Center to Improve Care of the Dying

GRIEF

Literature review

A literature review of the Medline database was conducted with the search terms bereavement, grief, instrument, measurement, and (palliative care or terminal care). We also searched the Cancerlit database. In addition, we selected references from the methods sections of the articles pulled, and the George Washington University online catalogue was searched for relevant books.

I. Listing of potential instruments

Based on the review of the literature, 6 potential instruments were found which are listed below.

a. Grief Resolution Index (GRI)- Remondet 1985.
b. Anticipatory Grief Scale (AGS)- Theut 1991.
c. The Revised Grief Experience Inventory (REGI)- Sanders 1985.
d. Texas Inventory of Grief (TIG)- Faschingbauer 1977.
e. Expanded Texas Inventory of Grief (ETIG)- Zisook 1982.
f. Grief Experience Questionnaire (GEQ)-Barrett 1989.

II. Review of potential instruments

I am recommending the Grief Resolution Index for inclusion into the Toolkit. Looking at the Table provided, you can see my review of each of the 6 scales. I will go into greater detail of the Grief Resolution Index and give my reasons for recommending it for the Toolkit.

A. Grief Resolution Index

i. Conceptual and Measurement model (Does the scale represent a single domain or do model scales measure distinct domains? Is the variability of the scale reported? What is the intended level of measurement i.e. ordinal, interval, ratio or category?)

The purpose of this Index is to identify individuals who are experiencing prolonged psychological distress associated with the death of a spouse. The Toolkit's purpose seems to best fit this Index rather than measuring grief itself. The grief process is complicated and cannot be properly measured by a short set of questions. This index of grief resolution seems to give a good indication of how well or poorly the individual was able to go through process of grief. It was able to predict both short-term and long-term adjustment to widowhood. This makes the index a useful longitudinal tool.

This is an interview assessment, but it could easily be self-administered. Although not reported, it should only take 5 minutes to complete. The statements were designed for use with widows, but, with the exception of item 2--Stopped saying "we"-- it could be worded to refer to any relationship. The 7 item index is a 5 point continuous scale from '1' being very poorly to '5' being very well. The mean score was 24.4 and standard deviation was 7.5. The factor loading for item '2' is the lowest of the 7 items in the index (.67). I believe it is reasonable to assume that dropping this item when using it with individuals other than spouses would not reduce the internal reliability of the index significantly. This of course would have to be bore out with analysis. As with all the measures of grief, the main focus of research has been with elderly widows. Very few studies have had samples of younger widows or of widowers for that matter. Studies that have had such samples, find differences in the intensity of grief, but the process or feelings are the same. Both young widows and men make the same types of statements about their feelings about the death of a loved one as older women and women. Therefore, I don't see why this index could not be used with men.

ii. Reliability (Did they address internal consistency? Did they address reproducibility?)

Internal consistency was assessed using Cronbach's Alpha. The Alpha level for the scale is good at .87.

iii. Validity (How did they address content related validity? Any information on construct related validity? Any information on criterion validity?)

Content validity was assessed by correlating the index with measures of short-term adjustment: Survival Expectations Index, Fear Index, Preparation Index and Desperation Index. The Grief Resolution Index was statistically significantly associated with all three short-term adjustment measures. Long-term adjustment, measured by: Beck Depression Inventory, Anxiety Index, Adjustment to Widowhood Index and Health Index. the GRI had statistically significant associations with all of these long-term adjustment measures. The index was constructed with input from members of the American Association of Retired Persons' widowed persons service which gives it face validity.

iv. Responsiveness (Any information? Has the scale ever been used as an outcome measure? If so, what populations?)

This scale is meant to be used as an outcome measure.

v. Interpretability (What populations has it been applied to? Is the score translated into a clinically relevant event? Does the score predict outcome events?)

This scale was used with a sample of widows, median age 75.5 and a median length of widowhood of 10.3 years. The women were recruited from senior adult recreation nutrition centers and were not randomly selected, nor, are they necessarily representative of all widows. Given these biases, I still feel that the index is a good tool for measuring psychological adjustment to a death.

vi. Burden (Any information on cost or time to administer? Does the instrument impact on the respondent? How long does the survey take to complete? Response rates? Any problems with missing data?)

This 7 item scale can be administered in 5 minutes.. No information about missing data was available.

vii. Alternative Forms (What are the modes of administration? Alternatives? Of alternatives exist, provide what is known for each of the above categories)

Intended as interview questions, however, I see no reason why it couldn't be self-administered.

viii. Cultural and Language Adaptations (Any information?)

The grief process is culturally defined and it would be not be reasonable to assume this scale or any of those mentioned in this report could be translated into different cultures. All the scales mentioned here measure the grief process in modern industrialized western cultures. Indicators of grief resolution may also be culturally defined, so this index would not be appropriate for other cultures.

ix. Conceptual Rational for Recommendation

After reviewing the literature, it is my opinion that, for clinicians, trying to quickly identify an individual having difficulty adjusting to the death of a loved one should be a higher priority than measuring the grief process itself. The GRI is a quick and effective way to assess difficulty in adjustment. The GRI is a good starting point for clinicians to begin the intervention process where more in depth assessment on the grief process and adjustment could take place.

x. Future Research

The downfall of the GRI is that it does not give any indication of adjustment with the different dimensions of grief. Constructing a resolution index that tapped into different dimensions of grief, would help in pinpointing what grief component the person is having difficulty with, and then interventions could be tailored to that component. It may be that some individuals are no longer preoccupied with thoughts about the loved one, but they have a high level of anger about the death which produces a high level of stress. Looking at the relationship and predictability of anticipatory grief in grief resolution is an area which has had a lot of theoretical work, but little empirical research. I would, therefore, like to recommend the Anticipatory Grief scale for be considered for inclusion into the Toolkit. I will go into further detail of the Anticipatory Grief Scale and give my reasons for recommending it along with the Grief Resolution Index below.

B. Anticipatory Grief Scale

i. Conceptual and Measurement model

This scale represents the major domains cited in the literature on grief. It is the first Anticipatory Grief Scale cited in the literature. Anticipatory Grief has usually been measured using 'time since death' as a proxy for this process. However, anticipatory grief has been cited in the literature as having important implications for outcomes after the death of a loved one, no one had attempted to quantify the process until now.

Grief is a multi-dimensional process, and, therefore a scale measuring the different aspects is important. The scale is a 27 item 5-point Likert scale: 1 = Strongly Disagree, 2= Disagree, 3 = Somewhat Agree, 4 = Agree, and 5 = Strongly Agree. No mean scores were reported.

This is a self-administered questionnaire and can be completed in 10 - 15 minuets. The questions are worded with reference to the dying patient and to the disease diagnosis. In this sample of women it was 'dementia', but the wording could be changed to include any diagnosis. It could also be re-worded to use a different point of reference (i.e. instead of relative, spouse, child could be used). It could also be used after the death by leaving out "since...was diagnosed" and replacing with "since the death of....". I will discuss this further later.

ii. Reliability

Internal consistency was assessed using Cronbach's Alpha. The Alpha level for the scale is good at .84.

iii. Validity

Content validity was assessed by correlating the scale with the SCL-90-R depression, anxiety and hostility sub-scales. These measures of psychological distress have been shown to be outcomes of grief. The correlations were not reported, but the p levels were and they showed statistical significance at the .001 level for each of the psychological distress measures.

iv. Responsiveness

This scale has not been used as an outcome measure.

v. Interpretability

This scale was used with a sample of wives whose spouses were diagnosed with dementia. The mean age of the respondent was 68.1 and the mean age of the patient was 72.3. The average time the patients had symptoms of dementia was 3.8 years.

The use of Likert scales have been widely used as outcome variables.

vi. Burden

This 27 item scale can be administered in 10 - 15 minutes. No information about missing data was available.

vii. Alternative Forms

Intended as a self-administered questionnaire. However, I see no reason why it could not be administered by an interviewer of intake person.

viii. Conceptual Rational for Recommendation

Since one of the purposes here is to find instruments which can be used in the last month of life and can help in the assessment of quality of care, I think that it is important to assess anticipatory grief. The advantages are that clinicians, social workers and counselors can identify problems an individual may be having before the death of the loved one so that interventions can take place to avert long term bad outcomes after the death. Someone who is having intense grief before the death may be at greater risk for abnormal grief after the death which leads to both psychological and physical bad health. At the same time, someone who does not experience any prior grief may have greater problems with acceptance after the death, which could also lead to poor outcomes.

Measuring anticipatory grief and then grief would allow for the assessment of areas which may need improvement for the care of family members of someone who dies. Using the Anticipatory Grief Scale as a predictor of Grief Resolution could give clinicians a handle on the effectiveness of their grief intervention programs.

ix. Future Research

This Anticipatory Grief Scale seems to be multi-dimensional; however, there was not any factor analysis done to assess the independent dimensions. I think this would be an important piece to be done. In the grief literature, time and time again we see that anger and guilt over the death are strong predictors of psychological distress. It would be beneficial if this scale could be scored on each dimension to be able to identify specific areas which need to be addressed with interventions. A person may have 'normal' feelings of sadness, but may have very intense feelings of anger or guilt.

Research on interventions with family members who are loosing a loved one has not been fully explored. Some studies which have been done have not shown interventions to be effective, but they have tended to only measure grief at two time periods after the death and have not measured grief before. Very little has been done in research to develop interventions before the death of a loved one.

Research also needs to be done with samples of younger women and men of all ages. Also, samples of widows and widowers whose loved one died from various medical conditions needs to be explored. Grief and resolution of grief may vary depending on the condition of the loved one. Little has been done in this area.

References

Barrett TW and Scott TB. Development of the Grief Experience Questionnaire. Suicide and Life- Threatening Behavior 1989;19:201-215.

Covinsky, KE, Goldman, L., Cook, E.F., et al. The Impact of Serious Illness on Patients' Families. JAMA 1994;272:1839-1844.

Faschingbauer TR, Devaul RA and Zisook S. Development of the Texas Inventory of Grief. Am J Psychiatry 1977;134:696-698.

Gallager, D., Rappaport, M., Benedict, A., et al. Burden: A Review of Measures Used Among Caregivers of Individuals with Dementia. Paper presented at the Annual Scientific Meeting of the Gerontological Society of America 1985, New Orleans.

Gerritsen JC and Van der Ende PC. The Development of a Care-giving Burden Scale. Age and Aging 1994;23:483-491.

Kosberg JI and Cairl, R. The Cost of Care Index: A Case Management Tool for Screening Informal Caregivers. Gerontologist 1986;26:273-278.

Kosberg, J.I., Cairl, R.E. and Keller, D.M. Components of Burden: Interventive Implications. Gerontologist 1990;30:236-242.

Lev E, Munro BH and McCorkie R. A shortened version of an instrument measuring bereavement. International Journal of Nursing Studies 1993:213-226.

Lundin T. Long-term Outcome of Bereavement. Br J Psychiatry 1984;142:424-428.

Montgomery RJV, Gonyea JG and Hooyman NR. Caregiving and the Experience of Subjective and Objective Burden. Family Relations 1985;34:19-26.

Novak M and Guest CI. Application of a Multidimensional Care-giver Burden Inventory. Gerontologist 1989;29:798-803.

Pratt C, Schmall VL, Wright S and Cleland M. Burden and Coping Strategies of Caregivers to Alzheimer's Patients. Family Relations 1985;34:27-33.

Remondet JH and Hanson RO. Assessing A Widow's Grief--A Short Index. J Gerontol Nursing 1987;13:30-34.

Robinson BC. Validation of A Caregiver Strain Index. J Gerontol 1983;38:344-348.

Schott-Baer D, Fisher L and Gregory. Dependent Care, caregiver burden, hardiness, and self-care agency of caregivers. Cancer Nursing 1995;18:299-305.

Siegel K, Raveis VH, Houts P and Mor V. Caregiver Burden and Unmet Patient Needs. Cancer 1991;68:1131-1140.

Stull DE, Kosloski K and Kercher K. Caregiver Burden and Generic Well-Being: Opposite Sides of the Same Coin? Gerontologist 1994;34:88-94.

Theut SK, Jordan L, Ross LA and Deutsch SI. Caregiver's Anticipatory Grief in Dementia: A Pilot Study. International Journal of Aging and Human Development 1991;33:113-118.

Yancey D, Greger HA and Coburn P. Determinants of Grief Resolution in Cancer Death. J Pall Care 1990;6:24-31.

Zarit SH, Reever KE and Bach-Peterson J. Relatives of the Impaired Elderly: Correlates of feelings of Burden. Gerontologist 1980;20:649-655.

Zisook S, Devaul RA and Click MA, Jr. Measuring Symptoms of Grief and Bereavement. Am J Psychiatry 1982;139:1590-1593.

Section prepared by Mary Jo Roach