Doctors’ and nurses’ predictions of ICU outcomes have variable accuracy

 

Clinical question: How accurate are doctors and nurses at predicting survival and functional outcomes in critically ill patients?

Background: Doctors have been shown to have moderate accuracy at predicting in-hospital mortality in critically ill patients; however, little is known about their ability to predict longer-term outcomes.

Study design: Prospective cohort study.

Setting: Five medical and surgical ICUs in three hospitals in Philadelphia.

Synopsis: Physicians and nurses predicted survival and functional outcomes for critically ill patients requiring mechanical ventilation or vasopressors. Outcomes predicted were in-hospital and 6-month mortality and ability to return to original residence, toilet independently, ambulate stairs, remember most things, think clearly, and solve problems.

Six-month follow-up was completed for 299 patients. Accuracy was highest when either physicians or nurses expressed confidence in their predictions; doctors confident in their predications of 6-month survival had a positive likelihood ratio of 33.00 (95% CI, 8.34-130.63). Both doctors and nurses least accurately predicted cognitive function (positive LR, 2.36; 95% CI, 1.36-4.12; negative LR, 0.75; 95% CI, 0.61-0.92 for doctors, positive LR, 1.50; 95% CI, 0.86-2.60; negative LR, 0.88; 95% CI, 0.73-1.06 for nurses), while doctors most accurately predicated 6-month mortality (positive LR, 5.91; 95% CI, 3.74-9.32; negative LR, 0.41; 95% CI, 0.33-0.52) and nurses most accurately predicted in-hospital mortality (positive LR, 4.71; 95% CI, 2.94-7.56; negative LR, 0.6; 95% CI,0.49-0.75).

Bottom line: Doctors and nurses were better at predicting mortality than they were at predicting cognition, and their predicted outcomes were most accurate when they expressed a high degree of confidence in the predictions.

Citation: Detsky ME, Harhay MO, Bayard DF, et al. Discriminative accuracy of physician and nurse predictions for survival and functional outcomes 6 months after an ICU admission. JAMA. 2017;317(21):2187-95.

Dr. Herscher is assistant professor, division of hospital medicine, Icahn School of Medicine of the Mount Sinai Health System.

 

Clinical question: How accurate are doctors and nurses at predicting survival and functional outcomes in critically ill patients?

Background: Doctors have been shown to have moderate accuracy at predicting in-hospital mortality in critically ill patients; however, little is known about their ability to predict longer-term outcomes.

Study design: Prospective cohort study.

Setting: Five medical and surgical ICUs in three hospitals in Philadelphia.

Synopsis: Physicians and nurses predicted survival and functional outcomes for critically ill patients requiring mechanical ventilation or vasopressors. Outcomes predicted were in-hospital and 6-month mortality and ability to return to original residence, toilet independently, ambulate stairs, remember most things, think clearly, and solve problems.

Six-month follow-up was completed for 299 patients. Accuracy was highest when either physicians or nurses expressed confidence in their predictions; doctors confident in their predications of 6-month survival had a positive likelihood ratio of 33.00 (95% CI, 8.34-130.63). Both doctors and nurses least accurately predicted cognitive function (positive LR, 2.36; 95% CI, 1.36-4.12; negative LR, 0.75; 95% CI, 0.61-0.92 for doctors, positive LR, 1.50; 95% CI, 0.86-2.60; negative LR, 0.88; 95% CI, 0.73-1.06 for nurses), while doctors most accurately predicated 6-month mortality (positive LR, 5.91; 95% CI, 3.74-9.32; negative LR, 0.41; 95% CI, 0.33-0.52) and nurses most accurately predicted in-hospital mortality (positive LR, 4.71; 95% CI, 2.94-7.56; negative LR, 0.6; 95% CI,0.49-0.75).

Bottom line: Doctors and nurses were better at predicting mortality than they were at predicting cognition, and their predicted outcomes were most accurate when they expressed a high degree of confidence in the predictions.

Citation: Detsky ME, Harhay MO, Bayard DF, et al. Discriminative accuracy of physician and nurse predictions for survival and functional outcomes 6 months after an ICU admission. JAMA. 2017;317(21):2187-95.

Dr. Herscher is assistant professor, division of hospital medicine, Icahn School of Medicine of the Mount Sinai Health System.

 

Clinical question: How accurate are doctors and nurses at predicting survival and functional outcomes in critically ill patients?

Background: Doctors have been shown to have moderate accuracy at predicting in-hospital mortality in critically ill patients; however, little is known about their ability to predict longer-term outcomes.

Study design: Prospective cohort study.

Setting: Five medical and surgical ICUs in three hospitals in Philadelphia.

Synopsis: Physicians and nurses predicted survival and functional outcomes for critically ill patients requiring mechanical ventilation or vasopressors. Outcomes predicted were in-hospital and 6-month mortality and ability to return to original residence, toilet independently, ambulate stairs, remember most things, think clearly, and solve problems.

Six-month follow-up was completed for 299 patients. Accuracy was highest when either physicians or nurses expressed confidence in their predictions; doctors confident in their predications of 6-month survival had a positive likelihood ratio of 33.00 (95% CI, 8.34-130.63). Both doctors and nurses least accurately predicted cognitive function (positive LR, 2.36; 95% CI, 1.36-4.12; negative LR, 0.75; 95% CI, 0.61-0.92 for doctors, positive LR, 1.50; 95% CI, 0.86-2.60; negative LR, 0.88; 95% CI, 0.73-1.06 for nurses), while doctors most accurately predicated 6-month mortality (positive LR, 5.91; 95% CI, 3.74-9.32; negative LR, 0.41; 95% CI, 0.33-0.52) and nurses most accurately predicted in-hospital mortality (positive LR, 4.71; 95% CI, 2.94-7.56; negative LR, 0.6; 95% CI,0.49-0.75).

Bottom line: Doctors and nurses were better at predicting mortality than they were at predicting cognition, and their predicted outcomes were most accurate when they expressed a high degree of confidence in the predictions.

Citation: Detsky ME, Harhay MO, Bayard DF, et al. Discriminative accuracy of physician and nurse predictions for survival and functional outcomes 6 months after an ICU admission. JAMA. 2017;317(21):2187-95.

Dr. Herscher is assistant professor, division of hospital medicine, Icahn School of Medicine of the Mount Sinai Health System.