Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Jul 1.
Published in final edited form as: Clin Gerontol. 2022 Aug 2;46(4):574–584. doi: 10.1080/07317115.2022.2106927

The Modified Caregiver Strain Index in Black and White Dementia Caregivers at Hospital Discharge

Ashley Kuzmik 1, Marie Boltz 2, Rhonda BeLue 3, Barbara Resnick 4, Joanie Scott 5, Jacqueline Mogle 6, Douglas Leslie 7, James E Galvin 8
PMCID: PMC9892362  NIHMSID: NIHMS1833012  PMID: 35916440

Abstract

Objectives:

This study aimed to examine psychometric properties of the Modified Caregiver Strain Index (MCSI) in Black and White caregivers of persons living with dementia at hospital discharge.

Methods:

This was a cross-sectional study using baseline data of 423 family caregivers recruited from a cluster randomized clinical control trial. Factor structure, measurement invariance, and concurrent validity of the MCSI was analyzed. The moderating role of race on the relationship between MCSI score and anxiety, depression, burden was also examined.

Results:

The two-factor model fit the data best and was invariant across race. Regarding concurrent validity, higher MCSI scores were significantly associated with higher scores on the (HADS-A; anxiety), (HADS-D; depression), (ZBI; burden). Race moderated the relationship between MCSI score and anxiety, depression, and burden.

Conclusions:

The MCSI is a valid tool to assess caregiver strain in Black and White caregivers of persons living with dementia during hospital discharge. Results suggest that the effect of MCSI score on anxiety, depression, and burden varies by race.

Clinical Implications:

MCSI can be used by clinicians and service providers to help support the needs of Black and White caregivers of persons living with dementia during post-hospital transition.

Keywords: strain, caregivers, dementia, hospital discharge, psychometrics

Introduction

There are more than 6.2 Americans living with Alzheimer’s disease and related dementias (ADRD), and this number could grow to 13 million by 2050 (Alzheimer’s Association, 2021). Black Americans are almost twice as likely to have ADRD compared to White Americans (Alzheimer’s Association, 2021). The majority of people living with dementia continue to reside in the community and are cared for by family members such as spouses and children (Kasper et al., 2015; Alzheimer’s Association, 2021). Caring for persons living with dementia is often more challenging compared to other caregivers due to the intensity and long-term support of the disease (Kapoor et al., 2020; Ory et al., 1999). Persons living with dementia require more intense supervision, have more behavioral and psychological symptoms (i.e., agitation, anger, and depression), and generally need more assistance in activities of daily living and instrumental activities of daily living (Kales et al., 2015; Alzheimer’s Association, 2021). On average, caregivers of persons living with dementia provide more hours of care compared to non-dementia caregivers (Alzheimer’s Association, 2021).

Given the prolonged period of intense care, caregivers of persons living with dementia are at increased risk of greater levels of caregiver strain than caregivers of people without dementia (Capistrant, 2016; Ory et al., 1999). Prior studies have found that caregiver strain among persons living with dementia are associated with caregiver anxiety (Boltz et al., 2015a), depression (Jennings et al., 2015), and burden (Zhang et al., 2018). In addition, relative to White caregivers of persons living with dementia, Black caregivers often spend more time providing care (Alzheimer’s Association, 2021), report greater care demands (Rote et al., 2019), and experience more financial burden (R. Liu et al., 2021), yet, indicate higher psychological well-being (C. Liu et al., 2020).

Caregivers in the Hospital

Persons living with dementia have two to four times more hospitalizations than persons living without dementia (Bynum et al., 2004; Feng et al., 2014; Phelan et al., 2012). In the hospital setting, caregivers of persons living with dementia experience increased strain and anxiety due to increased caregiver demands and serious disruption to their caregiver routine (Boltz et al., 2015a, 2015b). Caregivers of persons living with dementia express worry about their relative’s illness and the associated consequences from the hospitalization, including increased care dependency (Bronson & Toye, 2015; Toye et al., 2019). During post-hospital transition, caregivers have reported lack of engagement in discharge planning process, and lack of preparedness, further contributing to their stress (Dewing & Dijk, 2016; Mockford, 2015). Evaluating caregiver strain at hospital discharge is needed to help support the needs of caregivers of persons living with dementia during the transitional period after hospitalization (Boltz et al., 2015a, 2015b).

Racial disparities could influence caregiver strain among caregivers of persons with dementia in the hospital setting. For example, compared to White caregivers of persons with dementia, greater care-related responsibilities in Black caregivers along with intensified stress in the hospital environment may increase their risk for health problems since Blacks tend to develop age-related conditions earlier than Whites (Thorpe et al., 2016). Additionally, Black caregivers of persons with dementia have historically experienced unique race-stressors in the healthcare setting such as poor communication with providers and increased financial strain compared to White caregivers (Alexander et al., 2022). Therefore, understanding racial differences in caregiver strain among caregivers of persons living with dementia at hospital discharge may help identify specific cultural needs that will assist post-hospital transition.

Caregiver Strain Index

The Caregiver Strain Index (CSI) has been widely used to assess perceived caregiver strain. It was originally developed by Robinson (1983) to evaluate strain in caregivers among older adults with heart disease or hip fracture upon discharge from the hospital. The initial CSI (dichotomous response; yes/no) had evidence of internal consistency (α = 0.86) and construct validity for three domains of strain (perception of caregiving, care-recipient characteristics, and emotional status). Similarly, in an exploratory factor analysis of the CSI, Rubio et al. (1999) identified three factors of strain (inconvenience, adjustment, and upsetting behavior) which were comparable to the areas of the strain reported in the original scale. In another study, exploratory factor analysis of the CSI among caregivers of cancers patients yielded four factors of strain (adaption, upsetting, inconvenience, and overwhelm), and acceptable validity (W = 0.41) and test-retest reliability (r = 0.75) was demonstrated (Ugur & Fadiloğlu, 2010).

Modified Caregiver Strain Index

The CSI was later developed into a modified version (MCSI; Thornton & Travis, 2003) which included a middle response category “sometimes” to better evaluate long-term caregiver experiences. The internal reliability of the MCSI (α = 0.90) was higher than the original CSI. The MCSI has demonstrated adequate internal consistencies in various populations including caregivers of persons with Parkinson’s disease in Malaysia (α = 0.75; Ramasamy et al., 2017) and women during puerperium in Spain (α = 0.71; Feligreras-Alcalá et al., 2021). However, no validation study of the MCSI has been conducted among caregivers of persons living with dementia.

Only two studies have examined the MCSI’s psychometric properties including the factorial structure, but the findings were inconsistent. Chan et al. (2013) identified a one-factor model of the MCSI in Chinese caregivers of patients with chronic illnesses and Ribeiro et al. (2021) revealed a two-factor structure of the scale in Portuguese caregivers of dependent older adults. These studies did not include measurement invariance across race; thus, equivalence of the MCSI across racial groups is limited.

The purpose of this study was to investigate the psychometric properties of the MCSI in Black and White caregivers of persons living with dementia during hospital discharge. Specifically, we analyzed (1) factor structure (i.e., one-factor model and two-factor model) and measurement invariance across race; (2) concurrent validity; and (3) moderating role of race on the relationship between MCSI score and anxiety, depression, burden. We hypothesized that: (1) the two-factor model would best fit the data and measurement invariance will be supported; (2) MCSI would be significantly associated with caregiver anxiety, depression, and burden; and (3) race will moderate the relationship between MCSI and anxiety, depression, and burden.

Methods

Study design, sample, and setting

This secondary analysis was a cross-sectional study using baseline data of 423 family caregivers recruited from a cluster randomized clinical control trial (ClinicalTrials.gov identifier: NCT03046121). This trial tested the efficacy of Family-centered Function-focused Care (Fam-FFC) on patient and caregiver-centered outcomes in community dwelling hospitalized patients with dementia (Boltz et al., 2018). Ethical approval was obtained from the university institutional review board. Prior to data collection, all participants provided written informed consent. Data was collected between November 2017 through August 2021.

The sample included family caregivers of hospitalized patients from six medical units in three Pennsylvania hospitals (two units per hospital). One hospital, a community teaching hospital, was located in an urban setting with a racially diverse population. Another small, non-teaching, community hospital served primarily White suburban residents. The third hospital, a large academic medical center served a racially diverse, primarily rural population. Family caregivers were eligible to participate if they (1) were 18 years or older; (2) spoke English or Spanish; (3) related to the patient by blood, marriage, adoption, or affinity as a significant other (defined so by the patient/legally authorized person as the primary person providing oversight and support on an ongoing basis). Family caregivers were excluded from the study if they (1) were unable to recall at least two words of a three-word recall.

Hospitalized patients were eligible to participate if they (1) were 65 years or older; (2) spoke English or Spanish; (3) had a family caregiver; (4) screened positive for dementia on the Montreal Cognitive Assessment (MoCA; ≤ 25) (Nasreddine et al., 2005) (5) had a score of ≥ 2 on the AD8 Dementia Screening Interview (Galvin et al., 2006); (6) had a diagnosis of very mild to moderate stage dementia confirmed by a score of 0.5 to 2.0 on the Clinical Dementia Rating Scale (CDR) (Morris, 1997); (7) indicated impaired function based on the Functional Activities Questionnaire (FAQ; ≥ 9) (Pfeffer et al., 1982). Patients were excluded from the study if they (1) were admitted from a nursing home; (2) were enrolled in Hospice; (3) had no family caregiver to participate; (4) had a significant neurological condition associated with cognition other than dementia (e.g., brain tumor); (5) had a major acute psychiatric disorder.

Procedures

Family caregiver data collection was completed by trained and experienced research evaluators. Data was collected within 72 hours of the patient’s hospital discharge at the caregiver’s home, or via electronic survey or telephone.

Additionally, part of the intervention in the parent study was that the nurse interventionist provided education and support related to the care of the person with dementia for 60 days on a weekly basis. The nurse interventionist also referred caregivers to appropriate resources that included psychological care.

Measures

Descriptive data of the family caregivers included age, gender, race, ethnicity, education, and marital status. Caregivers’ relationship with patient, living status with patient, employment status outside home, and the number of hours worked outside of home were also collected.

Modified Caregiver Strain Index

Caregiver strain was assessed using the 13-item Modified Caregiver Strain Index (MCSI), which includes five domains associated with caregiver strain such as financial, physical, psychological, social, and personal (Thornton & Travis, 2003). Each item is scored from 0 to 2 (0= “no”, 1= “yes, sometimes”, and 2= “yes, on a regular basis”), and total scores can range from 0 to 26. Any positive response may indicate a need for intervention in that area. Higher scores on the MCSI indicate greater caregiver strain; a score of 7 or higher indicates a high level of stress. The MCSI has good psychometric properties: internal reliability (α = 0.90) and test-retest reliability (r = 0.88) (Thornton & Travis, 2003).

Emotional Distress

The Hospital Anxiety and Depression subscale anxiety (HADS-A) and depression (HADS-D) (Herrmann, 1997) was used to measure caregiver anxiety and depression. Each subscale includes seven questions, and items are scored from 0 to 3. A total score of 0-7 was considered normal, 8-10 borderline abnormal, and 11-21 abnormal for each subscale. Abnormal scores indicated depression or anxiety. Prior testing of the HADS-A and HADS-D have provided evidence of good psychometric properties (Gough & Hudson, 2009).

Burden

Caregiver burden was evaluated using the Short Form Zarit Burden Interview (ZBI-12; (Ballesteros et al., 2012), a 12-item tool that measures two domains including personal strain and role strain. Items are scored from 0 (never) to 4 (almost always). Total scores range from 0 to 48; higher scores indicate greater levels of caregiver burden. Prior research has provided evidence of internal consistency and concurrent validity (Lin et al., 2017).

Data Analysis

SPSS and AMOS versions 25 were used for all statistical analyses (IBM Corp, Armonk, NY). Descriptive statistics were conducted to describe the sample and items of the MCSI. Data was examined for normality, outliers, and missing values. We determined overall factor structure by performing a confirmatory factor analysis (CFA) using maximum likelihood estimation procedure to test the one-factor model and two-factor model of the MCSI identified in prior studies (Chan et al., 2013; Ribeiro et al., 2021). The one-factor model tested all thirteen items of the MCSI loading onto a single factor. The two-factor model tested seven items loaded onto the “Individual Experiences of Burden” factor (first factor), while the other six items were loaded onto the “Repercussions on the Caregiver’s Life” factor (second factor). Multiple indices of goodness-of-fit were examined for model evaluation: comparative fit index (CFI), root mean square error of approximation (RMSEA), and standardized root mean square residual (SRMR). Cut-off values that have been previously recommended were used. Specifically, CFI > 0.90 is acceptable, RMSEA < 0.08 is reasonable, and SRMR < 0.08 is adequate model fit (Bentler, 1990, 1992; Hu & Bentler, 1999). We also reported the chi-square value (χ2), but due to its sensitivity to sample size it was not used to determine significance (Bentler & Bonett, 1980). To enhance goodness-of-fit of the model, modification indices were also considered.

To evaluate measurement invariance by race we compared factor level data, using multigroup CFA. We examined a series of nested models (configural, metric, and scalar invariance) to test measurement invariance by race: (1) configural invariance checked that the factor pattern/structure was equivalent across the groups; (2) metric invariance assessed the equivalence of factor loadings among the groups; (3) scalar invariance explored equivalence of item thresholds between the two groups. The following criteria was used to test invariance for each model: ΔCFI ≤ 0.01 and ΔRMSEA ≤ 0.015 (Cheung & Rensvold, 2002; Chen, 2007). The Δχ2 values were reported, but not further interpreted due to high sensitivity to large samples.

To assess concurrent validity we performed three separate linear regression models to determine if the MCSI scores were significantly associated with anxiety (HADS-A), depression (HADS-D), and burden (ZBI-12). Family caregiver’s gender, age, education and living status were controlled for in the model. Race and length of hospital stay were not significantly associated with the outcome variables, therefore, were not controlled for in the model. A p-value ≤ 0.05 was considered statistically significant.

To investigate moderation by race we conducted three multiple regression analyses on the association between MCSI score and anxiety (HADS-A), depression (HADS-D), and burden (ZBI-12). Each regression model included an interaction term (MCSI score*race). The moderating effect was determined by a significant interaction term. We controlled for the same covariates as in the models testing concurrent validity. Continuous predictor variables were mean centered. Plotted graphics were presented to better visualize the significant interactions. All moderation analyses were conducted using PROCESS macro for SPSS version 4.1 (Hayes, 2017).

Results

Sociodemographic Characteristics of Sample

As shown in Table 1, the majority of caregivers (N=423) were female (72%) and either non-Hispanic White (66%) or non-Hispanic Black (34%) with a mean age of 61.9 ± 14.0. Most caregivers were married (61%) and had some college or technical school (30%). Among the caregivers, 52% were the children of the patient, 60% lived with the patient, and 45% were employed outside the home averaging 40.2 ± 14.1 hours of outside work per week. Additionally, there was a statistically significant relationship between MCSI score in Black and White caregivers and their gender and education (p=0.02 and p=0.004, respectively). Item responses on the MCSI by race are included in Supplemental Table S1.

Table 1.

Sociodemographic characteristics of caregivers and Modified Caregiver Strain Index score (N=423)

Sociodemographic Characteristics Overall n (%) or mean ± SD MCSI Total mean ± SD
Black White P-Value
Gender Female 306 (72) 8.39 ± 6.90 8.25 ± 6.29 0.02
Male 117 (28) 8.37 ± 6.91 5.86 ± 5.45
Race White 279 (66) 8.42 ± 6.44 7.61 ± 6.62 0.38
Black 144 (34) - -
Ethnicity Not Hispanic or Latino 409 (97) 8.24 ± 6.45 7.58 ± 6.55 0.17
Hispanic or Latino 14 (3) 14.0 ± 0 10.0 ± 7.88
Education Less than high school 33 (8) 6.45 ± 6.13 7.08 ± 4.42 0.004
High school graduate 106 (25) 6.94 ± 5.38 5.63 ± 5.06
Some college or technical school 125 (30) 9.70 ± 6.74 7.99 ± 7.06
College graduate 90 (21) 9.50 ± 7.42 8.36 ± 7.0
Graduate or professional school 61 (14) 7.08 ± 6.68 9.06 ± 7.42
Marital status Married 259 (61) 7.97 ± 6.69 9.23 ± 6.85 0.21
Single 74 (17) 5.57 ± 6.24 7.37 ± 5.65
Divorced 47 (11) 7.54 ± 7.10 7.62 ± 6.03
Widowed 29 (7) 7.50 ± 6.0 6.15 ± 5.93
Separated 9 (2) 6.25 ± 5.38 15.0 ± 8.0
Relationship with patient Son/daughter 218 (52) 9.20 ± 6.45 7.79 ± 6.68 0.24
Spouse 124 (29) 8.26 ± 7.58 7.61 ± 6.75
Nephew/niece 15 (4) 8.86 ± 6.49 10.13 ± 6.81
Brother/sister 17 (4) 6.38 ± 5.32 9.67 ± 6.42
Granddaughter/grandson 13 (3) 7.40 ± 4.90 11.671 ± 9.02
Goddaughter 2 (<1) 0.0 ± 0 3.88 ± 3.94
Friend 8 (2) 2.80 ± 1.64 4.0 ± 5.29
Daughter-in-law 9 (2) 0.0 ± 0 14.0 ± 0
Paid Caregiver 8 (2) 4.33 ± 2.31 6.20 ± 4.97
Partner 6 (1) 7.0 ± 9.90 3.25 ± 4.72
Other (e.g., sister in law, ex, cousin) 3 (1) 10.50 ± 7.78 13.0 ± 0
Living with Patient 253 (60) 8.09 ± 6.69 8.05 ± 7.0 0.52
Employed outside of home 189 (45) 8.63 ± 6.37 8.19 ± 7.0 0.20
Age 61.93 ± 14.0 58.84 ± 14.41 63.50 ± 13.55 -
Hours of employment outside of home (per week) 40.20 ± 14.10 40.71 ± 15.21 39.91 ± 13.43 -
Open in a new tab

Confirmatory Factor Analysis

As shown in Table 2, the one-factor model indicated poor model fit on all indices except for SRMR: CFI = 0.851, RMSEA = 0.117, and SRMR = 0.039. The two-factor model showed better model fit, but RMSEA was still high: CFI = 0.915, RMSEA = 0.084, and SRMR = 0.029. Modification indices indicated improvement in model fit with two correlated residuals between items 4 and 12 and items 9 and 10. The modification indices did not alter other parameters of the model; thus, the items were allowed to correlate. This modified model exhibited acceptable fit among all indices: CFI = 0.932, RMSEA = 0.076, and SRMR = 0.027. All standardized factor loadings of the modified two-factor model were statistically significant, and values ranged from 0.52 to 0.75 (Figure 1)

Table 2.

Confirmatory Factor Analysis of Modified Caregiver Strain Index with Fit Indices (N=423)

Model χ2 df CFI RMSEA SRMR
One-factor Model 310.337 65 0.851 0.117 0.039
Two-factor model 252.908 64 0.915 0.084 0.029
Final Model
Two-factor model with error covariance* 213.170 62 0.932 0.076 0.027
Open in a new tab

Note: Two pairs of error covariance between items 4 and 12, and items 9 and 10

Figure 1.

Figure 1.

Open in a new tab

Two-Factor Confirmatory Analysis of Modified Caregiver Strain Index with Factor Loadings

Note: Factor_1 = Individual Experiences of Burden; Factor_2 = Repercussions on the Caregiver’s Life”

Measurement Invariance by Race

Table 3 shows the results from the measurement invariance analysis by race. The configural model had satisfactory fit statistics: CFI = 0.916 and RMSEA = 0.061. When comparing the metric and scalar models, change in model fit were below the cutoffs: ΔCFIs (≤0.01) and ΔRMSEAs (≤0.015).

Table 3.

Model Fit Indices for Measurement Invariance of Modified Caregiver Strain Index (N=423)

χ2 df CFI RMSEA Δχ2 ΔCFI ΔRMSEA
Configural Invariance 317.975 124 0.916 0.061 - - -
Metric Invariance 337.761 135 0.912 0.060 19.786 0.004 0.001
Scalar Invariance 351.293 138 0.908 0.061 13.532 0.004 0.001
Open in a new tab

Concurrent Validity

As shown in Table 4, after controlling for family caregiver’s gender, age, education and living status, higher MCSI scores were significantly associated with higher scores on the HADS-A (β = .57, t = 13.89(418), p<.001), HADS-D (β = .54, t = 12.96(418), p<.001), and ZBI (β = .77, t = 23.95(418), p<.001).

Table 4.

Regression models for predication of outcome measures (N=423)

B β t (p) 95% Confidence Level
Lower Upper
HADS-A .40 .57 13.89 (<.001) .35 .46
HADS-D .31 .54 12.96 (<.001) .39 1.71
ZBI 1.09 .77 23.95 (<.001) 1.0 1.18
Open in a new tab

Note: HADS-A = Hospital Anxiety and Depression subscale for Anxiety; HADS-D = Hospital Anxiety and Depression subscale for Depression; ZBI = Zarit Burden Interview

Moderation Analysis

Table 5 presents the results from the moderation analysis. The interaction between MCSI and race was a significant predictor of anxiety (HADS-A; B = −0.13, p = 0.03), depression (HADS-D; B = −0.10, p = 0.04), and burden (ZBI; B = −0.27, p = 0.01); thus, moderation was supported in each model. Figure 2 (A, B, and C, respectively) shows that White caregivers with higher MCSI scores reported higher levels anxiety, depression, and burden compared to Black caregivers.

Table 5.

Moderation Analysis

Variable HADS-Ab B (SE) HADS-Dc B (SE) ZBId B (SE)
MCSI Score 0.58 (0.09)* 0.44 (0.07)* 1.45 (0.14)*
Race −0.76 (0.40) −0.23 (0.33) −0.60 (0.63)
MCSI x Race −0.13 (0.06)* −0.10 (0.05)* −0.27 (0.10)*
Controls
FCGa Gender 0.06 (0.42) 0.07 (0.34) −0.76 (0.66)
FCGa Age −0.02 (0.01) 0.03 (0.01)* 0.01 (0.02)
FCGa Education −0.26 (0.14) 0.001 (0.12) 0.10 (0.22)
FCGa Living Status 0.86 (0.41) 0.98 (0.34)* 0.96 (0,64)
Open in a new tab

Note:

*

p < 0.05

a.

FCG: Family Caregiver

b.

Dependent Variable: HADS-A = Hospital Anxiety and Depression subscale for Anxiety

c.

Dependent Variable: HADS-D = Hospital Anxiety and Depression subscale for Depression

d.

Dependent Variable: ZBI = Zarit Burden Interview

Figure 2.

Figure 2.

Open in a new tab

Visual representation of the moderation of Modified Care Strain Index by race on anxiety, depression, and burden

Note: *MCSI score was mean centered

Discussion

This study aimed to examine the factor structure, measurement invariance, and validity of the MCSI in a sample of caregivers of persons living with dementia at hospital discharge. Our study is the first psychometric testing of the MCSI in Black and White caregivers of persons living with dementia during the post-hospitalization period. The findings supported our hypothesis that the two-factor model fit the data best. Few studies have examined the factorial structure of the MCSI, and the factors found are inconsistent with one study showing two factors (Ribeiro et al., 2021) and another study suggesting one factor (Chan et al., 2013). Our results further support the two-factor structure of the MCSI, future studies should examine the factorial structure of the MCSI in samples of more diverse caregivers in varied settings to further confirm the two factors.

As hypothesized, the multigroup CFA showed that the MCSI was invariant across race, confirming measurement invariance. The results demonstrated that configural and metric invariances supported the data, showing the same factor pattern (i.e., two-factor structure) and equivalence of factor loadings across the groups. In addition, scalar invariance was established, providing evidence that the item thresholds between the two groups are equivalent. Since different types of caregiver strain vary across cultural groups (e.g., financial burden and emotional burden) (C. Liu et al., 2020; R. Liu et al., 2021), future research is recommended to further assess the MCSI item responses across other racial and ethnic groups.

Regarding concurrent validity, the MCSI scores predicted anxiety, depression, and burden, supporting our hypothesis. Specifically, higher levels of caregiver strain were significantly associated with higher levels of caregiver anxiety, depression, and burden. The results are consistent with other studies where caregiver strain was associated with caregiver anxiety (Boltz et al., 2015a), depression (Jennings et al., 2015), and burden (Zhang et al., 2018). Future research should continue to examine caregiver strain across different cultures and populations.

Confirming our hypothesis, results from the moderation analysis indicated that race moderated the relationship between MCSI score and anxiety, depression, and burden. Our findings showed that White caregivers with higher MCSI scores had more anxiety, depression, and burden compared to Black caregivers. Prior literature demonstrated that Black caregivers had less psychological comorbidity compared to White caregivers (C. Liu et al., 2020). Future research should examine other factors that could influence caregiver strain such as the caregiver’s physical health, access to resources, and socioeconomic status. Additionally, future research is warranted to confirm whether the findings of this study are generalizable to the wider caregiver population.

In this study, family caregivers reported significant levels of strain. This finding, which is consistent with other research, underscores the advisability of assessing caregiver strain upon discharge of the hospitalized person with dementia. An evaluation of strain and management of contributing factors may support caregiver efficacy and promote the well-being of both members of the dyad.

Limitations

Limitations of this study include recruitment of participants in only four hospitals from a single state. Our sample also lacked representation of other ethnicities including Hispanics. Additionally, since patients with severe dementia were not included in our sample, the results may not be generalizable to all caregivers of persons living with dementia. The MCSI is a self-reported instrument and data collection took place via electronic survey or telephone, thus, the findings may have been biased due to social desirability. For example, participants may have minimized their levels of caregiver strain to reduce concerns and make them appear as better caregivers via telephone, whereas participants who completed the electronic survey (i.e., no potential influencer) may have reported more accurate levels of caregiver strain. Further, the three-word recall is a reductive and broad form for measuring the mental state of the caregiver and could induce bias. Lastly, because this study was a secondary analysis, we did not have access to other potentially salient variables which could affect caregiver strain such as caregiver’s living environments, physical health status, and socioeconomic status.

Conclusions

Racial disparities in caregiver strain exist among caregivers of persons living with dementia in the hospital setting. For example, Black caregivers experience greater financial strain compared to White caregivers (Alexander et al., 2022). Thus, it is critical to establish the validity of the MCSI to target culturally specific areas of caregiver strain that need most support during transition from hospital to home. Despite limitations, this study provides support for the MCSI as a valid instrument to assess caregiver strain in Black and White caregivers of persons living with dementia at the time of hospital discharge. Our findings further support the two-factor model and provide evidence of measurement invariance of the MCSI in Black and White caregivers of persons living with dementia. Concurrent validity of the MCSI was also established, demonstrating that this questionnaire can be used to identify caregiver strain among Black and White caregivers of persons living with dementia at hospital discharge. Specifically, the concurrent validity analyses of the MCSI are important to provide relevant data specific to Black and White caregivers needs that will guide meaningful partnerships with clinicians and service providers. In addition, the moderating role of race on the relationship between MCSI score and anxiety, depression and burden indicate that culturally tailored approaches are needed to reduce or manage caregiver strain in Black and White caregivers of persons living with dementia during hospital discharge. Overall, results suggest that the MCSI is a valid and appropriate tool that can be used by clinicians and service providers to help support the needs of Black and White caregivers of persons living with dementia during post-hospital transition. Future research is warranted to further test the psychometric data of the MCSI on diverse populations such as caregivers of persons living with dementia from multiple cultures and geographic regions.

Supplementary Material

Supplemental Table 1

Clinical Implications.

  • The MCSI is a valid and useful tool to evaluate caregiver strain in Black and White caregivers of persons with dementia at hospital discharge.

  • Findings support the two-factor structure and measurement invariance across race of the MCSI.

  • Race moderated the relationship between MCSI score and anxiety, depression, and burden; White caregivers with higher MCSI scores demonstrated more anxiety, depression, and burden compared to Black caregivers.

Funding:

This study was supported by the National Institute of Aging (NIA) under grant R01AG054425. The contents of the article are solely the responsibility of the authors and do not necessarily represent the official views of the NIH/NIA.

Footnotes

Disclosure statement: No potential conflict of interest was reported by the author(s).

Contributor Information

Ashley Kuzmik, Pennsylvania State University, Ross and Carol Nese College of Nursing, 306 Nursing Sciences Building, University Park, PA 16802.

Marie Boltz, Pennsylvania State University, Ross and Carol Nese College of Nursing, 306 Nursing Sciences Building, University Park, PA 16802.

Rhonda BeLue, Department of Public Health, University of Texas at San Antonio, San Antonio, Texas 78249.

Barbara Resnick, University of Maryland, School of Nursing, 655 West Lombard Street, Baltimore, MD 21201.

Joanie Scott, Pennsylvania State University, Ross and Carol Nese College of Nursing, Pennsylvania State University, University Park, PA 16802.

Jacqueline Mogle, Department of Psychology, Clemson University, Clemson, South Carolina 29634.

Douglas Leslie, Department of Public Health Sciences, Pennsylvania State University, Hershey, PA 17033.

James E. Galvin, Comprehensive Center for Brain Health, University of Miami, Miller School of Medicine, Boca Raton, FL 33433.

Data Availability Statement:

The data presented in this study are available on request from the corresponding author.

References

  1. Alzheimer’s Association. 2021. Alzheimer’s disease facts and figures. Alzheimer’s Dementia, 17(3). https://www.alz.org/media/documents/alzheimers-facts-and-figures.pdf [DOI] [PubMed] [Google Scholar]
  2. Alexander K, Oliver S, Bennett SG, Henry J, Hepburn K, Clevenger C, & Epps F (2022). “Falling between the cracks”: Experiences of Black dementia caregivers navigating U.S. health systems. Journal of the American Geriatrics Society, 70(2), 592–600. 10.1111/jgs.17636 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Ballesteros J, Santos B, González-Fraile E, Muñoz-Hermoso P, Domínguez-Panchón AI, & Martín-Carrasco M (2012). Unidimensional 12-item Zarit Caregiver Burden Interview for the assessment of dementia caregivers’ burden obtained by item response theory. Value in Health: The Journal of the International Society for Pharmacoeconomics and Outcomes Research, 15(8), 1141–1147. 10.1016/j.jval.2012.07.005 [DOI] [PubMed] [Google Scholar]
  4. Bentler PM (1990). Comparative fit indexes in structural models. Psychological Bulletin, 107, 238–246. 10.1037/0033-2909.107.2.238 [DOI] [PubMed] [Google Scholar]
  5. Bentler PM (1992). On the fit of models to covariances and methodology to the Bulletin. Psychological Bulletin, 112(3), 400–404. 10.1037/0033-2909.112.3.400 [DOI] [PubMed] [Google Scholar]
  6. Bentler PM, & Bonett DG (1980). Significance tests and goodness of fit in the analysis of covariance structures. Psychological Bulletin, 88(3), 588–606. 10.1037/0033-2909.88.3.588 [DOI] [Google Scholar]
  7. Boltz M, Chippendale T, Resnick B, & Galvin JE (2015a). Anxiety in family caregivers of hospitalized persons with dementia: Contributing factors and responses. Alzheimer Disease and Associated Disorders, 29(3), 236–241. 10.1097/WAD.0000000000000072 [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Boltz M, Chippendale T, Resnick B, & Galvin JE (2015b). Testing family-centered, function-focused care in hospitalized persons with dementia. Neurodegenerative Disease Management, 5(3), 203–215. 10.2217/nmt.15.10 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Boltz M, Kuzmik A, Resnick B, Trotta R, Mogle J, BeLue R, Leslie D, & Galvin JE (2018). Reducing disability via a family centered intervention for acutely ill persons with Alzheimer’s disease and related dementias: Protocol of a cluster-randomized controlled trial (Fam-FFC study). Trials, 19(1), 496. 10.1186/s13063-018-2875-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Bronson M, & Toye C (2015). Providing information for family carers of hospital patients experiencing dementia. Dementia, 14(2), 267–272. 10.1177/1471301214560240 [DOI] [PubMed] [Google Scholar]
  11. Bynum JPW, Rabins PV, Weller W, Niefeld M, Anderson GF, & Wu AW (2004). The relationship between a dementia diagnosis, chronic illness, medicare expenditures, and hospital use. Journal of the American Geriatrics Society, 52(2), 187–194. 10.1111/j.1532-5415.2004.52054.x [DOI] [PubMed] [Google Scholar]
  12. Capistrant BD (2016). Caregiving for Older Adults and the Caregivers’ Health: An Epidemiologic Review. Current Epidemiology Reports, 3(1), 72–80. 10.1007/s40471-016-0064-x [DOI] [Google Scholar]
  13. Chan WCH, Chan CLF, & Suen M (2013). Validation of the Chinese version of the Modified Caregivers Strain Index among Hong Kong caregivers: An initiative of medical social workers. Health & Social Work, 38(4), 214–221. 10.1093/hsw/hlt021 [DOI] [PubMed] [Google Scholar]
  14. Dewing J, & Dijk S (2016). What is the current state of care for older people with dementia in general hospitals? A literature review. Dementia (London, England), 15(1), 106–124. 10.1177/1471301213520172 [DOI] [PubMed] [Google Scholar]
  15. Feligreras-Alcalá D, Cazalilla-López M. del P., del-Pino-Casado R, & Frías-Osuna A (2021). Validity and Reliability of the Caregiver Strain Index Scale in Women during the Puerperium in Spain. International Journal of Environmental Research and Public Health, 18(7), 3602. 10.3390/ijerph18073602 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Feng Z, Coots LA, Kaganova Y, & Wiener JM (2014). Hospital and ED use among Medicare beneficiaries with dementia varies by setting and proximity to death. Health Affairs (Project Hope), 33(4), 683–690. 10.1377/hlthaff.2013.1179 [DOI] [PubMed] [Google Scholar]
  17. Galvin JE, Roe CM, Xiong C, & Morris JC (2006). Validity and reliability of the AD8 informant interview in dementia. Neurology, 67(11), 1942–1948. 10.1212/01.wnl.0000247042.15547.eb [DOI] [PubMed] [Google Scholar]
  18. Gough K, & Hudson P (2009). Psychometric Properties of the Hospital Anxiety and Depression Scale in Family Caregivers of Palliative Care Patients. Journal of Pain and Symptom Management, 37(5), 797–806. 10.1016/j.jpainsymman.2008.04.012 [DOI] [PubMed] [Google Scholar]
  19. Hayes AF (2017). Introduction to Mediation, Moderation, and Conditional Process Analysis, Second Edition: A Regression-Based Approach. Guilford Publications. [Google Scholar]
  20. Hu L, & Bentler PM (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling: A Multidisciplinary Journal, 6(1), 1–55. 10.1080/10705519909540118 [DOI] [Google Scholar]
  21. Jennings LA, Reuben DB, Evertson LC, Serrano KS, Ercoli L, Grill J, Chodosh J, Tan Z, & Wenger NS (2015). Unmet needs of caregivers of individuals referred to a dementia care program. Journal of the American Geriatrics Society, 63(2), 282–289. 10.1111/jgs.13251 [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Kales HC, Gitlin LN, & Lyketsos CG (2015). Assessment and management of behavioral and psychological symptoms of dementia. BMJ (Clinical Research Ed.), 350, h369. 10.1136/bmj.h369 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Kapoor A, Kapoor A, Manoharan SVRR, Streim J, & Mavandadi S (2020). CAREGIVER-RELATED FACTORS ASSOCIATED WITH THE DESIRE TO INSTITUTIONALIZE CARE RECIPIENTS WITH DEMENTIA. The American Journal of Geriatric Psychiatry, 28(4), S86–S87. 10.1016/j.jagp.2020.01.113 [DOI] [Google Scholar]
  24. Kasper JD, Freedman VA, Spillman BC, & Wolff JL (2015). The Disproportionate Impact Of Dementia On Family And Unpaid Caregiving To Older Adults. Health Affairs, 34(10), 1642–1649. 10.1377/hlthaff.2015.0536 [DOI] [PMC free article] [PubMed] [Google Scholar]
  25. Lin C-Y, Wang J-D, Pai M-C, & Ku L-JE (2017). Measuring burden in dementia caregivers: Confirmatory factor analysis for short forms of the Zarit Burden Interview. Archives of Gerontology and Geriatrics, 68, 8–13. 10.1016/j.archger.2016.08.005 [DOI] [PubMed] [Google Scholar]
  26. Liu C, Badana ANS, Burgdorf J, Fabius CD, Roth DL, & Haley WE (2020). Systematic Review and Meta-Analysis of Racial and Ethnic Differences in Dementia Caregivers’ Well-Being. The Gerontologist, 61(5), e228–e243. 10.1093/geront/gnaa028 [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Liu R, Chi I, & Wu S (2021). Caregiving Burden Among Caregivers of People With Dementia Through the Lens of Intersectionality. The Gerontologist, gnab146. 10.1093/geront/gnab146 [DOI] [PubMed] [Google Scholar]
  28. Mockford C. (2015). A review of family carers’ experiences of hospital discharge for people with dementia, and the rationale for involving service users in health research. Journal of Healthcare Leadership, 7, 21–28. 10.2147/JHL.S70020 [DOI] [PMC free article] [PubMed] [Google Scholar]
  29. Morris JC (1997). Clinical dementia rating: A reliable and valid diagnostic and staging measure for dementia of the Alzheimer type. International Psychogeriatrics, 9 Suppl 1, 173–176; discussion 177-178. 10.1017/s1041610297004870 [DOI] [PubMed] [Google Scholar]
  30. Nasreddine ZS, Phillips NA, Bédirian V, Charbonneau S, Whitehead V, Collin I, Cummings JL, & Chertkow H (2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53(4), 695–699. 10.1111/j.1532-5415.2005.53221.x [DOI] [PubMed] [Google Scholar]
  31. Ory MG, Hoffman RR, Yee JL, Tennstedt S, & Schulz R (1999). Prevalence and impact of caregiving: A detailed comparison between dementia and nondementia caregivers. The Gerontologist, 39(2), 177–185. 10.1093/geront/39.2.177 [DOI] [PubMed] [Google Scholar]
  32. Pfeffer RI, Kurosaki TT, Harrah CH, Chance JM, & Filos S (1982). Measurement of functional activities in older adults in the community. Journal of Gerontology, 37(3), 323–329. 10.1093/geronj/37.3.323 [DOI] [PubMed] [Google Scholar]
  33. Phelan EA, Borson S, Grothaus L, Balch S, & Larson EB (2012). Association of Incident Dementia With Hospitalizations. JAMA, 307(2), 165–172. 10.1001/jama.2011.1964 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Ramasamy MS, Ibrahim R, Madon Z, & Arshat Z (2017). A Validity Study of Malay-translated Version of the Modified Caregivers Strain Index Questionnaire (M-CSI-M). Journal of Business and Social Review in Emerging Economies, 3(2), 217–226. [Google Scholar]
  35. Ribeiro O, Brandão D, Oliveira AF, Martín I, Teixeira L, & Paúl C (2021). The Modified Caregiver Strain Index: Portuguese version. Journal of Health Psychology, 26(10), 1597–1608. 10.1177/1359105319883933 [DOI] [PubMed] [Google Scholar]
  36. Robinson BC (1983). Validation of a Caregiver Strain Index. Journal of Gerontology, 38(3), 344–348. 10.1093/geronj/38.3.344 [DOI] [PubMed] [Google Scholar]
  37. Rote SM, Angel JL, Moon H, & Markides K (2019). Caregiving Across Diverse Populations: New Evidence From the National Study of Caregiving and Hispanic EPESE. Innovation in Aging, 3(2), igz033. 10.1093/geroni/igz033 [DOI] [PMC free article] [PubMed] [Google Scholar]
  38. Rubio DM, Berg-Weger M, & Tebb SS (1999). Assessing the validity and reliability of well-being and stress in family caregivers. Social Work Research, 23(1), 54–64. 10.1093/swr/23.1.54 [DOI] [Google Scholar]
  39. Thornton M, & Travis SS (2003). Analysis of the reliability of the modified caregiver strain index. The Journals of Gerontology. Series B, Psychological Sciences and Social Sciences, 58(2), S127–132. 10.1093/geronb/58.2.s127 [DOI] [PubMed] [Google Scholar]
  40. Thorpe RJ, Fesahazion RG, Parker L, Wilder T, Rooks RN, Bowie JV, Bell CN, Szanton SL, & LaVeist TA (2016). Accelerated Health Declines among African Americans in the USA. Journal of Urban Health : Bulletin of the New York Academy of Medicine, 93(5), 808–819. 10.1007/s11524-016-0075-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Toye C, Slatyer S, Quested E, Bronson M, Hill A, Fountaine J, Uren HV, Troeung L, & Maher S (2019). Obtaining information from family caregivers to inform hospital care for people with dementia: A pilot study. International Journal of Older People Nursing, 14(1), e12219. 10.1111/opn.12219 [DOI] [PubMed] [Google Scholar]
  42. Ugur O, & Fadiloğlu C (2010). “Caregiver Strain Index” validity and reliability in Turkish society. Asian Pacific Journal of Cancer Prevention: APJCP, 11(6), 1669–1675. [PubMed] [Google Scholar]
  43. Zhang M, Chang Y-P, Liu YJ, Gao L, & Porock D (2018). Burden and Strain among Familial Caregivers of Patients with Dementia in China. Issues in Mental Health Nursing, 39(5), 427–432. 10.1080/01612840.2017.1418034 [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Supplemental Table 1
NIHMS1833012-supplement-Supplemental_Table_1.docx (14.5KB, docx)

Data Availability Statement

The data presented in this study are available on request from the corresponding author.

RESOURCES