The Hard Facts about Survival Rates
Helping patients and families understand CPR and feeding evidence with honesty, care, and room for their questions.
When the question is really, “Will we still be caring for her?”
A family member asking about CPR or a feeding tube may be asking whether the patient still has a chance. They may also be asking whether declining an intervention would mean letting someone down. A hospice nurse needs accurate evidence, but the way that evidence is shared matters just as much as the numbers.
Begin by finding out what the family hopes the treatment will accomplish. “What are you most hoping this would help your mother do?” is a different opening from immediately reciting a survival rate. It gives you a chance to understand whether they are thinking about more time, recovery, comfort, or an obligation they feel toward the patient.
The purpose of evidence is to help people make informed choices. It should not become a way to win an argument or obtain a DNR order. If you hear fear or guilt, acknowledge it and allow the person to respond before giving more information. The nurse can reinforce the team’s explanation, identify misunderstandings, and bring questions requiring an individualized recommendation to the responsible clinician.[2]
Use the POLST form as a guide to the discussion
Our practical approach begins at the bottom of the form and moves upward. In Box D, identify the person who would help communicate the patient’s wishes if needed and review any advance directive. In Box C, discuss artificial nutrition. In Box B, explore treatment during serious illness before arrest. In Box A, discuss CPR. This is the authors’ conversational approach, not a required sequence.
Do not rush from one box to the next because an answer has been recorded. Ask what that answer means to the patient. “I want everything” may mean “I want you to keep helping me,” while “no machines” may reflect a specific experience the patient has never explained. Neither phrase is sufficient by itself to determine every treatment choice.
The step-by-step POLST article provides sample questions for each section. The evidence here supports that conversation. California’s guidance also clarifies that POLST is voluntary, complements an advance directive, and allows DNR alongside Full Treatment before arrest.[5]
Three outcomes that families may hear as one
When someone says a patient “survived a code,” the listener may picture that person returning home. Clinically, several different outcomes need to be distinguished. Return of spontaneous circulation means that circulation has returned after resuscitation. Survival to hospital discharge means the patient left the hospital alive. Neurologic or functional recovery describes additional aspects of the outcome.
You might explain, “Getting a heartbeat back is an important first step. It does not tell us by itself whether someone will survive the rest of the hospital stay or what they will be able to do afterward.” Pause and ask which of those outcomes matters most to the patient.
Do not multiply a discharge-survival rate measured among all arrests by a separate arrest-survival percentage. The discharge rate already includes the earlier stages. A calculation that combines incompatible denominators can produce a frightening number that does not represent the evidence.
What the AHA population figures show
The 2025 American Heart Association guidelines report the following outcomes. The setting and data year are part of the finding and should stay attached to the number.[1]
| Population and data year | Outcome | Rate |
|---|---|---|
| EMS-treated adult out-of-hospital cardiac arrest, 2024 | Survival to hospital discharge | 10.5% |
| Same out-of-hospital population | Favorable neurologic survival | 8.2% |
| Adult in-hospital cardiac arrest, 2023 | Survival to hospital discharge | 23.6% |
These are not rates specifically for hospice patients, people with advanced dementia, or all older adults. They also do not predict one person’s outcome. Arrest rhythm, whether the event is witnessed, response time, illness severity, and baseline function affect the clinical context. Age alone is not an individualized prognosis.
For a family that wants numbers, you can say, “In that out-of-hospital group, about 10 or 11 people out of 100 left the hospital alive. The figure does not tell us exactly what would happen to your father.” Explain how the treating clinician views the patient’s circumstances rather than applying a broad statistic as a personal forecast.
Give the information in a way the person can absorb
Ask how much detail would be helpful. Some patients want the percentages and study context. Others want a plain explanation of what the intervention is likely to involve. Offer both without assuming that a person who prefers simple language is less capable of making a decision.
Share one point, then check what was heard. “What does that information make you think about?” may reveal that the person has focused on a possibility you did not expect. If they say, “Then there is no hope,” clarify what the evidence does and does not establish, and ask what they are hoping for.
Do not use “favorable neurologic survival” as shorthand for a guaranteed return to the person’s previous life. A study outcome may not capture the specific ability a patient values, such as speaking with a grandchild or living in a familiar setting. Ask about those priorities and bring them into the clinical discussion.
If emotion interrupts the explanation, let the explanation wait. VitalTalk’s resources emphasize acknowledging and exploring emotion rather than simply adding facts.[6] A nurse might respond, “Hearing this seems to have brought up something important. Would you like to tell me about it?” Use your own words and allow silence.
Keep CPR separate from care before arrest
A patient who declines CPR may still want treatment for a potentially reversible illness. Box B deserves its own discussion. Ask what the patient would want the team to consider if breathing became difficult or an infection developed while the patient still had circulation. Have the appropriate clinician explain the likely benefits and burdens of the available options.
For example, a patient might say, “I do not want chest compressions, but I want you to treat an illness that could get better.” Clarify the intended plan rather than interpreting DNR as refusal of all treatment. Conversely, someone who prioritizes comfort may want the team to consider transfer if distress cannot be managed where they live.
Discuss the possible treatment that follows an initially successful resuscitation, including intensive care and breathing support when appropriate. Avoid the opposite overstatement that all survivors remain unconscious or inevitably die shortly afterward. Balanced explanations make room for uncertainty without either minimizing burdens or declaring outcomes that cannot be known.
When feeding becomes an expression of love and worry
In many families, preparing a meal has been a way of caring for the patient for years. When eating becomes difficult, a relative may feel helpless or responsible. “I cannot just stop feeding him” deserves a compassionate response before a technical discussion of feeding routes.
You might say, “You have been caring for him through food for a long time. Let us talk about what is happening with his eating and how we can continue to care for him.” Ask what the family has noticed, what the patient appears to enjoy or find distressing, and what they understand about the illness.
Assess the problem rather than assuming its cause from the hospice diagnosis. Communicate changes in swallowing, intake, alertness, symptoms, and treatment tolerance through the care team. The question is what approach is appropriate for this patient now, not whether a family is willing to show enough love.
What the advanced-dementia feeding study adds
A prospective cohort published in 2025 enrolled 283 patients with advanced dementia and feeding problems; 176 were included in the propensity-matched analysis. Nasogastric feeding did not show a statistically significant one-year survival advantage compared with careful hand feeding: adjusted hazard ratio 0.85, 95% confidence interval 0.59–1.23. Pneumonia risk was higher in the nasogastric group: adjusted hazard ratio 1.60, 95% confidence interval 1.06–2.42.[3]
This was an observational comparison. Matching addresses measured differences but does not eliminate all possible confounding or establish causation. The study concerns nasogastric feeding in advanced dementia; it does not answer every question about gastrostomy, other diagnoses, or different treatment goals. A hazard ratio is not the percentage of patients who experienced an outcome.
At the bedside, begin with the clinical meaning: “For the people with advanced dementia in this study, feeding through a tube in the nose did not show longer survival than careful hand feeding. The clinician can help us consider what that means for your mother.” Do not tell the family that the study proves identical outcomes for every patient or that all forms of artificial nutrition are ineffective.
Describe the care that continues
If careful hand feeding is selected, make the plan visible. Who will assist? What positioning, texture, pace, and other instructions are appropriate to the individualized assessment? How will caregivers recognize distress, and when should they call? Coordinate with the relevant clinicians and swallowing or nutrition professionals when indicated.
Do not give a family the impression that declining a tube leaves them with no work worth doing. Comfort, mouth care, appropriate feeding assistance, observation, and companionship remain important. At the same time, do not promise that careful hand feeding or a tube can eliminate aspiration. Describe the uncertainties honestly and explain how the team will respond.
Ask about the patient’s own experience. Does a familiar taste bring pleasure? Is the person interested in food, or turning away? Follow the individualized care plan and avoid pressuring the patient to meet someone else’s expectations. Help the family understand that being present and attentive remains a way to care, even when the relationship with food changes.
For a patient who has a tube, continue to support comfort, social contact, and the patient’s preferences. Tube feeding does not mean the person must be isolated, nor does every patient with a tube have the same swallowing restrictions. Clarify the actual orders and avoid assumptions about what the person may or may not do.
Stroke-related swallowing difficulty needs a different discussion
Recovery potential changes the decision. The 2026 AHA/ASA acute ischemic stroke guideline recommends enteral nutrition within seven days when indicated, generally with nasogastric feeding initially. Gastrostomy is reasonable when unsafe swallowing is expected to persist beyond two to three weeks.[4] Findings in advanced dementia should not be applied automatically to someone recovering from stroke.
A family may ask why a tube was recommended for one relative but not another. Explain that the diagnosis, expected course, treatment goal, and patient’s wishes may differ. “We are asking what this treatment could realistically help this person achieve” is more useful than presenting tube feeding as universally good or universally harmful.
Help the family identify the question that needs an answer: whether swallowing may recover, what a trial would involve, or how benefit would be assessed. Then bring that question to the clinician responsible for the recommendation. A nurse can make the consultation more useful by communicating what the family is actually worried about.
Make a time-limited trial a shared plan
When a trial is appropriate, explain what the team hopes to learn and when the situation will be reviewed. The goals might concern swallowing recovery, tolerance, comfort, or another outcome relevant to the patient. The prescriber and team should specify the intervention, meaningful measures of benefit, and options if the goal is not achieved.
Ask the patient or surrogate to describe the plan back in their own words. A family may have heard “trial” as “a promise that this will work,” or as a decision that cannot be changed. Clarify that reassessment is part of the plan and that the discussion will consider what has actually happened.
Document the patient’s priorities, the questions answered, the remaining uncertainty, and who will follow up. A conversation about difficult evidence should end with a dependable next step. Families do not need every uncertainty removed before they can feel supported; they need to know that uncertainty will not leave them alone.
References & resources
- AHA, 2025 Guidelines: Executive Summary, sections 2.2–2.3
- AHA, 2025 CPR and ECC Guidelines: Ethics
- Prospective matched cohort: nasogastric feeding versus careful hand feeding, 2025
- AHA/ASA, 2026 Acute Ischemic Stroke Guideline, section 5.3
- Coalition for Compassionate Care of California: POLST Quick Reference
- VitalTalk: Responding to Emotion