| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
Correspondence: Address correspondence to Yuri Jang, PhD, Department of Aging and Mental Health, Florida Mental Health Institute, University of South Florida, 13301 Bruce B. Downs Blvd., Tampa, FL 33612. E-mail: yjang{at}fmhi.usf.edu
 |
Abstract |
|---|
 TOP Abstract MethodsResultsDiscussionReferences |
|---|
Key Words: Caregiver intervention • Neuroticism • Depression • Alzheimer's disease
Empirical evidence has shown that psychosocial and educational interventions can lower caregivers' depressive symptoms and enhance mental health (Marriott, Donaldson, Tarrier, & Burns, 2000; Mittelman et al., 1995; Mittelman, Roth, Coon, & Haley, 2004). However, the effects of interventions vary greatly in relation to the contents of programs, characteristics of caregivers and patients, and other contextual factors (Cooke et al., 2001; Sörensen, Pinquart, & Duberstein, 2002). In order to maximize the benefits of interventions, it is essential to identify characteristics of caregivers who are likely to be most responsive and to tailor these interventions to the specific needs of the caregivers.
Among the important factors that may influence the effectiveness of an intervention are the caregiver's personality traits, particularly neuroticism. Neuroticism is defined as an individual's propensity to experience negative emotions (Costa & McCrae, 1985). Individuals who are high in neuroticism have been characterized as "worrying, nervous, emotional, insecure, inadequate, and hypochondriacal" (Costa & McCrae, 1985, p. 2). They also have been shown to appraise situations as more stressful or threatening and to be more susceptible to physical problems and psychological distress (Jang, Mortimer, Haley, & Graves, 2002; Watson & Pennebaker, 1989). In the caregiving literature, caregivers who score high on neuroticism have been shown to report higher levels of functional impairment and problematic behaviors of their care recipients, and to experience more strain and depressive symptoms than those who score low on neuroticism (Bookwala & Schulz, 1998; Hooker, Monahan, Bowman, Fraizer, & Shifren, 1998). These studies suggest that neuroticism may be associated not only with negative caregiving outcomes but also with the way in which caregivers respond to interventions. Because the cross-sectional association of neuroticism and depression does not necessarily imply a causal role for personality, it is important to assess whether neuroticism predicts longitudinal changes in caregiver depression. To our knowledge, none of the studies of caregiving and neuroticism to date have examined this issue.
Recently, there has been an interest in the effects of personality characteristics on response to interventions or treatment outcomes. In clinical studies with depressed patients, high neuroticism has been found to predict worse response to various treatments, including psychotherapy, behavioral therapy, and pharmacological interventions (Mulder, 2002; Taylor & McLean, 1993). The role of neuroticism in treatment response should be addressed with intervention studies of nonclinical populations. The identification of personality characteristics affecting treatment outcomes would provide valuable information for designing and implementing more effective interventions for caregivers.
When neuroticism is considered as a determinant of the effectiveness of an intervention using changes in depressive symptoms as an outcome index, several conceptual and methodological issues should be addressed. One potential concern is that the two constructs—neuroticism and depression—would overlap, sharing so much variance that they would be essentially indistinguishable. However, studies have shown that personality predisposition for depression is modest, and neuroticism and depression are generally considered to be correlated but distinctive constructs (Enns & Cox, 1997).
Another important issue in assessing the effects of neuroticism on changes in depression in a longitudinal study is the extent to which each of the constructs consists of both state and trait components. There is a general consensus that neuroticism is a stable and enduring personality trait (Costa & McCrae, 1992), whereas the state–trait issue in depression is more complex and debatable. Recent longitudinal studies have shown evidence for both state and trait characteristics in depression, demonstrating both episodic variability with onset and termination and trait-like proneness or vulnerability (Davey, Halverson, Zonderman, & Costa, 2003; Enns & Cox, 1997). Because many caregiving studies utilize depression as an outcome, it is of interest to see whether change in depression varies as a function of the trait of neuroticism.
In the present investigation, we utilized the NYU Spouse-Caregiver Intervention Study (where NYU stands for New York University) in assessing the role of one personality characteristic, neuroticism, in determining the effectiveness of an intervention. The treatment program is a comprehensive psychosocial intervention with three major components (individual and family counseling, support group participation, and ad hoc consultation), and it has previously proved to be effective for reducing depression in spouse-caregivers (Mittelman et al., 1995, 2004) and postponing or preventing institutionalization of AD patients (Mittelman et al., 1993, 1996). Given this evidence of efficacy, in the present analysis we specifically examined the relationship of caregiver neuroticism to change in caregiver depression over time, and to the degree of benefit earned from the intervention, using caregivers' depressive symptoms as an index of change. We hypothesized that caregivers high in neuroticism would experience higher levels of depression that would persist over time, and that these caregivers would also be less likely to respond to the enhanced psychosocial intervention compared with caregivers low in neuroticism.
|
Methods |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
Participating caregivers were randomly assigned either to an enhanced treatment group or to a usual-care control group. Caregivers in the enhanced treatment group were provided with a comprehensive program including individual and family counseling, support group participation, and ad hoc consultation. Individual and family counseling sessions were tailored to each specific situation, with a major purpose of increasing support for the spouse-caregivers from other family members. Upon enrolling, caregivers agreed that they would join weekly support groups 4 months later, after two individual and four family counseling sessions were completed. In addition, ad hoc counseling was available to caregivers at any time they need additional counseling or advice. More detailed information on the content and procedure of the program has been reported elsewhere (Mittelman, Epstein, & Pierzchala, 2002). The control group in the NYU Spouse-Caregiver Intervention Study is a usual-care group rather than a pure control group with no treatment. The same highly skilled counselors were available to both treatment and control groups, and a greater level of information and support was accessible to caregivers at the NYU-ADC compared with that available in most medical and community settings. General services provided to all families of patients at the NYU-ADC were available to these caregivers, including access to ad hoc counseling for the purpose of information about or referral to resources. Consequently, many caregivers in the usual-care group received more help than might have been typically available to caregivers who did not participate in the study. They also could join support groups on their own. However, no special efforts were made to make sure that they utilized these resources, and they did not receive the formal counseling provided to the treatment group.
During a 9
-year period, a total of 406 spouse-caregivers of AD patients were enrolled in a randomized controlled trial with two successive cohorts, one with enrollment that started in 1987 and the other in 1991. We have recently demonstrated that both cohorts of caregivers showed similar benefits in depressive symptomatology from intervention (Mittelman et al., 2004). After about 1 year into the recruitment of the first cohort, the NEO Personality Inventory (NEO-PI; Costa & McCrae, 1992) was added to the intake assessment battery. It was administered to all caregivers who enrolled thereafter until mid-1995. Thus, of these 406 spouse-caregivers, 320 completed the NEO-PI; all analyses reported in the present paper are based on those 320 caregivers. Follow-up interviews were conducted every 4 months for the first year after intake into the study, resulting in four evaluations that were used in the present analyses (i.e., the baseline, 4-month, 8-month, and 1-year assessments). Among the total study sample of 406 caregivers, 23 (5.7%) dropped out within 1 year of enrollment, of whom 5 (1.2%) died, 6 (1.5%) became too ill to participate, and 12 (3.0%) refused to continue.
Measures Used in the Analysis
Neuroticism was measured with the 48 items from the NEO-PI (Costa & McCrae, 1992). The NEO-PI has shown excellent reliability and validity in numerous studies. We asked individuals to indicate how much they agreed with statements such as "I often feel tense and jittery," and "I often get angry at the way people treat me," using a 5-point scale ranging from strongly agree to strongly disagree. Positively worded items were reverse coded, and all responses were summed for total scores for neuroticism. The range was from 0 (low neuroticism) to 192 (high neuroticism).
The Geriatric Depression Scale (GDS; Yesavage et al., 1983) was used to measure symptoms of caregiver depression. The GDS was designed specifically for the assessment of depressive symptoms in older populations, excluding items related to somatic complaints. Numerous studies have tested the reliability and validity of the scale with various samples and verified its psychometric properties (Stiles & McGarrahan, 1998). The GDS includes 30 items with a yes–no format, such as "Do you feel that your life is empty?" and "Are you in good spirits most of the time?" We calculated the total score by counting the number of responses that suggest probable depression. Scores can range from 0 (no depressive symptoms) to 30 (severe depressive symptoms).
Analytic Strategy
We conducted mixed model growth curve analyses using SAS Proc Mixed (Littell, Milliken, Stroup, & Wolfinger, 1996) to examine changes in depression during the first year after enrollment in the study. These growth curve models offer important advantages over more traditional repeated measures analyses (Francis, Fletcher, Stuebing, Davidson, & Thompson, 1991). Using this approach, we fit growth curves for each individual caregiver's data based on the amount of data that this person provided, and this allowed caregivers who discontinued the study prior to the 1-year assessment to be included in the analytic models without imputation of data for the missing observations. We also included variability in the actual time of the assessments in the growth curve models by analyzing time as a random effect.
We modeled individual growth curve parameters as a function of baseline depression score, gender, the number of years spent in caregiving, neuroticism, treatment group (enhanced treatment vs. usual care), and the time since baseline that each particular observation was made. We used restricted maximum likelihood estimation and specified an unstructured covariance structure. We analyzed the depression scores obtained after treatment onset (i.e., 4 months, 8 months, and 1 year) as repeated observations of the dependent variable, with mean-centered baseline depression scores serving as a covariate. We also entered neuroticism and time since baseline (in weeks) as mean-centered, continuous predictor variables. We obtained these mean-centered scores by subtracting the mean across all caregivers on a particular variable from each caregiver's observed score on that variable, resulting in scores with overall means of zero. Consequently, these mean-centered scores represent deviations from the mean on each respective predictor variable.
We estimated the time-since-baseline effect by using a logarithmic growth model. In this model, we used the natural logarithm of the number of weeks since baseline as the time-since-baseline predictor. In previous analyses of the effect of treatment on the depression data from the first year after intake (Mittelman et al., 2004), both linear and logarithmic growth models were examined. The logarithmic growth model was found to fit slightly better than the linear model according to the Akaike information criterion (AIC), so we used the logarithmic growth model here in our analyses of the impact of the caregivers' neuroticism levels on depression as a measure of treatment response.
In addition to main effects, we tested two-way interaction effects among study variables in the mixed model growth curve analyses. The main effect for treatment group and the Treatment group x Time interaction effects reported here on a restricted sample are similar to those already reported by Mittelman and colleagues (2004) for the entire group of study participants. The uniqueness of the present analyses is in examining the effects of neuroticism on changes in depression (i.e., the neuroticism main effect, the Neuroticism x Treatment group interaction effect, and the Neuroticism x Time interaction effect).
|
Results |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
|
Effects of Neuroticism on Changes in Depression
Table 2 presents the results from the logarithmic growth model for the first year after enrollment into the study. The significant predictors of depressive symptoms over the first year after enrollment are the baseline depression score, neuroticism, treatment group, the Neuroticism x Time-since-baseline interaction effect, the Treatment group x Time interaction effect, and the Baseline depression x Time interaction effect. The baseline depression covariate effect, treatment group main effect, Treatment group x Time interaction effect, and Baseline depression x Time interaction effect were all previously found to be significant and reported by Mittelman and colleagues (2004). These findings generally represent the positive effects of the enhanced intervention condition in alleviating caregivers' depressive symptoms over time. Of interest here are the main effect for neuroticism and the neuroticism interaction effects.
|
|
The Neuroticism x Time interaction effect (p =.0029) indicates that depression scores tended to increase across the first year for caregivers who were high in neuroticism, but decreased across time for caregivers who were low in neuroticism. Combining this interaction effect with the Treatment group x Time interaction effect (p =.0317) leads to the interesting pattern observed in Figure 1. It is clear that the high neuroticism caregivers in the usual-care group tended to show the greatest increases in depression, although, again, this represents additive and not synergistic effects. We ran a subsequent model that added a three-way interaction effect among neuroticism, treatment group, and time since baseline, but this effect was not statistically significant (p =.34). Consequently, both treatment group and neuroticism exerted additive effects in predicting depression levels, and we found the intervention to be equally effective in reducing depression for both low and high neuroticism caregivers.
With a supplementary analysis, we confirmed that neuroticism was not associated with caregiver attrition in either treatment group. Our findings from a logistic regression model of dropout rate showed no effects for treatment group (p =.48), neuroticism (p =.56), or Treatment group x Neuroticism (p =.92).
|
Discussion |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
In the present study, our focus was on exploring the role of neuroticism in qualifying the demonstrated effectiveness of the NYU Spouse-Caregiver Intervention program (Mittelman et al., 1995, 2004) in reducing caregivers' depressive symptoms in the first year after enrollment. The results of growth curve modeling showed interesting patterns of relationships between neuroticism and changes in depression for caregivers in the enhanced treatment group and the usual-care control group.
Our analysis showed that, regardless of treatment group membership, caregivers who scored high in neuroticism had higher levels of depression than their counterparts who scored low on neuroticism. The observed interaction between neuroticism and time since baseline demonstrated that caregivers with high neuroticism were likely to experience an increase in symptoms of depression during the 1-year period, whereas those with low neuroticism would experience a decrease in depression in both groups. The predisposition to depression associated with the personality trait of neuroticism seems to play an interactive role with the ongoing caregiving situation, further exacerbating depressive symptoms as time goes on. We found no interaction effects between neuroticism and treatment group, suggesting that the enhanced treatment was effective in managing depression across all levels of neuroticism. The enhanced intervention not only reduced the symptoms of depression among caregivers with low neuroticism, as expected, but also helped those with high neuroticism avoid increasing levels of depression. It is notable that caregivers who were high in neuroticism adhered to the treatment at high rates and benefited from it. The high adherence rate for an at-risk population adds to the clinical significance of these findings.
The findings from this study are important for several reasons. First, the results demonstrate that neuroticism is a risk factor for increased caregiver distress over time. Although neuroticism and other dimensions of personality have been shown to be associated with caregiver appraisal and well-being (Bookwala & Schulz, 1998; Hooker et al., 1998), to our knowledge this is the first longitudinal demonstration of the effects of neuroticism on intervention outcomes. The results clearly indicate that the relationship between personality and depressive symptoms is more than just a confound of the two constructs, and that personality can modify the longitudinal course of adjustment to the caregiving role. This finding suggests that caregivers high in neuroticism, who can easily be identified by brief assessments (e.g., NEO Five Factor Inventory; Costa & McCrae, 1992), could be targeted for intervention as a high-risk group.
Our results also show that individuals both high and low in neuroticism benefit from the enhanced intervention, although caregivers with high neuroticism receiving the usual care had higher depression scores over time. This is contrary to the previous literature on treatment of depression in which individuals with personality disorders or high scores on neuroticism responded more poorly to pharmacological, psychological, and behavioral interventions and psychotherapies (Mulder, 2002; Taylor & McLean, 1993). The comparable amount of benefit of the enhanced treatment to all caregivers suggests that even those with difficult traits may benefit from certain psychosocial interventions. Our positive findings show that personality need not be a barrier to successful intervention, at least with programs similar to the NYU enhanced treatment package.
Several features of the NYU program contribute to the favorable outcomes. Although not designed with caregiver personality as a target of intervention, the enhanced treatment used in the present study includes a number of features that may be ideal for working with caregivers whose personalities predispose them to more negative outcomes. The NYU enhanced treatment includes individually tailored interventions that utilized individual counseling (including structured in-person sessions and ad hoc phone sessions), as well as structured family interventions and encouragement to use support groups. Counselors, who all had at least master's degree preparation and many years of previous experience, thus had the opportunity to provide individually designed approaches to caregivers and their families, within the context of the intervention protocol, that may have been useful in overcoming the negativity, relatively poor ability to handle stress, and interpersonal strain characteristic of individuals with high neuroticism. For example, counselors in the enhanced treatment program may have been able to directly intervene with family members to elicit their support for the caregiver when a direct approach by the caregiver might have been less effective. In addition, the counselors took into account the caregivers' personalities in helping them set realistic goals for themselves.
On the basis of our findings and the existing literature, we expect that other psychosocial interventions for caregivers that rely less on interventionists with considerable clinical training, and interventions that do not allow for individualized tailoring of treatment, will not be as effective with caregivers with high neuroticism. Future studies of other caregiver interventions—particularly those such as support groups—should address this question of whether less structured and individualized interventions also can be effective across a broad spectrum of personality traits.
It should be noted that most participants in the NYU Spouse-Caregiver Intervention Study had at least a high school education and the sample included few non-Whites. There is a need for further studies on the issues of personality and response to interventions that take into consideration other personality characteristics and include participants with a wider range of backgrounds.
The present study suggests that longitudinal studies of caregiving and caregiver intervention studies should pay closer attention to personality variables as potential predictors of the longitudinal course of caregiver depression. Our findings lend support to the usefulness of enhanced, individualized treatment intervention programs for caregivers, indicating that the programs should be more widely disseminated.
|
Footnotes |
|---|
1 Department of Aging and Mental Health, University of South Florida, Tampa.
2 Department of Psychology, University of Alabama at Birmingham.
3 Department of Biostatistics, University of Alabama at Birmingham.
4 School of Aging Studies, University of South Florida, Tampa.
5 Department of Psychiatry, New York University School of Medicine, New York.
Decision Editor: Linda S. Noelker, PhD
Received for publication May 28, 2003. Accepted for publication December 22, 2003.
|
References |
|---|
|
TOPAbstractMethodsResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  J. E. Gaugler, R. L. Kane, and R. Newcomer Resilience and Transitions From Dementia Caregiving J. Gerontol. B. Psychol. Sci. Soc. Sci., January 1, 2007; 62(1): P38 - P44. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. L. Lewis, K. Hepburn, S. Narayan, and L. N. Kirk Relationship matters in demantia caregiving American Journal of Alzheimer's Disease and Other Dementias, November 1, 2005; 20(6): 341 - 347. [Abstract] [PDF]
|
|
|
|
 |
|
 G. Berger, T. Bernhardt, E. Weimer, J. Peters, T. Kratzsch, and L. Frolich Longitudinal Study on the Relationship Between Symptomatology of Dementia and Levels of Subjective Burden and Depression Among Family Caregivers in Memory Clinic Patients J Geriatr Psychiatry Neurol, September 1, 2005; 18(3): 119 - 128. [Abstract] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
