Psychosomatic Medicine Faster Service from Outside North America
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Steptoe, A.
Right arrow Articles by Marmot, M.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Steptoe, A.
Right arrow Articles by Marmot, M.
Related Collections
Right arrow Other Cardiovascular Medicine
Right arrow Social Support
Right arrow Quality of Life
Right arrow Stress and Coping
Psychosomatic Medicine 65:1029-1037 (2003)
© 2003 American Psychosomatic Society

ORIGINAL ARTICLES

Burden of Psychosocial Adversity and Vulnerability in Middle Age: Associations With Biobehavioral Risk Factors and Quality of Life

Andrew Steptoe, DPhil and Michael Marmot, PhD, FRCP

From the Department of Epidemiology and Public Health, University College London, UK.

Address for correspondence: Andrew Steptoe, Department of Epidemiology and Public Health, University College London, 1-19 Torrington Place, London WC1E 6BT, UK. Tel: (44) 20 7679 1804 Fax: (44) 20 7916 8542 E-mail: a.steptoe{at}ucl.ac.uk


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
OBJECTIVES: Numerous psychosocial factors are associated with disease risk. This study investigated the possibility that a combination of chronic stress from exposure to multiple sources and absence of protective psychosocial resources would be related to heightened emotional distress, health risk behavior, biological risk factors, and impaired quality of life, independently of socioeconomic position (SEP).

MATERIALS AND METHODS: Data were analyzed from 227 men and women aged 47 to 59 years from the Whitehall II epidemiological cohort. A psychosocial adversity and vulnerability index (PAVIX) was constructed from high scores on measures of job demands, neighborhood stress, and financial strain, low emotional support, limited social networks, low active coping, and low sense of control.

RESULTS: The measures making up the PAVIX were relatively independent of one another. Scores on the PAVIX were greater in lower SEP participants, and in single, separated, or divorced than married participants. The PAVIX was positively associated with psychological distress, depression, hopelessness, sleep problems, hostility, low self-esteem and loneliness, independently of age, sex, SEP, and marital status. There were no associations with health behaviors, but relationships were observed with glycohemoglobin, plasma fibrinogen, plasma viscosity, and body mass (women), that were again independent of covariates. Individuals with high PAVIX scores also reported impaired health-related quality of life.

DISCUSSION: The accumulated burden of life stress coupled with limited protective psychosocial resources is associated with adverse psychological, biological, and quality of life outcomes. This integrated approach to the investigation of psychosocial factors may prove valuable in understanding etiological processes.

Key Words: psychosocial factors, • stress, • socioeconomic position, • quality of life, • coronary heart disease.

Abbreviations: SEP = socioeconomic position;; BMI = body mass index;; CRP = C-reactive protein.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Research in the fields of psychosocial epidemiology and behavioral medicine has established that several psychosocial factors are associated with disease risk and impaired quality of life, including chronic stress exposure, poor social supports, and limited social networks (1). Psychological characteristics such as hostility and a low sense of control have also been related to health outcomes (2,3). These psychosocial factors are thought to influence physical health through behavioral pathways (unhealthy patterns of health behavior and lifestyle), and psychobiological pathways, with the activation of stress-related neuroendocrine, autonomic, and immunological responses (4). Research is commonly conducted on individual psychosocial variables, rather than on the overall burden of exposure. This analysis outlines an approach to the measurement of the cumulative burden of psychosocial adversity and vulnerability, and describes associations with socioeconomic status (SEP), biobehavioral risk factors, and quality of life.

Combining psychosocial measures relevant to chronic disease is not new, because different sources of stress have been aggregated to assess the influence of cumulative adversity in previous studies (5,6). For example, Evans and English (7) recently assessed an array of physical stressors such as noise and crowding, and interpersonal stressors like family conflict and community violence, in children from low- and middle-income families. The extension of this approach to include both cumulative adversity and unfavorable psychosocial resource profiles has been more limited, although there are instances in which measures of life stress have been combined with social supports or coping strategies to assess stress buffering effects (8,9). The evidence that vulnerability factors such as low social support act in a buffering capacity is equivocal (10), so we decided to create a single index incorporating both measures of exposure to adversity and the absence of protective social and psychological resources.

This study involved middle-age employed men and women, and this dictated the ways in which adversity and vulnerability factors were operationalized. Exposure to chronic adversity was assessed by combining markers of potential stress from three sources—high job demands, financial strain, and neighborhood stress. Each of these has individually been related to biological risk factors, impaired quality of life, and increased incidence of serious disease outcomes (11,12). We reasoned that the impact of these adversities would be accentuated if they are combined with a lack of protective social and psychological resources. Both emotional support and social networks are associated with favorable biological risk profiles and survival in prospective studies (13). Consequently, low emotional support and limited social networks constitute vulnerability factors. Evidence relating psychological characteristics with objective health outcomes is less conclusive. However, it is widely acknowledged that coping responses are critical to adjustment to life stress (5,14). Aspinwall and Taylor (15) have argued that proactive coping facilitates accommodation to the hazards of everyday life, and that an active, problem-solving approach to the management of stressors may be generally more adaptive than avoidant coping strategies. Another relevant psychological characteristic is fatalism, or strong beliefs in the play of chance. Low perceived control is a determinant of heightened stress responsivity and poor adaptation (16). We therefore reasoned that low preferences for active coping, and strong beliefs in chance, are psychological vulnerability factors.

The main objective of this research was to determine whether a combined index of the burden of adversity and vulnerability would be a useful analytic tool for the investigation of psychosocial factors and health. A psychosocial adversity and vulnerability index (PAVIX) was constructed by summing high ratings of job demands, financial strain, neighborhood stress, and chance beliefs, and low scores on measures of emotional support, social networks, and active coping responses. We examined the composition of the index to determine whether particular measures or combinations of measures were selectively represented among participants with high scores, or whether each measure was equally represented. We predicted that scores on the PAVIX would be greater in people of lower SEP, since a number of the individual constituents have previously been associated with SEP (17,18).

The relationship between the PAVIX and the following four classes of outcome was investigated: (1) psychological well-being. It was predicted that a higher burden of psychosocial adversity and vulnerability would be positively associated with heightened perceived stress, distress as measured with standardized psychiatric measures, sleep problems, and with psychological characteristics that might in turn compromise health, including low self esteem, greater loneliness and hostility; (2) health behavior. If behavioral factors are responsible in part for the impact of negative life experiences on health, then PAVIX scores should be positively related to smoking, alcohol consumption and lack of physical activity; (3) biological risk factors. We hypothesized that scores on the PAVIX would be positively associated with body mass, waist-to-hip ratio, glycohemoglobin (an index of impaired glucose homeostasis), markers of vascular inflammation (fibrinogen and C-reactive protein), and rheological measures implicated in hemostasis (plasma viscosity and blood viscosity); and (4) health-related quality of life. It was hypothesized that the PAVIX be positively associated with impaired quality of life in multiple domains.

Low SEP is associated with several of these outcome variables (19–21). The relationship between vulnerability and adversity and the four sets of outcomes were therefore adjusted for SEP as defined by occupation, so as to evaluate independent associations.


   MATERIALS AND METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
Participants
Data were collected from 129 men and 111 women, drawn from the Whitehall II study. Details of this study have been published previously (22,23). Volunteers were recruited on the following criteria: white European origin, aged 45 to 59 years, not planning to retire for at least 3 years, no history of coronary heart disease or cancer, and no previous diagnosis or treatment for hypertension or endocrine-related disorder. Participants were drawn from higher (administrative and professional), intermediate (senior executive officers), and lower (executive officers, clerical, office support) employment grades. Grade of employment in this population is strongly associated with other markers of SEP such as educational attainment and income. The response rate among those invited to take part was 55% and was greater in higher than lower status participants.

Measures of Psychosocial Adversity and Vulnerability
The psychosocial adversity and vulnerability index (PAVIX) consisted of three measures of chronic stress (job demands, financial strain, and neighborhood stress), and four measures of psychosocial resources (emotional social support, social network size, sense of control, and use of active coping).

Job demands were assessed with a measure developed from the demand/control model, as described by Bosma et al. (24). Four items (eg, "Do you have to work very intensively?") were administered, each of which was rated on a 4-point scale ranging from 0 (often) to 3 (never/almost never). Scores were converted to a scale from 0 to 100, in which 100 indicates maximum demands. The Cronbach {alpha} for the scale in this study was 0.70.

Financial strain was assessed with an adaptation of a widely used economic strain measure (25). This measure assesses difficulty paying one’s bills, being able to replace items such as furniture or a car when needed, and being able to provide for one’s family in terms of food, clothing, and medical care. Eight items were presented, with response options ranging from 1 = no difficulty to 3 = very great difficulty (Cronbach {alpha}=0.86).

Neighborhood stress was measured with the neighborhood problems scale, an inventory assessing exposure to community-wide stressors (26). Participants were asked to indicate the degree to which 10 neighborhood conditions (eg, litter in the streets, noise from traffic) constituted a problem, with ratings ranging from 1 = not a problem to 3 = serious problem. The internal consistency of the scale in the present sample was 0.74.

Emotional social support and social networks were both measured using the Close Person’s Questionnaire (27). For emotional support, respondents were asked to designate the single individual to whom they had felt closest over the previous 12 months, and then to rate a series of statements concerning the support provided by this relationship on 4-point scales from not at all to a great deal. Eight items contributed to the emotional support scale (eg, "How much in the past 12 months did this person make you feel good about yourself?" Cronbach {alpha}=0.86), with higher scores indicating greater support. Social networks were measured with a series of questions concerning contact with family, friends, and relatives. Respondents were given points toward a social isolation score if they lived alone, if they had no relatives outside their household, never visited or were never visited by relatives or friends, or had no relatives or friends who they saw at least once per month. These scores could range from 0 (no social isolation) to 3 (maximum isolation); 46.6% of participants had social isolation scores of 1 to 3, so they were designed as the high social isolation group.

Sense of control was assessed using the chance locus of control scale developed by Levenson (28). This consisted of eight items (eg, "To a great extent my life is controlled by accidental happenings") rated on 6-point scales ranging from 0 (strongly disagree) to 5 (strongly agree). The alpha for the scale was 0.76. Ratings could range from 0 to 40, with higher scores indicating greater belief in chance control.

Active coping was measured with the four-item active coping scale from the COPE inventory of coping strategies (29). Each item (eg, "I concentrate my efforts on doing something about it") was rated on a 4-point scale ranging from 1 (never) to 4 (always). Average scores were calculated (range: 1–4), with higher values indicating greater use of active coping (Cronbach {alpha} = 0.71).

Additional Psychosocial Measures
Subjective SEP was assessed with the "ladder" measure described by Adler et al. (30). Participants were shown a drawing of a ladder with 10 rungs, representing where people stand in society. They were told that at the top of the ladder are the people who are best off—those who have the most money, most education, and best jobs. At the bottom are the people who are the worst off, have the least money, least education, and the worst jobs or no job. They were asked to place themselves on the rung on which they felt that they stood.

Perceived stress was assessed with the 10 item Perceived Stress scale (31). Respondents were asked how they had thought or felt over the past month, and each item (eg, "How often have you felt that you were on top of things?") was rated on a 5-point scale ranging from 0 = never to 4 = very often. Ratings were summed so that scores could range form 0 to 40 (Cronbach {alpha} = 0.87).

Psychological distress was assessed with the General Health Questionnaire 30 (GHQ 30), a well-established screening questionnaire for psychiatric disorder suitable for population studies (32). The Cronbach {alpha} in the present sample was 0.93. The standard scoring system was used, so scores could range 0 to 30, with higher scores indicating greater distress. To assess specific associations with depressive symptoms, we used the method described by Stansfeld et al (33) of selecting four items from the GHQ 30 that are also present in the depression subscale of the GHQ 28, scoring them on a Likert scale from 1 to 4. The Cronbach {alpha} for this GHQ depression scale was 0.89.

Hopelessness was assessed with the two-item scale used in the Kuopio study (34). Each item (eg, "The future to me seems to be hopeless, and I cannot believe that things are changing for the better"), each of which was scored on a 5-point scale from absolutely disagree to absolutely agree. Ratings were averaged to generate a hopelessness score ranging from 1 to 5.

Sleep problems were assessed with the sleep problems scale described by Jenkins et al. (35). This consists of four items, assessing how often in the past month the individual has woken up several times in the night, had trouble staying asleep etc. There were six response options, ranging from 0 = not at all to 5 = 22 to 31 days. The total was computed, so scores could range from 0 to 20. The measure has been used extensively in previous research on healthy populations and patient groups. The Cronbach {alpha} in the present sample was 0.73.

Self-esteem was measured with the 10 item Rosenberg scale (36). Items were rated on a 4-point scale, and scores could range from 0 to 30 with higher ratings indicating greater self-esteem. The Cronbach {alpha} was 0.89.

Cynical hostility was assessed with a 22-item version of the Cook-Medley scale (37). The scale primarily taps cognitive components of hostility including negative beliefs about and attitudes toward others, including cynicism and hostile attributions. Scores could range from 0 to 22, with higher ratings indicating greater hostility. The Cronbach {alpha} for this scale was 0.83.

Loneliness was assessed with the revised UCLA Loneliness scale (38). This consists of 20 items (eg, "I feel that no one knows me really well"), and participants were asked to rate how often they felt that way on a scale from 1 = never to 4 = often. Ratings were summed to produce a total score ranging from 20 (low) to 80 (high). The internal consistency in this study was 0.91.

Biobehavioral Measures and Quality of Life
The measures of health behavior were based on those used in earlier reports from the Whitehall II study (17). Participants were asked whether they were current smokers and a dichotomous score was used to indicate smoking status. Alcohol intake was measured by the number of servings of spirits, wine, and beer consumed in the past week. Physical activity was assessed with two measures. First, participants were asked how many times per week they engaged in vigorous activity of sufficient intensity to make them breathless. Second, total walking time was computed from separate estimates of walking on weekdays and the weekends.

Physical measurements included height, weight, and waist and hip circumference, assessed with standard methods (23). Waist-to-hip ratio and body mass index (BMI) were computed. Information concerning hormone replacement treatment was collected from women.

Measures from plasma were taken from blood drawn after 30 minutes of rest. Venous blood samples were collected in citrated (0.109 mol/L; 9:1 v:v) and K2-EDTA (1.5 mg/ml) tubes, centrifuged at room temperature, and the supernatant plasma was snap frozen at -70°C within 1 hour. Plasma glycohemoglobin was assayed using boronate affinity chromatography, a combination of boronate affinity, and liquid chromatography at a laboratory that conforms with the Diabetes Control and Complications Trial (DCCT) standardized method (39). Clottable fibrinogen was measured from citrated frozen samples by an automated Clauss assay in a MDA-180 coagulometer (Organon Teknika, Cambridge, UK) using the manufacturer’s reagents and the International Fibrinogen Standard (40). Hematocrit was assessed immediately after each blood sample was drawn using a microhematocrit centrifuge and reader (Hawksley Gelman, Lancing, Sussex, UK).

C-reactive protein was detected using a sensitive, two-site ELISA with antibodies from Dako diagnostics (Ely, Cambs, UK), as detailed previously (41). The interassay and intraassay coefficients of variation (CV) were less than 10%. Plasma viscosity was measured in the K2-EDTA samples using a semi-automated capillary viscometer (Coulter) at 37°C. Blood viscosity was calculated from hematocrit and plasma viscosity as described by Lowe et al. (42).

Quality of life was assessed using the short form 36 Health Survey (SF-36). Normative data for the general UK population are available (43), and the measure has also been applied to the complete Whitehall II cohort (20). Seven dimensions of functioning were assessed: physical functioning or activities of daily living, role impairment owing to physical problems, role impairment caused by emotional problems, social functioning, mental health, general perceptions of health, and vitality or energy. Scores were coded and transformed to a scale in which 0 = worst possible health and 100 = best possible health.

Procedure
Psychosocial measures were completed before the laboratory session, and took approximately 30 minutes to complete. Participants attended individually either in the morning or afternoon for the laboratory session. They were instructed not to have drunk any caffeinated beverages or to have smoked for at least 2 hours before the testing, and not to have consumed alcohol or exercised on the evening before or the day of testing. Those who had recently or currently had from any minor illness, or had used over-the-counter antiinflammatory or antihistamine medication over the previous 2 days, were rescheduled. After anthropometric measurement, a venous cannula was inserted and the participant rested quietly for 30 minutes. A blood sample then was drawn for analysis of inflammatory, hemostatic, and rheological markers.

Statistical Analysis
Complete data for all seven components of the PAVIX were available for 227 participants, so analyses were restricted to these individuals. They did not differ from participants with missing data in age, sex, or grade of employment. The PAVIX was constructed by dividing scores on each of the seven constituent measures by median split into high and low categories, except for social isolation which was categorized as detailed. The division points for the scales were as follows: job demands (66–75), financial strain (8–9), neighborhood stress (13–14), emotional support (23–24), chance control (15–16), and active coping (2.5–2.75). The negative categories were then summed; one unit was added to the PAVIX for high job demands, high financial strain, high neighborhood stress, social isolation, low emotional support, high chance control beliefs, and low use of active coping methods. Each individual had a single score between 0 and 7 on the PAVIX, with higher scores reflecting greater burden of psychosocial adversity and lack of psychosocial resources.

Associations between the burden of psychosocial adversity, objectively defined SEP, and sex, were assessed by comparing PAVIX scores for participants in the higher, intermediate, and lower grade of employment groups, and by comparing men and women using analysis of variance. Associations with subjective SEP were assessed by correlating PAVIX scores with ladder ratings, and by comparing PAVIX scores in individuals in the three tertiles of ladder ratings using analysis of variance. For the remaining analyses, participants were divided on the basis of PAVIX scores into low (0–2), medium (3, 4), and high (5–7) categories. These three categories were then compared on psychosocial, biobehavioral, anthropometric, and quality of life measures using analysis of variance for linear trends and {chi}2 as appropriate. Sex was included as a factor in all analyses, so as to assess possible interactions between sex and PAVIX category. The pattern of statistical findings was similar when the PAVIX was modeled as a continuous variable in regression analyses. The categorical results are presented here as being more readily comprehensible. Plasma viscosity results were skewed, log-transformed before analysis, and geometric means are presented. Factors known to be associated with biological risk markers were included as covariates, as detailed in the Results section.

The possibility that associations between PAVIX scores and outcomes could be caused by the dominant influence of one particular component was tested in two ways. First, we assessed the extent to which high scores on each component contributed to the total PAVIX measure. Second, when associations between the PAVIX and any of the principle dependent variables was significant, analyses were repeated with each of the component measures removed in turn from the composite PAVIX. If the removal of any one of the seven components made a difference to the pattern of results, this would indicate that this component was primarily responsible for associations with the outcome in question. Data are presented as means ± standard error of the mean (SEM).


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 
The 123 men and 104 women in these analyses had an average age of 52.3 (range: 47–59 years). Eighty-eight (38.8%) were from the higher, 74 (32.6%) from the intermediate, and 65 (28.6%) from the lower grades of employment. A minority (35.2%) had university degree level education. There were 21 (9.3%) current smokers. Most of the participants (167 or 73.6%) were married or in stable relationships, but 40 (17.6%) were single, and 20 (8.8%) were divorced, separated, or widowed.

Psychosocial Adversity and Vulnerability Index
Mean scores on the PAVIX averaged 3.2 and were normally distributed, including the full possible range from 0 to 7. Most (115 or 50.7%) participants reported high scores on three or four of the measures of adversity and vulnerability, so they were classified into the medium PAVIX category. Thirty-six (15.9%) had scores of 5 to 7, and 76 (33.5%) had fewer than three high scores, so they were allocated to the low PAVIX category. PAVIX scores did not vary with age or sex.

The averages of each component measure for participants in the low, medium, and high PAVIX categories are shown in Table 1. As expected, there were significant linear effects for all seven measures [F (2, 221) = 7.23 to 33.4, all p< 0.001]. Participants in the high PAVIX category reported greater job demands, financial strain, and neighborhood stress, less emotional support, greater social isolation, stronger beliefs in chance, and less use of active coping than respondents in the lower PAVIX categories. All seven components were equally represented at each level of the total PAVIX. For example, the proportion of high PAVIX (5–7) respondents who scored above the median for each component measure was as follows: high job demands 25.6%, high financial strain 24.2%, high neighborhood stress 27.9%, low emotional support 27.7%, high social isolation 28.4%, strong beliefs in chance 27.8%, and little use of active coping 28.7%. The proportion of low PAVIX (0–2) participants who scored below the median for the component measures was: low job demands 24.4%, low financial strain 18.5%, low neighborhood stress 18.0%, high emotional support 16.8%, low social isolation 14.7%, weak beliefs in chance 15.7%, and extensive use of active coping 9.9%. The seven components were relatively independent of one another; correlations ranged from -0.21 to +0.20, with the highest negative association being between financial strain and active coping, and the highest positive association between neighborhood stress and chance beliefs. The internal consistency of the PAVIX was low ({alpha} = 0.12), reflecting the independence of the components.


View this table:
[in this window]
[in a new window]
  		TABLE 1. Components of the Psychosocial Adversity and Vulnerability Index
 
Socioeconomic Position and Demographic Correlates
There was a strong association between objective SEP (grade of employment) and PAVIX scores. The mean score on the index was 2.89 ± 0.15 in the higher, 3.33 ± 0.16 in the intermediate, and 3.56 ± 0.18 in the lower grade of employment groups [F (1, 220) = 4.13, p= 0.006]; 40.9% of the higher SEP participants were in the lowest PAVIX category, compared with 26.2% of the lower SEP participants. By contrast, only 10.2% of the higher SEP group were in the high PAVIX category, compared with 20% of the lower SEP participants.

Scores on the ladder measure of subjective SEP averaged 6.61, and ranged from 1 to 9.5. Ladder SEP ratings were negatively correlated with PAVIX scores (r= -0.25, p< 0.001). When ladder SEP ratings were divided into tertiles, the mean PAVIX score was 3.57 ± 0.18, 3.27 ± 0.17, and 2.93 ± 0.15 across the low, medium, and high subjective SEP groups [F (1, 220) = 3.68, p= 0.008]. Thus the burden of psychosocial adversity and vulnerability was associated with both objective and subjective indices of socioeconomic position.

PAVIX scores differed with marital status [F (2, 227) = 11.7, p< 0.001]. Mean scores were lower for married participants (2.93) compared with those who were single (4.05) or divorced/separated/widowed (3.65); 32.5% of the single and 20.6% of the divorced/separated/widowed were in the high PAVIX category, compared with 11.4% of the married participants ({chi}2 = 5.80, p= 0.016). These patterns did not differ in men and women.

Psychological Factors
The associations between scores on the adversity and vulnerability index and psychological factors are summarized in Table 2. Age, sex, grade of employment, and marital status were included as covariates. The three PAVIX categories differed on all measures of psychological distress, including the perceived stress scale [F (2, 220) = 33.0, p< 0.001], the GHQ 30 [F (2, 220) = 6.09, p= 0.003], the GHQ depression scale [F (2, 220) = 6.60, p< 0.001], and the hopelessness scale [F (2, 219) = 4.45, p= 0.003]. For all measures, psychological distress was greatest in the highest PAVIX category, and least in participants with low PAVIX scores. These effects remained significant in analyses of the PAVIX in which each of the components was omitted in turn from the composite measure. High levels of adversity and vulnerability were also associated with low self-esteem, greater hostility, and greater loneliness [F (2, 220) = 8.98, 5.29, and 16.9, respectively, p< 0.001]. It therefore appears that the cumulative burden of adversity and vulnerability was associated with psychological distress and a profile of unfavorable psychological characteristics independently of age, sex, socioeconomic status, and marital status. There were no interactions between PAVIX category and sex in these analyses.


View this table:
[in this window]
[in a new window]
  		TABLE 2. Psychological Correlates of the Burden of Adversity and Vulnerability
 
There was a marked association between the PAVIX and scores on the sleep problems questionnaire [F (2, 219) = 7.88; p< 0.001]. Sleep problem scores were greater in participants reporting high adversity and vulnerability, independently of covariates. The association was unaltered when each component of the PAVIX was omitted in turn from the composite measure.

Biobehavioral Factors
The proportions of participants with low, medium, and high PAVIX scores who smoked were 7.9%, 11.4%, and 5.6%, respectively (p= 0.92), indicating that the index was not associated with cigarette smoking. Participants had drunk an average of 9.60 servings of alcohol over the past week (range: 0–55), but this was not related to PAVIX scores; the mean number of servings adjusted for age and grade of employment was 9.64, 9.67, and 8.81 for low, medium, and high PAVIX categories respectively [F (2, 219) = 0.11, p= 0.68). The number of minutes walking per week was 346.9 (range 0–2520 min), and averaged 354.7, 354.2, and 268.1 in the low, medium, and high PAVIX categories [F (2, 217) = 1.24, p= 0.17]. The number of episodes of vigorous physical activity sufficient to get out of breath over the past week averaged 1.5 ± 0.2, and was also not related to PAVIX scores.

The analyses of BMI and waist circumference both showed interactions between PAVIX category and sex [F (2, 219) = 3.78 and 4.11, respectively, p< 0.025]. However, the differences in waist circumference were no longer significant after body mass index was taken into account, and there were no differences in waist-hip ratio. In the case of BMI, an association with PAVIX category was present in women but not men. In women, BMI (adjusted for age and grade of employment) averaged 24.1, 25.4, and 27.0 in the low, medium, and high PAVIX categories [F (2, 99) = 3.23, p= 0.013]. The association was not significant in men (mean BMI 26.2, 25.6, and 24.9 for the three categories).

There was a significant linear difference across PAVIX categories in glycohemoglobin [F (2, 216) = 2.52, p= 0.037]. Glycohemoglobin levels were positively associated with the adversity and vulnerability index, independently of age, sex, grade of employment, BMI, and smoking status (Table 3). A similar positive relationship was recorded for plasma fibrinogen [F (2, 214) = 7.75, p< 0.001], with an average difference of 0.40 g/L (95% CI: 0.20–0.61) between lowest and highest PAVIX categories, after adjustment for covariates. Both these associations remained significant when each component of the PAVIX was omitted in turn from the composite score.


View this table:
[in this window]
[in a new window]
  		TABLE 3. Biological Risk Factors and the Burden of Adversity and Vulnerability
 
Three individuals were excluded from the CRP analysis because they had very high levels suggestive of acute infection. There was no association between the PAVIX and CRP concentration [F (2, 201) = 0.09, p= 0.92]. Hematocrit was also unrelated to the PAVIX. However, plasma viscosity was positively associated with PAVIX category after adjustment for age, sex, grade of employment, BMI, and smoking [F (2, 214) = 3.34, p= 0.022]. The relationship for blood viscosity was similar, but did not reach statistical significance [F (2, 213) = 1.75, p= 0.065]. There were no interactions between PAVIX and sex in these analyses. When analyses of plasma viscosity were repeated with each component of the PAVIX omitted in turn from the composite measure, the association became nonsignificant in all cases except one; the composite PAVIX that excluded neighborhood stress remained associated with plasma viscosity (p= 0.034), while none of the others were. Thus in the case of plasma viscosity, neighborhood stress was not important compared with other components of the PAVIX.

Health-Related Quality of Life
The results for the seven scales of the SF 36 included in this study are detailed in Table 4. There was no association between the PAVIX and role impairment because of physical problems. But for the other six scales, there were significant linear effects across the PAVIX categories, with lower quality-of-life ratings in those experiencing greater psychosocial burdens [F (2, 220) = 3.92–15.5]. These effects were independent of age, sex, grade of employment, and marital status. The associations between the six SF 36 scales and the PAVIX remained significant when each component of the PAVIX was removed in turn from the composite score.


View this table:
[in this window]
[in a new window]
  		TABLE 4. Quality of Life and the Burden of Adversity and Vulnerability
 

   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
ACKNOWLEDGMENTS
 REFERENCES
 
This study was an attempt to take a more integrated approach to the investigation of the impact of psychosocial factors on biological risk and well-being than is typical in the literature. The combined index included measures of chronic adversity, social resources, and psychological characteristics. It can be argued that other factors could have been represented in the index. Our choice was orientated toward the sample of middle-age men and women in predominantly white-collar occupations, and other factors would certainly be more relevant at other stages of life. The three measures of adversity were selected to reflect different aspects of chronic life stress for the spheres of work, living conditions, and personal economic situation. Chronic stress stemming from marital and family conflict might have been incorporated (had data been available), since these factors have an impact on health and biological risk factors (44,45). Low job control rather than high demands might have been used as an indicator of strain in the work domain (24). We decided that a general measure of sense of control was preferable, because it would be relevant to spheres of life outside as well as inside the work domain, and might therefore reflect more fundamental individual differences. Both emotional support and social networks were represented, because they have somewhat distinct influences on health (13).

It is important to note that the index of psychological adversity and vulnerability had low interitem correlations and low internal consistency, indicating that the components were relatively independent. We did not, for example, find that individuals reporting high financial strain or neighborhood stress were particularly socially isolated or lacking in emotional support. Each of the items was equally represented in the high, medium, and low categories of the combined index. We also tested whether any single component of the PAVIX had a dominant influence on associations with psychological distress, impaired quality of life and biological risk, by performing analyses with each component of the composite index removed in turn.

The relationship between the PAVIX and SEP was as hypothesized. People with less privileged positions in society experience greater chronic stress from diverse sources, and have fewer protective psychological and social resources. The association was apparent using both objective and subjective indices of SEP. The overall pattern was expected, because lower SEP has previously been associated with a greater financial strain, greater neighborhood stress, social isolation, low emotional support, and low sense of control (17,26,46). Although job demands are generally rated as greater in low status work, in the Whitehall II cohort it is the higher grade civil servants who report greater job demands (17). In the light of these associations between the burden of adversity and vulnerability and SEP, all other analyses controlled for objective socioeconomic position, to determine whether the PAVIX was independently related to outcome variables.

Interestingly, the burden of adversity and vulnerability was similar in men and women in this cohort. Age was also unrelated to the PAVIX, but the narrow age range (47–59 years) may be an explanation. We had anticipated that divorced, separated, or widowed participants might report greater vulnerability and adversity, because these life experiences are known to lead to strain and loss of resources (45). But single participants also had higher PAVIX scores on the psychosocial adversity and vulnerability index. The explanation is not clear. Single people lack the possibility of social support through marriage, and difficulty in establishing a life partnership might increase perceived adversity, but a selection effect cannot be ruled out.

The hypotheses relating the burden of adversity and vulnerability with psychological stress and other psychological factors were generally confirmed. Individuals scoring high on the PAVIX had greater scores on the GHQ 30, the depression subscale of the GHQ, and Kuopio hopelessness scale (Table 2). It has recently been shown in the Whitehall II cohort that the GHQ 30 predicts increased 5-year incidence of self-reported coronary heart disease and new electrocardiographic abnormalities, independently of age, grade of employment, and health behaviors (48), while the hopelessness scale predicts the progression of subclinical atherosclerosis (34). The effects observed here may therefore be significant for cardiovascular as well as mental health outcomes. There were no interactions between gender and the PAVIX in these analyses, so we did not detect any tendency for men and women to react differently to chronic adversity.

Associations between the burden of adversity and vulnerability and psychological factors such as low self-esteem, hostility, and loneliness are interesting, because these factors have in turn been linked to disturbances in neuroendocrine and autonomic function, and to health outcomes (2,49). Because this study was cross-sectional, the causal pathways are uncertain. It is possible that chronic stress coupled with ineffective coping responses and social support deficits led to distress, lowered self-esteem, loneliness, and a cynical outlook on life. Equally plausibly, however, characteristics such as hostility and low self-esteem may increase exposure to sources of chronic stress and promote social isolation.

This study found no relationship between the burden of psychosocial adversity and vulnerability and health behaviors relevant to chronic disease risk including cigarette smoking, two measures of alcohol consumption and three measures of regular physical activity. The inference is that in this sample, associations between psychosocial factors and disease risk are unlikely to be mediated through behavioral pathways. The number of cigarette smokers was small, and this reduced the likelihood of observing associations. Variations in alcohol consumption and physical activity were large, so range restriction is unlikely to be the explanation for lack of effects. However, the measures of health behavior were not comprehensive, and it is possible that more comprehensive investigation would uncover associations. We also observed very limited associations between the PAVIX and anthropometric variables. The index was positively associated with body mass in women but not men, and there were no differences in waist-to-hip ratio after body mass and SEP had been taken into account. These results are not in agreement with the hypothesis that chronic stress causes abdominal obesity (50). The range of waist-to-hip ratio and body mass index was somewhat limited, and this will have reduced the possibility of observing associations.

By contrast, higher burden of adversity and vulnerability was associated with greater glycohemoglobin concentration (Table 3). Glycohemoglobin is an indicator of impaired glucose tolerance in nondiabetic populations, and has been shown to predict all cause and cardiovascular mortality independently of other risk factors (51). The observation of a positive association between glycohemoglobin and psychosocial adversity and vulnerability is consistent with the evidence relating other indices of glucose metabolism with locus of control and social support (52, 53). It should be noted that none of the participants in this study had levels of glycohemoglobin indicative of type 2 diabetes.

We hypothesized that the burden of psychosocial adversity and vulnerability would be positively associated with concentration of the acute phase reactant CRP, a marker of the inflammatory processes underlying atherosclerosis that has been shown in prospective studies to be a robust predictor of future coronary heart disease (54). No such association was observed. Similarly, we did not uncover consistent relationships between the PAVIX and inflammatory cytokines such as IL-6 (data not shown). By contrast, fibrinogen concentration was positively associated with the PAVIX. Fibrinogen is thought to heighten cardiac risk through increasing plasma and whole blood viscosity, infiltration of the arterial wall, stimulation of atherogenic cell proliferation, and promotion of platelet aggregation (54). The positive association between the PAVIX and plasma viscosity suggests that rheological mechanisms may be significant, because they act both on atherogenesis and thrombosis (55). Importantly, these associations were independent of SEP and factors such as smoking and body mass index. Although causal conclusions cannot be drawn, the findings are consistent with the notion that the burden of psychosocial adversity and vulnerability increases cardiovascular risk through disturbances in glucose metabolism and impairment of blood flow due to red cell aggregation and increased viscosity at low sheer rates.

In the final set of analyses, we demonstrated that the PAVIX was positively associated with low ratings on several dimensions of quality of life. These relationships persisted after adjustment for SEP which is also associated with impaired quality of life (20). The findings suggest that the experience of high levels of chronic stress coupled with limited psychosocial resources may have generalized effects on health and well-being, some of which may not be identified by the specific markers of psychological distress or biological risk measured in this investigation.

A major concern in these analyses was whether associations between the index of adversity and vulnerability and outcome variables were caused by the dominant influence of one particular component. For example, life stress and social relationships have been related to psychological well-being and biological risk factors in previous studies, so might be responsible for the effects observed here. However, we found that in the analyses of psychological well-being, sleep problems, glycohemoglobin, fibrinogen, and quality of life scales, no single component was necessary for associations to be observed. When each of the seven constituents was removed in turn from the composite PAVIX, statistical effects were unchanged. The one exception to this pattern was plasma viscosity, where it emerged that all components bar one (neighborhood stress) were necessary for significant associations to emerge. These subsidiary analyses therefore strengthen the argument that combining measures of adversity and vulnerability provides incremental information about psychosocial factors that is not supplied by analyses of separate components.

There are important limitations to this analysis. The study was performed with middle-age employed men and women of white European origin without known serious health problem or disability, so results may not generalize to other groups. The items included in the PAVIX were selected pragmatically from the range of variables collected in this study, and other markers might prove more sensitive in this or other samples. We decided to sum the factors rather than generate a measure of the balance between adversities and psychosocial vulnerabilities. However, it is plausible that a balance measure would provide further insight into the associations of psychosocial factors with risk indicators. The relatively homogenous nature of the population studied led to restricted variability in some outcome measures, and stronger associations might emerge in a broader population. Nonetheless, the results show a rather consistent pattern of negative psychological, biological, and quality of life outcomes in people experiencing a high cumulative burden of chronic adversity and psychosocial vulnerability. This approach may prove useful in future research through providing a more integrated perspective on the global experience of strain than is achieved by the measurement of single psychosocial factors.


   ACKNOWLEDGMENTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
REFERENCES
 
This research was supported by the Medical Research Council, UK. We are grateful to Sabine Kunz-Ebrecht, Pamela J. Feldman, Gonneke Willemsen, Natalie Owen, and Bev Murray for their assistance in data collection. The CRP analyses were performed by Vidya Mohamed-Ali (Department of Medicine, University College London), and the fibrinogen and plasma viscosity assays by Gordon D. O. Lowe and Ann Rumley (Department of Medicine, University of Glasgow).

Received for publication November 5, 2002.


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS AND METHODS
 RESULTS
 DISCUSSION
 ACKNOWLEDGMENTS
 REFERENCES
 

  1. Institute of Medicine. Health and Behavior. The interplay of biological, behavioral, and societal influences. Washington DC: National Academy of Sciences; 2001.
  2. Miller TQ, Smith TW, Turner CW, et al. A meta-analytic review of research on hostility and physical health. Psychol Bull 1996; 119: 322–348.[CrossRef][Medline]
  3. Bosma H, Schrijvers C, Mackenbach JP. Socioeconomic inequalities in mortality and importance of perceived control: cohort study. BMJ 1999; 319: 1469–1470.[Free Full Text]
  4. Steptoe A. Psychophysiological bases of disease. In: Johnston M, Johnston D, ed. Comprehensive Clinical Psychology Volume 8: Health Psychology. New York: Elsevier Science; 1998.
  5. Lamont D, Parker L, White M, et al. Risk of cardiovascular disease measured by carotid intima-media thickness at age 49–51: lifecourse study. BMJ 2000; 320: 273–278.[Abstract/Free Full Text]
  6. Kendler KS, Gardner CO, Prescott CA. Toward a comprehensive developmental model for major depression in women. Am J Psychiatry 2002; 159: 1133–1145.[Abstract/Free Full Text]
  7. Evans GW, English K. The environment of poverty: multiple stressor exposure, psychophysiological stress, and socioemotional adjustment. Child Dev 2002; 73: 1238–1248.[CrossRef][Medline]
  8. Rosengren A, Orth-Gomér K, Wedel H, et al. Stressful life events, social support, and mortality in men born in 1933. BMJ 1993; 307: 1102–1105.
  9. Turner Cobb JM, Steptoe A. Psychosocial stress and susceptibility to upper respiratory tract illness in an adult population sample. Psychosom Med 1996; 58: 404–412.[Abstract/Free Full Text]
  10. Wade TD, Kendler KS. Absence of interactions between social support and stressful life events in the prediction of major depression and depressive symptomatology in women. Psychol Med 2000; 30: 965–974.[CrossRef][Medline]
  11. Evans GW, Kantrowitz E. Socioeconomic status and health: the potential role of environmental risk exposure. Annu Rev Public Health 2002; 23: 303–331.[CrossRef][Medline]
  12. Belkic K, Landsbergis P, Schnall P, et al. Psychosocial factors: review of the empirical data among men. Occup Med 2000; 15: 24–46.
  13. Berkman LF, Glass TA. Social integration, social networks, social support and health. In: Berkman LF, Kawachi I, eds. Social Epidemiology. New York: Oxford University Press; 2000.
  14. Penley JA, Tomaka J, Wiebe JS. The association of coping to physical and psychological health outcomes: a meta-analytic review. J Behav Med 2002; 25: 551–603.[CrossRef][Medline]
  15. Aspinwall LG, Taylor SE. A stitch in time: self-regulation and proactive coping. Psychol Bull 1997; 121: 417–436.[CrossRef][Medline]
  16. Steptoe A. The significance of personalcontrol in health and disease. In: Steptoe A, Appels A, eds. Stress, PersonalControl and Health. Chichester: John Wiley; 1989: 309–318.
  17. Marmot MG, Davey Smith G, Stansfeld S, et al. Health inequalities among British civil servants: the Whitehall II study. Lancet 1991; 337: 1387–1393.[CrossRef][Medline]
  18. Lynch JW, Kaplan GA, Shema SJ. Cumulative impact of sustained economic hardship on physical, cognitive, psychological, and social functioning. N Engl J Med 1997; 337: 1889–1895.[Abstract/Free Full Text]
  19. Brunner E Davey Smith G, Marmot M, et al. Childhood social circumstances and psychosocial and behavioural factors as determinants of plasma fibrinogen. Lancet 1996; 347: 1008–1013.[CrossRef][Medline]
  20. Hemingway H, Nicholson A, Stafford M, et al. The impact of socioeconomic status on health functioning as assessed by the SF-36 questionnaire: The Whitehall II Study. Am J Public Health 1997; 87: 1484–1490.[Abstract/Free Full Text]
  21. Wamala SP, Murray MA, Horsten M, et al. Socioeconomic status and determinants of hemostatic function in healthy women. Arterioscl Thromb Vasc Biol 1999; 19: 485–492.[Abstract/Free Full Text]
  22. Steptoe A, Feldman PM, Kunz S, et al. Stress responsivity and socioeconomic status: A mechanism for increased cardiovascular disease risk? Euro Heart J 2002; 23: 1757–1763.[Abstract/Free Full Text]
  23. Steptoe A, Kunz S, Owen N, et al. The influence of socioeconomic status and job control on plasma fibrinogen responses to acute mental stress. Psychosom Med 2003; 65: 137–144.[Abstract/Free Full Text]
  24. Bosma H, Marmot MG, Hemingway H, et al. Low job control and risk of coronary heart disease in Whitehall II (prospective cohort) study. BMJ 1997; 314: 558–565.[Abstract/Free Full Text]
  25. Pearlin LI, Menaghan EG, Lieberman MA, et al. The stress process. J Health Soc Behav 1981; 22: 337–356.[CrossRef][Medline]
  26. Steptoe A, Feldman PJ. Neighborhood problems as sources of chronic stress: development of a measure of neighborhood problems, and associations with socioeconomic status and health. Ann Behav Med 2001; 23: 177–185.[CrossRef][Medline]
  27. Stansfeld S, Marmot M. Deriving a survey measure of social support: the reliability and validity of the Close Persons Questionnaire. Soc Sci Med 1992; 35: 1027–1035.
  28. Levenson H. Differentiating among internality, powerful others, and chance. In: Lefcourt HM, ed. Research with the Locus of Control Construct Vol 1: Assessment Methods. New York: Academic Press; 1981: 15–63.
  29. Carver CS, Scheier MF, Weintraub JK. Assessing coping strategies: A theoretically based approach. J Pers Soc Psychol 1989; 56: 267–283.[CrossRef][Medline]
  30. Adler NE, Epel ES, Castellazzo G, et al. Relationship of subjective and objective social status with psychological and physiological functioning: preliminary data in healthy white women. Health Psychol 2000; 19: 586–592.[CrossRef][Medline]
  31. Cohen S, Kamarck T, Mermelstein R. A global measure of perceived stress. J Health Soc Behav 1983; 24: 386–396.
  32. Goldberg D. Manual of the General Health Questionnaire. Windsor: NFER-Nelson; 1972.
  33. Stansfeld SA, North FM, White I, et al. Work characteristics and psychiatric disorder in civil servants in London. J Epid Comm Health 1995; 49: 48–53.[Abstract]
  34. Everson SA, Kaplan GA, Goldberg DE, et al. Hopelessness and 4-year progression of carotid atherosclerosis. The Kuopio Ischemic Heart Disease Risk Factor Study Arterioscler Thromb Vasc Biol 1997; 17: 1490–1495.[Abstract/Free Full Text]
  35. Jenkins CD, Stanton BA, Niemcryk SJ, et al. A scale for the estimation of sleep problems in clinical research. J Clin Epidemiol 1988; 41: 313–321.[CrossRef][Medline]
  36. Rosenburg M. Society and the Adolescent Self-Image. Princeton: Princeton University Press; 1965.
  37. Cook WW, Medley DM. Proposed hostility and pharisaic-virtue scales for the MMPI. J Appl Psychol 1954; 238: 414–418.[CrossRef]
  38. Russell D, Peplau LA, Cutrona CE. The revised UCLA Loneliness Scale: concurrent and discriminant validity evidence. J Pers Soc Psychol 1980; 39: 472–480.[CrossRef][Medline]
  39. Diabetes Control and Complications Trial Research Group. The effect of intensive treatment of diabetes on the development and progression of long-term complications in insulin-dependent diabetes mellitus. N Engl J Med 1993; 329: 977–986.[Abstract/Free Full Text]
  40. Gaffney PJ, Wong MY. Collaborative study of a proposed international standard for plasma fibrinogen measurement. Thromb Haemost 1992; 68: 428–432.[Medline]
  41. Steptoe A, Willemsen G, Owen N, et al. Acute mental stress elicits delayed increases in circulating inflammatory cytokine levels. Clin Sci 2001; 101: 185–192.[Medline]
  42. Lowe G, Rumley A, Norrie J, et al. Blood rheology, cardiovascular risk factors, and cardiovascular disease: the West of Scotland Coronary Prevention Study. Thromb Haemost 2000; 84: 553–558.[Medline]
  43. Jenkinson C, Layte R, Wright L, et al. The UK. SF-36: An Analysis and Interpretation Manual. Oxford: Health Services Research Unit; 1996.
  44. Orth-Gomér K, Wamala SP, Horsten M, et al. Marital stress worsens prognosis in women with coronary heart disease: The Stockholm Female Coronary Risk Study. JAMA 2000; 284: 3008–3014.[Abstract/Free Full Text]
  45. Kiecolt-Glaser JK, Newton TL. Marriage and health: his and hers. Psychol Bull 2001; 127: 472–503.[CrossRef][Medline]
  46. Lachman ME, Weaver SL. The sense of control as a moderator of social class differences in health and well-being. J Pers Soc Psychol 1998; 74: 763–773.[CrossRef][Medline]
  47. Deleted in proof.
  48. Stansfeld SA, Fuhrer R, Shipley MJ, et al. Psychological distress as a risk factor for coronary heart disease in the Whitehall II Study. Int J Epidemiol 2002; 31: 248–255.[Abstract/Free Full Text]
  49. Cacioppo JT, Hawkley LC, Crawford LE, et al. Loneliness and health: potential mechanisms. Psychosom Med 2002; 64: 407–417.[Abstract/Free Full Text]
  50. Bjorntorp P. Do stress reactions cause abdominal obesity and comorbidities? Obes Rev 2001; 2: 73–86.[CrossRef][Medline]
  51. Khaw KT, Wareham N, Luben R, et al. Glycated haemoglobin, diabetes, and mortality in men in Norfolk cohort of European prospective investigation of cancer and nutrition (EPIC-Norfolk). BMJ 2001; 322: 15–18.[Abstract/Free Full Text]
  52. Horsten M, Mittleman MA, Wamala SP, et al. Social relations and the metabolic syndrome in middle-aged Swedish women. J Cardiovasc Risk 1999; 6: 391–397.[Medline]
  53. Ravaja N, Keltikangas-Jarvinen L, Viikari J. Life changes, locus of control and metabolic syndrome precursors in adolescents and young adults: a three-year follow-up. Soc Sci Med 1996; 43: 51–61.
  54. Danesh J, Collins R, Appleby P, et al. Association of fibrinogen, C-reactive protein, albumin, or leukocyte count with coronary heart disease: meta-analyses of prospective studies. JAMA 1998; 279: 1477–1482.[Abstract/Free Full Text]
  55. Danesh J, Collins R, Peto R, et al. Haematocrit, viscosity, erythrocyte sedimentation rate: meta-analyses of prospective studies of coronary heart disease. Eur Heart J 2000; 21: 515–520[Free Full Text]



This article has been cited by other articles:


Home page
 Palliat MedHome page
B. Hanratty, P. Holland, A. Jacoby, and M. Whitehead
Review article: Financial stress and strain associated with terminal cancer a review of the evidence
Palliative Medicine, October 1, 2007; 21(7): 595 - 607.
[Abstract] [PDF]

Home page
 Psychosom. Med.Home page
V. K. Tsenkova, G. D. Love, B. H. Singer, and C. D. Ryff
Socioeconomic Status and Psychological Well-Being Predict Cross-Time Change in Glycosylated Hemoglobin in Older Women Without Diabetes
Psychosom Med, October 1, 2007; 69(8): 777 - 784.
[Abstract] [Full Text] [PDF]

Home page
 Psychosom. Med.Home page
E. M. Friedman, G. D. Love, M. A. Rosenkranz, H. L. Urry, R. J. Davidson, B. H. Singer, and C. D. Ryff
Socioeconomic Status Predicts Objective and Subjective Sleep Quality in Aging Women
Psychosom Med, September 1, 2007; 69(7): 682 - 691.
[Abstract] [Full Text] [PDF]

Home page
 J Aging HealthHome page
J. S. Ziembroski and M. J. Breiding
The Cumulative Effect of Rural and Regional Residence on the Health of Older Adults.
J Aging Health, October 1, 2006; 18(5): 631 - 659.
[Abstract] [PDF]

Home page
 Psychosom. Med.Home page
P. P. Vitaliano, R. Persson, A. Kiyak, H. Saini, and D. Echeverria
Caregiving and Gingival Symptom Reports: Psychophysiologic Mediators
Psychosom Med, November 1, 2005; 67(6): 930 - 938.
[Abstract] [Full Text] [PDF]

Home page
 J. Gerontol. B Psychol. Sci. Soc. Sci.Home page
L. K. George
Socioeconomic Status and Health Across the Life Course: Progress and Prospects
J. Gerontol. B. Psychol. Sci. Soc. Sci., October 1, 2005; 60(suppl_Special_Issue_2): S135 - S139.
[Full Text] [PDF]

Home page
 Biol Res NursHome page
S. L. Szanton, J. M. Gill, and J. K. Allen
Allostatic Load: A Mechanism of Socioeconomic Health Disparities?
Biol Res Nurs, July 1, 2005; 7(1): 7 - 15.
[Abstract] [PDF]

This Article
Right arrow Abstract Freely available
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
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Steptoe, A.
Right arrow Articles by Marmot, M.
Right arrow Search for Related Content
PubMed