How is Prognosis Estimated and Communicated for People Facing Serious Illness?

Introduction and Scope of the Problem

Prognostication—the process of addressing “what to expect” for an individual’s disease course—is essential for meaningful decision making at all stages of serious illness. Estimating prognosis is most commonly applied to estimating life expectancy, but estimates may also be applied to other important markers in an illness trajectory, including when and if treatment will lead to a cure, the development of disability, or nursing home placement.

A patient’s expected prognosis can help guide clinical practice. Determination of a prognosis can be central to accessing health benefits; for example, eligibility for hospice services among those with a prognosis of 6 months or less or planning for nursing home care in older adults with dementia who may lose independence in the community. Prognostication can provide context for clinical decisions in relation to competing medical priorities or risks. The time to benefit (TTB) from a clinical intervention is the time between the clinical intervention (when complications and harms are most likely) and the time when health outcomes improve. For example, in an individual with spinal metastases the average TTB for surgery to relieve spinal cord compression is approximately 3 months, suggesting that those with a prognosis of less than 3 months may benefit from discussions about whether conservative management of pain might be preferred to undergoing surgery for spinal cord compression in the last months of life. As another example, an individual with a prognosis limited to months may have little to gain from QTc interval monitoring when titrating methadone as compared with the inconvenience of frequent testing and clinic visits. In addition, an understanding of prognosis might guide the provision of counseling for anticipatory grief and mental health needs.

There is a growing body of evidence suggesting strong patient preferences for timely, balanced, and sensitive conversations around prognosis to help inform shared decision making and life choices. In a study of 60 older adults from diverse racial and ethnic backgrounds, 75% of individuals stated they would want to know if their doctor thought they had less than 1 year to live. Rationale provided by participants included being able to prepare financially, having time to say “goodbye,” and addressing spiritual needs. Similarly, patients’ prognostic understandings can influence willingness to undergo chemotherapy or other cancer-directed treatments. Communicating prognosis can be thought of as a key component of patient-centered care which is endorsed by the Institute of Medicine and the National Priorities Partnership. Patient-centered care integrates the patient’s experiences of illness, values, and preferences with the physician’s clinical recommendations.

Barriers to Prognostic Estimates

There are several barriers to prognostic estimates and communication. A culture of “ritualized optimism” in medicine where clinicians are uncomfortable discussing prognosis out of concern for inappropriately “taking away hope” is an important barrier. Evidence remains mixed on the extent of patient distress with prognostic discussions. One study of patients with metastatic lung and colorectal cancer found that disclosure that chemotherapy would not be curative was associated with lower patient ratings of clinician satisfaction. Other studies have indicated that patients want accurate prognostic information even if it is upsetting as it can help individuals and families prepare. And, in many circumstances, this information does not harm emotional well-being, but can actually lower anxiety and depression and improve the doctor–patient relationship. One study of 265 patients with advanced cancer found prognostic discussions with oncologists resulted in small but statistically significant improvements in patient ratings of the therapeutic alliance in the days and months following the conversation. Taken together, evidence suggests prognostic discussions should incorporate robust responses to emotional needs and psychological counseling.

Additional factors that may hinder estimating and communicating prognosis include prognostic uncertainty, clinician discomfort with accurate estimates, patient factors (cultural norms, language differences, and communication preferences), or family-related barriers (different beliefs or acceptance in prognostic information in comparison to the patient). While barriers can be complex and multifactorial, even within individual patient scenarios, improved estimates and communication are possible. Indeed, having a shared understanding of prognosis between the clinical team and the patient is a critical prior step to shared decision making and meaningful goals of care discussions. In the following sections we provide a framework for (1) estimating prognosis and (2) communicating prognosis.

Estimating Prognosis

General Principles

In recent years, the science of estimating prognosis has improved substantially, due to increasingly available data from national surveys, clinical trials, and the development of prognostic calculators. Prognostic calculators combine multiple clinical indicators to yield an overall prognostic estimate. They are particularly useful as clinician estimates alone have a higher frequency of errors and can be erroneously optimistic. Accurate prognostication should carefully consider how to integrate and tailor data from prognostic indices to individual clinical scenarios. Prognostic indices used in combination with clinical intuition are likely to yield less biased and more accurate estimates of prognosis. We present an algorithm for this clinical approach in Fig. 31.1 .