Family Refusal to Consent Donation: Retrospective Quantitative Analysis of Its Increasing Tendency and the Associated Factors Over the Last Decade at a Spanish Hospital

Background. Organ and tissue recovery remains limited by several factors. This study retrospectively analyzes the factors associated with family refusal to consent to donation at a highdonor-volume Spanish hospital. Methods. Data regarding the annual number of potential donors and family refusal rates at hospital and regional levels were retrieved from 2008 to 2017. Descriptive, bivariate, and multivariate analyses were performed to detect those factors independently associated with family refusal. Results were cross-validated using the data from years 2018 and 2019 as the validation group. To explore potential inter-relations between factors a Multiple Correspondence Analysis was performed. Results. A total of 601 family interviews for petition of consent were conducted between 2008 and 2017, 531 (88.4%) resulted in acceptance and 70 (11.6%) resulted in refusal of the donation. Lesser experience of the interviewers (odds ratio [OR], 2.980; P = .001), donation after brain death (OR, 2.485; P = .013), number of interviews conducted per family (OR, 1.892; P < .001), age of the main decision maker (OR, 1.025; P = .045), and high or middle attributed cultural levels (OR, 0.142; P < .001 and OR, 0.199; P < .001 respectively) were observed to be independently associated with the family final decision. The logistic regression model displayed good predictive power for both derivation and validation cohorts, with an overall predictive accuracy of 80.9% (95% confidence interval, 0.747-0.870; P < .001) and 74.4% (95% confidence interval, 0.635-0.854; P = .001), respectively. Conclusions. Transplant coordination team members having a thorough knowledge of the family decision mechanisms may be a key factor in donation process optimization.

Automated summary

This study found that factors such as interviewer experience, donation after brain death, number of interviews conducted per family, age of the main decision maker, and cultural levels of the family were independently associated with family refusal to consent to donation. The logistic regression model displayed good predictive power for both derivation and validation cohorts, with an overall predictive accuracy of 80.9% and 74.4%, respectively.