“How much time?” A question that some would ask and want to know while others would rather not ask or know. This question is not unusual for those patients with an incurable terminal illness. The most common patient group would be those with incurable cancer. However, this question is also relevant to many patients with non-cancerous conditions such as end-stage lung disease, cardiac failure or neurological conditions such as multiple sclerosis. So how accurate is the doctor’s prognostication / prediction?
A study from University Hospital Dusseldorf, Germany studied aspects of subjective and objective prediction of survival in 216 cancer patients who were referred for palliative radiotherapy (J Clin Oncol 2007; 25: 3313 – 3320). The survival of these patients was estimated according to 3 categories - < 1 month, 1 – 6 months and > 6 months – by two physicians (A & B) and the institutional tumour board (C). After 6 months, the accuracy of these estimates was assessed. The correct progneses were correctly predicted in 61% by A, 55 % by B and 63% by C. In general, statistical examination of the study results showed fair agreement of the estimates which proved to be overly optimistic. In particular the survival of patients who died in < 1 month was over-estimated by 96% by A, 71% by B and 87% by C. Overall, the accuracy of the three estimates did not improve with increasing professional experience.
A 50 – 60 % accuracy rate in predicting an outcome is not much better than flipping a coin. Statistically, if you flip a coin enough times, you will get heads half of the time and tails the other half. Predicting when the final moment will be is a tall order. The only person who could predict accurately is the Almighty up there. Doctors are probably more accurate at predicting if the time is very near – within a few days. Any attempt to predict in terms of weeks or months would be close to pure guess work. There are objective parameters which helps a doctor to be more guarded with his prediction. For example, patients who have colorectal and breast cancer have better prognosis than those with brain metastases. Patients who are too frail and weak to perform daily activities at home tend to do less well. Consequently, these patients have less time on their hand.
When faced with the question “How long?”, a doctor can only come up with an average estimation based on published studies of similar patients. It can act as a guide, albeit not an absolute one. To the patient as well as the family members this piece of information is important. Knowing that it is only months and not years would allow the patient to set his or her own affairs in order.
When I am asked the question, I give a range, rather than a definite number of weeks or months. This is because I know that not only have I no power over such matters and hence cannot be accurate but also to prevent patients having to ask the difficult question “Doctor X told me I only have 3 months to live, can it be longer?”. After all, doctors are human, they do not hold all the answers!
