Written and maintained by CASRAI Editorial Board
Last updated
The Richmond Agitation-Sedation Scale (RASS) is the most widely used sedation and agitation instrument in adult intensive care, and it is a two-way scale: it measures excess agitation as well as excess sedation on the same 10-point line. It was developed by a critical-care team at Virginia Commonwealth University Health System in Richmond, Virginia — led by Curtis N. Sessler — and validated in two published studies — Sessler et al. (2002) and a follow-up multicenter reliability/validity study by Ely et al. (2003) — and it now sits at the center of the Society of Critical Care Medicine’s PADIS guidelines and the ABCDEF liberation bundle. This page covers the full 10-point scale, the verbal-then-physical assessment steps, target ranges by clinical context, the reliability evidence, and the specific rule that ties RASS to CAM-ICU delirium screening.
What RASS is, and why it replaced older sedation scales
Before RASS, ICU sedation was commonly scored with instruments like the Ramsay Scale or the Sedation-Agitation Scale (SAS), which either compressed agitation into a single top category or used vaguer behavioral anchors. RASS was designed and validated specifically to give agitation the same granularity as sedation — four distinct levels above the calm/alert midpoint and five distinct levels below it — and to standardize exactly how a clinician elicits the response, not just how they interpret it. Sessler CN et al., “The Richmond Agitation-Sedation Scale: Validity and Reliability in Adult Intensive Care Unit Patients,” American Journal of Respiratory and Critical Care Medicine 166(10):1338–1344 (2002) is the original validation, conducted across medical, surgical, and trauma ICU patients, ventilated and non-ventilated, sedated and unsedated. A second, larger multicenter study — Ely EW et al., “Monitoring Sedation Status Over Time in ICU Patients: Reliability and Validity of the Richmond Agitation-Sedation Scale (RASS),” JAMA 289(22):2983–2991 (2003) — confirmed the instrument held up across a broad mix of rater types (bedside nurses, physicians, and research staff) and ICU settings, and it is this second study that anchored RASS into the sedation-monitoring literature that followed, including the CAM-ICU delirium tool.
The 10-point scale, +4 to -5
Each level has both a one- or two-word label and a specific behavioral description — the description, not the label, is what a rater actually scores against.
| Score | Term | Description |
|---|---|---|
| +4 | Combative | Overtly combative or violent; immediate danger to staff |
| +3 | Very agitated | Pulls or removes tube(s) or catheter(s); aggressive toward staff |
| +2 | Agitated | Frequent nonpurposeful movement; fights the ventilator |
| +1 | Restless | Anxious or apprehensive but movements not aggressive or vigorous |
| 0 | Alert and calm | Spontaneously attentive to the examiner without prompting |
| -1 | Drowsy | Not fully alert, but has sustained (>10 seconds) eye contact and awakening to voice |
| -2 | Light sedation | Briefly (<10 seconds) awakens with eye contact to voice |
| -3 | Moderate sedation | Any movement (but no eye contact) to voice |
| -4 | Deep sedation | No response to voice, but any movement to physical stimulation |
| -5 | Unarousable | No response to voice or physical stimulation |
The assessment procedure: observe, then voice, then touch
RASS is deliberately administered as a fixed sequence, not a holistic impression — the same three-step order every time is what makes it reproducible across raters:
- Observe the patient. If the patient is spontaneously alert, calm, and attentive without any prompting, score 0. If the patient shows agitation, restlessness, or combative behavior on observation alone, score +1 through +4 against the descriptions above and stop — no further stimulation is needed or appropriate.
- If not alert on observation, apply verbal stimulation. State the patient’s name in a normal voice and ask them to open their eyes and look at the examiner, repeating once if needed. Sustained eye contact lasting more than 10 seconds scores -1. Eye contact that occurs but is not sustained beyond 10 seconds scores -2. Any movement in response to the voice that stops short of eye contact scores -3.
- If there is no response to voice, apply physical stimulation. The standard sequence is to shake the patient’s shoulder first, and if there is still no response, follow with a sternal rub. Any movement in response to physical stimulation scores -4. No response to either voice or physical stimulation scores -5.
The assessment is meant to take well under a minute at the bedside, and because it is criterion-based rather than impressionistic, it is designed to be repeatable by whichever clinician is at the bedside at a given hour — not just the nurse who has been with the patient all shift.
Target RASS ranges by clinical context
RASS is a monitoring instrument, not a treatment protocol by itself, but the Society of Critical Care Medicine’s Pain, Agitation/sedation, Delirium, Immobility, and Sleep disruption (PADIS) guidelines and the associated ABCDEF liberation bundle both frame sedation depth as a deliberate target to be reassessed regularly, not a side effect of whatever analgesic or sedative infusion happens to be running:
- Light sedation, generally RASS -1 to -2, is the default target for most adult mechanically ventilated ICU patients under current light-sedation-first guidance, unless a specific clinical indication argues against it. Light sedation is associated with shorter ventilator time and shorter ICU stay compared with deep sedation as a routine default.
- Deeper sedation (RASS -3 and below) is reserved for specific indications rather than used as a default — severe ARDS requiring prone positioning or neuromuscular blockade, refractory intracranial hypertension, status epilepticus, or active therapeutic hypothermia are the recurring examples in the critical-care literature, and even then the target should be an explicit, reassessed RASS number, not an open-ended infusion.
- A RASS target should be written as a specific order (for example, “target RASS -1 to 0”) and reassessed at a fixed interval, so sedation gets titrated to the number rather than the number simply being recorded after the fact.
Interrater reliability and validation
RASS’s reliability evidence is a large part of why it displaced earlier sedation scales in ICU practice. The original Sessler et al. (2002) validation found strong agreement between raters (a weighted-kappa statistic in the high range) across different rater types — staff nurses, research nurses, and physicians — and across different patient populations, including ventilated and non-ventilated, sedated and unsedated patients. The Ely et al. (2003) follow-up, run at two academic medical centers across a broader mix of ICU types and rater professions (nurses, physicians, and research staff), reported similarly strong agreement (a high intraclass correlation coefficient), which is what gave RASS the multicenter evidence base needed to be adopted as the sedation instrument underlying the CAM-ICU delirium tool described below. In practical terms, this reliability evidence is the reason a RASS score charted by one clinician can be trusted by the next shift without re-examining the patient from scratch.
RASS and CAM-ICU: why sedation depth gates delirium screening
The Confusion Assessment Method for the ICU (CAM-ICU) is the standard bedside delirium-screening tool for mechanically ventilated and non-verbal ICU patients, and it is built directly on top of RASS rather than run independently of it. A valid CAM-ICU assessment requires the patient to be responsive enough to attend to simple commands, so the protocol requires checking RASS first: CAM-ICU should only be attempted when the current RASS score is -3 or higher (that is, -3, -2, -1, 0, or a positive score). If the RASS score is -4 or -5 — deep sedation or unarousable — the patient cannot be meaningfully assessed for the presence or absence of delirium, and the standard instruction is to stop, document the patient as unable to be assessed at that time, and re-check RASS (and reattempt CAM-ICU) later, rather than forcing a delirium call on a patient who cannot follow the CAM-ICU commands in the first place. This sequencing — RASS first, CAM-ICU only if RASS clears the -3 threshold — is why the two instruments are taught and charted together rather than as separate, unrelated tools.
Documentation and practical use
Because RASS is designed to be reassessed on a fixed schedule (commonly with each set of vital signs, and again after any sedative or analgesic dose change), the value of a single RASS reading is limited — what a chart audit or a quality review actually looks for is a trend: whether the score stayed inside its ordered target range, and whether an out-of-range score triggered a documented response (a sedation adjustment, a physician notification, or a repeat assessment) rather than simply being recorded and left alone. Assessing RASS accurately also depends on documenting confounders that could distort the score independent of true sedation depth — pre-existing neurologic injury, deafness or language barriers affecting the verbal-stimulation step, and neuromuscular blockade, which makes the motor-response components of the scale unusable and requires a different monitoring approach entirely (such as processed EEG) until the blockade is reversed.
Frequently asked questions
What does a RASS score of 0 mean?
A RASS score of 0 means the patient is alert and calm — spontaneously attentive to the examiner without any need for verbal or physical prompting. It is the scale’s midpoint, sitting between the four agitation levels (+1 to +4) and the five sedation levels (-1 to -5).
What is the target RASS score for most ICU patients?
Under current light-sedation-first ICU guidance, the typical target for most mechanically ventilated adult patients is RASS -1 to -2, reassessed regularly and titrated to that specific target rather than left to whatever the sedative infusion happens to produce. Deeper targets are reserved for specific clinical indications, not used as a default.
Why does CAM-ICU require a RASS of -3 or higher?
CAM-ICU depends on the patient being able to attend to simple commands, which a patient at RASS -4 or -5 (deep sedation or unarousable) cannot reliably do. Attempting CAM-ICU below RASS -3 does not produce a meaningful delirium result — the correct step is to document the patient as unable to be assessed and re-check RASS later, not to force a score.
How often should RASS be reassessed?
RASS does not specify its own reassessment interval — that is a unit-policy decision — but it is commonly checked with each routine vital-signs round and again after any sedative or analgesic dose change, since the whole point of a numeric target is to catch drift quickly rather than after a full shift has passed.
Can RASS be used on a patient receiving neuromuscular blockade?
Not reliably. Neuromuscular blockade removes the motor response that most of the scale’s levels depend on, so RASS cannot distinguish true sedation depth from pharmacologic paralysis in that state; guidance instead points to objective monitoring such as processed EEG until the blockade is reversed.
Braden Scale and Morse Fall Scale are the same kind of validated, point-scored bedside instrument applied to pressure-injury and fall risk rather than sedation — see also the Johns Hopkins Fall Risk Assessment Tool for how a second fall-risk instrument differs from Morse in structure and scoring. Back to the CASRAI Patient Safety hub for surveillance definitions, root cause analysis, credentialing, and the rest of the hospital patient-safety and infection-prevention library.
