Ageing populations and a higher incidence of chronic conditions with advanced age have resulted in increasing numbers of older patients with end-stage renal disease (ESRD) [1, 2]. This trend is supported by a growing group of older adults considered eligible for and undergoing kidney transplantation (KT) [3–6]. According to records from the Swiss Transplant Cohort Study (STCS), 21% of all KT recipients in Switzerland—where there is no age limit prohibiting access to KT—are ≥65 years of age at time of transplantation [7]. In this context, age always refers to chronological age, i.e., the age counted in years since date of birth. Recently published guidelines recommend considering all patients with chronic kidney disease who are likely to progress to ESRD for KT regardless of their age [8]. KT is considered the preferred treatment option compared to hemodialysis, as it provides better results in terms of survival, cost effectiveness and patient reported quality of life [8–11]. The demand for KT at the same time significantly exceeds the number of available donor organs, thus, studies focusing on predictors for outcomes in older KT recipients present an important research area to support clinical decision and policy making.

Older patients often present with conditions such as disability, functional and cognitive decline and increased numbers of comorbidities such as cardiopulmonal diseases, diabetes or cancer, which can result in adverse health outcomes. Most studies point at an increased post-KT mortality in older KT recipients, an expected finding when comparing outcomes with study results from the general population [12–16]. On the contrary, a number of studies reported mortality rates similar to or lower than in adults of younger age [17–19]. Moreover, patients undergoing KT show a lower mortality risk compared to similar patients remaining on the waitlist and on dialysis [3, 20–22]. Inconsistencies also exist for graft loss, with studies showing higher rates in the older cohort [14, 17, 23] or alternatively a non-significant or protective effect by increasing age [15, 18, 24, 25].

Further, to better understand the effectiveness of healthcare interventions, the inclusion of patient reported outcome measures (PROMs) in addition to more commonly studied biomedical outcomes such as mortality and graft loss, is increasingly being acknowledged in transplant research [26]. In KT PROMs, like quality of life and self-rated health status, have been found to improve pre- to post-transplant in all age groups. Prospective longitudinal data from larger data sets, however, are scarce [9, 26]. Thus, studies in KT that include PROMs to better evaluate health outcomes over time are needed.

In previous studies, two methodological limitations in the field of KT point to the need to improve applied methods in future research. First, age has been frequently used as a categorical variable to facilitate interpretation of study findings, with varying age cut-offs across studies, thus assuming non-linear relationships. However, no one has investigated whether this holds true and where such a relationship would divert from linearity. Second, mortality and graft loss in KT are commonly analyzed using standard survival analysis (e.g., Kaplan-Meier survival curves and Cox’ proportional hazards regression). These methods only take into account one type of outcome per analysis, whereas KT recipients are simultaneously at risk for several adverse events. When graft survival is analyzed, a patient can experience death with a functioning graft, without altering the probability of graft loss, typically resulting in overestimated outcome probabilities [27].

Clinicians and policymakers have to rely on a limited body of evidence to guide organ allocation as well as pre- and post-KT management for older recipients. Thus, prospective multi-center research is essential to provide insights regarding causal relationships between patient’s age and post-KT outcomes, with potential for generalizability [5, 6, 8, 11, 20, 28]. In particular, since age is still associated with lower odds to be waitlisted for and access to KT [12, 24, 29]. The aim of this study was to analyze the relationship between age at time of KT with mortality, graft loss incidence and post-transplant self-rated health status in adult KT recipients while controlling for bio-psychosocial risk factors.

The objective of this prospective nationwide cohort study was to analyze the relationship between age at time of KT with mortality, graft loss incidence and self-rated health status in adult KT recipients. Age at the time of KT predicted mortality in a linear fashion but a non-linear relationship between age and graft loss was detected. Our analysis indicates that by taking into account the competing risk of mortality in estimating probabilities of graft loss, the risk of both outcomes is fairly independent of each other. Thus, graft loss probabilities can be reasonably well estimated using Cox’ regression analysis without applying a competing risk analysis. It should be noted that only 2.7% of our sample experienced both outcomes, and a higher overlap may result in larger differences between the two analysis methods. The self-rated health status during the assessment period from pre-KT up to 8 years post-KT showed higher scores at all follow-up time-points compared to the pre-KT status, regardless of their age. No relationship between age at time of KT and post-KT self-rated health status was found, age therefore did not predict this outcome.

Our results support those of previous studies that reported increased mortality after KT in older recipients compared to younger ones; however, this is consistent with data from the general population [12–16]. Whereas the linear relationship between age and patient survival in the current study does not back the assumptions of previous studies that stepwise mortality risk changes across age groups exist. The result of our analysis does hence not support the use of age as a categorical variable to interpret study findings of patient survival in KT. In contrast, the nonlinear prediction of graft loss by age at the time of transplantation does not reflect conclusions of studies that found a linear increase in post-KT graft loss with older patients [14, 17, 23]. Patients between 35 and 55 years of age presented with a lower risk of experiencing graft loss in our study, whereas older and younger recipients showed a higher probability. Pre-transplant drug non-adherence is a proposed factor that can negatively influence adherence to immunosuppressive regimen in transplant candidates. Several studies reported that non-adherence to the immunosuppressive regimen has a negative effect on graft and patient survival in the population of KT recipients [36–38, 48–51]. Evidence shows that younger adults are at greater risk for drug non-adherence in KT [37]. Concurrently, mild cognitive impairment and the presence of additional comorbidities are common in ESRD and KT recipients. They are also found to be associated with older age in these populations [49, 50]. Furthermore, mild cognitive impairment is associated with decreased medication adherence as well as health literacy in KT recipients [51]. This evidence may support our findings that middle-aged adults after KT have a lower risk of experiencing graft loss than their younger and older counterparts.

Chronological ageing alone has been described as an inaccurate representation of patients’ functional ability and individuals of similar age can show diverse physical and cognitive conditions [28, 52]. Biological age in turn was found to be a strong independent predictor for adverse health outcomes such as mortality and graft loss in KT recipients [5, 8, 28, 52]. Physical frailty is currently proposed as an indicator for biological age [5, 8, 28]. The inclusion of frailty measurements to determine the biological age of a KT recipient could hence be a valuable addition to the single determination of age counted in years in the KT population to predict adverse health outcomes. Relevant associations and organizations increasingly acknowledge the importance of the inclusion of frailty assessments in clinical practice guidelines for evaluating and managing candidates for KT [5, 8, 28, 53].

Besides biological age, psychosocial factors can independently predict poor post-KT outcomes and are increasingly valued in transplant research [31, 54]. International transplant societies endorse a comprehensive bio-psychosocial evaluation prior to transplantation and include them in their clinical guidelines [14, 20, 28]. With a low median HADS score of 4 points, our study participants reported fewer depressive symptoms than described in other studies [33, 55]. However, our study showed that 28.6% of the KT recipients were non-adherent to their medication before transplantation and 19.6% were smoking. These figures reflect the results of previous research [32, 34, 56] but only current smoking status was determined as a significant covariate for the mortality outcome event in our sample. No other psychosocial covariate was identified as significant in our analysis. These results could be due to the fact that in our current study only a limited number of psychosocial factors could be considered as covariates, since routine data collection of a comprehensive set of variables has only been added more recently. The STCS Psychosocial Questionnaire is self-administered and not conducted as a face-to-face interview.

Regardless of age, the self-rated health status during the whole follow-up period was rated notably higher post-KT. This shows that the effect of the intervention from a patient perspective was influencing their health status positively in a sustainable way and therefore KT presents a longtime advantage compared to the pre-transplant status. This finding concurs with previous smaller studies over shorter time periods showing an increase in quality of life and self-rated health post-KT [9, 26] and can be used in clinical practice for counselling particularly of older potential KT recipients. To include the patient perspective on health outcomes by assessing PROMs such as self-rated health status in the pre-KT evaluation and decision making should therefore be considered.

The strengths of this extensive study are its longitudinal, prospective design as well as the application of competing risk analysis. The nationwide multi center design in a European setting including a comprehensive sample of KT recipients with an extensive follow-up time, provides insights regarding relationships between patient’s age and the post-KT outcomes of patient and graft survival. The application of competing risk analysis allows the prognosis of graft loss in the presence of mortality as a competing risk. Despite its’ strong and rigorous study design, a notable weakness of this study is that only 2.7% of our sample experienced both outcomes of mortality and graft loss. A higher overlap may result in larger differences between the competing risk and standard survival analysis. Thus, in samples with a higher overlap, the probabilities of graft loss may not be sufficiently well estimated if only Cox’ regression analysis is used without applying a competing risk analysis. A further weakness of our study is that we assessed only a limited number of pre-KT psychosocial factors.

This study revealed that age at the time of KT predicted mortality in a linear fashion concurring with records from the general population in the same country. In contrary, a non-linear relationship between age and graft loss was detected showing that KT recipients aged between 35 and 55 years presented with the lowest risk of experiencing a graft loss event. Taking into account the competing risk of mortality in estimating probabilities of graft loss, the risk of both outcomes was fairly independent from each other. Thus, graft loss probabilities can be estimated using Cox’ regression analysis. Self-rated health status during the follow-up period was indicated notably higher post-KT, regardless of age. No relationship between age at time of KT and post-KT self-rated health over the entire follow-up time was found. Therefore, age alone seems to be an inaccurate measure to guide risk prediction in KT. This underlines the importance of exploring further aspects such as biological age as a valuable addition to existing KT-guidelines aiming to provide pre-tailored and effective guidance particularly as individuals of similar age can show substantially diverse conditions.

The numbers of older patients considered eligible for and undergoing kidney transplantation are increasing. However, the effect of age at time of transplantation on health outcomes in kidney transplantation remains inconclusive. The objective of our study was to analyze the relationship between age at time of kidney transplantation with mortality, graft loss and self-rated health status in adult kidney transplant recipients. We used data from the prospective Swiss Transplant Cohort Study and included data of 2366 kidney transplant recipients who received a single-organ kidney transplant between 2008 and 2017. Age at transplantation linearly predicted mortality. It was also predictive for graft loss, though nonlinearly, showing that recipients aged between 35 and 55 years presented with the lowest risk of experiencing graft loss. Self-rated health status during the follow-up period was indicated notably higher post-transplantation, regardless of age. No relationship of age with self-rated health status was detected. Therefore, age alone seems to be an inaccurate measure to guide risk prediction and clinical decision making in kidney transplantation. This underlines the importance of exploring further aspects such as biological age including cognition, psychosocial factors, PROMs and physical functioning as a valuable addition to existing KT-guidelines aiming to provide pre-tailored and effective guidance particularly as individuals of similar age can show substantially diverse conditions.