Why don’t they want to donate? Cultural and psychological factors influencing the organ donation intention among Hong Kong University students Authors: Wanming Liang, Chi-Shing Tse Published: 2025 ABSTRACT Despite significant advances in transplantation technologies, the global gap between organ availability and demand continues to widen. Hong Kong, in particular, reports notably low organ donation rates among developed regions. Understanding the psychological barriers to donation is essential for developing effective interventions, particularly among young adults who represent a crucial demographic for establishing lifelong donation commitments. The present study examined the complex interplay of psychosocial and emotional factors influencing organ donation intentions among Hong Kong university students. A cross-sectional survey was conducted with 280 university students in Hong Kong, using validated instruments to assess multiple psychological constructs: depressive thinking, altruism, death anxiety, self-efficacy, perceived social support, Big Five personality traits, and Buddhist and Karmic beliefs. Hierarchical regression and moderated mediation analyses were performed to identify direct and indirect predictors for organ donation intentions. Results showed that higher death anxiety (β = −.133, p = .036) and stronger Buddhist beliefs emphasizing bodily integrity (β = −.22, p < .001) were associated with lower donation intentions, while greater self-efficacy (β = .22, p < .001) and perceived social support (β = .28, p < .001) positively predicted donation intentions. Notably, dispositional characteristics including altruism, depressive thinking, personality traits, and Karmic beliefs did not directly predict donation intentions. However, depressive thinking exerted an indirect negative effect on donation intentions through elevated death anxiety, and this mediational pathway was stronger among participants with lower self-efficacy (index of moderated mediation = .035, 95% CI [.0002,.0013]). These findings highlight the critical roles of emotional and situational factors, rather than stable personality characteristics, in determining organ donation intentions among Hong Kong university students. Public health interventions aiming to reduce death anxiety and strengthen self-efficacy may be particularly effective in promoting organ donor registration among young adults in Hong Kong. Future initiatives should prioritize culturally sensitive approaches that address existential concerns while building confidence in donation decision-making. Introduction Organ transplantation remains one of the most transformative medical advances of the twentieth century, offering life-saving treatment for patients with end-stage organ failure [1]. Despite technical progress and rising public awareness, a persistent shortage of donor organs continues to challenge health systems worldwide. In Hong Kong, this problem is particularly acute: as of September 30, 2025, only 409,020 residents out of a population exceeding 7.4 million had registered as potential organ donors, while over 2,000 patients awaited transplants daily [2]. Given that deceased donors constitute the primary source of transplantable organs in Hong Kong, understanding the psychosocial and emotional determinants of posthumous donation intentions is crucial for developing effective interventions to alleviate this shortage. Prior research has identified multiple psychosocial determinants influencing organ donation intentions, including relational ties, religious beliefs, cultural values, personality traits, altruism, family dynamics, trust in healthcare systems, and demographic factors [3–10]. However, few studies have explicitly investigated these factors within Hong Kong’s unique sociocultural context [11]. To address this crucial research gap, the present study aims to provide a deeper understanding of the psychosocial factors shaping organ donation intentions among Hong Kong university students. Psychosocial determinants of organ donation intentions Previous findings on psychosocial predictors of organ donation intentions, such as knowledge, social atmosphere, personality trait, altruism, self-efficacy, family dynamic, and religious belief, remain inconsistent, particularly within Chinese populations. Among these factors, self-efficacy, defined as the confidence in one’s ability to perform specific behaviors successfully [12], has emerged consistently as a significant predictor of donation intentions [13,14]. For example, Japanese university students who expressed greater self-efficacy regarding their ability to complete organ donation procedures were more likely to register as donors [15]. Table 1 synthesizes the primary areas of contradiction identified in prior research, which form the backdrop for the current investigation. 10.1371/journal.pone.0338201.t001Table 1Summary of inconsistent findings in prior research on psychosocial predictors of organ donation intentions.FactorCommonly Held Belief/ Supporting FindingContradictory Evidence from Prior ResearchKey ResourcesKnowledge about Organ DonationA higher level of knowledge and understanding about the donation process logically leads to increased willingness to donate.Multiple studies report finding no significant relationship between the level of donation-related knowledge and the intention to register as a donor.[16–19]Family dynamicsStrongly influence donation decisions in collectivistic cultures.Family views often represent significant barriers or facilitators.[20–22]AltruismAs an inherently altruistic act, organ donation is presumed to be strongly and positively predicted by an individual’s general altruistic disposition.The direct link between altruism and donation intention can become non-significant when other factors, such as social norms or self-efficacy, are statistically controlled.[6,11,14,23,24]Personality TraitsCertain personality traits, such as Agreeableness and Conscientiousness, are expected to be reliable predictors of prosocial behaviors like organ donation.While Agreeableness shows a generally positive correlation, the effect of Conscientiousness is inconsistent across different studies and populations.[10,25]Buddhistic & Karmic BeliefsBeliefs in Karma, which emphasize the consequences of one’s actions, should logically promote prosocial acts like donation to generate positive future outcomes.The specific relationship between Karmic beliefs and organ donation remains underexplored and unclear, while other religious concerns (e.g., bodily integrity) act as significant barriers.[1,26–28] Summary of inconsistent findings in prior research on psychosocial predictors of organ donation intentions. These inconsistencies highlight a critical insight: the decision to donate organs is not the result of a simple, linear calculation. It is a deeply personal and psychologically complex choice influenced by a dynamic interplay of cognitive, emotional, and social factors. A factor that appears decisive in one study may be neutralized or even reversed by another variable in a different context. This underscores the necessity of an integrated approach that examines not just the direct effects of individual factors, but also their interactions. Interplay among emotional and psychosocial determinants: self-efficacy, altruism, death anxiety, and depressive thinking While psychosocial predictors have been extensively studied, emotional determinants such as death anxiety and depressive thinking have received far less attention, particularly in non-Western populations. Death anxiety, refers to negative emotional responses specifically triggered by mortality awareness [14], may influence organ donation in complex ways. According to Terror Management Theory [29], individuals employ psychological defenses to manage existential anxieties. Such defense involves engaging in prosocial behavior like organ donation to achieve symbolic immortality strategies or cultural worldview affirmations. In other words, heightened death anxiety may motivate altruistic actions, such as organ donation, which can serve as a buffer against the discomfort associated with mortality awareness [30]. However, empirical findings remain inconclusive, with studies reporting both negative associations (heightened death anxiety reduces willingness to donate due to avoidance of death-related topics, e.g., [31–33] and no significant relationship [34]. Furthermore, self-efficacy may moderate death anxiety effects. Jessop and Wade [35] indicates that individuals with higher self-efficacy are less susceptible to the negative effects of death anxiety, whereas those with lower self-efficacy were more likely to avoid action. Given the potential moderating role of self-efficacy, the present study examines how self-efficacy interacts with death anxiety to shape organ donation intentions among Hong Kong university students, an area warranting deeper empirical investigation. Depressive thinking, characterized by persistent feelings of sadness, dissatisfaction, or self-critical thoughts, often precedes clinical depression [36]. This study specifically targets young adults who occasionally experience such thoughts but do not meet clinical criteria for depressive disorder. Depressive states often correlate with reduced altruistic engagement and prosocial behaviors [37,38], with helping others promoting self-healing of depression [38]. However, recent studies suggest that depressive experiences may stimulate empathy and altruistic responses via processes akin to posttraumatic growth or “altruism born of suffering” [39]. An illustrative social media post, “I don’t deserve a good life, but others do” [40], reflects how depressive cognition may prompt altruistic behaviors like organ donation. Given these mixed evidence, clarifying how depressive thinking interacts with altruism and death anxiety to influence organ donation intentions is essential. The present study Although prior research has examined these psychosocial and emotional determinants separately, their unique and interactive influences remain poorly understood. To address this gap, the present study investigates the interplay among psychosocial and emotional variables influencing organ donation intentions in Hong Kong university students (aged 18–25), a demographic previously identified as more positively disposed toward organ donation compared to older age groups (aged 31–70) in Chinese contexts [1,41]. The Theory of Planned Behavior [42] and Theory of Reasoned Action [43] provide robust frameworks for investigating organ donation behaviors, positioning individual intentions as key predictors of donation behaviors, influenced by attitudes, subjective norms, and perceived behavioral control [44]. Guided by these frameworks, the current study explores the predictive roles of demographic, psychosocial, and emotional variables in predicting organ donation intentions among Hong Kong university students. Demographic factors, including age, gender, educational background, family socioeconomic status, and personal income, were considered, as they were found to be associated with donation willingness [10,45,46]. Specifically, we address two overarching research questions RQ1: Do psychosocial and emotional factors (knowledge, social atmosphere, personality traits, self-efficacy, altruism, family dynamics, Buddhism/Karma beliefs, depressive thinking, death anxiety) uniquely predict organ donation intentions after controlling for other variables? RQ2: How do depressive thinking, altruism, death anxiety, and self-efficacy interact to influence organ donation intentions? Based on these questions, we derive 15 hypotheses. We expect that organ donation intentions will be higher for those with greater knowledge about organ donation (H1), stronger perceived social support from family and friends (H2), greater social media usage (H3), higher conscientiousness (H4), higher agreeableness (H5), elevated self-efficacy (H6), increased altruism (H7), stronger Buddhist beliefs (H8), stronger Karmic beliefs (H9), healthier family dynamics (H10), higher depressive thinking (H11), and lower death anxiety (H12). In addition to these direct relationships, we propose three hypotheses examining moderation and mediation effects (see supporting information - S1 Appendix or in https://osf.io/ajb65/files/dpax9 for conceptual models). First, altruism is expected to moderate the positive relationship between self-efficacy and organ donation intentions, such that this association will be stronger for highly altruistic individuals (H13). Second, altruism and death anxiety are expected to mediate the relationship between depressive thinking and organ donation intentions (H14). Depressive thinking may enhance altruism, thereby increasing donation intentions, or it may heighten death anxiety, thereby reducing donation intentions. Third, we propose a moderated mediation model wherein self-efficacy moderates the indirect effects of depressive thinking on organ donation intentions through altruism and death anxiety (H15). These indirect pathways would vary according to levels of self-efficacy. Overall, the present study aims to provide a more integrated understanding of how psychosocial and emotional variables jointly shape posthumous organ donation intentions among Hong Kong university students. While prior research on this population has been limited, with one notable study focusing primarily on bodily integrity concerns [47], there remains a lack of updated, multifaceted investigations that consider a broader range of integrated psychosocial and emotional predictors. By addressing this gap, the present study offers timely insights into the complex determinants underlying organ donation intentions among a critical yet understudied demographic, ultimately informing targeted health communication strategies and evidence-based interventions in Hong Kong. Method Participants A total of 280 undergraduate students (74.5% female; Mₐ₉ₑ = 20.00, SD = 1.00) were recruited from the Chinese University of Hong Kong between February and April 2025. Although relying solely on university students may limit generalizability to wider young adult populations, this sample provides a relevant starting point for examining organ donation intentions in detail. A priori power analysis using G*Power [61] indicated a required sample size of 263 to detect a medium effect (f² = 0.15) [48] in a multiple regression model (fixed model, R² increase), with 16 predictors (knowledge, two components of social atmosphere, five personality traits, altruism, self-efficacy, family dynamics, beliefs in Buddhism and Karma, depressive thinking, death anxiety, and the interaction between self-efficacy and altruism), 6 covariates (gender, age, education, individual monthly income, family monthly income, parental education), α = .05, and power = .99. To account for potential attrition or incomplete data, the target sample was increased to 280 participants. Participants received monetary compensation (120 HKD) for their participation. The study received ethical approval from the Survey and Behavioural Research Ethics Committee of the Chinese University of Hong Kong (Ref. No. EDU2025−034). Written informed consent was obtained electronically before participation. A pilot study (N = 20) was conducted to assess questionnaire comprehensibility, and feedback informed minor adjustments to item phrasing and response formats. Materials and procedure Given our focus on depressive thinking, as referred to as negative thought patterns distinct from clinical depression, participants reporting a prior clinical depression diagnosis were excluded. Data were collected via an online questionnaire provided bilingually (Chinese and English). Established measures adapted from prior research included the Family Health Questions and Assignment Criteria [49], Big Five Inventory (BFI-44) [50], and Automatic Thoughts Questionnaire – Negative (ATQ-N) [51]. Other scales underwent rigorous translation and back-translation procedures. Specifically, two bilingual translators independently translated original English items into Chinese. Researchers reviewed these translations to produce a consensus version, and subsequently back-translated into English by two additional bilingual speakers unaware of the original items. Final translations were carefully reviewed for semantic equivalence and cultural appropriateness (full questionnaire available in Supporting information - S2 Appendix, https://osf.io/ajb65/files/6mtk2). Further refinements were implemented based on pilot study findings. The instrument also included demographic items (age, gender, education level, individual and family monthly income, parental education levels, and previous clinical diagnosis of depression) and measures assessing organ donation intentions and registration status, adapted from Chen et al. [1]. Organ donation decision-making was assessed by two items adapted from established measures [1,10,52–54]: (a) intention (“When your life cannot be saved, are you willing to donate organs?”), rated on a 7-point Likert scale (1 = extremely unlikely, 7 = extremely likely); and (b) registration status (“Have you ever registered for organ donation?”), with a binary response (0 = No, 1 = Yes). Psychosocial and emotional factors were measured as follows: Organ Donation Knowledge: Assessed via a 10-item true-or-false scale. Correct responses were summed, yielding scores from 0 to 10, with higher scores indicating greater knowledge (α = .74, [54]. Organ Donation Knowledge: Assessed via a 10-item true-or-false scale. Correct responses were summed, yielding scores from 0 to 10, with higher scores indicating greater knowledge (α = .74, [54]. Social Atmosphere: Included two components: (a) Social interaction/support (6 items adapted from D’Alessandro et al. (8; α = .77), measuring perceived social support and interactions regarding organ donation on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree), with higher scores indicating greater social interaction/support of organ donation; and (b) Social media usage (items adapted from Gong et al. [55]; α = .70), assessing frequency of social media exposure using a 5-point scale (1 = never, 5 = almost every day). Participants also reported the number of days they were exposed to various media in a week (0–7). Scores were summed, with higher totals indicating greater media usage. Social Atmosphere: Included two components: (a) Social interaction/support (6 items adapted from D’Alessandro et al. (8; α = .77), measuring perceived social support and interactions regarding organ donation on a 5-point Likert scale (1 = strongly disagree; 5 = strongly agree), with higher scores indicating greater social interaction/support of organ donation; and (b) Social media usage (items adapted from Gong et al. [55]; α = .70), assessing frequency of social media exposure using a 5-point scale (1 = never, 5 = almost every day). Participants also reported the number of days they were exposed to various media in a week (0–7). Scores were summed, with higher totals indicating greater media usage. Personality Traits: Evaluated with the 44-item Big Five Inventory (BFI), assessing neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness [56,57] on a 5-point Likert scale (α = .70−.80) [50]. Personality Traits: Evaluated with the 44-item Big Five Inventory (BFI), assessing neuroticism, extraversion, openness to experience, agreeableness, and conscientiousness [56,57] on a 5-point Likert scale (α = .70−.80) [50]. Self-efficacy: Three items adapted from Brug et al. [58] assessed perceived ease of organ donor registration, rated on a 5-point Likert scale. Responses were summed, with higher scores denoting greater self-efficacy (α = .87; [11]. Self-efficacy: Three items adapted from Brug et al. [58] assessed perceived ease of organ donor registration, rated on a 5-point Likert scale. Responses were summed, with higher scores denoting greater self-efficacy (α = .87; [11]. Family Dynamics: Nine items adapted from Family Health Questions and Assignment Criteria measured perceived family relationship quality [10,49], using a 5-point Likert scale Higher scores indicated stronger perceived familial trust and connectedness (α = .83; [10]. Family Dynamics: Nine items adapted from Family Health Questions and Assignment Criteria measured perceived family relationship quality [10,49], using a 5-point Likert scale Higher scores indicated stronger perceived familial trust and connectedness (α = .83; [10]. Beliefs in Buddhism: 3 items adapted from Lam and McCullough’s [59] Belief in Buddhism scale was used. Given noted contradictions regarding organ donation within Buddhist teachings, all items were included to capture overall religious beliefs. Items were rated on a 5-point Likert scale with higher scores indicating stronger beliefs in Buddhism. Beliefs in Buddhism: 3 items adapted from Lam and McCullough’s [59] Belief in Buddhism scale was used. Given noted contradictions regarding organ donation within Buddhist teachings, all items were included to capture overall religious beliefs. Items were rated on a 5-point Likert scale with higher scores indicating stronger beliefs in Buddhism. Beliefs in Karma [Beliefs in Buddhism and Karma were retained separately because they represent distinct cognitive–spiritual orientations: Buddhism beliefs emphasize bodily integrity and postmortem peace, whereas Karmic beliefs focus on moral causation and reciprocity. Although related, both have unique implications for donation willingness in East Asian contexts.]: Beliefs were measured using 16 items from White et al.’s [60] Beliefs in Karma [Beliefs in Buddhism and Karma were retained separately because they represent distinct cognitive–spiritual orientations: Buddhism beliefs emphasize bodily integrity and postmortem peace, whereas Karmic beliefs focus on moral causation and reciprocity. Although related, both have unique implications for donation willingness in East Asian contexts.]: Beliefs were measured using 16 items from White et al.’s [60] Belief in Karma Questionnaire, with higher scores indicating stronger beliefs in Karma (α = .94). Altruism: Eleven items adapted from Morgan and Miller [61] assessed altruistic attitudes. Negatively phrased items were reverse-scored. Items were rated on a 5-point Likert scale with higher summed scores indicating greater altruism (α = .84; [62]). Altruism: Eleven items adapted from Morgan and Miller [61] assessed altruistic attitudes. Negatively phrased items were reverse-scored. Items were rated on a 5-point Likert scale with higher summed scores indicating greater altruism (α = .84; [62]). Depressive Thinking: Measured by the 30-item Automatic Thoughts Questionnaire – Negative (ATQ-N) [36] that assesses frequency of automatic negative thoughts characteristic of depressive thinking [51], capturing habitual negative cognitions distinct from clinical depression. Items were rated on a 5-point frequency scale (1 = not at all; 2 = sometimes; 3 = moderately often; 4 = often; 5 = all the time), with higher summed scores indicating more frequent depressive thinking (α = .95). Depressive Thinking: Measured by the 30-item Automatic Thoughts Questionnaire – Negative (ATQ-N) [36] that assesses frequency of automatic negative thoughts characteristic of depressive thinking [51], capturing habitual negative cognitions distinct from clinical depression. Items were rated on a 5-point frequency scale (1 = not at all; 2 = sometimes; 3 = moderately often; 4 = often; 5 = all the time), with higher summed scores indicating more frequent depressive thinking (α = .95). Death Anxiety: Evaluated using the 17-item Death Anxiety Inventory (DAI) [14] rated on a 5-point Likert scale. Higher summed scores indicated greater death anxiety (α = .89). Death Anxiety: Evaluated using the 17-item Death Anxiety Inventory (DAI) [14] rated on a 5-point Likert scale. Higher summed scores indicated greater death anxiety (α = .89). Data analyses All analyses were conducted using IBM SPSS Statistics (Version 29) with the PROCESS macro (Version 4.2) [63]. Prior to analysis, assumptions of normality, linearity, and multicollinearity were examined. Descriptive statistics and bivariate correlations were computed for all study variables. To assess the relative contribution of key predictors to organ donation intention, hierarchical multiple regression analyses were performed. Demographic variables (e.g., age, gender, religious) were entered in Step 1, followed by psychological and emotional variables (depressive thinking, death anxiety, altruism, self-efficacy, and Buddhist/Karmic beliefs) in Step 2. To further test the hypothesized relationships among variables, a series of mediation and moderation analyses were performed using the PROCESS macro. A separate moderation analysis (Model 1) examined whether altruism moderated the relationship between self-efficacy and donation intention, controlling for death anxiety, Buddhist beliefs, and perceived social interaction/support. Model 4 was used to examine whether death anxiety and altruism mediated the effect of depressive thinking on organ donation intention. Model 14 was applied to test a moderated mediation model, in which self-efficacy moderated the indirect effects of depressive thinking on organ donation intention via death anxiety and altruism. All continuous variables were standardized before computing interaction terms. Bootstrapping with 5,000 resamples was used to estimate 95% bias-corrected confidence intervals for indirect and interaction effects, which were considered statistically significant if zero was not included in the confidence interval. Results Demographic characteristics All data, code, and materials are available at https://osf.io/ajb65/. The majority reported family monthly household incomes between 10,000 and 49,999 HKD, with varied parental education levels. Over 80% reported individual monthly incomes below 4,000 HKD (see Table 2 for detailed demographic characteristics). Predictors of Organ Donation Intentions (H1-H12) Organ Donation Intentions (Q1). To identify psychosocial and emotional predictors of organ donation intentions, we conducted hierarchical regression analyses with standardized (z-scored) variables. Step 1(demographic covariates: gender, age, family income, parental education, and personal income) did not significantly predict donation intention, R² = .010, F (6,273)=.437, p = .854. In Step 2, adding psychosocial and emotional predictors significantly improved the model fit, R² = .370, F (21,258)=7.216, p < .001, indicating substantial explanatory power beyond demographics. Multicollinearity was assessed using variance inflation factor (VIF) values. All VIF values in the final model ranged from 1.071 to 1.959, indicating no significant multicollinearity concern. In this model, social interaction/support (B = .10, SE = .02, β = .28, p < .001) and self-efficacy (B = .12, SE = .03, β = .22, p < .001) emerged as strong positive predictors of donation intention, supporting H2 and H6. In contrast, Buddhist beliefs toward organ donation (B = −.14, SE = .04, β = −.22, p = .002) and death anxiety (B = −.015, SE = .007, β = −.133, p = .033) were significant negative predictors, supporting H12. Narratively, these findings suggest that individuals with higher perceived social support and self-efficacy are more willing to donate organs, whereas stronger Buddhist beliefs and higher death anxiety are associated with lower willingness to donate. Correlation and regression coefficients are presented in Tables 3 and 4. 10.1371/journal.pone.0338201.t002Table 2Participants’ demographic variables.CountN %GenderMale6623.6%Female21476.4%Grade LevelYear 16723.9%Year 25921.1%Year 36222.1%Year 47025.0%Year 5227.9%Father’s education levelLower secondary or below9132.5%Higher secondary/ diploma11942.5%Bachelor’s degree4716.8%Postgraduate degree or above238.2%Mother’s education levelLower secondary or below10336.8%Higher secondary/ diploma12845.7%Bachelor’s degree3913.9%Postgraduate degree or above103.6%Family household income<10,000 HKD186.4%10,000–29,999 HKD10637.9%30,000–49,999 HKD8329.6%50,000–69,999 HKD3913.9%70,000–89,999 HKD176.1%>=90,000 HKD176.1%Individual monthly income<1,000 HKD8931.8%1,000–3,999 HKD13849.3%4,000–5,999 HKD4014.3%6,000–7,999 HKD10.4%8,000–9,999 HKD72.5%>=10,000 HKD51.8% Participants’ demographic variables. 10.1371/journal.pone.0338201.t003Table 3Pearson correlation among variables (N = 280). M SD 12345678910111213141516171.Donation Intention5.580.36--2.KlgOD7.561.280.11--3.SocSup18.213.950.43**0.09--4.SocTyp----0.090.040.04--5.SocUse10.521.460.17**0.14*0.090.25**--6.Ext25.613.160.050.020.23**−0.080.16**--7.Agr27.913.490.13*0.040.20**−0.07.014*−.40**--8.Con30.353.20−0.010.13*0.10−0.110.15*0.44**0.31**--9.Neu26.283.08−0.04−0.05−0.07−.0380.09.0370.28**0.30**--10.Ope30.055.210.02−0.040.18**−.045−0.010.42**0.20**0.25**0.14*--11.SelEff7.062.750.44**0.13*0.35**.1090.08−0.07−0.01−0.04−0.14*−0.12*--12.FamDyn32.425.300.15*0.050.29**.0060.18**0.25**0.110.12*−0.12*0.21**0.16**--13.BudOD6.542.14−0.29**−0.19**−0.08−0.15*−0.13*0.070.090.110.15**0.15*−.025**0.02--14.BelKar42.2812.23−0.05−0.12*0.06−0.09−0.080.08.0700.110.090.07−.015.0830.58**--15.Altru36.095.860.23**−0.010.33**−0.010.13*0.49**0.36**0.32**0.040.29**0.18**0.41**0.030.15*--16.DepThi77.8024.89−0.030.01−0.09−0.10−0.12−0.31**0.04−0.050.31**−.175**−0.06−0.42**.044−0.00−0.29**--17.DeaAnx41.1912.43−0.32**−0.05−0.20**−0.08−0.13*0.040.12*0.110.34**0.1−0.36**−0.13*0.38**0.31**−0.040.19**--Note. KlgOD = Knowledge of Organ Donation SocSup = Social interactional/support; SocTyp = Social media type; SocUse = Social media use; Ext = Extraversion; Agr = Agreeableness; Con = Consciousness; Neu = Neuroticism; Ope = Openness; SelEff = Self-efficacy; FamDyn = Family dynamics; BudOD = Buddhist beliefs toward organ donation; BelKar = Kamric beliefs; Altru = Altruism; DepThi = Depressive thinking; DeaAnx = Death anxiety; Correlation is significant at the 0.01 level (2-tailed). *p < .05. **p < .01. ***p < .001 Pearson correlation among variables (N = 280). Note. KlgOD = Knowledge of Organ Donation SocSup = Social interactional/support; SocTyp = Social media type; SocUse = Social media use; Ext = Extraversion; Agr = Agreeableness; Con = Consciousness; Neu = Neuroticism; Ope = Openness; SelEff = Self-efficacy; FamDyn = Family dynamics; BudOD = Buddhist beliefs toward organ donation; BelKar = Kamric beliefs; Altru = Altruism; DepThi = Depressive thinking; DeaAnx = Death anxiety; Correlation is significant at the 0.01 level (2-tailed). *p < .05. **p < .01. ***p < .001 10.1371/journal.pone.0338201.t004Table 4Results of hierarchical regression analysis on psychological and emotional factors influencing organ donation intentions.ModelStandardized Coefficients β t F ChangeR² Change1(Constant)5.38***0.440.01Gender−0.04−0.73Age−0.05−0.75Family household income0.010.12Father’s education level−0.02−0.28Mother’s education level0.020.23Individual monthly income−0.06−1.002(Constant)6.54***9.84***0.36***Gender−0.08−1.54Age−0.04−0.71Family household income0.0110.19Father’s education level−0.03−0.43Mother’s education level−0.03−0.48Individual monthly income−0.07−1.25Organ donation knowledge0.010.19 Social interaction/support 0.28 4.83*** Social media use0.111.944Extraversion−0.05−0.69Agreeableness0.060.96Consciousness−0.07−1.07Neuroticism0.071.14Openness0.030.60 Self-efficacy 0.23 3.80*** Family dynamics−0.01−0.12 Buddhist beliefs toward organ donation −0.22 −3.20** Karmic beliefs0.111.72Altruism0.091.38Depressive thinking0.040.62 Death anxiety −0.13 −2.15* Note. *p < .05; **p < .01; ***p < .001 Results of hierarchical regression analysis on psychological and emotional factors influencing organ donation intentions. Note. *p < .05; **p < .01; ***p < .001 Organ Donation Registration Status (Q2). Only 7.9% of participants (22 of 258) reported being registered organ donors. Due to the small number of registered donors, logistic regression analyses failed to converge. Thus, subsequent analyses focused exclusively on donation intentions rather than actual registration status. Moderation analysis: altruism, self-efficacy, and organ donation intention (H13) A moderation analysis PROCESS Model 1 examined whether altruism moderated the relationship between self-efficacy and donation intention, controlling for death anxiety, Buddhist beliefs, and social interaction/support. The overall model was significant, F(6,273)=22.86, p < .001, R² = .33. However, the self-efficacy × altruism interaction was nonsignificant (B = 0.06, SE = 0.05, t = 1.29, p = .20, 95% CI [−0.03, 0.16]; ΔR² = .004, F(1,273)=1.67, p = .20), indicating that altruism did not moderate the relationship between self-efficacy and organ donation intention. Thus, H13 was not supported. Mediation analysis: roles of altruism and death anxiety in linking depressive thinking and donation intentions (H14) Parallel mediation analyses (PROCESS Model 4; 5,000 bootstrap samples) [63] tested whether altruism and death anxiety mediated the relationship between depressive thinking and donation intention, controlling for self-efficacy, Buddhist beliefs toward organ donation, and social interaction/support. Results showed that depressive thinking predicted lower altruism (B = −.26, SE = .05, t = −4.83, p < .001, β = −.26), and higher death anxiety (B = .16, SE = .05, t = 3.00, p = .003, β = .16). In turn, altruism positively predicted donation intention (B = .11, SE = .05, t = 2.06, p = .005 β = .11), while death anxiety negatively predicted donation intention (B = −.12, SE = .06, t = −2.11, p = .036, β = −.12). The direct effect of depressive thinking on donation intentions was nonsignificant, B = .07, SE = .05, t = 1.30, p = .192, 95% CI [−.035,.172], β = .07. The total indirect effect through altruism and death anxiety combined was significant (effect = −.049, BootSE = .022, 95% CI [−.096, −.001]). However, when examined individually, neither indirect pathway alone reached significance (altruism effect = −.020, BootSE = .019, 95% CI [−.070,.003]; death anxiety effect = −.020, BootSE = .013, 95% CI [−.048,.001]. Overall, the results suggest that depressive thinking indirectly reduced organ donation intention primarily through increasing death anxiety and reducing altruistic motivation, although the direct path from depressive thinking to donation intention was nonsignificant. Moderated mediation analysis: self-efficacy moderating indirect effects of depressive thinking (H15) A moderated mediation analyses (PROCESS Model 14) was tested to examine whether self-efficacy moderated the indirect paths linking depressive thinking to donation intentions via altruism and death anxiety, controlling for social interaction/support and Buddhist beliefs toward organ donation. The overall model significantly predicted donation intention, F(8,271)=21.94, p < .001, R² = .39. Depressive thinking significantly predicted lower altruism (B = −.27, SE = .05, t = −4.86, p < .001) and higher death anxiety (B = .16, SE = .05, t = 3.02, p = .003), but did not directly predict donation intentions (B = .07, SE = .05, t = 1.30, p = .194). Altruism positively predicted donation intentions (B = .13, SE = .05, t = 2.29, p = .041), while death anxiety did not directly predict intention in this moderated model (B = −.07, SE = .06, t = −1.16, p = .247). Self-efficacy directly predicted intention (B = .23, SE = .05, t = 4.27, p < .001). Importantly, the death anxiety × self-efficacy interaction was significant (B = .22, SE = .04, t = 4.95, p < .001, 95% CI [.130,.302]), contributing a significant incremental variance (ΔR² = .055, F(1,271)=24.55, p < .001). In contrast, the altruism × self-efficacy interaction was not significant (p = .116). This suggests that the moderating effect of self-efficacy was specific to the pathway involving death anxiety. Probing the significant death anxiety × self-efficacy interaction showed that the indirect effect via death anxiety was significant only at low level of self-efficacy (−1SD, Effect = −.046, 95% CI [−.0911, −.0118]), but not at mean or high level (+1SD). Thus, self-efficacy serves as a buffer against the detrimental impact of death anxiety on donation intentions. The index of moderated mediation was significant for death anxiety, Index = .035, 95% CI [.011,.065], confirming moderation by self-efficacy only along the death anxiety pathway and partially supporting H15 (see Figs 1–2). No moderation was observed via altruism (Index = −0.02, 95% CI [−.057,.017]). Moderated mediation analysis of depressive thinking on organ donation intentions through death anxiety, moderated by self-efficacy. Self-efficacy as a moderator between death anxiety and organ donation intentions. In sum, our findings highlight death anxiety and self-efficacy as critical factors shaping organ donation intentions among Hong Kong university students. Depressive thinking reduces willingness to donate indirectly by elevating death anxiety, while high self-efficacy mitigates this adverse effect. These results underscore the importance of strengthening individual self-efficacy to counteract existential anxieties and promote prosocial health behaviors. Discussion Summary of major findings The present study examined psychosocial and emotional determinants of organ donation intentions among Hong Kong university students, emphasizing the mediating roles of altruism and death anxiety, and the moderating role of self-efficacy. Aligning with Morgan and Miller’s [61,64] organ donation framework, our findings indicate that organ donation intentions are not solely rational decisions but are substantially influenced by emotional factors, particularly death anxiety. Heightened death anxiety was associated with reduced intentions to donate organs [32,65]. Self-efficacy emerged as a robust positive predictor, both directly increasing intentions and buffering against the negative effect of death anxiety. Social support also significantly predicted higher donation intentions, reflecting the importance of interpersonal validation in Hong Kong’s relational-oriented culture. In contrast, altruism and knowledge did not directly predict intentions, suggesting that prosocial disposition or factual understanding alone may be insufficient to overcome existential and cultural barriers. Interpretation using established frameworks Drawing on Terror Management Theory [49], contemplation of organ donation may activate existential fears, eliciting avoidance responses that reduce willingness to donate. Cultural beliefs regarding bodily integrity and reincarnation, particularly within Buddhist frameworks, further influence donation intentions. While younger cohorts in Hong Kong show increasing acceptance of organ donation, our findings indicate that culturally ingrained concerns about bodily completeness remain salient, consistent with prior work on Chinese populations [47]. Our results also revealed a negative relationship between Buddhist beliefs emphasizing bodily integrity and organ donation intentions. The measure explicitly highlighted beliefs regarding maintaining bodily intactness after death and implications for reincarnation (beliefs about the next life). While prior work indicated a gradual shift toward greater acceptance of organ donation in Hong Kong and traditional beliefs regarding bodily integrity were gradually waning [66], our findings suggest that culturally ingrained beliefs about bodily integrity and reincarnation persist among younger cohorts, particularly regarding afterlife concerns [66–68]. Similar concerns regarding reincarnation and bodily completeness were observed among Chinese American populations [69], underscoring cross-cultural consistencies in this barrier. Corroborating previous findings [6,10], perceived social support significantly associates with greater donation intentions. In Hong Kong, a society where interpersonal relationships and social harmony are highly valued, endorsement from family and peers likely provides essential validation, facilitating openness toward donation. These findings highlight the importance of fostering supportive interpersonal environments and open dialogue as integral components of organ donation promotion strategies [70,71]. Contrary to our hypothesis and previous findings [14,61,72], altruism neither directly predicted donation intentions nor moderated the relationships examined. This unexpected result challenges the view that prosocial dispositions reliably translate into willingness to donate, particularly in contexts involving existential concerns or bodily interventions. Nevertheless, our findings align with emerging perspectives suggesting altruistic traits alone may be insufficient to motivate donation intentions when emotional barriers such as death anxiety are salient [73,74]. Similarly, factual knowledge about organ donation did not predict donation intention, despite generally high knowledge levels in our sample (M = 7.56 out of 10). This finding diverges from earlier studies linking increased knowledge to greater willingness [6]. The limited variance (SD = 1.28) in knowledge scores among our educated sample likely attenuated its predictive power. Importantly, this highlights the persistent knowledge-action gap, suggesting that informational interventions alone may inadequately address underlying emotional and existential barriers, such as death anxiety. Further, personality traits measured by the Big Five Inventory were unrelated to donation intentions, contrasting with prior research identifying agreeableness and conscientiousness as significant predictors [75]. Similarly, while Buddhist beliefs significantly predicted donation intention, karmic beliefs showed limited effects. These observations suggest situational and emotional factors may overshadow dispositional personality traits and general religious worldviews in predicting donation intentions within this demographic and cultural context. Hence, intervention strategies emphasizing context-specific psychological determinants, particularly self-efficacy and social support, may be more effective than those focused exclusively on stable personality or broad religious beliefs. Depressive thinking indirectly reduced intentions via heightened death anxiety and diminished altruism, neither pathway alone reached full statistical significance. However, the indirect effect through death anxiety approached significance, corroborating prior evidence suggesting depressive cognition may exacerbate existential fears, thereby inhibiting prosocial action [76]. These findings underscore the complexity of emotional responses to mortality in decisions about death-related prosocial behaviors. Supporting this idea, Blackie and Cozzolino [77] found that mortality salience could enhance prosocial intentions, such as blood donation, yet noted this effect diminishes with avoidance tendencies common in depressive states. However, Zeng and Tse [78] argued that stable cultural orientations and relational self-esteem more strongly moderate mortality salience effects on prosocial behaviors than immediate emotional reactions. Xiao et al. [79] found that Chinese participants did not show a reduced willingness to donate organs under mortality salience, diverging from findings in Western samples where avoidance and defensive reactions to death reminders often lead to lower prosocial intentions [80]. Together, these mixed findings highlight the need for deeper exploration of how stable traits and cultural worldviews interact with existential emotions to influence organ donation intentions. Self-efficacy emerged as a robust positive predictor of donation intention, affirming its critical role in facilitating health-related behaviors [14]. Increased self-efficacy may enhance individuals’ perceptions of agency and control over end-of-life decisions, thereby mitigating existential anxieties and encouraging a proactive approach to organ donation. Indeed, higher self-efficacy was associated with lower death anxiety, r = −.36, p < .001, reflecting a sense of personal empowerment and reduced vulnerability towards existential concerns [81]. More importantly, self-efficacy significantly moderated the negative impact between death anxiety and donation intention: individuals with high self-efficacy appeared buffered against the detrimental impact of death anxiety, highlighting its protective role. Public health and practical implications Our results emphasize several modifiable factors that can inform intervention strategies. First, enhancing self-efficacy appears critical. Individuals with higher self-efficacy are more confident in their ability to make informed decisions about organ donation and are less influenced by existential anxieties. Intervention programs could include skills-building workshops that educate students about organ donation procedures, clarify misconceptions, and guide participants through decision-making exercises. For example, interactive simulations or decision aids could help individuals practice making donation-related choices, thereby strengthening their sense of control and personal agency. Moreover, campaigns could emphasize stories of peers successfully registering as donors to model positive behaviors and reinforce efficacy. Second, fostering supportive social environments is essential, especially in culturally relational contexts like Hong Kong where family and peer opinions carry considerable weight [82]. Programs could encourage family discussions about organ donation, incorporating culturally sensitive materials that respect beliefs about bodily integrity and afterlife concerns while providing clear explanations of donation processes. Peer-led initiatives and mentorship programs could further normalize donation as a prosocial behavior, taking advantage of the influence of social networks to create a sense of collective endorsement. Public health campaigns could combine these approaches, integrating informational content, emotional support strategies, and social endorsement, to reduce fear, mitigate existential anxiety, and increase willingness to donate. By targeting these modifiable psychosocial factors, interventions can move beyond simple knowledge dissemination toward empowering individuals and creating socially supportive contexts that facilitate actual registration and donation behaviors. Limitations and future directions Despite its contributions, this study has limitations. First, its cross-sectional design precludes causal inferences. Future longitudinal or experimental studies are necessary to determine whether reducing death anxiety or increasing self-efficacy causally enhances organ donation intentions and actual registration behaviors. Second, reliance on self-report measures introduces potential biases, such as social desirability or recall inaccuracies. Future research could integrate behavioral tasks, implicit measures, or experimental manipulations to validate and extend present findings. Third, our mediation and moderation analyses focused primarily on depressive thinking, death anxiety, altruism, and self-efficacy. Future research should explore other potential mediators (e.g., empathy, relational self-esteem) linking personality traits to donation intentions [75]. Finally, the substantial intention-action gap warrants further investigation into psychological, social, and contextual factors hindering translation of positive intentions into donor registration behaviors. Importantly, these limitations also present opportunities for applied interventions and policy initiatives. For instance, public health campaigns or university-based donation drives could specifically aim to reduce death anxiety, enhance self-efficacy, and leverage peer and family support to facilitate donor registration. Cross-cultural comparisons with regions exhibiting higher donation rates (e.g., Spain or the United States) could further identify systemic and cultural facilitators, such as opt-out policies, that might be adapted to the Hong Kong context. In conclusion, addressing these limitations offers a roadmap for translating theoretical insights into practical strategies that increase organ donation rates. Supporting information Conceptual models of the study. (PDF) Full questionnaire of the study. (PDF)