One challenge in the care of critically ill patients is delirium, an acute confusional state characterized by an acute onset or fluctuating course, attention deficits, and disorganized thinking.1 Delirium affects nearly half of patients in the intensive care unit (ICU)2,3,4 and is associated with adverse outcomes in critically ill patients (longer ICU/hospital stay, increased mortality, and cognitive impairment after discharge)5 and in family members who witness symptoms of delirium (emotional distress and feelings of anxiety and helplessness).6,7,8 Previous studies report that family participation in patient care is acceptable to patients and their family members, and that families report fewer symptoms of anxiety and depression and increased satisfaction with care following participation.9,10,11 Published guidelines for family-centred care include the provision of family support through family education programs, which show beneficial effects for ICU family members by reducing anxiety, depression, and post-traumatic stress disorder.12 Herein lies an opportunity for families to participate in delirium prevention, detection, and management. Although previous research has shown that family members of critically ill patients want to assist with nonpharmacological delirium prevention activities,13,14 most families lack adequate delirium knowledge to be effective partners.

Family members of critically ill patients are knowledgeable about the pre-ICU mentation, routines, preferences, and signs of discomfort of their loved ones. As such, a family member educated about ICU delirium may be a useful resource for detecting delirium symptoms and preventing and managing delirium using nonpharmacological strategies, preparing them to partner with the ICU care team in delirium-centred patient care. Family education through leaflets or informational videos has been identified as a strategy to reduce registered nurse (RN)-led education regarding patient care activities, thus minimizing additional burden that would otherwise be placed on nursing staff.15,16,17 The objective of this study was to evaluate the effectiveness of an ICU Family Education on Delirium (iFAM-ED) intervention that prepares family members to partner with the ICU care team to detect delirium symptoms and prevent and manage delirium using nonpharmacological strategies.

Methods

The study was approved by the Conjoint Health Research Ethics Board at the University of Calgary (reference number: REB18-0331) in April 2018. Written informed consent was acquired from each family participant. Using a pre-test post-test quasi-experimental study design, the study was conducted between 14 January and 31 October 2019 using a convenience sample of family members of critically ill patients admitted to the 28-bed general systems adult ICU at Foothills Medical Centre in Calgary, Canada (catchment population 1.8 million). Family members eligible for inclusion were adults (≥ 18 yr) who could provide informed consent and understand English. More than one family member per patient could participate.

A research assistant approached eligible family members to participate using a script that included the general requirements of informed consent and a standardized description of delirium. Baseline questionnaires were completed after enrollment and in the presence of the study team. Baseline questionnaires included paper versions of a demographic questionnaire (relationship to patient, date of birth, sex, gender, ethnicity, and educational background), the Generalized Anxiety Disorder-seven (GAD-7)18 scale, and the previously validated Caregiver ICU Delirium Knowledge Questionnaire (CIDKQ)19 to quantify their baseline knowledge of delirium.

Intervention

The ICU Family Education on Delirium (iFAM-ED) intervention was developed based on a study that educated family members to recognize delirium symptoms in older adults20 and adapted for critically ill adults using relevant ICU delirium5,21,22,23 and family delirium education literature.5,20,21,24,25,26 A multidisciplinary team including patient partners (i.e., past ICU patients and family members who are part of our research team), delirium researchers, physicians, and members of a delirium working group (RNs, pharmacist, allied health) iteratively reviewed the iFAM-ED intervention. The intervention included two parts: 1) a six-minute iFAM-ED video module (script found in Appendix 1) and 2) previously validated case vignettes of hypothetical ICU patients (Appendix 2). The iFAM-ED video module included a definition of delirium, a description of the possible symptoms of delirium, delirium risk factors, symptoms that distinguish delirium from dementia, nonpharmacological treatments to prevent and manage delirium, and how to communicate delirium symptoms to the ICU care team. The iFAM-ED video module was presented at a grade six reading level (Flesch-Kincaid grade level). It was assessed for understandability independently by two research assistants (unfamiliar with the study) using the Patient Education Materials Assessment tool for audiovisual materials (PEMAT-A/V).27 The iFAM-ED video module met the criteria for understandability and actionability with scores of 85% and 100%, respectively (wherein a score of ≥ 70% is considered understandable or actionable).28 The iFAM-ED case vignettes selected for each participating family member were case-matched for the patient’s age category, sex, and reason for admission and randomly selected for which features of delirium were present/absent for a total of 80 unique case vignettes (Table 1). After watching the iFAM-ED video module, family members practiced detecting delirium in the provided case vignettes using two family-administered delirium detection questionnaires: the Family Confusion Assessment Method (FAM-CAM)29 and Sour Seven.30 Each case vignette was evaluated for the four features of delirium: inattention (feature 1), sudden onset or fluctuating course (feature 2), altered level of consciousness (feature 3), and disorganized thinking (feature 4). If features 1, 2 and either 3 or 4 were present, the hypothetical patient had delirium.1,31,32

Table 1 Case vignettes of hypothetical ICU patients
Full size table

Family members were given a choice to complete the iFAM-ED module at their convenience using their own devices, or while at the bedside using a study tablet or following along with a printed booklet while a research assistant read the video script. Directly following the iFAM-ED intervention, the family member completed the CIDKQ again to quantify any immediate change in delirium knowledge. To quantify the short-term retention of the knowledge, family members completed an online version of the CIDKQ (generated using Qualtrics survey software, hosted by the University of Calgary [Qualtrics, Provo, UT]) two weeks following provision of the iFAM-ED intervention (Fig. 1). We used evidence-based cohort retention follow-up protocols to reduce attrition (https://www.improvelto.com/).33,34

Fig. 1
figure 1

Schematic of the ICU Family Education on Delirium (iFAM-ED) intervention. ICU = intensive care unit

Full size image

Measures

Caregiver ICU Delirium knowledge questionnaire (CIDKQ)

A 21-item multiple choice (yes/no/don’t know) questionnaire was used to measure family members’ knowledge about ICU delirium and to assess the effect of the education provided. The CIDKQ addressed the three dimensions of delirium knowledge: risk factors (items 1–10), actions (items 11–16) and symptoms (items 17–21). The CIDKQ score ranges from 0 to 21, with a higher score indicating more ICU delirium knowledge. The internal consistency reliability (Cronbach’s alpha) for the CIDKQ is 0.79.19

Family confusion assesement method (FAM-CAM)

This is an 11-item delirium assessment tool derived from the Confusion Assessment Method1 and designed to be administered by family members.29 The FAM-CAM is considered positive if the following features are present: acute onset or fluctuating course (questions 1, 9, 10), inattention (question 2), and either disorganized thinking (questions 3,5,6) or altered level of consciousness (question 4). The FAM-CAM was shown to have a high sensitivity (88%), specificity (98%), and reliability (k = 0.85) in general hospital settings.29

Sour seven

This is a 7-item delirium assessment tool designed to be administered by informal caregivers or untrained nurses.30 Sudden disturbances in level of awareness or attentiveness, fluctuations in awareness and attentiveness, and disordered thinking may indicate delirium. Scores ≥ 9 out of 18 are indicative of delirium and have 100% positive predictive value in hospitalized seniors.30

Generalized anxiety disorder-seven (GAD-7)

This is a 7-item scale following the DSM-IV criteria for GAD.18 The items are scored from 0 (not at all) to 3 (nearly every day) based on the frequency of symptoms, with higher scores indicating more anxiety. A score ≥ 10 out of 21 is indicative of clinically significant GAD; scores of 5, 10, and 15 represent mild, moderate, and severe levels of anxiety, respectively.

Statistical analysis

Sample size was calculated using pre-test post-test effect size and variance data from a study educating family members of older adults to recognize delirium.20 Assuming a 95% confidence interval (Zα = 1.96), 80% power (Zβ = 1.96), standard deviation (SD; σ = 8.5) and mean difference \(\left( {\bar{x}_{post} - \bar{x}_{pre} = 2.5} \right),\)20 a minimum sample size of 21 family members was required.

$$Sample\, size = \frac{{2 \times \left( {Z_{\alpha } + Z_{\beta } } \right)^{2} \times \sigma^{2} }}{{\left( {\bar{x}_{post} - \bar{x}_{pre} } \right)^{2} }}$$

Data analysis was conducted using Stata version 14 (StataCorp., College Station, TX, USA). Descriptive statistics were examined for all study variables as mean, median, or number and percentage, where appropriate. Paired t tests (pre-test/post-test, pre-test/two weeks post-test) were used to examine the difference in delirium knowledge (i.e., CIDKQ score) before and after the iFAM-ED module. A P < 0.05 was considered statistically significant. Kappa was calculated as a measure of agreement between the family-administered delirium detection using the FAM-CAM or Sour Seven and the true delirium status in each case vignette (the reference standard), wherein agreement was interpreted as fair (0.21–0.40), moderate (0.41–0.60), substantial (0.61–0.80), and almost perfect (0.81–1.00).35 We analyzed GAD-7 and CIDKQ scores as continuous (GAD-7: 0–15, CIDKQ: 0–21) and categorical (GAD-7: < 10 or ≥ 10 [clinically significant GAD]; CIDKQ: < mean or ≥ mean) variables. We used Pearson correlation coefficients to evaluate the correlation between GAD-7 and CIDKQ as continuous variables, wherein a Pearson r value of 0.10, 0.30, or 0.50 was interpreted as small, medium, and large effect sizes, respectively.36

Results

Between 14 January and 31 October 2019, 99 family members were recruited, with a participation rate of 59% (99/168) (Fig. 2). Family members had a mean (SD) age of 50.8 (14.5) yr, were mostly female (43/63, 68%), and were most commonly the spouse (24/63, 38%) (e.g., husband, wife) or adult children (22/63, 35%) of the patient (Table 2). Most family members experienced mild (38%, 24/63) or clinically significant (30%, 19/63) symptoms of anxiety. There was no correlation between symptoms of anxiety and delirium knowledge when analyzed as binary variables (P = 0.872) or continuous variables (Pearson correlation coefficient < 0.10).

Fig. 2
figure 2

Participant flow

Full size image
Table 2 Characteristics of family members who completed the full study
Full size table

Of the 99 enrolled family members, 81 completed the iFAM-ED intervention (81/99, 82%) and 63 completed all questionnaires at two-week follow-up (63/81, 78%). Most family members (73/99, 74%) used the study tablet to watch the iFAM-ED video module. Of the 24 family members who chose to watch the iFAM-ED video module on their own device, 37% (9/24) completed the iFAM-ED intervention. Completion of the iFAM-ED intervention and associated questionnaires took about 20 min and no families reported being overwhelmed by their participation in the study.

Family members’ ICU delirium knowledge improved significantly (P < 0.001) following provision of the iFAM-ED intervention (pre-CIDKQ, 13.8; 95% CI, 13.0 to 14.6; post-CIDKQ, 16.7; 95% CI, 16.3 to 17.2; P < 0.001) and was retained two weeks after the intervention (CIDKQ two weeks, 16.4; 95% CI, 15.8 to 17.1; P < 0.001) (Table 3). Family members’ knowledge of the three dimensions of ICU delirium (risk, actions, and symptoms) also improved significantly (P < 0.001) (Table 3). This included improvement in knowledge of delirium risk factors, actions to prevent/manage delirium, and the ability to distinguish between symptoms of dementia and delirium (Table 4).

Table 3 Delirium knowledge before, immediately after, and two weeks after the iFAM-ED module (n = 63)
Full size table
Table 4 Items of the Caregiver ICU Delirium Knowledge Questionnaire (CIDKQ) correctly answered before, immediately after, and two weeks after the iFAM-ED module (n = 63)
Full size table

After watching the iFAM-ED video module, family members correctly classified delirium in 78% (95% CI, 69 to 85) and 92% (95% CI, 86 to 96) of the provided case vignettes using the FAM-CAM or Sour Seven, respectively. The agreement between the family member’s detection of delirium in the case vignette was moderate using the FAM-CAM (kappa = 0.44) and substantial using the Sour Seven (kappa = 0.75). The features of delirium presented in the case vignettes that family members frequently missed were altered level of consciousness (“He seems more tired and falls asleep while you are talking with him”) and disorganized thinking (“He does not know that today is Sunday or that he is in the hospital”) (Table 5).

Table 5 Proportion of family members who correctly identified the features of delirium in the provided case vignettes (n = 63)
Full size table

Discussion

The current study showed that provision of the iFAM-ED intervention effectively improves family members’ ICU delirium knowledge. More participants recognized ICU delirium risk factors (new medication, mechanical ventilation, change in surroundings), actions (patient orientation, sleep/wake cycle), and symptoms (sudden confusion). This included 78% and 92% of participants correctly classifying delirium in the provided case vignettes using the FAM-CAM and Sour Seven, respectively. Furthermore, family members showed short-term retention of their ICU delirium knowledge.

Several studies have reported family member delirium education in cardiac surgery,37 hospitalized older adults,24,26 or palliative care38,39,40 populations. We compared the effectiveness of the iFAM-ED intervention with the delirium education intervention provided to family members of elective hip and knee surgery patients in the study by Bull et al.20 Similarly to the study by Bull et al., our study showed a statistically significant improvement in delirium knowledge scores from pre-intervention to two weeks post-intervention.20 Since knowledge is maintained after two weeks, family members may better detect delirium onset after their loved ones are discharged from the ICU (e.g., hospital ward, home). In contrast to our iFAM-ED intervention, the delirium education intervention in the study by Bull et al. was telephone-based and was delivered to family members by a nurse over a period of three weeks prior to their loved ones’ hospital admission (given the elective nature of their admission). The current study employed a short, video-based intervention that included comparable content to the Bull study, but was just as effective in a shorter amount of time (time to complete: 20 min as opposed to three weeks).

The iFAM-ED intervention has several benefits for educating family members of critically ill patients on detecting delirium symptoms and preventing and managing delirium using nonpharmacological strategies. Previous studies have shown that families benefit from the provision of understandable and consistent information.41,42,43 The current study supports the provision of understandable, consistently presented ICU delirium education. The iFAM-ED intervention was understandable; it was presented at a grade six reading level (as recommended by the National Institutes of Health and the American Medical Association),44,45 was created by a multidisciplinary team (including patient and family partners), and scored 85% (≥ 75% considered understandable) on the PEMAT-A/V tool. In a clinical setting, the iFAM-ED intervention is video-based and, as such, is a more consistent method for presenting ICU delirium education than nurse-provided ICU delirium education interventions that may vary in delivery and content and, based on nurse workload, may not be consistently provided.

Further refinement of the iFAM-ED intervention could improve a family member’s ICU delirium knowledge and ability to identify symptoms of delirium. For instance, after the iFAM-ED intervention, more participants distinguished symptoms of delirium from symptoms of dementia, but fewer participants identified dementia as a risk factor for delirium. It is possible that, after the iFAM-ED intervention, participants viewed delirium and dementia as two separate syndromes that could not occur concurrently. Further refinement of the iFAM-ED intervention could include an explanation of delirium superimposed on dementia. Family members who understand the difference between delirium and dementia may be helpful partners or resources to the ICU care team to recognize delirium superimposed on dementia.46

Further refinement of the iFAM-ED intervention could include additional description of disorganized thinking, (e.g., “This person acts different than they did before they were admitted to the ICU” or “If you ask this person a question, they may reply with an answer that does not make sense”) to improve family member’s identification of disorganized thinking in provided case vignettes. In addition, the readability of the FAM-CAM (Flesch-Kinkaid reading level ~11th grade) and the Sour Seven (Flesch-Kinkaid reading level ~12th grade) are higher than the suggested grade six level. Efforts to improve the readability of the FAM-CAM and Sour Seven may improve family member’s understanding of the questionnaires and may improve their identification of disorganized thinking and altered level of consciousness.

A recent qualitative study reported that family members wish to be involved with nonpharmacological delirium prevention, but would like delirium education at the beginning of the patient’s stay.13 Nevertheless, this may not be the case for all family members, who may feel stressed or overwhelmed when first introduced to the ICU environment. A video-based ICU delirium education intervention provides families an opportunity to engage, decline, or defer their participation in delirium prevention, detection, and management. This flexibility is an important consideration for any family education intervention because, as with all family engagement strategies, participation in patient care may not be desired by all family members (or patients). Another benefit of video-based ICU delirium education is that it allows families to watch the video on their own time or multiple times. Although the current study showed that video-based ICU delirium education was effective on its own, previous studies report that family members find communication with the bedside RN to be important.13,14,47,48 As such, video-based delirium education should be accompanied by ICU care team engagement. This need was illustrated by the small proportion of family members who completed the iFAM-ED module on their own, using their own device, compared with family members who used the study tablet with a research assistant nearby to answer any questions.

There were several limitations of this study. First, the study was conducted in one ICU located at an academic centre. Upon entry to the ICU, as standard of care, the family is provided a pamphlet on delirium (risk factors, symptoms, and actions). The ICU regularly screens for delirium using the Intensive Care Delirium Screening Checklist49 and discusses delirium at multidisciplinary rounds. As such, these results may not be generalizable to family members in other ICUs with different processes for delirium screening, monitoring, and management. Second, the current study may not be representative of family members who were overwhelmed or were highly anxious and declined to participate or did not complete all follow-up. Third, the current study lacked a control group. A future randomized-controlled trial is needed to compare the effects of the iFAM-ED intervention with a family-group who receives no education on delirium and ability to detect delirium symptoms in their loved one. Despite these limitations, the current study has several strengths. First, the iFAM-ED intervention was created together with a multidisciplinary team, including patient and family partners. Although the study occurred at a single centre, Foothills Medical Centre serves a diverse population with a catchment area of 1.8 million. Family-administered delirium detection, prevention, and management is not a replacement for established delirium screening, monitoring, and management, but may be helpful in settings with limited nurse staffing resources or where healthcare providers do not have familiarity with a patient’s cognitive status prior to admission.50

Conclusions

A video-based ICU delirium education intervention was effective in educating family members of critically ill patients on the detection of delirium symptom and the prevention and management of delirium using nonpharmacologic strategies. This delirium education may act as a primer for family members to partner with the ICU care team in delirium-related patient issues or empower family members to participate in delirium-focused discussions. This study supports future research evaluating whether ICU delirium education improves family members’ ability to detect symptoms of delirium and to prevent and manage delirium in their loved one during their ICU stay or if family participation in delirium prevention and management improves patient outcomes.