This page is written for the person who runs an inpatient unit. It is in the order the argument actually goes: what fifteen-minute checks cover, what a fixed rhythm does, what one-to-one observation costs, what the evidence says it buys, and only then what the sensor measures.
It records that somebody looked and what they saw in that moment. It does not record heart rate, breathing, or what changed between one check and the next.
Continuous measurement does not replace the check or the clinician. It fills the interval with something that can be reviewed afterwards: a rate, a trend, and the times a patient was out of bed.
Checks should be performed at varying intervals of less than fifteen minutes, so that the patient cannot predict the exact time of the next check.
Chammas, Januel and Bouaziz, Frontiers in Psychiatry, 2022 · source
A national case series in England and Wales recorded 113 in-patient suicides that happened while the patient was under observation, about sixteen a year. Most were under intermittent observation, and five happened under constant observation.
Flynn et al., Psychological Medicine, 2017 · source
Both papers are about observation as a practice, not about any device. The XK300 measures heart rate and breathing rate and monitors presence and movement. It is not an alarm system and it does not detect any event.
Some estimates report that an acute care hospital in the United States may spend more than $1 million annually on patient safety attendants, an expenditure often not reimbursed.
Shields, Lawson and Flanders, Nursing, 2021 · source
The Academy of Consultation-Liaison Psychiatry has cited an estimate that US hospitals spend $2 billion a year on sitters. The estimate is widely repeated and has no primary source attached to it, so it is written here as what it is.
ACLP News, February 2019 · source
The most careful costing study of sitter use declined to put a dollar figure on it at all, because hospitals do not have the systems to track the expense. It found a median of 18 hours of sitter time per patient stay.
Kramer and Schubert, PLOS ONE, 2023 · source
One 533-bed teaching hospital reported cutting its annual sitter cost by $894,000 from a $1.2 million baseline. Another reported a third off its monthly spend, and a third reported a quarter fewer one-to-one hours.
Case reports collected in ACLP News, February 2019 · source
Figures like these count sitter hours saved. They do not subtract what the replacement costs to buy, install and run, so read them as a reduction in one line rather than as net savings.
One-to-one sitters is a costly intervention, and evidence that it is effective is uncertain. The review found no randomised trials, and no studies reporting the cost-effectiveness of sitters.
Shekelle et al., “One-to-One Observation: A Systematic Review”, VA Evidence Synthesis Program, August 2019 · source
That finding is the reason this page exists, and it is worth being precise about what it does and does not say. It does not say observation is useless. It says that after twenty studies, none of them randomised, the evidence that it works is not established, and that nobody has costed it properly.
Whatever fills that gap has to be judged the same way. Measuring a patient continuously is not proof of a better outcome either. What it does give you is a record of physiological status across the whole admission, which observation alone cannot produce.
Video monitoring has the most evidence of any alternative to sitters, with eight studies showing either no change or a decrease in falls alongside a large drop in sitter use. The same review notes that reported savings count sitter hours only, and leave out the cost of acquiring, installing and maintaining the technology.
VA Evidence Synthesis Program, 2019 · source
The Joint Commission’s position is that video monitoring alone does not satisfy the requirement for a patient at high risk of suicide: that calls for continuous visual observation by an assigned staff member who can intervene immediately.
Joint Commission FAQ position, as summarised by Barrins & Associates · source
Bathrooms, by policy and by dignity. Rooms where hardware is removed because glass and housings become instruments of harm. Darkness, and anything under a blanket. Radar reads through all of them and produces no image.
Among 98 inpatients with severe mental illness on general adult psychiatric wards, 14 had type II diabetes, 11 had hypertension, 12 had COPD, 22 were obese and 70 smoked.
Hyland, Prime and Carter, BJPsych Open, 2021 · source
In a five-year audit of medical emergencies in an acute mental health unit, the most common trigger for calling the team was simply that staff were worried about the patient, at 34 per cent, ahead of low blood pressure at 23 per cent. The authors call for research into how often vital signs are actually measured in mental health units, as an essential part of catching deterioration early.
Porter et al., International Journal of Mental Health Nursing, 2018 · source
At one Washington state medical centre, about 80 per cent of psychiatric inpatients are covered by Medicaid, the average psychiatric stay is about 16 days, and costs far exceed reimbursement. Across that system’s sixteen psychiatric units, operating income runs at about minus 37 per cent.
Reported by Stateline, March 2026 · source
Continuously, when the patient is still, through bedding and in complete darkness.
Pacing against lying still, and how long either has continued.
Estimated from distance when the sensor is placed near the bed, plus movement in the room.
The sensor produces neither. There is nothing to review, leak or subpoena.
“The Vital Sign Monitoring Sensor (Model XK300) is intended to measure heart rate and respiration rate in adult patients in a general care hospital environment including nursing homes. The Vital Sign Monitoring Sensor can be used for home healthcare for data collection to inform patient care but not to acutely treat a patient. XK300 monitors presence or absence of a patient in a detection area of within 7 meters. The XK300 also monitors the length of continuous patient motion or absence of patient motion.”
Adults. The clearance is for adult patients. Use with anyone younger is the hospital’s own clinical decision, and this page does not make a claim about it.
Device statements, clinical decisions. XK describes what the sensor measures and displays. Your programme decides who reviews it, what is shown, and how it sits alongside your observation policy.
Hourly checks wake people. A patient woken every hour is groggier and more agitated the next day, which is a clinical cost as well as a staffing one. Measuring does not require entering the room.
The sensor sits above the ceiling tile or behind a wall panel and reads through it. No housing, no glass, no fixing, nothing that becomes an anchor point.
It runs on its own cellular connection, so there is no hospital network access, no security review and no infrastructure to provision. Installation is mounting and power.
6.5 to 8.5 GHz impulse ultra-wideband, a 135 degree field of view and 65 degrees of azimuth, covering a single room from one position.
The suicide risk reduction goal requires an environmental risk assessment and action to minimise physical features that could be used in an attempt. Its seven revised elements took effect on 1 July 2019 for hospitals and behavioural health organisations, and on 1 July 2020 for critical access hospitals.
Joint Commission R3 Report Issue 18 · source
From 1 January 2026 general and critical access hospitals work to National Performance Goal 8 under the Accreditation 360 restructure, while behavioural health care and human services organisations continue under National Patient Safety Goal 15.01.01.
Joint Commission Accreditation 360, effective January 2026
The XK300 is FDA 510(k) cleared, K202464, as a Class II prescription device to measure heart rate and respiration rate in adult patients. It is not an alarm, it does not detect self-harm or any other event, and it does not replace observation or clinical judgement.