Published on June 2016 | Categories: Documents | Downloads: 57 | Comments: 0 | Views: 718
of 10
Download PDF   Embed   Report



Lindsey Davis 2, October 2004RESEARCH BIOLOGICALand others / Biopsychological Markers 10.1177/1099800404267353 FOR NURSING Vol. 6, No.

Biopsychological Markers of Distress in Informal Caregivers
Linda Lindsey Davis, RN, PhD Michael Weaver, RN, PhD, FAAN Edward Zamrini, MD Alan Stevens, PhD Duck-Hee Kang, RN, PhD, FAAN C. Richard Parker Jr, PhD Background. Thirty caregiving wives participated in a study of caregiving distress and negative mood (depressive symptoms) by making diary entries on stressful caregiving situations and collecting saliva samples 4 times a day. At the end of the 7-day study period, caregivers’ salivary cortisol production was compared with their diary entries and correlated with pencil and paper self-report scores of caregiver distress and depressive symptoms. Findings. Despite the inability to control a number of factors thought to confound cortisol production (exercise, smoking, alcohol ingestion, and prescription medications), there was a statistically significant difference between No Caregiving and Caregiving cortisol, F(1,739) = 7.67, P = 0.006, with cortisol production higher when caregiving events occurred. However, efforts to code specific types of caregiving situations (e.g., 1 = indirect care; 4 = AD problem behavior care) did not further differentiate cortisol production. Although caregivers’ self-reports for the same 7-day period indicated they were depressed, pencil-and-paper measures of distress and negative affect were not significantly correlated with cortisol production. Conclusions and Recommendations. The finding that this caregiving group was significantly stressed by caregiving, as evidenced by increased cortisol producBIOLOGICAL RESEARCH FOR NURSING Vol. 6, No. 2, October 2004, 90-99 DOI: 10.1177/1099800404267353 Copyright © 2004 Sage Publications

tion during caregiving episodes, verifies the importance of further exploration of specific caregiving situations as contributory factors in caregiver health and well-being. In that saliva is a relatively economical and comparatively noninvasive biological data source for community-based stress studies, methodological limitations of the study are identified and 5 recommendations are made for future biological marker studies of caregiver distress in community-based settings. Key words: dementia, informal caregiving, home care, caregiving stress, biological markers, cortisol

Linda Lindsey Davis, RN, PhD, is a professor, Michael Weaver, RN, PhD, FAAN, is a professor, and Duck-Hee Kang, RN, PhD, FAAN, is an associate professor from the School of Nursing, University of Alabama at Birmingham. Edward Zamrini, MD, is an assistant professor from the School of Medicine, University of Alabama at Birmingham. Alan Stevens, PhD, is an associate professor from the School of Medicine, Center for Aging, University of Alabama at Birmingham. C. Richard Parker, Jr., PhD, is a professor from the Department of Obstetrics and Gynecology, University of Alabama at Birmingham. Address for correspondence: Linda Lindsey Davis, RN, PhD, School of Nursing, University of Alabama at Birmingham, NB 410, 1530 3rd Ave South, Birmingham, AL 35294-1210; phone: (205) 934-0026; fax: (205) 975-6142; e-mail: [email protected]. This article was supported by grants from the Southeast Center for Excellence in Geriatric Medicine and the University of Alabama at Birmingham School of Nursing.

Lindsey Davis and others / Biopsychological Markers 91

More than 52 million family members and friends

are caregivers for someone with a chronic illness (National Family Caregiver Association [NFCA] 2000). Acting as an informal caregiver for a person with Alzheimer’s disease (AD) is considered particularly stressful and has been correlated with increased psychological and physical disorders as well as increased incidence of personal health problems (Kiecolt-Glaser and Glaser 1999; Schulz and Beach 1999; Vedhara and others 2000; Farran and others 2001). Cortisol production has been explored under laboratory conditions as an indicator of hypothalamicpituitary-adrenal (HPA) axis response to stress in informal caregivers as well as healthy people (Cummins and Gevirtz 1993; van Eck and others 1996). Cortisol levels are known to increase within minutes of a stressful event and have a half-life of about 1 h after the event (Vining and others 1983; Kugler and others 1996; Pike and others 1997; Earle and others 1999; Stewart and Seeman 2000). When compared with noncaregivers, caregivers also have lessened antibody responses to virus vaccines, reduced cytokine production, and reduced natural killer cell activities (Castle and others 1995; Kiecolt-Glaser and others 1996; Vedhara and others 1999). Depressed caregiving spouses have been shown to have decreased lymphocyte proliferation responses (Castle and others 1995). Although cortisol responses to the chronic stress of caregiving have been explored in AD caregivers (Kiecolt-Glaser and others 1996; Irwin and others 1997; Vedhara and others 1999), the impact of acute caregiving events on caregivers’ stress responses has not been reported. This exploratory study used a repeated-measures design to explore acute cortisol responses in AD caregivers. The goals of this study were to identify specific caregiving situations that are stressful to AD caregivers; clarify the relationship between acute, situational stressors and chronic stress in AD caregivers; and provide empirical data on the feasibility and utility of collecting a biological measure of caregiver distress in the home setting. Two study aims were proposed: • Aim 1 was to compare AD caregivers’ cortisol levels in different caregiving situations in the home.

Aim 2 was to determine if AD caregivers’ cortisol levels were correlated with their selfreported caregiving distress and negative affect.

Materials and Methods
Participants Because cortisol production is influenced by gender (Earle and others 1999) and age (Reid and others 1992; Kiecolt-Glaser and others 1996), the study sample was restricted to women age 60 and older caring for a husband with AD. Men with AD and their wives were recruited from 1 university medical center memory clinic and surrounding home care agencies. Men with AD were required to (1) be 60 or older; (2) have a physician-determined diagnosis of AD-type dementia; (3) have mild to moderate cognitive impairment, as evidenced by a Mini Mental State Exam (MMSE) score less than 24 and greater than 9 (Folstein and others 1975); (4) have 4 or more behavior problems, as evidenced by a score of 4 or greater on the 24-item problem occurrence frequency scale A of the Revised Memory and Behavior Problem Checklist (Teri and others 1992); and (5) be willing to participate in a functional assessment in the home. Dyads were excluded if the AD person had a DSM-IV diagnosis of alcoholism or schizophrenia or a history of traumatic brain injury. In addition to the age requirement, caregiving wives were required to (1) live with the AD spouse; (2) have unpaid, primary caregiving responsibility for the spouse for a minimum of 4 h in the home each day; (3) be oriented and able to speak and read English and respond to questions; (4) reside in Alabama, within an 80-mile radius of the university; and (5) be willing to participate in the 7-day study protocol. Although hormonal and psychotropic medications are believed to influence cortisol production (Kirschbaum and Hellhammer 1989), the high incidence of use of these medications among women over age 60 provided rationale for tracking the incidence of hormone and psychotropic medication use in participants rather than excluding caregiving wives who were using these medications.

92 BIOLOGICAL RESEARCH FOR NURSING Vol. 6, No. 2, October 2004

Data Collection and Instruments Data were collected in the home setting. Descriptive data on the dyad, including age, health problems, medications, months of caregiving, use of community services, and other caregiver support, were collected. Caregivers were asked to rank their own health on a 5point scale ranging from excellent to poor. Caregivers also were asked to name 3 of their most challenging or stressful caregiving situations. As a measure of the AD-related functional loss, caregiving wives were asked to score their husbands’ level of independence in performing 14 self-care activities using the 8-item Physical Self-Maintenance (PSM) and the 6-item Instrumental Activities of Daily Living (IADL) instruments (Lawton and Brody 1969; Lawton 1988). To measure caregiving burden, 2 instruments were used: scale A of the Revised Memory and Behavior Checklist (Teri and others 1992), which measures the frequency of the occurrence of 24 ADrelated memory, mood, and problem behaviors, and the 42-item Caregiving Hassles Scale (Kinney and Stephens 1989). In addition, 4 pencil-and-paper measures commonly used to assess caregiver distress and negative mood (depressive symptoms) over a 1-week period were administered at baseline and repeated after completion of the 7-day study: • Revised Memory and Behavior Problem Checklist–Scale B: Caregiver Reaction Scale (RMBPC-B). This scale assesses the level of caregiver distress around caregiving problems over the past week (sample item: My husband is aggressive to others verbally). Respondents score each item on a Likert-like scale from 0 (not at all) to 4 (extremely). Factor analysis of the RMBPC indicates 3 1st-order factors: memory loss, depression, and disruptive behaviors. Subscale alphas for the caregiver reaction scale ranged from 0.89 to 0.90. Validity has been demonstrated through comparison of RMBPC-B scores with various caregiver burden measures (Teri and others 1992). In the current study, Cronbach’s alpha = 0.84 for caregiver reaction. Screen for Caregiver Burden–Scale B (SCBB). The SCB-B was used as a 2nd measure of

caregiver distress. It asks how upsetting particular caregiving situations were over the past week (sample item: I have little control over my husband’s behavior), with a Likert-like scale from 0 (never) to 3 (severely upsetting). Validity and reliability (alphas of 0.88-0.89 for the scale) have been demonstrated in various caregiving studies (Vitaliano and others 1991). Center for Epidemiological Studies–Depression (CES-D). The 16-item, 4-point, Likertlike self-report measure CES-D was used to measure caregiver affect over the past week (sample item: I felt depressed) (Radloff 1977). The scale for individual items went from 0 (none of the time or rarely) to 3 (most of the time). Total scores of 16 or greater indicate clinical depression. This measure has demonstrated reliabilities (alphas) of 0.75-0.88 in recent caregiver studies (Knight and others 1997). In the current study, Cronbach’s alpha was 0.78. Geriatric Depression Scale (GDS). This 15item version of the dichotomous choice (Yes = 1, No = 0) GDS is often used to ask older individuals about their affect (sample item: Do you often feel helpless?) over the past week (Yesavage and others 1983). Scores of 5 or greater are considered indicative of clinical depression. It takes 10 min or less to administer and has demonstrated alphas of 0.80 or greater with older adults (D’Ath and others 1994), with 90% sensitivity and 70% specificity (Meara and others 1999).

Procedures Human subject protection procedures were reviewed and approved by the university institutional review board. After written consent was secured from caregivers, cognitive loss and problem behavior occurrence in the AD person were verified with the MMSE. Upon completion of baseline measures, staff explained procedures for filling out the daily caregiving situation diary and collecting saliva samples. A 1-page summary of diary and saliva collection procedures was left with each caregiver. Caregivers were asked to

Lindsey Davis and others / Biopsychological Markers 93

practice with the diary and saliva collection system over a 2-day run-in period. Caregivers were considered to have completed the practice protocol if at least 75% of the daily diary entries and associated saliva specimens (3 of 4 possible entries) were collected each day over the 2-day run-in. If caregivers were not successful in making diary entries and collecting saliva specimens during this phase, they were paid US$15 for their time and dropped from the study. Those who were successful continued the diary entries and saliva collection for the 7-day study period. At the end of that period, staff visited in the home, repeated the distress and depression measures, collected the diary and saliva specimens and paid caregiving wives US$50. Caregiving diary data. Using a procedure based on the one described by Smyth and colleagues (1998) to collect daily diary and saliva data in the field, each caregiver was provided with a 7-day caregiving-situation diary. Caregivers were asked to make an entry in the diary and collect saliva within 30 min after a “difficult or challenging” caregiving episode 4 times each day for 7 days. An example of an entry would be “Getting him to take a bath.” Saliva data. Caregiving wives were asked to refrain from smoking, caffeine and alcohol intake, and vigorous exercise during the data-collection period. Each caregiver received 28 individual saliva collection tubes in a storage container. Caregivers were directed to rinse their mouths with water and then chew a compressed cotton pellet for 2 min after making a diary entry. Saliva-saturated pellets were to be placed in rubber-stopper test tubes, numbered and cross-referenced with diary entries, and then stored in the provided container in the refrigerator/freezer. To ensure the completion of 4 diary entries and saliva collections each day, caregivers also received a programmed wristwatch to wear. To avoid earlymorning and late-evening diurnal variations in cortisol production (Schmidt-Reinwald and others 1999), caregivers were asked at baseline to describe a “typical” day. Their individual wristwatches were then programmed to signal 4 times beginning 2 h after their reported time of arising and ending 2 h before their reported time of retiring for the night. If no stressful caregiving episode had occurred within the last 3 h,

Table 1.

Alzheimer’s Disease Person Characteristics at Baseline Respondents (n) 30 Statistics x = 76.9 (s = 6.9) Range: 64-92 years 30 76.7% (n = 23) Caucasian 23.3% (n = 7) African American 30 x = 11.6 (s = 4.2) Range: 1-20 years 30 x = 12.9 (s = 3.0) Range: 9-21 x = 3.5 (s = 3.6) Range: 0-14 28 28 x = 6.3 (s = 2.6) Range: 1-12 30 x = 15.7 (s = 4.4) Range: 4-23

Characteristics Age Ethnicity Years of education

Mini Mental State Exam score (MMSE)a Independence in instrumental activities of daily living (IADL)b Independence in physical selfmaintenance (PSMs)c Number of problem behaviors RMBPCScale Ad

a. Possible range 0-30, lower scores = more cognitive loss. b. Possible range 0-16, lower scores = more IADL dependence. c. Possible range 0-12, lower scores = more physical self-care dependence. d. Possible range 0-24, higher scores = greater frequency of problems.

caregivers were instructed to respond to the next wristwatch signal by making a diary entry and collecting a saliva specimen. Caregivers were instructed to code diary entries made in response to these wristwatch signals as “no caregiving.” Salivary cortisol was calculated in micrograms cortisol/100 mL (µg/dL). Cortisol levels were determined in duplicate by radioimmunoassay using commercially available 125I-cortisol kits with minor modification. The sensitivity of the assay is reported to be 12 ng/dL, with a mean intra-assay coefficient of variation of 4.8% (van Eck and others 1996). Sample Given the degree of reported correlation (r = 0.63) between morning cortisol response and adrenocortical response to ACTH reported by Schmidt-Reinwald and colleagues (1999), a sample size of 30 subjects was required for correlations of 0.50 or greater to ensure a power of at least 0.80 for testing the null hypothesis:

94 BIOLOGICAL RESEARCH FOR NURSING Vol. 6, No. 2, October 2004 Table 2. Age Ethnicity Years of education Caregiver Characteristics at Baseline Respondents (n) 30 30 30 Statistics x = 71.4 (s = 6.7) Range: 64-88 years 76.7% (n = 23) Caucasian 23.3% (n = 7) African American x = 12.1 (s = 2.6) Range: 8-17 years x = 69 months (s = 28.9) Range: 12-132 months x = 11.6 h (s = 6.9) Range: 4-24 h n = 7 (27%) n = 12 (46%) x = 17.2 (s = 4.8) Range: 7-28 x = 34.0 (s = 16.4) Range: 2-67 x = 31.6 (s = 11.1) Range: 1-45 x = 23.3 (s = 6.3) Range: 11-36 x = 6.1 (s = 2.3) Range: 2-10


Months of caregiving 30 experience Hours/day spent in 29 caregiving Number taking 26 hormone replacement medication Number taking 26 psychotropic medication Number of reported 29 daily caregiving hassles (DCH)a Caregiver distress and mood Caregiver distress 30 (RMBPC-Scale B)b Caregiver distress 30 (SCB-Scale B)c Caregiver mood 30 (CES-D)d Caregiver mood (GDS)e 30

mean of 11.6 years of education, and more than 23% were African American. Cognitive loss in these men was significant: the mean MMSE score was 12.9. As reported by their wives at baseline, the 30 men were limited in self-care independence and had demonstrated an average of more than 15 problem behaviors (out of a possible 24) over the past week. As shown in Table 2, wives were younger and had more education than their spouses. They were experienced caregivers, having provided almost 12 h of daily care for more than 5 years. These caregiving wives reported an average of 17 caregiving hassles each day (out of a possible 42) and had a mean CES-D score of 23.3 and a mean GDS score of 6.1, indicating a high number of self-reported depressive symptoms. Of the 26 caregiving wives who responded to questions about their drug usage, 12 (46%) were taking prescribed psychotropic medications; however, 24 of the 30 wives (80%) rated their own personal health as “good” to “excellent.” Caregiving wives identified 87 caregiving situations as challenging or problematic (Table 3). Of these, 30 (34%) were descriptors of problem behavior situations, 12 involved caregivers’ reports of their feelings around caregiving situations, and 5 were unclear in meaning and were not coded. Data Analysis Aim 1, to compare AD caregivers’ cortisol levels in different caregiving situations in the home, was addressed using a linear mixed model (LMM) approach (Bryk and Raudenbush 1992; Hox 2000). Caregiving diaries indicated a wide range of daily schedules, ranging from entries that began at 7:30 AM and ended around 8 PM (for a wife whose reported day began at 5:30 AM and ended around 10 PM) to another whose diary entries began around 5 PM and ended around 4 AM. Because 2 separate laboratories were used for salivary cortisol analyses, the LMM included a between-subjects lab effect and within-subjects effects for situation, time, and a situation-by-lab interaction. The time effect was modeled as a random effect, to allow for individual slope estimates of the repeated measures over time. An unstructured, autoregressive moving average (ARMA) covariance structure was used to model repeated-measures dependencies, as

a. Possible range 0-42, higher scores = more daily hassles. b. Possible range of scores 0-72, higher scores = more caregiver distress. c. Possible range of scores 0-96, higher scores = more caregiving distress. d. Possible range of scores 0-60, higher scores = more depressive symptoms. e. Possible range of scores 0-15, higher scores = more depressive symptoms.

Rho = 0. Over a 15-month period, 34 spousal caregivers were recruited. Of these, 4 (12%) dropped out during the run-in phase: 1 cited the “unpleasant” nature of collecting and storing her own saliva, 2 dropped out because they concluded that collection of saliva and recording of daily diary entries were too timeconsuming, and the 4th dropped out because she decided to admit her husband to a nursing home. Tables 1-3 provide a baseline profile of the 30 caregiving dyads who completed the 7-day study. As shown in Table 1, the ages of men with AD ranged from 64 to 92 years; they had completed a

Lindsey Davis and others / Biopsychological Markers 95 Table 3. Type of Situation Instrumental activities of daily living Physical care Stressful Caregiving Situations Reported by Caregivers Number of Reports 8 14 Examples Managing our money Having total financial responsibility Taking care of everything by myself Helping him eat Helping him go to the bathroom Incontinence/unable to take care of his own bowel movements Communication problems Explaining things over and over Not following instructions Not understanding Ensuring he is safe Can’t leave him alone Not knowing where he is Repeating things over and over Getting him to let me help him Getting him to bathe Getting him to go to bed/get out of bed Hearing things and thinking someone is trying to get in Hiding/hoarding/losing things Taking things apart Arguing with others Trying to go to work Trying to drive the car Wandering Getting up at night and wandering around Sundowning Arguing with the TV Loss of him as a person Loss of personal time Loss of social lifestyle Decline in my own health Can’t get out for exercise Having to stay at home Total responsibility Seeing him go down Difficulty in life Doing things Public socializing Keeping good help

Communication 18

Problem behaviors


Caregiver feelings




that produced the best model fits. Potential confounding effects from using different labs for cortisol measurements and repeated measurements were statisti-

cally controlled through their inclusion in the model as covariates (Little and others 2002). Caregiving situations described in the 30 wives’ diaries were analyzed using 2 coding methods. The 1st method coded each situation as 0 = no caregiving or 1 = caregiving. The 2nd method employed 5 categories of caregiving situation: 0 = no caregiving, 1 = IADL caregiving (e.g., household maintenance, financial affairs), 2 = physical care caregiving (e.g., feeding, bathing, dressing, toileting), 3 = communication caregiving (e.g., AD person arguing, not cooperating), or 4 = problem behavior caregiving (e.g., managing agitation, wandering, delusions). Model-derived residuals were examined for assumption conformity and appropriate transformations were applied where necessary. To reduce the inflationary effect of multiple testing on type I error rate, a 0.01 significance level was employed. Daily total cortisol concentration was estimated by calculating the area under the curve (AUC) formed by the 4 measured salivary cortisol levels using the trapezoid formula (Rowland and Tozer 1995; Schmidt-Reinwald and others 1999). Total and mean cortisol concentrations across the 7-day period were calculated. Based on conclusions reported by Stone and colleagues (2001) that more than half of “normal” individuals will show either flat or inconsistent cycles, cycle rhythm dysfunction measurement was rejected in favor of total cortisol. Aim 2, to determine whether AD caregivers’ cortisol levels were correlated with their self-reported caregiving distress and affect, was addressed using Pearson’s product-moment correlations. From a possible total of 840 saliva specimens that could have been collected over the 7-day study period, only 31 (less than 4%) were missing or unusable. However, 14 of the 30 caregiving wives were missing at least 1 specimen. To avoid a potential downward bias of the sum because of missing specimens while maximizing power for hypothesis testing through use of all subjects, mean cortisol AUC was calculated in addition to total cortisol AUC and correlated with caregivers’ pencil-and-paper self-reported distress (caregivers’reported feelings of distress and depressive symptoms) scores at the end of the 7-day data collection period. Again, a conservative P value of P < 0.01 was used for evaluating the test of Rho = 0.

96 BIOLOGICAL RESEARCH FOR NURSING Vol. 6, No. 2, October 2004 Table 4. Bivariate Correlations between Caregiver’s Cortisol AUC and Self-Report Distress and Affect Measure Scores at the End of the Study Period Daily Cortisol Level (AUC) 1.00 1.00 0.199 0.444 0.062 –0.495 Mean AUC 1.00 0.001 0.149 0.064 –0.364 RMBPC-B Score SCB-B Score CES-D Score GDS Score

Variable/Number of Respondents Daily cortisol level (AUTTC) (n = 16) Mean AUC (n = 30) RMBPC-B score (n = 28) SCB-B score (n = 27) CES-D score (n = 27) GDS score (n = 28)

1.00 0.647* 0.311 0.349

1.00 0.715* 0.471

1.00 0.180


NOTE: AUC = Area under the curve; RMBPC-B = Revised Memory and Behavior Problem Checklist–Scale B; SCB-B = Screen for Caregiver Burden–Scale B; CES-D = Center for Epidemiological Studies-Depression; GDS = Geriatric Depression Scale. *P < 0.01.

Aim 1 The raw least squares means (LSM) cortisols were 0.2252 mcg/dl (SE = 0.01963) for “no caregiving” and 0.2516 mcg/dl (SE = 0.02120) for “caregiving.” The distribution of cortisol concentration was highly skewed, and a log transformation was applied to the raw cortisol values to normalize the distribution. There was a statistically significant difference in mean logtransformed cortisol between no caregiving and caregiving, F(1,739) = 7.67, P = 0.006, with cortisol production higher when caregiving events occurred. The overall test for differences in mean log-transformed cortisol values between caregiving situation types (coded 0-4) was significant, F(4,737) = 4.72; P < 0.001, indicating that at least 1 mean was different. Post hoc contrasts comparing mean log-transformed cortisol between pairs of situations identified no caregiving (LSM = –0.9291) as having lower cortisol levels (t[737] = –2.68; P = 0.008) than basic self-care caregiving (LSM = –0.8033). Additionally, the differences between no caregiving (LSM = –0.9291) and problem behavior caregiving (LSM = –0.8362) and between physical caregiving (LSM = –0.8033) and communication caregiving (LSM = –0.9008) tended toward significance (t[737] = –2.10; P = 0.036) and (t[737] = 2.09; P = 0.037), respectively. To determine if there might be a lag in cortisol response following a caregiving event, events were plotted against lagged cortisol values, and the LMMs were rerun, using the lagged log-transformed cortisol val-

ues. Neither the plots nor the LMM provided evidence of an association between caregiving event and lagged log-transformed cortisol value, F(1,738) = 0.19; P = 0.665. Aim 2 A matrix of correlations between caregiving wives’ total AUC, mean AUC, and the 2 distress and 2 affect measures was constructed to address Aim 2 (Table 4). Although caregivers exhibiting higher levels of depressive symptoms tended to also exhibit higher levels of distress, no statistically significant bivariate correlations were found between caregiver AUC cortisol values and the self-report measures of distress and depression. As expected, correlations between the 2 measures of distress (RMBPC-B and SCB-B) were significant. Unexpectedly, there was no significant association between the 2 depression measures (CES-D and GDS), and only 1 of the distress measures (SCBB) was significantly correlated with a measure of depression (CES-D), though its correlation with the GDS approached statistical significance (P = 0.015).

Although both cortisol production and pencil-andpaper measures indicated that these wives experienced distress and negative affect around caregiving, there were few significant relationships between the different measures. There are a number of alternative explanations for these findings. First, cortisol responses may have been flattened by medication use by some of

Lindsey Davis and others / Biopsychological Markers 97

the wives in the sample. Of the 12 wives who reported prescription psychotropic medication use, 9 were on selective serotonin reuptake inhibitors and 5 were taking various pain medications. Second, although cortisol production in caregivers differed significantly in this study between situations coded “no caregiving” and those coded “caregiving,” the further breakdown of caregiving situations according to type of problem (e.g., IADL caregiving, ADL caregiving) failed to demonstrate significant differences in acute cortisol responses to these different situations. This failure of the schema for subcoding caregiving situations to reveal differences in cortisol production may be indicative of caregiver burnout, which is often purported to result from long-term AD caregiving (Stuckey and others 1996; Quayhagen and others 1997; Dyer and others 2000; Kasuya and others 2000). After an average of 5 years of caregiving experience, these caregiving wives may have had muted biological responses to selected caregiving stressors because of caregiving burnout. Third, failure to control for selected lifestyle variables may have obscured differences. Although Smyth and colleagues (1998) reported that statistical control of lifestyle variables (smoking, caffeine, alcohol, and exercise) did not significantly influence the relationship between daily stressors and salivary cortisol in their community-based study of healthy subjects, differential adherence of some caregivers in controlling selected lifestyle variables in this study may have masked cortisol responses in the sample group. Although caregivers were asked to refrain from vigorous exercise, smoking, caffeine, and alcohol ingestion during the 7-day data collection period, adherence with this request could not be verified. If this bias was operant in the study, the finding that responses to no caregiving situations differed from the responses to physical care situations may have been an artifact of specimen collection following a physically active care activity such as bathing. Finally, the relative impact of caregivers’ personal characteristics and stress-reducing strategies in specific situations should be acknowledged. Although it was beyond the scope of the study to collect data on factors moderating the caregiver–care recipient relationship (e.g., marital satisfaction) or caregiver personality factors thought to mediate or buffer caregiver burnout, such as caregivers’problem-solving abilities,

cognitive self-efficacy, and mastery (Bookwala and Schulz 1998; Steffen and others 2002), the potential effects of these factors must be acknowledged.

Because of the reportedly devastating impact of progressive dementia on family health and well-being, providing care for a person with AD is frequently selected as an exemplar of caregiving distress. The profile of these 30 caregiving wives suggests that they were similar to other highly stressed AD caregivers. As evidenced by the MMSE and RMBPC-A screening scores of the men with AD, their caregiving wives were managing mid-stage dementia, when confusion and problem behaviors are most common (Lieberman and Fisher 1995; Doriswamy 1996; Dyer and others 2000). Wife caregiver affect scores (both CES-D and GDS) indicated most were clinically depressed, as is often reported in the AD caregiver literature (Miller and others 1995; Bookwala and Schulz 1998). Although not significantly correlated with pencil-andpaper tests of caregiver distress and depression, cortisol levels in these caregiving wives increased with caregiving events. Design limitations noted above suggest 5 refinements for future community-based studies of biological markers of distress in caregivers: • Because the subcoding schema failed to demonstrate differences in caregiver cortisol production in response to different types of caregiving situations, future studies should include a situational stress rating scale. For example, the caregiving diary might be revised to include a Likert-like scale for each situation for the caregiver to indicate how stressful she perceived a situation to have been. To determine whether there is a duration factor influencing cortisol production, caregivers might also be asked to note the amount of time spent in each caregiving situation. The relationship between acute and chronic stress around caregiving needs more exploration. Kiecolt-Glaser and Glaser (1999), Vedhara and colleagues (2000), and Mills and colleagues (1999) have studied long-term immunologic outcomes of caregiving, but they have not clarified the interaction of acute and

98 BIOLOGICAL RESEARCH FOR NURSING Vol. 6, No. 2, October 2004

chronic stress and its impact on caregivers’subsequent health and well-being. To further explore the biopsychological link between chronic long-term caregiving stress and acute, episodic caregiving stress, future studies might include selected immune system variables proposed by these and other investigators as biological markers of chronic stress in caregivers. Mediators—such as situational mastery, hardiness, resilience, social support, and problem solving—hypothesized to buffer caregiving stress should also be included in future studies. All have been proposed as possible contributory factors in caregiver distress (cf. Miller and others 1995; Bookwala and Schulz 1998; Rabin 2000; Yee and Schulz 2000; Steffen and others 2002). Identification of coping strategies that differentially buffer situational caregiving stress may provide direction for future caregiver interventions. Psychosocial moderators of caregiving stress—such as other family relationships (e.g., adult children), length of caregiving career, quality of caregiver-care recipient relationship, caregiver neuroticism, and self-esteem—also should be explored. Although less susceptible to intervention, such variables also may influence caregiving stress responses. To explore the caregiving burnout phenomenon, both diurnal cycle rhythm and cortisol AUC should be calculated. To accomplish this, caregivers might be asked to make diary entries and collect saliva for 1 week only when caregiving situations have not occurred followed by a 2nd week of data collection in response to caregiving situations.

improving the ecological validity of caregiver studies and justifies future studies in community-based settings.

Bookwala J, Schulz R. 1998. The role of neuroticism and mastery in spouse caregivers’ assessment of and response to a contextual stressor. J Gerontol Psychol Sci 53(3):155-64. Bryk A, Raudenbush S. 1992. Hierarchical linear models: applications and data analysis methods. Newbury Park (CA): Sage. Castle S, Wilkins S, Heck E, Tanzy K, Fahey J. 1995. Depression in caregivers of demented patients is associated with altered immunity: impaired proliferative capacity, increased CD8+, and a decline in lymphocytes with surface signal transduction molecules (CD38+) and a cytotoxicity marker (CD56+ CD8+). Clin Exp Immunol 101(3):487-93. Cummins S, Gevirtz R. 1993. The relationship between daily stress and urinary cortisol in a normal population: an emphasis on individual differences. Behav Med 19:129-34. D’Ath P, Katona P, Mullan E, Evans S, Katona C. 1994. Screening, detection and management of depression in elderly primary care attenders: I. The acceptability and performance of the 15item Geriatric Depression Scale (GDS15) and the development of shortversions. Family Pract 11(3):260-6. Doriswamy PM. 1996. Current cholinergic therapy for symptoms of Alzheimer’s disease. Prim Psychiatry 3(56):56-68. Dyer C, Pavlik V, Murphy K, Hyman D. 2000. The high prevalence of depression and dementia in elder abuse or neglect. J Am Geriatr Soc 48(2):205-8. Earle T, Linden W, Weinberg J. 1999. Differential effects of harassment on cardiovascular and salivary cortisol stress reactivity and recovery in women and men. J Psychosom Res 46:125-41. Farran C, Loukissa D, Hauser P, McCann J, Swanson B, Zeller J. 2001. Psychoneuroimmunological outcomes in dementia caregiver studies: an idea whose time has come? Online J Knowledge Synthesis Nurs 8(6). Retrieved November 22, 2003, from Folstein M, Folstein S, McHugh P. 1975. Mini-mental state: a practical method for grading the cognitive state of patients for the clinician. J Psychiatr Res 12:189-98. Hox J. 2000. Multilevel analyses of grouped and longitudinal data. In: Little T, Schnabel K, Baumert J, editors. Modeling longitudinal and multilevel data. Practical issues, applied approaches, and specific examples. Mahwah (NJ): Lawrence Erlbaum. Irwin M, Hauger R, Patterson TL, Semple S, Ziegler M, Grant I. 1997. Alzheimer caregiver stress: basal natural killer cell activity, pituitary-adrenal cortical function, and sympathetic tone. Ann Behav Med 19(2):83-90. Kasuya R, Polgar-Bailey P, Takeuchi P. 2000. Caregiver burden and burnout. Postgrad Med 108(7):119-23. Kiecolt-Glaser J, Glaser R. 1999. Editorial: chronic stress and mortality among older adults. J Am Med Assoc 283(23):225960.

Although this exploratory study demonstrated the feasibility of subject-initiated biopsychological data collection in the home setting, interpretation of findings was complicated by problems in verifying participant adherence with data-collection protocols and omission of possible confounding variables. Despite these challenges, the ability to collect data from caregivers in the midst of difficult and problematic caregiving situations (rather than from retrospective recall of situations or through exposure to artificially induced stressors in the laboratory) holds promise for

Lindsey Davis and others / Biopsychological Markers 99 Kiecolt-Glaser J, Glaser R, Gravenstein S, Malarkey W, Sheridan J. 1996. Chronic stress alters the immune response to influenza vaccine in older patients. Proc Natl Acad Sci U S A 93:3043-7. Kinney J, Stephens M. 1989. Caregiving hassles scale: assessing the daily hassles of caring for a family member with dementia. Gerontologist 29(3):328-32. Kirschbaum C, Hellhammer D. 1989. Salivary cortisol in psychobiological research: an overview. Neuropsychobiology 22:150-69. Knight R, Williams S, McGee R, Olaman S. 1997. Psychometric properties of the Centre for Epidemiologic Studies Depression Scale (CES-D) in a sample of women in middle life. Behav Res Ther 35(4):373-80. Kugler J, Reintjes F, Tewes V, Schedlowski M. 1996. Competition stress in soccer coaches increase salivary immunoglobulin A and salivary cortisol concentrations. J Sports Med Phys Fit 36:117-20. Lawton M, Brody E. 1969. Assessment of older people: self-maintenance instrumental activities of daily living. The Gerontologist 9:179-86. Lawton MP. 1988. Instrumental activities of daily living (IADL) scale: original observer-rated version. Psychopharmacol Bull 24:785-7. Lieberman M, Fisher L. 1995. The impact of chronic illness on the health and well-being of family members. Gerontologist 35:94-102. Little R, Stroup W, Freund R. 2002. SAS for linear models. Cary (NC): SAS Institute. Meara J, Mitchelmore E, Hobson P. 1999. Use of the GDS-15 geriatric depression scale as a screening instrument for depressive symptomatolgy in patients with Parkinson’s disease and their carers in the community. Age Ageing 28:35-8. Miller B, Campbell R, Farran C, Kaufman J, Davis L. 1995. Race, control, mastery and caregiver distress. J Gerontol Soc Sci 50B:S374-82. Mills P, Yu H, Ziegler M, Patterson T, Grant I. 1999. Vulnerable caregivers of patients with Alzheimer’s disease have a deficit of circulating CD62L-T lymphocytes. Psychosom Med 61:16874. National Family Caregiver Association. 2000. Caregiver survey2000. Kensington (MD): Author. Pike J, Smith T, Hauger R, Nicassio P, Patterson T, McClintock J, and others. 1997. Chronic life stress alters sympathetic, neuroendocrine, and immune responsivity to an acute psychological stressor in humans. Psychosom Med 59:447-57. Quayhagen M, Quayhagen M, Patterson T, Irwin M, Hauger R, Grant I. 1997. Coping with dementia: family caregiver burnout and abuse. J Mental Health Aging 3(3):357-64. Rabin B. 2000. Social networks and dementia. Lancet 356:76-7. Radloff L. 1977. The CES-D scale: a self-report depression scale for research in the general population. Appl Psychol Measure 1:385-401. Reid J, Intrieri R, Susman E, Beard J. 1992. The relationship of serum and salivary cortisol in a sample of healthy elderly. J Gerontol Psychol Sci 47:176-9. Rowland M, Tozer T. 1995. Clinical pharmacokinetics, concepts and applications. London: Williams and Wilkins. Schmidt-Reinwald A, Pruessner J, Hellhammer D, Federenko I, Rohleder N, Schürmeyer T, and others. 1999. The cortisol response to awakening in relation to different challenge tests and a 12-hour cortisol rhythm. Life Sci 64:1653-60. Schulz R, Beach S. 1999. Caregiving as a risk factor for mortality: The Caregiver Health Effects Study. J Am Med Assoc 282:2215-19. Smyth J, Ockenfels M, Porter L, Kirschbaum C, Hellhammer D, Stone A. 1998. Stressors and mood measured on a momentary basis are associated with salivary cortisol secretion. Psychoneuroendocrinology 23(4):353-70. Steffen AM, McKibbin C, Zeiss AM, Gallagher-Thompson D, Bandura A. 2002. The revised scale for caregiving self-efficacy: reliability and validity studies. J Gerontol Psychol Sci 57(1):74-86. Stewart J, Seeman T. 2000. Salivary cortisol measurement. Consensus Conference report, McArthur Research Network Meeting, December 1999, Rockefeller University. Retrieved August 20, 2002, from cort./html. Stone AA, Schwartz JE, Smyth J, Kirschbaum C, Cohen S, Hellhammer D, Grossman S. 2001. Individual differences in the diurnal cycle of salivary free cortisol: a replication of flattened cycles for some individuals. Psychoneuroendocrinology 26:295-303. Stuckey JC, Neundorfer MM, Smyth KA. 1996. Burden or wellbeing: the same coin or related currency? Gerontologist 36(5):686-93. Teri L, Logsdon R, Uomoto J, Zarit S, Vitaliano P. 1992. Assessment of behavioral problems in dementia: the revised memory and behavior problems checklist. Psychol Aging 7(4):622-31. van Eck M, Berkhof H, Nicolson N, Sulon J. 1996. The effects of perceived stress, traits, mood states, and stressful daily events on salivary cortisol. Psychosom Med 58:447-58. Vedhara K, Cox NK, Wilcock GK, Perks P, Hunt M, Anderson S, and others. 1999. Chronic stress in elderly carers of dementia patients and antibody response to influenza vaccination. Lancet 20:353:627-31. Vedhara K, Shanks N, Anderson S, Lightman S. 2000. The role of stressors and psychosocial variables in the stress process: a study of chronic caregiver stress. Psychosom Med 62:374-85. Vining R, McGinley R, Symons R. 1983. Hormones in saliva: mode of entry and consequent implications for clinical interpretation. Clin Chem 29(10):1752-6. Vitaliano P, Russo J, Young H, Becker J, Maiuro R. 1991. The screen for caregiver burden. Gerontologist 31:76-83. Yee J, Schulz R. 2000. Gender differences in psychiatric morbidity among family caregivers: a review and analysis. Gerontologist 40:147-64. Yesavage J, Brink R, Rose T, Lum W, Huang V, Adey M, and others. 1983. Development and validation of a geriatric depression screening scale. J Psychiatr Res 17:37-49.

Sponsor Documents

Or use your account on


Forgot your password?

Or register your new account on


Lost your password? Please enter your email address. You will receive a link to create a new password.

Back to log-in