Identifying Palliative Care Needs Using the Double Surprise Question: Intuitive Knowledge of General Practitioners in Six European Countries is not Influenced by Their Characteristics. A Vignette Study

IntroductionThe Surprise Question ("Would I be surprised if this patient were to die in the next 12 months?") helps to identify patients' who may need palliative care, but its accuracy is limited. Adding a Second Surprise Question ("Would I be surprised if this patient were still alive after 12 months?"), together called the Double Surprise Question, appeared promising.AimTo determine which patient - and GP-characteristics influence responses to the Double Surprise Question.DesignA cross-sectional study of participants, who were invited to complete an online survey with 20 case vignettes. Outcome was what characteristics of the vignettes and of the GPs influence the answers to the Double Surprise Question, using the association between the answers and the estimated chance of dying within one year as a reference. One-way ANOVA and multilevel logistic regression were used. Trial registration: NCT03697213.Setting/participantsGPs from six European countries.Results250 GPs completed the survey. GPs were more likely to answer 'yes' to the Second Surprise Question when the estimated chance of dying was higher, the Australian Modified Karnofsky Performance Scale was lower and for patients with cancer compared to organ-failure. GP characteristics did not influence the answer.ConclusionsThis study indicates that the Double Surprise Question is broadly applicable by GPs as no associations were found between measured GP characteristics and responses to the Double Surprise Question. We recommend prospective studies in real-world populations to relate the answers to the Double Surprise Question to actual death to further demonstrate its clinical efficacy.