This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation

Development and Initial Validation of an Inventory to Assess Grief in Caregivers of Persons With Alzheimer's Disease

^a Department of Psychology, University of Missouri, St. Louis
^b Alzheimer's Disease Research Center, Department of Neurology, Washington University School of Medicine, St. Louis, MO

Correspondence: Thomas M. Meuser, PhD, Alzheimer's Disease Research Center, Washington University School of Medicine, 4488 Forest Park Ave., Suite 130, St. Louis, MO 63108. E-mail: meusert{at}abraxas.wustl.edu.

Decision Editor: Laurence G. Branch, PhD

	Abstract

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 
Purpose: The purpose of this study was to develop an empirically based, psychometrically sound instrument for the assessment of grief in caregivers of persons with Alzheimer's disease. Design and Methods: A total of 184 statements addressing personal grief reactions were obtained from 45 adult child and 42 spouse caregivers in 16 focus groups representing early, middle, late, and postdeath stages. These were presented to a second sample of 83 adult child and 83 spouse caregivers, who rated them according to their current experience. Also administered were the Beck Depression Inventory, the Anticipatory Grief Scale, the Caregiver Strain Index, the Caregiver Well-Being Scale–Basic Needs, and the Perceived Social Support–Family Questionnaire. Results: Factor analysis resulted in a 50-item scale containing three factors: Personal Sacrifice Burden, Heartfelt Sadness and Longing, and Worry and Felt Isolation. Cronbach's alpha scores ranged from .90 to .96, indicating high internal consistency reliability for each factor as well as for their combined total. Correlations of each factor with other measures provide evidence of validity, especially when assessed in the context of the caregiver grief model. Implications: Results suggest that caregiver grief is neither a unitary nor a static construct and that a scale such as this may be appropriate for use in supportive, clinical, and research settings.

Key Words: Alzheimer's disease • Dementia • Caregiver grief • Questionnaire

The recognition of grief as a definable and measurable concept pertaining to persons with Alzheimer's disease and their caregivers has only recently appeared in the professional literature. Since the first significant references to grief associated with dementia (Mace and Rabins 1981; Rabins 1984), only a handful of authors have addressed the concept (for current reviews, see Doka 2000, and Meuser and Marwit 2001), and even fewer have attempted to systematically study it. The majority literature has associated caregiver grief with the more frequently discussed concepts of caregiver depression, burden, stress, and coping.

Snyder 2001 summarized an emerging literature concerning the "rapidly growing awakening to the lived experience of Alzheimer's" (p. 8). She identified a trend toward qualitative research that uses interviews, video recordings, and narratives to identify common experiential themes. Although her work primarily addresses experiences of persons with early–mild Alzheimer's disease, others have used similar techniques to identify the communal themes of caregivers. Within this latter literature, a picture of caregiver grief has begun to emerge. For example, Farran, Keane-Hagerty, Salloway, Kupferer, and Wilken 1991, using an interview procedure, quantified a number of grief-related losses. Feeling the loss of the relationship was most frequently endorsed (occurring in 52% of their sample), followed by the loss of familiar communication (32%), the loss of personal freedom (31%), the loss of enjoying future planning (30%), the loss of the care receiver's mental capacities (29%), and simply mourning losses in general (42%). Similarly, Loos and Bowd 1997 identified four grief-related themes among dementia caregivers: loss of social and recreational interaction, loss of control over life events, loss of well-being, and loss of occupation. Others have noted, in more general ways, the existence of grief as a significant and universal experience among dementia caregivers (Bowd and Loos 1996; Walker and Pomeroy 1996, Walker and Pomeroy 1997; Walker, Pomeroy, McNeil, and Franklin 1994).

A few studies have attempted to look at caregiver grief at different points in the progression of the disease. Bass, Bowman, and Noelker 1991 and Mullan 1992 examined grief before and after the physical death of the care receiver, and Gilhooly, Sweeting, Whittick, and McKee 1994 looked at grief before and after the social death of the care recipient; that is, before and after the time at which the care recipient lost his or her capacity for meaningful social interaction. Each study had provided some evidence that a caregiver's predeath experience may affect his or her postdeath grief adjustment. Rudd, Viney, and Preston 1999 looked at home care versus nursing home care and concluded that there is a stronger grief response following nursing home placement, particularly for female caregivers. Ponder and Pomeroy 1996 proposed a curvilinear model of grief experience following the course of the disease such that grief intensity is high in the earliest stages of caregiving (approximately the first 2 years), decreases in the middle stages (next 2 years), and then rebounds (Year 5 and beyond).

Apart from these few stage-related studies, there has not been, until recently, any systematic investigation of grief across the entire course of the disease, including postdeath, or any study that has simultaneously taken into account the relationship of the care provider to the care receiver. Meuser and Marwit 2001 attempted to do this by means of a combined qualitative–quantitative methodology similar to that recently endorsed for grief research by Neimeyer and Hogan 2001. Their participants were dementia caregivers assigned to 1 of 16 2-hr, semistructured focus groups determined by their caregiver status (adult child, spouse) and by the care recipient's stage of disease progression (early, middle, late, or postdeath). Stage of active disease was determined by caregivers' filling out the informant interview portion of the Clinical Dementia Rating (CDR) Scale (Berg 1988; Hughes, Berg, Danziger, Coben and Martin 1982; Morris 1993), a semistructured interview that measures the six cognitive–functional domains of memory, orientation, judgment and problem solving, community affairs, home and hobbies, and personal affairs. Participants also completed the Anticipatory Grief Scale (Theut, Jordan, Ross, and Deutsch 1991) and a grief questionnaire specifically designed for their study. Their results indicated that there are significant and definable differences in the grief responses of adult child and spouse caregivers at every stage of the disease, including postdeath, and that predeath anticipatory grieving relieves short-term but not long-term postdeath grief.

The present study used direct statements from caregivers in early, middle, late, and postdeath stages who participated in Meuser and Marwit 2001 focus groups, along with additional measures such as the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock, and Erbaugh 1961), the Anticipatory Grief Scale (AGS; Theut et al. 1991), the Caregiver Strain Index (CSI; Robinson 1983), the Perceived Social Support–Family Questionnaire (Procidano and Heller 1983), and the Well-Being Scale–Basic Needs subscale (Berg-Weger, McGartland, Rubio, & Steiger Tebb, 2000; Tebb 1995) to develop and initially validate a much-needed caregiver grief inventory that takes into account the caregiver's relationship to the care recipient and the care recipient's stage of disease. This work is guided by the general expectation that caregiver grief is a definable, stage-determined, internally consistent construct; that it is measurable in its totality as well as in its component parts; that it bears some similarity to depression and caregiver strain, yet retains sufficient unique characteristics; and that it has explanatory value within the context of existing models.

	Methods

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 
Participants
Participants (N = 166) were 83 adult child and 83 spouse caregivers of persons with progressive dementia (see Appendix A, Note 1). Participants were recruited through the Memory and Aging Project (the clinical research arm of the Alzheimer's Disease Research Center at Washington University School of Medicine), the St. Louis Chapter of the Alzheimer's Association, and the Alzheimer List (an e-mail–based support group; www.adrc.wustl.edu/alzheimer). The demographic distribution and care-related characteristics of this sample are presented in Table 1 and Table 2 . This sample represents 81% of the 205 caregivers who initially agreed to participate. Those not included either dropped out after receiving the questionnaire or left too many questions unanswered. No data are available on these 39 caregivers.

Demographic and Care-Related Information
A two-page questionnaire was used to elicit the information found in Table 1 and Table 2 .

Informant Impairment Rating
Table 3 shows the CDR scale scoring table (Berg 1988; Hughes et al. 1982; Morris 1993). It was administered as a stand-alone measure of care recipient impairment. Participants were asked to read the impairment descriptions and make their best guess about their loved ones' current status for each cognitive–functional domain.

BDI
The BDI (Beck et al. 1961) is designed to measure key symptoms and attitudes associated with clinical depression (Beck et al. 1961). This well-validated inventory is composed of 21 forced-choice questions (four answer options in each, scored from 0 to 3 points) assessing current emotional health. The BDI has demonstrated high internal consistency in both psychiatric ( = .86) and community-dwelling ( = .81) populations (Beck, Steer, and Garbin 1988). See Table 4 for statistics based on the present sample.

AGS
The AGS (Theut et al. 1991) is designed to be specific to dementia caregivers and is composed of 27 grief-related items responded to on a 5-point Likert scale (e.g., "I have felt very much alone since the diagnosis of dementia was made"). The AGS was originally validated on a sample of 27 female spouse caregivers of dementia patients and has demonstrated respectable internal consistency (.84). Four items were excluded in this study due to questionable face validity and unclear direction for scoring (e.g., "I feel close to my relative who has dementia"). See Table 4 for statistics on the 23-item version based on the present sample.

Caregiver Grief Items
One hundred eighty-four grief-related items were generated from statements made by 45 adult child and 42 spouse caregivers during 16 focus groups conducted during the first phase of this project (see Meuser and Marwit 2001). Forty percent of the items were direct or modified quotations from individual focus-group participants (e.g., "I have this empty, sick feeling knowing that my loved one is gone"). Sixty percent were paraphrased quotations or manufactured items intended to reflect the range of issues discussed by focus group participants and represented in our caregiver grief model (e.g., "I feel this constant sense of responsibility and it just never leaves"). All items were rated on a 5-point Likert scale ranging from strongly disagree to strongly agree, with some items constructed for reverse scoring.

Caregiver Well-Being Scale—Basic Needs Subscale
This scale (Berg-Weger et al. 2000; McGartland Rubio, Berg-Weger, and Tebb 1999; Tebb 1995) asks caregivers to rate the degree to which 22 strengths-based needs are met in their current life situation. Needs such as "feeling loved," "enjoying sexual intimacy," and "having meaning in your life" are rated on a 5-point Likert scale ranging from never or almost never to almost always. Berg-Weger and colleagues 2000 reported a coefficient alpha for the Basic Needs subscale of .91. They claimed known-groups validity by demonstrating an inverse correlation (-.60) with Radloff 1977 Center for Epidemiological Studies–Depression Scale. Further validation was established by McGartland Rubio, Berg-Weger, and Tebb 1999 using structural equation modeling.

CSI
This 13-item scale (Robinson 1983) asks participants to circle "yes" or "no" responses to items related to the potential difficulties of caregiving, such as "Is your sleep disturbed (e.g., because your spouse or parent is in and out of bed or wanders around all night)?" or "Have there been changes in personal plans (e.g., because you had to turn down a job; could not go on vacation)?" This dichotomous scale was originally tested on a population of 86 caregivers of persons who had had heart or hip operations (Cronbach's = .86).

Perceived Social Support Questionnaire–Family subscale
This 20-item scale (Procidano and Heller 1983) is "designed to measure the extent to which an individual perceives that his/her needs for support, information, and feedback are fulfilled by family" (p. 2). It requires participants to rate items such as "My family gives me the moral support I need" or "Members of my family are good at helping me solve problems." The original study used a 3-point scale; the current study uses a 5-point scale ranging from never to always in order to give a wider range of responses. Procidano and Heller reported high test–retest reliability (r = .83 over a 1 month interval) and high internal consistency (Cronbach's = .90). More recent studies (e.g., Steffen, McKibbin, Zeiss, Gallagher-Thompson, and Bandura in press), have reported somewhat lower internal consistency (Cronbach's = .74).

	Results

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 
Factor Analysis of the Caregiver Grief Items
A number of steps were taken to reduce the 184 caregiver grief items to a manageable number for use in an inventory to measure caregiver grief. First, we tested each individual item for skewness. Twenty items with skewness measures greater than 1.0 were dropped from further analysis. Next, an unrotated principal-components analysis was run, with eigenvalues set to be greater than 1.0. A total of 61 factors were suggested. A subsequent rotated principal-components model (varimax) suggested 39 factors (see Table 5 ). Examination of the eigenvalues and associated skree plot suggested that only 3 factors were distinct.

Orthogonal (varimax) and oblique rotations (oblimin, = 0.3) were generated. Both yielded similar results. Given that emotion-related measures tend to correlate with one another, we decided to continue with the oblique rotation. A final principal axis factoring model with oblique rotation was then generated. This final effort explained 34% of the item variance and yielded a 0.889 Kaiser-Meyer-Olkin measure of sampling adequacy (1.0 is perfect). Factor loadings at this stage ranged from .404 to .915 (see Table 6 ).

Reliability and Validity
The MM-CGI Total Grief (TG) score is the sum of all 50 items with some reverse coding involved. The TG score yielded an internal consistency reliability of .96 (Cronbach's ). A Guttman split-half analysis resulted in a similarly consistent score of 0.91 (see Appendix A, Note 3). Descriptive statistics for each of the three factors across CDR levels are listed in Appendix D and visually depicted in Fig. 1.

View larger version (65K): [in this window] [in a new window]   Figure 1. Marwit-Meuser Caregiver Grief Inventory (MM-CGI) scores by Caregiver Type and Clinical Dementia Rating (CDR) scale level. CDR 0.5 = very mild dementia, 1.0 = mild, 2.0 = moderate, 3.0 = severe; A, Post hoc Student-Newman-Kuels statistic (SNK) showed total grief scores at the CDR 0.5 level to differ significantly from the CDR 3.0 for both caregiver types; B, Post hoc SNK showed Factor 1 scores at the CDR 0.5 level to differ significantly from the CDR 3.0 scores for both caregiver types; C, Post Hoc SNK showed spouse caregiver scores on Factor 2 at CDR 0.5 to differ significantly from CDR 3.0 scores. Child caregiver scores at CDR 0.5 were shown to differ significantly from all remaining CDR levels (1–3); D, Differences were not statistically significant.

As can be seen in Table 7 and Table 8 , the correlation of each factor with the MM-CGI TG score is high (rs = .897, .833, and .856, respectively), as would be expected. The correlations of the MM-CGI TG score with the three depression measures are more modest, as would be expected if grief and depression, as some have hypothesized, are fundamentally different constructs that share some characteristics. For example, the MM-CGI TG and the BDI share 57% variance, leaving 43% as unique. The correlations of the MM-CGI TG with the depression measures range from .714 with the GDS to .798 with the AGS, the latter being somewhat higher, as expected, as it was specifically designed for populations coping with Alzheimer's disease. Correlations of the MM-CGI TG score with the CSI and the Well-Being Scale (inversely related) are more modest yet (rs = .656 and –.656, respectively), again as would be expected as these have even less direct relationship to grief than does depression. Finally, the correlation of MM-CGI TG to family support exists in an inverse relationship, as anticipated, but to a lesser degree than the other measures (r = -.360).

	Discussion

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 
Before the development of the current instrument, the only measure claiming to specifically address grief in caregivers of patients with Alzheimer's disease was the AGS (Theut et al. 1991). However, that measure was not empirically derived and was based on a limited sample of 27 wives of veterans diagnosed with dementia using Diagnostic and Statistical Manual of Mental Disorders (3rd ed., Rev.; American Psychiatric Association 1987) criteria. The AGS shows moderate internal consistency reliability (Cronbach's = .84) and was validated against a single instrument, the Hopkins Symptoms Checklist. In contrast, MM-CGI items were derived from caregiver statements obtained under controlled conditions (Meuser and Marwit 2001). MM-CGI items are psychometrically supported by factor-analytic techniques, they demonstrate high internal consistency reliability ( = .96), and they appear valid when measured against a number of instruments (including depression, caregiver strain, well-being, and perceived family support inventories) and against an existing model of caregiver grief (Meuser and Marwit 2001). Furthermore, the current study used a large caregiver participant population (N = 166) including both male and female care providers representing each main caregiver type (spouse and adult child) and stage of dementia progression (CDR Levels 0.5–3).

The final 50-item instrument consists of three factors: Personal Sacrifice Burden (Factor 1), Heartfelt Sadness and Longing (Factor 2), and Worry and Felt Isolation (Factor 3). Factor 1 items focus on individual losses in the caregiver's present life related to the caregiver role itself, such as loss of personal freedom (i.e., I'm no longer free to do things for me), loss of sleep, compromised health, and loss of energy. As might be expected, this factor is highly correlated with Robinson 1983 CSI and more highly correlated with this measure than are the other factors. Factor 3 captures the feeling of losing connections with, and support from, others. As expected, this factor was correlated (-.607) with Procidano and Heller 1983 Perceived Social Support Questionnaire–Family subscale and more highly correlated with this measure than are the other factors. Correlations of Factor 1 with caregiver strain and of Factor 3 with perceived family support provide consensual and construct validation for each.

Factor 2 expresses more of the intrapersonal emotional reactions that accompany caregiving (e.g., abject sadness, longing for what was, feelings of powerlessness, and unwillingness to accept the obvious). This factor appears to be more closely related to traditional concepts of grief, that is, one's personal, internal, emotional reactions to actual or impending loss. Unfortunately, standard grief instruments were not included among the measures used and so the relationship of Factor 2 to such measures could not be assessed. We are now investigating studies to address this shortcoming.

With further regard to Factor 2, it is interesting to note that, of the three factors, it had the lowest correlation with the BDI (.588; 35% shared variance) and the GDS (.487; 24% shared variance). If this factor is the strongest indicator of traditional grief, then these modest correlations become understandable in light of the literature, which differentiates grief from depression (e.g., Prigerson et al. 1995, Prigerson et al. 1996; Walker and Pomeroy 1997). That literature suggests that these two affective states share some expressed symptoms, but also have enough different qualities to warrant individual labels (e.g., depression's being more pervasive, responsive to a genetic vulnerability, and more responsive to selective serotonin reuptake inhibitors; involving lower self-worth; and being more likely to reoccur spontaneously; grief being more reactive to specific loss situations, etc.). Thus, it makes sense that there would be some shared variability, but modestly so. In this context, it also makes sense that Factors 1 and 3, which reflect more of the depression-related qualities of felt loss and social disconnection, would be more highly correlated with standard depression measures. In other words, the items comprising Factors 1 and 3 appear to be somewhat more cognitively based and may lend themselves to those cognitive–behavioral techniques often associated with changing depressive thinking, whereas the items in Factor 2 appear to be more personally affectively based and may respond best to empathic support.

To summarize up to this point, (a) all three factors appear to be grief-related factors; (b) all share some variability with depression, strain, and perceived family support; and (c) each retain certain unique grief-specific components. The intercorrelations among the three factors show significant, yet modest overlap, thereby supporting both their relatedness to the general concept of grief as well as their separateness as independent components.

The validity of the MM-CGI can also be addressed by comparing the responses of the adult child and spouse caregivers in the present study with the model of adult child and spouse caregiver grief presented by Meuser and Marwit 2001. This can be done because the characteristics of these two study populations are essentially equivalent. The current sample consisted of 50% adult child caregivers, predominantly female (88%), with a mean age of 51.81 (SD = 8.05) and a mean education of 15.09 years (SD = 2.54), and 50% spouse caregivers, predominantly female (74%), with a mean age of 71.47 (SD = 8.93) and a mean education of 14.17 years (SD =2.81). Meuser and Marwit 2001 sample consisted of 51% adult child caregivers, predominantly female (93%), with a mean age of 51.60 (SD = 9.60) and a mean education of 16 years (SD = 3.00), and 49% spouse caregivers, predominantly female (60%) and with a mean age of 71.80 (SD = 9.00) and a mean education of 14.00 years (SD = 2.81). The two samples were also equivalent on race, marital status, employment status, and religious affiliation. When making these comparisons, it must be kept in mind that the present study used four CDR levels (0.5, 1.0, 2.0, and 3.0), whereas Meuser and Marwit 2001 used three (0.5/1.0 combined, 2.0, and 3.0).

Fig. 1 is helpful in addressing the conformity of the current data with those of Meuser and Marwit 2001 model. Fig. 1 shows the mean MM-CGI TG scores for adult child and spouse caregivers at CDR levels 0.5, 1.0, 2.0, and 3.0. The patterns depicted are similar to those that would be predicted by the model. Meuser and Marwit stated that "for adult-child caregivers, expressed grief (the composite of all grief-related reactions) is almost curvilinear; minimal at CDR 1 (CDR .5 & 1 combined in Fig. 1), most intense and CDR 2, and moderated at CDR 3" (p. 667). In contrast, Marwit and Meuser's (2001) spouse caregivers showed greater acceptance of felt grief (higher grief) at earlier CDR stages than was true for adult child caregivers with "expressed grief increasing linearly from CDRs 1–3" (p. 667).

When the overall grief response is examined in terms of individual factors (Fig. 1), the current data again support Meuser and Marwit 2001 caregiver grief model. Fig. 1 depicts the curves for Factor 1, Personal Sacrifice Burden. For adult child caregivers, the grief responses associated with loss of freedom, loss of sleep, and escalating demands increase from CDR 0.5 through CDR 2 and then decrease at CDR 3, the latter being that stage most closely associated with nursing home placement. Meuser and Marwit stated,

For the prototypic adult-child caregiver in this study, there is a sense of emotional relief that comes with nursing home placement. Their burden is lifted and the intensity of feelings of the previous CDR 2 stage gives way to a mellowing of feelings. (p. 667)

In contrast, for spouse caregivers, Meuser and Marwit noted, "Although nursing home placement relieved the physical burdens of care, the emotional struggles and sense of responsibility remain and may be even more prominent than before" (p. 666, italics added), thus an increase in the factor from CDR 2 to CDR 3. Again, the MM-CGI data and the Meuser and Marwit model are in agreement.

Fig. 1 depicts the flow of Factor 2, Heartfelt Sadness and Longing, that is, the factor considered closest to the traditional concept of grief. Here grief-related responses continue to increase, almost linearly, from CDR 0.5 through CDR 3. Again, this conforms to the Meuser and Marwit 2001 model. Meuser and Marwit stated,

At stage 3, adult-child caregiver discussions are marked by a very different quality of grief than either of the first two stages. It may be best characterized as a true "grief of the soul." With these individuals, the anger of stage 2 is replaced by a deep sadness for that which can never be again. The predominant theme is the final and absolute loss of the parent as parent. Members talk about grief being at its highest point early in stage 3. (p. 665)

With Factor 2, the same holds true for spouse caregivers. Sadness and longing for the past are described as highest at CDR 3.

Last, in Fig. 1, depicting the curves for Worry and Felt Isolation (Factor 3), most noteworthy is the deescalation of the intensity of this factor at CDR 3 for adult child caregivers and the escalation at CDR 3 among spouse caregivers. In contrast, Meuser and Marwit 2001 stated,

What stands out most for spouse caregivers in this advanced stage is a sense of being stuck and unsure how to proceed with life... The togetherness of the past has given way to an uneasy individuality... Living as a single person is a frightening concept. (p. 666)

Again, the MM-CGI factors and the caregiver grief model are in agreement.

The MM-CGI is a new instrument. Its three factors show high internal consistency reliability and preliminary validity when assessed in the context of existing inventories as well as in the context of the caregiver grief model delineated by Meuser and Marwit 2001. Although the current study provided initial support for the instrument, more needs to be done to fully establish construct validity. A number of weaknesses in this study will need to be addressed in future research, including the validation of the MM-CGI as a stand-alone measure with a new sample of caregivers. The validity correlations reported here relate to 50 items derived from a larger pool of 184 items, all administered to the same sample. It is unclear how responding to a larger pool of items may have influenced 50-item scores. Additional correlations with other and related measures (e.g., a standard measure of death-related grief) will be essential to further establish construct validity. With regard to informant-based assessment of dementia stage, we believe that our use of the CDR as a self-report measure may need reexamination in the future. As noted in Appendix A, Note 2, there was a significant difference between self-reported impairment and actual clinician-determined impairment (i.e., caregivers tended to downplay slightly the level of impairment in the care recipient). Furthermore, we have no data on the stability of these impairment ratings as they were collected from participants at one point in time. Similar or different findings may emerge with the use of alternative staging methods.

It is hoped that by presenting the scale in full in this article, others will contribute to this endeavor and will keep us apprised of their findings. We believe that the present data add to the emerging literature suggesting that caregiver grief is a viable concept for study in the larger body of Alzheimer's disease research. These findings, along with those of Meuser and Marwit 2001, suggest that caregiver grief cannot, and should not, be approached as either a unitary or a static concept. Rather, caregiver grief needs to be understood in terms of its component parts and in terms of its interaction with caregiver type and stage of disease. In this regard, we hope that the MM-CGI will prove useful not only as a research instrument but as a clinical diagnostic tool for identifying areas amenable to constructive intervention (see Appendix C).

	Acknowledgments

 
This research was supported by Grant 1999-PRG-1730 from the Alzheimer's Association. We are grateful to Dr. John C. Morris, co-director of the Washington University Alzheimer's Disease Research Center and director of the Memory & Aging Project, and his capable staff (particularly Joanne Norton, MSN, RN, CS) for their support of this project through facility usage, participant recruitment, and Clinical Dementia Rating scoring assistance. We are thankful to the Alzheimer's Association, St. Louis Chapter, for its kind assistance in participant recruitment. We also thank Dr. John Chibnall of St. Louis University Medical School and Dr. Kim Powlishta of Washington University School of Medicine for their help with statistical analysis.

Drs. Meuser and Marwit contributed equally to all phases of this project.

Received for publication January 12, 2002. Accepted for publication April 15, 2002.

	Appendix A

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 
Notes
1. Eighteen percent of the participants had also taken part in focus groups held a year earlier (Meuser and Marwit 2001).

2. Seventy-nine participants in this study also take part in the Memory & Aging Project's longitudinal study of aging and cognition, and this allowed for a comparison of the two assessment methods (i.e., table only vs. full CDR). The partial correlation of these, while controlling for time of assessment, is 0.67 (p < .0001, n = 79). Although similar in terms of shared variance, there was nonetheless a significant mean difference between methods (full CDR mean of 1.46 > caregiver CDR mean of 1.24, p < .01, n = 82). Caregivers underestimate actual clinical impairment using the table-only method.

3. The MM-CGI was mailed to the original sample nine months later. Sixty-one caregivers responded (37%). Correlation of the three factor scores and the total score (50-items, Times 1 and 2) were as follows: Factor 1, .71, Factor 2, .73, Factor 3, .75, Total, .71 (all ps < .01). These correlations suggest stability of the measure, with all factors equally affected, in part, by the passage of time and progression of the disease.

4. Percentage of care was defined as the proportion of care tasks that the caregiver completes individually, ranging from 0% to 100%.

	Appendix B

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 
Marwit–Meuser Caregive Grief Inventory

View larger version (247K): [in this window] [in a new window]   Figure 2. Marwit–Meuser Caregive Grief Inventory

	Appendix C

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

The MM-CGI is composed of three subscales (factors) and a total grief (TG) score:

Factor 1: Personal Sacrifice Burden (What "I" as the caregiver have had to give up in my life)

Factor 2: Heartfelt Sadness and Longing (true grief factor including sadness and longing for how life used to be)

Factor 3: Worry and Felt Isolation (uncertainty of how things will turn out coupled with isolation from others)

How May the Inventory Be Used?

The MM-CGI is appropriate for use in any clinical or supportive care setting where individual attention is available. Although the inventory can be self-scored, we believe it is important for caregivers to discuss and understand their scores in the context of their larger life situation. Personal attention (either one on one or in a group format) from a lay or professional counselor is essential for such discussion and understanding.

How Can the MM-CGI Scores Be Interpreted?

Until further research data become available, interpretation of the MM-CGI total and subscale scores will necessarily be a bit subjective. As noted in the brief interpretive comments on the scale itself (see Appendix B), "high scores may indicate a need for formal intervention or support assistance to enhance coping." Low scores may be indicative of adaptive coping or denial of grief-related issues. Denial can often be adaptive early in a grief process or just after a new loss is recognized. It is when denial continues for months or years that problems can occur. For example, a caregiver who denies aspects of grief may become depressed as a result. Talking through the underlying grief issues may help to reduce or even resolve the mood disturbance. It is up to the lay or professional counselor to sort this out on an individual level. Scores in the average range (1 SD on either side of the mean) are common responses to loss in the caregiving experience.

Special Cases for Interpretation: Gender, Caregiver Type, and Clinical Dementia Rating (CDR) Stage

Women tend to report greater emotional distress on questionnaires than do men. This is true for four scales in the present study: the Beck Depression Inventory, the Anticipatory Grief Scale, the Caregiver Strain Index, and the MM-CGI Factor 2, Heartfelt Sadness and Longing (see Table 8 ). When working with men, it is important to recognize that their reactions may be somewhat greater than they actually report. Men in older age cohorts were socialized to keep their emotional needs in the background. It is up to the counselor to gradually tease these out and offer an appropriate level of support. Forcing someone to express feelings is never a good idea, however.

Adult child caregivers reported greater strain from caregiving relative to their spousal counterparts. This is due, in part, to the fact that children of demented persons must meet two or more sets of competing demands: maintaining a home life while also caring for their parent. Although the physical burdens of care may be greater for children, the emotional burdens of Personal Sacrifice Burden (MM-CGI Factor 1) and Heartfelt Sadness and Longing (Factor 2) may be somewhat greater for spousal caregivers. There is a shift in focus of loss that occurs over time for adult children and spouse caregivers: Children focus on their own personal losses first, then shift to grieving what the parent has lost, whereas the opposite is true for spouse caregivers.

Grief is also different across CDR stages of dementia progression (0.5 = very mild, 1 = mild, 2 = moderate, 3 = severe). As shown in Fig. 1, there was a steady rise in total grief across CDR stages for spouse caregivers, but not for adult children. Nursing home placement brings high levels of heartfelt sadness and felt isolation for spouse caregivers (MM-CGI Factors 2 and 3). Adult child caregivers reach their peaks in personal sacrifice burden and felt isolation at CDR 2, months or years before placement. Adult child caregivers demonstrate an overall decline in grief by CDR 3.

How to Respond?

Each person recognizes losses and expresses grief differently. An individual approach to provide grief-related support often works better than a less focused group intervention. It is important to consider the individual characteristics of the caregiver when assessing the qualities of their grief and coping resources. Men may express less raw emotion, yet need the same level of support as women. Adult children have a very different grief pattern than do spouse caregivers. What is high or low for one person may not translate well to another. "Good" grief-related support usually involves more listening than talking, an honest dialogue about emotional and life–death issues (especially anger and guilt), and a comfortable pace set to match the caregiver's openness and coping capabilities. We would appreciate hearing your comments on use of the MM-CGI.

	Appendix D

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

 

	References

TOP Abstract Methods Results Discussion Appendix A Appendix B Appendix C Appendix D References

• American Psychiatric Association1987. Diagnostic and statistical mannual of mental disorders 3rd ed., rev.. Author, Washington, DC.
• Bass D. M., Bowman K., Noelker L. S., 1991. The influence of caregiving and bereavement support on adjusting to an older relative's death. The Gerontologist 31:32-42. [Abstract]
• Beck A. T., Steer R. A., Garbin M. G., 1988. Psychometric properties of the Beck Depression Inventory: 25 years of evaluation. Clinical Psychology Review 8:77-100.
• Beck A. T., Ward C. H., Mendelson M., Mock J., Erbaugh J., 1961. An inventory for measuring depression. Archives of General Psychiatry 4:561-571.
• Berg L., 1988. Clinical dementia rating. Psychopharmacology Bulletin 24: (4) 637-639. [Medline]
• Berg-Weger M., McGartland Rubio D., Steiger Tebb S., 2000. The Caregiver Well-Being Scale revisited. Health & Social Work 25: (4) 255-263. [Medline]
• Bowd A. D., Loos C. H., 1996. Needs, morale and coping strategies of caregivers for persons with Alzheimer's disease in isolated communities in Canada. American Journal of Alzheimer's Disease May: (June) 32-39.
• Brink T. L., Yesavage J. A., Lum O., Heersema P., Adey M. B., Rose T. L., 1982. Screening tests for geriatric depression. Clinical Gerontologist 1:37-44.
• Doka K. J., 2000. Mourning psychosocial loss: Anticipatory mourning in Alzheimer's, ALS, and irreversible coma. Rando T. A., , ed.Clinical dimensions of anticipatory mourning: Theory and practice in working with the dying, their loved ones, and their caregivers 477-492. Research Press, Champaign, IL.
• Farran C. J., Keane-Hagerty E., Salloway S., Kupferer S., Wilken C. S., 1991. Finding meaning: An alternative paradigm for Alzheimer's disease family caregivers. The Gerontologist 31:483-489. [Abstract]
• Gilhooly M. L. M., Sweeting H. N., Whittick J. E., McKee K., 1994. Family care of the dementing elderly. International Review of Psychiatry 6:29-40.
• Hughes C. P., Berg L., Danziger W. L., Coben L. A., Martin R. L., 1982. A new clinical scale for the staging of dementia. British Journal of Psychiatry 140:566-572. [Abstract/Free Full Text]
• Loos C. H., Bowd A. D., 1997. Caregivers of persons with Alzheimer's disease: Some neglected implications of the experience of personal loss and grief. Death Studies 21:501-514. [Medline]
• Mace N. L., Rabins P. V., 1981. The 36 hour day Johns Hopkins University Press, Baltimore.
• McGartland Rubio D., Berg-Weger M., Tebb S.S., 1999. Assessing the validity and reliability of well-being and stress in family caregivers. Social Work Research 23:54-64.
• Meuser T. M., Marwit S. J., 2001. A comprehensive, stage-sensitive model of grief in dementia caregiving. The Gerontologist 41:658-670. [Abstract/Free Full Text]
• Morris J. C., 1993. The Clinical Dementia Rating (CDR): Current version and scoring rules. Neurology 43:2412-2414.
• Mullan J., 1992. The bereaved caregiver: A prospective study of changes in well-being. The Gerontologist 32:673-683.
• Neimeyer R., Hogan N., 2001. Quantitative or qualitative? Measurement issues in the study of bereavement. Stroebe M. S., Hansson R. O., Stroebe W., Schut H., , ed.Handbook of bereavement research: Consequences, coping, and care 89-118. American Psychological Association, Washington, DC.
• Ponder R. J., Pomeroy E. C., 1996. The grief of caregivers: How pervasive is it?. Journal of Gerontological Social Work 27: (1/2) 3-21.
• Prigerson H. G., Bierhals A. J., Kasl S. V., Reynolds C. F., III Shear M. K., Newsom J. T., et al. 1996. Complicated grief as a disorder distinct from bereavement-related depression and anxiety: A replication study. American Journal of Psychiatry 153:1484-1486. [Abstract/Free Full Text]
• Prigerson H. G., Frank E., Kasl S. V., Reynolds C. F., III Anderson B., Zubenko G. S., et al. 1995. Complicated grief and bereavement-related depression as distinct disorders: Preliminary empirical validation in elderly bereaved spouses. American Journal of Psychiatry 152:22-30. [Abstract/Free Full Text]
• Procidano M. E., Heller K., 1983. Measures of perceived social support from friends and from family: Three validation studies. American Journal of Community Psychology 11:1-24. [Medline]
• Rabins P. V., 1984374–375. Management of dementia in the family context. Psychosomatics 25: (5) 369-371. [Abstract/Free Full Text]
• Radloff L. S., 1977. The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measurement 1:385-401.
• Robinson B. C., 1983. Validation of a Caregiver Strain Index. Journal of Gerontology 38:344-348. [Abstract/Free Full Text]
• Rudd M. G., Viney L. L., Preston C. A., 1999. The grief experienced by spousal caregivers of dementia patients: The role of place of care of patient and gender of caregiver. International Journal of Aging & Human Development 48: (3) 217-240.
• Sheikh J. I., Yesavage J. A., 1986. Geriatric Depression Scale (GDS): Recent evidence and development of a shorter version. Clinical gerontology: A guide to assessment and intervention 165-173. Haworth Press, New York.
• Snyder L., 2001. The lived experience of Alzheimer's: Understanding the feelings and subjective accounts of persons with the disease. Alzheimer's Care Quarterly 2: (2) 8-22.
• Steffen A. M., McKibbin C., Zeiss A. M., Gallagher-Thompson D., Bandura A., in press. The Revised Scale for Caregiving Self-Efficacy: Reliability and validity studies Psychological Sciences, Journal of Gerontology.
• Tebb S., 1995. An aid to empowering: A caregiver well-being scale. Health & Social Work 20: (2) 87-92. [Medline]
• Theut S. K., Jordan L., Ross L. A., Deutsch M. D., 1991. Caregiver's anticipatory grief in dementia: A pilot study. International Journal of Aging and Human Development 33: (2) 113-118. [Medline]
• Walker R. J., Pomeroy E. C., 1996. Depression or grief? The experiences of caregivers of persons with dementia. Health & Social Work 21: (4) 247-254. [Medline]
• Walker R. J., Pomeroy E. C., 1997. The impact of anticipatory grief on caregivers of persons with Alzheimer's disease. Home Health Care Services Quarterly 16: (1/2) 55-77.
• Walker R. J., Pomeroy E. C., McNeil J. S., Franklin C., 1994. Anticipatory grief and Alzheimer's disease: Strategies for intervention. Journal of Gerontological Social Work 22: (3/4) 21-39.
• Yesavage J. A., Brink T. L., Rose T. L., Lum O., Huang V., Adey M. B., et al. 1983. Development and validation of a geriatric depression screening scale: A preliminary report. Journal of Psychiatric Research 17: (1) 37-49.

This article has been cited by other articles:

				K. Betts Adams, M. J. McClendon, and K. A. Smyth Personal losses and relationship quality in dementia caregiving Dementia, August 1, 2008; 7(3): 301 - 319. [Abstract] [PDF]

				J. B. Frank Evidence for Grief as the Major Barrier Faced by Alzheimer Caregivers: A Qualitative Analysis American Journal of Alzheimer's Disease and Other Dementias, January 1, 2008; 22(6): 516 - 527. [Abstract] [PDF]

				C. H. Ott, S. Sanders, and S. T. Kelber Grief and Personal Growth Experience of Spouses and Adult-Child Caregivers of Individuals With Alzheimer's Disease and Related Dementias Gerontologist, December 1, 2007; 47(6): 798 - 809. [Abstract] [Full Text] [PDF]

				M. C. Reilly, N. R. Relkin, and A. S. Zbrozek Development and testing of a new outcome measure of relationship between patients with Alzheimer's disease and their partners. American Journal of Alzheimer's Disease and Other Dementias, August 1, 2006; 21(4): 249 - 257. [Abstract] [PDF]

				K. B. Adams and S. Sanders Alzheimer's Caregiver Differences in Experience of Loss, Grief Reactions and Depressive Symptoms Across Stage of Disease: A Mixed-Method Analysis Dementia, June 1, 2004; 3(2): 195 - 210. [Abstract] [PDF]

This Article Abstract Full Text (PDF) Services Download to citation manager Citing Articles Citing Articles via HighWire PubMed PubMed Citation