Medical Decision-Making: At the Heart of Care for Older Patients With Cancer

A Conversation With William Dale, MD, PhD

Get Permission

William Dale, MD, PhD

William Dale, MD, PhD

Although oncology has moved toward a personalized approach to care, medical decision-making within the context of multidisciplinary care has broad public health implications. To shed light on this and other relevant issues, The ASCO Post recently spoke with William Dale, MD, PhD, of the University of Chicago. He is an international speaker with expertise in medical decision-making, behavioral economics, and quality-of-life issues in older patients with cancer, particularly prostate cancer, as well as a founding editor of the Journal of Geriatric Oncology.

Interest in Geriatrics and Palliative Medicine

Please tell the readers a bit about your background and current position/work.

I am Section Chief of Geriatrics and Palliative Medicine at the University of Chicago. I am also the founder and Director of the Specialized Oncology Care And Research in the Elderly, or ­SOCARE Clinic, a unique program that offers interdisciplinary, individualized, and integrated treatment for older patients with cancer. Finally, I am a health-services researcher studying medical decision-making in older adults with cancer, especially the psychology of decision-making in patients with prostate and pancreatic cancers. I will be co-leading the Annual Meeting for the Society of Medical Decision Making in 2017.

Components of Medical Decision-Making

Medical decision-making covers a wide spectrum of issue. Would you discuss medical decision-making within optimal strategies in caring for patients with cancer, particularly given the relatively new focus on financial toxicity.

Medical decision-making does cover a wide range of issues! To set the context, there are three broad areas within the field of medical decision-making: (1) the psychology of medical decision-making; (2) shared decision-making (especially decision aids); and (3) comparative-effectiveness analysis. The concept of “financial toxicity” has focused on the issue of medication costs, especially new chemotherapies, confronting patients making choices about their care. Interestingly, along with medical decision-making, it falls into all three categories.

First, under the psychology of decision-making, how does the high cost of drugs affect patients’ decisions about treatments? If one worries about going into debt to pay for a treatment that might not significantly extend life, with significant toxicities, perhaps a different choice is made.

Second, for shared decision-making, is it appropriate for a physician to discuss costs with a patient considering treatments? Patients are not truly informed about a treatment choice if they don’t have all the relevant information about it. The out-of-pocket costs, which can be very high for new treatments, are surely relevant information, just as much as the likelihood of developing hearing loss or a heart valve problem as a result of treatment.

Third, from a policy perspective, how do we properly include treatment costs into the risk/benefit calculation of cancer care for patients, families, and payers, including the government? Should a medication that costs a great deal of money and on average extends life a few weeks or months be approved? Elucidating such trade-offs is central to medical decision-making.

Life Expectancy, Gleason Score, and PSA

Prostate cancer, from screening to treatment, still conjures up heated debate. Please discuss medical decision-making in older patients with prostate cancer.

Prostate cancer diagnosis and treatment remain hugely controversial, especially for older patients. From a medical decision-making perspective, the heart of the matter swirls around two numbers: the over 200,000 men diagnosed with prostate cancer each year and the approximately 30,000 men who die of it. In essence, a very large number of (mostly older) men have prostate cancer, but a relatively small percentage of them will die of it; yet, this remains the second leading cause of cancer deaths in men. How do we avoid overtreating those with prostate cancer who will die of something else while also avoiding undertreating those who will die of this potentially curable cancer?

It is crucial not just to know someone’s age, but his or her level of fitness. Fitness, measured objectively with appropriate assessments, typically called “geriatric assessments,” indicates someone’s life expectancy for his or her age. In general, if a man has a life expectancy less than 10 years, he is not a candidate for prostate cancer testing with prostate-specific antigen (PSA).

Assessing patient preferences is most important during times of great uncertainty in care: for instance, if a treatment improves the likelihood of living longer but also increases the likelihood of short-term morbidity or mortality.
— William Dale, MD, PhD

Tweet this quote

Next, one should know the “riskiness” of the cancer for progressing. The best measure of this is the Gleason score for disease grade. It is unlikely for a man to die of a Gleason 6 (or less) cancer, called low-grade prostate cancer, and therefore a deliberate approach to treating it is appropriate. Conversely, Gleason 8 or higher disease, called high-grade prostate cancer, will grow rapidly and is much more likely to cause problems and lead to death. With these two pieces of information—fitness to assess life expectancy and Gleason grade to assess cancer severity—a rational approach to prostate cancer decision-making is possible.

Another important aspect of decision-making is the role of the blood-based biomarker for prostate cancer, PSA. This is still the most commonly used biomarker of disease. Unfortunately, it is a challenging biomarker to use. For screening, it is a very “noisy” marker, with a lot of variation associated with it; consequently, there are lots of false-positive results associated with it in a screening context. This is primarily why the large trials of PSA for screening show little benefit for mortality gains. However, after treatment such as surgery or radiation, it is an excellent marker for disease recurrence, since there should be little to no PSA in a man after treatment. PSA should be used appropriately based on these characteristics and the clinical circumstances.

Unfortunately, PSA often becomes the primary focus for patients, leading to a great deal of anxiety. (It is sometimes referred to as “prostate-specific anxiety” marker!) As a result, many patients, and sometimes even doctors, make decisions due to anxiety from a PSA value that is higher than expected. A great deal of overtreatment of prostate cancer can be attributed to anxiety from PSA, which doesn’t reflect a rational assessment of the overall clinical picture. Our own research has shown that a primary reason for the overuse of androgen-deprivation therapy for prostate cancer is directly attributable to anxiety rather than a realistic assessment of the risks and benefits.

Cognitive Impairment

Cognitive impairment is not uncommon in the geriatric setting. Is there a structured, guideline-driven approach to assess and deal with this challenge in the oncology setting?

Cognitive impairment is indeed a major problem in treating older adults with cancer. Although I am not aware of any guidelines for patients with cancer, there are several well-validated tools for assessing cognitive impairment in older adults. The most well known are the Mini-Cog, the Mini Mental Status Examination (MMSE), and the Montreal Cognitive Assessment (MoCA); each is a short screening tool that can be used to identify cognitive impairment in older adults, including those with cancer. A patient who scores in the abnormal range on these tests should undergo more complete assessment for cognitive impairment or dementia.

The most important settings in which a cognitive assessment should occur are prior to consent for treatment or a trial; after treatment that can impact cognition, such as whole-brain radiation or chemotherapy; and during hospitalization, when delirium can dangerously impact outcomes. Patients with cognitive impairment require additional supports to safely participate in most cancer treatments.

Ask the Patient!

The emphasis on patient involvement in care decisions introduces the question of how do we measure patient preferences and incorporate them into a decision?

This is a huge question, and it depends on the context of the decision being made. Assessing patient preferences is most important during times of great uncertainty in care: for instance, if a treatment improves the likelihood of living longer but also increases the likelihood of short-term morbidity or mortality. A good example is consideration of a bone marrow transplant: It may be a possible way to cure a disease, but it also raises the risks of earlier death from that very treatment. Cancer care is full of such difficult and uncertain trade-offs.

The best way to measure preferences is also the most traditional—Ask the patient! There are more formal ways to do it, such as through a time-tradeoff or standard gamble methodology. Also, there are decision aids, which are formalized practical tools for eliciting patient preferences, although such formal techniques are mostly for a research setting.

In a busy clinical setting, one has to ask patients in a careful way whether they prefer a longer life or a shorter life with a higher quality of life. Exactly how to pose the question is something of an art more than a science, but it fundamentally involves empathy and exploration of past decisions. One of my personal favorites is to ask, “What kind of person are you when it comes to making choices?” Providing information to patients in an unbiased way is crucial to allowing fully informed, but shared, decisions for patients making difficult trade-offs.

The Challenge of Medicine

Would you like to share some closing thoughts?

Medical decision-making lies at the heart of care for patients with cancer, especially older patients. Almost all of our treatment modalities involve trading one set of potential outcomes for others. Often, these outcomes are uncertain, statistical distributions based on evidence from a group of selected patients in a trial setting. Applying them to an unselected patient in the office is fraught with difficulties. Doing so in a thoughtful, humane way is the challenge of medicine. Patients don’t want to know “the statistics”—They want to know what will happen as individuals.

In the recent book When Breath ­Becomes Air, 36-year-old neurosurgery resident and author Paul Kalanathi captures this sentiment perfectly after being diagnosed with metastatic lung cancer: “It occurred to me that my relationship with statistics changed significantly as soon as I became one.” ■

Disclosure: Dr. Dale reported no potential conflicts of interest.