Title
Author
DOI
Article Type
Special Issue
Volume
Issue
Non-pharmacological delirium care in ICU: a systematic review
1Medical Intensive Care Unit, University Medical Centre Maribor, 2000 Maribor, Slovenia
2Faculty of Health Sciences, University of Maribor, 2000 Maribor, Slovenia
*Corresponding Author(s):andrej.cerni@ukc-mb.si (Andrej Černi)
† These authors contributed equally.
| History | Submitted: 10 July 2025 | Accepted: 25 September 2025 | Published: 08 January 2026 |
| Copyright: | ©2026 The Author(s). Published by MRE Press. |
Background: Delirium is a frequent and serious complication in critically ill patients, particularly in intensive care units (ICUs). It is associated with prolonged hospital stays, increased morbidity, and long-term cognitive impairment. Non-pharmacological interventions are considered safe alternatives to pharmacological treatments and align with principles of person-centered care. The aim of this review is to evaluate and synthesize evidence on the effectiveness of non-pharmacological interventions in preventing and managing delirium in critically ill adult ICU patients. Methods: A systematic review was conducted using PubMed, Web of Science, and the Cochrane Library. Included studies assessed non-pharmacological interventions in adult ICU patients, reporting on delirium-related outcomes. Thematic synthesis and quality appraisal using the Grading of Recommendations Assessment, Development, and Evaluation (GRADE) framework were performed. Results: Multicomponent interventions—particularly ABCDE/ABCDEF bundles—were most consistently effective in reducing delirium incidence and duration. Early mobility and structured physical activity showed positive effects when implemented early and consistently. Family engagement interventions, especially structured visitation and reorientation, were also effective. Other strategies, including cognitive stimulation, music therapy, and environmental modifications, showed mixed results. The quality of evidence ranged from low to high. Conclusions: Non-pharmacological interventions, especially when multicomponent and protocol-driven, are effective in managing delirium in critically ill patients. Their integration into ICU practice can reduce the delirium burden and improve patient outcomes. The PROSPERO Registration: CRD420251151545.
Cite this article
Andrej Černi, Andrej Markota, Leona Cilar Budler. Non-pharmacological delirium care in ICU: a systematic review. Signa Vitae. 2026; 22(1): 20-33. doi: 10.22514/sv.2025.182
1. Introduction
Delirium is a prevalent and serious complication in critically ill patients, particularly within intensive care units (ICUs). Characterized by acute disturbances in attention, cognition, and perception, delirium affects a significant proportion of ICU patients, with estimates suggesting that up to 70% of mechanically ventilated patients may experience this condition during their stay [1]. The incidence of delirium can vary widely, influenced by factors such as the severity of illness, use of sedatives, and pre-existing conditions [2, 3]. Notably, delirium is associated with adverse clinical outcomes, including increased morbidity, prolonged hospital stays, and elevated mortality rates [4, 5].
The implications of delirium extend beyond immediate clinical outcomes. Research indicates that patients who experience delirium are at a heightened risk for long-term cognitive impairment and diminished quality of life following hospitalization [6]. For instance, sedative-associated delirium has been identified as a modifiable phenotype that can significantly impact long-term cognitive function and disability [6]. Furthermore, caregivers of patients with delirium often report experiencing psychological distress, including symptoms of depression and anxiety, highlighting the broader emotional toll of this condition [7].
Despite its high prevalence and serious consequences, delirium often remains underdiagnosed and inadequately managed in ICU settings. Effective screening tools, such as the Confusion Assessment Method for the ICU (CAM-ICU), have been developed to facilitate the early detection of delirium [8]. However, the implementation of evidence-based management strategies, particularly non-pharmacological interventions, has been inconsistent across healthcare settings [9]. Multicomponent interventions that address the multifactorial nature of delirium have shown promise in reducing its incidence and duration, yet many ICUs have been slow to adopt these practices [9, 10].
Non-pharmacological interventions are increasingly recognized as essential components of delirium prevention and management in critically ill patients. These strategies target modifiable risk factors, such as sensory deprivation, immobility, disrupted sleep-wake cycles, and social isolation—factors that are often exacerbated in the ICU environment. Interventions such as early mobilization, orientation protocols, family engagement, noise and light control, cognitive stimulation, and multicomponent care bundles (e.g., ABCDE and ABCDEF) have demonstrated effectiveness in both reducing delirium incidence and shortening its duration. Compared with pharmacological approaches, non-drug interventions offer a safer alternative with fewer adverse effects and are aligned with the principles of person-centered and holistic care [11, 12]. Nonetheless, variability in implementation, staff training, and institutional culture remains a barrier to their widespread use.
Given the growing body of evidence supporting these approaches, there is a pressing need to systematically evaluate and synthesize current findings to guide clinical practice and policy. This systematic review aims to consolidate the available evidence on non-pharmacological strategies for delirium prevention and management in ICU patients, with the goal of informing evidence-based guidelines and improving the quality of care for this high-risk population.
The aim of this systematic review is to evaluate and synthesize current evidence on the effectiveness of non-pharmacological interventions for the prevention and management of delirium in critically ill patients admitted to intensive care units. By identifying the most effective strategies and examining their implementation contexts and outcomes, this review seeks to inform the development of evidence-based clinical practice guidelines and support the integration of non-drug interventions into routine ICU care.
2. Materials and methods
The systematic review was carried out following the Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) [13] approach: (1) formulating the research question, (2) conducting an initial literature search, (3) constructing the search string and establishing inclusion and exclusion criteria, (4) performing a comprehensive literature search and analysis, (5) synthesizing the gathered literature, (6) evaluating the quality of the studies and identifying potential biases, and (7) interpreting the results and formulating recommendations, as seen in Fig. 1. PRISMA 2020 Checklist is provided in Supplementary material.
Fig. 1.PRISMA 2020 flow diagram.
The following research question was developed using PIO framework [14]: Which non-pharmacological interventions (I) are most effective in the management of delirium (O) in critically ill patients (P)?
The literature search was performed between 13 and 24 March 2025 across multiple electronic databases, including PubMed, Web of Science, and the Cochrane Library. The search was limited to studies published in English involving adult patients aged over 18 years. All inclusion and exclusion criteria in the literature search process are presented in Table 1. Search string was as follows: (“ICU” OR “intensive care” OR “critically ill”) AND ((“non-pharmacologic*” OR “nonpharmacologic*”) AND (“delirium prevention” OR “delirium management”)) AND (“delirium” OR “cognitive function” OR “functional recovery” OR “confusion”) AND (“cognitive stimulation” OR “early mobilization” OR “music therapy” OR “family involvement” OR “reorientation”).
| Inclusion criteria | Exclusion criteria | |
| Intervention | Non-pharmacological strategies for the management of delirium in critically ill patients. | Only pharmacological treatment of delirium. |
| Study type | Randomized controlled trials, cohort studies, systematic reviews and meta-analyses. | Literature reviews without systematic methodology. |
| Outcome | Relevant clinical outcomes (incidence, duration or severity of delirium). | Not involving critically ill patients. |
| Population | Adult patients (>18 years). | Pediatric or neonatal population. |
| Study access | Fully available articles. | Articles without available full texts. |
We included randomized controlled trials (RCTs) and cohort studies as the primary sources of evidence, given their strength in establishing causal inference and generalizability. However, due to the limited number of RCTs available in this field, we also incorporated quasi-experimental studies, systematic reviews, and meta-analyses to provide a broader overview of current knowledge. This inclusive approach allowed us to capture a wider scope of evidence on non-pharmacological interventions for delirium management in critically ill patients. The decision to include multiple study designs is consistent with approaches recommended in fields where RCT evidence is scarce, but where clinically relevant insights may be gained from alternative study types.
Two independent reviewers screened the titles and abstracts based on predefined inclusion and exclusion criteria. Full-text articles of eligible studies were reviewed, and any discrepancies were resolved by consensus or consultation with a third reviewer. Data extraction included reference, study design, sample, intervention details, outcomes measured, and key findings.
The quality of the data was evaluated using the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) system. This approach assesses five key aspects: study limitations, imprecision, inconsistency, indirectness, and publication bias. The GRADE system categorizes quality levels as high (++++), moderate (+++), low (++), or very low (+) [15].
A narrative synthesis was conducted for all the included studies. Results were synthesized by the authors.
3. Results
The systematic literature review of 47 studies reveals a growing body of evidence supporting the use of non-pharmacological interventions for preventing and managing delirium in critically ill patients in ICUs. These studies employed diverse methodological approaches, including RCTs, quasi-experimental designs, prospective and retrospective cohort studies, systematic reviews and meta-analyses, as well as qualitative and mixed-methods research. The target populations varied, encompassing adult ICU patients with different clinical profiles, including those who were mechanically ventilated, post-surgical, or elderly.
Across the included studies, 22 (47%) reported positive effects of non-pharmacological strategies, 9 (19%) reported no significant effect, 14 (30%) demonstrated mixed or unclear findings, and 3 (6%) were qualitative or survey-based and did not provide direct effectiveness data. Interventions with the most consistent evidence of benefit were multicomponent bundles and family engagement strategies, both associated with significant reductions in delirium incidence, severity, or duration. Early mobilization and structured rehabilitation also showed favorable outcomes, particularly in RCTs targeting elderly or ventilated patients. By contrast, single-component interventions, such as music therapy, mindfulness, or environmental modifications, produced more variable results, with some smaller studies suggesting potential benefit but larger trials often reporting no significant differences.
A detailed overview of study characteristics, including interventions, assessment tools, and effectiveness, is provided in Table 2 (Ref. [16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35, 36, 37, 38, 39, 40, 41, 42, 43, 44, 45, 46, 47, 48, 49, 50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60]).
Based on a thematic analysis of 47 studies, six major themes were identified. These themes reflect the most recurrent and evidence-supported approaches to delirium care, and they provide a structured framework for understanding both the diversity of interventions and their respective strengths in terms of clinical effectiveness and evidence quality.
| Reference | Study design | Sample | Intervention | Results | Quality of the evidence (GRADE) | Effectiveness |
| Álvarez et al. [40], 2017 | RCT, pilot | 140 non-intubated ICU patients ≥60 years | Early and intensive OT (2×/day for 5 days); polysensory & cognitive stimulation, BADLs, family engagement | Lower incidence and severity of delirium compared to control | Low (pilot design, specific population) | Mixed/Unclear |
| Arbabi et al. [16], 2018 | Quasi-experimental | 148 general ICU patients >18 years | Multicomponent bundle (education, environment, early mobility) | Reduced incidence of delirium; better awareness among staff | Moderate (implementation project, not randomized) | Positive—reduced incidence |
| Balas et al. [17], 2014 | Pre-post prospective cohort | 296 ICU patients ≥19 years | ABCDE bundle (awakening/breathing trials, delirium monitoring, early mobility) | Reduction in delirium prevalence and ICU days with delirium | Moderate (large sample, good assessment frequency) | Positive—reduced prevalence |
| Bannon et al. [18], 2018 | Qualitative study (focus groups) | 68 ICU staff, 12 survivors, 2 family members | Multicomponent bundle for delirium prevention | Bundle deemed acceptable; barriers included resource limitations | Very low (qualitative, no outcome data) | Qualitative/Indirect |
| Bannon et al. [19], 2019 | Systematic review and meta-analysis | 15 RCTs, 2812 patients | Various non-pharmacologic strategies | No significant effect in most trials; one showed benefit of voice reorientation | Low to very low (heterogeneity, inconsistent results) | Mixed/Unclear |
| Bounds et al. [20], 2016 | Retrospective study | 159 ICU patients ≥18 years, ICU stay >24 h | ABCDE bundle (medication choice, mobility, coordination) | Lower prevalence of delirium in intervention group | Low (retrospective design) | Positive—lower prevalence |
| Bryczkowski et al. [21], 2014 | Pre-post prospective cohort | 123 SICU patients >50 years | Multicomponent bundle (sedation, sleep hygiene, education) | Reduced duration of delirium and improved satisfaction | Moderate (applied bundle, not RCT) | Positive—reduced incidence |
| Campbell [36], 2014 | Evidence-based project | 58 ICU patients ≥18 years on ventilation ≥48 h | Early mobility protocol post-sedation interruption | No significant change in delirium incidence | Low (small sample, limited design) | Null—no change |
| Chai [22], 2017 | Pre-post, quasi-experimental QI | 301 mixed ICU patients >18 years | ABCDEF bundle (pain, awakening, mobility, family engagement) | Reduction in delirium incidence | Moderate (quality improvement context) | Positive—reduced incidence |
| Colombo et al. [23], 2012 | Quasi-experimental | 314 (Control = 170; Intervention = 144) | Multi-component: reorientation, visual/acoustic stimulation, sensory deprivation | Incidence: 35.5% vs. 22% (p = 0.02) | Moderate | Positive—reduced incidence |
| Damshens et al. [50], 2018 | RCT | 80 ICU trauma patients >15 years | Music therapy (2×/day, 45 min, instrumental) | No significant difference in incidence | Low (small RCT) | Null—no significant difference |
| Eghbali-Babadi et al. [43], 2017 | RCT | 68 post-surgical ICU patients (18–70 years) | Family visitation (30–40 min, with sensory aids, reorientation) | Significant reduction in delirium incidence | Moderate (well-controlled, focused design) | Positive—reduced incidence |
| Eldean et al. [24], 2024 | Quasi-experimental | 60 mechanically ventilated ICU patients | Daily application of ABCDEF bundle for 7 days | Delirium incidence: 20% (study) vs. 70% (control); p = 0.001 | Moderate—statistically significant results in a real-world ICU; lacks RCT control | Positive—reduced incidence |
| Fallahpoor et al. [56], 2016 | Action research | 100 CABG ICU patients >18 years | 3-phase management model (before, during, after surgery) | Reduction in delirium incidence | Moderate (complex multi-phase design) | Mixed/Unclear |
| Giraud et al. [41], 2016 | RCT, pilot time-cluster | 223 ICU patients ≥70 years after cardiac surgery | Structured mirror usage during awakening, nursing care, procedures | No significant effect on incidence or duration of delirium | Low (pilot design, innovative method) | Null—no effect |
| Gómez Tovar et al. [57], 2024 | RCT | 213 ICU patients (≥18 years old) | DyDel nursing model: structured care plan delivered each shift | Delirium incidence: 5.6% vs. 14.8%; p = 0.037 | High—well-designed RCT with significant clinical outcomes and adequate sample | Positive—reduced incidence |
| Guo et al. [25], 2016 | RCT | 160 oral cancer patients aged 65–80 | Multicomponent non-pharm bundle (reorientation, sleep, sensory aids, music) | Significant reduction in incidence of delirium | Moderate (well-structured design) | Positive—reduced incidence |
| Hamzehpour et al. [26], 2017 | RCT | 100 ICU patients >18 years | Roy adaptation model (fluid balance, sleep, nutrition, oxygen, monitoring) | Significant improvement in confusion scores (NEECHAM) | Moderate (structured nursing model, measurable effects) | Positive—improved scores |
| Johnson et al. [44], 2024 | Mixed-methods pilot study | 15 patients, 15 family members | Family voice reorientation intervention | Feasible, acceptable; suggests psychological benefit | Low (pilot study, small sample) | Qualitative/Indirect |
| Karadas & Ozdemir, 2016 | RCT | 94 ICU patients ≥65 years | ROM exercises once daily (10 reps/exercise, passive-active) | No significant difference in delirium duration or incidence | Low (elderly population, physical therapy focused) | Null—no effect |
| Kersten & Reith, 2016 | Narrative review | Not applicable | Summary of guidelines and delirium management in ICU | Highlights the importance of delirium diagnosis and management | Very low (review article, not original research) | Qualitative/indirect |
| Khan et al. [58], 2014 | Pre-post implementation study | 702 ventilated ICU patients ≥18 years | “Wake Up and Breathe” protocol (daily sedation interruption & breathing trial) | No significant difference in incidence or prevalence | Low (large sample, non-randomized) | Null—no effect |
| Khan et al. [51], 2020 | RCT | 52 ICU patients on mechanical ventilation | Personalized music, slow-tempo music, or audiobook control | Music well tolerated; no significant difference in delirium/coma-free days | Low (pilot nature, small sample, no significant effect) | Null—no effect |
| Kram et al. [27], 2015 | Pre-post implementation study | 83 ICU patients | ABCDE bundle (awakening/breathing trials, coordination, early mobility) | Lower delirium prevalence observed | Moderate (implementation with measurable outcome) | Mixed/Unclear |
| Li et al. [52], 2025 | Systematic Review and Network Meta-Analysis | Multiple RCTs covering postoperative ICU patients | Non-pharmacological sleep interventions | RR = 0.32 to 0.61 depending on method; improved sleep | High—network meta-analysis with strong effects and large pooled data set | Positive—bundled care effective |
| Lisann-Goldman et al. [59], 2019 | Mixed-methods pilot study | 25 cardiac surgery patients ≥40 years | Mindfulness exercises (pre- & post-op Langerian approach) | No delirium observed in either group | Low (pilot study, small sample) | Mixed/Unclear |
| Lundström et al. [28], 2005 | Quasi-experimental | 400 (Control = 200; Intervention = 200) | Multi-component: staff education, patient-allocation, nursing guidance | No difference in incidence (31% vs. 31.5%, p = 0.91); duration significantly lower (59.7% vs. 30.2% on day 7, p = 0.001) | Moderate | Mixed/Unclear |
| Ma et al. [45], 2024 | Qualitative Study | 17 participants (6 patients, 11 family) | Auditory stimulation and family involvement | Improved emotional support and orientation | Moderate (Qualitative)—rich thematic insight but lacks generalizability and control | Mixed/Unclear |
| Mailhot et al. [46], 2017 | RCT | 30 patient/family caregiver dyads post-cardiac surgery | Mentoring-based intervention involving family in delirium management | Improved psycho-functional recovery; similar delirium severity in both groups | Low (small sample size, pilot study design) | Positive—reduced severity |
| Martinez et al. [29], 2012 | Quasi-experimental | 287 (Control = 60; Intervention = 227) | Multi-component: reorientation, sensory deprivation, sleep hygiene, family, early mobilization | Incidence: 38% vs. 24% (p = 0.03); Duration: 5.6 vs. 3.5 days (p = 0.13) | Moderate | Positive—reduced incidence |
| Matsuura et al. [30], 2022 | Network meta-analysis | 11 studies, 2549 patients | Multicomponent non-pharmacologic interventions | SP-CS-EM-PC-AS and SP-CS effective (OR 0.46–0.47) | Moderate to high (systematic method, consistent findings) | Positive—multicomponent effective |
| Moon & Lee [31], 2015 | RCT | 123 ICU patients ≥18 years | Multicomponent bundle (reorientation, early ambulation, sensory aids, etc.) | Significant reduction in delirium incidence | Moderate (structured, controlled) | Positive—reduced incidence |
| Munro et al. [42], 2017 | RCT | 30 ICU patients >18 years | Automated reorientation messages 8×/day | Increased delirium-free days in the intervention group | Low (small but innovative approach) | Mixed/Unclear |
| Parry et al. [37], 2014 | Case-matched control study | 16 ICU sepsis patients ventilated >48 h | Functional electrical stimulation (FES) cycling (20–60 min/day) | Reduced delirium duration (not statistically significant) | Low (small sample, case-control design) | Mixed/Unclear |
| Pun et al. [32], 2019 | Prospective multicenter cohort | 10,840 ICU patients | ABCDEF bundle (full vs. partial implementation) | Significant reduction in delirium with full implementation | High (large sample, strong prospective design) | Positive—reduced incidence |
| Rivosecchi et al. [33], 2016 | Pre-post QI project | 483 medical ICU patients | M.O.R.E. bundle (music, orientation, reorientation, eye/ear care) | Reduced incidence and duration of delirium | Moderate (implementation design, good outcomes) | Positive—reduced incidence |
| van Rompaey et al. [53], 2012 | RCT | 136 (Control = 67; Intervention = 69) | Use of earplugs at night | Higher NEECHAM score in intervention (26 vs. 24, p = 0.04), suggesting lower incidence | Moderate | Positive—reduced incidence |
| Rosa et al. [47], 2017 | Pre-post prospective study | 286 ICU patients ≥18 years | Extended visitation hours (12 h/day, family participation) | Significant reduction in incidence and duration of delirium | Moderate (clear intervention, measurable benefit) | Positive—reduced incidence |
| Schweickert et al. [38], 2009 | RCT | 104 (Control = 55; Intervention = 49) | Early exercise and mobilization (daily physical & occupational therapy) | Duration: 4.0 vs. 2.0 days (p = 0.03) | Moderate | Positive—reduced duration |
| Simons et al. [54], 2016 | RCT | 734 ICU patients ≥18 years | Dynamic Lighting Application (blue-white light in AM, dim in PM) | No significant effect on delirium incidence or duration | Moderate (large sample, novel environmental control) | Null—no effect |
| Smithburger et al. [48], 2017 | Cross-sectional survey | 60 nurses, 58 physicians, 60 family members | Survey on opinions about family involvement | High willingness from families and providers; gaps in communication noted | Very low (descriptive, no intervention tested) | Mixed/Unclear |
| Sonia et al. [49], 2024 | Umbrella Review | 12 studies | Multicomponent bundles including family and mobility | Multicomponent strategies most effective; family role crucial | High—synthesis of multiple systematic reviews; consistent findings for ICU interventions | Positive—reduced severity |
| Spies et al. [55], 2024 | Prospective Observational Pilot Study | 74 ICU patients in 2 ICU room types | ICU room design with dynamic lighting | Modified room reduced delirium severity; serum melatonin linked | Moderate—pilot study with significant effect, but small sample and exploratory design | Mixed/Unclear |
| Sullinger et al. [34], 2017 | Pre-post, retrospective observational | 89 ICU patients ≥18 years with acute delirium | Management bundle + nursing education (mobility, sensory support, massage, etc.) | Significant reduction in delirium days | Moderate (promising results, retrospective design) | Positive—non-pharma moderately effective |
| Veronese et al. [35], 2024 | Umbrella Review of RCTs | 59 systematic reviews, 110 meta-analytic estimates | Multiple non-pharmacologic and pharmacologic approaches | Non-pharma methods moderately effective; variable evidence quality | Moderate to High—umbrella review with consistent effects; based on 485 RCTs, though quality varied | Positive—reduced incidence |
| Zhang et al. [60], 2017 | Pre-post prospective study | 278 post-CABG ICU patients ≥18 years | Delirium risk factor screening & modification (pain, comfort, reorientation, sleep) | Significant reduction in incidence; shorter delirium duration | Moderate (structured, measurable effect) | Positive—reduced incidence |
| Zhou et al. [39], 2025 | Systematic Review and Meta-Analysis | 18 studies (n = 1794 intervention, n = 2129 control) | Early mobilization | Pooled OR: 0.65 (CI 0.49–0.86); p = 0.003 | High—meta-analysis with consistent effect across studies, moderate heterogeneity | Positive—pooled OR 0.65 |
| RCT: randomized controlled trials; GRADE: Grading of Recommendations, Assessment, Development, and Evaluation; ICU: intensive care units; OT: Occupational Therapy; BADLs: Basic Activities of Daily Living; QI: Quality Improvement; CABG: Coronary Artery Bypass Grafting; ROM: Range Of Motion; RR: Relative Risk; SP-CS-EM-PC-AS: Sleep Promotion, Cognitive Stimulation, Early Mobilization, Pain Control, and Assessment; OR: Odds Ratio; M.O.R.E.: Music, Orientation, Reorientation, Eye/ear care; NEECHAM: Neelon and Champagne Confusion Scale; CI: Confidence Interval; SICU: Surgical Intensive Care Unit. |
3.1 Multicomponent bundles
The use of multicomponent care bundles—most notably the ABCDE and ABCDEF protocols—emerged as the most robust and consistently effective strategy. These bundles integrate various evidence-based elements including pain management, daily sedation interruption, spontaneous breathing trials, delirium assessment, early mobilization, and family engagement. Studies [16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 31, 32, 33, 34, 35] demonstrated significant reductions in delirium prevalence, duration, and ICU length of stay, particularly when the bundles were comprehensively implemented. The evidence supporting these interventions ranged from moderate to high quality, with larger cohort studies and multicenter prospective designs contributing to their credibility. Implementation barriers included staff training demands and workflow integration, suggesting the need for institutional support to optimize adherence.
3.2 Early mobility and physical activity
Interventions focusing on early mobilization and physical activity represented a second major theme. These interventions included passive and active range-of-motion exercises, early physical therapy, and functional electrical stimulation (FES). While the strength of evidence was mixed, studies with structured and early application [36, 37, 38, 39] showed reductions in delirium duration and incidence. Conversely, studies relying solely on passive movement or late-stage intervention tended to report null effects. The overall quality of evidence in this domain ranged from low to high, contingent upon study design and intensity of the intervention.
3.3 Cognitive and sensory stimulation
A number of studies examined interventions designed to enhance cognitive engagement and sensory orientation, such as the use of mirrors during awakening, automated verbal reorientation messages, and structured environmental cues. This theme, though theoretically promising, yielded less consistent results. Studies [40, 41, 42] provided evidence of modest improvements in delirium-free days or confusion scores; however, the small sample sizes and short intervention durations limited generalizability. The evidence quality for this theme was predominantly low.
3.4 Family engagement and visitation models
Structured family involvement, including extended visitation hours, participation in care routines, and sensory-based reorientation using familiar voices, was associated with improved outcomes in multiple studies. Studies [43, 44, 45, 46, 47, 48] indicated that family engagement can significantly reduce the incidence and duration of delirium. These findings were further reinforced by an umbrella review [49], which underscored the central role of families in delirium prevention. The quality of evidence for this theme ranged from moderate to high, with high feasibility and acceptability reported across clinical settings.
3.5 Music and auditory interventions
Auditory stimulation through music therapy or structured voice messages constituted another thematic cluster. The results were variable: while some studies demonstrated minor improvements in patient orientation or psychological comfort, others found no significant effect on delirium-related outcomes [50, 51]. The evidence was generally of low to moderate quality, with most trials limited by small sample sizes, heterogeneity in music selection, and short intervention periods. Nonetheless, music interventions were consistently reported as safe, well-tolerated, and acceptable to patients and families, suggesting potential adjunctive value in broader multimodal programs.
3.6 Sleep promotion and environmental modifications
The final theme encompassed interventions aimed at optimizing the ICU environment to promote sleep and circadian rhythm regulation. These included the use of earplugs, dynamic lighting systems, noise reduction protocols, and environmental redesign. Studies [52, 53] supported the efficacy of sleep-promoting interventions in reducing delirium risk, particularly when implemented as part of broader care bundles. Isolated use of lighting interventions [54] and dynamic lighting [55] yielded mixed results. The quality of evidence for this category ranged from moderate to high, particularly in studies with rigorous methodology and longer observation periods.
A short summary of these major themes, with typical effectiveness, quality of evidence and other remarks are presented in Table 3.
| Theme | Typical Effectiveness | Quality of Evidence | Remarks |
| Multicomponent Bundles | High | Moderate to High | Most effective when fully implemented |
| Mobility/Physical Activity | Moderate to High | Low to High | Early, structured activity yields best outcomes |
| Cognitive/Sensory Stimulation | Low to Moderate | Low | Promising, needs stronger evidence |
| Family Engagement | Moderate to High | Moderate to High | Consistently beneficial across contexts |
| Music and Auditory Interventions | Mixed | Low to Moderate | Indirect benefits may exist |
| Sleep/Environmental Modifications | Moderate | Moderate to High | Best in combination with other interventions |
Positive effects were observed in 22 studies, showing reduced incidence, duration, or severity of delirium—especially with multicomponent bundles, family engagement, and the ABCDEF protocol. No Effect was reported in 9 studies, typically in cases involving music therapy, range of motion (ROM) exercises, lighting changes, or pilot interventions. Qualitative/Indirect Insight was found in 3 studies, highlighting factors such as feasibility, acceptability, or emotional impact without measuring clinical outcomes directly. Unclear effects were identified in 14 studies, mostly due to ambiguous result wording or lack of clear outcome reporting.
4. Discussion
This systematic review demonstrates that non-pharmacological interventions play a crucial role in preventing and managing delirium among critically ill patients in the ICU. The evidence indicates that multicomponent bundles and structured family engagement are the most consistently effective strategies, while interventions such as early mobilization show promise but are more context-dependent, and single-component approaches like music therapy or isolated environmental modifications yield less consistent outcomes. These findings emphasize that delirium, as a multifactorial syndrome, is best addressed through comprehensive, bundled care rather than isolated interventions.
4.1 Quality appraisal
Fig. 2 presents the distribution of study outcomes across four levels of evidence quality as assessed by GRADE (Very Low, Low, Moderate, High). The majority of studies reporting a positive effect were rated with Moderate quality (n = 15). A substantial number also came from Low quality studies (n = 7), and a few from Very Low (n = 1) and High (n = 1). Interventions supported by moderate-quality evidence appear most consistently effective. However, positive results also occur in studies with lower rigor, raising questions about potential bias or overestimation. Very low and low GRADE levels are linked to both uncertainty and isolated reports of limited effectiveness, suggesting that these findings should be interpreted with caution.
Fig. 2.Study results by quality of evidence. GRADE: Grading of Recommendations, Assessment, Development, and Evaluation.
4.2 Interpretation of main findings
Across the six themes identified, multicomponent bundles (such as the ABCDE and ABCDEF protocols) and family engagement strategies emerged as the most consistently effective interventions, producing reductions in delirium incidence, duration, and severity. Early mobilization showed benefit when applied systematically and early, while isolated physical activity protocols were generally less effective. By contrast, single-component interventions, such as cognitive stimulation, music therapy, and environmental modifications, yielded more mixed or uncertain outcomes, often due to methodological weaknesses or variable implementation fidelity. Together, these results emphasize that delirium, as a multifactorial syndrome, is best addressed through comprehensive, bundled strategies rather than isolated approaches.
4.3 Alignment with guidelines
Our findings are consistent with existing international guidance, including the Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption (PADIS) 2018 recommendations and the ICU Liberation (ABCDEF) bundle, both of which highlight delirium monitoring, daily sedation interruption, early mobilization, and family engagement as central to ICU care. Our review reinforces these recommendations and provides further evidence that integrating bundles into everyday ICU practice is likely to have the greatest impact on patient outcomes.
4.4 Heterogenity of findings
The variability in outcomes for interventions such as cognitive stimulation and music therapy can be attributed to several important sources of heterogeneity. First, differences in study populations played a major role. Studies conducted in elderly ICU patients often demonstrated more consistent benefits from cognitive stimulation, likely due to their higher baseline vulnerability to delirium, whereas studies focusing on post-surgical or mixed ICU populations reported less pronounced effects. Similarly, the inclusion or exclusion of patients with pre-existing cognitive impairment influenced the observed outcomes.
Second, the timing and intensity of interventions varied widely across studies. Some trials implemented daily structured sessions delivered by trained personnel, while others employed shorter or less frequent interventions, often delivered by non-specialists. Interventions with greater frequency and personalization tended to yield more favorable outcomes.
Third, considerable variability was observed in the choice of outcome measures. While some studies assessed delirium incidence, others focused on duration or severity. For example, music therapy appeared more effective in reducing delirium severity scores than in preventing new-onset delirium, leading to apparent discrepancies across studies.
Finally, methodological differences must be acknowledged. Smaller pilot studies or quasi-experimental designs often reported stronger effects, whereas larger, well-powered RCTs tended to show more modest or null results. Variability in blinding, control group design, and fidelity of intervention delivery further contributed to differences in reported effectiveness.
Taken together, these factors highlight the complex nature of evaluating single-component interventions for delirium management in critically ill patients. Future trials should strive for greater methodological consistency, including standardized outcome measures, clear definitions of intervention protocols, and sufficiently powered study designs to reduce heterogeneity and enhance comparability across studies.
4.5 Limitations
One of the major limitations of this review is the heterogeneity of included study designs. While RCTs and cohort studies were prioritized, the review also incorporated quasi-experimental studies, systematic reviews, and meta-analyses. Although this approach broadened the evidence base and allowed us to include important insights, it inevitably reduced comparability across studies and increased the risk of bias. The variability in methodological rigor, intervention protocols, and outcome measures must therefore be considered when interpreting the findings. Future research should focus on high-quality RCTs to strengthen the evidence regarding the effectiveness of nurse-led non-pharmacological interventions in ICU delirium management.
4.6 Clinical implications and future directions
From a clinical perspective, these findings emphasize the central role of ICU nurses in implementing non-pharmacological interventions. Nurses are well positioned to lead efforts in mobilization, reorientation, environmental optimization, and family engagement, thereby embedding evidence-based delirium prevention into routine care. Future research should prioritize high-quality, adequately powered RCTs, include predefined subgroup analyses (e.g., elderly and ventilated patients), and address barriers to implementation, such as staff training and workflow integration.
5. Conclusions
The current body of evidence indicates that multicomponent non-pharmacological intervention bundles offer the most consistent benefits in the prevention and management of delirium in critically ill patients. These interventions are particularly effective when they integrate key elements, such as sleep promotion, cognitive reorientation, and early mobilization. The convergence of multiple supportive strategies appears to address the multifactorial nature of delirium more comprehensively than single interventions.
Family-led interventions have also emerged as a promising and acceptable approach within critical care settings. These strategies foster patient-centered care and have been well-received by both healthcare professionals and families. Nevertheless, the existing evidence, while supportive, is primarily based on pilot or small-scale studies. There is a clear need for larger, rigorously designed RCTs to establish their clinical effectiveness and to delineate the mechanisms through which family involvement may influence patient outcomes.
Music therapy and voice reorientation interventions have demonstrated feasibility and high levels of patient and staff acceptability. However, the current literature does not yet provide robust evidence to confirm their effectiveness in reducing the incidence or severity of delirium. Although these interventions show promise, their clinical utility remains to be validated through further empirical investigation.
Future research in this field should prioritize the implementation of large-scale, methodologically rigorous RCTs that employ standardized and validated outcome measures for delirium. Moreover, studies should adopt implementation science frameworks to evaluate the real-world feasibility, sustainability, and scalability of non-pharmacological interventions across diverse ICU environments. This approach is essential to bridge the gap between research findings and clinical practice, ensuring that effective interventions can be reliably translated into routine care.
Abbreviations
CAM-ICU, Confusion Assessment Method for the ICU; GRADE, Grading of Recommendations, Assessment, Development, and Evaluation; ICU, Intensive Care Unit; OT, Occupational Therapy; RCT, Randomized Controlled Trial; FES, functional electrical stimulation; PRISMA, Preferred Reporting Items for Systematic reviews and Meta-Analyses; PIO, Population, Intervention, Outcome; PADIS, Pain, Agitation/Sedation, Delirium, Immobility, and Sleep Disruption; BADLs, Basic Activities of Daily Living; QI, Quality Improvement; CABG, Coronary Artery Bypass Grafting; ROM, Range of Motion; RR, Relative Risk; SP-CS-EM-PC-AS, Sleep Promotion, Cognitive Stimulation, Early Mobilization, Pain Control, and Assessment; OR, Odds Ratio; M.O.R.E., Music, Orientation, Reorientation, Eye/ear care; NEECHAM, Neelon and Champagne Confusion Scale; CI, Confidence Interval; SICU, Surgical Intensive Care Unit.
Availability of data and materials
The authors declare that all data supporting the findings of this study are available within the paper and any raw data can be obtained from the corresponding author upon request.
Author contributions
AČ and LCB—designed the research study, performed the research and analysed the data. AM—provided help and advice on data analysis and synthesis. All authors wrote the manuscript. All authors contributed to editorial changes in the manuscript. All authors read and approved the final manuscript.
Ethics approval and consent to participate
Not applicable.
Acknowledgment
Not applicable.
Funding
This research received no external funding.
Conflict of interest
The authors declare no conflict of interest.
Supplementary material
Supplementary material associated with this article can be found, in the online version, at https://oss.signavitae.com/mre-signavitae/article/1991426647536877568/attachment/Supplementary%20material.docx.
