"Suicide watch" covers a range of activity. Someone may be getting a visual check once an hour, once every 30 minutes, or once every 15 minutes. That may be as

"Suicide watch" covers a range of activity. Someone may be getting a visual check once an hour, once every 30 minutes, or once every 15 minutes. That may be as simple as looking through a hatch / peephole to see if the person isn't obviously ligatured, or it may include calling to the person and hearing a response. There's not much evidence that 15 minute observations work, and some evidence they cause harm. If someone's risk is high enough for them to be on observations they probably need to be on continuous observations. Continuous observations are tricky. Imagine another person being with you 24 hours of the day, watching you while you eat, bathe, defecate. We want to avoid them if possible because it doesn't build the patient's / prisoner's safety. It keeps them safe for those few days, but it may significantly increase their risk afterwards. And this is if the observations actually happen. Sometimes staff lie. https://twitter.com/KirstenSjovoll/status/1067382479336742912?s=20 https://twitter.com/KirstenSjovoll/status/106738247933674291... 15 minute obs disturb sleep and don't work: https://www.cambridge.org/core/journals/bjpsych-bulletin/article/against-the-stream-intermittent-nurse-observations-of-inpatients-at-night-serve-no-purpose-and-cause-sleep-deprivation/BF2CD76D8F8970581ED48C7594E9A3C4#.XGPGNer5dXM.twitter https://www.cambridge.org/core/journals/bjpsych-bulletin/art...