach’s α = 0.92). For each subscale, item scores were coded, averaged, and transformed into a 0–100 scale, where higher scores indicated better health. A score of 50 serves as the reference value [43].

The Patient Health Questionnaire (PHQ) [44] was employed to assess depression, anxiety, and somatization. The depression module (PHQ-D; α = 0.85) includes nine items, reflecting the nine diagnostic criteria for depression from the Diagnostic and Statistical Manual of Mental Disorders, Fourth Edition [45]. Each item is rated from 0 (not at all) to 3 (nearly every day), producing a total score between 0 and 27. The anxiety module (PHQ-A; α = 0.93) consists of seven items [46], rated on a 3-point Likert scale (0 = not at all, 2 = more than half the days), yielding a total score ranging from 0 to 21. The somatization module (PHQ-S; α = 0.79) includes 13 items [47], rated from 0 (not bothered at all) to 2 (bothered a lot), with a total score ranging from 0 to 26. Scores of 5, 10, and 15 marking the cutoffs for mild, moderate, and severe depression, anxiety and somatization respectively.

Ad-hoc items assessed whether donors assumed a caregiving role for the recipient (“Have you been in charge of the recipient since the donation?” yes/no) and the subjective caregiving burden (“If yes, how much burden was for you taking care of the recipient?” 10-point scale). Additional questions evaluated donors’ subjective feelings of responsibility for the recipient’s illness or death (“Somehow I feel responsible for the recipient’s death or bad condition, please leave unchecked, if that does not apply to the recipient” yes/no).

All statistical analyses were performed using IBM SPSS Statistics, Version 27.0 (SPSS Inc., Chicago, IL, USA). Categorical variables are presented as frequencies (n) and percentages (%), whereas continuous variables are expressed as means (M) and standard deviations (SD). Clinical variables were compared across groups using one-way analysis of variance (ANOVA) with post hoc comparisons, as well as Student's t-test, as appropriate. General linear models were used to examine the associations between donor–recipient relationship type and physical and mental wellbeing. In addition, multiple linear regression analyses were performed to investigate the associations of the caregiving role or responsibility for the recipient’s health with the endpoints of interest (QoL, anxiety, depression, somatization), adjusting for relevant covariates. To account for multiple testing, p-values from the primary multivariable analyses were adjusted using the Benjamini–Hochberg false discovery rate (FDR) procedure. The remaining analyses are reported in the Supplementary Material/Appendix as exploratory findings and are presented without FDR adjustment. To ensure transparency and minimize the risk of selective reporting, both significant and non-significant exploratory results are provided. Statistical significance was set at p < 0.05.

Table 2 demonstrates the link between donor-recipient relationship type on donors’ physical and mental quality of life (PCS, MCS). Although an exploratory ANOVA analysis found a significant relationship between PCS and relationship type, the association was no longer statistically significant after adjusting for confounding variables such as sex, age, months since donation and center. After adjustment, those donated at Hospital Clinic reported higher PCS than those donating at Charité [β = 4.63 (1.37), I (1.93–7.34), p < 0.001]. Older age was related to lower PCS (β = −0.18 (0.07), [−0.31, −0.04], p = 0.010) and longer time since donation [β = −0.06 (0.015), 95% CI [−0.09, −0.03], p < 0.001] was also significantly linked to lower PCS.

A general linear model was conducted to examine the association between type of relationship with the recipient and MCS scores, adjusting for sex, age, months since donation, and transplant center. No significant overall association was observed between relationship type and MCS scores across groups (all adjusted p-values >0.05). Exploratory pairwise comparisons of adjusted means suggested a difference between spouses (M = 75.59, SE = 1.07) and children/other genetically related donors (M = 83.27, SE = 2.89); however, this finding should be interpreted with caution given the absence of a significant overall effect. Controlling the covariates, female donors had lower MCS scores compared with male donors [β = −2.79 (1.41), (−5.56, −0.03), p = 0.048]. Older age at donation was associated with higher MCS scores [β = 0.23 (0.07), (0.10–0.36), p < 0.001], whereas a longer time since donation was associated with lower MCS scores [β = −0.04 (0.015), (−0.07, −0.01), p = 0.014]. In addition, donors recruited from Hospital Clinic had higher MCS scores than those from Charité [β = 3.91 (1.39), (1.18–6.63), p = 0.005]. Results of the independent-samples t-tests and simple regression analyses are presented in Supplementary Tables 3–5.

In 25.4% of cases, donors assumed a caregiving role post-donation, while 68% did not. Neither adjusted or unadjusted analysis found that being a caregiver donor was related to donors’ PCS. When adjusting for covariates the relationship between caregiving role and PCS remained insignificant, though center [β = 5.53 (1.42), (2.74, 8.32), p < 0.001], donor’s age [β = −0.19 (0.07), (−0.32,-0.06), p = 0.004] and time since donation [β = −0.06 (0.02), (−0.09, −0.03), p < 0.001] as covariates were found to be significantly related to PCS. Those donating at Hospital Clinic reported higher PCS than those donating at Charité, whereas older donors and donors who had undergone kidney donation longer ago presented lower PCS scores.

The donor caregiving role for the recipient was linked to lower MCS. Notably, after adjusting for age, sex, center and time since donation, the model explained 8.4% of variance of MCS compared to 1.5% found in an exploratory unadjusted regression.

Even after adjusting for covariates, the burden of caregiving was significantly associated with anxiety rates accounting for 14% of the variance. Alongside, center as a covariate demonstrated a significant association with anxiety, with those donated at Hospital Clinic reporting lower anxiety than those at Charité [β = −1.50 (0.68), (−2.84, −0.16), p = 0.03].

The burden of caregiving was significantly and positively associated with somatization rates (Table 2). After controlling for confounders, the burden of being a caregiver donor was still related to somatization rate accounted for 16.6% of variance. Among covariates sex [β = 2.07 (0.85), (0.39, 3.76), p = 0.016] and months since donation [β = 0.02 (0.01), (0.001, 0.40), p = 0.041] with female donors and longer time since donation being associated with higher somatization rate.

Only 41.6% of donors replied to the question regarding feelings of responsibility for the recipient’s death or health deterioration, with the question having the option of leaving it unchecked if not relevant. In total 3.2% of donors expressed feelings of responsibility, whereas 38.4% did not. Both unadjusted and adjusted analysis confirmed a significant association between feelings of responsibility and depression rates (Table 2). Importantly, the association between feelings of responsibility and depression remained significant even when center [β = −1.17 (0.55), (−2.25, −0.09), p = 0.033] and age [β = −0.06 (0.02), (−0.10, −0.02), p = 0.006] as confounders found to be significantly linked to depression.

This study has several limitations. Its cross-sectional design increases recall bias risk, and the absence of pre-donation data limits our ability to attribute outcome variations specifically to kidney donation. The small number of unrelated donors (n = 9) restricts generalizability for this subgroup. Sample selection bias is another concern, as our study is lacking data on non-responders, and donor with adverse outcomes may be less likely to participate in follow-up [58], likely overrepresenting donors with more favourable outcomes. Finally, underrepresentation of donors from culturally diverse donors due to language barriers limits insights into culturally specific experiences and outcomes. Future studies should pursue more inclusive sampling, accounting for cultural diversity and non-responder profiles.

Another key issue is the lack of standardized psychosocial assessment tools in living donation research. Despite clinical recommendations [59], no consensus or robust evidence on such tools exists, resulting in wide variability in screening practices [60]. In our study, clinical variables were assessed using internationally accepted psychometric instruments. However, their accuracy and adequacy proved ambivalent, as our independent variables failed to consistently predict mental wellbeing. The selection or development of appropriate tools standardized in donation specific outcome measures is therefore essential for future research, a finding consistent with Massey et al. [61].

The findings emphasize the importance of long-term psychosocial follow-up and support for living kidney donors, especially for a specific minority that seems to be affected. Special attention should be given to high-risk groups carrying caregiving roles and associated burden. Additionally, donor concerns and feelings of responsibility related to the recipient’s health should be carefully assessed, and addressed through appropriate counselling interventions. Enhancing donor understanding and providing targeted support are critical to safeguarding donor wellbeing.