Development of the AP-DONOR Score for Psychological Barriers to Organ Donation: A Preliminary Psychometric Study Among Romanian Medical Students

1. Introduction

Solid organ transplantation is an established life-saving treatment for patients with end-stage organ failure. However, the persistent shortage of donor organs remains a major global challenge and continues to limit access to transplantation [1]. Decisions regarding organ donation are influenced by psychological, social, cultural, religious, and healthcare-related factors [1].

Healthcare professionals play an important role throughout the donation process. Communication between intensive care physicians and families can influence consent for organ donation [2]. Family decisions may also be affected by attitudes toward brain death, perceptions of the healthcare team, and concerns surrounding the donation process [3]. Bereaved families therefore require clear information, effective communication, and appropriate psychological support when confronted with decisions concerning organ donation [4].

Psychological factors are particularly relevant to donation attitudes. Death anxiety has been associated with attitudes toward organ donation among nursing students and practicing nurses [5]. Fear, prejudice, and cultural perceptions may also influence willingness to consider donation [6]. These observations suggest that donation decisions cannot be explained by medical knowledge alone [5,6].

Religious and cultural beliefs represent another important dimension. Religious interpretations have been examined in relation to organ donation in Muslim populations in Pakistan [7]. Catholic religious background has also been investigated as a determinant of donation attitudes among African residents in Spain [8]. Awareness and attitudes toward transplantation have similarly been studied in Saudi Arabia, where religious and sociocultural considerations form part of the broader decision-making context [9].

Medical students are a particularly relevant population because their current knowledge and attitudes may later influence professional involvement in transplantation and communication with patients and families. A Turkish study reported generally favorable attitudes among medical students, with more than 70% expressing willingness to donate [10]. In contrast, university students in western China showed substantially lower willingness despite measurable donation-related knowledge [11]. These differences suggest that knowledge alone may not fully explain willingness to donate [10,11].

Professional and experiential factors may also influence donation attitudes among healthcare personnel [12]. In critical-care settings, knowledge of brain death and the ability to support families are central components of the donation pathway [13]. Recent nationwide evidence from Romania identified communication difficulties, educational needs, family consent, and organizational preparedness as relevant components of brain-death determination and organ-donation practice among intensive care physicians [14]. These findings reinforce the need to address communication, emotional preparedness, and professional competence alongside factual knowledge [13,14].

Previous instruments have assessed general attitudes toward posthumous organ donation and commitment to donate [15]. Death anxiety has also been measured using established tools such as the Templer Death Anxiety Scale [16]. These constructs, however, do not specifically isolate apprehension arising from uncertainty about brain death determination, emotional burden, family acceptance, institutional integrity, societal preparedness, and professional preparedness for donation procedures.

To address this gap, we developed the Assessment of Psychological Apprehension toward Organ Donation (AP-DONOR). Its conceptual development was informed by previous work on organ-donation attitudes [15] and death anxiety [16], but no items from those instruments were directly translated or adapted. AP-DONOR was deliberately designed to be brief, easy to administer, and straightforward to score, so that it could be used in educational and research settings without imposing substantial respondent burden.

The instrument contains seven items organized into two conceptually related dimensions. Diagnostic Fear and Anxiety (DFA) addresses concerns regarding the certainty of brain death diagnosis, the emotional burden associated with brain death determination, and difficulties experienced by grieving families in understanding and accepting brain death. Institutional and Social Mistrust (ISM) addresses concerns regarding potential abuse of the donation process, societal acceptance of brain death, societal preparedness for organ donation, and professional preparedness for donation-related procedures.

The primary objective of this Phase I study was to develop a brief AP-DONOR scoring framework and evaluate its preliminary internal psychometric structure and reliability among Romanian medical students. The secondary objective was to examine cross-sectional associations between AP-DONOR dimensions and willingness to donate organs post-mortem and to explore cognitive, sociodemographic, and experiential correlates of willingness. We hypothesized that higher DFA and ISM scores would be associated with lower donation willingness. External and prospective validation in independent populations is reserved for Phase II.

2. Materials and Methods

2.1. Study Design, Setting, and Sample

An analytical cross-sectional observational study was conducted at the Faculty of Medicine and Pharmacy, “Dunărea de Jos” University of Galați, Romania. Second-year medical students were recruited across six successive academic enrollment cohorts between March 2023 and May 2026. The study was designed as the derivation and preliminary internal evaluation phase of the AP-DONOR project rather than as definitive external validation.

All cohorts received the same standardized educational content before survey administration. The questionnaire was completed online 7–10 days after the module. This timing was selected to permit cognitive consolidation while limiting immediate recall effects. Because the survey was administered after the educational exposure, the results represent a post-module cross-sectional snapshot rather than unprimed baseline attitudes or longitudinal change.

2.2. Standardized Educational Context

The standardized module was developed by an interdisciplinary team of intensive care specialists, transplant surgeons, bioethicists, and psychologists. It covered: (1) Romanian legal and clinical criteria for brain death determination; (2) organ procurement procedures, allocation pathways, and consent frameworks; and (3) standardized and empathetic communication with donor families. Content and delivery were kept consistent across cohorts to reduce curricular variability.

2.3. Study Design, Setting, and Sample

2.3.1. Item Generation and Pilot Testing

The AP-DONOR instrument consists of seven statements rated from 0 (Strongly Disagree) to 4 (Strongly Agree). Conceptual domains were informed by the organ-donation attitude literature and death-anxiety literature [15,16], but the instrument is not a translation or adaptation of an existing scale. An interdisciplinary panel drafted items addressing diagnostic uncertainty, emotional burden, family acceptance, institutional integrity, perceived abuse risk, societal preparedness, and professional training. Face validity and semantic clarity were examined through forward–backward translation procedures and pilot testing in 77 third-year medical students who were excluded from the final analytical sample. Quantitative content-validity indices were not calculated and are planned for Phase II.

2.3.2. Factor Structure and Simple Scoring Framework

Exploratory factor analysis identified two dimensions: Diagnostic Fear and Anxiety (DFA; I1–I3) and Institutional and Social Mistrust (ISM; I4–I7). Factor loadings were used to characterize the latent structure, not to impose a complex bedside scoring algorithm.

For practical use, the proposed primary scoring framework uses unweighted sums: DFA Subscore = I1 + I2 + I3 (0–12); ISM Subscore = I4 + I5 + I6 + I7 (0–16); AP-DONOR Total Score = I1 + I2 + I3 + I4 + I5 + I6 + I7 (0–28). Higher values indicate greater apprehension. Mean subscale scores on the original 0–4 metric were also used for group comparisons. Factor-weighted scoring was examined as a sensitivity approach and showed very high concordance with the simple scoring metrics (r>0.94).

For descriptive presentation only, mean subscale scores were categorized as low (<1.50), moderate (1.50–2.50), or high (>2.50). These categories were defined a priori and are not validated diagnostic, prognostic, or screening thresholds.

2.4. Additional Variables and Outcome Definition

Sociodemographic variables included age, gender, marital status, and religious affiliation (Eastern Orthodox vs. other/undisclosed). Cognitive and ethical variables included perceived clarity of brain death, support for a national donor registry and donor cards, and views regarding financial incentives.

The behavioral endpoint was willingness to donate one’s own organs post-mortem following brain death confirmation. Responses of 3 (To a great extent) or 4 (To a very great extent) were classified as Willing; responses of 0–2 were classified as Unwilling/Undecided. This dichotomization was used for exploratory association modeling and does not represent a clinical criterion.

2.5. Data Collection, Ethics, and Anonymization

From an eligible population of 608 students, 585 were invited and 561 accessed the survey. Twelve incomplete submissions were excluded before analysis, leaving 549 participants and a final response rate of 93.8%. Participation was voluntary and uncompensated.

The study was conducted in accordance with the Declaration of Helsinki and the General Data Protection Regulation (GDPR; Regulation (EU) 2016/679). According to the determination of the University Ethics Committee, the study did not require formal ethics committee approval because it consisted exclusively of anonymous online questionnaires and involved no experimental procedures, medical interventions, or physical manipulation of human participants. The study procedures respected the principles of research ethics, confidentiality, anonymity, and protection of participants’ personal data. Electronic informed consent was obtained from every participant before access to the questionnaire.

2.6. Statistical Analysis

Analyses were performed using SPSS version 26.0 and R version 4.2.1. Two-sided p<0.05 was considered statistically significant. Categorical variables are reported as counts and percentages; continuous variables as mean±SD or median (IQR), as appropriate.

Internal consistency was assessed using Cronbach’s alpha. The full cohort was randomly partitioned into a training subsample (n₁=275) for exploratory factor analysis using principal axis factoring with Promax rotation and a validation subsample (n₂=274) for confirmatory factor analysis using maximum likelihood estimation. Sampling adequacy was evaluated with the Kaiser–Meyer–Olkin statistic and Bartlett’s test. CFA fit was assessed using χ²/df, CFI, TLI, RMSEA, and SRMR. Composite reliability (CR) and average variance extracted (AVE) were calculated from standardized loadings. Discriminant validity was examined using the Fornell–Larcker criterion.

Group comparisons used χ² or Fisher’s exact tests for categorical variables and t tests or non-parametric alternatives for continuous variables, as appropriate. The Phase I analysis focused on internal psychometric structure, reliability, and cross-sectional associations with donation willingness. Because this study was designed for score development rather than clinical prediction, no definitive predictive model or screening threshold was claimed.

3. Results

This section may be divided by subheadings. It should provide a concise and precise description of the experimental results, their interpretation, and the experimental conclusions that can be drawn.

3.1. Participant Characteristics

The final sample included 549 students. Mean age was 22.3±4.8 years (range 19–46 years); 79.2% (n=435) were female, 60.5% (n=332) were single, and 91.3% (n=501) identified as Eastern Orthodox Christians. Overall, 77.0% (n=423) were classified as willing to donate their organs post-mortem.

Age, gender, marital status, and religious affiliation were not significantly associated with donation willingness (Table 1).

3.2. Attitudes Toward Brain Death and Organ Donation

Regarding brain-death comprehension, 51.0% (n=280) considered the concept clear but difficult to explain to relatives, 26.2% (n=144) considered it fundamentally difficult for families to understand, and 22.8% (n=125) considered it clear and easy to communicate. Students who perceived brain death as clear were more frequently willing to donate than those who perceived it as unclear (84.6% vs. 62.1%, χ²=31.84, p<0.001).

Religious beliefs were considered an important determinant of donation decisions by 83.1% (n=456). An opt-in national registry was supported by 82.4%, and donor cards by 67.1%. Organ donation after brain death was considered the preferred clinical approach by 68.9%; 65.6% acknowledged a severe emotional burden on certifying physicians; 63.8% considered understanding brain death important for family authorization; and 63.8% considered financial incentives ethically unacceptable (Table 2).

3.3. AP-DONOR Item-Level Differences

All seven AP-DONOR items differed significantly between willing and unwilling/undecided participants (all p<0.001). Unwilling/undecided students consistently reported greater diagnostic fear, emotional concern, family-acceptance difficulty, perceived abuse risk, conceptual concern regarding brain death, societal unreadiness, and concern about inadequate professional training.

Table 3. AP-DONOR item scores by donation willingness.

Item	Willing mean±SD	Unwilling/ Undecided mean±SD	Statistic Test*
I1 False-positive brain-death diagnosis	1.12±0.89	2.08±0.95	t=10.35
I2 Emotional burden on professionals	1.08±0.84	1.95±0.92	t=9.88
I3 Family difficulty accepting brain death	1.25±0.81	1.99±0.91	t=8.52
I4 Abuse/manipulation/trafficking concern	0.98±0.85	1.98±0.98	t=10.92
I5 Brain death not accepted as biological death	0.95±0.80	1.82±0.96	t=9.84
I6 Societal unreadiness	1.31±0.95	2.05±1.04	t=7.45
I7 Insufficient professional training	0.94±0.78	1.72±0.92	t=9.15

¹ *All between-group comparisons in Table 3 had p<0.001.

Table 3. AP-DONOR item scores by donation willingness.

Item	Willing mean±SD	Unwilling/ Undecided mean±SD	Statistic Test*
I1 False-positive brain-death diagnosis	1.12±0.89	2.08±0.95	t=10.35
I2 Emotional burden on professionals	1.08±0.84	1.95±0.92	t=9.88
I3 Family difficulty accepting brain death	1.25±0.81	1.99±0.91	t=8.52
I4 Abuse/manipulation/trafficking concern	0.98±0.85	1.98±0.98	t=10.92
I5 Brain death not accepted as biological death	0.95±0.80	1.82±0.96	t=9.84
I6 Societal unreadiness	1.31±0.95	2.05±1.04	t=7.45
I7 Insufficient professional training	0.94±0.78	1.72±0.92	t=9.15

¹ *All between-group comparisons in Table 3 had p<0.001.

3.4. Subscale Scores and Descriptive Stratification

Across the full cohort, the simple summed DFA subscore was 4.02±2.04 (range 0–12), the ISM subscore was 4.92±2.56 (range 0–16), and the AP-DONOR Total Score was 8.94±4.07 (range 0–28). On the original 0–4 mean metric, DFA was 1.34±0.68 and ISM was 1.23±0.64. Unwilling/undecided participants had higher summed DFA (5.97±2.07 vs. 3.45±1.65), ISM (7.52±2.60 vs. 4.16±2.04), and Total Score (13.49±3.95 vs. 7.61±3.20); all p<0.001.

Descriptive stratification also differed markedly by donation willingness. For DFA, 71.4% of willing participants were in the low tier, 24.8% in the moderate tier, and 3.8% in the high tier; corresponding proportions among unwilling/undecided participants were 28.6%, 52.4%, and 19.0% (χ²=122.45, p<0.001). For ISM, the corresponding distributions were 78.3%, 19.4%, and 2.4% among willing participants versus 34.1%, 50.0%, and 15.9% among unwilling/undecided participants (χ²=138.12, p<0.001). For the AP-DONOR Total Score, low/moderate/high descriptive tiers were 72.1%/24.1%/3.8% among willing participants and 27.8%/52.4%/19.8% among unwilling/undecided participants (χ²=135.67, p<0.001). These tiers are descriptive only and are not validated cut-offs (Table 4).

3.5. Factor Structure, Reliability, and Construct Evidence

EFA in the training subsample (n₁=275) supported a two-factor solution explaining 56.8% of total variance (KMO=0.812; Bartlett’s χ²(21)=642.18, p<0.001). Standardized loadings ranged from 0.49 to 0.62 for DFA and from 0.52 to 0.61 for ISM.

CFA in the independent validation subsample (n₂=274) supported the two-factor structure: χ²=21.32, df=13, χ²/df=1.64, CFI=0.972, TLI=0.955, RMSEA=0.048 (90% CI 0.012–0.081), and SRMR=0.038.

Cronbach’s alpha was 0.74 for DFA, 0.78 for ISM, and 0.72 for the seven-item total. CR calculated directly from the standardized loadings was 0.57 for DFA and 0.66 for ISM. AVE was 0.31 and 0.32, respectively. Thus, internal consistency by alpha was acceptable, whereas CR and AVE remained below conventional benchmarks for stronger convergent evidence. Discriminant separation was supported by the Fornell–Larcker comparison because the squared inter-factor correlation (r²=0.144) was lower than both AVE estimates. These findings support a preliminary two-factor structure but also identify clear targets for Phase II refinement.

The slightly lower Cronbach’s alpha for the total seven-item score compared with the two sub scores is consistent with the intended bidimensional structure of AP-DONOR, in which DFA and ISM represent related but distinguishable constructs (Table 5).

3.6. Cross-Sectional Associations with Donation Willingness

Donation willingness was strongly associated with both AP-DONOR dimensions. Mean DFA was 1.15±0.55 among willing participants and 1.99±0.69 among unwilling/undecided participants (p<0.001). Mean ISM was 1.04±0.51 and 1.88±0.65, respectively (p<0.001). Perceived clarity of brain death was also associated with willingness: 84.6% of students who perceived the concept as clear were willing to donate compared with 62.1% among those who perceived it as unclear (χ²=31.84, p<0.001). These are cross-sectional associations; they do not establish predictive validity, causality, or individual-level screening performance.

Table 6. Principal Phase I associations with donation willingness.

Variable	Willing	Unwilling/Undecided/comparison
DFA mean score	1.15±0.55	1.99±0.69; p<0.001
ISM mean score	1.04±0.51	1.88±0.65; p<0.001
Clear brain-death concept	84.6% willing	62.1% willing when unclear; p<0.001

Table 6. Principal Phase I associations with donation willingness.

Variable	Willing	Unwilling/Undecided/comparison
DFA mean score	1.15±0.55	1.99±0.69; p<0.001
ISM mean score	1.04±0.51	1.88±0.65; p<0.001
Clear brain-death concept	84.6% willing	62.1% willing when unclear; p<0.001

4. Discussion

The principal finding of this Phase I study is that a brief seven-item instrument captured two distinct but related dimensions of psychological apprehension toward organ donation. Both dimensions were strongly associated with donation willingness. The study therefore provides an initial derivation framework for AP-DONOR rather than definitive external validation.

DFA reflects concern about false-positive brain-death determination, the emotional burden on healthcare professionals, and difficulty experienced by families in accepting neurological death. ISM reflects apprehension regarding abuse or trafficking, societal acceptance of brain death, societal preparedness, and professional training. The two-domain structure is clinically and educationally interpretable because it separates concerns that may require different interventions.

Death-related anxiety has previously been associated with less favorable organ-donation attitudes [17]. Fear of personal death can also reduce willingness to commit to post-mortem donation [18]. Multidimensional approaches to organ-donation attitudes further suggest that willingness is not a single construct and may reflect interacting cognitive, affective, and social processes [19]. AP-DONOR extends this literature by focusing specifically on diagnostic apprehension and institutional-social mistrust.

Psychological strain is relevant in medical education more broadly. Burnout and related psychological variables have been documented among medical students [20], and depressive symptoms and distress have also been reported in Romanian medical-student populations [21]. These findings do not establish that burnout or depression causes donation reluctance, but they support the broader rationale for considering affective responses when designing professional education.

Family acceptance is another central component of donation decisions. Qualitative work has shown that perceptions of death, consent, and family meaning shape post-mortem donation decisions [22]. Religious worldviews can also influence donation attitudes [23]. Family refusal studies identify multiple reasons for declining donation, including concerns related to brain death and trust [24]. Conceptual work on transplantation systems likewise emphasizes trust in clinicians, diagnostic accuracy, and allocation processes [25], while qualitative syntheses describe persistent uncertainty and emotional consequences among donor relatives [26].

Professional identity and cultural context further shape donation-related reasoning. Nursing students describe interactions between faith, culture, and professional identity when considering organ donation [27]. Family-centered decision-making remains influential in diverse populations [28]. Medical students may support transplantation in principle while retaining concerns about brain death or donation procedures [29]. Knowledge and favorable attitudes therefore do not always translate into personal willingness [30,31], and similar gaps have been described in other student populations [32]. International meta-analysis confirms substantial variability in medical-student willingness across educational settings [33].

Educational interventions remain promising. Web-based and structured educational programs can improve donation knowledge, attitudes, or willingness [34,35]. At the system level, successful donation after brain death depends on timely donor identification, effective family approach, and institutional processes [36]. Current curricular initiatives also highlight the need for more standardized organ-donation education in medical schools [37]. Knowledge deficits regarding donation and transplantation remain visible even among senior healthcare students [38].

Cross-national evidence reinforces the importance of context. Medical-student attitudes vary across Spain [39], Poland [40], and other settings. Public and student willingness may be shaped by cultural and religious factors, as illustrated in the United Arab Emirates [41]. Recent Singaporean data also show that knowledge and awareness do not eliminate practical communication barriers [42]. Prosocial orientation alone may not fully explain donation behavior [43], while death attitudes have been associated with donor-card possession [44].

Within the present cohort, perceived clarity of brain death was associated with greater donation willingness. This finding is particularly relevant because the survey was conducted after standardized teaching. Apprehension therefore remained measurable even after a common educational exposure. The result suggests that factual teaching may be necessary but not sufficient; learners may also need opportunities to work through diagnostic uncertainty, family communication, ethical concerns, and institutional trust.

The AP-DONOR framework was deliberately designed for simplicity. Seven items can be summed into a 0–28 total score, while the 0–12 DFA and 0–16 ISM subscores preserve domain-specific information. This structure is intended to support rapid use in educational research. It is not yet a validated clinical screening system. The factor loadings, modest CR values, and low AVE estimates show that Phase II should test whether the present items remain stable, whether any item requires refinement, and whether a simple total score provides adequate criterion performance in independent samples.

4.1. Strengths and Limitations

Strengths include the relatively large homogeneous sample, a high response rate, standardized educational exposure, pilot testing, and independent split-sample EFA and CFA. The use of separate exploratory and confirmatory subsamples strengthened the internal structural evaluation while avoiding the stronger claim of external validation.

Several limitations are important. First, this was a single-center cross-sectional study in a predominantly female and Eastern Orthodox cohort, limiting generalizability. Second, survey administration 7–10 days after the educational module may have introduced priming and does not permit estimation of pre-to-post change. Third, willingness was self-reported and was not verified against donor registration or later behavior. Fourth, quantitative content-validity indices and test–retest reliability were not obtained. Fifth, although Cronbach’s alpha was acceptable, CR was 0.57 for DFA and 0.66 for ISM and AVE was 0.31–0.32, so convergent evidence remains limited. Sixth, the descriptive low/moderate/high categories were not externally calibrated. Finally, the present Phase I analyses establish cross-sectional associations and internal structural evidence only; standalone discrimination, calibration, and externally justified thresholds for the 0–28 AP-DONOR Total Score remain to be evaluated in Phase II.

4.2. Phase II Validation Strategy and Implications

Phase II should prospectively evaluate AP-DONOR in independent universities and, where feasible, among other healthcare trainees and professionals. Planned validation should include test–retest reliability, external CFA, measurement invariance across sex, academic centers, and cultural groups, criterion validity against established related measures and objective donation-related behavior, and assessment of calibration and standalone discrimination of the 0–28 total score. Empirically derived thresholds should be considered only after external criterion-based evaluation.

For medical education, the two subscores may provide a practical framework for intervention design. High DFA may indicate a need for structured teaching on neurological death determination, simulation, and difficult family conversations. High ISM may indicate a need for transparent discussion of allocation ethics, regulatory safeguards, organ-trafficking concerns, and institutional accountability. These applications remain educational hypotheses until validated prospectively.

5. Conclusions

This study represents the development and preliminary internal psychometric evaluation of AP-DONOR, a brief seven-item scoring framework designed to quantify psychological and institutional apprehension toward organ donation. Two related dimensions—Diagnostic Fear and Anxiety and Institutional and Social Mistrust—were identified, and both were strongly associated with post-mortem donation willingness. Clearer understanding of brain death was also associated with greater willingness.

The Phase I findings support the feasibility and interpretability of a simple AP-DONOR framework, but they do not establish an externally validated screening instrument or definitive cut-off values. Phase II will determine reproducibility, external factor structure, criterion validity, measurement invariance, calibration, standalone discriminative performance, and empirically justified thresholds in independent populations.