Psychosomatic Medicine
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
Published online before print October 17, 2007, 10.1097/PSY.0b013e318157466f
This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tsenkova, V. K.
Right arrow Articles by Ryff, C. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tsenkova, V. K.
Right arrow Articles by Ryff, C. D.
Related Collections
Right arrow Social Class
Right arrow Diabetes
Right arrow Other Cardiovascular Medicine
Psychosomatic Medicine 69:777-784 (2007)
© 2007 American Psychosomatic Society

ORIGINAL ARTICLES

Socioeconomic Status and Psychological Well-Being Predict Cross-Time Change in Glycosylated Hemoglobin in Older Women Without Diabetes

Vera K. Tsenkova, MA, Gayle Dienberg Love, PhD, Burton H. Singer, PhD and Carol D. Ryff, PhD

From the Department of Psychology (V.K.T., C.D.R.) and Institute on Aging (G.D.L., B.H.S., C.D.R.), University of Wisconsin-Madison, Madison, Wisconsin; Office of Population Research (B.H.S.), Princeton University, Princeton, New Jersey.

Address correspondence and reprint requests to Vera K. Tsenkova, Department of Psychology, University of Wisconsin-Madison, W. J. Brogden Hall, 1202 W. Johnson Street, Madison, WI 53706-1696. E-mail: tsenkova{at}wisc.edu


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Objective: To investigate whether socioeconomic status and psychological well-being (eudaimonic and hedonic aspects) predicted nondiabetic levels of glycosylated hemoglobin (HbA1c) over time, after adjusting for covariates and baseline level of HbA1c.

Methods: These questions were investigated with a longitudinal sample (n = 97; age = 61–91 years) of older women without diabetes. Socioeconomic status, well-being, and health behaviors were assessed using self-administered questionnaires. Fasting blood samples for assays of HbA1c were obtained before 7 AM during the respondents’ overnight stay at the General Clinical Research Center at the University of Wisconsin-Madison. All measurements were obtained at baseline and 2-year follow-up.

Results: Regression analyses showed that higher income and positive affect predicted lower levels of HbA1c, after controlling for baseline HbA1c and health factors. Additionally, three well-being measures (purpose in life, personal growth, and positive affect) moderated the relationship between income and HbA1c.

Conclusion: These results suggest that psychological well-being and socioeconomic status interact in important ways in influencing nondiabetic glucose metabolism.

Key Words: eudaimonic well-being • hedonic well-being • socioeconomic status • income • glycosylated hemoglobin

Abbreviations: HbA1c = glycosylated hemoglobin; SES = socioeconomic status; MASQ = Mood and Anxiety Symptom Questionnaire; GCRC = General Clinical Research Center; WHR = waist-to-hip ratio.


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Glycosylated hemoglobin (HbA1c), an indicator of glycemic control, gives an estimate of the average plasma glucose over the previous 3 months. Glycemic control is essential for the management of Type 1 and Type 2 diabetes, as increased glycemic levels have been reliably linked to diabetes-related complications (1). However, recent research has highlighted the importance of HbA1c to the health of people without diabetes: in epidemiological studies, 1% increase in nondiabetic HbA1c levels was associated with 40% increase in coronary heart disease, 16% increase in cardiovascular disease, and 26% increase in mortality, underscoring the importance of HbA1c as a progressive cardiovascular risk factor (2–5). Given the importance of glucose metabolism to diabetes and cardiovascular health, different indices of nondiabetic glucose metabolism have also been linked to psychosocial and sociodemographic factors, such as depression, hostility, and anger (6), coping (7,8), socioeconomic status (SES) (7), and neighborhood crime (9). The goal of this study was to examine the interplay between SES and multiple aspects of positive psychological functioning in predicting cross-time changes in HbA1c.

SES, Psychosocial Factors, and HbA1c
SES and Glycemic Control
The prevalence of Type 2 diabetes increases with low SES, as indexed by poverty income ratio or employment grade (10–12). Furthermore, low SES is associated with poor metabolic control (13) even in countries with universal health care system, despite greater adherence to preventive healthcare measures (14). These effects are only partly explained by known risk factors associated with low SES such as smoking, obesity, and hypertension.

Consistent with the larger literature on diabetic HbA1c levels, Feldman and Steptoe (7) documented an inverse relationship between SES, indexed by grade of employment, and nondiabetic HbA1c in a subsample of the Whitehall II epidemiological cohort of British civil servants. Additionally, Brummett et al. (9) found that worse neighborhood characteristics moderate the effects of caregiving on HbA1c levels in caregivers without diabetes. Specifically, caregivers and noncaregivers in good neighborhoods had similar HbA1c levels, whereas caregivers with worse neighborhood characteristics had significantly higher HbA1c levels than noncaregivers with the same neighborhood characteristics.

SES and Well-Being
The scientific study of well-being distinguishes between eudaimonic well-being, which entails purposeful engagement and self-development, and hedonic well-being, such as happiness and contentment (15,16). Although both approaches assess well-being, they address different features of what it means to be well (17). Evidence has also shown that eudaimonic and hedonic well-being are empirically distinct and clarified that various combinations of them are differentially linked to age and education (18). In general, eudaimonic well-being has been positively related to education and occupational status (19,20), although prior work has also documented resilience among people with low socioeconomic standing or high life adversity (16,21–24). With regard to hedonic well-being, higher income is also related to greater happiness, but effects are generally small (25). In addition, increases in income are not associated with increases in well-being (26,27).

Psychosocial Factors and HbA1c
Relevant to the present study, psychosocial factors have been linked to HbA1c in people without diabetes. Work by Feldman and Steptoe (7) documented an inverse association between problem-focused coping and HbA1c and also linked a cumulative measure of psychosocial adversity and vulnerability to increased HbA1c (28). Tsenkova, Love, Singer, and Ryff (8) showed that higher levels of problem-focused coping and positive affect predicted cross-time decline in HbA1c levels in older women without diabetes, after controlling for baseline HbA1c and sociodemographic and health factors. Furthermore, positive affect was found to moderate the effects of problem-focused coping, such that the adverse effects of low problem-focused coping on cross-time changes in HbA1c were amplified among those who also had low levels of positive affect. To our knowledge, no study has linked both measures of eudaimonic and hedonic well-being to HbA1c, at the same time considering their interplay with SES.

Study Aims
Drawing on previous literature, we aimed to assess the links between SES, eudaimonic and hedonic well-being with levels of HbA1c in a sample of aging women without diabetes. By using a nondiabetic sample, it becomes possible to demonstrate that the relationships observed are not mediated through diabetes-related responsibilities (e.g., checking glucose levels, taking medications, monitoring diet and exercise), but rather reflect more general processes related to the biological correlates of well-being and SES.

Using a longitudinal sample, which allows us to predict changes in HbA1c, we sought to extend previous research in four key ways: a) test the effects of SES on HbA1c and investigate whether the links depend on the SES measure used (income or years of education); b) investigate the independent effects (i.e., whether one occurs net of the other) of eudaimonic and hedonic well-being on HbA1c; c) test for possible interactive influences between SES and well-being; and d) investigate whether all hypothesized effects were independent of the possible influence of negative affectivity. The latter is responsive to recent observations about the importance of investigating whether the health benefits attributed to positive affect were not mere reflections of the absence of negative affect (29,30).

With regard to hypotheses, we drew on the prior literature to predict that higher SES, measured by income and years of education, would predict lower cross-time levels of HbA1c. Furthermore, we expected that higher eudaimonic and hedonic well-being would also be independently linked with lower cross-time levels of HbA1c.

Finally, based on previous work on resilience, we tested whether well-being moderates the relationship between SES and glycemic control. Given our interest in positive health (8,16,31), we were particularly interested in whether high well-being might compensate for the risks associated with low SES, rendering their HbA1c levels comparable with those individuals with high SES. Conversely, low well-being was expected to have detrimental effects, serving to amplify the adverse effects of low SES, and thereby contribute to worse glycemic control. Finally, we also hypothesized that all of the above effects would hold after controlling for negative affect in the models.


   METHOD
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 
Participants
Respondents were older women without diabetes from a prior four-wave longitudinal study of aging women who were undergoing community relocation (n = 301). About 5 years later, additional research support allowed for a fifth wave of data collection that included psychosocial assessments and baseline biomarkers on approximately half of the original sample (n = 135). Among those from the original study who did not participate, 16% were no longer eligible (due to death, morbidity, or moving out of the area); another 42% declined to participate (for no stated reason, or due to concerns about health and demands of the study). The newly recruited sample was not significantly different from the original sample with regard to chronic conditions, health symptoms, income, or marital status, although they did score significantly higher on four of the six eudaimonic well-being measures (environmental mastery, personal growth, purpose in life, and self-acceptance) than the original sample. No comparisons on positive affect were available as those scales were added to the biomarker study.

Two years later, parallel psychological and biological measures were obtained on 115 of these women. Attrition analyses showed no significant differences on any variable of interest (HbA1c, positive and negative affect, eudaimonic scales) between women who participated in both waves of biomarker collection and those who dropped out. Of the 115 women who provided data in both waves of biological data collection, 97 met the inclusion criteria of having no history of diet or pharmacologically controlled diabetes and HbA1c levels <7.0%. The first wave of biological data collection took place between February 2000 and January 2002; the second wave was completed between April 2002 and March 2004; the biomarker supplement was approved by the Institutional Review Board, Protocol 1996-446. Women in the current study were predominantly white (97% were white and 3% black) and ranged in age from 61 to 91 years (mean = 73.86 years) when they first participated in comprehensive psychosocial and biological assessments. The sample income was comparable to the US Census data for older adults, which found that between 2000 and 2004, 10% of adults >65 years lived in poverty, 28% had low income, 35% had middle income, and 27% had high income. For our participants, these numbers are respectively 12%, 26%, 42%, and 20% (32). Participants were paid $100 compensation in addition to all travel expenses. Participants signed the informed consent form when they arrived for an overnight stay at the General Clinical Research Center (GCRC) located within the University of Wisconsin Hospital and Clinics. Table 1 shows the descriptive characteristics of the study sample.


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

  		TABLE 1. Descriptive Statistics for Demographic and Health-Related Variables (n = 97)

 

Measures
All psychosocial and biological measures were collected at baseline and 2-year follow-up. Self-administered questionnaires were sent to respondents 3 to 4 weeks before their visit to the University of Wisconsin-Madison campus for the biomarker assessments. These questionnaires were completed independently and returned to investigators at the time of their campus visit. Demographic data such as marital status, income, education, and age were also obtained.

Socioeconomic Status
SES was operationalized by two variables: pretax household income and years of education completed. Data were obtained from self-administered questionnaires completed before the biological data collection. A measure of wealth or assets for the respondents was not available.

Psychosocial
Eudaimonic well-being refers to the realization of personal potential (16) and was measured with Ryff’s psychological well-being scales (33). This instrument incorporates six 14-item scales that represent different dimensions of well-being based on theoretical integration of numerous formulations of positive functioning and include autonomy, environmental mastery, personal growth, positive relations with others, purpose in life, and self-acceptance. Internal consistency for the six scales ranged from 0.85 to 0.92 (Table 2). Previous publications have documented the reliability and validity of the scales (33,34). The hypothesized 6-factor structure of well-being has been supported by multiple confirmatory factor analytic studies, including some involving nationally representative samples (34–39). Recent work has also documented that eudaimonic well-being is empirically distinct from, yet related to, hedonic well-being (17).


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

  		TABLE 2. Descriptive Statistics and Reliabilities for Eudaimonic and Hedonic Well-Being

 

Hedonic well-being was assessed using the positive affect scale of the short form Mood and Anxiety Symptom Questionnaire (MASQ) (40). The MASQ-Short Form uses five scales to measure a respondent’s mood and gather information about problems and experiences that the respondent may have encountered over the past week. The high positive affect subscale includes 14 items that capture more hedonic, joy-in-living aspects of positive affect and ask respondents how much they had felt various positive emotions (e.g., happy, cheerful, optimistic, having fun) in the past week. The other four subscales—general distress-depressive symptoms (12 items), general distress-anxious symptoms (11 items), loss of interest (8 items), and anxious arousal (17 items)— tap symptoms indicative of negative affectivity. Respondents were asked to indicate the extent to which each statement described their feelings over the past week. Response options ranged from 1 (not at all) to 5 (extremely) (Table 2).

Physical and Biological Measures
After completing the self-administered questionnaires, participants were admitted to the GCRC. A nurse or physician took the respondent’s medical history and conducted a physical health examination. GCRC nursing staff obtained blood samples. Use of prescription and over-the-counter medications was recorded during the GCRC visit. Fasting blood samples for assays of HbA1c were obtained before 7 AM during the respondents’ overnight stay by the nursing staff. Assays were carried out using a high-performance liquid chromatography, a boronate affinity HbA1c method that is not subject to analytical interference by hemoglobin variants (41). The coefficient of variation for this method is 3.4% for people without diabetes. All assays were conducted at the University of Wisconsin Hospital and Clinics Clinical Laboratory.

Statistical Method
Frequency distributions for all continuous measures were first examined and normalized as needed. Specific items related to overall health or conditions with which HbA1c has been linked were then selected for preliminary correlational analyses. These included smoking and drinking history, family history of diabetes, medications (blood pressure, depression, ß blockers, cholesterol, corticosteroid), age, income, years of education, marital status (coded as married versus all others), heart disease or problems, and waist-to-hip ratio (WHR). Only those showing significant correlations (p < .05) with HbA1c were kept in the regression model as covariates. This final set of covariates consisted of age, marital status, income, WHR, and cholesterol medications.1

Multiple regression analyses were conducted in which the final set of covariates was included at the first step of the multivariate model. Baseline measure of HbA1c was then added at the second step of the model. After including covariates and baseline HbA1c levels, two types of models were created. First, main effects models consisted of separate regression models run for each SES or psychosocial factor (income, years of education, positive affect, psychological well-being). Second, interaction models included a measure of SES and well-being at step 4 as well as their interaction term at step 5. All continuous independent variables were mean-centered.

Further analyses were built on the above analytic models to test the independence of eudaimonic and hedonic well-being from each other as well as net of effects of negative affect. Specifically, to assess whether the effects of hedonic well-being are independent of eudaimonic well-being, measures of the latter were included as covariates, both as individual controls (i.e., only one eudaimonic well-being subscale), and all six subscales together. In the models testing the independence of eudaimonic well-being, hedonic well-being was added as a covariate in each model.

To test the independence of the positive and negative measures used, the measures of negative affect, both as individual controls (i.e., only one MASQ negative affectivity subscale) and all four subscales together were added as further covariates to all models. Finally, since 12% of our sample was taking antidepressant medications (Table 1), the effects of diagnosis of depression were controlled as well.

Missing Values
There was one missing value in baseline HbA1c measurement. In all analyses, pairwise deletion was used and only cases that did not have data on a variable used in the current calculation were omitted.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
DISCUSSION
 NOTES
 REFERENCES
 
All regression models described below included the full set of control variables and a baseline measure of HbA1c.

Main Effect Models
Table 3 shows the results of hierarchical regression analyses of different SES and well-being measures predicting follow-up HbA1c.


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

  		TABLE 3. Hierarchical Regression Analysis for SES and Well-Being Measures Predicting T2 HbA1c

 

SES
The first hypothesis tested whether baseline measures of SES predicted cross-time changes in HbA1c levels. Pretax household income was a significant predictor of lower cross-time HbA1c levels (R2 = .399; ß = –0.186; p < .05). Years of education did not predict HbA1c levels.

Well-Being
The second hypothesis investigated the effects of eudaimonic and hedonic well-being on HbA1c. None of the scales of eudaimonic well-being was a significant predictor of HbA1c. However, hedonic well-being, as measured by positive affect, was a significant predictor of lower HbA1c levels over time (R2 = .429; ß = –0.243; p < .01).

Interaction Models
SES, Well-Being, and HbA1c
Our third hypothesis investigated whether well-being moderated the relationship between SES and HbA1c (Table 4). Three significant interactions were obtained: both eudaimonic (purpose in life and personal growth) and hedonic (positive affect) well-being moderated the relationship between HbA1c and income. The interactions were graphed according to the established procedures (42), followed by tests of whether the slope of the simple regression line was significantly different from zero at 1 standard deviation above and below the mean of well-being. Results showed a consistent pattern: low levels of positive affect (ß = –0.023; SE = 0.006; p < .01), purpose in life (ß = –0.020; SE = 0.006; p < .01), and personal growth (ß = –0.020; SE = 0.006; p < .01) amplified the adverse effects of low income on cross-time changes in HbA1c.


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

  		TABLE 4. Hierarchical Regression Analysis for Psychosocial Factors Predicting T2 HbA1c

 

This pattern, illustrated in Figure 1, shows that among participants with low income, low well-being amplified the detrimental effect, contributing to ever higher levels of HbA1c. The simple slopes for high well-being were not significantly different from zero, thus illustrating that the HbA1c levels of these individuals did not vary depending on whether they were economically advantaged or disadvantaged. The same pattern of effects was also obtained for the measures of personal growth and positive affect.


Figure 111
View larger version (8K):
[in this window]
[in a new window]

  		Figure 1. Interactive effect of psychological well-being (PWB): Personal growth (PG) and income on cross-time HbA1c levels (p < .05).

 

Independence Models (results not shown)
Statistically controlling for hedonic well-being did not affect the nonsignificant relationship between eudaimonic well-being and HbA1c. Similarly, including measures of eudaimonic well-being—individually or altogether—did not change the significant relationship between positive affect and HbA1c. This independence of the relationships between the two measures of well-being and HbA1c levels was observed in main effect and interaction models.

Finally, statistically controlling for diagnosis of depression and negative affectivity (depressive symptoms, loss of interest, anxious arousal, and distress-anxious)—individually or altogether—did not alter any of the relationships observed in main effect and interaction models. None of the measures of negative affectivity showed a significant relationship to HbA1c in our models.


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
NOTES
 REFERENCES
 
The key aims of this investigation were to assess the independent and interactive influences of SES and psychological well-being on cross-time changes in HbA1c in a sample of aging women. First, we found that low SES predicted, as hypothesized, higher levels of HbA1c over time. The SES measures we used—income and years of education—were moderately related (r = .321; p < .001), and both were normally distributed in our sample, yet only income was related to cross-time HbA1c levels. Multiple studies have documented a consistent inverse relationship between SES and incidence of diabetes as well as subsequent glycemic control, with both income and education being implicated. Achieving good glycemic control for people with diabetes involves a complex regimen of responsibilities (e.g., meal preparations, exercise, medication regimens) that may require both knowledge (i.e., education) and resources (i.e., income). These possible mediating influences are not operative among nondiabetics, which makes it difficult to explain why income, but not educational standing, was linked to HbA1c levels after adjusting for other factors (WHR, age, and medications). Perhaps the differences reflect sampling limitations—namely, the study was based exclusively on older, white women. Thus, a critical requirement in interpreting the differential main effects is determining whether the finding holds when assessed with a more diverse (by age, gender, and SES) sample.

We also observed differential patterns with regard to the effects of eudaimonic and hedonic well-being on changes in HbA1c; we found such effects for hedonic well-being (positive affect), independent of eudaimonic influences and other covariates, thereby extending our previous findings (8). Thus, these results add good glycemic control to the growing body of evidence on the health benefits of positive affect, where others have documented longer healthy life expectancy and reduced risk of physical disease as well as reduced risk of mortality, disability, and stroke in older adults (29,44–47). However, a parallel main effect benefit of eudaimonic well-being was not evident. Our prior research (43) has shown that eudaimonic and hedonic aspects of well-being have largely distinct biological correlates (except for high-density lipoprotein cholesterol). These new findings underscore that the two aspects of well-being are also not equivalent in how they relate to HbA1c. Such differences call for more precise theoretical formulation of how and why these distinct patterns are evident, an issue to which we return after considering the evidence for moderating effects.

Most previous studies have examined single factor influences on glycemic control, thereby failing to take account of combinations of sociodemographic and psychological factors that might better predict varying levels of HbA1c (for an exception, see work by Brummett et al. (9)). We found support for the interplay of income with three different aspects of well-being in predicting HbA1c levels. Specifically, two eudaimonic measures (purpose in life and personal growth) and hedonic well-being (positive affect) moderated the relationship between income and HbA1c. In all cases, we found that lower well-being amplified the adverse effects of lower economic standing, thus contributing to ever higher levels of HbA1c. High well-being, in contrast, predicted comparable levels of HbA1c among those with lower as well as higher levels of income. Taken together, these findings underscore the importance of putting together sociodemographic factors as well as individual differences in various aspects of psychological well-being in predicting cross-time changes in levels of HbA1c.

Such moderating effects also extend prior work (23), which has shown that those with economic disadvantage but compensating good quality relationships had reduced risk of high allostatic load, relative to those who had both SES and relationship vulnerabilities. Additionally, the findings build on previous research linking positive affect to HbA1c (8) by showing that low positive affect amplifies the adverse effects of low income on glycemic control. Regarding the moderating influences of purpose in life and personal growth, we would also note that these same two dimensions of well-being were previously linked to other biomarkers (e.g., salivary cortisol) in the same sample (16), thereby suggesting that these particular aspects of well-being are linked to diverse biological systems, perhaps modulating the effects of a larger sociodemographic environment. Such linkages may be particularly relevant in later life, when purpose and growth have been shown to decline on average, although concomitant risk factors for chronic disease are accumulating.

An important part of the present investigation was establishing independence of the various psychosocial predictors. Specifically, we documented that the effects of eudaimonic well-being were independent of the effects of hedonic well-being, and vice versa. This finding is consistent with prior work proposing the related yet distinct nature of eudaimonic and hedonic well-being (17). Additionally, all documented relationships were unaffected by adjusting for negative affect and depression, suggesting that the lower biological risk conferred by positive psychological factors was not attributable to the deleterious effects of negative factors. Evidence is thus mounting that the biological costs linked to low well-being or the benefits linked to high well-being are not merely the flip side of what has been previously documented regarding correlates of psychological distress, but constitute independent influences in their own right.

The mechanisms through which SES and well-being exert their effects on glucose metabolism are not well understood, although hypothalamus-pituitary-adrenal axis activity and immune functioning are likely pathways (16,43–45). More specifically, one possible mechanism explaining the main effect relationship between the positive affect and HbA1c could be associations with neurotransmitters such as epinephrine and norepinephrine that have been previously linked to positive affect (44–46) as well as glycemic control (47). This pathway may not, however, be relevant for understanding the link between eudaimonic well-being, where we observed that the links to HbA1c are only evident under conditions of socioeconomic disadvantage. Such a pattern extends our prior work showing that SES is positively correlated with purposeful life engagement and personal growth, with the effect especially strong for women (48).

The mechanistic processes underlying these eudaimonic effects remain to be identified, although they may involve more behavioral pathways (e.g., those who see their lives as purposeful and growth-producing may practice better nutrition, get more exercise, better sleep, and more closely monitor their health status). To assess these differing possibilities, including both behavioral and biological mechanisms, future research on glycemic control must include additional biomarkers as well as assessment of health behaviors, along with the variables of focus in the present investigation.

Our study is limited by the age, gender, and ethnic status (white) of the respondents. Samples with more socioeconomic, ethnic, and age diversity will be needed to determine the relevance of the obtained income and well-being interactions for other social groups. Including more temporal measurements of HbA1c levels would be advantageous for tracking dynamic change over time, including the important question of whether the patterns observed herein constitute early warning signals for who transitions to disease outcomes.

Glycemic control, we underscore, is essential in diabetes and cardiovascular disease—two major chronic conditions in the US and other industrial nations; hence, the attention given to the diabetic epidemic (49), together with the related need for increased attention to HbA1c as a primary indicator of long-term glycemic utilization. Furthermore, HbA1c is an independent predictor of cardiovascular events, regardless of diabetes status, and as such, is considered an important indicator of cardiovascular risk (2).

Considering issues of application, we submit that psychosocial resources, such as aspects of well-being, constitute potentially modifiable factors (50,51) that people bring to their life stressors, including the challenges of social inequality. Thus, interventions to promote well-being, particularly among those who most need it, constitute important future directions related to our findings that well-being moderates the effects of socioeconomic standing on cross-time changes in HbA1c. Pending confirmation from larger, more diverse samples, our work suggests that glycemic control, which is implicated in multiple health outcomes (diabetes, cardiovascular disease), may itself be partially shaped by standing in the socioeconomic hierarchy as well as by subjective levels of contentment and engagement in life.


   NOTES
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
REFERENCES
 
1Because the use of univariable screening of control variables for inclusion in the regression models may result in lack of stability in small samples (52), we did a principal components analysis (PCA) with the original set of 12 possible covariates and another PCA using only the final set of five covariates included in the manuscript. The PCA of all possible covariates produced five factors (across various rotational schemes), and as such, offered no gain over the number of covariates we already had in the model (i.e., 5). However, the PCA of final covariates did provide a reduction from five variables to two factors. Therefore, we extracted factor scores for each individual and reran the regression models, using these scores (instead of the five prior covariates). Results were consistent with our previous findings; i.e., no differences were evident in significance levels and only slight differences existed in the regression coefficients in main effect models, interaction models, and their subsequent simple slope analyses. Thus, we found no evidence of instability in the findings and kept the original covariates because the information about them is more easily interpretable than using the two extracted components. Back

Received for publication February 7, 2007; revision received June 26, 2007.

This research was supported by Grants R01-AG08979 and P01-AG020166 from the National Institute on Aging, Grant P50-MH61083 from the National Institute of Mental Health, and Grant M01-RR03186 from the National Institutes of Health to the University of Wisconsin General Clinical Research Center.

DOI:10.1097/PSY.0b013e318157466f


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 METHOD
 RESULTS
 DISCUSSION
 NOTES
 REFERENCES
 

  1. Alam T, Weintraub N, Weinreb J. What is the proper use of hemoglobin a1c monitoring in the elderly? J Am Med Dir Assoc 2005;6:200–4.[CrossRef][Medline]
  2. Gerstein HC. Glycosylated hemoglobin: finally ready for prime time as a cardiovascular risk factor. Ann Intern Med 2004;141:475–6.[Free Full Text]
  3. Khaw KT, Wareham N, Bingham S, Luben R, Welch A, Day N. Association of hemoglobin A1c with cardiovascular disease and mortality in adults: the European prospective investigation into cancer in Norfolk. Ann Intern Med 2004;141:413–20.[Abstract/Free Full Text]
  4. Muntner P, Wildman RP, Reynolds K, Desalvo KB, Chen J, Fonseca V. Relationship between HbA1c level and peripheral arterial disease. Diabetes Care 2005;28:1981–7.[Abstract/Free Full Text]
  5. Vitelli LL, Shahar E, Heiss G, McGovern PG, Brancati FL, Eckfeldt JH, Folsom AR. Glycosylated hemoglobin level and carotid intimal-medial thickening in nondiabetic individuals. The atherosclerosis risk in communities study. Diabetes Care 1997;20:1454–8.[Abstract]
  6. Suarez EC. Sex differences in the relation of depressive symptoms, hostility, and anger expression to indices of glucose metabolism in nondiabetic adults. Health Psychol 2006;25:484–92.[CrossRef][Medline]
  7. Feldman PJ, Steptoe A. Psychosocial and socioeconomic factors associated with glycated hemoglobin in nondiabetic middle-aged men and women. Health Psychol 2003;22:398–405.[CrossRef][Medline]
  8. Tsenkova V, Love G, Singer B, Ryff C. Coping and positive affect predict longitudinal change in glycosylated hemoglobin. Health Psychol, In Press.
  9. Brummett BH, Siegler IC, Rohe WM, Barefoot JC, Vitaliano PP, Surwit RS, Feinglos MN, Williams RB. Neighborhood characteristics moderate effects of caregiving on glucose functioning. Psychosom Med 2005;67:752–8.[Abstract/Free Full Text]
  10. Kumari M, Head J, Marmot M. Prospective study of social and other risk factors for incidence of type 2 diabetes in the Whitehall II study. Arch Intern Med 2004;164:1873–80.[Abstract/Free Full Text]
  11. Robbins JM, Vaccarino V, Zhang H, Kasl SV. Socioeconomic status and type 2 diabetes in African American and non-Hispanic white women and men: evidence from the third national health and nutrition examination survey. Am J Public Health 2001;91:76–83.[Abstract]
  12. Maty SC, Everson-Rose SA, Haan MN, Raghunathan TE, Kaplan GA. Education, income, occupation, and the 34-year incidence (1965–99) of type 2 diabetes in the Alameda County Study. Int J Epidemiol 2005;34:1274–81.[Abstract/Free Full Text]
  13. Hassan K, Loar R, Anderson BJ, Heptulla RA. The role of socioeconomic status, depression, quality of life, and glycemic control in type 1 diabetes mellitus. J Pediatr 2006;149:526–31.[CrossRef][Medline]
  14. Jotkowitz AB, Rabinowitz G, Raskin Segal A, Weitzman R, Epstein L, Porath A. Do patients with diabetes and low socioeconomic status receive less care and have worse outcomes? A national study. Am J Med 2006;119:665–9.[CrossRef][Medline]
  15. Ryan RM, Deci EL. On happiness and human potentials: a review of research on hedonic and eudaimonic well-being. Annu Rev Psychol 2001;52:141–66.[CrossRef][Medline]
  16. Ryff CD, Singer BH, Dienberg Love G. Positive health: connecting well-being with biology. Philos Trans R Soc Lond B Biol Sci 2004;359:1383–94.[Abstract/Free Full Text]
  17. Keyes CL. The mental health continuum: from languishing to flourishing in life. J Health Soc Behav 2002;43:207–22.[CrossRef][Medline]
  18. Keyes CL, Shmotkin D, Ryff CD. Optimizing well-being: the empirical encounter of two traditions. J Pers Soc Psychol 2002;82:1007–22.[CrossRef][Medline]
  19. Ryff CD, Magee WJ, Kling KC, Wing EH. Forging macro-micro linkages in the study of psychological well-being. In: Ryff CD, Marshall VW, editors. The Self and Society in Aging Processes. New York: Springer; 1999.
  20. Ryff CD, Singer B. Know thyself and become what you are: a eudaimonic approach to psychological well-being. J Happiness Studies, In Press.
  21. Markus HR, Ryff CD, Curhan KB, Palmersheim KA. In their own words: well-being at midlife among high school-educated and college educated adults. In: Brim OG, Ryff CD, Kessler RC, editors. How Healthy are We? A National Study of Well-Being at Midlife. Chicago: University of Chicago Press; 2004.
  22. Singer BH, Ryff CD. Racial and ethnic equalities in health: environmental, psychosocial, and physiological pathways. In: Devlin B, editor. Intelligence, Genes, and Success: Scientists Respond to the Bell Curve. New York: Springer; 1997.
  23. Singer BH, Ryff CD. Hierarchies of life histories and associated health risks. Ann N Y Acad Sci 1999;896:96–115.[CrossRef][Medline]
  24. Singer BH, Ryff CD, Carr D, Magee WJ: Life histories and mental health: a person-centered strategy. In: Raftery A, editor. Sociological Methodology. Washington, DC: American Sociological Association; 1998.
  25. Diener E, Suh EM, Lucas RE, Smith HL. Subjective well-being: three decades of progress. Psychol Bull 1999;125:76–302.
  26. Diener E, Sandvik E, Seidlitz L, Diener M. The relationship between income and subjective well-being: relative or absolute? Social Indicators Research 1993;28:195–223.[CrossRef]
  27. Diener E, Suh ME. Subjective well-being and age: An international analysis. Annual Review of Gerontology and Geriatrics 1997;17:304–24.
  28. Steptoe A, Marmot M. Burden of psychosocial adversity and vulnerability in middle age: associations with biobehavioral risk factors and quality of life. Psychosom Med 2003;65:1029–37.[Abstract/Free Full Text]
  29. Pressman SD, Cohen S. Does positive affect influence health? Psychol Bull 2005;131:925–71.[CrossRef][Medline]
  30. Steptoe A, Wardle J. Positive affect and biological function in everyday life. Neurobiol Aging 26 Suppl 2005;1:108–12.
  31. Ryff CD, Singer B. The contours of positive human health. Psychological Inquiry 1998;9:1–28.[CrossRef]
  32. Bureau USC: Current Population Survey. Annual Social and Economic Supplement, 1975–2005, 2002.
  33. Ryff CD. Happiness is everything, or is it? Explorations on the meaning of psychological well-being. J Pers Soc Psychol 1989;57:1069–81.[CrossRef]
  34. Ryff CD, Keyes CL. The structure of psychological well-being revisited. J Pers Soc Psychol 1995;69:719–27.[CrossRef][Medline]
  35. Cheng ST & Chan ACM. Measuring psychological well-being in the Chinese. Pers Indiv Diff 2005;38:1307–16.[CrossRef]
  36. Clarke PJ, Marshall VW, Ryff CD, Wheaton B. Measuring psychological well-being in the Canadian study of health and aging. Int Psychogeriatr 2001;13(Suppl 1):79–90.[CrossRef][Medline]
  37. Ryff CD, Singer B. Best news yet on the six-factor model of well-being. Social Science Research 2006;35:1102–18.
  38. Springer KW, Hauser RM. An assessment of the construct validity of Ryff’s scales of psychological well-being: method, mode, and measurement effects. Social Science Research 2006;35:1079–110.
  39. Van Dierendonck D. The construct validity of Ryff’s scales of psychological well-being and its extension with spiritual well-being. Personality and Individual Differences 2004;36:629–43.[CrossRef]
  40. Watson D, Weber K, Assenheimer JS, Clark LA, Strauss ME, McCormick RA. Testing a tripartite model: I. Evaluating the convergent and discriminant validity of anxiety and depression symptom scales. J Abnorm Psychol 1995;104:3–14.[CrossRef][Medline]
  41. Klenk DC, Hermanson GT, Krohn RI, Fujimoto EK, Mallia AK, Smith PK, England JD, Wiedmeyer HM, Little RR, Goldstein DE. Determination of glycosylated hemoglobin by affinity chromatography: comparison with colorimetric and ion-exchange methods, and effects of common interferences. Clin Chem 1982;28:2088–94.[Abstract/Free Full Text]
  42. Aiken LS, West SG: Multiple Regression: Testing and Interpreting Interactions. Newbury Park, CA: Sage Publications; 1991.
  43. Ryff CD, Dienberg Love G, Urry HL, Muller D, Rosenkranz MA, Friedman EM, Davidson RJ, Singer B. Psychological well-being and ill-being: do they have distinct or mirrored biological correlates? Psychother Psychosom 2006;75:85–95.[CrossRef][Medline]
  44. Berk LS, Tan SA, Fry WF, Napier BJ, Lee JW, Hubbard RW, Lewis JE, Eby WC: Neuroendocrine and stress hormone changes during mirthful laughter. Am J Med Sci 1989;298:390–6.[Medline]
  45. Codispoti M, Gerra G, Montebarocci O, Zaimovic A, Raggi MA, Baldaro B: Emotional perception and neuroendocrine changes. Psychophysiology 2003;40:863–8.[CrossRef][Medline]
  46. Cohen S, Doyle WJ, Turner RB, Alper CM, Skoner DP. Emotional style and susceptibility to the common cold. Psychosom Med 2003; 65:652–7.[Abstract/Free Full Text]
  47. Black PH. The inflammatory consequences of psychologic stress: relationship to insulin resistance, obesity, atherosclerosis and diabetes mellitus, type II. Med Hypotheses 2006;67:879–91.[CrossRef][Medline]
  48. Ryff CD, Singer B. From social structure to biology: Integrative science in pursuit of human health and well-being. In: Snyder CRL, Lopez SJ, editors. Handbook of Positive Psychology. New York: Oxford University Press; 2002.
  49. Steinbrook R. Facing the diabetes epidemic—mandatory reporting of glycosylated hemoglobin values in New York City. N Engl J Med 2006;354:545–8.[Free Full Text]
  50. Fava GA, Ruini C, Rafanelli C, Finos L, Salmaso L, Mangelli L, Sirigatti S. Well-being therapy: conceptual and technical issues. Psychother Psychosom 1999;68:171–9.[CrossRef][Medline]
  51. Fava GA, Ruini C, Rafanelli C, Finos L, Salmaso L, Mangelli L, Sirigatti S. Well-being therapy of generalized anxiety disorder. Psychother Psychosom 2005;74:26–30.[CrossRef][Medline]
  52. Harrell FE. Regression Modeling Strategies With Applications to Linear Models, Logistic Regression, and Survival Analysis. New York: Springer; 2001.




This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrowRequest Permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tsenkova, V. K.
Right arrow Articles by Ryff, C. D.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tsenkova, V. K.
Right arrow Articles by Ryff, C. D.
Related Collections
Right arrow Social Class
Right arrow Diabetes
Right arrow Other Cardiovascular Medicine

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS