Family Support Strategies During Intensive Care Unit: A Systematic Review

Abstract

Family members of critically ill patients in intensive care units (ICUs) often experience significant psychological distress due to uncertainty and fear of adverse outcomes. Structured nursing interventions may help mitigate these challenges by providing emotional support and improving communication. To assess the effectiveness of nursing interventions, compared to standard care, in enhancing communication with healthcare professionals, supporting decision-making, and strengthening emotional attachment among family members of adult ICU patients. This systematic review followed the PRISMA 2020 guidelines. Included studies focused on adult ICU patients and evaluated nursing-led interventions aimed at supporting families in communication, decision-making, or emotional bonding. Studies not involving nursing-led interventions or conducted outside the ICU setting were excluded. A comprehensive search was conducted across PubMed, Scopus, CINAHL, SciELO, and Web of Science for articles published between 2015 and 2025. The quality of included studies was appraised using the Johns Hopkins Nursing Evidence-Based Practice Model and the ROB 2 tool. A narrative synthesis was used to summarize and compare study outcomes. Eighteen studies were included, involving diverse populations, with a predominance of female participants. Interventions ranged from in-person programs—such as structured communication during rounds, open visitation policies, and family involvement in patient care—to technology-assisted approaches that facilitated decision-making and orientation. These interventions were associated with increased family satisfaction and reduced psychological distress. The evidence suggests that family-centered nursing interventions are effective in addressing the emotional and informational needs of families during ICU stays. However, variability in intervention design, small sample sizes, and moderate risk of bias in some studies limit the generalizability of findings. Further high-quality, mixed-methods and controlled studies are recommended to strengthen the evidence base.

Highlight

• Interventions are feasible within ICU settings and support patient- and family-centered care principles. Nurses hold a central role by fostering empathetic, clear communication, ensuring understanding, and guiding families through complex decision-making processes, enhancing engagement and emotional well-being throughout critical care.

Introduction

Family members of critically ill patients hospitalized in intensive care units (ICUs) experience profound periods of uncertainty that affect their psychological wellbeing. They often fear potential outcomes, particularly death, especially when complications arise during hospitalization. ¹

In the field of nursing, we identified 3 crucial conceptual perspectives regarding the situation of families of critically ill adult patients hospitalized in ICUs. First, the needs of family members, several studies have focused on understanding this experience, and the results have highlighted the need for security, information, proximity, and support.^2-4 Unsurprisingly, these demands are even greater when patients are semi- or unconscious. ⁴

ICU admission causes more anxiety and disturbance for family members than patients. ⁵ Family members feel lost, navigating an unfamiliar environment while wondering whether they will see their family member again, how long survival will last, and under what conditions the patient will return home.^6,7 This distress often leads family members to neglect their physical and psychological needs as they prioritize the patient. Previous studies comparing nurses’ and families’ perceptions reveal both groups prioritize providing the best possible care.^3,8

Second, humanized ICU care, though complex, significantly impacts families’ experiences. ICU staff primarily focus on patient recovery, often overlooking family members, reporting only when the patient’s condition worsens or once daily, and limiting discussion to the patient’s treatment. Recognizing families’ concerns is crucial, as they play a vital role in comprehensive care and decision-making.^9,10 Despite advances in humanized ICU care during the global pandemic, deficiencies persist. Humanizing ICUs enables us to recognize the family as the central axis of all care actions. A family involved in the care process is healthier and better equipped to combat the despair associated with ICU stays. ¹¹

Lastly, patient- and family-centered care (PFCC) is widely recognized as an integral component of high-quality healthcare. This approach fosters collaboration with family members by emphasizing the provision of information, emotional support, and both their participation and involvement in the care of the patient, especially in decision-making and direct care activities within the ICU setting. The technical nature of critical care risks dehumanizing both patients and families. ¹² Therefore, PFCC is recommended in nursing to encourage family involvement across healthcare settings, bringing care closer to the lived experiences of ICU patients’ families. It may also support the development of strategies to engage families more effectively, ultimately improving ICU processes and outcomes. ¹³

The literature highlights the need for ICU nurses to establish collaborative relationships with patients’ families, addressing their needs and helping them manage distress. ¹⁴ Family members of ICU patients describe these units as cold and distant places with restricted and limited access, where illness severity and complexity are paramount. Therefore, family members and friends are often excluded to ensure better clinical care for critically ill patients. ¹⁵ ICUs are hospital units that family members fear the most, either due to the patient’s condition or the lack of humanized care.

In recent years, systematic reviews have shown that interventions for families of patients in ICUs focus on involvement outcomes, such as communication, decision-making, satisfaction, or health outcomes related to family trauma, well-being, and patient outcomes. ¹⁶ Stress-reduction interventions are classified as active, which facilitate expression, and passive, characterized by receiving care. ¹⁷ However, family-centered care is often noted to have no effect on family members’ anxiety, depression, or post-traumatic stress disorder. ¹⁸

Even though the growing body of literature, there is a lack of synthesis regarding the specific strategies used to support families during ICU stays, particularly those that integrate educational, emotional, and technological components. This gap limits the ability of healthcare professionals to implement evidence-based interventions that address the complex needs of families in critical care settings.

Objective

To evaluate the effectiveness of nursing interventions aimed at supporting families in ICU settings, with particular focus on communication, emotional outcomes, and participation in care.

Methods

This systematic narrative review was conducted based on Preferred Reporting Items for Systematic reviews and Meta-Analyses (PRISMA) ¹⁹ guidelines and registered in PROSPERO with registration number CRD42023415141.

The research question was formulated according to the following Population, Intervention and Outcome (PIO) framework: P = family members of adult ICU patients; I = effectiveness of nursing intervention programs; O = communication, emotional outcomes, and involvement in care. Question: For family members of adult ICU patients, how effective are nursing intervention programs in improving communication with the healthcare team, emotional well-being, and active involvement in patient care?

Eligibility Criteria

This study included scientific articles published in Spanish, Portuguese, and English with controlled trial designs, quantitative or mixed studies, intervention studies, and observational studies, whether cohort or case-control studies with a follow-up of 8 years from 2015 to 2025. The timeframe was selected to encompass developments before, during, and after the COVID-19 pandemic, a period that significantly influenced ICU family-centered care practices worldwide. We excluded from this study literature published in books, articles not supported by scientific evidence, comments, expert panel opinions, scientific societies, or the World Health Organization. Studies without a description of the results broken down by sex or gender variables and studies on neonatal or pediatric intensive care were also excluded from the sample.

Information Sources

The following bibliographic databases were searched from 2015 to 2025: PubMed Scopus, CINAHL, SciELO, and Web of Science (WOS).

Search Strategy

We adapted for each database a combination of controlled vocabulary and free-text keywords. Specifically, Medical Subject Headings (MeSH) and Health Sciences Descriptors (DeCS) were used to ensure comprehensive coverage of the topic. Boolean operators (AND, OR, NOT) were applied to combine search terms appropriately. The search was conducted in English. The last search was conducted in 2025 (Description available in Appendix 1).

Selection and Data Collection

This process was carried out in 4 phases. First, all articles identified from the databases were imported into Rayyan Desktop for reference management, duplicate removal and assessment. Next, 2 researchers independently screened the titles and abstracts of the remaining records to evaluate their relevance based on the clinical question of the research and the predefined inclusion and exclusion criteria.

Studies that passed the initial screening were then subjected to a peer critical appraisal process. This evaluation was supported by a panel of nurses with extensive expertise in Intensive Care Unit practice (See Figure 1). The panel used the assessment tool recommended by Johns Hopkins Nursing Evidence-Based Practice Model. ²⁰ Annex E specifically addresses the critical appraisal of evidence. Following these steps, relevant data from the studies included—such as author, publication year, study design, intervention characteristics, outcomes measured, findings, and limitations—were compiled into the evidence summary table. In total we included 18 studies.

Figure 1.

PRISMA flow diagram.

*Title and abstract analysis based on research questions and inclusion/exclusion criteria.

**Restricted access.

Bias Analysis

Three researchers independently assessed the risk of bias in experimental studies including the Cochrane Risk of Bias tool RoB 2 tool (Beta Version 7). The evaluation considered the following domains: randomization process, deviations from intended interventions, missing outcome data, measurement of the outcome, and selection of the reported result. ²¹ Each domain was rated at risk of bias, and these assessments were documented in a standardized form. Although some studies exhibited potential biases, they were not excluded, as they contributed valuable insights into clinical practice strategies and the interpretation of evidence (Description available in Appendix 2). To minimize bias during the study inclusion phase, we implemented several strategies, including the involvement of an expert panel to assist with the critical appraisal of eligible articles and the use of a validated tool to assess risk of bias specifically in randomized trials. The results of the bias assessment informed the narrative synthesis, with greater emphasis placed on studies demonstrating lower risk of bias.

Synthesis Methods

We conducted a narrative synthesis without meta-analysis, ²² structured around key interventions. We used thematic analysis, and the results were grouped according to their primary focus, such as the integration of information and communication technologies (ICT) in educational practices; strengthening family support and involvement in decision-making; promoting family participation during rounds and visits; as well as the involvement in patient care. The results presented as supporting evidence drawn from each included study, providing a comprehensive understanding of the strategies used to support families in intensive care units. We extracted quantitative data to characterize the study populations using descriptive statistics.

Results

This section presents findings from the reviewed studies, with a focus on risk of bias, participant demographics characteristics, and the reported effectiveness of PFCC-based programs in the ICU. It provides a summary of the evidence, and a thematic analysis of the studies gathered in their methodological quality, outcomes, to inform the evaluation and application of evidence-based practice.

Risk of Bias Assessment

The results show that while many studies received low-risk ratings in the domains randomization process and deviations from intended interventions, a considerable proportion showed “some concerns” or “high risk” in critical areas such as:

• a) Selection of the reported result (27.8% high risk)

• b) Measurement of the outcome (44.4% with some concerns)

• c) Missing outcome data (20% with some concerns)

One major challenge in this type of research, especially studies evaluating long-term psychological symptoms, is the longitudinal nature of follow-up. As demonstrated in this study, declining survey response rates over time can reduce statistical power and reduce the potential intervention effects. Moreover, different outcomes are shaped by a complex interplay of personal, contextual, and environmental factors, making it difficult to isolate the impact of a single intervention. The results are visually represented in Figure 2.

Figure 2.

Risk of bias assessment.

Source. Analysis of researchers.

Sociodemographic Characteristics of the Study Population

The analyzed studies revealed that women comprised 61.18% of the intervention group and 54.47% of the control group. Participants in both groups were predominantly over 50 years old. The most common relationship to ICU patients was that of a spouse—28.69% in the intervention group and 35.81% in the control group—followed closely by children (28.76% and 32.10%, respectively). Notably, the intervention group had a significantly higher proportion of “other relatives” (17.41%) compared to the control group (6.15%), suggesting differences in family structure or participant selection. Authors consistently noted that women are more engaged in caregiving roles. A key limitation is that 27.78% of the included studies lacked complete demographic data on family members.

Resume of Evidence

Table 1 summarizes evidence using elements from the Johns Hopkins Nursing Evidence- Based Practice Tool. It presents 18 studies on PFCC interventions for families of critically ill adult patients hospitalized in ICUs. The studies are categorized by strategies commonly used in clinical practice. This summary helps identify effective interventions related to the outcomes addressed in the research question and highlights areas where the evidence is either strong or limited.

Table 1.

Individual Evidence Summary.

Author and year	Design	Type of intervention	Outcome(s) measured	Key findings	Identified strategy	Quality of evidence level
Suen et al (2021) 23	Two-arm, single-blind, patient-level randomized trial	Digital tools (IG) Control: Usual care enhanced with scheduled family meetings (CG)	System Usability Scale (SUS) Quality of Communication (QOC) Questionnaire Shared Decision-Making Questions (subset of QOC)	Surrogates rated the tool as usable (SUS). Communication quality was rated slightly higher in the intervention group, though the difference was not statistically significant. Quality of Communication: IG: (n = 24): Mean = 8.9 ± 1.6. CG: (n = 23): Mean = 8.0 ± 2.7 t-value: 0.19 No significant improvement was observed in shared decision making.	Promote the integration of Information and Communication Technologies (ICTs) into educational practices	LEVEL I Medium Quality
Chiang et al (2017) 24	A randomized control trial	Tablet-based family education (EF-T). Control: Education of families by routine (EF-R).	The Depression Anxiety Stress Scale − Chinese (C − DASS) Stress: The Family Needs Assessment Questionnaire’s Communication and Physical Comfort Scale (CPCS)	C–DASS: scores were slightly lower in the EF-R group compared to EF-T. Effect sizes were small for stress (η² = .07) and depression (η² = .09); effect size for anxiety was not reported Satisfaction: No statistically significant differences were found between EF-T and EF-R groups.	Promote the integration of Information and Communication Technologies (ICTs) into educational practices	LEVEL I Medium Quality
Hoffmann et al (2023) 25	Multicenter double-blind, randomized, placebo-controlled trial (ICU Families Study) in Austria and Switzerland.	Website included ICU education with experts, stress reduction counseling by family, hygiene, and patient experiences. Control: Placebo	Impact of Event Scale (IES) Hospital Anxiety and Depression Scale (HADS-Deutsch)	At baseline (T0), 50% of participants showed PTSD symptoms (IES > 27). The web-based intervention had no differential effect on PTSD at T1, T2, or T3. Depressive symptoms (HADS-D) were high at baseline (mean = 16.4 ± 6.6) but decreased over time: T1 = 12.2, T2 = 10.6, T3 = 9.5. No significant differences in depressive symptom reduction were observed between intervention and control groups at any time point.	Promote the integration of Information and Communication Technologies (ICTs) into educational practices	LEVEL I Medium Quality
Cox et al (2024) 26	Randomized parallel-group clinical trial	Mobile application for 7 days, which guided doctors to identify the needs of family members. Control: usual care and meetings when the doctor considered it necessary.	NEST Score (Nursing Effectiveness Score Tool): Assessed from days 1-3 and 3-7; also reported as median (IQR) on days 3 and 7 for patients remaining in the ICU. Observable Outcome Measures: Palliative care consultations Family meetings	NEST scores showed no significant differences: Days 1-3 (intervention = −3.1, control = −2.0; diff = −1.3), Days 3-7 (intervention = −2.3, control = −2.2; diff = 0.0). Median NEST Day 3:24.5 vs 27.5; Day 7:22.0 vs 28.0. Palliative care consultations were higher in the intervention group (38.2% vs 4.0%, P < .001). Family meetings were more frequent (87.0% vs 64.0%, P = .001).	Promote the integration of Information and Communication Technologies (ICTs) into educational practices	LEVEL I Medium Quality
Drop et al (2025) 27	Multicenter randomized controlled trial Control: included weekly meetings with the medical team and access to ICU diaries, without VR exposure	A 14-min first-person virtual reality (VR) experience explained ICU procedures, equipment, and daily care from a patient’s perspective. Control: received standard care.	IES-R: Assessed posttraumatic stress severity. HADS-A/HADS-D: Measured anxiety and depression (cutoff ≥ 8). SF-36 (MCS & PCS): Evaluated mental and physical quality of life. ICU-VR Satisfaction Survey: Applied only to the intervention group.	At 6 months, no significant differences were found between groups in PTSD (23% vs 18%), anxiety (22% vs 30%), depression (17% vs 23%), or combined symptoms (38% vs 30%). Symptom severity (IES-R, HADS-A, HADS-D) and quality of life (mental: 50.2 vs 52.6; physical: 56.1 vs 54.3) were similar. ICU-VR satisfaction: 90% recommend, 81% preferred, 76% understood better, 52% processed ICU experience.	Promote the integration of Information and Communication Technologies (ICTs) into educational practices	LEVEL I Medium Quality
Yuan et al (2023) 28	Randomized controlled trial	Daily video calls with nurses. Control: routine in-person visits.	Anxiety and Depression Scales Delirium Incidence Satisfaction Surveys	No significant group differences were found in patient or family anxiety, depression, or delirium. However, satisfaction with video visits was significantly higher in both patients and family members in the intervention group (P < .05).	Promote the integration of Information and Communication Technologies (ICTs) into educational practices	LEVEL I Medium Quality
White et al (2018) 29	Multicenter, stepped, cluster-randomized trial	Communication training for nurses and implementation of a family support plan with regular meetings within 48 h of enrollment in the protocol. Control: received usual care.	Hospital Anxiety and Depression Scale (HADS): Assessed at 6 months. Quality of Communication (QOC): Measured using the PPPC score. Impact of Event Scale (IES): Used to assess posttraumatic stress symptoms. Modified Decision-Making Scale: ICU Length of Stay: Recorded in days.	No significant difference in HADS scores (11.7 vs 12.0; β = −.34; 95% CI: −1.67 to 0.99; P = .61). QOC (PPPC score) was slightly better in controls (1.8 vs 1.7; β = −.15; 95% CI: −0.26 to −0.04; P = .006). IES scores were similar (21.2 vs 20.3; β = .90; 95% CI: −1.66 to 3.47; P = .49). Surrogate decision-making improved (69.1 vs 62.7; β = 6.39; P = .001). ICU stay was shorter (6.7 vs 7.4 days; P = .045).	Strengthen Communication and family support	LEVEL I Medium Quality
Azoulay et al (2018) 30	Randomized controlled trial	Structured support with assessment tools. Control: standard ICU care without structured assessments.	Family Comprehension Assessment (Day 5) Family-Staff Meeting Metrics Satisfaction and Psychological Symptoms	Adequate in both intervention and control groups (P = .30). Median number of meetings per patient: 6; Median total meeting time: 72.5 min; Median number of questions asked: 20 (including 11 from the list) No significant differences between groups. Satisfaction and Anxiety/Depression: No significant differences between intervention and control groups.	Strengthen Communication and family support	LEVEL I Medium Quality
Curtis et al (2016) 31	Parallel-group randomized trial	Structured communication support for families during the ICU stay (follow up 3 and 6 months). Control: received usual care.	Generalized Anxiety Disorder Scale (GAD-10). Posttraumatic Stress Scale. SOFA Patient Health Questionnaire. Mortality and life-sustaining treatment data: Collected to assess unadjusted mortality and time to withdrawal of life-sustaining treatments.	At 3 months, anxiety (GAD-10) decreased by 1.96 points in the intervention group vs. a 0.76-point increase in controls (P = .096). At 6 months, reductions were 2.42 vs 0.42. Posttraumatic stress scores were lower in the intervention group at 3 months (29.81 vs 31.57) and 6 months (27.09 vs 30.52). No mortality difference. Time to withdrawal of life-sustaining treatments was shorter (P = .001).	Strengthen Communication and family support	LEVEL I Medium Quality
Torke et al (2016) 32	A randomized controlled pilot intervention trial	Structured support program for families, including meetings, rounds, and follow-up with an interdisciplinary team. Control: received usual care.	Mean Sequential Organ Failure Assessment (SOFA). Patient Health Questionnaire (PHQ-9). Generalized Anxiety Disorder 7-item Scale (GAD-7). Impact of Event Scale – Revised (IES-R). Decision Conflict Scale. Decision Regret Scale.	Patient Health Questionnaire (PHQ-9): No significant differences in depressive symptoms between groups (Intervention = 9.1 ± 5.1; Control = 5.2 ± 5.3; P = .073). Generalized Anxiety Disorder Scale (GAD-7): No significant differences in anxiety symptoms (Intervention = 8.1 ± 5.6; Control = 4.3 ± 5.2; P = .072). Some measure reported were not informed.	Strengthen Communication and family support	LEVEL I Low Quality
Liu Xinchun et al (2022) 33	Multicenter randomized trial	Structured family meeting focused on improve family–staff communication and support. Control: standard care provided	Family satisfaction, level of depression, and patients’ quality of life. Physicians’ perceptions Communication Skills and Compassion Assessment Device (CSCAD). Daily medical cost and length of stay in the intensive care unit.	The intervention group showed significantly higher family satisfaction (statistical significance: P = .0001), lower depression levels (P = .005), and better patient quality of life (P = .0007) compared to the control group. Physicians in the intervention group had more positive CSCAD scores than those in the control group (P < .05). There was no significant difference in daily medical costs between the groups. The intervention group had a shorter ICU stay than the control group (P = .0004).	Strengthen Communication and family support	LEVEL I Medium Quality
Thinnes et al (2023) 34	A randomized controlled study	Participants received daily written updates (via email or hard copy) detailing ICU problems, their causes, and related diagnostics or treatments. Control: standard care provided.	Although the study employed structured surveys and chart reviews as data collection tools, the authors did not explicitly identify or label these tools as instruments. Subgroup analysis based on ICU visitation policies (restricted vs. open).	Daily written patient care updates as the intervention. Medical comprehension assessment: Identification of 6 common ICU problems (shock, renal failure, weakness, respiratory failure, encephalopathy, liver failure).	Foster rounds and visits	LEVEL I Low Quality
Debay et al (2023) 35	Study a pilot randomized controlled trial	Families joined daily rounds in person or virtually, receiving orientation, engagement, questions, and follow-up. Control: usual care	FS-ICU (Family Satisfaction in the Intensive Care Unit) Survey FAME Tool (Family Engagement Measurement Tool) Rounds Acceptance Tracking	Satisfaction with Care: Intervention group: 87.3, Control group: 74.7 P = .03 Satisfaction with Decision-Making: Intervention group: 75.6 Control group: 69.3 P = .27	Foster rounds and visits	LEVEL I Medium Quality
Weber et al (2018) 36	Pre- and postimplementation (pre-I and post-I) design	Families met with ICU staff twice weekly during scheduled visits. Control: Families received standard care without scheduled visits.	4-item Validated ICU Family Satisfaction Questionnaire FS-ICU 24 used pre- and post-intervention; participation in rounds tracked.	Mean Family Satisfaction Scores: Pre-intervention: 89.2 ± 11.2 Post-intervention: 87.4 ± 14.2 P = .6. Secondary Analysis (FS-ICU 24): No difference between pre-intervention respondents and 39.0% of post-intervention respondents who participated in family rounds. Emotional Support: Pre-intervention: 75% post-intervention 54% P = .002. Significant decrease. Coordination of Care: Pre-intervention: 82% Post-intervention 68% P = .03. Significant decrease Communication Frequency: Pre-intervention: 60% Post-intervention 43% P = .03. Significant decrease	Foster rounds and visits	LEVEL II Medium Quality
Liang et al (2023) 37	A prospective, assessor-blind, parallel-group randomized controlled trial	Daily 30-min auditory and visual stimulation sessions for up to 7 days in ICU. Control: Standard ICU care only.	PTSD Score Assessment Tool Depressive Symptoms Scale Delusional Memories Scale Family Satisfaction with Care Scale Anxiety Scale for Family Caregivers	Patients in the intervention group showed: Lower PTSD scores: 21.92 ± 6.34 vs 27.62 ± 10.35, P = .001. Lower depressive symptoms: 3.76 ± 3.99 vs 6.78 ± 4.75, P = .001. Fewer delusional memories: 0.47 ± 0.92 vs 0.82 ± 1.23, P = .00. No depressive symptoms at 1-month post-intervention. No effect on: ICU length of stay 30-day mortality (all P > .05). Family caregivers in the intervention group showed: Higher satisfaction with care: 127.12 ± 14.14 vs 114.38 ± 21.97, P = .001. Lower anxiety levels compared to the control group	Enhance family proximity and participation in care	LEVEL II Medium Quality
Mitchell et al (2017) 38	A randomized controlled trial	Intervention: Families preovided daily memory cues, sensory aids, and cognitive stimulation. Control: Patients received standard ICU care only.	Delirium Assessment Tool ICU Delirium Days Measurement	Delirium Prevalence: Pre-randomization: 50%, Intervention group: 59%, Non-intervention group: 56%, No statistically significant differences (P = .87, p = .98). Delirium Rate Comparison: Intervention: 50%, Non-intervention: 54%, Not statistically significant (P = .80) ICU Delirium Days: Intervention: M = 1.34, SD = 1.57, control: M = 1.03, SD = 1.12. Effect size: Cohen’s d = 0.23. Sample Size Estimation: To detect a significant effect with power = 0.80 and α = .05, a future study would need n = 596 participants.	Enhance family proximity and participation in care	LEVEL I Medium Quality
Sottile et al (2015) 39	A randomized controlled trial	Daily 30-min physical therapy (PT) after stabilization. Control: standard care without structured PT sessions.	Comparing patient and family groups across various dimensions using the 7-point Functional Independence.	Need: Family members reported significantly higher needs than patients (P = .003). Enjoyment: Patients reported significantly more enjoyment than family members (P = .03). Difficulty and Satisfaction: Borderline significance observed (P = .05). Intervention effectiveness partially supported.	Enhance family proximity and participation in care	LEVEL II Medium Quality
Marshall et al (2023) 40	A multi-center, randomized, parallel group superiority Phase II randomized controlled trial.	Family-focused intervention or decision support or nutrition optimization, or usual care protein and energy intake during ICU and hospital stay (nutritional intervention) and family satisfaction (decision support)	Randomization Tracking: Measured as rate per month: 0.5 (0.13-1.53) Family Feedback Survey Decision Support and Nutritional Support Tools	Randomization rate: 0.5 per month (range: 0.13-1.53) Emotional burden: 46% of families (58/126) felt overwhelmed Intervention delivery: Decision support: 76.9%-100%. Nutritional support: 44.8%-100% No significant differences between groups in: Nutritional adequacy, Decisional conflict, Satisfaction with decision-making, Overall family satisfaction with ICU	Enhance family proximity and participation in care	LEVEL II Medium Quality

Promote the Integration of Information and Communication Technologies (ICTs) Into Educational Practices

The studies grouped under this theme focused on the implementation of PFCC programs using digital tools^23-28 to enhance information access, communication, decision-making, psychological well-being, and satisfaction among families of ICU patients.

Specifically, 4 studies evaluated digital strategies that aim to improve psychological well-being in ICU patients and their families. Validated scales were used to assess outcomes related to anxiety, depression, and posttraumatic stress severity.^24,25,27,28 However, these interventions generally did not result in statistically significant reductions in anxiety, stress, or depression scores compared to control conditions. In fact, the use of tablet-based education (EF-T) showed no significant improvements in psychological symptoms relative to routine education (EF-R), despite being rated positively for usability and acceptability. ²⁴ Nonetheless, some indirect benefits were observed. ICU nurses using structured digital tools, such as tablets and mobile apps reported being able to provide emotional support, facilitate emotional expression, and deliver consistent information to families. ²⁴ Similarly, the ICU Virtual Reality (ICU-VR) intervention failed to reduce psychological distress but improved participants’ understanding of the ICU environment and was associated with high satisfaction scores. ²⁷

About the study on Web-Based Interventions for Family Engagement included ICU orientation content, expert chats, videos, and stress-reduction resources and family reported high levels of user engagement. Although underpowered due to recruitment and retention challenges, ²⁵ participants in the intervention group accessed the platform 3 times more often than controls. This suggests that such low-cost, accessible platforms may promote family engagement, even in the absence of measurable changes in clinical outcomes.

In matter of real-time communication, the study that implemented daily WeChat video calls to connect ICU patients with their families, adopting nurse-coordinated dyadic communication model. ²⁸ The intervention ensured privacy and interaction. Over 70% of participants reported that staff met their needs during these calls, and satisfaction with both staff communication and the virtual visiting method was significantly higher in the intervention group. These findings underscore the role of video-based communication in supporting emotional connection and trust in the ICU context.

Two additional studies assessed digital tools aimed at improving communication and shared decision-making.^23,26 One study used the System Usability Scale (SUS) to evaluate surrogate perceptions of usability, finding slightly better communication quality in the intervention group. However, these differences were not statistically significant, and no improvements were observed in shared decision-making. ²³ Another study employed short videos and interactive questions to prepare surrogates for clinician-family meetings. ²⁶ While the intervention did not significantly reduce family distress, it improved the structure of communication. Notably, the intervention group had a higher number of palliative care consultations and family meetings, suggesting a potential improvement in care coordination, despite the limited follow-up period.

The main limitations in the studies were in sample sizes, study power, and follow-up duration warrant cautious interpretation of effectiveness and call for more robust, large-scale trials. However, the findings highlight the value of digital tools in addressing relational and informational needs, even when measurable health outcomes remain unchanged.

Ethical Considerations in the Use of ICTs

The use of digital platforms—such as personal devices, videoconferencing tools, and web-based data collection systems—requires careful attention to issues of data privacy, confidentiality, and informed consent, particularly when involving vulnerable populations like critically ill patients and their families. We noted that only one study ²⁸ explicitly provided more discussion to protect patient privacy during video calls. This points to a broader gap in the literature regarding the ethical safeguards implemented in this kind of intervention.

Strengthen Communication and Family Support

This theme encompasses 5 studies that implemented interventions centered on communication skills and family support.^29-33 The investigations analyzed typically regular meetings that emphasized empathy, active listening, kindness, and understanding, with the goal of improving the quality of interactions between families and healthcare teams. However, it was evident substantial differences regarding design, temporality, and structural components of the interventions. The consistency in the goal of the studies contrasts with the methodological and results heterogeneity.

Among the 5 studies reviewed, only two^29,33 explicitly reported outcomes related to the communication quality and decision-making processes one of them ²⁹ reported that the control group had slightly better perceptions of communication quality (PPPC: 1.8 vs 1.7; P = .006). However, the same study showed improvements in decision-making (P = .001) and a reduction in ICU stay (P = .045). One study ³² mentioned the use of the Decision Conflict Scale and the Decision Regret Scale but did not disclose the results which limit interpretability. All 5 studies assessed psychological outcomes such as anxiety, depression, and posttraumatic stress symptoms and 2 studies measured family satisfaction as an outcome.^30,33

About psychological outcomes,^29-33 the comparison between studies was difficult because of the use of different instruments (HADS, GAD-7/10, PHQ-9, IES-R). The improvements observed in some studies were uncertain and, in several cases, not statistically significant. It was observed in Curtis et al ³¹ that the reduction in anxiety in the intervention group was clinically relevant but not significant at month 3 (P = .096). The study by Liu et al ³³ was the only one that showed robust results in multiple dimensions: greater family satisfaction (P = .0001), less depression (P = .005), better patient quality of life (P = .0007), and better communication skills scores of the medical team (P < .05).

This evaluation shows that, while structured communication is a favorable approach, its actual impact on clinical, psychological, and decision-making indicators is not yet sufficiently established. The discrepancy between the expected effects and the results obtained represents a challenge about the effectiveness expected for this type of strategy.

Foster Rounds and Visits

The studies included in this theme^34-36 implemented various strategies to enhance information exchange, closeness, proximity, and interactions among family members, ICU patients, and nursing or healthcare teams. These investigations considered both open or restricted visitation policies and carried out specifics approach to improve engagement with family members. The communication processes included writing updates, patient care briefings, conversations, meetings, and/or video calls. The conceptual orientation of the programs focused on either facilitating time spent with ICU patients or sharing structured information to family members, especially when open visits were restricted, such as during pandemic time.

Regarding the instruments used to measure outcomes, only one study ³⁵ employed the FS-ICU and the FAME Tool, both of which have been widely implemented and demonstrate a clear factorial structure and high internal consistency. These tools are recommended for assessing family satisfaction and engagement in ICU settings. Another study ³⁶ used a 4-item validated ICU Family Satisfaction Questionnaire, probable adapted for local use.

However, instruments such as the structured surveys with chart reviews, ³⁴ Rounds Acceptance Tracking, ³⁵ and the FS-ICU ³⁶ did not report formal validation based on psychometric research or established clinical measurement standards. This lack of validation could limit the generalizability and comparability of the findings.

The outcomes related to effectiveness revealed disparities across studies.^34-36 The study by Debay et al ³⁵ demonstrated a statistically significant increase in family satisfaction with care (P = .03), although the improvement in satisfaction with decision-making did not reach significance (P = .27). Another intervention that reached statistical significance was the Written Communication Study by Thinnes et al, ³⁴ which reported an improvement in medical understanding among family members. This was under restricted visiting conditions, with the intervention group showing a significantly higher likelihood of correctly identifying ICU problems compared to the control group (aOR = 2.9, 95% CI: 1.9-4.2; P < .001), with a smaller, but still significant, effect under open visiting conditions (aOR = 1.8, 95% CI: 1.1-3.1; P = .02). In contrast, the study of Weber et al ³⁶ showed significant decreases in key indicators after the intervention. Although mean satisfaction scores did not differ significantly (P = .6), secondary outcomes revealed significant decreases in perceived emotional support (P = .002), care coordination (P = .03), and communication frequency (P = .03). ³⁶ These findings suggest that foster rounds or visits for family may provide an opportunity to facilitate the engagement between family and patient. However, implementation factors, such as limited frequency, lack of timeliness, and lack of staff training may negatively impact family members’ experiences.

Enhance Family Participation in Care and Decision-Making

The 4 studies included in this theme^37-40 demonstrated that family involvement either direct, or indirect in activities such as sensory stimulation, physical therapy, or nutritional support during ICU hospitalization can improve outcomes for both patients and their family members. Most of these interventions showed statistically significant effects, particularly in satisfaction-related outcomes.

Providing patients and family caregivers with a daily 30-min auditory and visual stimulation session from recruitment for up to 7 days during ICU stay increased satisfaction with care (P = .001), mitigated depressive and anxiety symptoms among family members, and decreased short-term delusional memories in patients (P = .00). ³⁷

In another study on family participation in care, family members provided patients with orientation to surroundings or memory clues, such as family photographs, and conducted both sensory checks (vision and hearing with glasses and hearing aids) and therapeutic or cognitive stimulation (discussing family life and reminiscing) daily. Despite this study aligning with the strategy of enhancing family proximity and engagement in care, the results showed no statistically significant differences in delirium prevalence, rate, or ICU delirium days because of the small effect size (Cohen’s d = 0.23). ³⁸

Patient and family perceptions of physical therapy for ICU patients were assessed in a study with a control arm and an early, intensity physical therapy arm. Patients were randomized to either arm after 4 days of mechanical ventilation. Whether on or off sedation, the patients needed to be alert, cooperative, and able to follow 3 of 5 simple commands. Physical therapy in the ICU included breathing exercises, resistance exercises, upper and lower extremity active range of motion, and marching in place over a 30-min session. If the patients still required vasopressors, interventions were limited to bed exercises only. Both patients and family members reported that physical therapy was necessary and beneficial to recovery, but family members underestimated the actual patient enjoyment of physical therapy. ³⁹

In a randomized controlled trial, 2 family-centered nutrition optimization (OPTimal) and decision support (My ICUGuide) interventions were compared with usual care to assess their effects on protein and energy intake during ICU and hospital stay and on family satisfaction, respectively. These interventions were designed to engage families in a shared decision-making process with healthcare professionals in which they supervise and support nutrition during patient recovery. Although the feasibility and increased acceptability of these interventions over time had been established in previous studies, the results from this trial failed to demonstrate the efficacy of either intervention. ⁴⁰

These studies^37-40 demonstrated the implementation of different approaches, from cognitive stimulation and physical therapy to decision-making and nutritional support. Regarding methodological quality, we can indicate that the studies were acceptable, as the effective results were heterogeneous. The interventions were generally well accepted and feasible but often lacked statistically significant results.

Discussion

This review highlights that 38.89% of family-centered interventions in ICUs utilized ICT platforms, either custom-designed or existing tools, to enhance communication. These technologies have demonstrated benefits, such as low cost, improved information quality, communication, trust, and family satisfaction, as well as increased involvement in care decisions. Real-time communication facilitates the sharing of patient values and preferences, fostering stronger family-professional partnerships^16,41-44

These findings address the multifaceted needs of ICU families, who often experience stress from the time of admission through post-discharge, influenced by patient condition, relationship dynamics, and emotional vulnerability.^45,46 High anxiety during admission diminishes families’ supportive role and may lengthen hospital stays.^47-50 Poor communication—characterized by unclear, contradictory, or insufficient information—erodes trust and increases family distress, especially when they feel excluded from clinical conversations.^51-53 Given the complexity of decisions involving patient outcomes, families should be empowered to participate meaningfully.^54-57

To meet this need, nurses—often the primary point of contact—must receive training in empathetic, clear communication, using language that is both honest and understandable. However, while technology supports accessibility and transparency, it must be implemented cautiously, respecting ethical, legal, and privacy standards. Digital interventions must not substitute the warmth and human presence that underpin family- and patient-centered nursing.^26,58,59

The reviewed interventions effectively reduced stress, anxiety, and emotional burden, particularly by involving families in care decisions and providing targeted education. However, implementation challenges remain, especially during life-support procedures when family presence can pose risks if not properly guided. Ongoing education for family members is essential to balance engagement and safety.

Methodologically, all studies were appraised using validated tools, with an expert panel guiding intervention selection. While the Johns Hopkins model offered a general quality overview, the ROB 2 tool allowed deeper insight into risk of bias. It revealed frequent limitations, including lack of sensitivity analyses, absent statistical plans, inadequate reporting on blinding, and unclear handling of participant attrition—highlighting the need for more robust study designs.

The study also faces several limitations. First, the lack of high-quality, consistently effective interventions applicable to real-world ICUs constrains the generalizability of findings. Many included studies had small samples and used non-randomized designs, limiting causal conclusions. Second, the wide variability in outcome measures and intervention designs introduced significant heterogeneity, precluding meta-analysis. Third, the use of non-validated interventions risks decreasing care quality when adopted in clinical settings. Finally, the ROB 2 analysis underscored methodological weaknesses, notably in follow-up strategies and statistical transparency.

These limitations reinforce the need for high-quality nursing research aimed at designing, implementing, and evaluating rigorously tested family-centered interventions. Future studies should emphasize methodological rigor, real-world applicability, and ethical integration of technology to support comprehensive care for both ICU patients and their families.

Conclusions

This systematic narrative review shows the wide range of strategies employed by researchers to strengthen family-centered interventions. The findings reveal considerable variability in effectiveness. While some approaches—such as web-based platforms and structured communication protocols—appear to enhance family satisfaction and psychological outcomes, others, like family visitation programs, showed limited impact, often due to implementation barriers rather than conceptual shortcomings. Notably, ethical safeguards—particularly in ICT-based interventions—received relatively limited discussion, suggesting an area that could benefit from greater attention in future research.

From a nursing perspective, the analyzed studies are operationally feasible, as they integrate with available ICU staff and resources to provide meaningful support. However, the evidence base remains limited by methodological variability, potential measurement threshold effects, and a lack of studies in underrepresented regions, particularly in Latin America, where no eligible studies were identified during the review period.

Future research should prioritize rigorous, context-sensitive clinical trials that evaluate not only outcomes but also implementation processes. Emphasis should be placed on standardizing communication training, ensuring equitable access to digital tools, and adapting interventions to diverse cultural and institutional settings to enhance their relevance and impact.

Appendix 1. Search Equations

PubMed

(“Family Nursing”[MeSH Terms] OR “Family-Centered Care”) AND “Intensive Care Units”[MeSH Terms] NOT “Intensive Care Units, Neonatal”[MeSH Terms]

Scopus

(“family nursing” OR “family-centered care”) AND TITLE-ABS-KEY(“intensive care unit” OR ICU) AND NOT TITLE-ABS-KEY(neonatal)

Web of Sciences

TS=(“family nursing” OR “family-centered care”) AND TS=(“intensive care unit” OR ICU) NOT TS=(neonatal)

CINAHL

Family-Centered Care AND Intensive Care Unit

Scielo

(“family nursing” OR “family-centered care”) AND (“intensive care unit” OR ICU) NOT neonatal

Appendix 2.

Individual risk assessment with ROB2.

Low risk

Some concerns

High risk

D1 Randomization process

D2 Deviations from the intended interventions

D3 Missing outcome data

D4 Measurement of the outcome

D5 Selection of the reported result