Reply to S. Sommerlatte et al

We thank Sommerlatte et al¹ for their comments on our article on clinical decision making for older adults with cancer.² In particular, the authors focus on ethical concerns, expanding on two points inherent in our framework for decision making: (1) a concern about assigning decisional capacity prematurely and (2) the distinction between assistance with and substitution for capacity.¹ We are pleased by the degree of agreement between our views, and we believe our framework can accommodate the expressed concerns.

Regarding the first issue, we agree with the authors that appropriate effort should be made to assess an older patient's capacity to make a treatment decision before deeming them to lack that capacity and be in need of a surrogate decision maker (eg, health care proxy). We also agree that if the patient possesses a health deficit or disability that limits fulfilling one or more criteria of capacity (understanding, expressing a choice, appreciation, and reasoning³), efforts should be made to support the patient in overcoming that limitation before concluding that they lack capacity.⁴ This supported decision making, as referenced by the United Nations Convention on the Rights of Persons with Disabilities,⁵ can come in many forms depending on the specific deficit identified.

For example, older patients with hearing and/or visual impairment may not initially recognize the benefits and harms of a particular cancer treatment simply because they cannot hear and/or read them during the consent process. The mere presence of such conditions or disabilities is insufficient grounds by themselves for deeming one to lack capacity. If interventions can be implemented to address the disability and allow older patients to meet the criteria for capacity, eg, hearing aids or pocket talkers for hearing-impaired patients or a trusted caregiver for alternative means of communicating the information, these interventions should be used. As another example that we allude to in our review, cognitive impairment comes in many forms and stages of severity, and often cognitively impaired patients maintain capacity with adequate assistance to make cancer treatment decisions.^6,7 The need for supported decision making in the context of disability reinforces a fundamental point of our review: Optimal decision making for older adults with cancer requires a geriatric assessment to assess for prevalent aging-related vulnerabilities like sensory losses and cognitive impairment that often go undetected in routine oncology examinations. Sommerlatte et al¹ appropriately highlight that interventions targeting these deficits (eg, allowing extra time to discuss choices in an older adult with hearing or cognitive impairment) help ensure that older patients are able to receive, use, and communicate information regarding cancer treatments in the decision-making process.

The geriatric assessment also best allows clinicians to determine when health deficits that impair decision-making capacity are not modifiable—eg, an older patient with advanced dementia who can neither process nor recall the risks of intensive versus less intensive cancer treatment strategies. Sommerlatte et al¹ provide justifiable caution that invoking surrogate decision makers risks overlooking patient preferences and adding burden to those requested to make decisions on behalf of the impaired patient. However, moving forward with treatments without involving health care proxies in an older patient who truly does lack capacity equally risks missing patient preferences that can result in overtreatment or undertreatment. Accordingly, the caution expressed by Sommerlatte et al¹ implores us to commit time early and often toward eliciting health care preferences and values from older patients, while encouraging them to communicate these preferences and values to their health care proxy before impairment leads to incapacity. Indeed, Wendler and Rid's⁸ systematic review on the effects on surrogates of making treatment decisions for others found that knowing the patients' previously communicated preferences reduced stress on surrogates, providing motivation for older patients to complete advance directives.

Regarding the second point concerning the varying clinician involvement with an older patient's final treatment decision—ranging from a clinician recommending a particular treatment and a patient deciding on a treatment without a recommendation—we did not intend to imply that the limited evidence on which approach is best suggests that some universal recommendation is possible. Rather, we intended to raise awareness of these nuances within the limits of the article, with our primary recommendation being to avoid extremes. We support an approach that, on the one hand, avoids medical paternalism relying on unilateral clinician treatment recommendations that do not take into account patient preferences and that, on the other hand, avoids medical futility that appeases patient or surrogate requests for treatments that offer little benefit and may cause net harm.⁹ We agree with the emphasis by Sommerlatte et al¹ on individualizing approaches to exploring patient preferences that depend both on the oncologists' and older patient's communication styles.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Reply to S. Sommerlatte et al

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).