| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
| ||||||||||||||||||||||||||||||||
RESEARCH ARTICLE |
Department of Psychiatry, University of Rochester Medical Center, New York.
Address correspondence to Paul R. Duberstein, Laboratory of Personality and Development, Box PSYCH, University of Rochester Medical Center, 300 Crittenden Blvd., Rochester, NY 14642. E-mail: Paul_Duberstein{at}urmc.rochester.edu
 |
Abstract |
|---|
 TOP Abstract MethodResultsDiscussionReferences |
|---|
A deeper understanding of the determinants of poor perceived health is needed, given its association with major health outcomes, including level of functioning (Bosworth et al., 1999
), health care service utilization (Hansen, Fink, Frydenberg, & Oxhoj, 2002), and mortality (Benyamini & Idler, 1999; Idler & Benyamini, 1997). Because judgments about health are affectively charged (Benyamini, Idler, Leventhal, & Leventhal, 2000), it is not surprising that personality traits that reflect long-standing patterns of emotional experience and expression such as high neuroticism and low extraversion contribute to poor perceived health (Duberstein et al., 2003; Hooker, Monahan, Bowman, Frazier, & Shifren, 1998; Kempen, Jelicic, & Ormel, 1997). Much of the research has examined the influences of personality on global scales of perceived health, not responses to individual items. However, health care practitioners in busy primary care settings are far more likely to ask patients a few select questions in a face-to-face conversation rather than administer a long battery of items. Similarly, epidemiological studies might incorporate one or two items tapping perceived health, not lengthier scales.
On the basis of the premise that different personality traits predict responses to items of differential valence (positive vs negative) and temporal phrasing (past, present, future), we examined the associations between personality traits, depression, physical illness, and responses to four self-report items drawn from the Medical Outcomes Survey Short Form-36 (SF-36), a widely used measure in health services research. From a clinical perspective, understanding the extent to which personality influences individuals' health ratings—independent of actual medical burden—may provide practitioners with another valuable piece of information in the difficult task of evaluating the accuracy and severity of patient health ratings.
We designed the present study to test three hypotheses. First, we hypothesized that higher Neuroticism and lower Extraversion scores would be associated with reports of poorer perceived health on all four items, over and above the effects of physical morbidity (Goodwin & Engstrom, 2002). Second, given the documented associations between Extraversion and future-related constructs such as hopelessness and optimism (Marshall, Wortman, Vickers, Kusulas, & Hervig, 1994), we hypothesized that future expectations of poor health ("I expect my health to get worse") would be associated with low Extraversion scores more so than with high Neuroticism scores. Third, we hypothesized that controlling for the effects of active depressive symptoms, or "state depression," would diminish or eliminate the influence of Neuroticism on items phrased in the present tense ("I seem to get sick a little easier than other people," "I am as healthy as anybody I know," and "My health is excellent"), because prior work has demonstrated a powerful link between active depressive symptoms and contemporaneous perceptions of health (Miller et al., 1996).
|
METHOD |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
), oversampling (i.e., 76 of 156) those who scored above a cutoff of 21 (Lyness et al., 1997) but also including a random sample of those who scored below 21 (i.e., 229 of 507). Of the 305 participants, 39 did not complete the personality inventory and were excluded from the analyses; they did not differ from the other participants in terms of age and gender distribution, but they were less educated: M = 13.4, SD = 2.8 versus M = 12.1, SD = 2.9, t(301) = 2.81, p =.005. Age ranged from 60 to 94 years (M = 71.2, SD = 7.5); Whites constituted a majority of the sample (n = 245, 93.2%). Slightly more than half (n = 132, 50.3%) of the participants were married, and most (n = 217, 82.5%) were retired; 41 had been prescribed an antidepressant.
Instruments
NEO-Five Factor Inventory
We operationalized personality by use of the NEO-Five Factor Inventory (NEO-FFI; Costa & McCrae, 1992), a 60-item self-report questionnaire measuring each of the Big Five personality factors (Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness). The coefficient alpha for the five scales in the current study ranged from 
= 0.67 (Openness) to = 0.88 (Neuroticism). The NEO inventories have demonstrated reliability and validity in older samples (Duberstein et al., 2003; Hooker et al., 1998).
Cumulative Illness Rating Scale
To ensure that the relationships between personality and subjective health were not confounded by objective medical burden, we quantified the amount of physical disease in each organ system at the time of study entry by using the Cumulative Illness Rating Scale (CIRS; Linn, Linn, & Gurel, 1968), a validated (Conwell, Forbes, Cox, & Caine, 1993) physician-rated index derived by means of patient history as well as physical examination and laboratory findings. Higher scores indicate greater disease burden.
Hamilton Depression Rating Scale
We controlled for depressive symptomatology by using the Hamilton Depression Rating Scale (HDRS; Williams, 1988), a 24-item examiner-rated, structured assessment of the presence and severity of depressive symptoms in the week prior to interview. Higher HDRS scores reflect greater depressive symptoms. Cronbach's alpha in the present study was = 0.84.
General Health Perceptions Scale, SF-36
We dichotomized four items from the General Health Perceptions scale of the SF-36 (Ware & Sherbourne, 1992), two positively valenced items ("I am as healthy as anybody I know" and "My health is excellent") and two negatively valenced questions ("I seem to get sick a little easier than other people" and "I expect my health to get worse"). Respondents who endorsed "neutral," "definitely true," or "mostly true" in response to the positive items were categorized as reporting good perceived health. The remainder were categorized as reporting poor perceived health. Respondents who responded "definitely false" and "mostly false" to the negative items were categorized as having good perceived health; the others were categorized as having poor perceived health.
Data Analysis
For each of the outcomes variables, we conducted two logistic regression models. In Model 1, predictors included Neuroticism, Extraversion, Openness, Agreeableness, and Conscientiousness. Covariates were age, gender, and the CIRS score. Model 2 was identical to Model 1 except it also included the HDRS score as a covariate.
|
RESULTS |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
|
|
|
DISCUSSION |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
Although Neuroticism has been associated with poorer perceived health in prior research with older populations (Duberstein et al., 2003; Hooker et al., 1998; Kempen et al., 1997), the present results suggest two interesting qualifiers. First, the association of Neuroticism and response to the item "I am as healthy as anybody I know" appears to be driven more by current depressive symptoms than by long-standing, trait negative affect, although we recognize that it is difficult to disentangle these two constructs in a cross-sectional design and trait negative affect has been shown to amplify the risk for mood disorders (Ormel, Oldehinkel, & Vollerbergh, 2004). Because this item seems to refer to immediate rather than general circumstances, responses may be strongly influenced by "within the past week" depressive somatic symptoms captured by the HDRS, such as sleeplessness and anergia. Second, Extraversion, rather than Neuroticism, influences responses to the item "I expect my health to get worse." Thus, individuals scoring higher on Neuroticism are likely to report worse health in response to items that do not require future forecasts. One implication is that judgments about poor health that reflect dispositional worry, anxiety, and somatic vigilance might be met with interventions aimed at least partly at allaying emotional distress, rather than an immediate launch into costly and time-consuming laboratory tests and medical procedures (Epstein et al., 2005).
The association between Extraversion and responses to the item "I expect my health to get worse" is robust. It is not surprising that individuals lacking gregariousness, vigor, and positive affect render more gloomy health prognostications. Marshall and colleagues (1994) found significant positive associations between high Extraversion and a factor they labeled "optimistic control." Another study reported that low Extraversion was associated with worry about one's health in the future (Caramela-Miller, Sterns, & Murphy, 1999), and in depressed patients Extraversion is associated with hopelessness, independent of Neuroticism (Duberstein, Conner, Conwell, & Cox, 2001). Because positive affect (more than negative affect) underlies thoughts about the future (Tellegen, 1985), it too may drive Extraversion's association with denials of future decline.
Prudence dictates a balanced interpretation of these results. Although Extraverted individuals render more optimistic predictions of health, Extraversion probably does not inexorably serve a protective function. A level of optimism that underestimates risks may be markedly maladaptive, and the present finding highlights the need to investigate Extraversion's health behavior correlates. Clinically, health care practitioners may wish to be mindful of the extent to which highly Extraverted individuals appear able to realistically anticipate future health risks. Reminders about standard health risks related to aging may be warranted, and greater attention to preventive interventions may reduce health risks and costs that would otherwise accrue from failures to notice or seek treatment for developing problems.
The current cross-sectional design limits causal interpretations. Worse perceived health may lead to an increase in Neuroticism, and the expectation of worsening health could lower Extraversion. The relationship between perceived health and affectively oriented personality traits over time is likely complex and reciprocal. Generalizability to other geographic regions, demographic groups, or disease-specific populations cannot be guaranteed. Relationships between personality and perceived health may also be moderated by the presence of a particular disease or symptom. Participants were oversampled for depression, not randomly sampled. We did not control for alcohol intake, and data on other health behaviors, such as diet, exercise, and smoking, were unavailable. Finally, few people reported poor health on the first item.
The study's strengths include its use of rigorous controls for depression and physical morbidity, and reliance on a comprehensive personality taxonomy, although alternative conceptions to the Five Factor Model also hold considerable promise for understanding the links between personality, aging, and health (Hooker & McAdams, 2003). Whereas the current study highlights the contribution of Neuroticism and Extraversion to perceived health, the findings should not be interpreted to mean that health care providers working with older adults should ignore other personality traits.
In summary, the present cross-sectional findings suggest that personality traits influence responses to individual items tapping perceived health. Given that perceived health has prognostic implications for a host of health outcomes (Benyamini & Idler, 1999; Idler & Benyamini, 1997), rigorous, prospective research on the determinants and modifiability of perceived health is needed. The role of personality in perceived health requires careful consideration in future research.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
Received for publication November 16, 2005. Accepted for publication April 5, 2006.
|
References |
|---|
|
TOPAbstractMethodResultsDiscussionReferences |
|---|
This article has been cited by other articles:
 |
|
 |
|
  S. Sorensen, P. R. Duberstein, B. Chapman, J. M. Lyness, and M. Pinquart How Are Personality Traits Related to Preparation for Future Care Needs in Older Adults? J. Gerontol. B. Psychol. Sci. Soc. Sci., November 1, 2008; 63(6): P328 - P336. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. P. Chapman, J. M. Lyness, and P. Duberstein Personality and Medical Illness Burden Among Older Adults in Primary Care Psychosom Med, April 1, 2007; 69(3): 277 - 282. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||
| HOME | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|---|
