Decision-making in Chemotherapy Treatment

It is hard to decide how to treat cancer patients. It is hard for the oncologist and hard for the patient, too, who is often (and should be) involved in decisions about treatment. The patient can rely on the oncology team for guidance but even experienced doctors and nurses often struggle with plotting a course of action.  There are over a hundred types of cancer. Treatment options include surgery, radiation, chemotherapy, immunotherapy, and hormone therapy. And even within chemotherapy there are over a hundred drugs, and many ways of combining and sequencing chemotherapy medications and methods of administration. There are decisions about salvage chemotherapy and palliative care, decisions about other medical issues to be addressed, decisions that involve payment and whether the patient can afford treatment.

Empirically it has been found that when making treatment decisions, patients consider not only efficacy (chance of cure or disease response) but also quality of life, toxicity, convenience, and cost – but different patients weigh those factors differently and understand the costs and benefits differently.

Good medical practitioners know that involving a patient, and sometimes the patient’s family, is the best course in interventions involving long and complex treatments. The US government’s Agency for Healthcare Research and Quality (AHRQ) developed a program called the SHARE Approach to help decision-makers think through the issues.

SHARE is named for the first letter in each of the five steps:

  • Seek out the patient’s input.
  • Help the patient understand treatment options.
  • Assess the patient’s preferences.
  • Reach a decision together.
  • Evaluate the decision.

These are obviously directed to the medical professional and don’t address the nitty gritty of how to “reach a decision together”, but they are a good reminder to health care providers to involve patients in the process, and offer some structure for the process.

Patients come with a wide range of intelligence, medical knowledge, wealth, emotional maturity and personalities. Different ones have different support networks and temperaments. Some get along with oncology nurses and some are belligerent. One individual with advanced cancer may prefer length of overall survival above all else, whereas another might view minimization of symptoms as the highest priority. Working with the patient to chart a course of action is a challenge for medical teams.

This website is about treatment – and specifically chemotherapy – but decisions present themselves all along the cancer cycle: prevention, screening, diagnosis, treatment, survivorship, and end of life.

It is an unfortunate but real truth that in general humans are not great decision-makers. This includes the most experienced doctors, but the advantage medical professionals have over patients is experience working with many cancer cases and more emotional detachment. People are more emotional than rational when it comes to their bodies and the bodies of their loved ones.

Another problem is inadequate numeracy. Numeracy is the mathematical analogue or literacy. Many, if not most, people fail to completely appreciate risks and probabilities.

Cognitive biases influence decisions. If the patient has a friend or family member who has experienced a similar cancer, the patient may unduly draw from the story of the other person.

Further, people do not understand their own preferences in the moment and they are even worse at anticipating what they will prefer in the future.

Quality of Life

Quality of life isn’t just an abstract good; it’s something thinkers and policy-makers attempt to quantify and measure. Analysis of treatment strategies and questions of whether new drugs should be approved or paid for often affects expected effect on life quality. It’s not just ivory tower policy wonks who take this approach; sociologists have found ordinary patients take into account quality of life when making decisions about treatment.

Quality of life (often abbreviated QoL) refers to perception of life, health, social interactions, and interests in the framework of culture. It is subjective, of course, and different people have different values and hence judge quality of life differently, but it is so important that healthcare policy makers analysts feel it should factor into the treatment approval process. Hence QoL is often set as the main outcome measure in studies of effectiveness of treatment.

When clinical trials are established, the researchers are supposed to state the endpoints in advance (before the trials start). While such medically measurable times as Disease Free Survival (DFS) and Progression Free Survival (PFS) times can be employed, sometimes quality of life metrics (which are subjective) are used as endpoints. These are called Patient Reported Outcomes.

Further sharpening of the pencil has led to more nuanced metrics. A treatment that extends life expectancy is, all other things being equal, a good thing. But all other things are not equal. A few good years with high quality of life may be more than a slightly longer life in pain and misery. How to weigh the factors (and indeed which factors to choose) is not immediately obvious and requires thought and is subject to debate. And quantifying factors is also difficult.

But those caveats aside, analysts have developed the idea of the “quality-adjusted life year” (QALY). The American Society of Clinical Oncology put forth a “framework” for comparing treatment options.  Factors evaluated include toxicity and side effects, efficacy (expected increase in life expectancy), and costs.  A year in good health counts as 1 QALY. A year spent in some pain or discomfort counts as less than 1 QALY. The QALY number therefore takes into account quantity and quality of life. An intervention with a lower cost-to-QALY ratio would be preferred over an intervention with a higher ratio. Although the QALY can be adapted for individual decision making, it is not the purpose for which it is most commonly used.

oncologistIndividuals in the same situation may have different preferences. For example, one patient with cancer may choose life expectancy as the main goal, while another might consider minimization of symptoms as the highest priority.

The incremental cost-effectiveness ratio ICER is the ratio between the difference in cost and the difference in benefit of two interventions. QALYs are commonly used in assessing benefits when deriving an ICER, which is commonly expressed as incremental cost per QALY. Policy makers use ICER in cost-effectiveness analysis.

ASCO Guide

The American Society of Clinical Oncology (ASCO)’s Value in Cancer Care Task Force says the focus for evaluating the quality of care should be clinical benefit (efficacy), toxicity (safety), and cost (efficiency). The people who put out clinical trials don’t always break things down like this. They look mostly at efficacy, with toxicity being a consideration only if it is too high. That’s one reason some think the drug approval process is flawed. It looks at the wrong endpoint.

And this makes sense when viewed through the lens of patient-centered medicine, or more precisely, in how a patient would look at it. When they make treatment decisions, patients and families consider the chance of a cure to the extent they understand them. They also consider side effects, the burden of receiving treatment, cost, and overall quality of life. This is hard for professionals in the healthcare field to factor into their calculations. And of course, the calculation is not consistent from patient to patient. Different patients have different values.

Tools and Guides

The Value in Cancer Care Task Force attempts to define value in cancer care by emphasizing three elements: clinical benefit (efficacy), toxicity (safety), and cost (efficiency).  Two versions of the ASCO framework have been developed—one for advanced cancer and another for potentially curative treatment (adjuvant or neoadjuvant therapy).

In both the advanced disease and curative frameworks, points are awarded (or subtracted) in the categories of clinical benefit and toxicity. In the advanced disease framework, bonus points can be earned if a regimen shows statistically significant improvement in palliation of symptoms and/or treatment-free interval compared with the control treatment in a clinical trial. Clinical benefit and toxicity (and bonus points, in the advanced disease framework) are combined to generate a net health benefit (NHB) score.

The acronym SDM has gained currency. It stands for Shared Decision Making and the idea is becoming more prevalent in the medical world.

The American Psychological Association published a paper outlining models of decision-making in cancer treatment:

The Ottawa Hospital has a webpage on “patient decision aids.”

A survey of cancer patients in Canada found half remembered being offered choices in their treatment. The ones who thought they had choices were more active in their treatment.

An article published in STAT opined that despite the rhetoric of SDM, it wasn’t widely implemented to the degree it should be. The article called SDM “more myth than reality”.


If serious problems or risks are found in the clinical trials process, the new drug is usually not approved. An exception may be made if the drug is judged to have very high value.

Medicines already on the market are sometimes recalled if reports of adverse effects surface.

As part of its regulation of US pharmaceutical use, the FDA sometimes has the manufacturer include a “black box warning” in the package insert. Black box warnings are issued when a drug causes a risk of serious injury. The requirement for the black box warning may be placed when the FDA first approves the new medicine, or it may be added in response to reports of adverse effects from patients.

See also: Chemo in wider view of cancer treatment.