In the past decade, considerable research has been done that explicates the effects of stress on the mental and physical health of adults. Some of these studies have used older populations.1"8 Although several studies have attempted to explain gender differences in the stress/health relationship of elderly persons, few have examined the combined effect of marital status and gender on health.
Currently in the US, the majority of people over age 65 are female. Of these, most are widowed. Many, however, are married and living with retired elderly spouses. For these women, the stress of losing a mate has not occurred, yet their roles as wives in a marital environment combined with the events of aging may be more stressful and detrimental to their health than the role of widows. Put more succinctly, the crisis of widowhood, once endured and coped with, may be conducive to less stress or better coping and health. Because the experience of stress may affect an elderly married woman differently than an elderly widow, nurses who work with older persons need to modify their approaches to address these special circumstances.
This study was undertaken to explore the differential effects of stress on elderly married women. An additional goal was to identify interventions that could be used in gerontologicai nursing practice to promote effective coping in this same group.
Roy9 has suggested that interaction with the environment leads to the experience of stress. As a person experiences stress, his/her sense of personal competence is threatened, requiring an effort on the part of the individual to adapt to the Stressor and return to a state of self-adequacy and stability. According to Roy's theory, an individual has four behavioral modes for responding to Stressors: physiological, selfconcept, role performance, and interdependence.
Roy's theory has particular relevance for elderly married women. Menopause, which signals the end of the reproductive years, is a physiologic change that often precipitates other physical challenges to the musculoskeletal and circulatory systems. In addition to shaping an elderly woman's perceptions of physical wellbeing, menopause is an example of a physical alteration with the power to bring about changes in sexual identity and other psychological realms as well.
According to Roy,9 self-concept is a person's belief about himself/herself. Often, self-concept is closely intertwined with role performance, or behavior that is linked to feelings of social adequacy.9 Since married older women are the least likely group to have worked outside the home,10 they are in a dependent position. Although they do not have to deal with retirement first-hand, elderly wives must cope with the work cessation of their husband, a person they have relied on for financial security. A husband's employment may also have represented a source of socialization for the couple that will be lost after retirement. During this same period, the elderly wife may be witnessing other changes in her life situation that affect her selfconcept. The "empty nest syndrome" and grandparenting are other experiences that affect elderly women's views of themselves and their interactions with others, since "losing" one's children and becoming a grandparent both alter one's identity as mother and wife.
Roy defines interdependence as the need of all persons for nurturing relationships.9 As a group frequently called on to provide home care for an ill spouse, elderly wives often experience dramatic changes in their marital relationship as a consequence of the need to become caregivers. Even if both spouses are healthy and there is no immediate need for caregiving, the potential for illness of the husband is still a very real threat. Duvall11 has suggested that a developmental task for older families is preparing for the loss of a spouse. As the more dependent person in the marital relationship, and the one likely to survive the longest, this task can be extremely difficult for an elderly woman.
Many of the changes of later life are of a profoundly stressful nature. On a closer examination of these changes, it is older wives who are likely to undergo the most serious threats to selfconcept, role performance, and interdependence needs.
The data were collected as part of the second stage of a project titled "Service Delivery Arrangements for Rural Elderly in the Northeast." Eighteen minor civil divisions (MCDs) were randomly selected from six Northeastern states: Pennsylvania, New Jersey, Vermont, Maine, New Hampshire, and West Virginia. A combination of stratification, quota, cluster, and probability sampling was used to select these sites.12 Subjects were randomly selected through telephone exchanges (the first three digits of the number after the area code) in each site. One method used was random digit dialing. To accomplish this, telephone exchanges were identified for the MCD of concern.
A computerized list of randomly selected 4-digit numbers was generated for each exchange. Interviewers called each of the resulting 7-digit numbers, discarding those numbers that were not working or at which there were no elderly. In some instances, the telephone exchanges did not coincide with the MCD that had been selected for study; ie, the exchange overlapped into neighboring MCDs not identified for study. In these cases, a second method was used that involved the generation of a computer file of all telephone directory listings for each MCD. Commercial listings were then deleted and the remaining listings were randomly ordered by a computer, which assigned random numbers to each listing. Interviewers then called the numbers sequentially until elderly willing to be interviewed were identified.
Fifty respondents were interviewed in each selected MCD for a total of 900 non-institutionalized persons aged 65 years and older. In all, 6,585 telephone numbers were called to complete the 900 interviews. A refusal rate of 24% represented a combination of the number of elderly who refused to be interviewed and the number of family members who refused on behalf of the elderly living with them.
The dependent variable, health status, was measured subjectively. This decision was based on several studies13*16 that found subjective health measures (how subjects rate their own health) were as accurate as objective measures (physician's assessments) in defining an individual's health status. Using a broad definition of health as one that includes physical and mental components, subjects were asked to describe their physical health at the time of the interview as very good, good, fair, poor, or very poor. In addition, they were asked to identify whether their health inhibited their daily activity and, if it did, to what degree. As a measure of their mental health, elderly respondents were asked how frequently they felt a great deal of strain connected with their daily activities and how often they felt tense and nervous. Finally, respondents were asked, on a scale from O to 10 (O = very dissatisfied and 10 = very satisfied), how satisfied they were with life.
A factor analysis specifying two factors was then conducted on the six health measures. Only one factor emerged, which indicated that these items were associated with only one dimension. Cronbach's alpha was then used to test the reliability of all six measures. A value of .87 was found, indicating a high degree of reliability. Based on this information and using SPSS sub-program Facscore,17 the six health measures were combined into a single weighted factor score for each subject (mean = O, SD = 1).
Stress was measured subjectively based on research by Tausig18 and Dohrenwend et al,19 which suggests that, for undesirable life events, there is little difference between stress measured objectively and subjectively. Following a proposition by Dohrenwend and Dohrenwend20 that stressful events occurring in the past might not produce the same stress levels as recent events, respondents were asked to identify those events that had occurred only in the past year. Accordingly, elderly subjects were asked whether in the past year they had experienced stress with their families and friends. If respondents answered "yes" they were then asked to rate the degree to which these Stressors upset them (not very upsetting, somewhat upsetting, very upsetting). Information on more specific events was not solicited; ie, respondents were not asked more specific questions about family/friend stress. This rationale was based on pilot research that suggested that specific lists of life events considered stressful by the elderly (death of a spouse, death of a friend, problems with children or grandchildren) could be grouped into broader categories without weakening their predictive ability.21 This pilot study also indicated that the measurement of the degree to which the elderly were upset by these categories on a three point scale (not very upsetting, somewhat upsetting, and very upsetting) was more predictive than a scale that rated the intensity of stress from O to 10.
DISTRIBUTION OF HEALTH AND STRESS
THE STRESS/HEALTH RELATIONSHIP BY GENDER FOR UNMARRIED ELDERLY
Preliminary regression analysis of stress and health on all 900 subjects was conducted. Results indicate that, as in previous research, there was a negative relationship between stress and health for these respondents; ie, for all subjects, high stress was predictive of poor health.22
Accordingly, all 900 respondents were categorized into four groups: those with poor health and low stress, poor health and high stress, good health and low stress, and good health and high stress. A cross tabular analysis was conducted and the four stress/ health categories were analyzed to test the effects of gender and marital status on each category. Statistical significance was tested using chi squares for contingency with three degrees of freedom.
To better determine the marital conditions under which this relationship might be occurring, the researchers decided to group the data according to all possible combinations of stress and health and then combine them by marital status and gender.
Ages of the 900 respondents varied from 65 to 94 years. Sixty-five percent (n = 589) were women and 35% ( n = 3 1 1 ) were men . fbrty-f ive percent were married and 64% were retired. Fifty percent had less than a high school education whereas 20% were educated beyond secondary school. The majority of married elderly respondents (54%) were men. The majority of the non-married elderly respondents (91%) were women.
Table 1 indicates that most of these elderly were healthy (ratings of good or very good and no or very little limitations of health on daily activity), satisfied with their lives (73% indicated high satisfaction), and more than two thirds indicated that they had had little or no stress with their families and friends in the past year.
The results of the cross tabular analysis are presented in Tables 2 and 3. Findings suggest that for unmarried men and unmarried women, the stress/ health relationship is similar; ie, there are no significant differences for men and women by the four stress/health categories. For married elderly respondents, however, the situation is different. Here, it can clearly be seen that the health status of elderly married men is unrelated to stress, and that of all four groups of elderly persons in this sample, the married women experienced the poorest health and the highest stress. These findings support our hypotheses that the stress/health relationship varies in elderly populations due to the combined effects of gender and marital status.
DISCUSSION AND NURSING IMPLICATIONS
The data presented in this study indicate that, for the most part, elderly respondents were in good health and experienced relatively low levels of stress. The most significant findings of this study, however, were those that indicate that married women were in the poorest health and the most vulnerable to stress. Roy's9 theory that explicates four modes of response to Stressors (physiological, self-concept, role performance, and interdependence) provides a framework for the discussion of these findings and suggests a rationale for nursing interventions.
THE STRESS/HEALTH RELATIONSHIP BY GENDER FOR MARRIED ELDERLY
According to Roy ,9 nursing interventions begin with a clear statement of goals to be achieved. These goals reflect mutual efforts between the nurse and client to either maintain or enhance the level of adequacy, For elderly wives, an appropriate goal would be to identify ways to change coping behaviors needed for adaptation to the Stressors of later life. Roy's9 diagnostic category III, which refers to achieving awareness, developing an understanding of the stressful event, and selecting effective coping strategies, is most suitable for this group. Nursing actions such as collaboration (working with other health-care professionals to promote physiological coping), support (reinforcing client coping behaviors), teaching, and enabling (giving clients the strength and energy they may lack) are all used in working with each of the modes where elderly wives are likely to experience stress.
Although menopause is likely to have occurred for most women prior to the later years, other postmenopausal physical problems can begin at this time. Atherosclerosis and osieoporosis, two possible problems exacerbated by estrogen deficiency, are often experienced by women in the later years.23 Either of these changes cause very real physical discomforts that may lead to both physiologic and emotional stress in older women. Estrogen replacement therapy is increasingly used to offset postmenopausal complications. Nurses can inform women of the availability of this treatment, discuss the appropriateness of its use with the primary care provider, and monitor the effectiveness of the medication, if used. Health teaching about other measures that play a role in prevention of osteoporosis and heart disease, such as nutrition and exercise, can also be implemented by the nurse.
Although it is true that both married and unmarried women are likely to encounter postmenopausal difficulties, the marriage relationship itself may exert an influence on the way in which an older woman responds to these problems. Feigenbaum24 has remarked on differences in the termination of the reproductive phase of life between men and women. For men, this process is much more gradual, whereas for women, cessation in menstruation as well as concomitant signs of physical aging occur rather suddenly. Kinsey25 discovered that libido seems to fluctuate with age and sex as well. Whereas the men in his study desired intercourse more frequently in the earlier years, women reported an increased level of arousal in later life. For the married older woman, signs of the aging process that appear more visible in her than her spouse, possibly accompanied by sexual disinterest on the part of her husband, may have a profound negative effect on her self-esteem as well as physical well being.
Spier26 suggests that older adults may be reluctant to discuss topics of a sexual nature as a consequence of societal attitudes towards sexuality in the later years. Nurses who demonstrate a willingness to talk about a variety of topics related to physical and emotional love among older persons can serve as a valuable resource for elderly women experiencing such problems. Discussion groups dealing with sexual issues for older wives only can provide an open and relaxed atmosphere; however, husbands should be included in sexual counseling as well. Increasing longevity has also lengthened the marital relationship, therefore it is important to provide older couples with options for developing mutually satisfying interactions regardless of their age.
Self-Concept and Role Performance
For older married women, nurses should initiate discussions of the marital relationship and its impact on each spouse's life, Fengler27 has noted that the period after a husband's retirement is often one of great difficulty for wives who have not worked outside the home; this is a time when husbands are struggling to adjust to new roles and may also be physically present more often than in the past. In this case, nurses can help the older woman make a transition into new roles for both herself and her spouse. If a woman is particularly stressed by her husband's post-retirement adjustment, methods for dealing realistically with these struggles can be explored. Ideally, wives should be included in all pre-retirement counseling their spouses receive.
It is also important to explore the status of an older married woman's relationship with her children and grandchildren, if appropriate. Encouraging the expression of feelings about parenting, responses to children becoming independent, and the arrival of grandchildren are topics that can provide information on the elderly woman's feelings about herself as a mother. It may be useful to include children in these talks or to find ways in which the older woman can initiate such conversations on her own. An elderly mother may need to be given permission to verbalize negative feelings about grandparenting, since not every individual is enthusiastic about this role.28
In keeping with the findings of this research, health promotion and health education efforts geared at elderly persons needs to incorporate a focus on the role of social support. Assessing past patterns by which an individual achieved social intimacy will help identify interventions that can promote achievement of the same relationships in the later years. Since number of confidants contributes to good health, ways of maintaining old friendships and establishing new ones should be explored with both elderly men and women. Many aging persons experience a loss of important relationships, and nurses can help them deal with the grief process and suggest new social avenues to explore. Reminiscence groups can help elderly individuals explore feelings of loss and loneliness in relation to past friendships.29
Sokolovsky and Cohen30 found that elderly persons living alone (both men and women) often had larger than average social networks but had structured these networks in a way that allowed independence. These researchers make the point that being alone is not synonymous with being lonely. Likewise, being married does not mean that a woman's social life is fulfilling. Cool and McCabe31 propose that an elderly woman's powerlessness and lack of status is indicated by her relative invisibility in American society. For married elderly women, this statement may be even more true, because whatever recognition they may have gained is likely to have been achieved through their husbands.32 As spouses, many older women are unlikely to develop independent identities. Nurses, who are part of a predominantly female profession, can convey attitudes that either value or devalue the contribution of older women to society. Acceptance of the older woman as an individual separate from her husband is an important distinction that nurses need to make. Exploring skills that an older woman has or would like to develop can signal a belief in her ability to make worthwhile contributions to society.
The later years are also often a time when women are called on to provide physical or emotional care to an ailing spouse. This task has been shown to exert a detrimental effect on psychological health in a number of studies33-34 and its effect on physical health is still being investigated. Nursing interventions, such as participation in support groups and respite care programs as well as education, can help older wives learn to cope with providing care for an ill husband.
The nursing profession is in an ideal position to help improve the wellbeing of older wives. Engaging behaviors to help elderly married women achieve and strengthen their roles have special meaning in the case of aging wives, who can experience a positive sense of adaptation in their later years with the help of concerned professional nurses.9
In addition to implementing collaborative, supportive, and teaching interventions to improve the coping effectiveness of elderly women, gerontological nurses can educate others about coping and stress of married women in the later years. Nurse investigators should continue research that further explores the dynamics of the marital relationship for older individuals.
Bowman35 has identified long-term care as a particularly relevant issue for all women. As the number of older persons continues to increase in this country, nurses need to be the group of health-care professionals who aggressively advocate on behalf of the elderly. Promotion of healthy coping and adaptation in elderly wives offers an excellent opportunity for such advocacy.
- 1 . Arling G. Strain, social support, and distress in old age. J Gerontology. 1987; 42:107113.
- 2. Cohen Cl. Teresi J, Holmes D. Social networks, stress, and physical health: A longitudinal study of an inner city elderly population. J Gerontology. 1985; 40:478-486.
- 3. Fenwick R, Barresi CM. Health consequences of marital status change among the elderly: A comparison of cross-sectional and longitudinal analysis. J Health Soc Behav. 1981; 22:106-116.
- 4. Gray RM, Baker J. Kesler JP. Stress and health in later maturity. J Gerontologv. 1965; 20:65-68.
- 5. Krause N. Understanding the stress process: Linking social support with locus of control beliefs. J Gerontology. ?987;42:589-593.
- 6. Norris FH, Murrell SA. Protective function of resources related to life events, global stress, and depression in older adults. J Health Soc Behav. 1984; 25:424-437.
- 7. NorrisFH.MurrellSA.Transitoryimpactof life event stress on psychological symptoms in older adults. J Health Soc Behav. 1987; 28:197-211.
- 8. Thomas H. Expected unchangeability of life stress in old age - A contribution to a cognitive theory of aging. Human Development. 1981; 24:229-239.
- 9 . Roy C . Introduction to Nursing: An Adaptation. Englewood Cliffs, NJ: Prentice-Hall, Ine; 1984.
- 10. McConnell C, Manesco M. Gerontological Nursing: Concepts and Practice. Philadelphia: WB Saunders; 1988.
- 11. Duvall EM. Marriage and Family Relationships, 5th ed. Philadelphia: JB Lippincott; 1977.
- 12. Crawford CO. An Examination of Selected Dimensions of Independent Living Among a Sample of 900 Older Persons (Bulletin No 864). University Park, Penn: Pennsylvania State University; 1987.
- 13. Fillenbaum, GG. Social context and selfassessments of health among the elderly. J Health Soc Behav. 1979; 20:45-51.
- 14. Jylha M, Leskinen E, Alanen E, Leskinen A, Heikkinen E: Self-rated health and associated factors among men of different ages. J Gerontology. 1 986; 41:710-717.
- 15. Maddox GL, Douglass B. Self-assessment of health: A longitudinal study of elderly subjects. J Health Soc Behav. 1973; 14:8792.
- 16. Tessler R, Mechanic D. Psychological distress and perceived health status. J Health Soc Behav. 1978; 19:254-262.
- 17. Nie NH, Hull CH1 Jenkins JG, Steinbrenner K, Bent DH. Statistical Package for the Social Sciences. New York: McGraw HUl; 1975.
- 18. Tausig M. Measuring life events. J Health Soc Behav. 1982; 23:52-64.
- 19. DohrenwendBS, Askenasy AP, Krasnoff L, Dohrenwend BP. Exemplification of a method for scaling life events: The Peri Life Events Scale. J Health Soc Behav. 1978; 19:205-229.
- 20. Dohrenwend BS, Dohrenwend BP. Some issues in research on stressful life events. J NervMentDis. 1987:42:107-113.
- 21. Mansfield PK, Preston DB, Crawford CO. Health Stress and Coping Patterns in a Sample of Rural Elderly in Central Pennsylvania. Proceedings of the Second Annua! Blueprint Conference; 1983: Philadelphia.
- 22. Preston DB. The Effects of Marital Status. Sex Roles. Stress. Coping, and Social Support on the Health of the Elderly. University Park, Penn: Pennsylvania State University; 1984.
- 23. Hogstel M. Nursing Care of the Older Adult. New York: John Wiley and Sons; 1981.
- 24. Feigenbaum E. Sexual behavior in the later years. In: Kalish R, ed. The Later Years: Social Applications of Gerontology. Monterey, Calif: Brooks-Cole Publishing Co; 1977.
- 25. Kinsey A. Sexual Behavior in the Human Female. Philadelphia: WB Saunders; 1953.
- 26. Spier B. Developmental tasks of the aged. In: Yurik A, Robb S, Spier B, Ebert N, eds. The Aged Person and the Nursing Process. New York: Appleton-Century-Crofts; 1980.
- 27. ftngler AP. Attitudinal orientation of wives toward their husband's retirement. Int J Aging Hum Dev. 1975; 6:139.
- 28. Ebersole P, Hess P. Toward Healthy Aging: Human Needs and Nursing Response. St. Louis: CV Mosby; 1981.
- 29. Burnside L Working With the Elderly: Group Process and Techniques. North Scituate, Mass: Duxbury Press; 1978.
- 30. Sokotovsky J, Cohen C. Networks and adaptation: The cultural meaning of being a "ioner" among the inner city elderly. In: Sokotovsky J, ed. Growing Old in Different Societies: Cross Cultural Perspectives. Belmont, Calif: Wadsworth Publishing Co; 1983.
- 31. Cool L, McCabe J. The "scheming hag" and the "dear old thing": The anthropology of aging women. In: Sokolovsky J, ed. Growing Old in Different Societies: Cross Cultural Perspectives. Belmont, Calif: Wadsworth Publishing Co; Î983.
- 32. Murray R. Huelskoetter M. Driscoll O. The Nursing Process in Later Maturity. Englewood Cliffs, NJ: Prentice Hall Ine; 1980.
33. ArchbaldP. Impact of parent caring on middle-aged offspring. Journal of Gerontologie Nursing. 1980:6:78-85.
- 34. Cantor M. Strain among caregivers: A study of experience in the United Slates. Gerontologisl. 1983:23:597-604.
- 35; Bowman J. Long-term care is a woman's issue. American Nurse. 1988; 6:6- 13.
DISTRIBUTION OF HEALTH AND STRESS
THE STRESS/HEALTH RELATIONSHIP BY GENDER FOR UNMARRIED ELDERLY
THE STRESS/HEALTH RELATIONSHIP BY GENDER FOR MARRIED ELDERLY