Demographic and prosocial intrapersonal characteristics of biobank participants and refusers: the findings of a survey in the Netherlands

Genetic and biomedical research depend on voluntary sharing of personal data, while rising data demands have increased privacy and security concerns that may suppress willingness to participate. Prior work links biobank participation intent to demographic factors (e.g., higher education, having a partner, being non-religious, better self-reported health), though age and gender effects are inconsistent. Psychological characteristics—particularly prosocial orientations like societal trust, social engagement, and donor registrations—show more stable positive associations with willingness to participate. Domain-specific trust and attitudes toward biobanking and biomedical research also matter. Psychological traits may better explain explicit refusal than demographics, with refusal associated with concerns about privacy, potential stigma or discrimination, and commercial use of data. Trust reduces perceived risk and decision complexity, suggesting that competing values and trust dynamics shape participation thresholds. This study compares individuals who participated versus those who explicitly refused participation in a Dutch population-based biobank to assess the independent roles of demographic and prosocial intrapersonal characteristics in participation decisions.

The paper synthesizes evidence that demographic correlates (education, partnership status, non-religiosity, better self-reported health) often predict biobank participation intentions, whereas age and gender findings are mixed. Psychological factors consistently predict willingness: higher societal trust, social engagement, and prior donor registrations; stronger domain-specific trust in biobanks and positive attitudes toward biobanking and biomedical research. Prosocial motives frequently underlie participation decisions, and similar patterns appear in other cooperative domains (e.g., voluntary work, smart energy systems), with societal trust and domain-specific values (e.g., environmental concern) as robust predictors. Psychological characteristics may better distinguish refusers from non-respondents than demographics, with refusal associated with concerns about health-related stigma, discrimination risks, and commercial data use. Trust functions to reduce perceived risks and complexity in decisions. These insights motivate examining both demographic and prosocial intrapersonal characteristics concurrently to explain participation and refusal.

Design and setting: Cross-sectional online survey conducted in the Northern provinces of the Netherlands in 2018. Samples: Total n = 3019 comprising biobank participants (n = 2615) and refusers (n = 404). Participants were recruited via the Lifelines cohort (in May 2018) with a 22.2% response rate and no incentives; Lifelines originally recruited a representative sample (n = 167,000) from 2006–2013. Refusers were recruited via the DirectResearch EUpanel (n = 30,600; August 2018) with small financial incentives and a 41.4% response rate. Both sources applied similar recruitment protocols and stratified for sex and age. Definition of refusers: From panel respondents who indicated non-participation in large-scale medical scientific studies (n = 1738), refusers were those affirming “I did not want/would not want to participate in this type of research” (n = 404). Sample characteristics: Mean age participants 56 (SD ≈ 15.88), 50.5% male; refusers mean age 56 (SD ≈ 14.18), 48.5% male. Measures: - Demographics and self-reported health: age, gender, marital/registered partner status, education (low/moderate/high), religiosity, residence (urban/rural), work status (as SES indicator), self-reported general health (1 very poor to 5 very good). - Prosocial orientation and values: Actual prosocial behaviors (organ donor status, blood donor status, frequency of charitable donations). Social Value Orientation (SVO) using 6 key items (α > 0.90). Values assessed with 11 items based on basic values theory and adapted Environmental Personal Values Questionnaire (E-PVQ) including subscales: hedonic (individual pleasure), biospheric (environmental concern), and added healthspheric (health/healthcare concerns); reliabilities α > 0.80; intercorrelated subscales. - Societal trust: Six items (5-point Likert) reflecting trust in government and other citizens (competence, good intentions, trustworthiness), α = 0.86. - Domain-specific trust in data management and handling (TDM&H): Five items on trust in organizations (hospital, government, university, large commercial enterprises, SMEs) and four items on trust in specific researchers (hospital, university, market, polling). Principal component analysis (oblimin rotation) yielded three factors explaining 71.90% variance: (1) TDM&H/government (4 items; 45.74%; α = 0.83), (2) TDM&H/commercial enterprises (2 items; 14.24%; α = 0.65), (3) TDM&H/commercial researchers (2 items; 11.92%; α = 0.47); low-to-moderate inter-factor correlations (r < 0.40). Analysis: Descriptive statistics with chi-square and independent t-tests. Multivariate binary logistic regression to predict participation vs. refusal with manual entry strategy, including psychological predictors while controlling for demographics. Assumptions checked: no multicollinearity (VIF < 10), Box–Tidwell tests for nonlinearity with adjustments where improving fit, adequate sample size per Peduzzi rule-of-thumb; analyses in SPSS v25.

Descriptives and group differences: - Demographics and health: Participants more often had a registered partner (79.5% vs. 66.6%, p < 0.01); employment showed a positive trend (55.8% vs. 50.7%, p = 0.07). Participants reported better general health (mean 3.97 vs. 3.71 on 1–5 scale, p < 0.01). Refusers were more highly educated (high: 46.8% vs. 41.4%, p < 0.01), more urban-dwelling (25.2% vs. 20.5%, p < 0.01), and slightly less religious (34.7% vs. 37.2%, p < 0.01). - Prosocial behaviors and orientations: Participants donated to charities more frequently (2.68 vs. 2.44, p < 0.01), and were more often blood donors (11.7% vs. 5.4%, p < 0.01) and organ donors (55.8% vs. 36.7%, p < 0.01). Participants had higher SVO scores (32.25 vs. 29.94, p < 0.01). Value priorities: participants scored higher on healthspheric (3.99 vs. 3.73, p < 0.01) and hedonic values (4.25 vs. 4.09, p < 0.01), but lower on biospheric values (3.76 vs. 3.83, p < 0.01). - Trust: Participants reported higher societal trust (3.25 vs. 3.02, p < 0.01). They trusted government data management and handling more (TDM&H/government: 3.82 vs. 3.62, p < 0.01). Trust in commercial enterprises’ data handling did not differ (2.73 vs. 2.74, p = 0.81). Participants trusted commercial researchers’ data handling less (3.34 vs. 3.44, p < 0.01). Multivariate logistic regression (participation vs. refusal): Model fit χ²(26) = 615.414, p < 0.01; Pseudo R² = 0.40. Prosocial intrapersonal characteristics primarily explained differences; demographics added limited explanatory power. - Significant demographic predictors: Having a registered partner (positive); lower education (low vs. moderate/high) increased odds of participation; high vs. moderate/low had a smaller but significant effect; residence non-significant after adjustment (OR = 0.76, p = 0.10); higher self-reported health predicted participation (OR = 1.28, p = 0.02). - Values and behaviors: Healthspheric values were the strongest predictors of participation. Blood donor status OR = 2.31 (p < 0.01), organ donor status OR = 1.94 (p < 0.01). Lower biospheric values among participants remained robust. SVO (OR = 1.00, p = 0.37) and frequency of charitable donations (OR = 1.10, p = 0.20) were not significant in the multivariate model. - Trust: Societal trust and TDM&H/government independently predicted participation. Lower trust in TDM&H/commercial researchers was a key predictor of participation (OR = 0.57, p = 0.01), indicating participants’ comparatively lower trust in commercial researchers’ data handling compared to refusers.

The study demonstrates that prosocial intrapersonal factors, particularly health-related values and context-specific trust, are stronger determinants of biobank participation than demographics. Healthspheric values and donor statuses (blood, organ) distinctly characterize participants, suggesting that valuing health and healthcare is central to the decision to participate. Conversely, higher biospheric values among refusers indicate different value prioritizations that may reduce the appeal of participation framed around health benefits. Although participants showed higher SVO and donation frequency in univariate comparisons, these effects did not persist after adjusting for other variables, indicating no evidence for a broadly stronger general prosociality among participants beyond the biobanking context. Trust functions as both facilitator and barrier depending on context: greater societal and government data-handling trust supports participation, whereas higher trust in commercial researchers’ data handling is associated with refusal, highlighting skepticism toward commercial data users among participants. These insights suggest that recruitment strategies emphasizing health-related benefits and trustworthy governance can reduce participation thresholds, whereas unaddressed concerns about commercial data use may deter participation. Overall, the findings underscore that participation decisions are context-dependent, shaped by specific value orientations and nuanced trust profiles rather than general prosocial tendencies alone.

This study contributes evidence that among Dutch adults, health-oriented values and domain-specific trust profiles distinguish biobank participants from refusers more strongly than demographic factors. Emphasizing health-related values and clearly communicating robust, trustworthy data governance—particularly in government or public research settings—may enhance recruitment. Addressing concerns regarding commercial researchers’ data handling appears critical, for example by increasing transparency, oversight, and public benefit assurances in public–private collaborations. Future research should examine how prosocial intrapersonal characteristics influence both participation and withdrawal over time, and how these interact with contextual attributes of studies and data users to shape trust and value alignment.