Publication

• Title: The FICUS cluster randomized controlled trial of a family support intervention in adult intensive care units: mental health and family functioning outcomes
• Acronym: FICUS
• Year: 2026
• Journal published in: Intensive Care Medicine
• Citation: Riguzzi M, Jeitziner MM, Rufer M, von Felten S, Verweij L, Oesch S, et al. The FICUS cluster randomized controlled trial of a family support intervention in adult intensive care units: mental health and family functioning outcomes. Intensive Care Med. 2026 May 18.

Context & Rationale

• Background

  Family members of critically ill adults are not passive observers of critical illness. They frequently act as surrogate decision-makers, informal caregivers, interpreters of patient preferences, and emotional supports, while simultaneously experiencing uncertainty, fear, bereavement risk, and later caregiving burden. One-year observational data show that caregivers of critically ill patients have substantial adverse psychological, social, and quality-of-life consequences after ICU admission.¹ Post-intensive care syndrome-family encompasses anxiety, depression, post-traumatic stress symptoms, prolonged grief, and family disruption after critical illness.² Modern family-centred ICU guidelines recommend better communication, family presence, shared decision-making, and structured support, but the certainty of evidence for many interventions remains limited.³
• Research Question/Hypothesis

  The FICUS programme asked whether a nurse-led, interprofessional, multicomponent family support intervention could improve family care during ICU admission and whether these proximal care improvements would translate into better family functioning, resilience, life satisfaction, well-being, quality of life, distress, anxiety, depression, and post-traumatic stress over 12 months. The protocol hypothesised improvement in family satisfaction, communication, nurse support, family management of critical illness, and individual mental health compared with usual care.⁴
• Why This Matters

  FICUS tests a highly plausible and humane intervention in a pragmatic ICU context: dedicated family nurses, structured family conversations, psychoeducation, relationship-focused support, and interprofessional family meetings. Its importance lies less in a single clinical endpoint and more in the mechanistic question it addresses: can better ICU family care prevent later psychological morbidity, or are later family mental health outcomes driven by broader trajectories such as bereavement, patient disability, social adversity, and pre-existing vulnerability?

Design & Methods

• Research Question: Among adult family members of critically ill adults treated in Swiss ICUs, does a nurse-led, interprofessional family support intervention, compared with usual care, improve family functioning and mental health over the first year after ICU treatment?
• Study Type: Multicentre, parallel-group, cluster-randomised, controlled, superiority, hybrid effectiveness-implementation trial in 16 adult ICUs across 12 hospitals in German-speaking Switzerland.
• Population:
  • Family members aged ≥18 years.
  • Family member was a significant other, close support person, or surrogate decision-maker of the critically ill patient.
  • Patients were expected to have ICU length of stay ≥48 hours and to be at high risk of death, long-lasting functional impairment, or prolonged mechanical ventilation.
  • Family members had to be cognitively able to participate and able to complete German-language measures.
  • Only one family member per family was enrolled.
  • Refusal of general patient consent for use of routine clinical data led to exclusion.
  • ICUs were eligible if they operated ≥8 beds and admitted ≥300 patients per year with ICU length of stay ≥48 hours.
  • ICUs with an existing protocolised interprofessional family support programme were excluded.
• Intervention:
  • A structured family care pathway delivered by designated ICU family nurses in collaboration with the interprofessional ICU team.
  • Core component 1: engaging and liaising, including relationship-building, continuity, coordination, referrals, and transition support.
  • Core component 2: supporting, including family assessment, psychoeducation, relationship-focused support, and therapeutic family conversations.
  • Core component 3: communicating, including structured interprofessional family meetings, listening to concerns, information provision, discussion of treatment options, and shared decision-making.
  • Pathway began within 96 hours of patient ICU admission and continued through ICU discharge or death and into the post-ICU phase.
  • A minimum of five intervention sessions reflecting all three components was prespecified, with dose and frequency increased according to family need.
  • Two to three ICU-certified nurses per intervention ICU assumed the family nurse role; they received 5 days of training, monthly case conferences, and refresher training.
  • Implementation support was tailored to local barriers and included leadership support, champions, team information, implementation support persons, and external facilitation.⁵
• Comparison:
  • Non-standardised usual ICU family care as established before trial start.
  • Usual care included local practices for visitation, written and oral information, family-clinician interaction, family meetings, support structures, and referral to auxiliary services.
  • Control ICUs could respond to family needs, but could not introduce a new protocolised family support intervention, family nursing role, or structured family support pathway.
• Blinding: Unblinded. Clinicians, interventionists, data assessors, and family participants were not masked because the intervention changed visible care processes and introduced a family nurse role. This is important because most outcomes were family-reported and therefore potentially vulnerable to expectation, gratitude, social desirability, and response bias.
• Statistics: The sample-size calculation was based on the primary FICUS outcome, family satisfaction with ICU care. A total of 16 clusters, 8 per arm, with approximately 56 family members per cluster, was planned to detect a 5.5-point difference in FS-ICU-24 total score, assuming SD 16.3, ICC 0.03, coefficient of variation in cluster size 0.2, 80% power, and two-sided α=0.05.⁴ Longitudinal secondary outcomes were analysed using linear mixed-effects models with random intercepts for cluster and family member, autoregressive residual correlation, baseline outcome adjustment, visit, treatment-by-visit interaction, and covariates including patient age, admission type, mechanical ventilation, SAPS-2, SOFA, family relationship, and prior ICU experience. The statistical analysis plan was published before final analysis.⁶
• Follow-Up Period: Outcomes were assessed at ICU admission, ICU discharge or death, and 3, 6, and 12 months after ICU discharge.

Key Results

This trial was not stopped early. Family members were enrolled from May 2022 to January 2024, and the mental health/family functioning report analysed outcomes over 12 months post-ICU.

• The central finding is a dissociation between process outcomes and distal outcomes: the intervention improved satisfaction, communication, and perceived support, but did not improve family functioning or mental health over 12 months.
• The direction of several 6-month psychological outcomes slightly favoured usual care, although these complete-case comparisons were not statistically significant; the multiple-imputation finding for lower life satisfaction in the intervention group should be interpreted cautiously given missingness and baseline imbalance.
• The trial identifies clinically important risk markers for family distress: patient mechanical ventilation, spouse/partner relationship, and prior ICU experience.

Internal Validity

• Randomisation and Allocation: Cluster randomisation was appropriate because the intervention altered ICU-level practice and contamination would have been likely with individual randomisation. Randomisation used minimisation stratified by ICU certification and site, and allocation occurred after cluster-level baseline assessment.
• Small Number of Clusters: Only 16 ICUs were randomised. This is a major limitation for cluster trials because chance imbalance in patient case mix, ICU culture, staffing, baseline family-care practice, and organisational processes can materially affect results.
• Dropout and Exclusions: Follow-up completion fell from 83.2% at discharge to 67.0% at 12 months. The main adjusted models included only 609–613 of 885 participants because late or missing baseline questionnaires could not be used as baseline covariates. This creates vulnerability to informative missingness, even though attrition was broadly similar between arms.
• Missing Data: Baseline outcome missingness was approximately 26%. Multiple imputation was undertaken and broadly supported the complete-case findings, but the imputation model itself had limitations: outcomes were often missing as a whole questionnaire batch rather than sporadically, and some imputed models had convergence problems.
• Performance Bias: Blinding was impossible. ICU teams and family members knew whether a family nurse pathway was present. This is particularly important for satisfaction, communication, support, and other self-reported outcomes.
• Detection Bias: Outcomes were family-reported using validated instruments. This is appropriate for the constructs being measured, but it makes the trial sensitive to expectation effects, gratitude bias, participant burden, and differential interpretation of questionnaire items.
• Protocol Adherence: The intervention was delivered with meaningful contact volume, but full minimum pathway fidelity was low at 23.1%. The post-ICU therapeutic conversation was delivered within the prespecified window in only 43.9% of intervention families. This materially weakens any inference that the full intended intervention was ineffective.
• Baseline Characteristics: Family member characteristics were broadly similar, but patient-level imbalances favoured the control arm. Intervention-arm patients had more mechanical ventilation (60.3% vs 49.5%), more ICU death (19.9% vs 13.2%), longer median ICU stay (10 vs 8 days), and fewer unplanned admissions (77.1% vs 89.9%). Covariate adjustment helps but cannot exclude residual confounding.
• Heterogeneity: The intervention was intentionally adaptive, and ICUs differed in local family-care processes. This improves pragmatism but introduces heterogeneity of implementation, usual care, and treatment contrast.
• Timing: The intervention began early, within 96 hours of ICU admission, which is plausible for improving communication and family orientation. The weaker post-ICU delivery is important because bereavement, recovery, caregiving burden, and psychological symptoms often unfold after ICU discharge.
• Dose: Median exposure was 250 minutes per family, but this was distributed across a prolonged and heterogeneous illness trajectory. For prevention or treatment of anxiety, depression, PTSD, or prolonged grief, this may be too brief, too early, insufficiently psychologically specialised, or insufficiently sustained.
• Separation of the Variable of Interest: There was evidence of care-process separation. The primary FICUS report showed FS-ICU 81.78 versus 79.39, mean difference 2.39; communication mean difference 0.37; and cognitive/emotional support mean difference 8.71.⁷ This confirms that the intervention changed family care, even though it did not change distal psychological outcomes.
• Outcome Assessment: Measures were clinically relevant and psychometrically reasonable. Reported internal consistency was good for FAD-GF-12, BRS-6, SWLS-5, WHO-5, HADS-A, and HADS-D; IES-6 reliability was lower at baseline but improved across follow-up.
• Statistical Rigor: The use of mixed-effects models was appropriate for cluster randomisation and repeated measures. However, the trial was powered for family satisfaction, not for each 12-month mental health and family-functioning endpoint. Multiple outcomes, incomplete baseline adjustment, and secondary-outcome interpretation require caution.

Conclusion on Internal Validity: Internal validity is moderate. The trial was well designed, prospectively planned, multicentre, and analytically sophisticated, but small cluster number, unblinded subjective outcomes, baseline patient imbalance, substantial missing baseline data, and low full intervention fidelity limit causal certainty for the negative mental health findings.

External Validity

• Population Representativeness: The trial included a broad high-risk adult ICU population rather than only dying patients or surrogate decision-makers, increasing clinical relevance. However, only one family member per family was enrolled, and the experience of wider family systems was not directly measured.
• Language and Social Representativeness: German-language questionnaire completion was required. Non-German-speaking families, socially disadvantaged families, and those too burdened to participate are likely underrepresented.
• Healthcare System: The intervention was tested in Swiss adult ICUs, a high-resource environment with ICU-certified nurses, established clinical infrastructure, and capacity for training and implementation facilitation.
• Intervention Workforce: Family nurses were experienced ICU-certified nurses, with median work experience of 19 years. Translation to systems with limited specialist nursing capacity, lower staffing ratios, or less implementation support may be challenging.
• Usual Care Comparator: Usual care was non-standardised and may vary widely across countries and institutions. In ICUs where baseline family support is poor, the intervention effect on satisfaction and communication could be larger; in mature family-centred ICUs, ceiling effects may reduce measurable benefit.
• Clinical Applicability: The findings are most applicable to high-resource adult ICUs considering a dedicated family nurse model to improve communication and perceived support, rather than to settings seeking a standalone intervention to prevent PICS-F.

Conclusion on External Validity: External validity is moderate for high-resource adult ICUs with the capacity to train family nurses and implement a structured pathway. Generalisability is limited for non-German-speaking families, resource-limited systems, and populations with high social complexity or limited access to post-ICU support.

Strengths & Limitations

• Strengths:
  • Multicentre cluster-randomised design across 16 adult ICUs.
  • Cluster allocation reduced contamination risk for an ICU-level behavioural and organisational intervention.
  • Prospective protocol and statistical analysis plan.⁴⁶
  • Patient and family advisory group involvement in trial design and implementation.
  • Detailed intervention theory, implementation support, process evaluation, and fidelity measurement.
  • Longitudinal follow-up to 12 months, unusual for ICU family-support trials.
  • Use of validated family functioning, resilience, well-being, quality-of-life, distress, anxiety, depression, and post-traumatic stress measures.
  • Clear separation in proximal process outcomes in the primary FICUS report, confirming that the intervention did change family-care experience.⁷
• Limitations:
  • Only 16 clusters, with clinically important baseline patient imbalance.
  • Unblinded clinicians and participants.
  • Self-reported outcomes susceptible to response and expectation bias.
  • Substantial missing baseline data and incomplete 12-month follow-up.
  • Only about 70% of the enrolled sample entered the main adjusted longitudinal models.
  • Low combined fidelity to the full family care pathway, especially the post-ICU therapeutic conversation.
  • Secondary outcomes were not the basis of the original power calculation.
  • Non-German-speaking and highly burdened families likely underrepresented.
  • Patient post-ICU trajectories, subsequent disability, caregiving load, and access to mental health care were not sufficiently captured to explain later family outcomes.

Interpretation & Why It Matters

• Main Interpretation

  FICUS demonstrates that a dedicated nurse-led family support pathway can improve family satisfaction, communication, and perceived emotional/cognitive support, but these gains did not translate into improved family functioning or mental health over 12 months.
• Clinical Practice

  ICUs should not promise that a family nurse pathway will prevent anxiety, depression, PTSD, or family dysfunction. The stronger and more defensible claim is that such a pathway can improve the quality and humanity of family care during critical illness.
• Mechanistic Lesson

  Family psychological outcomes after ICU are distal and multifactorial. Communication quality is important, but later distress may depend on bereavement, patient survival with disability, caregiver role strain, prior mental health, social support, coping style, and access to specialist psychological care.
• Targeting

  The covariate findings suggest that spouses/partners, family members of mechanically ventilated patients, and family members with prior ICU exposure may warrant more intensive screening and follow-up.
• Implementation

  The low full-fidelity rate implies that real-world implementation of complex family pathways is difficult. Future models may require simpler core components, stronger post-ICU continuity, risk stratification, and escalation to mental health professionals.

Controversies & Other Evidence

• Clinically meaningful benefit versus statistical significance: The JAMA Internal Medicine editor’s note framed FICUS as a “high-touch, low-tech” intervention and emphasised that the primary satisfaction benefit was statistically significant but small and of uncertain clinical importance, particularly given high satisfaction in both trial arms. The same commentary highlighted the stronger effects on communication and perceived support.⁸
• Process outcomes may not be surrogates for mental health outcomes: FICUS supports a key methodological point: improved communication, support, and satisfaction should not be assumed to prevent PICS-F. The causal pathway from ICU communication to later psychological morbidity is plausible but incomplete.
• Ceiling effects: Baseline/usual-care satisfaction was already high in the primary report, with usual-care FS-ICU 79.39 out of 100. This restricts measurable improvement and makes a small mean difference harder to interpret clinically.
• Intervention dose and duration: The median intervention duration was 250 minutes, but full pathway fidelity was only 23.1%, and the post-ICU therapeutic conversation was delivered within the prespecified window in 43.9% of intervention families. A negative distal outcome trial with incomplete delivery should not be interpreted as proof that more intensive or sustained family mental health interventions are ineffective.
• Comparison with PARTNER: The PARTNER trial tested a nurse-led family-support intervention for surrogate decision-makers of patients with chronic critical illness and did not show consistent improvement in psychological outcomes, reinforcing that structured ICU family support does not reliably reduce later anxiety, depression, or PTSD.⁹
• Communication facilitator trials: A US communication facilitator trial reduced depressive symptoms at 6 months but not anxiety, whereas later nurse-facilitator work in France did not reproduce a consistent reduction in family psychological burden.¹⁰¹¹
• End-of-life specificity: A three-step support strategy for relatives of patients dying in ICU reduced prolonged grief outcomes, suggesting that interventions targeted to bereavement pathways may have clearer psychological benefit than broad, general ICU family support.¹²
• Four Supports trial: The Four Supports intervention was delivered with high fidelity but did not improve surrogate anxiety/depression at 6 months and was associated with higher post-traumatic stress symptoms in the intervention arm. An accompanying editorial argued that ICU surrogate psychological outcomes remain difficult to improve and may require more intensive psychological and coping support than communication-focused interventions alone.¹³¹⁴
• Caregiver Pathway contrast: The Norwegian Caregiver Pathway trial reported reduced PICS-F symptoms at 6 months among caregivers of ICU survivors, and its 12-month report suggested persistent benefit particularly among caregivers of survivors for PTSD and anxiety. This contrasts with FICUS and supports the hypothesis that structured post-ICU follow-up, survivor-specific trajectories, and targeted caregiver support may be crucial.¹⁵¹⁶
• Systematic review context: Systematic reviews show that many family-centred ICU interventions improve at least one family-centred or patient-centred outcome, but intervention heterogeneity, outcome multiplicity, and variable trial quality make it difficult to identify the active components needed to improve long-term family mental health.¹⁷¹⁸
• Qualitative signal: The FICUS qualitative evaluation found that families valued trusted continuity, proactive communication, and emotional, behavioural, and practical support. This reinforces that a trial can be negative for distal mental health outcomes while still identifying care processes that families experience as meaningful.¹⁹

Summary

• FICUS was a 16-ICU Swiss cluster-randomised trial of a nurse-led, interprofessional family support intervention for family members of critically ill adults.
• The intervention improved family satisfaction, communication quality, and cognitive/emotional support in the primary FICUS report, but did not improve family functioning, resilience, life satisfaction, quality of life, distress, anxiety, depression, or post-traumatic stress over 12 months.
• Mechanical ventilation, spouse/partner relationship, and prior ICU experience were associated with worse or more complex family outcomes, but did not identify a subgroup with clear intervention benefit.
• Internal validity is limited by small cluster number, unblinded subjective outcomes, baseline patient imbalance, missing baseline data, and low full fidelity to the intended pathway.
• The trial is best interpreted as evidence that improving ICU family-care processes is achievable and valuable, but insufficient on its own to prevent long-term PICS-F at the tested dose and duration.

Overall Takeaway

FICUS is an important pragmatic ICU family-care trial, but not a definitive practice-changing trial for preventing PICS-F. It shows that a nurse-led family support pathway can improve the experience and quality of family care, while also demonstrating that long-term family mental health outcomes are harder to modify and probably require risk-stratified, sustained, and psychologically specialised post-ICU support.

Bibliography

• 1.Cameron JI, Chu LM, Matte A, Tomlinson G, Chan L, Thomas C, et al. One-year outcomes in caregivers of critically ill patients. N Engl J Med. 2016;374:1831-1841.
• 2.Shirasaki K, Hifumi T, Nakanishi N, et al. Postintensive care syndrome family: a comprehensive review. Acute Med Surg. 2024;11:e939.
• 3.Hwang DY, Oczkowski SJW, Lewis K, Birriel B, Downar J, Farrier CE, et al. Society of Critical Care Medicine guidelines on family-centered care for adult ICUs: 2024. Crit Care Med. 2025;53:e465-e482.
• 4.Naef R, Filipovic M, Jeitziner MM, von Felten S, Safford J, Riguzzi M, et al. A multicomponent family support intervention in intensive care units: study protocol for a multicenter cluster-randomized trial (FICUS Trial). Trials. 2022;23:533.
• 5.Verweij L, Oesch S, Naef R. Tailored implementation of the FICUS multicomponent family support intervention in adult intensive care units: findings from a mixed methods contextual analysis. BMC Health Serv Res. 2023;23:1339.
• 6.von Felten S, Filipovic M, Jeitziner MM, Verweij L, Riguzzi M, Naef R. Multicomponent family support intervention in intensive care units: statistical analysis plan for the cluster-randomized controlled FICUS trial. Trials. 2024;25:568.
• 7.Naef R, Jeitziner MM, Riguzzi M, von Felten S, Verweij L, Rufer M, et al. Nurse-led family support intervention for families of critically ill patients: the FICUS cluster randomized clinical trial. JAMA Intern Med. 2025;185:1138-1149.
• 8.Stall NM, Mody L. Supporting families in the adult intensive care unit with high touch-low tech interventions. JAMA Intern Med. 2025;185:1150.
• 9.White DB, Angus DC, Shields AM, Buddadhumaruk P, Pidro C, Paner C, et al. A randomized trial of a family-support intervention in intensive care units. N Engl J Med. 2018;378:2365-2375.
• 10.Curtis JR, Treece PD, Nielsen EL, Gold J, Ciechanowski PS, Shannon SE, et al. Randomized trial of communication facilitators to reduce family distress and intensity of end-of-life care. Am J Respir Crit Care Med. 2016;193:154-162.
• 11.Kentish-Barnes N, Azoulay E, Reignier J, Cariou A, Lafarge A, Huet O, et al. A randomised controlled trial of a nurse facilitator to promote communication for family members of critically ill patients. Intensive Care Med. 2024;50:712-724.
• 12.Kentish-Barnes N, Chevret S, Valade S, Jaber S, Kerhuel L, Guisset O, et al. A three-step support strategy for relatives of patients dying in the intensive care unit: a cluster randomised trial. Lancet. 2022;399:656-664.
• 13.Butler RA, Seaman JB, Felman K, Stonehouse W, San Pedro R, Morse JQ, et al. Randomized clinical trial of the Four Supports intervention for surrogate decision-makers in intensive care units. Am J Respir Crit Care Med. 2025;211:370-380.
• 14.Courtright KR, Downar J. The struggle continues: improving outcomes for surrogate decision-makers after the ICU. Am J Respir Crit Care Med. 2025;211:309-311.
• 15.Watland S, Solberg Nes L, Ekeberg Ø, Rostrup M, Hanson E, Ekstedt M, et al. The Caregiver Pathway intervention can contribute to reduced post-intensive care syndrome among family caregivers of ICU survivors: a randomized controlled trial. Crit Care Med. 2025;53:e555-e566.
• 16.Watland S, Solberg Nes L, Ekeberg Ø, Rostrup M, Hanson E, Ekstedt M, et al. Effects of The Caregiver Pathway intervention on symptoms of post-intensive care syndrome among family caregivers to critically ill patients: long-term results from a randomized controlled trial. Intensive Care Med. 2025;51:2042-2053.
• 17.Wang G, Antel R, Goldfarb MJ. The impact of randomized family-centered interventions on family-centered outcomes in the adult intensive care unit: a systematic review. J Intensive Care Med. 2023;38:690-701.
• 18.Duong J, Wang G, Lean G, Slobod D, Goldfarb M. Family-centered interventions and patient outcomes in the adult intensive care unit: a systematic review of randomized controlled trials. J Crit Care. 2024;83:154829.
• 19.Walkling R, Verweij L, Oesch S, Naef R. Families’ experiences with a novel nurse-led family support intervention in adult intensive care units: a qualitative thematic analysis. Intensive Crit Care Nurs. 2026;92:104248.