Background Code status discussions are often inadequate, omitting key information on prognosis and resuscitation risks. In the randomized controlled CLEAR trial, use of a checklist to guide shared decision-making for code status discussions improved the quality of care by reducing patients' uncertainty and increasing their satisfaction with discussions and knowledge of this topic. Moreover, a higher rate of "do-not-resuscitate" (DNR) code status preference was observed during the index hospital stay. The present analysis evaluates whether these effects were sustained over time.Objective To evaluate the documented code status (primary outcome) and exploratory secondary outcomes, including patients' general knowledge about resuscitation measures after a mean follow-up time of 3.2 years, assessed by telephone interview.Design Long-term follow-up of a prospective, randomized controlled study.Participants Of the 1954 medical inpatients in the previous CLEAR trial, 604 had died and 10 were excluded due to missing information, leaving 1340 patients for analysis (mean age 65.0 (SD ± 16.3) years, 44.1% female).Interventions Code status discussions using a shared decision-making checklist including visual decision support addressing prognosis, resuscitation outcomes, and patient values compared to usual care.Main Measures Last documented code status at follow-up.Key Results Patients in the intervention group had a significantly higher rate of documented DNR status at follow-up compared to patients in the usual care group (251 36.1% vs. 191 29.8%; adjusted risk ratio 1.23, 95% CI 1.04 to 1.45, p = 0.02). No significant differences were observed in knowledge about resuscitation measures.Conclusions Several years after a shared decision-making code status discussion, patients recalled little specific knowledge about resuscitation measures and their prognostic outcomes. However, the intervention group maintained a higher documented preference for DNR. These results highlight the need for repeated discussions, including prognosis and expected outcomes, to support informed and consistent decision-making.
Bissmann et al. (Tue,) studied this question.