The ability to predict short- and long-term outcomes for cancer patients has become increasingly important. Changes in the way care is provided and paid for, along with a more consumerist attitude on the part of patients, have made this a more prominent issue.
Patients, the insurance system, and clinical investigators all rely on oncologists to provide reasonably accurate estimates of individual patients’ prognoses. However, as described in the article by Lamont and Christakis, this task is often undertaken with too little data or with psychological barriers that cause oncologists to be either overly optimistic or pessimistic about individuals’ prognoses. There is little doubt that we can do better and that technologic tools are at hand that can facilitate the task.
Types of Prognosis
As alluded to by the authors, prediction can come in two forms: (1) prediction of treatment value for the patient (percentage chance of cure or relapse) or (2) prediction of time to death. Cancer patients undergoing primary or adjuvant treatments are most likely to receive the first type of prognostication from their oncologist, end-stage patients, the second type. Nonetheless, both types of prognosis might be important to patients at any stage of the disease. Somewhat disappointingly, this article does not clearly explain the importance and relevance of this distinction.
Why Tell Patients Their Prognosis?
Patients often overestimate the benefits of treatment by as much as 60%.[1-4] Consequently, it is important that oncologists communicate clearly to patients about their prognosis. Possible reasons why patients are unclear about treatment benefits include denial, unclear communications with physicians, and lack of information. However, patients not only want this information[5-7] but also need it to make rational choices about their treatment.
Whereas the benefits of accurate prognostic information for earlier-stage cancer patients should be obvious, they may be less so for terminal patients. Despite the lack of curative treatment for end-stage patients, they are often offered (or ask for) salvage treatments. Patients often do not understand either the nature of these treatments or their potential beneficial and adverse effects. It has long been suspected that many end-stage cancer patients undertake phase I trials in the mistaken hope of deriving real therapeutic benefit. The results can be costly and painful. Futile care can leave patients and families angry and confused when hoped-for cures fail to materialize.
Communicating With Patients
Curiously, this article never delves into the process through which a prognosis is conveyed to patients—ie, how oncologists communicate with patients. As treatment options have proliferated, and as our arsenal of weapons against cancer has grown, it has become increasingly more challenging to discuss treatment and prognosis with our patients.
A precondition is a willingness on the part of oncologists to tell patients specific prognostic information. Research has demonstrated that numeric information, rather than more general verbal descriptions, is a more successful way to communicate prognosis. This is probably the next communication frontier of oncology.
Years ago, patients were rarely told they had cancer. Today, one would be hard pressed to find a cancer patient who did not know his or her diagnosis. However, the time for telling patients their diagnosis—and nothing more—has passed. We can now expect over 60% of all cancer patients to survive for 5 years or more.
Patients facing the most serious diagnoses experience even greater difficulties understanding and internalizing information. And yet, this is exactly the patient population with the most acute need for information and support from their oncologists. Despite the difficulties, some promising communication aids are under development.
The review by Lamont and Christakis demonstrates the difficulty physicians have in accurately calculating prognoses. Table 1 of the article illustrates that, in many situations, physicians have little more than a 50-50 chance of prognosticating correctly. Moreover, the pattern presented indicates that we have more difficulty predicting the failure of treatment (time to death) than treatment success (likelihood of cure). Thus, our most vulnerable patients, those with end-stage disease, are the least likely to obtain accurate prognostic information. Nonetheless, there are technologies that we can harness to help correct this situation.
Meta-analyses have been used to more accurately predict the effects of adjuvant therapy in patients,[11,12] and these can be applied to improving prognostication for end-stage patients. While having these data available would be invaluable, they need to be used explicitly by oncologists. To do this, we need to change our clinical culture so that it is the norm that prognostic estimates will be derived not simply from clinical judgment, but rather through the use of actuarial methods and data (as delineated in this article). Decision aid programs are currently being developed to help physicians and patients do just that.[13,14] Moreover, physicians must consider using visual aids to help patients understand and retain the information provided.
Finally, a single conversation may not be enough. Rather, an open, supportive ongoing relationship with the patient during the course of the illness will enhance the ability of physicians to more honestly address these difficult issues—first with themselves (foreseeing) and then with their patients (foretelling).
1. Siminoff L, Fetting J, Abeloff M: Doctor-patient communication about breast cancer adjuvant therapy. J Clin Oncol 7(9):1192-1200, 1989.
2. Siminoff LA, Fetting JH: Factors affecting treatment decision for a life-threatening illness: The case of medical treatment of breast cancer. Soc Sci Med 32(7):813-818, 1991.
3. Ravdin PM, Siminoff LA, Harvey JA: A survey of breast cancer patients who have received adjuvant therapy: Defining their parameters of knowledge and acceptance of treatment. J Clin Oncol 16(2):515-521, 1998.
4. Chan A, Woodruff RK: Communicating with patients with advanced cancer. J Palliat Care 13(3):29-33, 1997.
5. Davison B, Degner L, Morgan T: Information and decision-making preferences of men with prostate cancer. Oncol Nurs Forum 22(9):1401-1408, 1995.
6. Meredith C, Symonds P, Webster L, et al: Information needs of cancer patients in West Scotland: Cross sectional survey of patients’ views. Br Med J 313:724-726, 1996.
7. Cassileth BR, Volckman D, Goodman RL: The effects on radiation therapy patients’ desire for information. J Radiat Oncol 6:493-496, 1980.
8. Oken D: What to tell cancer patients: A study of medical attitudes. JAMA 175:1120-1128, 1961.
9. Novack DH, Plumer R, Smith RL, et al: Changes in physicians’ attitudes toward telling the cancer patient. JAMA 241(9):897-900, 1979
10. Ries LAG, Kosary CL, Hankey BF (eds), et al: SEER Cancer Statistics Review, 1973-1995, Bethesda, MD, National Cancer Institute, 1998.
11. Early Breast Cancer Trialists’ Collaborative Group: Tamoxifen for early breast cancer: An overview of the randomised trials. Lancet 351(9114):1451-1467, 1998.
12. Early Breast Cancer Trialists’ Collaborative Group: Polychemotherapy for early breast cancer: An overview of the randomised trials. Lancet 352(9132): 930-942, 1998.
13. Siminoff LA, Ravdin PM, et al: Impact of a computer-based tool for providing individualized estimates of outcomes breast cancer patients. Proc Am Soc Clin Oncol 17:105a, 1998.
14. Levine MN, Gafni A, Markham B, et al: A bedside instrument to elicit a patient’s preference concerning adjuvant chemotherapy for breast cancer. Ann Intern Med 117(1):53-58, 1992.