Guide-linked quiz · Medium Level

AI in Palliative Care and Mortality Prediction Quiz

Tests risk prediction versus prognosis, NICE uncertainty, supportive referral, validation, patient goals, and limits on treatment decisions.

Question 1 of 8

What does a mortality-risk model output represent?

Preview every question8 questions

Answers and explanations appear as you take the quiz.

  1. What does a mortality-risk model output represent?
    • A diagnosis that palliative care is required
    • A patient’s preference for treatment
    • A probability for a defined outcome window in the studied population
    • A definite date of death for a person
  2. How should a high-risk alert be used?
    • As a substitute for discussing preferences
    • As a prompt to review needs and consider a palliative conversation
    • As a reason to stop treatment automatically
    • As proof the person is in the last days of life
  3. What does NICE emphasize about recognizing the last days of life?
    • A model can identify the exact day reliably
    • All patients want to know a predicted timeline
    • A score should determine treatment withdrawal
    • Prognostic uncertainty and the need for clinical judgment and communication
  4. Why assess calibration?
    • It checks whether predicted risks align with observed outcomes
    • It measures only model speed
    • It guarantees subgroup fairness
    • It establishes patient preferences
  5. What does a retrospective risk model not prove?
    • That it predicts a defined outcome
    • That it may be evaluated in a cohort
    • That using its alert improves goal-concordant care
    • That it was trained on past records
  6. What should a service monitor after deploying alerts?
    • False alerts, missed referrals, alert burden, and care outcomes
    • Only the number of predictions made
    • Only the model’s training loss
    • Only the referral form design
  7. Which action would be inappropriate based only on a mortality score?
    • Assessing whether an alert is accurate
    • Restricting treatment or access to care
    • Offering a supportive discussion
    • Reviewing symptoms and preferences
  8. Why document a clinician’s reasoning after an alert?
    • It makes the prediction certain
    • It replaces informed consent
    • It guarantees the referral was wanted
    • It shows how context, uncertainty, and patient goals informed the response

Keep testing yourself

More quizzes picked for your level.

View all quizzes