Translation,Adaptation,and Criterion Validation of the Family Caregiver Assessment Tool for French-Speaking Cardiovascular Patients in Canada (FAM-CAM-Fr)

Abstract

Background

Delirium is a common yet underdiagnosed condition in hospitalized older adults, particularly challenging to detect early in cardiology settings. Although delirium assessment tools improve detection rates, observations by family caregivers of patients’ cognitive changes can offer valuable insights, supplementing assessments by healthcare professionals. However, validated French-language tools for family caregivers to assess delirium in acute care settings in Canada are lacking.

Purpose

Translate, culturally adapt, and validate the Family Confusion Assessment Method for French-speaking cardiovascular patients and their caregivers (FAM-CAM-Fr).

Methods

The translation and cultural adaptation of the FAM-CAM were conducted following the guidelines of Sousa and Rojjanasrirat (2011). Criterion validation involved 100 dyads of family caregivers and hospitalized cardiovascular patients. The FAM-CAM-Fr's performance was assessed by comparing it to the Confusion Assessment Method (CAM) and the DSM-5 diagnostic criteria for delirium. Measures of sensitivity, specificity, and agreement with the CAM were calculated.

Results

The FAM-CAM-Fr showed high specificity (92.6%) but low sensitivity (58%) in detecting delirium. Cohen's Kappa indicated a moderate agreement (>0.50) between the FAM-CAM-Fr and the CAM. Despite family caregivers using the tool without prior training, indicating its usability in real-world settings, sensitivity was lower compared to studies that included caregiver training, though specificity was similar.

Conclusion

The FAM-CAM-Fr is promising as a specific tool for screening delirium in cardiovascular patients. Despite its low sensitivity, its high specificity indicates that it is effective at ruling out delirium. Future research should focus on further validation across various settings.

Keywords

Delirium detection FAM-CAM-Fr cardiovascular patients cultural adaptation family-centered delirium assessment caregiver-supported screening

Introduction

Delirium significantly increases morbidity and mortality, while substantially affecting the recovery of hospitalized older adults who frequently experience this complication (Wilson et al., 2020). Its occurrence can also impair patients’ baseline cognitive function (Hayhurst et al., 2020). Although early detection of delirium is critical to mitigating its impact, it remains a major challenge (Aikawa et al., 2024). Studies consistently reveal that healthcare professionals accurately identify fewer than 30% of delirium cases in hospitalized older adults (Titlestad et al., 2024; Wand et al., 2014), with even lower recognition rates among those with underlying neurocognitive disorders, such as dementia (Fick et al., 2013, 2015).

The systematic use of delirium assessment tools by healthcare professionals can improve detection rates (Barr et al., 2013; Gélinas et al., 2018; Registered Nurses’ Association of Ontario, 2016). However, even with these tools, identifying delirium remains challenging, particularly in acute care settings or among patients with neurocognitive disorders such as dementia (Fick & Mion, 2008; Holly et al., 2014; Steis & Fick, 2008). The severely compromised clinical condition of patients, combined with their short hospital stays, makes it difficult for healthcare professionals to acquire the knowledge needed to recognize the subtle changes that signal delirium onset. Understanding patients’ baseline cognitive function and personality traits is crucial for detecting these subtle indicators (Fick et al., 2015; Rosgen et al., 2018).

In this context, family caregivers are a valuable source of information that could supplement the health professionals’ assessments and could help identify subtle changes in the usual cognitive functioning of the patient (Bull et al., 2016, 2017; Mailhot, 2017; Martins et al., 2014; Steis et al., 2012). Integrating family caregiver's knowledge of the patient and their observations of changes, using a structured tool, could help health professionals detect delirium more quickly and monitor it more accurately.

A recent systematic review identified three tools that could be independently completed by family caregivers (Rosgen et al., 2018). Among these three tools, the Family-Confusion Assessment Method (FAM-CAM) demonstrated strong psychometric properties and required minimal family training for its completion. The FAM-CAM is based on the most widely used delirium detection tool for healthcare professionals, i.e., the Confusion Assessment Method (Inouye et al.). More importantly, the FAM-CAM includes simple yes-no items along with an item that allows detailed qualitative description of families’ observations. Therefore, the FAM-CAM holds great potential in increasing delirium detection if supplemented to healthcare professionals’ systematic assessments. This tool has been validated in outpatients and long-term care settings and is available in English and other languages such as Portuguese and Chinese (Martins et al., 2014; Steis et al., 2012; Xing et al., 2024). The original American English version is available online at https://help.agscocare.org/login-required. The French version of the FAM-CAM is not available, thereby limiting the potentially significant contribution of families in the rapid detection of delirium in a Francophone population.

The absence of a validated French version of the FAM-CAM restricts the capacity of Francophone caregivers in Canada to engage in delirium screening. In the context of Canada's officially bilingual healthcare system, the availability of culturally and linguistically adapted tools is essential to ensuring equitable access to care for Francophone populations. Canadian healthcare guidelines recommend utilizing validated and culturally appropriate screening tools whenever possible (Registered Nurses’ Association of Ontario, 2016). Language barriers can hinder delirium detection. While professional interpreters may be available in certain contexts, a validated screening tool in the caregiver's preferred language would enhance accessibility and facilitate early detection. Modifying the FAM-CAM for French-speaking populations aligns with these recommendations and addresses a critical gap in delirium detection screening.

The purpose of this study was to create the version of the FAM-CAM for French speaking populations of Canada (FAM-CAM-Fr) and to perform an initial validation in cardiovascular patients in an acute care setting. The primary objective was to translate and culturally adapt the FAM-CAM into a French version for Canada according to the rigorous approach proposed by Sousa and Rojjanasrirat (2011). The secondary objective was to examine criterion validation of the FAM-CAM-Fr using the reference standard in the field of delirium (i.e., medical diagnosis according to DSM-5 criteria) and related criteria (i.e., CAM and keywords associated with delirium (encephalopathy, confusion, agitation, hallucination, somnolence, Haldol (Puelle et al., 2015)).

Methods

This study was conducted and reported in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) guidelines (Vandenbroucke et al., 2007). The reporting checklist is provided in Appendix 1.

Primary Objective: Translation and Cultural Adaptation

The original English version of the FAM-CAM was translated and culturally adapted to French for Canada using the method suggested by Sousa and Rojjanasrirat (2011). Following the copyright clearance from the author of the FAM-CAM for translation and use in this study, we conducted five steps to translate and culturally adapt the tool presented in Table 1. The table describes a detailed five-step translation and validation process for adapting the FAM-CAM tool from English to French. Step 1 consisted in the forward translation of the tool by two translators, both native French speakers, who independently translated the original English FAM-CAM into French, producing two initial French versions of the tool.

Table 1.

Translation Process.

1. FORWARD TRANSLATION
• Two translators, whose mother tongue was French, independently translated from English to French. This step resulted in two French versions of the FAM-CAM
2. COMPARISON OF THE TWO FRENCH VERSIONS
• A third person, not involved in step 1, compared the two French versions of the FAM-CAM and looked for discrepancies of words, sentences and meanings. Any ambiguities and discrepancies were discussed and resolved with a committee approach. This committee included members of the research team and translators involved in the Forward translation. This step resulted in a third French version of the FAM-CAM.
3. BLIND BACK TRANSLATION
• Two translators, whose mother tongue was English and who were blinded to the tool, independently translated from French to English. This step resulted in two versions of the FAM-CAM in English.
4. COMPARISON OF THE TWO ENGLISH VERSIONS
• A multidisciplinary committee compared the two English versions that resulted from step 3 to the original FAM-CAM. This committee looked for discrepancies of words, sentences and meanings. The two back translations were shared with the team of the Hospital Elder Life Program (HELP) from the Aging Brain Center - Hebrew SeniorLife, Institute for Aging Research. They were involved in discussions. Any ambiguities and discrepancies will be discussed and resolved with a committee approach. The committee then produced a pre-final French version of the FAM-CAM.
5. THE PRE-TEST
• A pre-test was conducted to assess clarity of each items of the French Version of the FAM-CAM. Eight experts from the research team and 10 family volunteers participated in the pre-test which led to the modification of two items of the French Version of the FAM-CAM. This resulted in a refined final French version of the FAM-CAM which was used in this validation study.

During step 2, a third individual, uninvolved in the initial translation, compared the two French versions to identify discrepancies in words, sentences, and meanings. A committee, comprising members of the research team and the original translators, discussed and resolved any ambiguities and differences, resulting in a unified third French version. Step 3 consisted in the blind back translation with two native English-speaking translators, who were blinded to the original FAM-CAM, independently translated the third French version back into English. This process generated two back-translated English versions for comparison. During step 4, a multidisciplinary committee compared the two back-translated English versions with the original FAM-CAM, carefully analyzing discrepancies in wording, sentence structure, and meaning. The Hospital Elder Life Program (HELP) team at the Aging Brain Center of Hebrew SeniorLife also reviewed these translations. Ambiguities were discussed, and discrepancies were resolved through a committee approach, leading to a pre-final French version.

The last step was a pre-test performed to assess the clarity and appropriateness. This pre-test was conducted with eight experts (i.e., researchers, clinicians, not members of the committee) from the research team and ten family volunteers who assessed the degree of clarity of the FAM-CAM-Fr version of the FAM-CAM. A rate of unclear of ≥ 20% resulted in the revision of the unclear item by the research team and committee members to further refine of the FAM-CAM-Fr. The FAM-CAM instructions were also translated and culturally adapted to the Canadian context during this process (Inouye et al., 2012). The FAM-CAM-Fr is covered under the original copyright and can be obtained from the HELP website (https://help.agscocare.org/login-required).

Secondary Objective: Criterion Validation of the FAM-CAM for French Speaking Populations of Canada

Study Sample

In this validation study, we included family caregivers and patients hospitalized in a specialized cardiovascular acute care hospital in Quebec, specifically on the cardiology unit. The family caregiver was defined as the individuals who provide support and with whom the patient has a significant relationship (Davidson et al., 2017). The family caregiver had to be 18 years old or older; understand and read French; have the capacity to provide informed consent; identify as a family caregiver (spouse, child, other family, friend) of an inpatient; have visited the patient at the hospital at least twice. The family caregiver also had to report having regular contact with the inpatient before hospitalization (≥ once a week). In the literature, this frequency of interaction was considered sufficient for the family member to report the usual cognitive and functional state of the patient, a knowledge necessary to be able to identify changes (Martins et al., 2014; Steis et al., 2012).

The hospitalized patient had to have the ability to provide informed consent or have a legal representative if their cognitive functioning was altered. Patients also had to present a stable clinical condition, allowing assessment of delirium.

Study Procedures

Following consent, data collection was performed at a time that was appropriate, i.e., either immediately following consent or later during hospitalization and included the completion of FAM-CAM-Fr by the family caregiver (described below) and CAM by the research nurse (described below). Both the FAM-CAM-Fr and CAM were to be completed within an hour of each other, independently and included information (i.e., changes in cognitive functioning or state of consciousness) over the last 24 h. – the family caregiver was asked to reflect on their interactions with the patient in the last 24 h and the project nurse completed the CAM based on their interview and on information from the charge nurse. The diagnosis and keywords were abstracted from the medical retrospectively. The project underwent scientific and ethics approval at the study hospital (2019–2426). The participants are identified by a code number. The key linking the name to the research file was kept by the lead researcher in an electronic file secured by a password.

Measures

Sociodemographic and Clinical Data

Among patients, the following data was collected from the medical chart by a research assistant: age, sex, education level, primary diagnosis for current hospitalization, medical history. Among family caregivers the following variables were collected from a self-reported questionnaire completed by families: Age, sex, mother tongue, level of education, relationship to the patient (friend, spouse, child, other), frequency of interactions with the patient and living situation with the patient (cohabitation, number of years), number and average duration of bedside visits during the present hospitalization.

Assessment of Delirium

Original English-version of the FAM-CAM. The FAM-CAM-Fr is a translation of the original American English version, developed by Inouye et al. (Inouye et al., 2012) based on the CAM. The original FAM-CAM is a tool that supports the detection of delirium that is completed by families in about five minutes. It includes the main features of the CAM (acute onset of observed change, fluctuation, inattention, disorganized thinking and altered level of consciousness). The tool contains 11 items: ten multiple-choice questions and one question answered with a description of the observed changes. The FAM-CAM can either be a positive score, suggestive of delirium, or a negative score according to the algorithm. In this study, family caregivers completed the FAM-CAM-Fr, independently from the research nurse, without knowledge of the CAM scores. Item 8 of the FAM-CAM allows family caregivers to qualitatively describe the changes they observed in their hospitalized relative.

Diagnosis. Medical chart was reviewed to identify a diagnosis of delirium within a timeframe of 12 h before or after the completion of FAM-CAM. A diagnosis of delirium corresponded to the criteria defined in the DSM-5 and was performed by attending physician (American Psychiatric Association, 2013).

Keywords associated with delirium. Keywords that may be associated with delirium were also retrieved from the medical chart. These keywords include encephalopathy, confusion, agitation, hallucination, somnolence, Haldol (Puelle et al., 2015).

CAM. The CAM was the reference tool completed by a trained research nurse. It consists of four items based on the diagnostic criteria for delirium (Inouye et al., 1990). The CAM items include acute change in cognitive state, impaired attention, disorganized thinking, and altered state of consciousness. Acute change, impaired attention coupled with the presence of disorganized thinking or altered state of consciousness results in a positive CAM score that suggests the presence of delirium. In this study, the project nurse completed the CAM without knowledge of the FAM-CAM-Fr scores.

Statistical Analysis

Sample Size

A sample size of 10 families for the primary objective and 100 dyads of family caregivers and patients for the secondary objective was deemed appropriate to attain the study objectives. Based on the results of a delirium study performed in the same clinical setting, it was estimated that 16% of subjects will present a positive diagnosis for delirium and 84% will be negative (Mailhot et al., 2017). We anticipated the sensitivity to be calculated on 16% of the total number of subjects. It was expected that the sensitivity could vary between 75% and 90% (Martins et al., 2014; Steis et al., 2012) we retained a sensitivity of 75% to be conservative. With a sample size of 100 (i.e., 16 true positives), the precision for the sensitivity estimate was of ± 21% with a 95% confidence interval. With this same sample size, the specificity was expected to be calculated on 84 subjects. Assuming a specificity of 85% (Martins et al., 2014; Steis et al., 2012), the precision for estimating the specificity would be 8% with a 95% confidence interval.

General Statistical Considerations

We used SPSS software 29 (Version 29.0.1.0; IBM SPSS Statistics for Windows Armonk, NY: IBM Corp.). Participants were presented using descriptive statistics with mean, standard deviation, median, minimum and maximum to describe the continuous variables, and frequencies and percentages to describe the categorical variables. The normality of continuous variables was assessed using the Shapiro–Wilk test. Sociodemographic and clinical data were contrasted according to CAM status. All statistical tests were bilateral and a significance level of 0.05 was used.

Descriptive statistics were also used to describe proportion of delirium based on the FAM-CAM-Fr, the diagnosis of delirium, the CAM and the keywords.

Primary Objective

Descriptive statistics were used to describe the results of the pretest on clarity of the FAM-CAM-Fr among the eight research team members and the ten family volunteers. The number of modified items for each round is also reported.

Secondary Objective

We assessed the sensitivity and specificity of the FAM-CAM-Fr by contrasting it to the reference standard (medical diagnosis according to DSM-5 criteria). Sensitivity was estimated by the proportion of FAM-CAM-Fr positives among the DSM-5 positives (true positives according to the reference standard) while the specificity was estimated by the proportion of negatives on the FAM-CAM-Fr among the negatives according to the DSM-5 (true negatives according to the reference standard). Sensitivity and specificity were presented with their 95% confidence intervals.

The following classification of assessments made with the FAM-CAM-Fr were also presented:

True positives (TP) were defined as being positive on the FAM-CAM-Fr matched with the presence of a delirium diagnosis;

False positives (FP) were defined as positive on the FAM-CAM-Fr matched with no delirium diagnosis;

True negatives (TN) were defined as negative on the FAM-CAM-Fr matched with no delirium diagnosis;

False negatives (FN) were defined as negative on the FAM-CAM-Fr matched with the presence of a delirium diagnosis.

The same procedure was used to contrast the FAM-CAM-Fr to the CAM and the delirium key words. For the keywords, the use of any of the keywords was considered as a «yes». Finally, to assess the reliability FAM-CAM-Fr against the CAM and the keywords, Cohen's kappa were calculated. Cohen's Kappa quantifies the level of agreement beyond chance and is particularly suitable for categorical data, making it an appropriate measure for evaluating diagnostic concordance.

Finally, the accuracy of the FAM-CAM-Fr was calculated by computing the following: (true positives + true negatives) / N.

Results

Primary Objective of Translation and Cultural Adaptation

Eight experts evaluated the clarity of the items and instructions in the initial version of the FAM-CAM-Fr tool, comprising four researchers and four clinicians specialized in cardiovascular care. All items were rated as clear by these experts.

Additionally, ten family members assessed the clarity of the tool and instructions. Items rated as unclear by more than 20% of the family members included item 6 (n = 2, 20% rated this item as unclear) and item 11 (n = 3, 30%) from the tool, as well as item 1 (n = 2, 20%) and item 5 (n = 2, 20%) from the instructions. These items were subsequently refined and incorporated into the final version of the FAM-CAM-Fr.

Secondary Objective of Criterion Validation

A total of 100 dyads participated in the study and were recruited between July 2018 and June 2019, thus 100 family caregivers completed the FAM-CAM-Fr while the research nurse assessed the 100 corresponding patients with the CAM. As shown in Table 2, 19 patients (19%) were diagnosed with delirium. More than half of the participants were male, with an average age of around 75 years, and there were no significant differences in age, sex, or education between the delirium and non-delirium groups. Family caregivers visited their loved one who presented delirium more often per week compared to the group with no delirium (Table 2).

Table 2.

Sociodemographic Characteristics of Participants Stratified by Delirium Diagnosis.

Characteristics	No delirium n = 81	Delirium n = 19	Overall n = 100	P value between D + and D-
Patients
Age, years, mean (SD)^a	74.6 (12.8)	80.2 (11.0)	75.57 (12.6)	0.567
Sex, male, n (%)^b	41 (51.2)	10 (58.8)	51 (52.6)¹	0.570
Education level, n (%)^c		0.301
Primary school	15 (19.5)	7 (41.2)	22 (23.4)	−
High school	32 (41.6)	7 (41.2)	39 (41.5)	−
CEGEP (publicly funded college)	15 (19.5)	1 (5.9)	16 (17.0)	−
University	14 (18.2)	2 (11.8)	16 (17.0)	−
Other	1 (1.3)	−	1 (1.1)	−
Family caregivers
Age	59.8 (15.3)	65.1 (17.2)	60.6 (15.6)	0.723
Gender^d				0.783
Female	57 (72.2)	11 (68.8)	68 (71.6)	−
Male	22 (27.8)	5 (31.3)	27 (28.4)	−
Education level^d				0.927
Primary school	3 (3.8)	0 (0.0)	3 (3.2)	−
High School	31 (39.2)	7 (43.8)	38 (40.0)	−
Cegep (community college)	19 (24.1)	4 (25.0)	23 (24.2)	−
University	25 (31.6)	5 (31.3)	30 (31.6)	−
Autre	1 (1.3)	0 (0.0)	1 (1.1)	−
Relationship with patient				0.935
Spouse	37 (46.8)	9 (56.3)	46 (48.4)	−
Son or daughter	26 (32.9)	5 (31.3)	31 (32.6)	−
Mother or Father	1 (1.3)	0 (0.0)	1 (1.1)	−
Friends	4 (5.1)	0 (0.0)	4 (4.2)	−
Sibling	5 (6.3)	1 (6.3)	6 (6.3)	−
Other	6 (7.6)	1 (6.3)	7 (7.4)	−
Frequency of weekly contact with patient				0.551
None	7 (14.3)	3 (30.0)	10 (16.9)	−
Once or twice	10 (20.4)	1 (10.0)	11 (18.70)	−
Three or four	5 (10.2)	1 (10.0)	6 (10.20)	−
More than four	27 (55.1)	5 (50.0)	32 (54.2)	−
Living with the patient				0.553
Yes	43 (54.4)	10 (62.5)	53 (55.8)	−
Non	36 (45.6)	6 (37.5)	42 (44.2)	−
How long, mean (SD), years	37.7 (18.1)	10 (46.8)	40.0 (17.0)	0.070
Frequency of visits per week				0.021
None	8 (12.5)	0 (0.0)	8 (10.1)	−
Once or twice	18 (28.1)	3 (20.0)	21 (26.6)	−
Three or four	4 (6.3)	2 (13.3)	6 (7.6)	−
More than four	34 (53.1)	10 (66.7)	44 (55.7)	−
Approximate duration of visits, hours	6.2 (4.1)	6.7 (3.7)	6.3 (4.0)	0.281
Frequency of visits at the patient's bedside since current hospital admission				0.616
First time	17 (21.5)	2 (11.8)	19 (19.8)	−
Once or twice	9 (11.4)	0 (0.0)	9 (9.4)	−
Three or four	12 (15.2)	3 (17.6)	15 (15.6)	−
More than four	41 (51.9)	12 (70.6)	53 (55.2)	−
Mother tongue^d				0.555
French	69 (87.3)	14 (87.5)	83 (87.4)	−
English	1 (1.3)	1 (6.3)	2 (2.1)	−
Italian	7 (8.9)	1 (6.3)	8 (8.4)	−
Spanish	2 (2.5)	−	2 (2.1)	−

Values are presented as number (percentage). Variables normally distributed are presented as mean standard deviation (SD). Variables not normally distributed are presented as median (interquartile range [IQR]).

Note: ^aMissing data for four participants. ^bMissing data for six participants. ^cMissing data for six participants. ^dMissing data for five participants.

D+: group of patients with delirium; D-: group of patients without delirium.

Table 3 presents the FAM-CAM-Fr and CAM results according to delirium diagnosis. Overall, the sample included 19 patients with a diagnosis of delirium, 17 patients received a positive score on the FAM-CAM-Fr and 20 with a positive score on the CAM. Keywords were mentioned more frequently in the group of patients with delirium (Table 3), with ‘confusion,’ ‘agitation,’ and ‘drowsiness’ being the most cited keywords in this group.

Table 3.

FAM-CAM-Fr and CAM Results According to Delirium Diagnosis.

	No delirium n = 81	Delirium n = 19	Overall n = 100	P value between D + and D-
FAM-CAM-Fr				<,001
Negative	75 (92.6)	8 (42.1)	83 (83.0)	−
Positive	6 (7.4)	11 (57.9)	17 (17.0)	−
CAM				<,001
Negative	78 (96.3)	2 (10.5)	80 (80.0)
Positive	3 (3.7)	17 (89.5)	20 (20.0)
Keywords mentioned
Encephalopathy	0 (0.0)	3 (15.8)	3 (3.0)	<,001
Confusion	6 (7.4)	13 (68.4)	19 (19.0)	<,001
Agitation	2 (2.5)	8 (42.1)	10 (10.0)	<,001
Hallucination	0 (0.0)	5 (26.3)	5 (5.0)	<,001
Drowsiness	2 (2.5)	7 (36.8)	9 (9.0)	<,001
Haldol	1 (1.2)	1 (5.3)	2 (2.0)	0.259

Note: FAM-CAM-Fr: Family Caregiver Assessment Tool for French-Speaking patients in Canada, CAM: Confusion Assessment Method.

The performance outcomes of the FAM-CAM-Fr are detailed in Table 4. Compared to the medical diagnosis, the FAM-CAM-Fr demonstrated low sensitivity (<60%) but high specificity (>90%), showing a similar performance when evaluated against the CAM. The Cohen's Kappa values for the agreement between the diagnosis of delirium and the FAM-CAM-Fr, as well as between the CAM and the FAM-CAM-Fr, were both greater than 0.50. Although the specificity of keywords was >90%, sensitivity was very low (<50%) and Cohen's kappa was the lowest. Finally, its accuracy was of 0.86.

Table 4.

Performance Outcome of the FAM-CAM-Fr.

	Sensitivity	Specificity	PPV	NPV	Cohen's kappa
FAM-CAM-Fr against :	Percentage (95% Confidence Interval)
DSM-5	58 (0.36; 0.78)	92.6 (0.85; 0.97)	64.7 (0.41; 0.84)	90.4 (0.83; 0.95)	0.53
CAM	60 (0.38; 0.79)	93.8 (0.87; 0.98)	70.6 (0.47; 0.88)	90.4 (0.83; 0.95)	0.57
Keywords	44 (0.26; 0.63)	92 (0.84; 0.97)	64.7 (0.41; 0.84)	83.1 (0.74; 0.90)	0.40

Note: FAM-CAM-Fr: Family Caregiver Assessment Tool for French-Speaking patients in Quebec and Canada, CAM: Confusion Assessment Method, PPV: Positive predictive value, NPV: Negative predictive value.

Discussion

The primary objective of this study was to translate and culturally adapt the original English version of the FAM-CAM for French-speaking cardiovascular patients in acute care settings in Canada. The FAM-CAM-Fr was developed through a rigorous process, involving expert input from translators, researchers, and family caregivers. Our results demonstrated low sensitivity but high specificity for the FAM-CAM-Fr, with moderate agreement with the reference standard and the CAM reference tool (McHugh, 2012). Notably, family caregivers administered the FAM-CAM-Fr without any specific training or detailed information about delirium provided by the research team. They relied solely on pamphlets available in the unit and the written instructions accompanying the tool. Despite this, the tool showed potential for enhancing delirium detection in clinical settings.

The performance metrics observed in our study are consistent with those reported for other versions of the FAM-CAM, though some variation exists. For example, studies by Steis et al. and Martins et al. reported higher sensitivity and specificity in different patient populations, indicating that contextual factors and caregiver training can influence the tool's effectiveness (Martins et al., 2014; Steis et al., 2012). Steis’ study included face-to-face sessions on how to use the FAM-CAM, offering education on each symptom and guidance on item scoring. In contrast, our study did not provide any training, aiming to test the FAM-CAM's validity under conditions more reflective of real-world cardiology practice. The improved sensitivity observed in studies with caregiver training suggests that structured education could enhance the tool's accuracy, particularly in high-risk populations such as older adults with pre-existing cognitive impairments. The rigorous validation of the German FAM-CAM, which demonstrated high sensitivity and specificity, further emphasizes the importance of cultural adaptation and methodological rigor in translating and validating tools (Greindl et al., 2022). While our study's sensitivity was lower, our specificity was comparable, suggesting the French FAM-CAM's robust potential in specific settings despite varying performance metrics across different populations and languages.

A systematic review and meta-analysis by Zhou et al. (2024) reported pooled sensitivity and specificity for the FAM-CAM at 0.74 and 0.91, respectively, based on data from 483 dyads across seven studies (Zhou et al., 2024). These studies primarily focused on older adults and critically ill populations. While our study's sensitivity was lower, the specificity was comparable, suggesting that the FAM-CAM can effectively rule out delirium across various populations and settings.

Studies included in Zhou et al., reported higher performance for various culturally adapted versions of the FAM-CAM (Table 5). For example, Steis et al. validated the original English version among families of older adults living in the community, while Martins et al. assessed the Portuguese version in a sample of critically ill patients. Xing et al. adapted and validated the Chinese version of the FAM-CAM in an ICU setting using a different reference standard. Bull et al. conducted their evaluation among non-cardiac surgery patients and reported levels of agreement comparable to our findings. Aside these translations, the FAM-CAM was translated into German. Greindl et al. used a prospective monocentric diagnostic study to validate the FAM-CAM in a 14-bed medical ICU of a university hospital in Germany (Greindl et al., 2022). The FAM-CAM was translated based on the Principles of Good Practice for the Translation and Cultural Adaptation Process for Patient-Reported Outcomes of the International Society for Pharmacoeconomics and Outcome Research (Wild et al., 2005) and involved three physicians specialized in ICU delirium. Cognitive debriefing interviews followed for the FAM-CAM's cultural adaptation with 28 participants, showing that the German FAM-CAM was understandable in the domain language and content. For the validation process, 50 dyads of critically ill patients and their family members were recruited. A summary of these studies, including the characteristics of the study populations, reference standards, and diagnostic performance metrics, is presented in Table 5.

Table 5.

Performance Outcome for the Original English Version of the FAM-CAM and for Translated and Culturally Adapted Versions of the FAM-CAM.

Reference	Study population	Validation standard	Sensitivity (95% CI)	Specificity (95% CI)	Kappa (95% CI)
Steis et al. (2012)	Older adults living in the community (n = 52)	DSM-5 diagnosis	88% (47–99%)	98% (86–100%)	Not reported
Martins et al. (2014)	Critically ill patients (n = 40)	DSM-5 diagnosis	75% (35–95%)	90% (74–97%)	Not reported
Bull et al. (2016, 2017)	Non-cardiac surgery patients	CAM vs. FAM-CAM (agreement)	Not reported	Not reported	0.47
Greindl et al. (2022)	ICU patients in a 14-bed medical ICU (n = 50)	CAM-ICU	95.5% (77.2–99.9%)	89.3% (71.8–97.7%)	0.84 (0.69–0.99)
Xing et al. (2024)	Critically ill patients (n = 190)	ICDSC	88.0%	89.0%	0.76

Note: CI: Confidence Interval; DSM-5: Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition; CAM: Confusion Asssessment Method; CAM-ICU: CAM-Intensive Care Unit; ICDSC: Intensive Care Delirium Screening Checklist.

These studies highlight the FAM-CAM's potential, though small sample sizes limit the precision of sensitivity and specificity estimates. Another key difference between our study and those by Steis and Martins is the level of training provided to families. For example, Steis’ study included face-to-face sessions on how to use the FAM-CAM, offering education on each symptom and guidance on item scoring. In contrast, our study did not provide any training, aiming to test the FAM-CAM's validity under conditions more reflective of real-world cardiology practice.

Despite advancements in delirium understanding, its detection in older cardiology patients, particularly those with pre-existing dementia, remains challenging (Shrestha & Fick, 2023). Delirium is often under-recognized in cardiology, despite its association with poor outcomes. There is growing interest in care models tailored to older adults or to adults presenting with delirium, which emphasize systematic mental status and delirium screening, alongside protocols for managing geriatric syndromes. Its integration into routine practice has the potential to inform national strategies aimed at improving delirium detection among French-speaking populations, thereby promoting equitable care across Canada. Given its ease of use, the FAM-CAM-Fr could be integrated into routine delirium screening in cardiology settings as a complementary tool alongside clinician-administered assessments. Implementing structured caregiver training programs could further optimize its accuracy while maintaining feasibility in clinical practice. Broad adoption of tools such as the FAM-CAM-Fr as part of standardized screening protocols could contribute to reducing the burden of delirium on the healthcare system, align with national geriatric care priorities and improve outcomes through earlier detection.

A limitation of this study is its conduct in a single cardiology unit, which may limit the generalizability of the findings. Additionally, we only included family members who had regular contact (≥ once a week) with the patient before hospitalization, potentially affecting the external validity of the FAM-CAM-Fr. Our patient population may have had undiagnosed cognitive impairments, complicating the detection of hypoactive delirium. Strengths of the study include the use of the CAM reference tool administered by trained staff blinded to caregiver responses on the FAM-CAM, providing a real-world validation of the tool in a cardiology setting. Moreover, the translation process followed a recognized and rigorous methodology.

Conclusion

In conclusion, our study resulted in the development of the FAM-CAM-Fr, a tool that may empower families to communicate observed changes in their hospitalized relative, thereby potentially enhancing delirium detection in cardiology settings. These findings are significant, as family caregivers completed the FAM-CAM without formal training or detailed information about delirium, underscoring the tool's potential to improve detection through the insights of family members, who are often most familiar with the patient's baseline cognitive function. Given the difficulties in identifying delirium, particularly in older cardiology patients with pre-existing dementia, the FAM-CAM offers a valuable and user-friendly screening option. In addition, structured family caregiver training programs should be considered to enhance screening accuracy and better incorporate the French version of the FAM-CAM into clinical workflows. Assessing the feasibility of implementing the French version of the FAM-CAM in routine practice would further strengthen its relevance in delirium detection strategies. Future research should prioritize multi-site validation of French version of the FAM-CAM across diverse clinical settings to enhance its generalizability and relevance. Investigating the impact of structured caregiver training on its diagnostic accuracy remains essential.

Supplemental Material

sj-doc-1-cjn-10.1177_08445621251350027 - Supplemental material for Translation, Adaptation, and Criterion Validation of the Family Caregiver Assessment Tool for French-Speaking Cardiovascular Patients in Canada (FAM-CAM-Fr)

Supplemental material, sj-doc-1-cjn-10.1177_08445621251350027 for Translation, Adaptation, and Criterion Validation of the Family Caregiver Assessment Tool for French-Speaking Cardiovascular Patients in Canada (FAM-CAM-Fr) by Tanya Mailhot, Zineb Bouaouina, Imène Khetir, Céline Gélinas, Judith Brouillette, Claudie Roussy, Stéphanie Jarry and Patrick Lavoie in Canadian Journal of Nursing Research

Footnotes

Author Contributions

T.M. contributed to the study concept and design, data acquisition, data analysis, interpretation of results, and manuscript preparation. Z.B. I.K. participated in data analysis and interpretation, as well as manuscript drafting. C.G., J.B., Z.B., I.K. and S.J. contributed to study design and critical revision of the manuscript. C.R., J.B. and P.L. contributed to study design, data interpretation, and provided critical revisions. All authors reviewed and approved the final version of the manuscript.

Sponsor's Role: NA.

Data Availability Statement

Data available on demand.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Ethical Approval and Informed Consent Statements

This project was approved by the Montreal Heart Institute ethics committee and all participants signed consent forms.

Funding

This study was supported by a grant from the Fond de Recherche du Québec- Santé, as well as institutional support from Montreal Heart Institute Foundation.

Preprint

This manuscript has not been previously published as a preprint.

ORCID iDs

Tanya Mailhot

Patrick Lavoie

Supplemental Material

Supplemental material for this article is available online.

Author Biographies

Tanya Mailhot is an Assistant Professor in Nursing Science. Her area of research focuses on the development of nursing interventions and the use of neuromonitoring technologies to prevent and manage delirium, particularly in cardiac surgery patients.

Zineb Bouaouina is a student in the Master of Nursing Science. Her area of research focuses on critical care nursing, with an emphasis on communication between families and healthcare professionals around delirium. Imène Khetir is a medical student and research assistant with an interest in tool development.

Céline Gélinas is a Professor in Nursing Science. Her area of research focuses on pain assessment and management in critically ill adults, with particular expertise in the development and validation of behavioral pain assessment tools for non-communicative patients.

Judith Brouillette is a Psychiatrist, Head of the Department of Psychiatry, and a Clinician-Scientist. Her area of research focuses on the neuropsychiatric manifestations of cardiovascular diseases, including anxiety.

Claudie Roussy is a Nurse Practitioner with expertise in cardiology. Her clinical and research interests focus on the management of complex cardiac conditions and the optimization of advanced nursing practices in cardiovascular care.

Stéphanie Jarry is a healthcare leader with a PhD in Clinical Pharmacology and a background in nursing. Her area of research focuses on clinical research and data analysis in cardiology and neuroscience, bridging scientific expertise with clinical practice to support evidence-based strategies in healthcare.

Patrick Lavoie is an Associate Professor in Nursing Science and holds the Canada Research Chair in Learning of Nursing Decision-Making. His research focuses on the education and training of nurses and other healthcare professionals in critical care settings. He specializes in clinical decision-making and judgment in life-threatening situations.

References

Aikawa

Ogata

Honda

Nagai

Murata

Morii

Anzai

Nishimura

Noguchi

(2024). Prolonged delirium during hospitalization is associated with worse long-term and short-term outcomes in patients with acute heart failure. International Journal of Cardiology, 399, 131776. https://doi.org/10.1016/j.ijcard.2024.131776

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders: DSM-5™ (5th ed.). American Psychiatric Publishing, Inc.

Barr

Fraser

G. L.

Puntillo

Ely

E. W.

Gélinas

Dasta

J. F.

Davidson

J. E.

Devlin

J. W.

Kress

J. P.

Joffe

A. M.

Coursin

D. B.

Herr

D. L.

Tung

Robinson

B. R. H.

Fontaine

D. K.

Ramsay

M. A.

Riker

R. R.

Sessler

C. N.

Pun

, … Jaeschke, R. (2013). Clinical practice guidelines for the management of pain, agitation, and delirium in adult patients in the intensive care unit. Critical Care Medicine, 41(1), 263–306. https://doi.org/10.1097/CCM.0b013e3182783b72

Bull

M. J.

Boaz

Maadooliat

Hagle

M. E.

Gettrust

Greene

M. T.

Holmes

S. B.

Saczynski

J. S.

(2017). Preparing family caregivers to recognize delirium symptoms in older adults after elective hip or knee arthroplasty. Journal of the American Geriatrics Society, 65(1), e13–e17. https://doi.org/10.1111/jgs.14535

Bull

M. J.

Boaz

Sjostedt

J. M.

(2016). Caregiver delirium knowledge questionnaire (CDKQ). Journal of Applied Gerontology, 35(7), 744–758. https://doi.org/10.1037/t53728-000

Davidson

J. E.

Aslakson

R. A.

Long

A. C.

Puntillo

K. A.

Kross

E. K.

Hart

Cox

C. E.

Wunsch

Wickline

M. A.

Nunnally

M. E.

Netzer

Kentish-Barnes

Sprung

C. L.

Hartog

C. S.

Coombs

Gerritsen

R. T.

Hopkins

R. O.

Franck

L. S.

Skrobik

, … Curtis, J. R. (2017). Guidelines for family-centered care in the neonatal, pediatric, and adult ICU. Critical Care Medicine, 45(1), 103–128. https://doi.org/10.1097/CCM.0000000000002169

Fick

D. M.

Inouye

S. K.

Guess

Ngo

L. H.

Jones

R. N.

Saczynski

J. S.

Marcantonio

E. R.

(2015). Preliminary development of an ultrabrief two-item bedside test for delirium. Journal of Hospital Medicine, 10(10), 645–650. https://doi.org/10.1002/jhm.2418

Fick

D. M.

Mion

L. C.

(2008). Delirium superimposed on dementia. AJN The American Journal of Nursing, 108(1), 52–60. https://doi.org/10.1097/01.NAJ.0000304476.80530.7d

Fick

D. M.

Steis

M. R.

Waller

J. L.

Inouye

S. K.

(2013). Delirium superimposed on dementia is associated with prolonged length of stay and poor outcomes in hospitalized older adults. Journal of Hospital Medicine, 8(9), 500–505. https://doi.org/10.1002/jhm.2077

10.

Gélinas

Bérubé

Chevrier

Pun

B. T.

Ely

E. W.

Skrobik

Barr

(2018). Delirium assessment tools for use in critically ill adults: A psychometric analysis and systematic review. Critical Care Nurse, 38(1), 38–49. https://doi.org/10.4037/ccn2018633

11.

Greindl

Weiss

Magnolini

Lingg

Mayer

Schaller

S. J.

(2022). 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. Journal of Advanced Nursing, 78(10), 3207–3216. https://doi.org/10.1111/jan.15227

12.

Hayhurst

C. J.

Marra

Han

J. H.

Patel

M. B.

Brummel

N. E.

Thompson

J. L.

Jackson

J. C.

Chandrasekhar

Ely

E. W.

Pandharipande

P. P.

Hughes

C. G.

(2020). Association of hypoactive and hyperactive delirium with cognitive function after critical illness. Critical Care Medicine, 48(6), e480–e488. https://doi.org/10.1097/CCM.0000000000004313

13.

Holly

Rittenmeyer

Weeks

S. M.

(2014). Evidence-based clinical audit criteria for the prevention and management of delirium in the postoperative patient with a hip fracture. Orthopaedic Nursing, 33(1), 27–34; quiz 35-26. https://doi.org/10.1097/nor.0000000000000020

14.

Inouye

S. K.

Puelle

Saczynski

Steis

(2012). The family confusion assessment method (FAM-CAM): Instrument and training manual. Boston, MA: Hospital Elder Life Program.

15.

Inouye

S. K.

van Dyck

C. H.

Alessi

C. A.

Balkin

Siegal

A. P.

Horwitz

R. I.

(1990). Clarifying confusion: The confusion assessment method. A new method for detection of delirium. Annals of Internal Medicine, 113(12), 941–948. https://doi.org/10.7326/0003-4819-113-12-941

16.

Mailhot

(2017). Étude pilote randomisée évaluant une intervention infirmière de mentorat d’aidants familiaux pour gérer le délirium post-chirurgie cardiaque. Université de Montréal, Papyrus.

17.

Mailhot

Cossette

Côté

Bourbonnais

Côté

M. C.

Lamarche

Denault

(2017). A post cardiac surgery intervention to manage delirium involving families: A randomized pilot study. Nursing in Critical Care, 22(4), 221–228. https://doi.org/10.1111/nicc.12288

18.

Martins

Conceição

Paiva

J. A.

Simões

M. R.

Fernandes

(2014). Delirium recognition by family: European Portuguese validation study of the family confusion assessment method. Journal of the American Geriatrics Society, 62(9), 1748–1752. https://doi.org/10.1111/jgs.12973

19.

McHugh

M. L.

(2012). Interrater reliability: The kappa statistic. Biochemia Medica (Zagreb), 22(3), 276–282. https://doi.org/10.11613/BM.2012.031

20.

Puelle

M. R.

Kosar

C. M.

Schmitt

Jones

R. N.

Marcantonio

E. R.

Cooper

Inouye

S. K.

Saczynski

J. S.

(2015). The language of delirium: Keywords for identifying delirium from medical records. Journal of Gerontological Nursing, 41(8), 34–42. https://doi.org/10.3928/00989134-20150723-01

21.

Registered Nurses’ Association of Ontario. (2016). Delirium, dementia, and depression in older adults: Assessment and care .

22.

Rosgen

Krewulak

Demiantschuk

Ely

E. W.

Davidson

J. E.

Stelfox

H. T.

Fiest

K. M.

(2018). Validation of caregiver-centered delirium detection tools: A systematic review. Journal of the American Geriatrics Society, 66(6), 1218–1225. https://doi.org/10.1111/jgs.15362

23.

Shrestha

Fick

D. M.

(2023). Recognition of delirium superimposed on dementia: Is there an ideal tool? Geriatrics (Basel), 8(1), 22. https://doi.org/10.3390/geriatrics8010022

24.

Sousa

V. D.

Rojjanasrirat

(2011). Translation, adaptation and validation of instruments or scales for use in cross-cultural health care research: A clear and user-friendly guideline. Journal of Evaluation in Clinical Practice, 17(2), 268–274. https://doi.org/10.1111/j.1365-2753.2010.01434.x

25.

Steis

M. R.

Evans

Hirschman

K. B.

Hanlon

Fick

D. M.

Flanagan

Inouye

S. K.

(2012). Screening for delirium using family caregivers: Convergent validity of the family confusion assessment method and interviewer-rated confusion assessment method. Journal of the American Geriatrics Society, 60(11), 2121–2126. https://doi.org/10.1111/j.1532-5415.2012.04200.x

26.

Steis

M. R.

Fick

D. M.

(2008). Are nurses recognizing delirium? A systematic review. Journal of Gerontological Nursing, 34(9), 40–48. https://doi.org/10.3928/00989134-20080901-12

27.

Titlestad

Haugarvoll

Solvang

S.-E. H.

Norekvål

T. M.

Skogseth

R. E.

Andreassen

O. A.

Årsland

Neerland

B. E.

Nordrehaug

J. E.

Tell

G. S.

Giil

L. M.

(2024). Delirium is frequently underdiagnosed among older hospitalised patients despite available information in hospital medical records. Age and Ageing, 53(2), afae006. https://doi.org/10.1093/ageing/afae006

28.

Vandenbroucke

J. P.

von Elm

Altman

D. G.

Gøtzsche

P. C.

Mulrow

C. D.

Pocock

S. J.

Poole

Schlesselman

J. J.

Egger

(2007). Strengthening the reporting of observational studies in epidemiology (STROBE): Explanation and elaboration. PLoS Medicine, 4, e297. https://doi.org/10.1371/journal.pmed.0040297

29.

Wand

A. P.

Thoo

Sciuriaga

Ting

Baker

Hunt

G. E.

(2014). A multifaceted educational intervention to prevent delirium in older inpatients: A before and after study. International Journal of Nursing Studies, 51(7), 974–982. https://doi.org/10.1016/j.ijnurstu.2013.11.005

30.

Wild

Grove

Martin

Eremenco

McElroy

Verjee-Lorenz

Erikson

(2005). Principles of good practice for the translation and cultural adaptation process for patient-reported outcomes (PRO) measures: Report of the ISPOR task force for translation and cultural adaptation. Value in Health, 8(2), 94–104. https://doi.org/10.1111/j.1524-4733.2005.04054.x

31.

Wilson

J. E.

Mart

M. F.

Cunningham

Shehabi

Girard

T. D.

MacLullich

A. M.

Slooter

A. J.

Ely

E. W.

(2020). Delirium. Nature Reviews Disease Primers, 6(1), 90. https://doi.org/10.1038/s41572-020-00223-4

32.

Xing

Zhu

Liu

Xia

Tong

(2024). An online delirium detection tool: Cross-cultural adaptation of a Chinese version of the Family Confusion Assessment Method. Australian Critical Care, 37(5), 740–746. https://doi.org/10.1016/j.aucc.2024.01.010

33.

Zhou

Wang

Zhou

(2024). Diagnostic accuracy of the Family Confusion Assessment Method for delirium detection: A systematic review and meta-analysis. Journal of the American Geriatrics Society, 72(3), 892–902. https://doi.org/10.1111/jgs.18692

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.10 MB