When my aunt was diagnosed with pancreatic cancer, the doctor told her she would live about three months. Unfortunately, she did not do that well and died about a month later. Is it common for the doctor's estimate to be so far off?
Q: When my aunt was diagnosed with pancreatic cancer, the doctor told her she would live about three months. Unfortunately, she did not do that well and died about a month later. Is it common for the doctor's estimate to be so far off?
A: Patients with advanced end-stage cancer usually want an open and truthful assessment of their condition. One survey showed that 98 percent wanted to have their questions answered honestly and for their doctor to be realistic.
There are many reasons people with advanced cancer benefit from an estimate of their life expectancy, and not just for things like in the movie "The Bucket List."
Terminal patients typically want to make end-of-life plans, such as ensuring their will and personal affairs are in order. They also want to be sure their family knows their wishes for end-of-life care, such as what palliative treatments they may want (they should make out a living will to document these decisions). In other words, they want to have a say in how best to balance the quality and quantity of life they have left.
The life expectancy of a terminally ill patient with advanced cancer depends on many factors:
The type of cancer: For example, the median survival of patients with advanced hepatocellular carcinoma is only one month; advanced pancreatic cancer is three months; colorectal adenocarcinoma is five months; but patients with advanced breast cancer have a median survival of more than 2 years. Potential palliative treatments also differ for different cancers.
The details of the patient's disease: The specific extent of their disease, including the size and location of tumors (and metastases), is important.
The patient's other medical conditions: This includes details of their diseases and how well controlled they are.
The patient's performance status: There are many ways this can be evaluated. For example, the Eastern Cooperative Oncology Group has developed a scale that ranges from 0 to 5, with 0 points for a normally active patient, 1 point for a patient with restricted strenuous activity (but who is able to do light housework or office work), 2 points if they can care for themselves even if they cannot work, 3 points if they can perform limited self-care (and are in bed about half the time), 4 points if they are completely disabled and 5 points once they have died. Other scales, such as the Karnofsky Performance Status, are similar in concept.
Other indicators: A patient's breathing status and their ability to eat and swallow are both also key factors.
In response to a patient's request, most doctors will attempt to give their best estimate of a patient's life expectancy. When this is based on the physician's experience and intuition, it is called a formulated prognosis.
Despite the physician's best intentions, however, a formulated prognosis is usually optimistic and inaccurate, typically estimating a patient's life expectancy as twice as long as it really is.
Multiple studies done from 1970 to 2000 show physicians' formulated prognoses between a factor of 1.2 and 5.3 times as long as the patient actually survives. A study by Glare et al found that physicians' estimates were only within four weeks of a patient's actual life span 61 percent of the time.
Physicians are, of course, just human, so it is not surprising they are optimistic about their patients' survival. In fact, consultants other than the patient's treating physician typically give more realistic (unfortunately, less optimistic) survival estimates.
There have been attempts to come up with better methods to estimate survival of patients with advanced end-stage cancer. One example is the Palliative Prognostic Index, a scoring system with points given for performance status, patient's oral intake (how well they can eat and drink), their respiratory status (whether they get short of breath with exertion or at rest) and their mental status. Most studies have shown that despite its inaccuracies, the physician's estimate of a patient's life expectancy is a crucial component of such systems.
Patients benefit from having some idea of their prognosis, and patients with advanced cancer should feel free to ask for an honest assessment from their physician.
Jeff Hersh, Ph.D., M.D., F.A.A.P., F.A.C.P., F.A.A.E.P., can be reached at DrHersh@juno.com.