Psychosomatic Medicine
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Engler, P.
Right arrow Articles by Stein, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Engler, P.
Right arrow Articles by Stein, M.
Related Collections
Right arrow Infectious Disease
Right arrow Stress and Coping
Right arrow HIV/AIDS
Psychosomatic Medicine 68:985-992 (2006)
© 2006 American Psychosomatic Society

ORIGINAL ARTICLES

Coping and Burden Among Informal HIV Caregivers

Patricia Engler, PhD, Bradley Anderson, PhD, Debra Herman, PhD, Duane Bishop, MD, Ivan Miller, PhD, Paul Pirraglia, MD, MPH, Jumi Hayaki, PhD and Michael Stein, MD

From Rhode Island Hospital, Division of General Internal Medicine, Providence, Rhode Island.

Address correspondence and reprint requests to Patricia Engler, PhD, Rhode Island Hospital, Division of General Internal Medicine, 593 Eddy Street, Providence, RI 02903. E-mail: pengler{at}lifespan.org


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Objective: The aim of this study was to examine the role of coping on caregiver burden among a heterogeneous group of caregivers of persons living with HIV during the era of highly active antiretroviral therapy.

Methods: Burden and coping were examined among 176 caregivers of persons living with HIV. Three styles of coping were examined using a 7-item scale: active-approach (task), blame–withdrawal (emotion), and distancing (avoidance).

Results: A total of 58.8% of the caregivers were women. They had a mean age of 42 years; 61.9% cohabited with the persons living with HIV who had a mean CD4 count of 401. All three styles of coping were significantly positively correlated with caregiver burden. After controlling for demographic variables and caregiver depression, active-approach coping and distancing coping independently moderated the relationship between perceived severity of HIV-related symptoms (stress) and caregiver burden; however, some caregivers experienced burden even at low levels of stress.

Conclusions: These results indicate that in the era of highly active antiretroviral therapy, coping mitigates the effect of stress on burden.

Key Words: caregiver burden • coping • HIV • symptom severity • stress

Abbreviations: AIDS = acquired immune deficiency syndrome; HAART = highly active antiretroviral therapy; HIPAA = Health Insurance Portability and Accountability Act; HIV = human immunodeficiency virus; PLWH = person living with HIV; CSI = Caregiver Strain Index.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Caregiver burden, the subjective experience of problems or strains linked to the caregiver role (1), is a well-studied phenomenon. Caregivers often experience instrumental and emotional stressors that result from physical, cognitive, and behavioral changes in their care recipients’ health status (e.g., (2)). However, the caregiving experience may vary widely depending on the nature of the care recipient’s illness or needs as well as the caregiver’s coping style.

Although the experience of the caregiver is documented for many illnesses, the research examining burden among caregivers of persons living with HIV (PLWH) is limited. The existing literature suggests that caring for a loved one with human immunodeficiency virus (HIV) presents unique demands for the caregiver. Because many HIV caregivers are sexual partners of the care recipients, they may struggle with concerns about their own health risks in addition to that of their partner (1,3). Also, the caregiver may receive less support from the partner’s family if there is conflict regarding the partner’s sexual orientation or health status (4).

The majority of the existing research on the impact of caregiving for a PLWH examines gay or bisexual men. In one study of both HIV-positive and HIV-negative gay/bisexual men caring for partners with acquired immune deficiency syndrome (AIDS), Land and colleagues (5) found greater levels of depression among HIV-positive caregivers. Younger age, greater role-related stress, and low self-esteem significantly predicted caregiver depression within both groups. When examined separately, poorer health and financial problems predicted caregiver depression among the HIV-positive caregivers, whereas greater stress associated with the caregiver role predicted greater depression among the HIV-negative caregivers.

Although depression has been linked to caregiving for a PLWH, caregiver burden has received less attention in the literature. Folkman and colleagues examined caregiver burden in both HIV-positive and -negative caregivers among gay men (6); they found that HIV-positive caregivers reported significantly greater caregiver burden. "Role overload," a construct conceptually similar to caregiver burden, is associated with caregiver depression in both correlation and regression analyses (7). In a review article, Prachakul and Grant (8) reported that caregiver burden resulting from role overload was significantly positively correlated with poor health in both HIV-positive and HIV-negative caregivers. Furthermore, they found that caregiver burden was significantly associated with both depression and suicidal ideation. Using the present data set of a large heterogeneous sample of HIV caregivers, Pirraglia and colleagues (9) found that among other variables, caregiver burden was associated with significantly greater odds of caregiver depression.

Informal caregivers differ in their experience of burden; thus, burden may be influenced by other variables such as severity of the illness and demographic factors. Certain patient variables such as the patient’s HIV-related problems and demographic variables such as cohabitation contribute to caregiver burden (10). More specifically, greater patient HIV-related problems were significantly, positively correlated with both caregiver burden and caregiver adjustment to the patient’s HIV illness, whereas cohabitation was significantly positively associated with caregiver burden. Both HIV symptoms of the PLWH and cohabitation were also found to be significantly associated with caregiver burden among a sample of caregivers for HIV-positive women (11). This research examining both the PLWH and the caregiver is informative; however, the relatively small sample sizes limit the generalizability of the findings (e.g., (10,12)), and these studies took place before the recent dramatic treatment improvements for HIV.

In addition to patient and demographic variables, caregiver characteristics such as coping style likely influence the experience of burden. Endler and Parker (13) distinguish among three basic styles of coping. Task-oriented coping refers to behaviors meant to deal actively with a problem, whereas emotion-oriented coping includes techniques intended to regulate one’s emotions (14). Avoidance-oriented coping includes behaviors that allow one to avoid a situation by seeking out other people or through distraction by engaging in some other task (13). Although the effectiveness of one’s coping style varies depending on the nature of the stressor, considerable research suggests that task-oriented coping may be more adaptive than certain types of emotion-oriented or avoidance-oriented coping (e.g., (15–18)), this may be particularly true with respect to chronic stress. This finding has been supported in the HIV literature (e.g., (16,17)). Although this classification of coping by its function has been used widely, it should be noted that other researchers argue that more comprehensive systems of coping better bridge instances of coping with adaptive processes (19).

Several studies have examined the relationship of coping with caregiving burden among the caregivers of a PLWH. Among a sample of gay men, HIV-positive and HIV-negative caregivers reported significantly different ways of coping (6). HIV-positive caregivers reported using significantly greater positive reappraisal (emotion-oriented), cognitive escape/avoidance (avoidance-oriented), and seeking social support forms of coping while experiencing significantly greater burden than their HIV-negative counterparts. In a sample of HIV-positive gay men and their caregivers, Pakenham, Dadds, and Terry (10) found that caregivers’ use of task-oriented coping strategies (i.e., "action" coping) was significantly negatively correlated with caregiver burden. Although not statistically significant, the caregivers’ use of emotion-oriented strategies (i.e., escape and fatalism) was positively correlated with global psychological distress. Despite the contribution of this research, some studies contained small homogenous samples, which limit the generalizability of the results, particularly in light of the recent dramatic improvements in HIV treatment. Additionally, more complex relationships between coping and stress should be examined more closely.

Research has suggested that coping may play a complex role in one’s health. Personal and social resources may act as buffers between stress and psychological adjustment within HIV people living with (e.g., (20)). Furthermore, it has been found that coping may moderate the relationship between factors related to stress and health. For instance, one study found that certain types of coping buffered the effect of heart rate reactivity on health status in a convenience sample (21). In another study of patients with chronic disease, coping competence moderated the relationship between stress and depression (22). That is, high coping competence buffered the effect of symptom stress on depression. As such, personal resources such as coping have been implicated as moderators of the relationship between stress and both health and psychological adjustment variables. However, these complex relationships within caregiver samples require further investigation.

The aim of the present study was to examine the role of coping on caregiver burden among a heterogeneous group of caregivers of PLWH during the era of highly active antiretroviral therapy (HAART). We hypothesized that a) active-approach coping (task-oriented) would be significantly, negatively correlated with caregiver burden; b) blame–withdrawal (emotion-oriented) coping would be significantly, positively correlated with caregiver burden; and c) coping would moderate the relationship between caregiver stress and caregiver burden.


   METHODS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Participants and Procedure
We use baseline data from a randomized clinical trial of a supportive psychoeducational telephone intervention for PLWH and their significant others serving as informal caregivers. A total of 514 individuals were approached between January 2001 and June 2004 during their visit to Brown University medical clinics and told about this Rhode Island Hospital Institutional Review Board-approved study for the PLWH and the families. Interested individuals were then screened to assess full eligibility for the study. Factors for inclusion included 18 years of age or older, English- or Spanish-speaking, regular access to a telephone (for the intervention component of a larger study), and competency to sign an informed consent form and participate in the research interview. During this initial screen, individuals were also asked to identify their "HIV support person," the person in their life who provides support for them around their HIV. If no informal caregiver was identified, a PLWH could enroll in the study alone. If the patient qualified for the study, he or she (and, if identified, his or her informal caregiver) was invited for a 90-minute baseline interview. During this interview, after completing informed consent and Health Insurance Portability and Accountability Act (HIPAA) forms, we assessed a variety of domains that were used for the larger investigation, including HIV-related medical experiences, coping, depression, family functioning, substance use, and social relationships. HIV-positive study participants and their informal caregivers were interviewed individually.

Of the 514 people living with HIV approached for participation in the larger, prospective study; 180 refused, 6 spoke languages other than English or Spanish, 40 did not have telephones, 6 had plans to move, and 17 were cognitively impaired, leaving the study sample of 265 people living with HIV and 176 informal caregivers. Study refusers were not significantly different from enrollees in race, ethnicity, or gender. The present study examines the data of the 176 informal caregivers.

At the conclusion of the baseline interview, eligible participants received a resource guide, which included local, national, and web-based resources and information about the study (i.e., follow-up appointment schedule, contact information, and so on). All participants were referred back to the Brown University clinics if a medical or psychiatric issue arose at any time during the course of the study.

Measures
The variables examined were similar to those reported by Pakenham and colleagues (10) to contribute to caregiver burden (e.g., coping, cohabitation, HIV-related symptoms).

The Caregiver Strain Index (CSI) is a 13-item measure that assesses the degree of caregiver strain or burden (23). Responses range from 0 to 13 with higher scores indicating greater burden. Internal consistency reliability was 0.84 in the study sample. Items include questions about issues such as sleep disturbance, inconvenience, financial strain, and physical strain associated with helping the care recipient.

Caregiver coping was measured using a 9-item HIV coping scale (24) developed in the HIV Cost and Service Utilization Study and tested with a large, nationally representative sample of HIV-positive Americans receiving medical care. The survey items are thought to reflect unique components of coping with HIV disease (e.g., "Involved yourself in volunteer work or in a community organization for people with HIV"), which is crucial given that a limitation of other coping measures is the absence of certain coping behaviors associated with HIV caregiving (17). For the caregivers, the instructions were adapted to reflect caring for a PLWH rather than a PLWH responding about their own coping efforts. Participants are asked, "In response to (patient) having HIV, how often during the last 4 weeks have you done each of the following?" Participants can use any or all of three forms of coping. The three forms of coping constitute subscales, which include items measuring active-approach (four task-related items; e.g., "asked other people for advice and information"; {alpha} = 0.55), blame–withdrawal (two emotion-related items; e.g., "criticized or lectured yourself"; {alpha} = 0.55), and distancing (three avoidance-related items; e.g., "tried to keep yourself from worrying about HIV/AIDS"; {alpha} = 0.71). Respondents select from a 6-point response scale ranging from "all of the time" to "none of the time." Two items from the active-approach subscale were omitted from the present analyses to increase the internal consistency of the subscale ({alpha} = 0.55 for the four-item subscale; {alpha} = 0.65 for the two-item subscale). Consistent with the scoring protocol described in Fleishman et al. (24), each coping item was linearly transformed to a range of 0 to 100 with higher scores reflecting greater use of coping. Subscales were calculated as the mean of the included items with possible scores ranging from 0 to 100.

Because patient-reported HIV-related symptoms are significantly positively correlated with both patient-reported and caregiver-reported stress (e.g., (25)) and we were interested in the caregivers’ perceptions, we used the caregiver’s report of the patient’s HIV symptoms as our measure of stress. An individual’s appraisal of a stressor is considered a key element in the stress process (26); therefore, we chose to use the caregivers’ perceptions of the PLWH’s symptoms rather than the PLWH’s reports of their own symptoms. The importance of subjective stress appraisal has been cited in other HIV-related research (e.g., (25)). This variable was assessed through a 13-item list of common HIV-related symptoms (e.g., headaches, fever, pain, diarrhea, weight loss). The caregivers were asked whether the PLWH had experienced any of the symptoms in the previous 6 months and, if present, rated the severity of each of those symptoms using a 5-point response scale. Responses were summed, resulting in a possible range of 13 to 65, with higher scores reflecting greater perceived number and severity of the caregivers’ perceptions of the PLWH’s symptoms. Internal consistency reliability was 0.80.

The Beck Depression Inventory (BDI-I (27)) was used to measure depressive symptoms. Scores range from 0 to 63 with higher scores signifying greater depression. Cronbach alphas range from 0.78 (nonpsychiatric sample) to 0.95 (psychiatric sample); test–retest reliabilities are strong (e.g., r = 0.90 for 2 weeks (28)).

The Symptom Factor-36 Scale is a multiitem measure that assesses eight "health concepts" (29). Items are measured on a Likert-type scale and have demonstrated clinical (e.g., (30)) and concurrent validity (31). The Physical Function Subscale was used in the present study as an indicator of the health status of the PLWH. It is reported to have a strong internal consistency (0.93 (32)).

Other covariates included age in years, race (an indicator variable coded 1 if white in multivariate analyses), gender (1 if male), and cohabitation (1 if the caregiver lives with the patient) with the patient. Two indicators of the PLWH’s health status were also included. These were the Symptom Factor-36 physical function score and CD4 count (scored 1 if <200, 0 otherwise).

Analysis Plan
The observed distribution of caregiver burden, the response variable, was positively skewed and distinctly nonnormal; the modal response category was at the lower limit of zero and the distribution was not amenable to normalizing power transformations (Fig. 1). Therefore, we report Spearman rank-order correlation coefficients to summarize bivariate associations and analyze caregiver burden as a count variable using negative binomial regression (33) to test our primary hypotheses.1 Tests of significance rely on robust standard error estimators as implemented in Stata version 8.2 (34). To facilitate interpretation, continuous predictor variables were standardized to zero mean and unit variance before analyses. Coefficients are exponentiated and give the expected factor change in the number of caregiver burden symptoms reported (33). To test the hypothesis that the effect of stress on caregiver burden was moderated by coping, we included the first-order stress by coping interaction terms. Because coping was conceptualized as a multidimensional construct, tests of first-order coping by stress interactions were conducted for each of the three coping measures.2 We report a Wald statistic that simultaneously tests the hypothesis that one or more of the three interaction coefficients are not equal to zero.


Figure 123
View larger version (9K):
[in this window]
[in a new window]

  		Figure 1. Observed distribution of caregiver burden.

 


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
Just over half (52.8%) of the informal caregivers were women; 40.9% were white, 27.3% were black, 23.9% were Hispanic, and 8.0% reported other ethnic origins (Table 1). Informal caregivers had a mean age of 42.12 (±13.09) years; 61.9% lived with the patient; 22.3% reported being HIV-positive. Nearly half of the informal caregivers were the romantic partners of the PWLH (47.2%). The remainder was divided between family members (33.5%) and friends (19.3%). Approximately 61.9% lived with the PWLH. On average, participants endorsed 4.17 (±3.39) of the 13 caregiver burden items; 27% scored in the "high burden" range (9).


View this table:
[in this window]
[in a new window]

  		TABLE 1. Background Characteristics of Caregivers (n = 176)

 

The PLWH had a mean CD4 count of 401.26 (±327.75) and 49 (27.8%) had CD4 counts of 200 or less. On average, the PLWH reported less than one HIV-related symptom (mean, 0.47 ± 0.83). Close to half of the people living with HIV reported intravenous drug use (46.0%) and 19.3% were men who had sex with men. Summary statistics for other selected correlates are given in Table 1.

Bivariate associations are summarized in Table 2. As expected, caregiver burden was positively and significantly correlated with blame–withdrawal coping (rs = 0.48, p < .05). Caregiver burden was also positively and significantly associated with active-approach coping (rs = 0.29, p < .05) and distancing coping (rs = 0.32, p < .05). Caregiver burden was also positively associated with increased stress (rs = 0.62, p < .05), Beck Depression Inventory scores (rs = 0.44, p < .05), older age (rs = 0.22, p < .05), caring for a patient with a CD4 count less than 200 (rs = 0.16, p < .05), and caring for a patient with poorer physical functioning as assessed by the SF-36 (rs = –0.30, p < .05). Caregiver burden was not associated with gender, race, or living with the patient. Active-approach coping was significantly associated with stress but not with depression. However, stress and depression were both positively and significantly associated with blame–withdrawal coping and distancing coping. Caregivers who lived with the patient reported significantly more blame–withdrawal (rs = 0.17, p < .05) and distancing coping (rs = 0.17, p < .05), but living with the patient was not associated with active-approach coping. Stress was significantly associated with caring for a PLWH with a CD4 count less than 200 (rs = 0.25, p < .05) and caring for a PLWH with poorer physical functioning (rs = –0.36, p < .05).


View this table:
[in this window]
[in a new window]

  		TABLE 2. Spearman Rank-Order Correlations (n = 176)

 

Table 3 summarizes results from seven negative binomial regression models. Model 1 includes main effects for all predictors, models 2 to 4 include main effects of one of the coping subscales and all other covariates, and models 5 to 7 test the first-order interaction of stress and each of the three coping subscales. To facilitate interpretation, all continuous predictor variables were standardized to zero mean and unit variance. Coefficients are exponentiated and give the expected factor change in the number of caregiver burden symptoms reported (33). Gender and race were not significant predictors of caregiver burden. Age, stress, and depression were significant positive predictors of caregiver burden. A one-standard deviation (SD) increase in stress increased the expected count of caregiver burden items endorsed by a factor of approximately 1.40 after adjusting for the other covariates included in the model (coefficients ranged from 1.39 to 1.45 in models 1 through 4). Similarly, a 1-SD increase in Beck Depression Inventory was associated with a significant increase in caregiver burden (the exponentiated coefficients ranged from 1.14 to 1.22 in models 1 to 4). Caring for a PLWH with better physical function, as assessed by the SF-36, tended (p values ranged from .04 to .06 across the seven estimated models) to be associated with lower caregiver burden. Patients’ CD4 counts were not associated significantly with caregiver burden.


View this table:
[in this window]
[in a new window]

  		TABLE 3. Negative Binomial Regression Models Predicting Caregiver Burden

 

In model 1, which included the main effects of all selected predictors, only blame–withdrawal coping was a statistically significant predictor of caregiver burden, although high collinearity among the coping measures may confound interpretation. The joint effect of all three coping measures evaluated simultaneously was statistically significant (Wald {chi}2 = 11.82, df = 3, p < .01). Models 2 through 4 give the adjusted effects of each individual coping measure on caregiver burden. Active-approach coping was not a statistically significant predictor (z = 1.56, p = .118) of caregiver burden when controlling for other covariates (see model 2). Blame–withdrawal coping (model 3) and distancing coping (model 4) positively and significantly predicted caregiver burden.

The joint effect of all three stress by coping interaction effects, evaluated simultaneously, is statistically significant (Wald {chi}2 = 12.40, df = 3, p < .01), indicating the effect of stress is moderated by one or more of the coping strategies. Models 5 through 7 test the hypothesis that the effect of stress is moderated by coping. Specifically, model 5 tests the first-order stress by active-approach coping interaction, model 6 tests the stress by blame–withdrawal coping interaction, and model 7 tests the stress by distancing coping interaction. The first-order perceived stress by coping interaction terms are statistically significant (p < .05) in two of the three models and directionally consistent in all three models. The coefficient for the first-order stress by active-approach coping interaction (model 5) is 0.88; this indicates the effect of stress on caregiver burden becomes weaker as active-approach coping increases. Graphs of the interactions are given in Figure 2. Using the results of model 5 to illustrate, 1-SD increase in stress increases the expected number of endorsed caregiver burden items by a factor of 1.56 when all other variables in the model are equal to 0 (white, male, does not live with patient, patient’s CD4 count >200, and all continuous variables including active coping are at the mean). If active-approach coping increases by 1 SD, the estimated effect of stress on caregiver burden decreases by a factor of 0.89 (e.g., 0.89 * 1.56 = 1.38). Conversely, if active-approach coping decreases by 1 SD, the estimated effect of stress on caregiver burden is 1.75. The first-order stress by blame–withdrawal coping (model 6) was not statistically significant at the 0.05 level (z = –1.78, p = .074), but the stress by distancing coping (model 7) interaction was a significant (z = –2.47, p = .028) predictor of caregiver burden. All of the coefficients indicate the effect of stress on caregiver burden decreases as coping increases.


Figure 223
View larger version (15K):
[in this window]
[in a new window]

  		Figure 2. Adjusted linear effect of stress on the predicted number of caregiver burden items participants endorsed evaluated at the mean, at +1 standard deviation above the mean, and at –1 standard deviation below the mean for three coping strategies.

 

We also looked at the first-order interaction of coping measures with other surrogate indicators of stress, including caregiver’s HIV status and living with the patient. The first-order active approach coping by living with the patient interaction effect on caregiver burden was statistically significant [instant rate ratio (IRR) = 0.81, z = –1.98, p < .05]; living with the patient increased the expected number of caregiver burden items endorsed by a factor of 1.36 when active approach coping was at the mean, 1.10 when active approach coping was 1 SD above the mean, and by a factor of 1.68 when active approach coping was 1 SD below the mean. Thus, the effect of living with the HIV-positive patient on caregiver burden is reduced as active-approach coping increases. The first-order blame–withdrawal coping (IRR = 0.79, z = –1.80, p = .071) and distancing coping (IRR = 0.80, z = –1.58, p = .114) by living with the HIV-positive patient interactions were not statistically significant but were directionally consistent with the hypothesis that increased coping decreases the effect of stressors on caregiver burden. First-order interaction effects of coping measures with caregivers’ HIV status were trivially small and not statistically significant.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
The present study examined the role of coping in the experience of caregiver burden among a heterogeneous sample of caregivers for PLWH. The results indicated that three types of coping—blame–withdrawal, active-approach, and distancing—were significantly positively correlated with caregiver burden. Furthermore, active-approach and distancing coping moderated the relationship between caregiver stress and caregiver burden even after controlling for demographic variables and caregiver depression; blame–withdrawal coping approached significance. Our results suggest that caregivers’ coping mitigated the effects of stress on burden; as coping efforts increased, the impact of stress on burden decreased. Still, some caregivers were likely to experience burden even at low levels of stress or caregiver demand. That is, despite perceiving few HIV symptoms in the PLWH, a subset of caregivers reported high levels of caregiver burden. This is an important finding given that the majority of the caregivers reported relatively low levels of stress from the PLWH’s symptoms in the era of HAART in which patient symptoms and immunosuppression are better controlled. Caregivers who experience burden despite low levels of stress should be studied further in an attempt to understand and alleviate burden. These results also seem to suggest that caregivers may benefit from any attempt to cope with stress. Healthcare providers may facilitate this process when working with people living with HIV and their informal caregivers.

The positive correlation between blame–withdrawal coping and caregiver burden is consistent with the expectation that this emotion-oriented form of coping is often thought to be maladaptive (e.g., (15,17)). However, the significant positive relationship between active-approach coping and caregiver burden was unexpected given that problem-oriented coping is often associated with better outcomes across a range of stressors. There are at least two explanations for this finding. First, although problem-oriented coping strategies are generally thought to be adaptive, they may be maladaptive in those situations when one can do nothing to improve the situation (35). Folkman and colleagues argue that problem-oriented coping strategies might be detrimental when they interfere with the use of emotion-oriented strategies such as acceptance, which might be more beneficial. Perhaps those caregivers who were engaging in active-approach coping experienced increased burden as a result of feelings of helplessness in alleviating the medical problems of the PLWH. Second, the positive relationship of active-approach coping and caregiver burden might be explained by the nature of the subscale. The two active-approach items used, "asked other people for advice or information" and "talked to someone about how you were feeling about (patient) having HIV," might be capturing the search for social support and therefore could be reflecting caregiver distress (35,36). That is, caregivers who are experiencing distress might search for social support in an effort to alleviate it resulting in a significant association between the two variables.

Caregiver burden was also significantly positively associated with stress as measured by the perceived severity of patient symptoms. It was also positively associated with caring for someone with poorer physical functioning and caring for someone with a CD4 count of less than 200. HIV symptoms are often reported to be a problem of frequent and significant concern for the PLWH (e.g., (3)). Fortunately, with improved antiretroviral therapies, prolonged periods of immune system preservation, and longer survival, HIV-related symptoms may be mitigated or postponed. The association of burden with age and depression has been noted in the caregiver literature for chronic illnesses (e.g., (37)), but with HIV, the relationships may be more complicated. Similarly, the lack of association between gender and caregiver burden in the present study might be explained by the greater variability in the nature of the caregiver–patient relationship (e.g., partner, family member, friend) among people living with HIV compared with those with other chronic illnesses.

The present study has the following limitations. First, although this study used a coping measure that was developed using a national sample of adults with HIV, the measure was adapted for the present study to capture the coping efforts of those caring for a PLWH. Additionally, the seven items used might not have adequately captured all relevant forms of coping with caregiving demands. For example, avoidance coping is thought to have two subtypes, distraction and social diversion (13), which may be differentially adaptive. Additionally, the internal consistency of the active-approach and blame–withdrawal subscales was low. Although this is partly a function of the small number of items comprising these scales, improved measurement of these constructs is called for in future research. Other researchers have recognized limitations with coping measures arguing that the coping literature is complex with multiple measures and poor consensus among coping subtypes (e.g., (16,21)). Additionally, some researchers have argued that emotion-oriented subscales of coping are confounded by psychopathology and thus falsely inflate associations with distress (38). The measure we used in the present study may suffer from this limitation. Second, the data were cross-sectional. Although this was an initial test of the moderating role of coping between caregiver stress and burden, longitudinal data may provide further information regarding the predictive relationships among caregiver stress, coping, and burden. This is particularly important given that coping is a complex construct and might be the cause or the consequence of distress within and across various events. Indeed, the possibility that highly burdened caregivers try multiple methods of coping to deal with stress cannot be ruled out with cross-sectional data. Third, an alternative measure of stress might yield different results. The caregiver’s view of the patient’s symptoms is only one way to conceptualize the stress construct. Perhaps another measure such as the patient’s self-report of symptoms or life events would yield alternative results. However, our finding that caregiver burden was also significantly positively associated with the PLWH low CD4 counts and with the PLWH reports of poor physical functioning would suggest that the findings may be similar. Fourth, the caregiver burden measure captured only the presence or absence of various aspects of burden; the degree of individual aspects of burden was not assessed in the present study.

Despite these limitations, the present study expands the literature by further elucidating the role of coping among a large, heterogeneous sample of caregivers for people living with HIV in the era of HAART. Directions for future research include examining the longitudinal relationships among variables and using alternative measures for the stress, coping, and burden constructs.

We thank Dr. Susan Folkman for her helpful discussion of our findings.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 
1The observed distribution of caregiver burden more closely approximated a count variable (overdispersed Poisson) than a normally distributed response. Regression diagnostics indicated the assumptions of ordinary least squares (OLS) regression were not well approximated. Therefore, we chose negative binomial regression with robust standard errors as a more appropriate model specification. Both OLS regression and iteratively reweighted least squares regression (one form of robust regression) using caregiver burden as a continuous response gave results concerning statistical tests consistent with those we report. Back

2A methodological justification for testing each interaction separately is the interpretational confounding that would be the result of the multicollinearity introduced by including three interaction terms all including a common component (stress). Back

This study was funded in part by the National Institutes of Health grants MH63051. Dr. Stein is a recipient of a NIDA Mid-Career Investigator Award K24 DA 00512.

DOI:10.1097/01.psy.0000245901.82935.cb


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Turner HA, Catania JA. Informal caregiving to persons with AIDS in the United States: caregiver burden among central cities residents eighteen to forty-nine years old. Am J Community Psychol 1997;25:35–59.[CrossRef][Medline]
  2. Pinquart M, Sorenson S. Associations of stressors and uplifts of caregiving with caregiver burden and depressive mood: a meta-analysis. J Gerontol Psychol Sci 2003;58B:112–28.
  3. Pakenham KI, Dadds MR, Terry DJ. Adaptive demands along the HIV disease continuum. Soc Sci Med 1996;42:245–56.[CrossRef][Medline]
  4. Wight RG, Aneshensel CS, Leblanc AJ. Stress buffering effects of family support in AIDS caregiving. AIDS Care 2003;15:595–613.[CrossRef][Medline]
  5. Land H, Hudson SM, Stiefel B. Stress and depression among HIV-positive and HIV-negative gay and bisexual AIDS caregivers. AIDS Behav 2003;7:41–53.[CrossRef][Medline]
  6. Folkman S, Chesney MA, Cooke M, Boccellari A, Collette L. Caregiver burden in HIV-positive and HIV-negative partners of men with AIDS. J Consult Clin Psychol 1994;62:746–56.[CrossRef][Medline]
  7. Wight RG. Precursive depression among HIV infected AIDS caregivers over time. Soc Sci Med 2000;51:759–70.[CrossRef][Medline]
  8. Prachakul W, Grant JS. Informal caregivers of persons with HIV/AIDS: a review and analysis. J Assoc Nurses AIDS Care 2003;14:55–71.[CrossRef][Medline]
  9. Pirraglia PA, Bishop D, Herman D, Trisvan E, Lopez RA, Torgersen CS, Van Hof AM, Anderson BJ, Miller I, Stein MD. Caregiver burden and depression among informal caregivers of HIV-infected individuals. J Gen Intern Med 2005;20:510–4.[CrossRef][Medline]
  10. Pakenham KI, Dadds MR, Terry DJ. Carers’ burden and adjustment to HIV. AIDS Care 1995;7:189–203.[CrossRef][Medline]
  11. Demi A, Bakerman R, Moneyham L, Sowell R, Seals B. Effects of resources and stressors on burden and depression of family members who provide care to an HIV-infected woman. J Fam Psychol 1997;11:35–48.[CrossRef]
  12. Irving GA, Bor R, Catalan J. Psychological distress among gay men supporting a lover or partner with AIDS: a pilot study. AIDS Care 1995;7:605–17.[CrossRef][Medline]
  13. Endler NS, Parker JDA. Multidimensional assessment of coping: a critical evaluation. J Pers Soc Psychol 1990;58:844–54.[CrossRef][Medline]
  14. Cattanach L, Rodin J. Psychosocial components of the stress process in bulimia. Int J Eat Disord 1988;7:75–88.
  15. Koff E, Sangani P. Effects of coping style and negative body image on eating disturbance. Int J Eat Disord 1997;22:51–6.[CrossRef][Medline]
  16. Martin SC, Wolters PL, Klaas PA, Perez L, Wood LV. Coping styles among families of children with HIV infection. AIDS Care 2004;16:283–92.[CrossRef][Medline]
  17. McCausland J, Pakenham KI. Investigation of the benefits of HIV/AIDS caregiving and relations among caregiving adjustment, benefit finding, and stress and coping variables. AIDS Care 2003;15:853–69.[CrossRef][Medline]
  18. Shatford LA, Evans DR. Bulimia as a manifestation of the stress process: a LISREL causal modeling analysis. Int J Eat Disord 1986;5:451–73.
  19. Skinner EA, Edge K, Altman J, Sherwood H. Searching for the structure of coping: a review and critique of category systems for classifying ways of coping. Psychol Bull 2003;29:216–69.
  20. Blaney NT, Goodkin K, Morgan RO, Feaster D, Millon C, Szapocznik J, Eisdorfer C. A stress moderator model of distress in early HIV-1 infection: concurrent analysis of life events, hardiness, and social support. J Psychosom Res 1991;35:297–305.[CrossRef][Medline]
  21. Connor-Smith JK, Comaps BE. Coping as a moderator of relations between reactivity to interpersonal stress, health status, and internalizing problems. Cognit Ther Res 2004;28:347–68.[CrossRef]
  22. Schroder KEE. Coping competence as predictor and moderator of depression among chronic disease patients. J Behav Med 2004;27:123–45.[CrossRef][Medline]
  23. Robinson BC. Validation of a caregiver strain index. J Gerontol 1983;38:344–8.[Abstract/Free Full Text]
  24. Fleishman JA, Sherbourne CD, Cleary PD, Wu AW, Crystal S, Hays RD. Patterns of coping among persons with HIV infection: configurations, correlates, and change. Am J Community Psychol 2003;32:187–204.[CrossRef][Medline]
  25. Pakenham KI, Rinaldis M. Development of the HIV/AIDS stress scale. Psychol Health 2002;7:203–19.[CrossRef]
  26. Lazarus RS, Folkman S. Stress, Appraisal, and Coping. New York: Springer; 1984.
  27. Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J. An inventory for measuring depression. Arch Gen Psychiatry 1961;4:561–71.[Abstract/Free Full Text]
  28. Beck AT, Steer RA, Garbin MG. Psychometric properties of the Beck Depression Inventory: twenty-five years of evaluation. Clin Psychol Rev 1988;8:77–100.[CrossRef]
  29. Ware JE, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). Med Care 1992;30:473–83.[Medline]
  30. McHorney CA, Ware JE, Baczek AE. The MOS-36-item short-form health survey (SF-36): II. Psychometric and clinical tests of validity in measuring physical and mental health constructs. Med Care 1993;31:247–63.[Medline]
  31. Calsyn DA, Saxon AJ, Bush KR, Howell DN, Baer JS, Sloan KL, Malte CA, Kivlahan DR. The Addiction Severity Index medical and psychiatric composite scores measure similar domains as the SF-36 in substance-dependent veterans: concurrent and discriminant validity. Drug Alcohol Depend 2004;76:165–71.[CrossRef][Medline]
  32. McHorney CA, Ware JE, Lu JFR, Sherbourne CD. The MOS 36-item short-form health survey (SF-36): III. Tests of data quality, scaling assumptions, and reliability across diverse patient groups. Med Care 1994;32:40–66.[Medline]
  33. Long JS. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage Publications; 1997.
  34. Stata Statistical Software: Release 8.2. College Station, TX: Stata Corp; 2003.
  35. Folkman S, Moskowitz JT. Coping: pitfalls and promise. Annu Rev Psychol 2004;55:745–74.[CrossRef][Medline]
  36. Charlton PFC, Thompson JA. Ways of coping with psychological distress after trauma. Br J Clin Psychol 1996;35:517–30.[Medline]
  37. Teri L. Behavior and caregiver burden: behavioral problems in patients with Alzheimer’s disease and its association with caregiver distress. Alzheimers Dis Assoc Disord 1997;11:S35–8.
  38. Stanton AL, Danoff-Burg S, Camerson CL, Ellis AP. Coping through emotional approach: problems of conceptualization and confounding. J Pers Soc Psychol 1994;66:350–62.[CrossRef][Medline]




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Engler, P.
Right arrow Articles by Stein, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Engler, P.
Right arrow Articles by Stein, M.
Related Collections
Right arrow Infectious Disease
Right arrow Stress and Coping
Right arrow HIV/AIDS

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS