| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
ORIGINAL ARTICLES |
From the Consultation-Liaison Service, Department of Psychiatry (S.L.S.), and Department of Medicine (H.P.H.), University of Texas Health Science Center at San Antonio, San Antonio, Texas; and Purdue Pharma (R.D.), Stamford, Connecticut.
Address reprint requests to: Stephen L. Stern, MD, Department of Psychiatry, Mail Code 7792, University of Texas Health Science Center at San Antonio, 7703 Floyd Curl Dr., San Antonio, TX 78229-3900. Email: sterns{at}uthscsa.edu
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
METHODS: A total of 795 persons aged 64 to 79 years completed an English or Spanish version of the 30-item Geriatric Depression Scale on entering the San Antonio Longitudinal Study of Aging, an epidemiologic survey, between 1992 and 1996. Women constituted 58% and Mexican Americans 54% of this randomly selected sample. Subjects who answered "no" to the item "Are you hopeful about the future?" were classified as hopeless.
RESULTS: As of August 1999, 29% of the 73 hopeless subjects had died, compared with 11% of the hopeful, a highly significant difference. The mortality rates for cardiovascular disease and cancer were significantly greater among the hopeless subjects (7%) than among the hopeful (3%). Hopelessness predicted all-cause mortality in a Cox proportional hazards model adjusted for age, ethnic background, current smoking status, number of comorbid medical conditions, self-rated health, and frequency of social contacts (risk ratio = 2.23, 95% confidence interval = 1.33 to 3.76, p = .0026). Neither sex nor probable depression was a significant predictor of mortality in this model.
CONCLUSION: These findings, together with those of others, suggest that hopelessness is a significant predictor of mortality in older and middle-aged adults of various ethnic backgrounds. Further research is needed to evaluate the mechanisms that underlie this phenomenon and the effects of treating hopelessness on the quality and duration of subjects’ lives.
Key Words: hopelessness • Mexican American • European American • mortality • cardiovascular disease • cancer.
Abbreviations: CI = confidence interval; EA = European American; GDS = Geriatric Depression Scale; HBA = home-based assessment; LOT = Life Orientation Test; MA = Mexican American; SAHS = San Antonio Heart Study; SALSA = San Antonio Longitudinal Study of Aging.
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
In the second study, Everson et al. (9) evaluated 2428 men from eastern Finland, aged 42 to 60 years at baseline, who participated in the Kuopio Ischemic Heart Disease Study. Subjects were followed for a mean of 6.0 years. Hopelessness at baseline was assessed by two questions, "I feel that it is impossible to reach the goals I would like to strive for" and "The future to me seems hopeless, and I can’t believe that things are changing for the better," each of which had five response options ranging from "absolutely agree" to "absolutely disagree." The authors combined the two items to construct a scale with a maximum score of 8; subjects with a score of 3 to 5 were considered moderately hopeless, and those who scored 
6, highly hopeless. Controlling for a variety of demographic and known risk factors by means of the Cox proportional hazards model, they found an increased risk of all-cause mortality for both men who were moderately hopeless (relative hazards = 1.82, 95% CI = 1.25–2.65) and those who were highly hopeless (relative hazards = 2.09, 95% CI = 1.30–3.34). They also found, after controlling for multiple risk factors, that moderately and highly hopeless subjects with a history of cardiovascular disease, diabetes, cancer, or respiratory disease had an increased risk of dying from both cardiovascular and noncardiovascular causes. They also found that mortality due to cancer was increased among moderately and highly hopeless men.
Several recent studies have used the Life Orientation Test (LOT) of Scheier and Carver (10) to assess the role of optimism and pessimism in health and mortality. In a group of 51 middle-aged men who had coronary artery bypass surgery, Scheier et al. (11) found that those who scored in the optimistic range on the LOT had better postsurgical outcomes. Schulz et al. (12), studying 238 female and male patients with advanced cancer, reported that pessimism scores on the LOT predicted mortality in those aged 30 to 59, but not in older subjects. Similarly, Robinson-Whelen et al. (13) noted that the LOT pessimism score predicted subsequent ill health in 113 predominantly female, middle-aged and older caregivers of persons with dementia as well as in 111 non–care-giving control subjects.
The results of these studies support the hypothesis that hopelessness and pessimism are associated with an increased risk of illness and mortality. Many unanswered questions remain, however. Of the two hopelessness studies, only that of Everson et al. (9) used questions specific to hopelessness, and it was limited to Finnish men aged 60 or younger at baseline. The LOT studies (11–13) also included primarily subjects of European origin. Thus, it is unclear whether hopelessness predicts mortality in older persons or individuals not of European origin. Studying these other groups could not only help in evaluating the generalizability of the hopelessness/health relationship but also might shed light on some of the mechanisms that underlie this phenomenon. The elderly would be very interesting to evaluate because the study of Schulz et al. (12) suggests that pessimism may not affect mortality in this age group. It is possible, as Schulz et al. suggest, that pessimism or hopelessness may be more normative in older persons. Studying members of various ethnic minority groups would also be of particular interest. Given the relative lower socioeconomic status of many minority group individuals, one might hypothesize that those who survive into old age would be a particularly hardy group and less susceptible to a health effect of hopelessness than similar members of the majority population.
To shed further light on the relationship between hopelessness and mortality, we evaluated data from a cohort of older, community-dwelling Mexican Americans (MAs) and European Americans (EAs). MAs are of special interest not only because they constitute the most rapidly growing segment of the elderly in the United States (14), but also because traditional MA culture is commonly viewed as having a more fatalistic (15, 16) and present-oriented (17) world view than EA culture. Although a fatalistic world view might well have a deleterious effect on health, present-time orientation could potentially mitigate some of these effects by making the future a lesser source of concern.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
To maximize sociocultural variation among MA participants in the SAHS, subjects were recruited from three types of neighborhood in the city of San Antonio: 1) low income, almost exclusively MA; 2) middle-income, ethnically balanced; and 3) high-income, predominantly EA. A random sample of households was selected in each type of neighborhood using the Polk directory, a city criss-cross directory. Only MAs were selected in the first type of neighborhood because so few EAs lived there. In the middle- and high-income neighborhoods, stratified random sampling was used to achieve enrollment of approximately equal numbers of the two ethnic groups. All 25- to 64-year-old men and nonpregnant women living in the selected households were considered eligible for the SAHS. For purposes of follow-up, only those individuals who completed a baseline clinic examination were considered part of the SAHS cohort.
From 1992 to 1996, efforts were made to contact all SAHS subjects who were currently or would shortly be between 65 and 79 years of age and to ask them to participate in SALSA. Of the 1247 subjects from the SAHS cohort who would have met these age criteria during the study period, 170 had died. Seventy-seven percent (833 of 1077 survivors) agreed to participate and completed the HBA as part of SALSA. A few subjects were interviewed slightly before their 65th birthday. Among the 23% of survivors (N = 244) who did not complete the HBA, 28 had moved out of the area, 26 could not be located, and 190 declined to participate. There were no significant differences between those who completed the HBA and those who did not in age, gender, perceived health (self-assessment of the quality of one’s health), or number of comorbid medical illnesses. However, completers were more educated (8.6 vs. 7.1 years of schooling, p 
.001) and, among Mexican Americans, were slightly more assimilated into American society.
Data collection in SALSA, consisting of the HBA plus a performance-based assessment given at the General Clinical Research Center of the University of Texas Health Science Center in San Antonio, was carried out between April 1992 and April 1996. All subjects were between 64 and 79 years old at the time of their HBA. All rating instruments, which were administered by an interviewer, were translated into Spanish using standard cross-cultural techniques and administered orally in either English or Spanish according to the participant’s stated language preference. Twenty-nine percent of the MAs were interviewed in Spanish.
Among the 833 persons who completed the HBA as part of SALSA, 795 subjects took the 30-item Geriatric Depression Scale (GDS) (21) and were included in this study. Those who responded "no" to the item "Are you hopeful about the future?" were classified as hopeless. The Spanish version of this question was "¿Tiene esperanzas acerca del futuro?," the literal translation of which is "Do you have hopes concerning the future?"
The data for this report are based on mortality statistics through August 1999, with a mean (SD) follow-up of 5.2 (1.5) years and a range of 0.2 to 7.4 years. To assess vital status, local newspaper obituaries were reviewed daily, vital statistics records of the San Antonio Metropolitan Health District were checked quarterly, and a search of the National Death Index was performed annually. Cause of death was determined by review of death certificates and was coded by a certified nosologist.
Data Analysis
A stepwise regression method was used to explore the relationship between hopelessness and all-cause mortality using Cox proportional hazards models. Four sets of data were entered into the model hierarchically. First, demographic variables (age, sex, ethnic background, years of formal education, and monthly household income) were entered. Second, significant demographic variables from step 1 (p .10) were entered along with potential health risk factors. These consisted of systolic and diastolic blood pressure, body mass index, number of comorbid medical illnesses (assessed by asking participants if they had ever been diagnosed by a doctor with any of seven common medical conditions), and perceived health. Third, significant variables from step 2 were entered along with lifestyle risk factors. These included current smoking, current drinking, probable alcoholism, and activity level (weekly energy expenditure based on self-estimated physical activity). Fourth, significant variables from step 3 were entered along with social protective or risk factors. These consisted of number of social contacts, social well-being (satisfaction with contacts), marital status, and whether the individual was living alone. Finally, hopelessness was added to the final model of significant variables in step 4 to determine whether it was a significant independent predictor of mortality. The final model before adding hopelessness included six variables: age, ethnic background, number of comorbid illnesses, perceived health, number of social contacts, and current smoking. All variables in this model were statistically significant (p < .05).
A separate stepwise regression was also conducted for the MA subjects.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
2 0.675, df = 1, p = .411). The prevalence of hopelessness among the four ethnic/gender groups ranged from 12.0% (22 of 184) for MA men to 9.9% (21 of 213) for EA women, 8.5% (21 of 248) for MA women, and 6.0% (9 of 150) for EA men (2 3.8, df = 3, p = .285). As shown in Table 1, the hopeful and hopeless participants varied significantly on a number of demographic and health variables but showed no difference on other important variables, such as blood pressure, body mass index, current drinking status, and probable alcoholism.
|
2 20.1, df = 1, p = .001). The mortality rate for the four gender/ethnic groups ranged from 21.7% for the MA men to 10.9% for the MA women, 9.3% for the EA men, and 8.0% for the EA women (2 20.5, df = 3, p = .001).
Among the subjects for whom death certificates were available (87 of 98), 25 died of cardiovascular disease and 25 of cancer. The mortality rates were identical for both diseases: 7.2% for the hopeless subjects and 2.8% for the hopeful (2 4.0, df = 1, p = .045).
Table 2 shows the final Cox proportional hazards model, in which hopelessness was a highly significant predictor of all-cause mortality.
|
There was insufficient power to conduct a separate analysis for the EAs because of the smaller number of deaths in this group.
Mortality among MAs did not differ significantly according to whether they were first-, second-, or third-generation: 19% (11 of 58) vs. 15% (40 of 271) vs. 17% (16 of 96), respectively (2 = 0.712, df = 2, p = .700). The mortality rate among the 29% of the MAs who were interviewed in Spanish was 20% (23 of 116), compared with 14% (44 of 316) for those interviewed in English (2 = 2.26, df = 1, p = .133).
One hundred thirty-nine subjects (17.5%) were classified as probably depressed based on a total GDS score of 11, the usual cutoff point for a major depressive syndrome (21). Their mortality rate (22.3%, 31 of 139) was significantly greater than that for nondepressed individuals (10.2%, 67 of 656) (2 15.5, df = 1, p = .001). Subjects with both hopelessness and probable depression had a significantly higher mortality rate than those with only one or none of these risk factors, as shown in Table 3. In a Cox model that controlled for age, sex, and household income, the risk of mortality increased significantly (p = .0001) as one went from category 1 (hopeful/not depressed) to category 2 (hopeful/depressed), category 3 (hopeless/not depressed), and finally category 4 (hopeless/depressed).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Similar to Anda et al., (8) we found a higher prevalence of hopelessness in a minority population, although the ethnic difference in a comparison with EAs was not significant. Consistent with the results of Fiscella and Franks (23) with African American subjects, we did not find that controlling for hopelessness significantly decreased the relative risk of mortality associated with MA ethnic background. Like Greene (24) but unlike Anda et al. (8), we did not find a higher prevalence of hopelessness in women. It is possible that the increased hopelessness in women observed by Anda et al. (8) may have been related to their assessment of this variable by means of a question that also included depressed mood.
Potential Limitations
We evaluated hopelessness with a single yes/no question, which prevented us from ascertaining whether severe as opposed to moderate hopelessness was associated with a greater increase in mortality, as suggested by Anda et al. (8) and Everson et al. (9). The hopeless subjects in our study, 9% of the total, may have been comparable in degree of hopelessness to Everson et al.’s (9) "severely hopeless" subjects, who constituted 11% of their population. Although the GDS has been reported to have high sensitivity and specificity for diagnosing a major depressive syndrome in a geriatric population (25), it is possible that the risk ratio for probable depression in our study might have been different had we diagnosed subjects by means of a psychiatric interview. Our findings with regard to hopelessness and mortality from cardiovascular disease and cancer need to be evaluated with caution in view of the fact that we assessed cause of death by review of death certificates, a method that may not always be reliable.
Relationships Among Hopelessness, Depression, and Mortality
Our finding that probable depression did not predict all-cause mortality after controlling for multiple risk factors differs from that of many authors (26–34) but is consistent with the findings of others (35–39). As Wulsin et al. (40) noted in their comprehensive review of the mortality of depression, the widely varying results in this literature may in part reflect many authors’ failure to control for key risk factors. Differences in the populations studied and divergent definitions of depression may also contribute to the disparate results.
It is of interest that Everson et al., (9, 41) Anda et al. (8), and we all found hopelessness to be a stronger predictor of mortality than depressive symptoms or probable depression. Everson et al. (41) noted that men with a score in the fourth quartile on the depression scale of the Minnesota Multiphasic Personality Inventory were at increased risk of cardiovascular mortality compared with those with scores in the first quartile, but this difference became only marginally significant after controlling for multiple risk factors. Similarly, in Anda et al.’s (8) data, a four-item depressed affect subscale of the General Well-Being Schedule (42) was of only borderline significance (relative risk = 1.5, 95% CI = 1.0–2.3) after adjusting for multiple risk factors. These results, together with our findings (see Table 3), suggest that those depressed individuals who are both depressed and hopeless may have the greatest risk of mortality.
In a recent study that found depressive symptoms to predict mortality in a geriatric population (34), item analysis showed that the effect was due to motivation-related items. These authors’ results may be consistent with our and others’ findings on hopelessness.
Directions for Future Research
Both our and Everson et al.’s (9) data suggest that hopelessness is associated with an increased risk of dying of both cardiovascular and noncardiovascular causes, including cancer, whereas Anda et al.’s (8) data support an increased risk of fatal ischemic heart disease. Further research in various populations to evaluate which causes of death are increased in hopeless individuals would be very useful. Such data would help shed light on the mechanisms by which hopelessness is related to mortality. One potential mechanism is suicide, which is clearly increased in psychiatrically ill persons with symptoms of hopelessness (43, 44). Suicide is much less common, however, in epidemiologic cohorts. Indeed, we had no cases of suicide among the 87 deaths in our sample for which causes of death were available. As Schulz et al. (12) suggest, hopeless or pessimistic individuals may, like those with depression (45, 46), be less likely to engage in health-promoting behaviors, such as eating well, exercising, following their doctor’s recommendations, and taking their medication as prescribed. They may also, like depressed persons (47), be more likely to engage in high-risk behaviors, such as substance-related behavioral disorders and unsafe sexual practices. Our data and that of Anda et al. (8) and Everson et al. (9) show, however, that hopelessness remains a significant predictor of mortality even after controlling for at least some key health-related behaviors, such as smoking and alcohol abuse. Similar to individuals suffering from depression (48–52), hopeless persons could also have biochemical and neurophysiologic abnormalities (such as abnormal platelet function, reduced heart rate variability, decreased immune function, decreased serotonergic function, and increased activity of the noradrenergic system and hypothalamic-pituitary-adrenal axis) that could help to explain their increased mortality. The report of Everson et al. (53) that hopeless men have greater progression of carotid atherosclerosis is a very promising finding in this area. Further imaging studies to evaluate changes in brain functioning associated with hopelessness, building on the work of Gottschalk et al. (54), could be very informative.
It would be useful, too, to explore further the relationships among hopelessness, optimism and pessimism, depression, and other proposed psychological risk factors for physical illness and mortality, such as vital exhaustion (55–57), hostility (58), and anxiety (59). Other useful areas for exploration include whether hopelessness predicts mortality in children, young adults, and persons over 80; the effect of hopelessness on the functional status and quality of life of physically ill persons; and the role of life experiences and genetic factors in the etiology of hopelessness. Assessing hopelessness by an instrument such as the Beck Hopelessness Scale (60), rather than by one or two questions, would likely facilitate future research.
More research into the causes of hopelessness is also needed. Individuals may experience hopelessness as part of a major depressive syndrome or as a relatively isolated symptom reflecting their current or past life situation. Economic hardship, social and cultural isolation, and loss of function due to poor health all may play a role in the etiology of hopelessness for some individuals, as might the loss of a spouse or a history of childhood abuse. It is possible that MAs, EAs, and other ethnic and racial groups might differ in the extent to which different etiologic factors play a role in the development of hopelessness. Tailoring the choice of treatment for hopelessness or the method of providing it to an individual’s cultural background might prove very useful.
Prospective studies to evaluate the effects of treating hopelessness on the quality and duration of subjects’ lives would be important. Both medication and cognitive-behavioral therapies, such as those of Beck et al. (61) and Seligman (3), would seem to be promising treatment modalities for hopeless individuals, though patients who suffer from hopelessness as part of severe melancholic depression might not be expected to respond well to cognitive-behavioral psychotherapies alone. It would be useful to explore whether depressed patients with and without hopelessness respond differentially to different therapeutic approaches. Both antidepressants and psychotherapy would be worth evaluating in hopeless individuals who have few other depressive symptoms.
Implications for Clinical Practice
The fact that a single item can provide relevant information about the risk of mortality suggests that "Are you hopeful about the future?" might be a useful screening question to include in the evaluation of older patients. Our data also lend further support to the time-honored advice that instilling hope is a key part of the clinician’s armamentarium.
|
CONCLUSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
Received for publication October 7, 1999.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONCONCLUSIONACKNOWLEDGMENTSREFERENCES |
|---|
This article has been cited by other articles:
 |
|
 |
|
  R. D. Lane Neural Substrates of Implicit and Explicit Emotional Processes: A Unifying Framework for Psychosomatic Medicine Psychosom Med, February 1, 2008; 70(2): 214 - 231. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 B. L. Egleston, D. O. Scharfstein, E. E. Freeman, and S. K. West Causal inference for non-mortality outcomes in the presence of death Biostat., July 1, 2007; 8(3): 526 - 545. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. J. Giltay, M. H. Kamphuis, S. Kalmijn, F. G. Zitman, and D. Kromhout Dispositional optimism and the risk of cardiovascular death: the zutphen elderly study. Arch Intern Med, February 27, 2006; 166(4): 431 - 436. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. E. Freeman, B. L. Egleston, S. K. West, K. Bandeen-Roche, and G. Rubin Visual Acuity Change and Mortality in Older Adults Invest. Ophthalmol. Vis. Sci., November 1, 2005; 46(11): 4040 - 4045. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Vale Psychosocial stress and cardiovascular diseases Postgrad. Med. J., July 1, 2005; 81(957): 429 - 435. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 J. L. Griffith and L. Gaby Brief Psychotherapy at the Bedside: Countering Demoralization From Medical Illness Psychosomatics, April 1, 2005; 46(2): 109 - 116. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 X. Zhang, S. L. Norris, E. W. Gregg, Y. J. Cheng, G. Beckles, and H. S. Kahn Depressive Symptoms and Mortality among Persons with and without Diabetes Am. J. Epidemiol., April 1, 2005; 161(7): 652 - 660. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G. I. Papakostas, T. Petersen, J. Pava, E. Masson, J. J. Worthington III, J. E. Alpert, M. Fava, and A. A. Nierenberg Hopelessness and Suicidal Ideation in Outpatients With Treatment-Resistant Depression: Prevalence and Impact on Treatment Outcome Focus, January 1, 2005; 3(1): 140 - 145. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. A. Everson-Rose, J. S. House, and R. P. Mero Depressive Symptoms and Mortality Risk in a National Sample: Confounding Effects of Health Status Psychosom Med, November 1, 2004; 66(6): 823 - 830. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. G. Schneider The Intersection of Mental and Physical Health in Older Mexican Americans Hispanic Journal of Behavioral Sciences, August 1, 2004; 26(3): 333 - 355. [Abstract] [PDF]
|
|
|
|
 |
|
 M. D. Sullivan Hope and Hopelessness at the End of Life Am J Geriatr Psychiatry, August 1, 2003; 11(4): 393 - 405. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. M. Schwarz, H. Schachinger, R. H. Adler, and S. M. Goetz Hopelessness Is Associated With Decreased Heart Rate Variability During Championship Chess Games Psychosom Med, July 1, 2003; 65(4): 658 - 661. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
