Prognostic disclosure to patients with cancer near the end of life

Citation
Eb. Lamont et Na. Christakis, Prognostic disclosure to patients with cancer near the end of life, ANN INT MED, 134(12), 2001, pp. 1096-1105
Citations number
36
Categorie Soggetti
General & Internal Medicine","Medical Research General Topics
Journal title
ANNALS OF INTERNAL MEDICINE
ISSN journal
00034819 → ACNP
Volume
134
Issue
12
Year of publication
2001
Pages
1096 - 1105
Database
ISI
SICI code
0003-4819(20010619)134:12<1096:PDTPWC>2.0.ZU;2-L
Abstract
Background: Patients' understanding of their prognosis informs numerous med ical and nonmedical decisions, but patients with cancer and their physician s often have disparate prognostic expectations. Objective: To determine whether physician behavior might contribute to the disparity between patients' and physicians' prognostic expectations. Design: Prospective cohort study. Setting: Five hospices in Chicago, Illinois. Patients: 326 patients with cancer. Intervention: Physicians formulated survival estimates and also indicated t he survival estimates that they would communicate to their patients if the patients insisted. Measurements: Comparison of the formulated and communicated prognoses. Results: For 300 of 311 evaluable patients (96.5%), physicians were able to formulate prognoses. Physicians reported that they would not communicate a ny survival estimate 22.7% (95% CI, 17.9% to 27.4%) of the time, would comm unicate the same survival estimate they formulated 37% (CI, 31.5% to 42.5%) of the time, and would communicate a survival estimate different from the one they formulated 40.3% (CI, 34.8% to 45.9%) of the time. Of the discrepa nt survival estimates, most (70.2%) were optimistically discrepant. Multiva riate analysis revealed that older patients were more likely to receive fra nk survival estimates, that the most experienced physicians and the physici ans who were least confident about their prognoses were more likely to favo r no disclosure over frank disclosure, and that female physicians were less likely to favor frank disclosure over pessimistically discrepant disclosur e. Conclusions: Physicians reported that even if patients with cancer requeste d survival estimates, they would provide a frank estimate only 37% of the t ime and would provide no estimate, a conscious overestimate, or a conscious underestimate most of the time (63%). This pattern may contribute to the o bserved disparities between physicians' and patients' estimates of survival .