Partnering With Family Members to Detect Delirium in Critically Ill Patients

A Comprehensive Intervention Program for Postoperative Delirium Prevention in Elderly Patients Undergoing Cardiac Surgery

PURPOSE

To explore the application effect of a comprehensive intervention program for postoperative delirium (POD) prevention.

DESIGN

Descriptive and comparative study.

METHODS

This study included 82 elderly patients who underwent cardiac surgery as the control group to receive routine perioperative care. Seventy seven elderly patients who underwent cardiac surgery were selected as the observation group and received a comprehensive intervention program for POD prevention. This study further compared the length of stay in the hospital, nutritional status, pain score, and sleep quality between the two groups after intervention.

FINDINGS

The postoperative length of stay in the hospital of the observation group was shorter than that of the control group (12.60 ± 2.97 vs 17.30 ± 7.10), and the incidence of POD was lower than that of the latter group (24.68% vs 47.56%, X2 = 8.976, P = .003). The pain scores of the observation group were lower than those of the control group (F = 195.381, P < .001). There was significant difference that the sleep quality score of the observation group was higher than that of the control group (F = 219.20, P < .001). The levels of albumin and hemoglobin in the observation group were higher than those in the control group (54.38 ± 5.87 vs 38.40 ± 3.11; 124.64 ± 13.18 vs 109.00 ± 10.20). The observation group had shorter mechanical ventilation duration, intensive care unit stay, and lower total hospital expenses compared with the control group (P < .05).

CONCLUSIONS

The comprehensive intervention program for POD prevention based on evidence-based medicine has a positive effect. In addition, it can also improve patients' pain, sleep quality, and nutritional status.

Optimizing ChatGPT's Interpretation and Reporting of Delirium Assessment Outcomes: Exploratory Study

BACKGROUND

Generative artificial intelligence (AI) and large language models, such as OpenAI's ChatGPT, have shown promising potential in supporting medical education and clinical decision-making, given their vast knowledge base and natural language processing capabilities. As a general purpose AI system, ChatGPT can complete a wide range of tasks, including differential diagnosis without additional training. However, the specific application of ChatGPT in learning and applying a series of specialized, context-specific tasks mimicking the workflow of a human assessor, such as administering a standardized assessment questionnaire, followed by inputting assessment results in a standardized form, and interpretating assessment results strictly following credible, published scoring criteria, have not been thoroughly studied.

OBJECTIVE

This exploratory study aims to evaluate and optimize ChatGPT's capabilities in administering and interpreting the Sour Seven Questionnaire, an informant-based delirium assessment tool. Specifically, the objectives were to train ChatGPT-3.5 and ChatGPT-4 to understand and correctly apply the Sour Seven Questionnaire to clinical vignettes using prompt engineering, assess the performance of these AI models in identifying and scoring delirium symptoms against scores from human experts, and refine and enhance the models' interpretation and reporting accuracy through iterative prompt optimization.

METHODS

We used prompt engineering to train ChatGPT-3.5 and ChatGPT-4 models on the Sour Seven Questionnaire, a tool for assessing delirium through caregiver input. Prompt engineering is a methodology used to enhance the AI's processing of inputs by meticulously structuring the prompts to improve accuracy and consistency in outputs. In this study, prompt engineering involved creating specific, structured commands that guided the AI models in understanding and applying the assessment tool's criteria accurately to clinical vignettes. This approach also included designing prompts to explicitly instruct the AI on how to format its responses, ensuring they were consistent with clinical documentation standards.

RESULTS

Both ChatGPT models demonstrated promising proficiency in applying the Sour Seven Questionnaire to the vignettes, despite initial inconsistencies and errors. Performance notably improved through iterative prompt engineering, enhancing the models' capacity to detect delirium symptoms and assign scores. Prompt optimizations included adjusting the scoring methodology to accept only definitive "Yes" or "No" responses, revising the evaluation prompt to mandate responses in a tabular format, and guiding the models to adhere to the 2 recommended actions specified in the Sour Seven Questionnaire.

CONCLUSIONS

Our findings provide preliminary evidence supporting the potential utility of AI models such as ChatGPT in administering standardized clinical assessment tools. The results highlight the significance of context-specific training and prompt engineering in harnessing the full potential of these AI models for health care applications. Despite the encouraging results, broader generalizability and further validation in real-world settings warrant additional research.

Translation, cultural debugging, and validation of the Chinese version of the Sour Seven Questionnaire: a cross-sectional study

Introduction

Intensive care unit delirium (ICUD) is an acute cerebral dysfunction accompanied by a change of level of consciousness, disorientation, and cognitive dysfunction, typically occurring over a short duration ranging from hours to days and resulting from underlying medical causes. Family members may sometimes detect changes in consciousness earlier than medical staff. The Sour Seven Questionnaire is a tool to assist family members in screening for delirium, but there is currently no Chinese version. This study aimed to translate and cross-culturally debug the Sour Seven Questionnaire and test the effectiveness of the Chinese version in screening for ICUD by family members.

Methods

To create the Chinese version of the questionnaire, the questionnaire was first translated and then culturally debugged through expert consultation and cognitive interviews. Patients and their family members admitted to three ICUs in a Chinese hospital were selected to test the Chinese version of the Sour Seven Questionnaire and the results were compared with those of the validated and recommended Confusion Assessment Method for the intensive care unit (CAM-ICU) assessment.

Results

A total of 190 ICU patients and their families were included in this study. Results of the CAM-ICU assessment showed that 73 (38.4%) patients developed ICUD compared to the 66 (34.7%) using the Chinese version of the Sour Seven Questionnaire, which had a Cohen's kappa coefficient of 0.853, a sensitivity of 0.863, and a specificity of 0.974. The positive predictive value was 0.954 and the negative predictive value was 0.919.

Discussion

The Chinese version of the Sour Seven Questionnaire is a valid assessment tool for helping families screen for ICUD, and it is effective in identifying altered consciousness in patients even during online visits.

An online delirium detection tool: Cross-cultural adaptation of a Chinese version of the Family Confusion Assessment Method

BACKGROUND

Intensive care unit (ICU) delirium is a common complication in older critically ill patients that has a significant impact. The Family Confusion Assessment Method (FAM-CAM) is a vital tool for assisting family members in identifying delirium; however, no study has yet been reported on the Chinese version of the scale.

OBJECTIVES

The objective of this study was to translate the FAM-CAM into a Chinese version and to verify its effectiveness for delirium detection in an online patient visit setting.

METHODS

This was a cross-sectional study. The FAM-CAM was translated to Chinese according to the International Society for Pharmacoeconomics and Outcomes Research guidelines. Patients and family members were recruited to participate in delirium assessments in three ICUs of one hospital. Family members then used the Chinese version of the FAM-CAM to assess for delirium via online visitation, and ICU nurses assessed patients for delirium using the Intensive Care Delirium Screening Checklist (ICDSC). Results were then compared between family members' and nurses' assessments.

RESULTS

Overall, 190 critically ill patients and 190 family members were included, of whom 117 (61.6%) were assessed for delirium using the Intensive Care Delirium Screening Checklist. The Cohen's kappa coefficient between the Intensive Care Delirium Screening Checklist and FAM-CAM was 0.759 (P < 0.01). The sensitivity of the Chinese version of the FAM-CAM was 0.880, specificity was 0.890, positive predictive value was 0.928, negative predictive value was 0.823, and area under the receiver operating characteristic curve was 0.881 (95% confidence interval: 0.872-0.935, P < 0.01).

CONCLUSION

The Chinese version of the FAM-CAM was shown to effectively help families detect delirium and was suggested as a crucial tool for assisting ICU nurses in the early identification of delirium. This tool may effectively be used to assess delirium during online visits.

Early mobility and family partnerships in the intensive care unit: A scoping review of reviews

BACKGROUND

Critical illness significantly impacts the well-being of patients and families. Previous studies show that family members are willing to participate in patient care. Involving families in early mobility interventions may contribute to improved recovery and positive outcomes for patients and families.

AIM

In this scoping review, we investigated early mobility interventions for critically ill patients evaluated in randomized controlled trials and the extent to which family engagement in those interventions are reported in the literature.

STUDY DESIGN

In this scoping review of reviews, EMBASE, CINAHL, PubMed and Cochrane Central databases were searched in October 2019 and updated in February 2022. Systematic reviews were included and assessed using A MeaSurement Tool to Assess Systematic Reviews (AMSTAR) 2. Data were synthesized using a narrative approach. PRISMA-ScR guidelines were adhered to for reporting.

RESULTS

Thirty-three reviews were included which described a range of early mobility interventions for critically ill patients; none explicitly mentioned family engagement. Almost half of the reviews were of low or critically low quality. Insufficient detail of early mobility interventions prompted information to be extracted from the primary studies.

CONCLUSIONS

There are a range of early mobility interventions for critically ill patients but few involve families. Given the positive outcomes of family participation, and family willingness to participate in care, there is a need to explore the feasibility and acceptability of family participation in early mobility interventions.

RELEVANCE TO CLINICAL PRACTICE

Family engagement in early mobility interventions for critically ill patients should be encouraged and supported. How to best support family members and clinicians in enacting family involvement in early mobility requires further investigation.

Patterns of interactions among ICU interprofessional teams: A prospective patient-shift-level survey approach

BACKGROUND

The Awakening, Breathing Coordination, Delirium monitoring and Early mobility bundle (ABCDE) is associated with lower mortality for intensive care unit (ICU) patients. However, efforts to improve ABCDE are variably successful, possibly due to lack of clarity about who are the team members interacting when caring for each patient, each shift. Lack of patient shift-level information regarding who is interacting with whom limits the ability to tailor interventions to the specific ICU team to improve ABCDE.

OBJECTIVE

Determine the number and types of individuals (i.e., clinicians and family members) interacting in the care of mechanically ventilated (MV) patients, as reported by the patients' assigned physician, nurse, and respiratory therapist (RT) each shift, using a network science lens.

METHODS

We conducted a prospective, patient-shift-level survey in 2 medical ICUs. For each patient, we surveyed the assigned physician, nurse, and RT each day and night shift about who they interacted with when providing ABCDE for each patient-shift. We determined the number and types of interactions, reported by physicians, nurses, and RTs and day versus night shift.

RESULTS

From 1558 surveys from 404 clinicians who cared for 169 patients over 166 shifts (65% response rate), clinicians reported interacting with 2.6 individuals each shift (physicians: 2.65, nurses: 3.33, RTs: 1.86); this was fewer on night shift compared to day shift (1.99 versus 3.02). Most frequent interactions were with the bedside nurse, attending, resident, intern, and RT; family member interactions were reported in less than 1 in 5 surveys (12.2% of physician surveys, 19.7% of nurse surveys, 4.9% of RT surveys).

INTERPRETATION

Clinicians reported interacting with 3-4 clinicians each shift, and fewer on nights. Nurses interacted with the most clincians and family members. Interventions targeting shift-level teams, focusing on nurses and family members, may be a way to improve ABCDE delivery and ICU teamwork.

Diagnostic accuracy of the Family Confusion Assessment Method for delirium detection: A systematic review and meta-analysis

BACKGROUND

Delirium is frequently disproportionately under-recognized despite its high prevalence, detrimental impact, and potential lethality. Informant-based delirium detection tools can offer structured assessment and increase the timeliness and frequency of detection. We aimed to examine the diagnostic accuracy of the Family Confusion Assessment Method (FAM-CAM) for delirium detection.

METHODS

We systematically searched the MEDLINE, EMBASE, PsycINFO, CINAHL, CNKI, WANFANG, and SinoMed databases from January 1988 to December 2022. Two reviewers independently screened studies and evaluated methodological quality using the revised quality assessment of diagnostic accuracy studies (QUADAS-2) tool. A bivariate random effects model was undertaken, and univariable meta-regression was carried out to explore heterogeneity.

RESULTS

Seven studies with 483 dyads of participants and family caregivers were identified. Pooled sensitivity and specificity were 0.74 (95% CI: 0.59, 0.86) and 0.91 (95% CI: 0.83, 0.95), respectively, with an area under curve (AUC) of 0.91. The positive likelihood ratio was 8.27 (95% CI: 3.97, 17.25), and the negative likelihood ratio was 0.28 (95% CI: 0.16, 0.50). Settings impacted specificity (p = 0.02).

CONCLUSIONS

Available evidence indicates that FAM-CAM exhibits moderate sensitivity and high specificity for delirium screening in adults. The FAM-CAM is concise and easy to use, making it appropriate for routine clinical practice, which might benefit early delirium detection and potentially foster delirium management.

PROSPERO REGISTRATION NUMBER

CRD42022378742.

Facilitators and barriers to evidence-based practice in central venous access device insertion and management in an intensive care unit: A qualitative study

INTRODUCTION

Research evidence and clinical practice guidelines exist on preventing central line-associated bloodstream infections. However, there is limited knowledge about the barriers and facilitators to evidence-based central venous access device care.

AIM

The aim of this study was to investigate the facilitators and barriers to evidence-based central venous access device care in the adult intensive care setting.

METHOD

This exploratory qualitative study involved focus groups and interviews with registered nurses and physicians involved in central venous access device insertion and management in a tertiary Australian intensive care unit. Purposive sampling was used to recruit staff (n = 26) with varying years of clinical experience and clinical positions. Six focus groups and three individual interviews were conducted. Interviews were audio recorded, transcribed verbatim and analysed using content analysis.

RESULTS

Three overarching categories emerged: work structures to support optimal performance; processes to optimise quality of care, and factors influencing staff members' behaviour. Perceived facilitators to optimal central venous access device care included explicit language use in procedure documents, work-system integrated strategies, research evidence dissemination, audit, and feedback. However, there was a lack of consistency in practices such as audit, feedback, and patient participation.

CONCLUSION

To bring about effective improvement in central venous access device care, future interventions should be tailored to address identified barriers, including integrating audit and feedback into clinicians' work processes. Additionally, future research is needed to explore the role of patients and their families in central venous access device care.

IMPLICATIONS FOR CLINICAL PRACTICE

When developing practice policies or procedure manuals, it is important to use explicit language to ensure clear communication of evidence-based recommendations to clinicians. Strategies integrated into work processes can enhance adherence to evidence-based practice. Large departments with limited educators should explore innovative methods like online education to ensure optimal central venous access device care.

Preliminary clinical testing to inform development of the Critical Care Pain Observation Tool for Families (CPOT-Fam)

Introduction

Many patients in the intensive care unit (ICU) cannot communicate. For these patients, family caregivers (family members/close friends) could assist in pain assessment. We previously adapted the Critical Care Pain Observation Tool (CPOT) for family caregiver use (CPOT-Fam). In this study, we conducted preliminary clinical evaluation of the CPOT-Fam to inform further tool development.

Methods

For preliminary testing, we collected (1) pain assessments of patients in the ICU from family caregivers (CPOT-Fam) and nurses (CPOT) and determined the degree of agreement (kappa coefficient, κ) and (2) collected openended feedback on the CPOT-Fam from family caregivers. For refinement, we used preliminary testing data to refine the CPOT-Fam with a multidisciplinary working group.

Results

We assessed agreement between family caregiver and nurse pain scores for 29 patients. Binary agreement (κ) between CPOT-Fam and CPOT item scores (scores ≥2 considered indicative of significant pain) was fair, κ = 0.43 (95% confidence interval [CI] 0.18-0.69). Agreement was highest for the CPOT-Fam items ventilator compliance/vocalization (weighted κ = 0.48, 95% CI 0.15-0.80) and lowest for muscle tension (weighted κ = 0.10, 95% [CI] -0.17 to 0.20). Most participants (n = 19; 69.0%) reported a very positive experience using the CPOT-Fam, describing it as "good" and "easy-to-use/clear/straightforward." We iteratively refined the CPOT-Fam over five cycles using the data collected until no further revisions were suggested.

Conclusion

Our preliminary clinical testing suggests that family involvement in pain assessment in the ICU is well perceived. The CPOT-Fam has been further refined and is now ready for clinical pilot testing to determine its feasibility and acceptability.

A randomised controlled trial of a nutrition and a decision support intervention to enable partnerships with families of critically ill patients

AIMS AND OBJECTIVES

The aim of the study was to investigate the effect of supporting family members to partner with health professionals on nutrition intakes and decision-making and to evaluate intervention and study feasibility.

BACKGROUND

Family partnerships can improve outcomes for critically ill patients and family members. Interventions that support families to engage with health professionals require evaluation.

DESIGN

A multi-centre, randomised, parallel group superiority Phase II randomised controlled trial.

METHODS

In nine intensive care units (ICUs) across three countries, critically ill patients ≥60 years, or those 55-59 years with advanced chronic diseases and expected ICU length of stay >72 h and their family member were enrolled between 9 May 2017 and 31 March 2020. Participants were randomised (1:1:1) to either a decision support or nutrition optimisation family-centred intervention, or usual care. Primary outcomes included protein and energy intake during ICU and hospital stay (nutrition intervention) and family satisfaction (decision support). Study feasibility was assessed as a composite of consent rate, intervention adherence, contamination and physician awareness of intervention assignment.

RESULTS

We randomised 135 patients/family members (consent rate 51.7%). The average rate of randomisation was 0.5 (0.13-1.53) per month. Unavailability (staff/family) was the major contributor to families not being approached for consent. Declined consent was attributed to families feeling overwhelmed (58/126, 46%). Pandemic visitor restrictions contributed to early study cessation. Intervention adherence for the decision support intervention was 76.9%-100.0% and for the nutrition intervention was 44.8%-100.0%. Nutritional adequacy, decisional conflict, satisfaction with decision-making and overall family satisfaction with ICU were similar for all groups.

CONCLUSIONS

Active partnerships between family members and health professionals are important but can be challenging to achieve in critical care contexts. We were unable to demonstrate the efficacy of either intervention. Feasibility outcomes suggest further refinement of interventions and study protocol may be warranted.

RELEVANCE TO CLINICAL PRACTICE

Interventions to promote family partnerships in critical illness are needed but require a greater understanding of the extent to which families want and are able to engage and the activities in which they have most impact.

REPORTING METHOD

This study has been reported following the Consolidated Standards of Reporting Trials (CONSORT) and the Template for Intervention Description and Replication (TIDieR) guidelines.

PATIENT OR PUBLIC CONTRIBUTION

Patients and caregivers were engaged in and contributed to the development and subsequent iterations of the two family-centred interventions use in this study.

CLINICAL TRIAL REGISTRATION NUMBER

Trial registration.

CLINICALTRIALS

gov, ID: NCT02920086. Registered on 30 September 2016. First patient enrolled on 9 May 2017 https://clinicaltrials.gov/ct2/results?cond=&term=NCT02920086&cntry=&state=&city=&dist=.

Patient Safety: Cognitive Assessment at Intensive Care Unit Discharge

Topic

Assessing functional cognition is a critical need for intensive care unit survivors transitioning to another level of care.

Clinical Relevance

Up to 62% of patients discharged from an intensive care unit have significant cognitive impairment that is not associated with severity of illness, number of comorbidities, or length of hospital stay. For more than 20 years, researchers have published an array of potentially effective interventions, including case management, patient and informal caregiver education, and home health care services.

Purpose

To describe the impact of and potential interventions for cognitive decline at intensive care unit discharge and discharge readiness on 30-day readmission rates.

Content Covered

Assessing the patient’s functional cognition assessment and advocating for appropriate resources are needed to improve patient and clinical outcomes.

Detection of delirium by family members in the intensive care unit: Translation, Cross-Cultural adaptation and validation of the Family Confusion Assessment Method for the German-Speaking area

AIM

The aim of this study was the translation, cross-cultural adaptation and validation of the Family Confusion Assessment Method in critically ill patients.

BACKGROUND

Delirium is a frequently unrecognized disorder in critically ill patients. Visiting family members might be the first to notice subtle changes in a patient's cognition and behaviour. The Family Confusion Assessment Method was developed to detect delirium by family members, but has not been available for the German-speaking area yet.

DESIGN

A prospective validation study was conducted between January 2020 and October 2020.

METHODS

The Family Confusion Assessment Method was translated into German according to the Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes. Subsequently, we compared the Family Confusion Assessment Method with the Confusion Assessment Method for the Intensive Care Unit in critically ill patients and their family members in a medical intensive care unit in Germany.

RESULTS

We included 50 dyads of critically ill patients and their family members. The prevalence of delirium measured by Confusion Assessment Method for the Intensive Care Unit was 44%. Cohen's kappa coefficient was 0.84. The German Family Confusion Assessment Method had a high sensitivity of 95.5% and specificity of 89.3%. The positive predictive value and negative predictive value were 87.5% and 96.2% respectively.

CONCLUSIONS

These findings suggest that the German Family Confusion Assessment Method is an accurate assessment tool for delirium detection in the intensive care unit by family members. Furthermore, the results indicate that family members may identify delirium by the Family Confusion Assessment Method without prior training.

IMPACT

Collaborating medical staff with patients' family members to detect delirium in the intensive care unit may lead to early recognition of delirium.

The impact of patient delirium in the intensive care unit: patterns of anxiety symptoms in family caregivers

BACKGROUND

The purpose of this study was to examine the association of patient delirium in the intensive care unit (ICU) with patterns of anxiety symptoms in family caregivers when delirium was determined by clinical assessment and family-administered delirium detection.

METHODS

In this cross-sectional study, consecutive adult patients anticipated to remain in the ICU for longer than 24 h were eligible for participation given at least one present family caregiver (e.g., spouse, friend) provided informed consent (to be enrolled as a dyad) and were eligible for delirium detection (i.e., Richmond Agitation-Sedation Scale score ≥ - 3). Generalized Anxiety Disorder-7 (GAD-7) was used to assess self-reported symptoms of anxiety. Clinical assessment (Confusion Assessment Method for ICU, CAM-ICU) and family-administered delirium detection (Sour Seven) were completed once daily for up to five days.

RESULTS

We included 147 family caregivers; the mean age was 54.3 years (standard deviation [SD] 14.3 years) and 74% (n = 129) were female. Fifty (34% [95% confidence interval [CI] 26.4-42.2]) caregivers experienced clinically significant symptoms of anxiety (median GAD-7 score 16.0 [interquartile range 6]). The most prevalent symptoms of anxiety were "Feeling nervous, anxious or on edge" (96.0% [95%CI 85.2-99.0]); "Not being able to stop or control worrying" (88.0% [95%CI 75.6-94.5]; "Worrying too much about different things" and "Feeling afraid as if something awful might happen" (84.0% [95%CI 71.0-91.8], for both). Family caregivers of critically ill adults with delirium were significantly more likely to report "Worrying too much about different things" more than half of the time (CAM-ICU, Odds Ratio [OR] 2.27 [95%CI 1.04-4.91]; Sour Seven, OR 2.28 [95%CI 1.00-5.23]).

CONCLUSIONS

Family caregivers of critically ill adults with delirium frequently experience clinically significant anxiety and are significantly more likely to report frequently worrying too much about different things. Future work is needed to develop mental health interventions for the diversity of anxiety symptoms experienced by family members of critically ill patients.

TRIAL REGISTRATION

This study is registered on ClinicalTrials.gov ( https://clinicaltrials.gov/ct2/show/NCT03379129 ).

Incorporating and evaluating citizen engagement in health research: a scoping review protocol

BACKGROUND

Citizen engagement in research is an emerging practice that involves members of the general public in research processes such as priority setting, planning, decision-making, research conduct, implementation, evaluation, and dissemination. Engaging citizens in research, particularly health research, increases the relevance of study findings, minimizes waste by facilitating stewardship over resources, and builds public trust in the research. While several existing frameworks guide the application of citizen engagement principles to health research, it is unclear how citizen engagement can be utilized to maximize benefits and minimize risks and challenges in health research. To address the gaps in knowledge around citizen engagement in health research, we propose a scoping review to synthesize the state of knowledge on methods to incorporate and evaluate citizen engagement in research. A protocol is presented in this manuscript.

METHODS

The methodology for our scoping review is guided by Arksey and O' Malley's framework for scoping reviews, and additional recommendations by Levac and colleagues. We will include peer-reviewed and gray literature that report on citizen engagement in health research (including biomedical, clinical, health systems and services, and social, cultural, environmental and population health) and report method(s) to conduct, measure, or evaluate citizen engagement. We will systematically search electronic databases (MEDLINE, EMBASE, CINAHL, JSTOR, PsycINFO, Scopus, and Science Direct) from inception onwards and search relevant organizations' websites for additional studies, frameworks, and reports on citizen engagement. Title and abstract and full-text citations will be screened independently and in duplicate. Data will be extracted independently and in duplicate, including document characteristics, citizen engagement definitions and goals, and outcomes of citizen engagement (e.g., barriers, facilitators).

DISCUSSION

This review will synthesize the definitions, goals, methods, outcomes, and significance of citizen engagement in health research, as well as any potential barriers, facilitators, and challenges outlined in existing literature. The findings will provide an evidence-based foundation for developing new or improved guidance for citizen engagement in health research. Overall, we anticipate that our scoping review will be a preliminary step to meaningful engagement of citizens in research and strengthen the relationship between the scientific community and the public through transparency and collaboration.

SYSTEMATIC REVIEW REGISTRATION

Open Science Framework  https://osf.io/hzcbr .

Natural language processing to measure the frequency and mode of communication between healthcare professionals and family members of critically ill patients

OBJECTIVE

To apply natural language processing (NLP) techniques to identify individual events and modes of communication between healthcare professionals and families of critically ill patients from electronic medical records (EMR).

MATERIALS AND METHODS

Retrospective cohort study of 280 randomly selected adult patients admitted to 1 of 15 intensive care units (ICU) in Alberta, Canada from June 19, 2012 to June 11, 2018. Individual events and modes of communication were independently abstracted using NLP and manual chart review (reference standard). Preprocessing techniques and 2 NLP approaches (rule-based and machine learning) were evaluated using sensitivity, specificity, and area under the receiver operating characteristic curves (AUROC).

RESULTS

Over 2700 combinations of NLP methods and hyperparameters were evaluated for each mode of communication using a holdout subset. The rule-based approach had the highest AUROC in 65 datasets compared to the machine learning approach in 21 datasets. Both approaches had similar performance in 17 datasets. The rule-based AUROC for the grouped categories of patient documented to have family or friends (0.972, 95% CI 0.934-1.000), visit by family/friend (0.882 95% CI 0.820-0.943) and phone call with family/friend (0.975, 95% CI: 0.952-0.998) were high.

DISCUSSION

We report an automated method to quantify communication between healthcare professionals and family members of adult patients from free-text EMRs. A rule-based NLP approach had better overall operating characteristics than a machine learning approach.

CONCLUSION

NLP can automatically and accurately measure frequency and mode of documented family visitation and communication from unstructured free-text EMRs, to support patient- and family-centered care initiatives.

Associations between caregiver-detected delirium and symptoms of depression and anxiety in family caregivers of critically ill patients: a cross-sectional study

BACKGROUND

Witnessing delirium can be distressing for family caregivers (i.e., relatives or friends) of critically ill patients. This study aimed to evaluate associations between caregiver-detected delirium in critically ill patients and depression and anxiety symptoms in their family caregivers.

METHODS

Consecutive adult patient-caregiver dyads were enrolled from a 28-bed medical-surgical intensive care unit. Patient delirium was screened for daily by family caregivers using the Sour Seven instrument. Family caregivers completed the Patient Health Questionnaire-9 (PHQ-9) and General Anxiety Disorder-7 (GAD-7) instruments daily to assess their own depression and anxiety symptoms. Response feature analysis was used to handle repeated measures. Descriptive statistics and regression analyses were completed.

RESULTS

One hundred forty-seven patient-caregiver dyads were enrolled. Clinically significant symptoms of depression and anxiety occurred in 27% and 35% of family caregivers, respectively. Caregiver-detected delirium occurred in 65% of patients, and was not associated with clinically significant caregiver depression (Odds Ratio [OR] 1.4, 95% Confidence Interval [95%CI] 0.6-3.1) or anxiety (OR 1.2, 95%CI 0.6-2.6) symptoms. When stratified by Sour Seven scores, scores 1-3 and 4-9 were associated with increased symptoms of anxiety (OR 3.1, 95%CI 1.3-7.0) and depression (OR 2.6, 95%CI 1.1-6.1) in family caregivers. Caregiver-detected delirium score was associated with severity of family caregiver anxiety symptoms (coefficient 0.2, 95%CI 0.1-0.4), but not depression symptoms (coefficient 0.2, 95%CI -0.0-0.3).

CONCLUSIONS

Caregiver-detected patient delirium was associated with increased depression and anxiety symptoms in family caregivers of critically ill patients. Further randomized research is required to confirm these associations.

Patients, family members and providers perceive family-administered delirium detection tools in the adult ICU as feasible and of value to patient care and family member coping: a qualitative focus group study

PURPOSE

While studies report on perceptions of family participation in delirium prevention, little is known about the use of family-administered delirium detection tools in the care of critically ill patients. This study sought the perspectives of patients, their family members, and healthcare providers on the use of family-administered delirium detection tools to detect delirium in critically ill patients and barriers and facilitators to using family-administered delirium detection tools in patient care.

METHODS

In this qualitative study, critical care providers (five physicians, six registered nurses) and participants from the Family ICU Delirium Detection Study (seven past patients and family members) took part in four focus groups at one hospital in Calgary, Alberta.

RESULTS

Key themes identified following thematic analysis from 18 participants included: 1) perceptions of acceptability of family-administered delirium detection (e.g., family feels valued, intensive care unit (ICU) care team may not use a family member's results, intensification of work load), 2) considerations regarding feasibility (e.g., insufficient knowledge, healthcare team buy-in), and 3) overarching strategies to support implementation into routine patient care (e.g., value of family-administered delirium detection for patients and families is well understood in the clinical context, regular communication between the family and ICU providers, an electronic version of the tool).

CONCLUSIONS

Patients, family members and healthcare providers who participated in the focus groups perceived family participation in delirium detection and the use of family-administered delirium detection tools at the bedside as feasible and of value to patient care and family member coping.

TRIAL REGISTRATION

www.ClinicalTrials.gov (NCT03379129); registered 15 December 2017.

Adaptation and Validation of a Chart-Based Delirium Detection Tool for the ICU (CHART-DEL-ICU)

OBJECTIVE

To adapt and validate a chart-based delirium detection tool for use in critically ill adults.

DESIGN

Validation study.

SETTING

Medical-surgical intensive care unit (ICU) in an academic hospital.

MEASUREMENTS

A chart-based delirium detection tool (CHART-DEL) was adapted for use in critically ill adults (CHART-DEL-ICU) and compared with prospective delirium assessments (i.e., clinical assessments (reference standard) by a research nurse trained by a neuropsychiatrist and routine delirium screening tools Confusion Assessment Method (CAM-ICU)) and (Intensive Care Delirium Screening Checklist (ICDSC)). The original CHART-DEL tool was adapted to include physician-reported ICDSC score (for probable delirium) and Richmond-Agitation Sedation Scale score (for altered level of consciousness and agitation). Two trained chart abstractors blinded to all delirium assessments manually abstracted delirium-related information from medical charts and electronic medical records and rated if delirium was present (four levels: uncertain, possible, probable, definite) or absent (no evidence).

RESULTS

Charts were manually abstracted for delirium-related information for 213 patients who were included in a prospective cohort study that included prospective delirium assessments. The CHART-DEL-ICU tool had excellent interrater reliability (kappa = 0.90). Compared to the reference standard, the sensitivity was 66.0% (95% CI = 59.3-72.3%) and specificity was 82.1% (95% CI = 78.0-85.7%), with a cut-point that included definite, probable, possible, and uncertain delirium. The AUC of the CHART-DEL-ICU alone is 74.1% (95% CI = 70.4-77.8%) compared with the addition of the CAM-ICU and ICDSC (CAM-ICU/CHART-DEL-ICU: 80.9% (95% CI = 77.8-83.9%), P = .01; ICDSC/CHART-DEL-ICU: 79.2% (95% CI = 75.9-82.6%), P = .03).

CONCLUSION

A chart-based delirium detection tool has improved diagnostic accuracy when combined with routine delirium screening tools (CAM-ICU and ICDSC), compared to a chart-based method on its own. This presents a potential for retrospective detection of delirium from medical charts for research or to augment routine delirium screening methods to find missed cases of delirium.

Association between delirium in the intensive care unit and subsequent neuropsychiatric disorders

Background

Patients in the intensive care unit (ICU) are known to be at increased risk of developing delirium, but the risk of subsequent neuropsychiatric disorders is unclear. We therefore sought to examine the association between the presence of delirium in the ICU and incident neuropsychiatric disorders (including depressive, anxiety, trauma-and-stressor-related, and neurocognitive disorders) post-ICU stay among adult medical-surgical ICU patients.

Methods

Retrospective cohort study utilizing clinical and administrative data from both inpatient and outpatient healthcare visits to identify the ICU cohort and diagnostic information 5 years prior to and 1 year post-ICU stay. Patients ≥ 18 years of age admitted to one of 14 medical-surgical ICUs across Alberta, Canada, January 1, 2014–June 30, 2016, and survived to hospital discharge were included. The main outcome of interest was a new diagnosis of any neuropsychiatric disorder 1 year post-ICU stay. The exposure variable was delirium during the ICU stay identified through any positive delirium screen by the Intensive Care Unit Delirium Screening Checklist (ICDSC) during the ICU stay.

Results

Of 16,005 unique patients with at least one ICU admission, 4033 patients were included in the study of which 1792 (44%) experienced delirium during their ICU stay. The overall cumulative incidence of any neuropsychiatric disorder during the subsequent year was 19.7% for ICU patients. After adjusting for hospital characteristics using log-binomial regression, patients with delirium during the ICU stay had a risk ratio (RR) of 1.14 (95% confidence interval [CI] 0.98–1.33) of developing any neuropsychiatric disorder within 1 year post-ICU compared to those who did not experience delirium. Delirium was significantly associated with neurocognitive disorders (RR 1.59, 95% CI 1.08–2.35), but not depressive disorders (RR 1.16, 95% CI 0.92–1.45), anxiety (RR 1.16, 95% CI 0.92–1.47), and trauma-and-stressor-related (RR 0.82, 95% CI 0.53–1.28) disorders.

Conclusions

The diagnosis of new onset of neurocognitive disorders is associated with ICU-acquired delirium. In this study, significant associations were not observed for depressive, anxiety, and trauma-and-stressor-related disorders.

A study protocol for a randomized controlled trial of family-partnered delirium prevention, detection, and management in critically ill adults: the ACTIVATE study

Background

Delirium is very common in critically ill patients admitted to the intensive care unit (ICU) and results in negative long-term outcomes. Family members are also at risk of long-term complications, including depression and anxiety. Family members are frequently at the bedside and want to be engaged; they know the patient best and may notice subtle changes prior to the care team. By engaging family members in delirium care, we may be able to improve both patient and family outcomes by identifying delirium sooner and capacitating family members in care.

Methods

The primary aim of this study is to determine the effect of family-administered delirium prevention, detection, and management in critically ill patients on family member symptoms of depression and anxiety, compared to usual care. One-hundred and ninety-eight patient-family dyads will be recruited from four medical-surgical ICUs in Calgary, Canada. Dyads will be randomized 1:1 to the intervention or control group. The intervention consists of family-partnered delirium prevention, detection, and management, while the control group will receive usual care. Delirium, depression, and anxiety will be measured using validated tools, and participants will be followed for 1- and 3-months post-ICU discharge. All analyses will be intention-to-treat and adjusted for pre-identified covariates. Ethical approval has been granted by the University of Calgary Conjoint Health Research Ethics Board (REB19–1000) and the trial registered. The protocol adheres to the Standard Protocol Items: Recommendations for Interventional Trials (SPIRIT) checklist.

Discussion

Critically ill patients are frequently unable to participate in their own care, and partnering with their family members is particularly important for improving experiences and outcomes of care for both patients and families.

Trial registration

Registered September 23, 2019 on Clinicaltrials.gov NCT04099472.