Journal of Gerontological Nursing

ALCOHOL USE Among Bereaved and Nonbereaved Older Persons

Barbara Valanis, DrPH; Rosalee C Yeaworth, PhD, RN; Marcia Resing Mullis, MSN, RN, CS

Abstract

There are an estimated IO million alcoholics in the United States and it is generally assumed that alcohol is a factor in 10% of all deaths.1 While the overall prevalence of drinking and alcohol abuse is generally lowest in older age groups, alcohol abuse does exist among older populations. Zimberg et al2 estimated that of persons 55 years and older living in the community, 10% to 15% abuse alcohol. A review of literature on alcohol problems in the elderly by Schuckit and Miller3 concluded that alcohol abuse probably affects between 2% to 10% of the general elderly population but that rates were higher among widowers, individuals with medical problems, and people who are in difficulty with the police. They estimated that perhaps as high as 20% of elderly individuals who are medical inpatients and 10% to 15% of elderly medical outpatients have serious life problems related to alcohol. A review of the literature on alcoholism reveals that few data-based studies are available. Much is based on estimates or opinions.

Tolerance for alcohol appears to decrease in old age. A decreased tolerance combined with age-related changes in balance, manual dexterity, and postural flexibility can increase the probability of falls, bums, or other accidents. Effects of alcohol can prevent serious heart conditions from being recognized. An additional concern is that while persons aged 65 or older make up about 10 percent of the U.S. population, this same group consumes 25% of the nation's prescription drugs.4 When drugs are combined with alcohol, there is potential for serious interaction, particularly when used by persons already in a vulnerable state of health because of physical decline. Alcohol/drug interactions lead to an estimated 2,500 deaths a year and 47,000 emergency room admissions a year.5

Only four bereaved subjects, however, reported drinking more than three drinks per day prior to the death of their spouse. Of these, only one did not score at least one on the SMAST, so he was categorized as a normal drinker and the category of heavy drinker was eliminated. One additional subject was classified as a problem drinker because of a score greater than one on the SMAST even though he did not report taking more than three drinks per day. (SMAST scores for problem drinkers range from one to three,) At the threemonth interview, 27% of bereaved subjects reported abstention from alcohol use. Sixty-seven percent were normal drinkers, and 6% (four subjects) were classified as problem drinkers.

No relationship was found between age, race, education, social resources, mental health, physical health, or economic resources ratings, and drinking patterns as reported for prior to bereavement. Sex was the only demographic variable related to drinking patterns at this time. Males comprised 6% of abstainers, 35% of normal drinkers and 75% of problem drinkers in a study population that was 30% male.

A statistically significant difference on use of alcohol [x2(1)=3.7; p<.05] between subjects with normal versus depressed scores on the Zung depression scale was observed (Table 3). Depressed subjects were less likely to drink than were subjects scoring normal on the depression scale.

Changes in Alcohol Use from Time 1 to Time 2

Change in drinking behavior between that reported for prior to bereavement and nine months of bereavement was assessed. No new cases of heavy drinking or problem drinking were found by nine months of bereavement. In fact, the prevalence of both decreased between that reported for prior to bereavement and for nine months postbereavement. Of the four subjects who were identified as problem drinkers prior to bereavement, two remained problem drinkers. One subject's SMAST score decreased from one to zero.

The…

There are an estimated IO million alcoholics in the United States and it is generally assumed that alcohol is a factor in 10% of all deaths.1 While the overall prevalence of drinking and alcohol abuse is generally lowest in older age groups, alcohol abuse does exist among older populations. Zimberg et al2 estimated that of persons 55 years and older living in the community, 10% to 15% abuse alcohol. A review of literature on alcohol problems in the elderly by Schuckit and Miller3 concluded that alcohol abuse probably affects between 2% to 10% of the general elderly population but that rates were higher among widowers, individuals with medical problems, and people who are in difficulty with the police. They estimated that perhaps as high as 20% of elderly individuals who are medical inpatients and 10% to 15% of elderly medical outpatients have serious life problems related to alcohol. A review of the literature on alcoholism reveals that few data-based studies are available. Much is based on estimates or opinions.

Tolerance for alcohol appears to decrease in old age. A decreased tolerance combined with age-related changes in balance, manual dexterity, and postural flexibility can increase the probability of falls, bums, or other accidents. Effects of alcohol can prevent serious heart conditions from being recognized. An additional concern is that while persons aged 65 or older make up about 10 percent of the U.S. population, this same group consumes 25% of the nation's prescription drugs.4 When drugs are combined with alcohol, there is potential for serious interaction, particularly when used by persons already in a vulnerable state of health because of physical decline. Alcohol/drug interactions lead to an estimated 2,500 deaths a year and 47,000 emergency room admissions a year.5

Two groups of elderly alcoholics have been identified by researchers.3,6-8 One group has been drinking most of their lives (early-onset alcoholics). The other group began drinking late in their adult lives (late-onset alcoholics).

The etiology of alcohol abuse and alcoholism in the elderly, as in younger people, is the subject of much debate. Horn, Wanberg and Adams,9 suggested that alcoholism is not a singular illness to be treated with a single form of therapy, but rather a label masking a variety of pathological conditions with different etiologies. Individual disciplines or professions have developed models and hypotheses in relation to alcoholism and then perpetuated these perspectives through their literature and research.10,11

Some of the more pervasive models of alcoholism are biomedica] models, psychological/psychiatric models, psychosocial models, and sociological models. The biomedicai conception of alcoholism views it as a progressive disease condition stemming from a deficit in body metabolism.12 This model has not been supported by research literature.13 One of the psychological/psychiatric models conceptualizes alcoholism as indicative of an underlying personality disorder.14-16 Problematic to this model is the fact that nearly every personality characteristic has been associated with alcoholism in one study or another.17

Another more psychosocial view is that one type of alcoholism is reactive to an overwhelming external stress.18 Closely related is a sociological model presenting alcoholism as a method of coping with pain and anger, used by people who are frustrated by the social system, in their attempts to attain their goals through socially acceptable means.19

Table

TABLE 1PERSONAL AND DEMOGRAPHIC CHARACTERISTICS OF BEREAVED AND NONBEREAVED SUBJECTS

TABLE 1

PERSONAL AND DEMOGRAPHIC CHARACTERISTICS OF BEREAVED AND NONBEREAVED SUBJECTS

The latter two models seem applicable to problems of the elderly. The elderly may be faced with many serious stresses: reduced economic resources, loneliness, poor health, and loss of significant others. Because our society provides few productive roles for its older citizens, it denies them the most usual, socially acceptable means of attaining the important goals of obtaining adequate income and feelings of usefulness. The literature on stress and aging has pointed to conjugal bereavement as the single greatest Stressor in terms of life change events.20 Thus, it would seem to follow that alcohol might be used by the elderly to deal with socially imposed frustrations and particularly to deal with the major stress of loss of a spouse.

Purpose of This Study

This study was part of a larger study of the reactions of older persons to the first year of widowhood.21 Data on the drinking behavior of older bereaved individuals and a comparison group of married older persons were analyzed for the purpose of answering the following questions:

1. What are the drinking patterns of older persons?

2. Is use of alcohol more frequent among recently widowed than among married older persons?

3. Does the amount and/or frequency of alcohol use increase subsequent to bereavement?

4. Are social resources, economic resources, mental health, physical health, or demographic characteristics associated with the drinking behavior of bereaved older persons?

Method

Subjects

The target population of bereaved older persons for this study was all surviving spouses of individuals dying during a five-month period who met the following criteria: a) the death was nonviolent and from natural causes; b) the deceased individual was 65 years of age or older, if male, and 60 years or older if female. (This allowed for the probability that men would be older than their spouses.)

Prospective subjects were identified through the death certificate of the deceased spouse recorded at the health department in a midwestern city where the study was conducted. Of the eligible subjects identified, 33% could not be located and 52% oY those located refused participation. This high rate of nonparticipation is consistent with that reported in other studies of newly bereaved individuals.22-24

Table

TABLE 2DISTRIBUTION OF DRINKING BEHAVIOR BY SEX FOR BEREAVED AND NONBEREAVED SUBJECTS

TABLE 2

DISTRIBUTION OF DRINKING BEHAVIOR BY SEX FOR BEREAVED AND NONBEREAVED SUBJECTS

To assess for systematic biases in characteristics of participants, data on race and sex of eligible subjects available from the death certificates were compared for participants and the total group of nonparticipants, including both those who could not be located and those who refused to participate. There was no. statistically significant difference by either race or sex.

The final study sample of bereaved older persons was comprised of 60 subjects, 42 women and 18 men. One man and eight women were black; the remaining 51 subjects were white. Subjects ranged in age from 58 to 83 years. Twenty-eight subjects (46.6%) were aged 70 years or older, a distribution closely resembling the population of this age group in the metropolitan setting where the study was conducted.

Nonbereaved subjects were noninstitutionalized, married couples, 58 years and older who were living together. Since bereaved subjects had been married until the last three months, married couples were thought to be a more appropriate comparison group than a cross-section of the general population which would include divorced and single persons. Nonbereaved subjects were identified through registry lists of three churches, two retirement centers, ten senior citizen centers and by personal referrals. Among 133 couples who were sent letters requesting participation, 27 could not be contacted by phone and 56 of those contacted refused to participate in the study, a refusal rate comparable to that of bereaved subjects. One woman refused the interview on the day of. appointment due to heat exhaustion, so she and her husband were also eliminated. The remaining 50 married couples comprised the study comparison group.

Table 1 shows the characteristics of both groups of subjects. Annual income of the subjects ranged from less than $1,999 to over $49,000. The median income category for bereaved subjects was between $5,000 and $6,999 while that for nonbereaved couples was between $7,000 and $8,999, a difference which might well be attributable to the loss of social security or retirement income of a spouse. As with income, the percentage owning their own homes was higher among the nonbereaved couples, although the difference was not statistically significant. A higher percentage of married couples were college graduates, although most of the college graduates were male. When we controlled for the effect of excess males among the two groups, they were comparable on all these variables.

Instruments

The mental health, physical health, social resources, and economic resources sections of the Duke University Older Americans' Resources and Services, Multidimensional Functional Assessment Questionnaire (OARS) served as the basic data collection instrument. These sections required approximately 30 to 45 minutes to administer. The OARS was tested extensively by its developers for validity, reliability, and ability to discriminate appropriately among groups. Ratings derived from the scales discriminated well and showed excellent agreement with ratings derived from both clinical interview data and professional examination. Interrater reliability among ten raters doing concurrent ratings ranged from .68 for the mental health rating to .83 for the physical health rating.25

Since depression is the major mental health problem anticipated in a bereaved population and some depressive symptomatology is commonly present with other psychiatric disorders26 the Zung Self-Rating Depression Scale (ZSDS) was used to quantify the intensity of depression in subjects. This scale, comprised of 20 items, rated one to four, is based on affective, psychological, and biological diagnostic criteria of depression. It takes about ten minutes to administer and produces a score range of 20 to 80. It is sensitive in differentiating depressive reactions from other disorders and correlates well (r=.79) with other scales, such as the Hamilton Rating Scale.27

The Short Michigan Alcoholism Screening Test (SMAST), a short version of the Michigan Alcoholism Screening Test, is a structured interview instrument reported to provide a consistent, quantifiable detection of alcoholism. These instruments were validated by obtaining independent evidence of problem drinking from records at medical facilities, social agencies, and police departments. The instruments do not produce false positives and the false negative rates are low, less than 5%.28

The SMAST was used in this research to identify "problem drinkers" in the bereaved study population. At three months of bereavement, subjects were asked about problems of drinking occurring before the death of their spouses. At nine months of bereavement, subjects were asked the same questions about problem drinking occurring after the death of their spouse. The type, frequency, and quantity of drinking, asked of both bereaved and nonbereaved populations, were measured by questions constructed similarly to those used by Cabalan et al29 in a national survey of drinking practices. This self-report of drinking was based on current drinking practices at the time of the interview.

Procedure

Prospective bereaved subjects who had been identified through the death certificates of their spouses were sent letters explaining the study and requesting their participation. The letter, mailed three months after the spouse's death, notified potential subjects that someone from the project would be contacting them within the week by telephone. A similar letter was mailed to prospective couples for the comparison group. Telephone contact was made by a graduate nursing student interviewer. Questions regarding the study were answered at this time and if the subject was willing to participate, a home interview was scheduled. Interviews for bereaved subjects were scheduled between three and four months after the spouse's death to avoid intrusion during the period of acute grief which immediately follows death. A follow-up interview, conducted only with bereaved subjects, was scheduled six months after the first interview.

Table

TABLE 3ZUNG DEPRESSION SCORES BY DRINKING CLASSIFICATIONS OF BEREAVED SUBJECTS

TABLE 3

ZUNG DEPRESSION SCORES BY DRINKING CLASSIFICATIONS OF BEREAVED SUBJECTS

Data were coded jointly by two nurse interviewers to increase the reliability of ratings. Subsequently, data were entered onto an SAS computer file for statistical analysis, using Chi- square tests, t-tests, and Spearman correlations. An alpha of 0.05 was accepted as statistically significant.

Results

Patterns of Alcohol Use Results

Prevalence of alcohol use among bereaved subjects was higher than that among the nonbereaved. Seventy-three percent of bereaved subjects used alcohol compared with 55% of the nonbereaved. Because alcohol use has been found in previous studies to be higher among males, drinking frequencies were compared in Table 2 for males and females in each group.

Differences in drinking patterns between bereaved males and females were statistically significant [x2(4) = 12.7; p<.02], but those between nonbereaved males and females were not. Only 5.6% of bereaved males abstained from alcohol use, contrasting with 46% of nonbereaved males [x2(4) = 13.96; p<0.01]. Although differences were not statistically significant, more abstainers were found among nonbereaved females than among bereaved females. However, while bereaved males who drank were more likely to drink daily, bereaved females who drank, only drank occasionally. Nonbereaved female drinkers, in fact, drank more frequently than the bereaved female drinkers. Drinking patterns for these nonbereaved females closely resembled those of their spouses. Not surprisingly, frequency of drinking was statistically significantly correlated with number of drinks in both groups (r=0.53;p<.001). Regular drinkers, in general, were also heavier drinkers.

Both bereaved and nonbereaved subjects preferred hard liquor to wine and beer. Fifty-three percent of the total combined sample drank predominantly hard liquor; 31% preferred beer and 16% wine. Bereaved and nonbereaved subjects did not differ significantly on type of alcohol used.

Factors Associated with Alcohol Use Among the Bereaved at Time 1

The following analyses involved only the bereaved subjects. Drinking behavior was classified originally into four categories: abstainers, normal drinkers, heavy drinkers, and problem drinkers, based on combined criteria using the quantity/frequency index and the SMAST. Normal drinkers were defined as drinking less than three drinks per day. Heavy drinkers were those reporting more than three drinks per day but scoring zero on the SMAST. Problem drinkers were those subjects scoring one or higher on the SMAST.

Table

TABLE 4CORRELATION COEFFICIENTS OF SELECTED VARIABLES OF BEREAVED SUBJECTS WITH CHANGES IN FREQUENCY OF DRINKING FROM PRIOR TO BEREAVEMENT TO 9 MONTHS POST LOSS

TABLE 4

CORRELATION COEFFICIENTS OF SELECTED VARIABLES OF BEREAVED SUBJECTS WITH CHANGES IN FREQUENCY OF DRINKING FROM PRIOR TO BEREAVEMENT TO 9 MONTHS POST LOSS

Only four bereaved subjects, however, reported drinking more than three drinks per day prior to the death of their spouse. Of these, only one did not score at least one on the SMAST, so he was categorized as a normal drinker and the category of heavy drinker was eliminated. One additional subject was classified as a problem drinker because of a score greater than one on the SMAST even though he did not report taking more than three drinks per day. (SMAST scores for problem drinkers range from one to three,) At the threemonth interview, 27% of bereaved subjects reported abstention from alcohol use. Sixty-seven percent were normal drinkers, and 6% (four subjects) were classified as problem drinkers.

No relationship was found between age, race, education, social resources, mental health, physical health, or economic resources ratings, and drinking patterns as reported for prior to bereavement. Sex was the only demographic variable related to drinking patterns at this time. Males comprised 6% of abstainers, 35% of normal drinkers and 75% of problem drinkers in a study population that was 30% male.

A statistically significant difference on use of alcohol [x2(1)=3.7; p<.05] between subjects with normal versus depressed scores on the Zung depression scale was observed (Table 3). Depressed subjects were less likely to drink than were subjects scoring normal on the depression scale.

Changes in Alcohol Use from Time 1 to Time 2

Change in drinking behavior between that reported for prior to bereavement and nine months of bereavement was assessed. No new cases of heavy drinking or problem drinking were found by nine months of bereavement. In fact, the prevalence of both decreased between that reported for prior to bereavement and for nine months postbereavement. Of the four subjects who were identified as problem drinkers prior to bereavement, two remained problem drinkers. One subject's SMAST score decreased from one to zero.

The other subject, who was identified as having problems related to drinking prior to bereavement, eventually refused to participate in the second interview at nine months of bereavement. The interviewer reported that his speech was slurred and that he sounded drunk each time she called to attempt to schedule the nine-month interview.

No new cases of heavy drinkers were identified by the quantity/frequency index at nine months postbereavement. The one who reported being a heavy drinker prior to bereavement reported a decrease in alcohol use (from three or four drinks daily to one or two drinks daily).

To evaluate overall changes in patterns of drinking reported prior to bereavement and that reported nine months after bereavement, a t-test for correlated measures was performed. Results indicated that there was not a significant change in the mean frequency of drinking or amount of drinking over time. However, the similarity in mean scores was largely due to equal change in both directions. Of the 55 subjects for whom data were available both prior to bereavement and nine months after, 57% did not change their use of alcohol; 16% increased use and 16% decreased use.

To determine if variables measured at three months postbereavement were correlated with these changes in frequency of drinking for subgroups, a Spearman's rank order correlation was calculated. Table 4 indicates these results. Three variables - sex, economic scores, and depression score - correlated significantly with changes in drinking. More males increased their frequency of drinking while more females decreased their frequency of drinking. Lower economic scores were associated with an increase in drinking. Finally, more subjects who were depressed increased their drinking.

Discussion

Prebereavement rates of alcoholic beverage use among bereaved female subjects in this study were identical (64%) to those reported in Barnes'30 study of widowhood and alcohol use in old age. The rate for the nonbereaved females in the current study (56%) was lower than for the bereaved, but still not as low as the rates reported in other surveys of older females. Johnson and Goodrich31 reported a rate of 47% drinkers among females while Rathborn-McCuan32 reported 38%.

Rates of alcohol use among bereaved males in this study (94.4%) were significantly higher than those reported in any other study including the widowed in Barnes' study (86%). 30 Rates of alcohol use among nonbereaved mates in this study (54%), however, were similar to those reported by Johnson and Goodrich31 and Rathborn-McCuan32 (56% and 61% respectively) for older male populations.

While the prevalence of alcohol use was higher among bereaved subjects than among nonbereaved, this study, like the study by Barnes,30 does not support the notion that bereavement is related to problem drinking. Also, like the Barnes study, this study provided no evidence that most bereaved subjects increase their use of alcohol during the first nine months after the loss of a spouse. Equal numbers of subjects increased and decreased alcohol use. Most reported no change.

By not supporting the notion that bereavement is related to problem drinking, these findings provide indirect support for the early-onset theory which suggests that drinking among the elderly is, in general, behavior that has developed over a lifetime in response to social norms, psychological and biological factors. This notion is further supported in the finding that moderate alcohol use was associated with depression scores in the normal range, while abstinence was associated with elevated scores.

It is probable that alcohol is enjoyed in the company of friends, a pattern of socializing which is likely to have developed in years prior to death of the spouse. In this case, drinking may reflect meaningful networks which contribute to keeping depression at bay. Such social drinking may be reflected in the similar drinking patterns observed among the husbands and wives in the comparison group.

The fact that bereaved men, those bereaved who were in poor economic straits, and those who scored positive for depression on the Zung depression scale at three months were more likely to increase their drinking in the subsequent six months suggests that there in fact, may be a high-risk subgroup of elderly who might become late-life alcohol abusers.

The finding of a subgroup likely to increase their alcohol use is consistent with Rosen and Glatt's (1971) findings thai while most of their alcoholic subjects evidenced long-standing personality problems, one third of them did begin excessive drinking in response to stresses of old age. Dupree, Broskowski and Schonfeld33 also reported a group of later-life onset users whose alcohol abuse increased following cumulative losses and stresses associated with aging. Further research is indicated to determine whether these factors consistently are associated with an increase in alcohol use. Additionally, a longer period of study is needed to ascertain if this increased drinking becomes problem drinking.

With the rapid growth of our elderly population, the increasing population of widowed elderly, and the high rate of alcohol abuse (10% to 15%) among the elderly who seek medical attention, continued research is particularly important. Research might focus on the natural history of drinking habits, looking particularly for what gives rise to changes, especially increased drinking in late life. Excessive alcohol use in a population which has a higher incidence of impaired health and which takes 25% of all prescription drugs, makes for a potentially lethal combination. Epidemiological research is badly needed to supplement the clinical observations about alcoholism and its treatment in the elderly.

In the meantime, nurses have an opportunity to identify potential problems associated with alcohol use among the elderly and to intervene to prevent their development. Use of alcohol should be routinely included in patient assessments. Patients need to be taught about possible interactions of their medications with alcohol, the decreasing tolerance for alcohol which often accompanies aging, and how alcohol can exacerbate prior physical conditions of the elderly.

Examples of prior physical conditions include impairments of musculoskeletal coordination, slowed reaction time, and balance problems. By providing calories and lessening appetite, heavy use of alcohol may substitute for a substantial portion of nutritional intake and contribute to a worsening nutrition status for older persons already experiencing poor appetite. On the other hand, moderate alcohol use in the company of friends at mealtime may contribute to an atmosphere of congeniality conducive to an improved appetite.

In regard to the elderly bereaved, assessment of coping response can identify those with abnormal grief reactions. Assessment should include: 1) the meaning of the spouse's death; 2) whether the loss is perceived realistically; 3) prior experience in coping with grief and effectiveness of coping responses in those instances; 4) whether the current episode triggered unresolved grief episodes from the past; and 5) the nature and quality of the available support system. Frequent ongoing complaints of a psychosomatic type or frequent visits to the doctor for treatment of such symptoms may be one sign of unresolved grief. An increasing tendency to withdraw from social interactions is another.

Support during the bereavement period can be offered through a variety of interventions including helping the widowed person to explore strengths and weaknesses, set short-term goals and by providing appropriate referrals. Such interventions may contribute to improved coping and less need for use of counterproductive coping methods such as use of alcohol to escape the psychic pain.

References

  • 1. National Institute on Alcohol Abuse and Alcoholism. Fourth Special Repon lo the US Congress on Alcohol and Health . US Dept of Health and Human Services, 1981.
  • 2. Zimberg S, Wallace J, Blume S: Practical approaches to alcoholism psychotherapy. New York, Plenum Press, 1978.
  • 3. Schuckit M, Miller P: Alcoholism in elderly men: A survey of a general medical ward. Annals of New York Academy of Sciences 1976:273:558-571.
  • 4. Basen M: The elderly and drugs - Problem overview and program strategy. Public Health Reports 1977; 92(1):43-48.
  • 5. Federal Drug Administration. FDA Drug Bulletin, 9, 10, 1979.
  • 6. Zimberg S: The elderly alcoholic. Gerontologist 1974; 14(4):221-224.
  • 7. Rosin A, Glatt M: Alcohol excess in the elderly. Quarterly Journal of Studies on Alcohol 1971; 32(1):53-59.
  • 8. Cohen S: Drug abuse in the aging patient. J Stud Alcohol 1976; 37:1455.
  • 9. Horn J, Wanberg K, Adams G: Diagnosis of alcoholism. Quarterly Journal of Studies on Alcohol 1974; 35(1): 147-174.
  • 10. Betemps E: Models of alcoholism in nursing textbooks. Unpublished master's thesis, University of Cincinnati, 1978.
  • 11. Keller M: Multidisciplinary perspectives on alcoholism and the need for integration. J Stud Alcohol 1975; 36(1):133-147.
  • 12. Jellenek EM: Phases of alcohol addiction. Quarterly Journal of Studies on Alcohol 1952; 13(4):673-684.
  • 13. Gitlow ES: Alcoholism: A disease, in Bourne, PG Fox, R (eds): Alcoholism Progress in Research and Treatment. New York, Academic Press, 1973, pp 1-9.
  • 14. Bennet AE: Alcoholism and the Brain. New York, Stratton Intercontinental Medical Book Corporation, 1977.
  • 15. Johnson CW: Drugs and alcohol abuse, in Johnson CW, Snibbe JR, Evans LA (eds): Basic Psychopathology: A Programmed Text. New York, Spectrum Publications, Inc, 1975, pp 270-297.
  • 16. Tamerin JS, Neumann CP: The alcoholic stereotype: Clinical appraisal and implications for treatment. Am J Psychoanal 1974; 34(4):315-323.
  • 17. Williams AF: The alcoholic personality, in Kissin B, Begleiter H (eds): Social Aspects of Alcoholism. The Biology of Alcoholism. New York, Plenum Press, 1976, vol 4, pp 243-274.
  • 18. Freedman AM, Kaplan HI, Sadock BJ: Comprehensive Textbook of Psychiatry. Baltimore, Williams and Wilkins, 1975.
  • 19. McKee M, Robertson J: Social Problems. New York, Random House, 1975.
  • 20. Amster L, Kraus H: The relationship between life crises and mental deterioration in old age. IM J Aging Hum Dev 1974; 5(1):51-55.
  • 21. Yeaworth RC, Valanis B: Health status and resources of recently bereaved older persons. Public Health Nursing 1985; 2(4):232-244.
  • 22. Clayton PJ: Mortality and morbidity in the first year of widowhood. Archives of General Psychology 1977; 30(6): 747-750.
  • 23. Maddison D, Viola A: The health of widows in the year following bereavement. J PsychosomRes 1968; 12(1):297-306.
  • 24. Parkes CM, Brown RJ: Health after bereavement: A controlled study of young Boston widows and widowers. Psychosom Mea 1972; 34(5):449-461.
  • 25. Fillenbaum J: Reliability and validity of the OARS multidimensional functional assessment questionnaire, in Pfeiffer E (ed): Multidimensional Functional Assessment: The OARS Methodology. Durham, NC, Duke University Center for the Study of Aging and Human Development, pp 25-34.
  • 26. Zung WW: From art to science: The diagnosis and treatment of depression. Arch Gen Psychiatry 1973; 29:328-337.
  • 27. Brown GL, Zung WW: Depression scales: Self on physician rating? A validation of certain clinically observable phenomena. Compr Psychiatry 1972; 13:361-367.
  • 28. Selzer M: The Michigan alcoholism screening test: The quest for a new diagnostic instrument. Am J Psychiatry 1971; 127, 89-94.
  • 29. Cabalan D, Cisin 1, Crossley H: American drinking practices: A national survey of drinking behaviors and attitudes. Monograph #6. New Brunswick, Rutgers Center for Alcohol Studies, 1969.
  • 30. Barnes G: Alcohol use among older persons: Findings from a western New York State general population survey. J Am Geriatr Soc 1979; 27(6):244-250.
  • 31. Johnson L, Goodrich C: Use of Alcohol by Persons 65 Years and Over, Upper East Side of Manhattan, (Report to the NIAAA, Mount Sinai School of Medicine). New York, City University of New York.
  • 32. Rathborn-McCuan E: Community Survey of Aged Alcoholics and Problem Drinkers. Report to NIAAA. Baltimore, Levindale Geriatric Research Center, 1976.
  • 33. Dupree LW, Broskowski H, Schonfeld L: The gerontology alcohol project: A behavioral treatment program for elderly alcohol abusers. The Gerontologist 1984; 24(5 ):510-516.
  • 34. National Institute on Aging. Aging and alcohol. US Dept of Health and Human Services, 1981.

TABLE 1

PERSONAL AND DEMOGRAPHIC CHARACTERISTICS OF BEREAVED AND NONBEREAVED SUBJECTS

TABLE 2

DISTRIBUTION OF DRINKING BEHAVIOR BY SEX FOR BEREAVED AND NONBEREAVED SUBJECTS

TABLE 3

ZUNG DEPRESSION SCORES BY DRINKING CLASSIFICATIONS OF BEREAVED SUBJECTS

TABLE 4

CORRELATION COEFFICIENTS OF SELECTED VARIABLES OF BEREAVED SUBJECTS WITH CHANGES IN FREQUENCY OF DRINKING FROM PRIOR TO BEREAVEMENT TO 9 MONTHS POST LOSS

10.3928/0098-9134-19870501-08

Sign up to receive

Journal E-contents