Consent to Medical Data Processing Among Adult Primary Care Patients: Associations with Vaccination Behaviour and...

Background: The secondary use of routinely collected medical data is increasingly important for healthcare research and public health. However, the validity and representativeness of consent-based datasets depend on patients' willingness to allow their medical information to be used. Although vaccination behaviour and trust in healthcare systems have both been extensively studied, little is known about their relationship with consent to medical data processing. Objective: This study aimed to identify factors associated with consent to medical data processing among adult primary care patients, with particular emphasis on vaccination status, vaccination attitudes, information-seeking behaviours, and trust-related factors. Methods: A cross-sectional study was conducted among 921 adult patients recruited from three primary healthcare centres in Poland representing different organizational and geographical settings. Data were collected using a standardized questionnaire. The primary outcome was participants' consent to the transfer of vaccination-history data (yes/no), referred to throughout the manuscript as "consent to medical data processing". Associations between consent status and socio-demographic characteristics, vaccination-related attitudes, information sources, and trust-related factors were assessed using chi-square tests and non-parametric methods. Hierarchical multivariable logistic regression models were constructed to identify factors independently associated with consent. Model performance was evaluated using the area under the receiver operating characteristic curve (AUC), Nagelkerke's R², and the Hosmer-Lemeshow goodness-of-fit test. Results: Overall, 565 participants (61.3%) provided consent to medical data processing and 356 (38.7%) declined consent. Among participants who provided consent, 88.5% were classified as vaccinated, compared with 69.1% among those who did not provide consent (p < 0.001). In multivariable analyses, vaccination status remained independently associated with consent across all hierarchical models. The fully adjusted model showed that vaccinated participants had more than twice the odds of providing consent compared with unvaccinated individuals (OR = 2.27; 95% CI: 1.32-3.92; p = 0.003). Older age was also positively associated with consent (OR = 1.02 per year; 95% CI: 1.01-1.03; p = 0.004). Significant and persistent differences were observed between healthcare centres, suggesting that organizational and contextual factors may play an important role. By contrast, vaccination attitudes, trust in the healthcare system, perceived accessibility of vaccination information, and information-source were not independently associated with consent after adjustment. The final model demonstrated good discrimination (AUC = 0.892), satisfactory calibration (Hosmer-Lemeshow p = 0.647), and moderate explanatory power (Nagelkerke's R² = 0.557). Conclusions: Consent to medical data processing was independently associated with vaccination status and healthcare centre among adult primary care patients. The findings suggest that vaccination behaviour and organizational context are important correlates of consent to medical data processing. Organizational characteristics may also play an important role in shaping consent behaviour in healthcare settings. Researchers using consent-based medical data should consider the potential impact of differential consent behaviour on the representativeness of study populations. Further research is needed to understand contextual and behavioural determinants of consent to medical data processing.