**Background:** Rapid response systems (RRS) and the National Early Warning Score (NEWS) are widely used to detect deteriorating ward patients, but cardiac arrests and deaths still occur. Nurses' clinical judgment and 'gut feeling' may identify at-risk patients earlier than vital sign thresholds alone. A nurse-led critical care outreach service (NLCCOS) was hypothesized to support ward nurses through bedside education, shared decision-making, and pre-emptive interventions, potentially reducing the need for Medical Emergency Team (MET) calls.
**Methods:** This mixed-methods pilot study was conducted at Zealand University Hospital, Koege, Denmark, from December 2018 to May 2019 in one orthopedic (30 beds) and one medical (30 beds) ward. A convenience sample of 100 patients rated as 'at-risk' by ward managers (based on nurse worry, difficulty stabilizing, or other clinical concerns, regardless of NEWS) was enrolled. Exclusion criteria were patients requiring MET or immediate ICU transfer. The NLCCOS (trained ICU nurses, usually part of the MET) visited the ward daily (Monday–Friday) after a morning conference. They assessed the patient jointly with the ward nurse, performed interventions (e.g., suction, CPAP, IV insertion), and provided bedside education. Data were collected via a Case Report Form (CRF) capturing vital signs, treatments, and care plans. Ward nurses completed a short questionnaire (pilot-tested on 10 people) after each visit, rating confidence and describing learning areas. Quantitative data were analyzed with SPSS (medians/IQR, means/SD, frequencies). Qualitative free-text responses were analyzed using inductive thematic analysis by two independent authors.
**Key Results:** Over six months, 100 patients were reviewed (51 medical, 49 surgical). Most patients (71%) had compromised respiratory status requiring supplemental oxygen; 23% used PEP and 29% used CPAP. Median vital signs included: respiratory rate 20, saturation 94%, heart rate 91 bpm, temperature 37°C. One third had a compromised conscious state (AVPU V,P,U: 33%). 73% were to receive full resuscitation; 18% had treatment limitations. After one month, 40% of patients had died. Sixty-one ward nurse surveys were returned (61% response rate; 27 surgical, 34 medical). Over 90% (n=55) of respondents reported learning something from the NLCCOS visit and felt more confident in caring for the patient afterward. Thematic analysis identified key educational areas: respiratory therapy strategies (suction, CPAP, PEP, oxygen), invasive procedures (arterial blood gas, gastric tube, catheter), medications (diuretics, analgesics, inhalations), and mobilization. Thirteen nurses reported no learning, citing reasons such as the patient had improved, the nurse already knew what to do, the patient was palliative, or the nurse was unable to participate.
**Clinical Implications:** The NLCCOS model appears feasible and acceptable in a single-site setting. It addresses gaps in ward nurses' clinical decision-making, particularly around respiratory management and fundamental care (mobilization, nutrition). The high one-month mortality (40%) suggests the NLCCOS may also help facilitate discussions about goals of care and appropriate treatment limitation. The study highlights the importance of combining track-and-trigger systems with skilled clinical judgment and 'nurse worry' to identify at-risk patients earlier. The authors recommend further research using randomized designs to test effectiveness and measure patient outcomes.