Guide-linked quiz · Hard Level

AI Fall Risk Prediction in Hospitals Quiz

Tests the Morse Fall Scale's limits, why rare falls give predictive models low precision, and how video bed-exit detection differs from pressure-pad alarms.

Question 1 of 8

How many items does the Morse Fall Scale score?

Preview every question8 questions

Answers and explanations appear as you take the quiz.

  1. How many items does the Morse Fall Scale score?
    • Four
    • Six
    • Eight
    • Ten
  2. Why can a fall prediction model with a high AUROC still produce mostly false alarms?
    • AUROC only measures processing speed
    • Hospitals cannot calculate AUROC for fall models
    • High AUROC means the model predicts every patient will fall
    • Falls are rare, so even good ranking yields low precision at usable thresholds
  3. How does computer vision bed-exit detection differ from a pressure-pad bed alarm?
    • It can detect posture changes like sitting up before weight leaves the mattress
    • It only works after the patient has fallen
    • It uses the patient's heart rate instead of movement
    • It requires the patient to press a call button
  4. What did a large randomized trial of increased bed alarm use find, according to the guide?
    • Falls fell by more than half
    • Alarms eliminated injurious falls
    • It did not find a reduction in falls
    • Alarms increased patient satisfaction but not staffing
  5. What CMS policy does the guide mention regarding falls?
    • CMS requires video monitoring in every room
    • Since 2008, CMS has not paid hospitals extra for falls with injury acquired during a stay
    • CMS bans the use of the Morse Fall Scale
    • CMS pays a bonus for each fall alert generated
  6. What limitation of the Morse Fall Scale does the guide highlight?
    • It is usually scored once a shift, misses changes between assessments, and labels many inpatients at risk
    • It can only be used in pediatric units
    • It requires laboratory data to calculate
    • It gives a single yes or no answer with no score
  7. After a patient is flagged as high fall risk, which action does the guide link to actually reducing falls?
    • Adding a second alarm with a louder tone
    • Moving the patient to a room farther from the nurses' station
    • Rescoring the patient every hour without other changes
    • Scheduled toileting rounds and a medication review
  8. Which privacy design do many computer vision fall systems use, according to the guide?
    • Streaming all footage to social media for review
    • Recording audio only and sharing it with families
    • Live viewing without recording, sometimes with blurring or outlines
    • Storing full-resolution video indefinitely for model training

Keep testing yourself

More quizzes picked for your level.

View all quizzes