Translation and cultural adaptation of an objective pain assessment tool for Finnish ICU patients

Linguistic Validation of Genomic Nursing Concept Inventory to Finnish Applying Mandysova's Decision Tree Algorithm

Background and Purpose: Currently, there is no available Finnish version of the Genomic Nursing Concept Inventory tool (GNCI). This study tested the validity, reliability, and clinical usability of a Finnish translation. Methods: A decision tree algorithm was used to guide the translation, as per International Society for Pharmacoeconomics and Outcomes Research guidelines. Item-Content Validity Index (I-CVI), modified kappa (k*) statistics, and Cronbach's alpha were calculated. Results: The I-CVI and k* values were "good" to "excellent" (I-CVI = 0.63-1.00, k* = 0.52-1.00), and Cronbach's alpha value was "good" (α = 0.816; 95% confidence interval: 0.567-0.956). Conclusion: The Mandysova's decision tree algorithm provided clear and rigorous direction for the translation and validity of the Finnish GNCI.

Translation and cultural adaptation of the Family Centered Care Assessment Scale (FCCAS) for Finnish pediatric nursing

PURPOSE

The purpose of this study was to translate and culturally adapt the Family Centered Care Assessment Scale (FCCAS) to evaluate family-centered care in the context of Finnish pediatric nursing.

DESIGN AND METHODS

The translation and cultural adaptation were done according guidelines of International Society for Pharmacoeconomics and Outcomes (ISPOR), which constitute a systematic process including ten phases. The presented research included cognitive interviews conducted with the end user of the scale: parents of hospitalized children.

RESULTS

The FCCAS was successfully translated and culturally adapted to the Finnish context. Translational, clinical and parental expertise were used to develop the scale. Conceptual equivalence was achieved in the translation. In the cultural adaptation, some of the items were modified based on experts' assessments to make them comprehensible and appropriate to the Finnish culture. The scale showed good evidence of content. Reporting of the study adheres to the COSMIN checklist.

CONCLUSIONS

Combining ISPOR guidelines and cognitive interviews are recommended to use in the translation and cultural adaptation process. Nursing staff and parents' involvement and awareness of family-centered care have been concretized. Systematic translation and cultural adaptation have prepared a Finnish version of the scale for psychometric testing.

PRACTICE IMPLICATIONS

The study outlines how rigorous methodological approaches can be applied to the translation and cultural adaptation of a measurement tool. The developed scale includes items which comprehensively cover family-centered care characteristics. In following study, it will be possible to evaluate the extent to which family-centered care is implemented in Finnish pediatric nursing.

Pain assessment of the adult sedated and ventilated patients in the intensive care setting: A scoping review

BACKGROUND

Pain is frequently encountered in the intensive care setting. Given the impact of pain assessment on patient outcomes and length of hospital stay, studies have been conducted to validate tools, establish guidelines and cast light on practices relating to pain assessment.

OBJECTIVE

To examine the extent, range and nature of the evidence around pain assessment practices in adult patients who cannot self-report pain in the intensive care setting and summarise the findings from a heterogenous body of evidence to aid in the planning and the conduct of future research and management of patient care. The specific patient cohort studied was the sedated/ ventilated patient within the intensive care setting.

DESIGN

A scoping review protocol utilised the Preferred Reporting Items for Systematic Reviews and Meta-Analysis extension for Scoping review checklist (PRISMA-ScR).

METHODS

The review comprised of five phases: identifying the research question, identifying relevant studies, study selection, charting the data and collating, summarizing, and reporting the results. Databases were systematically searched from January to April 2020. Databases included were Scopus, Web of Science, Medline via Ovid, CINAHL COMPLETE via EBSCO host, Health Source and PUBMED. Limits were applied on dates (2000 to current), language (English), subject (human) and age (adult). Key words used were "pain", "assessment", "measurement", "tools", "instruments", "practices", "sedated", "ventilated", "adult". A hand search technique was used to search citations within articles. Database alerts were set to apprise the availability of research articles pertaining to pain assessment practices in the intensive care setting.

RESULTS

The review uncovered literature categorised under five general themes: behaviour pain assessment tools, pain assessment guidelines, position statements and quality improvement projects, enablers and barriers to pain assessment, and evidence appertaining to actual practices. Behaviour pain assessment tools are the benchmark for pain assessment of sedated and ventilated patients. The reliability and validity of physiologic parameters to assess pain is yet to be determined. Issues of compliance with pain assessment guidelines and tools exist and impact on practices. In some countries like Australia, there is a dearth of information regarding the prevalence and characteristics of patients receiving analgesia, type of analgesia used, pain assessment practices and the process of recording pain management. In general, pain assessment varies across different intensive care settings and lacks consistency.

CONCLUSION

Research on pain assessment practices requires further investigation to explore the causative mechanisms that contribute to poor compliance with established pain management guidelines. The protocol of this review was registered with Open Science Framework (https://osf.io/25a6) Tweetable abstract: Pain assessment in intensive care settings lacks consistency. New information is needed to understand the causative mechanisms underpinning poor compliance with guidelines.

A Psychometric Analysis Update of Behavioral Pain Assessment Tools for Noncommunicative, Critically Ill Adults

This is an updated, comprehensive review of the psychometric properties of behavioral pain assessment tools for use with noncommunicative, critically ill adults. Articles were searched in 5 health databases. A total of 106 articles were analyzed, including 54 recently published papers. Nine behavioral pain assessment tools developed for noncommunicative critically ill adults and 4 tools developed for other non-communicative populations were included. The scale development process, reliability, validity, feasibility, and clinical utility were analyzed using a 0 to 20 scoring system, and quality of evidence was also evaluated. The Behavioral Pain Scale, the Behavioral Pain Scale-Nonintubated, and the Critical-Care Pain Observation Tool remain the tools with the strongest psychometric properties, with validation testing having been conducted in multiple countries and various languages. Other tools may be good alternatives, but additional research on them is necessary.

Translation into Spanish and Cultural Adaptation of the Critical-Care Pain Observation Tool

BACKGROUND

The Critical-Care Pain Observation Tool (CPOT) is recommended for evaluating pain behaviors in patients in the intensive care unit who are unable to report pain. The source of the only published Spanish version of the CPOT does not verify that it underwent a formal translation process.

OBJECTIVE

To describe the translation into Spanish and cultural adaptation of the original French version of the CPOT.

METHODS

Key persons in the translation process included one with a master's degree in translation, a critical care physician, nurse faculty members with vast experience in intensive care units, and the instrument's developer. This team followed the Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes Measures as a guide to translate and culturally adapt the CPOT.

RESULTS

The first Spanish-language version was back translated to French and was also compared with the English version. Revisions necessitated a second version, which was submitted to experts in critical care. Their modifications required a third version, which was back translated to French and discussed with the CPOT developer, after which a fourth version was created. Finally, a linguistic expert proofread the tool, and the translation leaders incorporated the recommendations, thereby obtaining a final Spanish version.

CONCLUSION

The Spanish version is ready to undergo validation with patients in the intensive care unit, which is the next step toward its use in assessing pain behaviors among patients in intensive care units where Spanish is spoken.

Validating a pain assessment tool in heterogeneous ICU patients: Is it possible?

BACKGROUND

Non-communicative adult ICU patients are vulnerable to inadequate pain management with potentially severe consequences. In German-speaking countries, there is limited availability of a validated pain assessment tool for this population.

AIM

The aim of this observational study was to test the German version of the Critical-Care Pain Observation Tool (CPOT) in a heterogeneous adult ICU population.

METHODS

The CPOT's feasibility for clinical use was evaluated via a questionnaire. For validity and reliability testing, the CPOT was compared with the Behavioural Pain Scale (BPS) and patient's self-report in 60 patients during 480 observations simultaneously performed by two raters.

RESULTS

The feasibility evaluation demonstrated high satisfaction with clinical usability (85% of responses 4 or 5 on a 5-point Likert scale). The CPOT revealed excellent criterion validity [agreement between CPOT and BPS 94.0%, correlation of CPOT and BPS sum scores r = 0.91 (P < .05), agreement of CPOT with patient self-report 81.4%], good discriminant validity [mean difference of CPOT scores between at rest and non-painful stimulus 0.33 (P < .029), mean difference of CPOT scores between at rest, and painful stimulus 2.19 (P < .001)], for a CPOT cut-off score of >2 a high sensitivity and specificity (93% and 84%), high positive predictive value (85%), and a high negative predictive value (93%). The CPOT showed acceptable internal consistency (Cronbach's α 0.79) and high inter-rater reliability [90% agreement, no differences in CPOT sum scores in 64.2% of observations, and correlation for CPOT sum scores r = 0.72 (P < .05)]. Self-report obtained in patients with delirium did not correlate with the CPOT rating in 62% of patients.

CONCLUSION

This is the first validation study of the CPOT evaluating all of the described validity dimensions, including feasibility, at once. The results are congruent with previous validations of the CPOT with homogeneous samples and show that it is possible to validate a tool with a heterogeneous sample. Further research should be done to improve pain assessment and treatment in ICU patients with delirium.

RELEVANCE TO CLINICAL PRACTICE

The German CPOT version can be recommended for ICUs in German-speaking countries.

Evaluation of Observational and Behavioural Pain Assessment Tools in Nonverbal Intubated Critically Adult Patients after Open - Heart Surgery: A Systematic Review

BACKGROUND

Over 70% of patients hospitalised in an intensive care unit (ICU) often experience moderate to severe pain due to pre-existing diseases, trauma, surgery, aggressive procedures, and routine ICU care. Many patients hospitalised in ICU are not able to speak and express their pain due to various causes, including mechanical ventilation, reduced consciousness, and administration of sedative drugs. Therefore, the use of observational and behavioural pain tools is recommended in this group of patients given their inability to express pain.

AIM

To examine the existing observational and behavioural tools for assessment of in Nonverbal Intubated Critically Adult Patients after Open-Heart Surgery.

METHODS

A systematic review of available observational and behavioural tools for assessment of pain was undertaken using the COSMIN checklist. A literature search was conducted using the following databases: Ovid, Science Direct, Scopus, PubMed, and CINHAL databases, Google Scholar search engine as well as Persian resources Sid, Magiran, Iran doc, and IranMedex up to the end of 2017 were reviewed.

RESULTS

A total of 47 studies that had examined five tools used in intensive care units after cardiac surgery in patients under mechanical ventilation were reviewed. Each of the five tools included behavioural and observational items, and only one tool had physiological items. All the tools had been evaluated regarding validity and reliability. In the three tools, sensitivity, specificity, responsiveness, and satisfaction were considered.

CONCLUSION

Based on available evidence and investigations, CPOT and BPS tools have good validity and reliability to be used in pain assessment in Nonverbal Intubated Critically Adult Patients after Open-Heart Surgery. The NVPS tool requires more studies to be further confirmed before the assessment of pain in this group of patients.

Pain measurement techniques: spotlight on mechanically ventilated patients

Background

Procedural pain is a frequent problem in intensive care units (ICUs). For that, pain assessment has been increasingly introduced to the ICU professional’s routine, and studies have been developed to show the relevance of measuring pain in critically ill patients.

Objective

This review aimed to describe pain measurement techniques for mechanically ventilated adult patients based on evidence and already published.

Method

Systematic literature search was performed on PubMed and Google Scholar. Keywords “pain”, “pain measurement”, “intensive care units” and “respiration, artificial” were combined to the Boolean operator AND. No language or publication year was limited in this search. The purpose and method of all papers were analyzed and only studies which described pain assessment in mechanically ventilated patients were included in this review.

Results

Objective methods were found in the literature to assess pain in mechanically ventilated adults. Behavioral scales were the most used method for pain measurement in noncommunicative patients. Vital signs were used, but the reliability of this method was questioned. Pupillometry, bispectral index and skin conductance were found and described as pain assessment methods.

Conclusion

This review showed that objective measures, as behavioral scales, are the gold standard tools to measure pain intensity in noncommunicative subjects. These data contribute to professionals’ knowledge about ICU pain measurement and emphasize its importance and consequences for adequate pain management.

Psychometric Assessment of Physiologic and Behavioral Pain Indicators in Polish Versions of the Pain Assessment Scales

BACKGROUND

There is an urgent need to prepare a reliable and accurate tool for pain assessment in patients who are unable to self-report. Translating pain assessment scales into foreign languages requires further validation testing.

AIM

The aim of the study was to carry out psychometric assessment of behavioral and physiological indicators of pain included in two Polish versions of pain assessment scales, the Behavioral Pain Scale (BPS) and the original Adult Non-Verbal Pain Scale (NVPS).

DESIGN

A prospective repeated-measure descriptive study was conducted.

SETTINGS AND PARTICIPANTS

Twenty-eight adult non-communicative mechanically ventilated ICU patients were included in the study. The study took place in five hospitals in Poland, one 15-bed general ICU of a university teaching hospital and four 6-bed medical ICUs of district hospitals.

METHODS

Pain assessment was conducted at rest, during non-painful and painful procedures independently by two observers.

RESULTS

Internal consistency of the Polish version of the scales was below the expected 0.7 value (Cronbach's alpha for the BPS 0.6883 and NVPS 0.6697). Principal component analysis showed that for the Polish version of the BPS, all three domains formed one separate factor (63.9%), while in the case of the NVPS two separate factors were found, one covering four domains of the NVPS (47.1%) and the other exclusively covering the category of Vital sign (20.2%). There was a significant difference between the pain scores with the NVPS (χ² = 228.95 p < .001) and the BPS (χ² = 236.46 p < .001) during three observation phases. There were no significant differences between scores obtained by different raters. The analysis of variance demonstrated a statistically significant difference in the values of physiological indicators of pain (SBP, DBP, MAP) between observation phases.

CONCLUSIONS

The Polish version of the BPS has better psychometric properties than the Polish version of the NVPS. It is necessary to define precisely the descriptors used in the scales and to implement a staff training program.

[Process for implementing a pain assessment scale for patients with brain injuries: description of a project methodology based on the Integrated Model of Consultation]

Background : in a neurosurgery unit, nurses selected the Critical Pain Observation Tool (CPOT) based on evidence to assess pain in brain-injured patients. However, months after implementation, nursing managers have observed an underutilization.Objectives : support a care team to overcome the pitfalls encountered during the implementation of the CPOT scale for brain-injured patients in neurosurgery unit.Methods : the Lescarbeau, Payette and St-Arnaud's Integrated Model of Consultation was selected. In addition to a scientific literature review, self-administered questionnaire and three interview guides were developed to gather the views of caregivers.Results : the process allowed to identify pitfalls at level of interprofessional collaboration CPOT scale and implementation processes. Improving interprofessional collaboration and adaptation of CPOT scale for brain-injured patients were withholding action priorities.Discussion : a rigorous methodology, the mutual recognition of clinical skills and the development of relationship of trust are prerequisites for the success of clinical innovation.Conclusion : the Integrated Intervention Model is a methodology of choice allowing to take into consideration both evidence and preferences of all actors at every step of the process led to informed choices and priorities setting for a successful implementation.

Reliability of three linguistically and culturally validated pain assessment tools for sedated ICU patients by ICU nurses in Finland

Background and aims:

Pain assessment in intensive care is challenging, especially when the patients are sedated. Sedated patients who cannot communicate verbally are at risk of suffering from pain that remains unnoticed without careful pain assessment. Some tools have been developed for use with sedated patients. The Behavioral Pain Scale (BPS), the Critical-Care Pain Observation Tool (CPOT) and the Nonverbal Adult Pain Assessment Scale (NVPS) have shown promising psychometric qualities. We translated and culturally adapted these three tools for the Finnish intensive care environment. The objective of this feasibility study was to test the reliability of the three pain assessment tools translated into Finnish for use with sedated intensive care patients.

Methods:

Six sedated intensive care patients were videorecorded while they underwent two procedures: an endotracheal suctioning was the nociceptive procedure, and the non-nociceptive treatment was creaming of the feet. Eight experts assessed the patients’ pain by observing video recordings. They assessed the pain using four instruments: the BPS, the CPOT and the NVPS, and the Numeric Rating Scale (NRS) served as a control instrument. Each expert assessed the patients’ pain at five measurement points: (1) right before the procedure, (2) during the endotracheal suctioning, (3) during rest (4) during the creaming of the feet, and (5) after 20 min of rest. Internal consistency and inter-rater reliability of the tools were evaluated. After 6 months, the video recordings were evaluated for testing the test-retest reliability.

Results:

Using the BPS, the CPOT, the NVPS and the NRS, 960 assessments were obtained. Internal consistency with Cronbach’s alpha coefficient varied greatly with all the instruments. The lowest values were seen at those measurement points where the pain scores were 0. The highest scores were achieved after the endotracheal suctioning at rest: for the BPS, the score was 0.86; for the CPOT, 0.96; and for the NVPS, 0.90. The inter-rater reliability using the Shrout-Fleiss intraclass correlation coefficient (ICC) tests showed the best results after the painful procedure and during the creaming. The scores were slightly lower for the BPS compared to the CPOT and the NVPS. The test-retest results using the Bland-Altman plots show that all instruments gave similar results.

Conclusions:

To our knowledge, this is the first time all three behavioral pain assessment tools have been evaluated in the same study in a language other than English or French. All three tools had good internal consistency, but it was better for the CPOT and the NVPS compared to the BPS. The inter-rater reliability was best for the NVPS. The test-retest reliability was strongest for the CPOT. The three tools proved to be reliable for further testing in clinical use.

Implications:

There is a need for feasible, valid and reliable pain assessment tools for pain assessment of sedated ICU patients in Finland. This was the first time the psychometric properties of these tools were tested in Finnish use. Based on the results, all three instruments could be tested further in clinical use for sedated ICU patients in Finland.

Video education for critical care nurses to assess pain with a behavioural pain assessment tool: A descriptive comparative study

AIM

To evaluate the impact of video education on critical care nurses' knowledge and skills in using a behavioural pain assessment tool for intensive care patients and to explore the nurses' experiences with video education.

METHODS

Forty-eight nurses in one intensive care unit watched an educational video on the use of the Critical-Care Pain Observation Tool, then assessed pain in two patients with the tool and took a knowledge test. The researcher made parallel pain assessments. Interrater reliability of patients' pain assessment between nurses and the researcher was determined to examine nurses' skills in using the tool after education. Twenty nurses were interviewed about their experiences with the video education. Interviews were analysed with deductive thematic analysis.

RESULTS

The knowledge test scores indicated that the nurses learned the principles of how to use the tool. The interrater reliability of pain assessments reached a moderate level of agreement during the painful procedure, with a weighted kappa coefficient value of 0.48, CL [0.37, 0.58]. The nurses perceived video education positively, but requested additional interaction.

CONCLUSIONS

Video education is useful in teaching the principles of using a pain assessment tool. Additional clinical training is required for nurses to reach adequate skills in using the tool.

[Validation of the Brazilian version of Behavioral Pain Scale in adult sedated and mechanically ventilated patients]

BACKGROUND AND OBJECTIVES

The Behavioral Pain Scale is a pain assessment tool for uncommunicative and sedated Intensive Care Unit patients. The lack of a Brazilian scale for pain assessment in adults mechanically ventilated justifies the relevance of this study that aimed to validate the Brazilian version of Behavioral Pain Scale as well as to correlate its scores with the records of physiological parameters, sedation level and severity of disease.

METHODS

Twenty-five Intensive Care Unit adult patients were included in this study. The Brazilian Behavioral Pain Scale version (previously translated and culturally adapted) and the recording of physiological parameters were performed by two investigators simultaneously during rest, during eye cleaning (non-painful stimulus) and during endotracheal suctioning (painful stimulus).

RESULTS

High values of responsiveness coefficient (coefficient=3.22) were observed. The Cronbach's alpha of total Behavioral Pain Scale score at eye cleaning and endotracheal suctioning was 0.8. The intraclass correlation coefficient of total Behavioral Pain Scale score was ≥ 0.8 at eye cleaning and endotracheal suctioning. There was a significant highest Behavioral Pain Scale score during application of painful procedure when compared with rest period (p≤0.0001). However, no correlations were observed between pain and hemodynamic parameters, sedation level, and severity of disease.

CONCLUSIONS

This pioneer validation study of Brazilian Behavioral Pain Scale exhibits satisfactory index of internal consistency, interrater reliability, responsiveness and validity. Therefore, the Brazilian Behavioral Pain Scale version was considered a valid instrument for being used in adult sedated and mechanically ventilated patients in Brazil.

A systematic review of observational pain assessment instruments for use with nonverbal intubated critically ill adult patients in the emergency department: an assessment of their suitability and psychometric properties

AIM AND OBJECTIVE

To examine the psychometric properties and suitability of the available observational pain instruments for potential use with nonverbal critically ill adult patients in the emergency department.

BACKGROUND

In the emergency department, assessing pain in critically ill patients is challenging, especially those unable to communicate the presence of pain. Critically ill patients are commonly unable to verbally communicate pain due to altered oral communication (e.g. endotracheal intubation) and/or diminished consciousness (e.g. sedation, delirium), placing them at great risk of inadequate pain management. Over half of intensive care critically ill intubated patients experience moderate-to-severe pain whilst intubated and mechanically ventilated.

DESIGN

Systematic review.

DATA SOURCES

The CINAHL, EMBASE, MEDLINE, ProQuest databases, and the Cochrane Library and the National Institute of Clinical Excellence were also searched from their date of inception to April 2016, with no language restrictions applied.

REVIEW METHOD

Studies were identified using predetermined inclusion criteria. Data were extracted and summarised and underwent evaluation using published classification of psychometric tests for consistency of interpretation.

RESULTS

Twenty-six studies evaluating five observational pain assessment instruments that had been used with critically ill intubated patients were identified. All five instruments included behavioural indicators, with two including physiologic indicators. All five instruments have undergone validity and reliability testing involving nonverbal critically ill intubated patients, three were examined for feasibility, and one instrument underwent sensitivity and specificity testing. None have been tested within the emergency department with nonverbal critically ill intubated adult patients.

CONCLUSION

The use of an appropriate and valid observational pain assessment instrument is fundamental to detecting and optimising pain management in nonverbal critically ill intubated patients in the emergency department. Of the observational pain assessment instruments reviewed, the Critical-Care Pain Observation Tool was identified as most appropriate for testing in a prospective trial in an emergency department setting.

Validation of the Danish version of the Critical Care Pain Observation Tool

BACKGROUND

Assessing pain in critically ill patients is a challenge even in an intensive care unit (ICU) with a no sedation protocol. The aim of this study was to validate the Danish version of the pain assessment method; Critical Care Pain Observation Tool (CPOT) in an ICU with a no sedation protocol.

METHODS

Seventy patients were included in this study. The patients were observed during a non-nociceptive procedure (wash of an arm) and a nociceptive procedure (turning). Patients were observed before, during, and 15 min after the two interventions (six assessments). Two observers participated in the data collection and CPOT scores were blinded to each other. Calculations of interrater reliability, criterion validity and discriminant validity were performed to validate the Danish version of CPOT.

RESULTS

The results indicated a good correlation between the two raters (all scores > 0.9 and P < 0.05). About 48 (68.6%) of the included patients were able to self-report pain. We found a significantly higher mean CPOT score at the nociceptive procedure than at rest or the non-nociceptive procedure (P < 0.05). No correlation was found between CPOT scores and physiological indicators. Patients self-reported pain and CPOT showed a significant correlation (P < 0.05). A CPOT score of ≥ 3 correlated with patients' self-reported pain (ROC AUC 0.83).

CONCLUSION

The Danish version of CPOT can be used to assess pain in critically ill patients, also when the ICU has a no sedation protocol. CPOT scores showed a good interrater reliability and correlates well with patient's self-reported pain.

Behavioral Pain Scale - translation, reliability, and validity in a Swedish context

BACKGROUND

Assessing pain in the intensive care unit (ICU) is challenging. Due to intubation and sedation, communication can be limited. International guidelines recommend assessing pain with instruments based on behavioral parameters when critically ill patients are unable to self-report their pain level. One of the recommended instruments, the Behavioral Pain Scale (BPS), has shown good validity and reliability in international studies. The aim of this study was to translate and adapt the BPS for critically ill intubated and non-intubated patients in a Swedish ICU context and to assess inter-rater reliability and discriminant validity.

METHODS

The BPS (both for intubated and non-intubated patients) was translated and adapted into Swedish using a translation method consisting of ten steps. The Swedish version was then tested for inter-rater reliability and discriminant validity on 20 critically ill patients (10 intubated and 10 non-intubated) before and directly after a potentially painful procedure (repositioning).

RESULTS

The Swedish version of the BPS showed inter-rater reliability with a percentage agreement of 85% when tested on a sample of critically ill patients. The instrument also showed discriminant validity between assessments at rest and after repositioning.

CONCLUSION

Results of the Swedish version of the BPS support its use in critically ill patients who cannot self-report their pain level. Still, additional studies are needed to further explore its reliability and validity in the Swedish ICU context.