Extended recovery room as a strategy to reduce inappropriate ICU admissions: an Italian hospital experience

Abstract

The limited operational hours of Recovery Rooms (RR) frequently result in potentially avoidable intensive care unit (ICU) admissions for intermediate-risk surgical patients. “Project Futura,” implemented at Santo Stefano Hospital (Prato, Italy) in March 2024, extended RR hours into the evening (until 24:00) and holidays by reorganizing existing ICU nursing staff without additional resources. This retrospective before-after study compared October–December 2023 (pre-intervention) with October–December 2024 (post-intervention). During the post-intervention period, 362 of 1519 surgical patients (23.83%) received extended RR monitoring; only 18 (1.18%) required subsequent ICU/intermediate care unit admission. ICU occupancy decreased from 91.79 to 73.78% (p < 0.0001), extraordinary emergency RR activations dropped from 22 to 0, estimated ICU costs decreased by 35.7%, and nursing on-call hours were reduced by 91.7%. These preliminary findings suggest that extending RR operational hours may represent a safe, cost-efficient strategy to improve the appropriateness of postoperative ICU utilization.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s44158-026-00430-7.

Dear Editor,

The recent literature addressing optimization of postoperative care pathways and reduction of inappropriate intensive care unit (ICU) admissions has highlighted a critical gap in perioperative care delivery [1]. We wish to share the results of “Project Futura,” an organizational quality-improvement intervention implemented at Santo Stefano Hospital in Prato, Italy, which suggests that extending Recovery Room (RR) operational hours may represent a feasible, clinically appropriate, and resource-efficient strategy to improve postoperative care allocation.

Setting and organizational challenge

Santo Stefano Hospital is a 400-bed acute care community hospital (DEA II level) serving the Prato district (approximately 260,000 inhabitants), with a 10-bed general ICU and a 6-bed Recovery Room. Surgical activity encompasses general surgery, orthopedics, urology, gynecology, vascular surgery, and otolaryngology, with approximately 6000 surgical procedures performed annually. Prior to the intervention, approximately 12% of surgical patients required postoperative ICU admission.

The RR historically operated only on weekdays during daytime hours (7:30–19:30). This limited availability created an organizational gap for surgeries concluding near closing time, emergencies occurring on holidays, and intermediate-risk patients requiring prolonged postoperative monitoring. Consequently, hemodynamically stable patients were frequently admitted to the ICU in the absence of alternative monitored settings, contributing to bed saturation (91.79% occupancy), elevated costs, and unjustified physical and psychological burden for patients and staff.

Intervention design

Initiated in March 2024, Project Futura reorganized existing ICU nursing staff to extend RR operational hours into the evening (until 24:00) and on holidays, without requiring additional financial or human resources. A nurse-to-patient ratio of 1:3 was maintained in the extended RR, with continuous anesthesiologist supervision. A multidisciplinary team of anesthesiologists and intensivists established prospective triage criteria to distinguish patients requiring direct ICU admission from those suitable for extended RR monitoring (minimum 2 h). The complete triage protocol, including inclusion and exclusion criteria, escalation triggers, monitoring standards, and discharge criteria, is provided in Supplementary Table S1. In brief, patients were considered eligible for extended RR monitoring if they were hemodynamically stable (with or without low-dose vasopressor support), had adequate respiratory function (not requiring mechanical ventilation), and had an anticipated monitoring need of less than 12 h. Patients requiring mechanical ventilation, high-dose vasopressor support, invasive hemodynamic monitoring, or continuous renal replacement therapy were admitted directly to the ICU. These criteria were shared with surgical teams through dedicated multidisciplinary meetings prior to implementation.

Patients who were not ready for ward discharge by 24:00 were transferred to the ICU for overnight monitoring and reassessed the following morning. This occurred in 18 cases (4.97% of extended RR patients) during the study period, and is included in the reported ICU/intermediate care unit transfer rate.

The extended RR functionally operated as an intermediate care setting, providing continuous multiparametric monitoring, capacity for low-dose vasoactive infusions, and one-to-one nursing assessment capability, but differing from a formal Intermediate Care Unit (IMCU) or Advanced Recovery Room Care (ARRC) model in that it utilized existing RR infrastructure and ICU nursing staff rather than dedicated IMCU beds or personnel [1, 2].

Study design and statistical methods

This is a single-center, uncontrolled before-after quality-improvement study. The pre-intervention period was defined as October–December 2023 and the post-intervention period as October–December 2024, using the same calendar trimester to minimize seasonal variation. The intervening period (March–September 2024) served as a run-in phase and was excluded from analysis. The primary endpoint was ICU bed occupancy rate, defined as (total ICU patient-days)/(available ICU bed-days) × 100. Secondary endpoints included the rate of subsequent ICU/intermediate care unit transfer after extended RR monitoring, total ICU patient-days, elective surgical ward occupancy, extraordinary emergency RR activations, nursing on-call hours, and estimated ICU costs. Patient characteristics, including age, sex, ASA physical status classification, surgical specialty, and type of surgery (elective vs. emergency), are reported in Supplementary Table S2. Definitions of all outcomes are provided in Supplementary Table S3.

Categorical variables were compared using the chi-square test or Fisher’s exact test as appropriate. Continuous variables were compared using the Student’s t-test or Mann–Whitney U test based on distribution. Statistical significance was set at p < 0.05. All analyses were performed using R software version 4.3.

This quality-improvement project utilized routinely collected, anonymized administrative and operational data (bed occupancy, patient flow, staffing hours, and cost estimates). No individual patient-level clinical data beyond standard administrative records were collected or analyzed. Under Italian legislation governing quality-improvement activities based on aggregated institutional data, formal ethics committee approval was not required. No identifiable patient information is reported.

Clinical and operational outcomes

During the post-intervention trimester, 362 patients (23.83% of 1519 surgical interventions) received extended RR monitoring. Of these, 273 utilized evening hours (19:30–24:00) and 20 utilized holiday coverage. Only 18 patients (4.97% of extended RR patients; 1.18% of total surgical patients) required subsequent ICU or intermediate care unit admission following RR monitoring, suggesting appropriate patient selection.

ICU occupancy decreased from 91.79 to 73.78% (p < 0.0001), with total ICU patient-days declining from 1689 to 1086. Elective surgical ward occupancy increased from 47.57 to 54.76% (p < 0.0001), consistent with successful reallocation of intermediate-risk patients to appropriate care settings. Average ICU length of stay remained unchanged (5.41 vs. 5.35 days, p = [value]), and short-stay ICU admissions (≤ 24 h) showed no significant increase (6.08% vs. 9.85%, p = 0.1148), suggesting that the ICU was more appropriately reserved for higher-acuity patients. Extraordinary emergency RR activations—previously triggered when ICU beds were saturated—decreased from 22 to 0.

We acknowledge that comprehensive safety endpoints, including 30-day postoperative mortality, unplanned ICU readmission within 48 h of ward transfer, Medical Emergency Team activations, cardiac arrest, reoperation, and major postoperative complications, were not systematically collected during this initial observation period. This represents an important limitation, and prospective collection of these outcomes has been incorporated into the ongoing evaluation of Project Futura.

Economic analysis

Estimated total ICU costs for the observed trimester decreased by 35.7% (from €2,838,516.51 to €1,825,120.74). Cost estimates were calculated from a hospital perspective using regional standard costs per ICU patient-day, applied to the observed reduction in ICU patient-days. Detailed methodology, including cost sources, separation of fixed and variable components, and inclusion of operational costs associated with extended RR hours, is provided in Supplementary Table S4. As the intervention utilized existing staff and infrastructure, no additional direct costs were incurred; however, opportunity costs related to redistribution of nursing workload are discussed in the Supplementary Material. Nursing on-call hours decreased by 91.7% (from 139 to 11.5 h).

Implications for nursing workforce

The reduction in on-call hours may contribute to improved nursing well-being, although this was not directly measured in the present study. The prevalence of shift work disorder among nurses is estimated at 45.5% (95% CI 39.2–51.8%), and nursing fatigue has been associated with increased medication administration errors [2]. The operational flexibility gained through Project Futura allowed reinvestment of nursing competencies into strategic clinical initiatives, including a Day-Care Nurse role, an ICU Follow-Up Clinic for Post-Intensive Care Syndrome, and a 24-h Medical Emergency Team. Whether these workforce changes translate into measurable improvements in nursing satisfaction, burnout, or patient safety outcomes remains to be evaluated.

Alignment with international evidence

Our findings are consistent with the emerging paradigm of Advanced Recovery Room Care (ARRC), although important differences in staffing models and patient populations should be noted. Ludbrook et al. demonstrated in a multicenter study that ARRC for intermediate-risk patients significantly increased days at home at 30 days (17 vs. 15 days, p = 0.04) and revealed that intensive early monitoring identified three times more medical emergency-level complications than usual care during the first 24 h (12.4% vs. 3.7%, p < 0.001), while patients subsequently experienced half the rate of complications after ward transfer (2.6% vs. 6.3%, p = 0.03) [1]. This temporal pattern suggests that intensive early monitoring may prevent progression of physiological derangements into serious complications—a mechanism that may also underlie the favorable outcomes observed in Project Futura, although our study was not designed to test this hypothesis.

Cost-effectiveness analysis of ARRC demonstrated 4.3 more days at home within 90 days and decreased hospital costs by $1081 per patient, with a 99.3% probability of increased days at home and 77.4% probability of cost dominance [3]. A 2025 meta-analysis of extended postoperative recovery units reported pooled mortality of 3% (95% CI 2–6%), with significantly lower mortality in enhanced perioperative care units (2%) compared with traditional ICU admissions (8%) [4].

The extended RR also shares functional characteristics with Intermediate Care Units. A nationwide Japanese study of over 2.2 million ICU patients demonstrated that hospitals with IMCUs had lower in-hospital mortality (aOR 0.94), lower ICU mortality (aOR 0.87), fewer ICU readmissions (aOR 0.92), and shorter ICU length of stay [5]. A Dutch single-center observational study demonstrated that IMCUs avoided ICU admission in 79.9% of patients with potential ICU treatment needs, with only a 6.9% transfer rate to the ICU, and reported cost differentials of €463/day for ward care, €1307/day for IMCU, and €2224/day for ICU [6]. The Society of Critical Care Medicine guidelines recommend that patients with risk factors for postoperative instability be monitored in a higher level of care unit than the ward during the immediate postoperative period [1].

Limitations

Several limitations should be acknowledged. This is a single-center, uncontrolled before-after study, and the observed associations cannot establish causality. The short post-intervention observation period (3 months) limits assessment of sustained effects. Although the same calendar trimester was used for comparison to minimize seasonal variation, we cannot exclude the influence of unmeasured confounders, including changes in surgical volume, case-mix complexity, emergency surgery burden, ICU admission policies, or broader hospital operational factors unrelated to the intervention. Comprehensive safety endpoints were not systematically collected. The economic analysis is based on estimated costs rather than actual expenditure tracking, and fixed ICU costs do not decrease linearly with reduced occupancy. Finally, generalizability to hospitals with different ICU/RR staffing models, surgical case-mix, or organizational structures may be limited.

Conclusion

Project Futura suggests that extending RR operational hours through strategic reorganization of existing resources is a feasible intervention associated with favorable clinical, operational, and economic outcomes. The observed reduction in ICU occupancy (from 91.79 to 73.78%), estimated cost savings (35.7%), and decreased nursing on-call burden (91.7%) support the potential value of extended Recovery Room care as a component of comprehensive perioperative care optimization. These preliminary findings warrant confirmation through longer observation periods, multicenter evaluation, and prospective collection of comprehensive safety and patient-centered outcomes.

Authors’ contributions

MS, SB, and IC conceptualized the study. MS collected and organised the data. VP and IC performed formal analyses. MS and IC drafted the manuscript. All authors contributed to the critical revision of the manuscript for important intellectual content. All authors approved the final version and agree to be accountable for all aspects of the work.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Competing interests

The authors declare no competing interests.