Emotional Well-Being of Living Kidney Donors: Findings From the RELIVE Study

Use of validated instruments

The Patient Health Questionnaire (PHQ-9) Depression Scale 23 is widely used to screen for depression in the medical and general population. The questionnaire contains nine items including questions on loss of interest, depressed mood, sleep, appetite and energy changes, low self-worth, difficulty concentrating, psychomotor activity changes and suicidal ideation. For each item, responses are scored between 0 and 3, ranging from “not at all” to “nearly every day”; overall scores can range from 0 to 27. Construct validity and criterion validity have been tested in primary care populations, with a PHQ-9 score ≥10 found to have 88% sensitivity and 88% specificity for clinical depression 23. Scores of 5, 10, 15 and 20 represent cutpoints for mild, moderate, moderately severe and severe depression. Any donor revealing potentially worrisome psychological responses was contacted by a clinician from the relevant center and referred for additional help if deemed necessary.

The Life Orientation Test-Revised 24 (LOT-R) identifies individuals who maintain positive expectations in adversity. This measure utilizes six items representing an optimistic trait or disposition toward optimism. Respondents indicate their agreement with items on a 0–4 scale (“strongly disagree” to “strongly agree”), with overall scores ranging from 0 to 24 and higher scores indicating optimism. Previous studies have demonstrated acceptable internal consistency, convergent and discriminant validity 24. Normative data were available for both medical and nonmedical populations.

The 36-Item Short Form Health Survey (SF-36, version 2) 25, is a validated and extensively used measure of quality of life (QOL) reflecting perceptions over the previous 4 weeks. It is divided into two broad scales of behavioral functioning, the Physical Component Score (PCS) and the Mental Component Score (MCS). MCS was not included in the analysis because of its strong correlation (r = −0.77) with the depression screening instrument, the PHQ-9. Normative scores exist for the general population and patients with medical conditions.

RELIVE study donation-specific questions

Utilizing questions described and used in the kidney donor literature 26, members of the RELIVE consortium developed a set of questions on the donation experience, attitudes about donation and donation-related relationships. Specific topics included, donor recovery time (e.g., “How long after donation surgery did it take you to return to your usual daily activities, such as walking, driving a car or shopping for groceries?” Less than 3 months, 3–6 months, more than 6 months or I never returned to my usual daily activities), psychological care before and since donation, family support, relationship with the recipient and recipient outcome and motivation for donation including moral obligation which has been described as a three step process that includes awareness of the effect of one's actions on the welfare of another person, ascribing responsibility to oneself rather than others and accepting the moral norm at issue 26 (study questions available on request). In a pilot study, the questionnaire was administered to 23 donors who donated within the previous 1–4 years at two sites (Mayo and UAB) to evaluate comprehension, flow and order of questions; modifications were made prior to beginning the full study.

Control subjects for comparison with RELIVE donors with respect to depression outcomes were derived from the publicly available National Health and Nutrition Examination Survey (NHANES) 2009–2010 data 27. NHANES participants were matched to RELIVE donors based on sex, race/ethnicity, marital status, educational attainment, using an iteratively relaxed match on age at the time of completing the survey (donors) or participating in NHANES (up to ±5 years) until at least one match was identified.

Statistical analysis

Descriptive information is reported as frequencies and percentages for sample characteristics, and means and standard deviations for the PHQ-9, LOT-R and PCS distributions. Scale scores were calculated if less than half of the items were missing. Pearson correlations were calculated between depression severity and optimism.

For regression analyses, all missing data were multiply imputed using IVEware (http://www.isr.umich.edu/src/smp/ive/) to generate 10 imputation sets. Thirty-eight donors had imputed values for the LOT-R, and 68 for the PHQ-9. We also imputed missing values in educational attainment at donation (n = 474), relationship to recipient (n = 7), BMI at donation (n = 49), history of psychological difficulties (n = 72) and marital status (n = 46).

We used logistic regression to examine differences between donors who were depressed at the time of questionnaire completion and those who were not, using PHQ-9 scores at or above 10 as the threshold for clinically relevant depression. Logistic regression was also used to examine differences between donors who reported experiencing emotional, psychological or substance abuse difficulties that they perceived to be a result of donation and those who did not, and to examine differences between donors who sought help for such difficulties and those who did not. For each of these three outcomes, models were identified using a best subsets approach, selecting the model with the highest likelihood score statistic in which all covariates were statistically significant at p < 0.05. A comparison of study participants to nonparticipating donors was also performed using logistic regression.

Our study received approval by the following institutional review boards: UAB, IRB approval number X070604010; UMN, IRB approval number 0905M66501; Mayo, IRB approval number 09-001345) and DCCs, IRB approval number CR00032674 and protocol number HUM00004345. Informed consent was provided by each participant.

Cohort demographics and psychiatric characteristics at the time of donation

The study cohort has previously been described 21. Of 6909 donors eligible for the study, 3470 acknowledged contact. Of these, 2455 agreed to participate. Thus, of the original cohort of potentially eligible subjects the response rate was 36% and for those who acknowledged contact (2455 of 3470) 71% completed the questionnaire. There were 3439 who did not acknowledge contact due to nonreceipt of communications or nonresponse to contact efforts. Of those contacted, 931 declined to participate, 5 had language barriers and 79 withdrew consent or did not return the questionnaire. RELIVE donors who were eligible but did not participate (n = 4454) had lower educational attainment at donation, donated more recently, were younger at donation, were less likely to have a history of depression before donation, and were more likely to be black or another nonwhite race, male, separated, divorced or widowed, and related to their recipient (all p < 0.05) compared to donors who did participate.

Of the 2455 participants who completed the psychosocial questionnaire (Table 1a, Table 1b; Supplemental Table S1): 93% were white, 61% were women and 52% were 40 years of age or older at donation. Over 41% were siblings and 87% were related either genetically or by marriage to their recipient. Of note, all categories of biologically related or spousal donors were less likely to have a predonation history of psychiatric difficulties than donors who were not related to their recipients (such as friends, coworkers or anonymous donors; 23% among unrelated donors compared to 8% of parents, 7% of children, 7% of siblings, 11% of spouses and 12% of other related donors) based on medical record review (all p < 0.05).

At donation, medical records indicated that over 90% of the sample had never been on antidepressant or antianxiety medication and had no history of alcoholism; 11% had past or current pain problems (Table 1a). Depression (8%) was the most common preexisting psychiatric condition followed by anxiety (3%), but notably, subjects with bipolar disorder (0.4%) had also donated (Table 1a). On the questionnaire, 11% reported that they had been treated for psychological problems prior to donation.

Psychiatric characteristics following donation

On the questionnaire, 4% noted psychiatric problems at some time after donation; 2% sought treatment, and 1% were in active treatment (Table 2a). Based on scores of 10 or greater on the PHQ-9, 8% of donors reported depressive symptoms at the time of the survey (Table 2b) with few reporting severe symptoms (Figure 1). The distributions of responses to PHQ-9 items are reported in Supplemental Figure S1. RELIVE donors were less likely to be taking medications for psychiatric issues including depression or anxiety (12%) at the time of the survey compared to 15% among matched NHANES controls (p = 0.011) (Table 2b).

Optimism

As a group, donor responses (LOT-R mean (M) = 17.8, standard deviation (SD) = 4.1) were similar to predonation liver donors (M = 17.8, SD = 3.1) and more optimistic than normative data from both a college-based sample (M = 14.3) and medically ill sample (M = 15.2) (Supplemental Figure S2) 24. The LOT-R was negatively correlated with depression, with those more optimistic less likely to report depression (r = −0.54, p < 0.001).

Donor characteristics associated with self-report of depression at the time of survey

Based on logistic regression (Table 3), depression in donors (PHQ-9 ≥ 10) was associated with a predonation history of depression (p < 0.001), longer postdonation recovery time (p = 0.009), greater financial burden (p = 0.013), stronger agreement with the statement “It was my moral obligation to donate” (p = 0.003), and emotional, psychological or substance abuse problems following donation (p = 0.010). After excluding donors who had a predonation history of depression, many of the same characteristics remained predictive of depression on the questionnaire.

The absence of depression symptoms (PHQ-9 ≤ 9) was associated with better physical health (measured as the age- and sex-adjusted PCS from the SF-36, p < 0.001), older age at donation (p = 0.002), higher optimism (measured as higher scores on the LOT-R, p < 0.001), being employed either full-time or part-time at the time of survey (p < 0.001) and being of white or European American race (p = 0.020). Among donors without a history of depression before donation, absence of depression symptoms was associated with better physical health (p < 0.001), higher optimism (p < 0.001) and being employed (p = 0.033).

In a separate analysis examining donation-specific questions, donors indicated whether they had experienced emotional, psychological or substance abuse concerns as a result of donation. Donors seeking help for these issues after donation were more likely to have a predonation history of depression (p = 0.022), and/or to have felt depressed after surgery (p < 0.001) (Table 3), and were more likely to report that their recipient's graft had failed (p = 0.007). Donors who reported emotional, psychological or substance abuse concerns after donation were more likely to have reported history of drug use at the predonation evaluation (p = 0.010), history of chronic pain prior to donation (p = 0.014), feeling that once the surgery was over they did not receive attention (p < 0.001), and were more likely to encounter postdonation re-hospitalization (p < 0.001) or medical complications not requiring hospitalization (p < 0.001). Nonsignificant findings for all models are provided in Supplemental Table S2.

Overall, most donors did not have a history of depression, did not report emotional, psychological or substance abuse difficulties after donation, and did not report depression (PHQ-9 < 9) at follow-up (Figure 2). However, donors who did report depression at the postdonation survey tended to report having difficulties after donation, and also tended to have a history of depression before donation.