What patients understand when you give them bad news
When you give bad news to a patient or immediate family, is their understanding likely to be accurate? Not necessarily, even when you are brutally frank about the poor prognosis, according to a recent study.
People tend to mentally soften the blow on hearing bad news, US researchers found in an experimental study involving 200 students who were asked to evaluate a range of prognoses. Even when presented with the stark statement that a patient “will definitely not survive”, participants in the study did not consider that as indicating a 100% likelihood of dying.
The researchers found that positive bias was accentuated the worse the prognosis was. Told that a patient was “very likely to survive”, participants rated the odds that the patient would survive at 89%; but when told that a patient was “very likely to die”, they estimated the odds of death at only 76%.
But they also found that using a more emotionally laden phrase to a prognosis could lessen the effects of positive bias. When told that “it is possible” that a patient would not survive, participants rated that as a 50/50 chance of survival. But if the physician used the phrase “I am concerned that [the patient] won’t survive”, participants downgraded the chances of survival to 35%.
However, the researchers didn’t find any difference in bias regarding the wording of the prognosis in terms of either dying or surviving. In other words, participants attached the same risk of death to the statement “He is unlikely to survive” as they did to “he is likely to die”.
The study authors say their research, along with previous work by other researchers, shows positive bias to be a universal defensive mechanism in response to negative information. But they add that putting numbers to the prognosis – for example telling patients or relatives that they have a 95% chance of dying within three months – is unlikely to counter the positive bias, as previous research has demonstrated that numerical prognoses are just as prone to bias.
“Practitioners should be aware of the ways in which commonly used non-numeric language may be understood in numeric terms during prognostic discussions, and check recipients’ understanding during consultations for accuracy and potential positive bias,” they conclude.
You can access the study here.
Interested in learning more? Professor Stewart Dunn will be moderating workshops in Sydney in 2018 on complex communication in health care. The workshops will cover open disclosure, breaking bad news, end-of-life conversations and dealing with conflict in the workplace. Read more about the workshops and sign up here.