Experience in coping with stroke: a survey of caregivers.
Families often lack consistent, comprehensive information to enable them to effectively cope with the family member recovering from a stroke after discharge from the hospital. The degree to which the family adapts can significantly affect the outcome. Studies of this nature are of increasing importance to all health care professionals as they face the graying of America with maturation of the baby boomers of the World War Il generation, and the concomitant increase in the prevalence of cardiovascular accident.
A family's ability to cope with a chronic disease can be an important factor in adaptation and rehabilitation. Power (1985) examined the family coping behaviors in multiple sclerosis and recommended that "What the family understands about the illness should be assessed early in the intervention for the imparting of appropriate information can become a useful coping mechanism for family members" (p. 82). But can this recommendation apply to stroke families, what kind of information is needed, and when is the most appropriate time to begin this process of family education? The relationship of family coping ability and the amount of information imparted to the family during hospitalization has not been investigated, nor has a comprehensive assessment of the needs of family members been conducted to determine the information needed to enhance the family's coping abilities.
The merit of family education about stroke residuals has been discussed but the delivery system was often inconsistently administered to patients and their families (Singler, 1975; McCormick and Williams, 1979; Crossman, et al. 1981). The merit of family education to stroke families in residential treatment programs has been well-documented as well (Strand, et al., 1985; Evans and Held, 1984; Dunn, et al., 1984; Tuckett and Williams, 1984; Jackson, 1984; Sahs and Hartman, 1976).
Other researchers have chronicled psychosocial difficulties which were experienced by persons with aphasia and their families (Biorn-Hansen, 1957; Malone, 1969). Families of persons with aphasia have been generally considered an integral part of the rehabilitation team because it was recognized that it was the family who had the greatest effect on the patient's motivation and attitude and who assumed the burden for long term care (Wepman, 1951). Malone (1969) observed, "The family cannot function as positive members of the rehabilitation team until they have been educated to the many and varied problems associated with aphasia and given some help in coping with these problems" (p. 151). While family education has been recommended, Linebough and Young-Charles (1978) conducted the only study to date that examined the informational needs of the families of aphasic patients. While the study was limited to families of patients with aphasia, Linebough and YoungCharles recommended farther investigation into the information needs of all stroke families so that differential counseling strategies may be developed. A study prepared by Field, Cordle, and Bowman (1981), challenged the positive relationship between information given to stroke family members during the patient's hospitalization and ability to cope with the problems associated with the return home. Field et al. (1981) concluded that success was limited to the practical management activities of care, and that psychological coping was not aided significantly. They Field et al. 1981) also recognized some methodological problems in this study, relating primarily to inadequate sample size, the plurality of staff involved in the study, and limited follow-up. More importantly, the content of the information given to the family members in the study was not specified.
If family information was not effective in facilitating coping or family adjustment, perhaps the content of the information was not geared toward the perceived needs of the families the scope of which has not been identified in the literature. The focus of this present study has been directed to these issues. Method
This study sought to determine if there is a correlation between family education for the family of a stroke patient prior to discharge from the hospital and family adjustment. A needs assessment study was designed to gather and report descriptive survey data from a group of family members of patients who had been hospitalized in an acute care facility. Data collected from the survey instrument was analyzed and explored for relationships between severity of stroke, information given during hospitalization, and coping with the situation after discharge from the hospital. Institutional setting
The formal study was conducted by mail from the Speech Pathology and Audiology Service of Hillcrest Baptist Medical Center, Waco, Texas. Attempts to extend the study to multiple locations were not successful due to legal difficulties, particularly with regard to protection of patient confidentiality, and difficulty locating facilities that are alike in aU other dimensions. Survey instrument
The primary design consideration involved the mode of administration and item construction.
The survey instrument used in the study was designed to be mailed--self-administered with no interviewer present. A mail survey was selected as it allowed family members anonymity. To help alleviate any ambiguity in the questionnaire, telephone contact with the researcher was encouraged in the letter of transmittal.
The questionnaire was constructed with the consultation of professional colleagues in the field of stroke rehabilitation. Items were selected from existing pertinent literature and were pretested to eliminate ambiguity.
Aside from content and length consideration with inherent time and cost constraints, the survey instrument utilized a combination of open and closed-ended questions so that the information could be coded and reduced into manageable categories. Pilot testing
A pilot test of 20 families and seven cross disciplinary colleagues involved with stroke rehabilitation was conducted. Following the recommendations of the respondents, the questionnaire was substantially revised by rephrasing questions, regrouping questions of a more logical sequence, and adding an entire section of questions on severity of stroke. Subjects
The population of this study was a sample of 166 stroke families whose affected member was hospitalized at Hillcrest Baptist Medical Center from September 1983, to September 1984, with a primary or secondary diagnosis of CVA, cerebral infarction, cerebral emboli, or cerebral hemorrhage. The primary service area of Hillcrest was McLennan County, Texas, accounting for over 85% of all admissions. According to data from the U. S. Department of Health, the rate of discharged patients with cerebrovascular disease was approximately 3.7/1,000 population (United States Department of Health and Human Services, 1984). Therefore, the estimated incidence of stroke for McLennan County for 1985 was approximately 450 patients for the twelve-month period under consideration. This study surveyed over one-third of that population and should be considered representative of a broader geographic area. The subjects of this study were those caregivers who responded to the survey (N = 86, Females: 65, Males: 21). Data gathering
The survey instrument, "Experience in Coping with Stroke," was used to gather the data from the stroke family members in the study. The questionnaire was mailed with a cover letter and a self-addressed, stamped envelope, requesting a reply within a designated two-week period. After two weeks, an attempt to contact all non-respondents by telephone was made. Of the 66 non-respondents, 37 were contacted by telephone: 28 indicated that they Will send in" the questionnaire, and 9 said they were "not interested". The remaining 29 non-respondents could not be reached by telephone and were mailed a second questionnaire. Of those questionnaires returned due to wrong address, alternate addresses of caregivers indicated in the patient's medical chart were obtained, and the questionnaires were forwarded. A follow-up letter, revised as a result of the information gathered from the non-respondents from the telephone interviews, and the questionnaire with a self-addressed, stamped envelope provided, was mailed, encouraging a reply within another designated two-week period. Data analysis and reporting
This needs assessment study was designed to provide descriptive data regarding coping experience of caregivers, and its relationship to the information provided during hospitalization to stroke family members.
The questionnaire yielded single items scores as wen as three index scores-coping ability, severity of stroke, and amount of information given. Some socio-demographic data were obtained independently from the stroke patient's medical record at the hospital to obtain a mailing list of current addresses, and to cross-reference some of the test items on the questionnaire. In analyzing the data, descriptive and inferential statistics were utilized in an attempt to transform raw data into a meaningful and organized set of measures including the development of the indices noted. To explore the relationship of the independent variable (amount of information given to the caregiver) and the dependent variable (coping ability or family adjustment) of the research hypothesis, a confirmatory factor analysis was performed, permitting the researcher to specify a hypothesized model of relationships between observed and latent variables and to assess the adequacy of the model using an established Goodness of Fit Index (GFI). Unlike other factor analysis techniques, confirmatory factor analysis permitted the researcher to obtain measures of reliability and construct validity. A high reliability measure (i.e., a coefficient Omega .80) implied that if the study were repeated, the same results would be obtained. A high measure of construct validity (i.e., a GFI .9) implied that a reliable and valid model was found. The researcher obtained estimates of the correlation between latent variable and estimates of how much each observed variable has contributed to the latent variable. The weight applied to the observed variables ranged between 0 and 1. The higher the weight, the more the observed variable has contributed to the latent variable. To report serendipitous findings, other statistical measures were employed, including statistical means, proportions, and analysis of variance. Response rate Of the 166 stroke families surveyed, 100 questionnaires were returned. Of the 100 respondents, 14 questionnaires were discarded due to: excessive missing data (7), denial of the occurrence of a stroke in the family (4), and stroke patients filling out the questionnaire themselves (3). Therefore, the total number of subjects in this study was 86, and the usable response rate of the sample surveyed was 52%. Socio-demographic characteristics of the subjects The first section of the questionnaire collected sociodemographic information of interest on the group of patients and their respective caregivers. An item analysis of the section was performed and the responses to each item converted into percentages. The results of the analysis of the sociodemographic items of the patients and their respondent caregivers appear in Table 1.
In summarizing, the majority (75%) of stroke patients in the study were 60-90 years of age, primarily female Caucasian, had at least a grade or high school education, and lived in a house. The single largest proportion of occupations represented in this sample was blue collar. The socio-demographic profile of the subjects in this study, the respondent caregivers, suggested that the majority were female Caucasians 65%), ranged in age from 23-87 years, had at least a high school education 82%), and lived in a house. Although 58% of the caregivers did not respond to the item of occupation, a large proportion of professionals were represented in the sample.
To get a better picture of the respondents, data was categorized according to gender and race. Missing data prohibited a display of other socio-demographic variables (education, occupation, and residence) in this manner. Missing data
As the socio-demographic characteristics of non-respondent caregivers were unknown, characteristics of patients with respondent caregivers and patients with non-respondent caregivers were obtained from the patient's medical record and were analyzed for statistical significance using a Chi Square Test of Independence (Table 2). Except for "race" and "services recommended upon discharge," the differences between the patients with respondent caregivers and patients with non-respondent caregivers were tested on the ChiSquare Test of Independence and found not to be significant at the.05 level. "Black" and "Hispanic" categories were combined to form a minority" category and then compared to "Caucasians" for statistical analysis. Similarly, the categories of "services recommended upon discharge" were aggregated and compared with the category of "none." Hypothesized findings
One formal hypothesis was tested. Is family education for the family of a stroke patient prior to discharge from the hospital related to family adjustment? This hypothesis was tested by computing linear combinations of variables to form indices of coping" and "amount of information given" to the caregiver. For each observation, a coping score and information score were found based on loadings (the correlations between the factors and the variables) from the confirmatory factor analysis, resulting in a coping factor and information given factor. The factor elements for coping and information were defined in Figure 1.
Pearson Product Moment Coefficient yielded a correlation of r = .76. High reliability measures of .80 and .88 were obtained for the coping and information factors, respectively (i.e., coefficient Omega .80). A Goodness of Fit Index (GFI) of .92 was obtained as a measure of fit of the entire model, suggesting a high measure of construct validity (i.e., GFI .9). A t-test was performed at the x = .05 level to equal -.70. There was no statistical evidence to indicate that the correlation between the coping factor and the information given factor was not equal to zero. (p >.05). A significant correlation between family education for the family of a stroke patient and family adjustment was not found.
Exploratory factor analysis was performed in an attempt to build a severity" index, but no significant correlations were identified with the other two indices of the study, coping and information given. Therefore, confirmatory factor analysis was not pursued with regard to severity. Serendipitous findings
The serendipitous findings of this study were divided into three general categories: (1) certain modal coping and information responses of the respondent caregivers, (2) response patterns of subgroups of respondent caregivers, and (3) content analysis of the open-ended question. Certain modal coping and information responses of the respondent caregivers
Although the null hypothesis could not be rejected, a majority of respondent caregivers indicated that they wanted more information, feeling that it would have helped them to cope better with their situation. A description of the certain modal responses to coping and information given items were presented in Table 3.
With regard to all the information items on the questionnaire, a large majority of the respondent caregivers suggested that it was the doctor who gave the information or no one at aH. Response patterns of subgroups of respondent caregivers
Analysis of variance was performed in order to identify possible response patterns of subgroups of respondent caregivers according to socio-demographic variables. An f-test was performed on aU of these items to show overall significance. On coping question 4, a Turkey-Kramer Test of Multiple Comparison was used. These results were presented in Table 4. Content analysis of open ended question
Content analysis was employed to examine respondent caregiver responses to the open question on the questionnaire. Using content analysis, the responses given on the open question were categorized according to the general thematic categories of description, emotional outpouring, specific suggestions/criticisms, and no response. Once categorized, the questionnaires were sorted by an independent rater. Interrater agreement was calculated and was found to be 98%. Responses to open-ended question are depicted in Figure 2, and further described in Table 5.
For the category of "no response," a statistical analysis was performed to see if the respondents to the open ended question were significantly different from non-respondents (no response) to this question according to their sociodemographic dimensions. Using a Chi-Square Test of Independence, (P = .7728), no statistical differences were found with regard to the only two characteristics with sufficient data, race, and gender.
Conclusions and Recommendations Conclusions from hypothesized findings
The hypothesized model of this study did not demonstrate a positive relationship between information given to the caregivers prior to discharge from the hospital and their ability to cope with the situation after discharge.
The estimated correlation between coping and information given was surprisingly low, -.076), but it cannot be concluded that there is no relationship between coping with stroke and the information given to the family. Several intervening variables, such as level of education, age, race, and other socio-demographic variables were not taken into account in the hypothesized model, and these variables may have explained why the estimated correlation between coping and information given was so low.
The high reliability scores of the coping questions suggested that the process of developing a viable construct to measure coping" has been initiated. Interestingly, those questions that scored high in the factor analysis were negative attributes, and, therefore, may have been only one aspect of coping." These identified characteristics were:
My own health problems have increased since the stroke. Since the stroke, I feel more impatient. Since the stroke, I feel more depressed. Since the stroke, I feel more over-protective.
Such characteristics may have been related to predisposing intrinsic personality attributes or physical condition that cannot be significantly changed regardless of the amount and/or quality of information given to the caregiver during hospitalization. There may have been other characteristics of coping ability that were not identified in this model.
Similarly, the high reliability scores on the information given questions suggests a viable construct of measuring information given during hospitalization was found in this study. Those identified characteristics were:
What is a stroke? The effects of a stroke on sexual activity. The stroke's effect on family relationships. The stroke's effect on relationships outside the home.
With the possible exception of the first item, these characteristics involved social interaction of the stroke patient inside and outside the family unit. Even the first item, may be a sub-element of what a stroke actually meant to the caregiver in terms of lifestyle changes.
The failure to reject the null hypothesis in the present study supported the findings of Field et al. (1981) which concluded their success with providing information to caregivers of aphasic patients was limited to the practical management aspects of care, and that psychological coping was not aided significantly. In this instance, severity of stroke may have been a significant factor as aphasic patients were considered severely involved. Severity as an important variable was not strictly controlled.
Perhaps the real issue underlying this study was not so much the quality or quantity of information given during hospitalization, but the perception of well-being - that ability to cope with a catastrophic situation that warranted further investigation. It may have been illusory to conceptualize that individuals can ever have sufficient information to cope with a stroke, particularly if the stroke is severe in nature.
Providing adequate information during hospitalization to family members of stroke patients may have been effective to select caregivers as a coping strategy. Thus, it may be possible that there is a multidimensional nature of subjective coping or adjustment where, as in so many fields of inquiry, methodological and semantic differences have generated differences of opinion which remain to be resolved.
Markides and Martin (1979) postulated: Health and subjective well-being appear to be inextricably interrelated. It would appear that one's health is a major influence on one's state of well-being, and that one's state of well-being is a major influence on one's health. It would appear that an alteration in either one is likely to affect a change in the other one.... If physical and psychological well-being are intimately intertwined, it is clear that being able to monitor both could be of considerable consequence (p. 683). If coping were independent of the quality and quantity of information given, but rather more closely related to predisposing intrinsic personality characteristics, future research could be directed toward identifying a constellation of factors (age, gender, severity of impairment, race, relationship to patient, etc.) which would permit the early identification of those caregivers who are likely to encounter difficulty adjusting to an altered life situation. Perhaps if one could predict which caregivers will require special help, health professionals would then be able to develop a more efficient allocation of staff and time, and more effectively deal with the problems confronting the caregivers.
The present study has uncovered a heuristic model to examine coping, which, with refinement, may lead to the development of other more comprehensive models. This possibility holds promise for caregivers of other chronic disabilities as a paradigm of coping may eventually emerge.
|Printer friendly Cite/link Email Feedback|
|Author:||Casas, Melody S.|
|Publication:||The Journal of Rehabilitation|
|Date:||Oct 1, 1989|
|Previous Article:||Relationship of rehabilitation client outcome to level of rehabilitation counselor education.|
|Next Article:||Ethics and rehabilitation supervision.|