‘Shared decision-making, preferences, risk-tolerance, and the cancer patient experience’

Abstract

Shared decision-making (SDM) is a best practice in personalized medicine. SDM is defined as ‘an approach where clinicians and patients share the best available evidence when faced with the task of making decisions, and where patients are supported to consider options, to achieve informed preferences’ [1]. It reflects the ideals of person-centered care, by virtue of its inherent respect and valuation of the patient perspective as a vital component in the treatment decisionmaking process [2]. Amid a shift from the unbridled medical paternalism of the mid-twentieth century, to the other extreme of unqualified patient self-determination and medical consumerism, SDM reflects a reasonable midpoint in the pendulum swing between these more polarized approaches. Accordingly, the National Academy of Medicine has called for us all to do SDM, and has been doing so for more than a decade [3]. Unfortunately, it is not clear that this long-standing recommendation has had much impact on clinicians’ practices, despite growing evidence as to the many benefits of an SDM approach in cancer care and beyond [4–6]. Data continue to show that most patients with advanced cancer fundamentally misunderstand their prognosis [7,8], and that many Americans die each year in ways that they would not want, such as in hospital or intensive care unit settings, or receiving ineffective chemotherapies in the last 2 weeks of their life [9–11]. These gaps seem to be even worse in cases of prognostic uncertainty, such as hematologic malignancies, where cure may still be possible in those with relapsed or refractory disease (albeit less likely) [12–18]. As we grapple with this ongoing problem of poor-quality end-of-life care in oncology, SDM must be seen as an essential tool to improve these outcomes. More broadly though, SDM and the elicitation of patient preferences that must accompany it are essential to improving patients’ experiences of illness when facing a cancer diagnosis. After all, patients are the experts on their own experiences. SDM is particularly necessary in cases where reasonable people might disagree about the ‘right’ path, wherein there is truly a ‘preference-sensitive decision’ at play. A classic example is the choice of whether to pursue adjuvant chemotherapy for an early stage, colorectal, breast, or lung cancer. There is no clear ‘right answer’ in these settings, but rather a series of trade-offs to consider, the value of which might be perceived differently by people with differing value systems and priorities. While these therapies are known to reduce the risk of cancer recurrence, in many cases these reductions are relatively modest, in the order of a few percentage points. Along with such modest reductions in recurrence risk, these therapies also come with some long-term risks and short-term toxicities. For example, cardiotoxicity may result from anthracycline-containing regimens often used for breast cancer treatment. Similarly, neuropathy may result from oxaliplatincontaining regimens often used as adjuvant therapy for colorectal cancer, or from taxanes as used to treat breast or lung cancer. Hopefully, we all agree that it is important for our patients to understand the risks and benefits of any therapy they consider, and that an honest, open conversation must take place about the right path for eachperson. Such is the spirit of SDM, and thosewho care for people with cancer generally seem to agree in principle about its importance. The irony, however, is that most of us have not actually been trained in how to operationalize SDM in our practices. Rather, under the traditional apprenticeship model of medical training, we espouse more of a ‘see one, do one, teach one’ approach. Sadly, this often means that someone who has never been trained in SDM ends up themselves somehowmodeling it for the next generation of clinicians, who then muddle through it on their own after seeing it just once or twice, perhaps never having seen a vanguard example of SDM in the flesh. The result is what I tend to hear most often when I observe these interactions in practice: relatively superficial, one-time discussions about treatment paths, emphasizing risks and toxicities, but missing any meaningful exploration and sharing of preferences, priorities, goals, and values. As such, we fail to recognize the importance of ‘diagnosing preferences,’ an essential component of true SDM [19,20]. The importance of ‘diagnosing preferences’ became abundantly clear to me recently, as I struggled to understand some surprising results from a research study our team conducted [21]. In short, we found that a sizeable minority of laypersons would choose a palliative-intent chemotherapy instead of one that could cure them, even when the chance of cure approaches 50%. We expected that the chance of cure would largely drive decisionmaking, and it certainly did in part, however amajor predictor of a