| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
REVIEW ARTICLE |
From the Department of Psychiatry (L.W.) and the Department of Environmental Health, Division of Epidemiology and Biostatistics (B.S.), University of Cincinnati, Cincinnati, Ohio.
Address reprint requests to: Lawson Wulsin, MD, 231 Albert Sabin Way, ML 559, Cincinnati, OH 45267-0559. Email: lawson.wulsin{at}uc.edu
ABSTRACT
OBJECTIVE: The objectives of this study were to systematically review the recent studies of the contribution of depression to the onset of coronary disease and to estimate the magnitude of the risk posed by depression for onset of coronary disease.
METHOD: We searched MEDLINE (1966–2000), PsychInfo (1967–2000), and cross references and conducted informal searches for all community studies of depression symptoms in samples with no clinically apparent heart disease at baseline. From these studies we selected all published cohort studies of 4 years or more follow-up that controlled for other major coronary disease risk factors and reported relative risks (or a comparable measure) of baseline depression for the onset of coronary disease. Following methods for the meta-analysis of epidemiologic studies, we used a random-effects model to estimate the combined overall relative risk.
RESULTS: Ten studies met our inclusion criteria. Relative risks ranged from 0.98 to 3.5. Nine studies reported significantly increased risk, including two with mixed results; one study reported no increased risk. The combined overall relative risk of depression for the onset of coronary disease was 1.64 (95% CI = 1.41–1.90).
CONCLUSIONS: This quantitative review suggests that depressive symptoms contribute a significant independent risk for the onset of coronary disease, a risk (1.64) that is greater than the risk conferred by passive smoking (1.25) but less than the risk conferred by active smoking (2.5). Future prospective community studies should examine the effect of severity and duration of depressive symptoms and disorders on the risk for the onset of coronary disease.
Key Words: coronary disease, • depression, • risk factors, • myocardial ischemia, • cardiovascular disease, • affective disorders.
Abbreviations: CES-D = Center for Epidemiological Studies-Depression Scale;; CI = confidence interval;; MMPI = Minnesota Multiphasic Personality Inventory;; RR = relative risk.
At least 10 controlled observational studies have found depression to contribute an independent risk for new events in the progression of existing coronary disease (1–10) . A number of narrative reviews have emphasized the growing evidence for the adverse effects of depressive symptoms or depressive disorders on specific aspects of existing coronary disease, such as cardiac events, utilization of medical services, level of physical and psychosocial functioning, and cardiovascular mortality (11–17) . In a recent review Musselman et al. (17) described the possible pathophysiological mechanisms by which depression may contribute to the progression of cardiovascular disease. These studies and reviews have begun to describe the contribution of depression to the progression of existing heart disease in general and coronary disease in particular.
A more recent group of studies has emerged that examine depression as a risk factor for the onset or development of coronary disease. Rozanski et al. (18) include most of these studies in their narrative review of five psychosocial factors that contribute to the pathogenesis and expression of coronary disease. Hemingway and Marmot (19) have systematically reviewed the literature on four psychosocial factors that show a possible etiological role in coronary disease, including a factor they call "depression and anxiety." All of the 11 eligible studies in this category reported that the exposure variable significantly increased adjusted relative risks for the development of coronary disease (RR = 1.6–5.4). Barrick (20), in a recent challenge to the Type A behavioral pattern research, selectively reviewed the literature on mood disorders as a risk factor for the pathogenesis of coronary disease. This qualitative review concluded that epidemiologic evidence supports "the notion that mood disorders confer risk for coronary disease, but it is premature to describe it [mood disorders] as a causative factor." Appels (21) also cautions against premature conclusions about a causal relationship for depression leading to coronary disease, citing gaps in the literature on the nature and duration of the depression, the biological mechanisms linking these two disorders, and the effect of antidepressant treatment on the risk for coronary disease.
The aim of this systematic review is to add the first quantitative estimate of the magnitude of the effect of depression on the onset of coronary disease. We addressed the question of how strong the effect of depression is on the development of coronary disease.
Although there is debate about the validity of various methods of meta-analysis of epidemiologic studies (22), there is adequate precedent for the rigorous application to observational studies of valid meta-analytic methods in the coronary disease literature (23–26) . In our selection of studies and data analyses, we followed the methods used to combine epidemiologic studies of smoking and coronary disease (24–26) . As Fliess and Gross (22) have demonstrated, the valid application of meta-analytic methods to epidemiologic studies requires attention to four issues: 1) heterogeneity across studies, 2) possible misclassification of subjects, 3) proper control for covariates or confounding factors, and 4) publication bias.
We confined our focus to coronary disease because the effect of depression on the various forms of heart disease (coronary disease, congestive heart failure, conduction disorders, etc.) and other cardiovascular disorders (hypertension, stroke) may vary with differences in the epidemiology and pathophysiology of these different cardiovascular disease processes. For example, the neurohumoral dysregulation associated with major depression may contribute to the plaque and thrombus formation of coronary disease (17) but have less effect on the loss of cardiac contractility in congestive heart failure from causes other than coronary disease. For conceptual clarity and comparability of measures for the exposure variable, we focused on studies that assess depressive symptoms or depressive disorders, excluding studies of related exposures such as anxiety, fatigue, or stress.
Why do we need a quantitative estimate of the effect of depression on the development of coronary disease? No psychosocial factor is currently recognized as a major risk factor for coronary disease. However, there is ample evidence that depression and coronary artery disease are strongly associated with each other (1–21) . An essential step in building an argument for depression as a major risk factor for coronary disease is the systematic examination of the evidence for and against depression preceding and independently predicting the onset of coronary disease. Quantitative estimates of the magnitude of depression’s effect on incident coronary disease will allow rough comparisons with the established risk factors, such as smoking, hypertension, hyperlipidemias, diabetes, obesity, and physical inactivity.
METHODS
In addition to informal searches and cross-referencing, we performed librarian-assisted MEDLINE (1966–2000) and PsychInfo (1967–2000) searches using multiple terms related to "depression" and "cardiovascular disease." For MEDLINE we crossed the expanded terms "depressive disorder," "bipolar disorder," and "depression" with "myocardial ischemia." For PsychInfo we crossed "depression emotion" and "affective disturbances" with "cardiovascular diseases." PsychInfo searches include dissertations and meeting abstracts that have not appeared as part of full published reports. (For simplicity, in this report the term "depression" includes depressive disorders and depressive symptoms unless otherwise specified.)
We included in our sample (Table 1) all English-language reports of prospective cohort studies that 1) examined community samples without clinically apparent heart disease at baseline, 2) measured depression by a standard self-report questionnaire or structured diagnostic interview, 3) conducted follow-up 4 years or more after baseline, 4) controlled for potential confounders such as coronary disease risk factors that are also related to depression (smoking, physical disease, etc.), and 5) reported a depression group’s relative risk (or measure that could be converted to relative risk) of developing coronary disease. These inclusion criteria were restrictive enough to meet the requirements for valid application of the random-effects model for combining epidemiologic studies but broad enough to include a substantial number of studies.
|
To ensure comparability of methods across studies, we qualitatively assessed the methodologic rigor of the studies included in our sample with respect to four areas: sample size, assessment method, comparison group, and factors controlled for, according to a modified (for this study) version of a published strength-of-evidence rating system (15) (Table 2). From our review of the literature, we selected the four most important confounders of the relationship between depression and coronary disease in addition to age and sex (ie, physical illness, smoking, hypertension, and physical inactivity), because all four are strongly related to both depression and coronary disease. Data on methods reported in Table 1 were selected by the first author and revised by the second author. Minor differences of opinion were all resolved through discussion. Although we evaluated the methods of the studies to organize our discussion of methodologic variations, we chose not to use quality scoring, which weights the contribution of each study to the meta-analysis on the basis of the quality score. The main criticism of incorporating quality scoring weights into meta-analyses is that there are no validated measures of quality and the use of subjective rating scales may lead to bias (27).
|
In combining the individual risk ratio estimates (RRi), we used a random-effects model according to DerSimonian and Laird (35). This model assumes heterogeneity across studies and has been recommended over the fixed-effects model for most meta-analyses of epidemiologic data (22, 26, 36) . There are several sources of heterogeneity among the studies included in this meta-analysis. For example, the 10 studies used 9 different measures of depression, four different definitions of coronary disease outcomes, and a range of follow-up from 4 to 40 years; they also controlled for different sets of confounders. Some studies included men and women and others men only. In addition, the chi-square test, as discussed by Cochran (37), indicated significant heterogeneity among studies, suggesting that the fixed-effects model would not be appropriate. These forms of methodologic heterogeneity are not unique to depression studies; they are similar to those found by van de Mheen and Gunning-Schepers (24) in their meta-analysis of smoking studies and by Booth-Kewley and Friedman (23) in their meta-analysis of psychological predictors of heart disease.
We assumed the natural logarithm of RRi (log RRi) to be asymptotically normal. The estimate for the combined log (RR) is a weighted average of the individual RRi values as follows:
|
|
The weights (wi) are the inverses of the sums of the within-study and the between-study variances. The within-study variance, var (log RRi), was estimated from the log-transformed 95% confidence interval reported for each study. The between-study variance is estimated by the method of moments according to DerSimonian and Laird (35).
To include the Ariyo et al. study (31), which does not report the confidence interval for the highest quartile of the CES-D (>14), we derived the confidence interval around the highest quartile by calculating the standard error from the log of the estimated relative risk for each 5-unit increase (1.15) and then took ln 1.45 ± (1.96 x SE) and then exponentiated that result to get the confidence intervals. We assumed that the standard error would not change over the range of CES-D scores.
We considered using meta-regression to more formally explore the sources of heterogeneity in this meta-analysis. We decided that this technique would not be useful because the small number of studies confer a serious lack of power to show a significant result.
RESULTS
Multiple comprehensive searches resulted in more than 500 citations, of which 10 studies met our inclusion criteria. Table 1 summarizes the methods and results of the 10 studies (28–34, 38–40) published between 1993 and 2000. These studies were comparable with respect to sample size (all >700) and comparison groups; they varied in depression measures (nine different measures), duration of follow-up, and in the confounders controlled for in the analyses. All studies controlled for at least two of the major confounders (physical illness, smoking, hypertension, and physical inactivity), and six studies controlled for three major confounders.
Figure 1 shows the results of the meta-analysis of the risk ratios for the 10 studies. Nine studies reported significantly increased risk (28–34, 39, 40) , including two that reported mixed results (33, 34) . One study was negative (38). Relative risks ranged from 0.98 to 3.5. Table 3 shows that based on the 13 data sets included, the combined overall relative risk of depression leading to the onset of coronary disease was 1.64 (95% CI = 1.41–1.90).
|
|
The possibility of publication bias in this analysis was investigated using a funnel plot and a formal test for publication bias by Begg and Mazumdar (41). Figure 2 shows a funnel plot in which effect size, measured as the log of relative risk (log RR), is plotted against the variance of log RR. A lack of symmetry around the average effect and a lack of studies of high variance and low log RR suggest the presence of publication bias. This impression was strengthened by the results of the formal test, which uses Kendall’s rank correlation procedure to measure the correlation between the variance-weighted average effect size and the variance from the 13 data sets (10 studies). The results show a rank correlation of 0.4 (p = .058). Given the low power of this test when a relatively small number of studies are included, we consider this to be evidence for the presence of publication bias in this meta-analysis.
|
Importance
As a group these 10 studies suggest that depression contributes a significant independent risk (RR = 1.64) for the development of coronary disease in community samples without clinically apparent heart disease. We believe this is the first analysis to quantify the risk posed by depression for the development of coronary disease across the available comparable community studies. The strengths of these studies include their large sample sizes, their selection of subjects without clinically apparent heart disease, their long durations of follow-up, their shared prospective cohort designs, and their controls for multiple major confounders.
The clinical importance of this finding is that it emphasizes the need for randomized, controlled trials of antidepressant treatments in depressed populations at high risk for coronary disease to see if effective antidepressant treatment reduces the risk of incident coronary disease.
Heterogeneity
Among the key issues affecting the validity of meta-analyses of epidemiologic studies, the heterogeneity of studies may be the most important. The heterogeneity of these studies, with respect not only to definitions of the exposure variable but also to definitions of the outcome variable, duration of follow-up, sample composition, and factors controlled for, are both a weakness and a strength in the argument for depression as a risk factor. In our selection of studies, we aimed for a balance between sufficiently restrictive criteria to allow for comparability of studies and sufficient breadth to include an adequate number of studies. We have chosen an analytic method, the random effects model, that takes into account the heterogeneity of the studies. In these ways we have followed the examples of some good meta-analytic studies of smoking and coronary disease. Though our included studies do not share the same definitions of exposure or outcome variables, the same durations of follow-up, or the same age groups at enrollment, the alternative of opting for more rigorous inclusion criteria (eg, shared measure of depression) to reduce heterogeneity would drop the number of included studies to three. On the other hand, if an effect of depression shows up in multiple studies despite variations in methods, it is more likely to be a real effect than an artifact of shared study methods.
The variability in the measures of depression raises the problem of how to interpret the classifications of depression across studies. Though the threshold for depression was set higher in the Pratt et al. (32) and Aromaa et al. (29) studies than in the seven studies that used only self-report measures, the cutoff for being classified as depressed in all 10 studies was consistent with at least mild clinical depression, that is, at least two depressive symptoms of moderate severity for at least 1 week.
The most reliable and clinically useful measures with respect to construct validation in this set of studies are the two structured diagnostic interviews, the Diagnostic Interview Schedule and the Present State Exam, used in the Pratt et al. (32) and Aromaa et al. (29) studies, respectively. The CES-D is the self-report measure in this set of studies with the best psychometric research background for epidemiologic studies and broadest use in clinical research. In a recent community study of the criterion validity of the CES-D (42), a cutoff score of 16 was associated with 100% sensitivity and 88% specificity for major depression by the Diagnostic Interview Schedule. A score of >14 in the Ariyo et al. (30) study is consistent with mild or minor depression.
Some studies commonly cited in the narrative reviews were excluded. Reluctantly we excluded the study by Mendes de Leon (43) because the article reported only relative risks per unit increase on the CES-D, but not relative risks and confidence intervals for the most depressed quartile, making it impossible to combine this relative risk with the relative risks of the other 10 studies. Correspondence with the lead author confirmed that the necessary data were not currently available. Wassertheil-Smoller et al. (44) reported on a sample of patients in a clinical trial for hypertension, not a community sample. The study by Hippisley-Cox et al. (45) used a case-control design of patients in general practice, some of whom had coronary disease at the time of enrollment. Because a full-length published report is not available for the study by Everson et al. (46), we excluded it to ensure that all reports included in our sample could be reviewed by readers. This exclusion represents the one example we know of the publication bias against negative findings in this area. This bias is highlighted by the publication of the related positive report by Everson et al. (47) on hopelessness and the onset of coronary disease in the same Kuopio Study. We also excluded studies that reported coronary disease deaths but not coronary disease incidence.
Measures of Depression
The second important issue, the possible misclassification of subjects, concerns primarily the quality of the measures of the exposure variable depression. This group of studies used nine different measures of depression, seven of which were self-report measures. Self-report measures are the least specific type of measure of depression and the most susceptible to misclassifying subjects. That is, people with other psychiatric disorders may score high on self-report depression measures. Structured diagnostic interviews are the most rigorous and specific method of assessment. However, self-report measures have been the standard method for assessing psychological symptoms in psychiatric epidemiology research for the past 20 years because of their ease of administration and scoring. The two studies that used the structured diagnostic interview method (29, 32) found high relative risks, whereas the one negative study (38) used a self-report measure of depression developed for the authors’ study. These observations suggest that the method of assessing depression may affect the risk estimate by either overestimating or underestimating the true risk. We know of no way of correcting for this limitation in the existing studies. The use of structured diagnostic interviews in addition to self-report measures in future studies would raise the standard of measurement and reduce the risk of misclassification.
Frequency of assessment and duration of follow-up may also affect classification of depression because depressive symptoms and disorders can be brief (days), recurrent, or chronic. In seven studies a depression measure was administered only once at baseline; in three studies the depression measure was repeated at least once. Five studies had follow-up periods of 10 years or less, and one had follow-up as long as 40 years. However, the five with short follow-up periods do not differ significantly from the five long follow-up studies with respect to combined relative risks (combined RR = 1.49, 95% CI =1.36–1.62 vs. RR = 1.66, 95% CI = 1.41–1.95, respectively). Thus it is impossible to infer from this group of studies what contribution the duration of depression makes to the risk of coronary disease.
Controls
The methods for controlling for differences in clinical populations within each study were roughly comparable across studies. Table 1 shows the differences across studies in the specific factors that were controlled. All studies controlled for at least two major confounders in addition to other factors; most controlled for three. Penninx et al. (34) and Mendes de Leon et al. (43) note the strong impact of controlling for physical disability in the elderly, which reduced the relative risk from significant to insignificant in at least one calculation in each study. It is likely that the strength of the relationship between depression and coronary disease becomes overshadowed by physical disability in the elderly. Alternatively, it is possible that coronary disease is the underlying cause of both the depression and the disability, a possibility proposed by Alexopoulos et al. (48) and others.
Because studies varied in the combination of confounding factors controlled for, it is also possible that in any given study some of the reported effect of depression on risk for coronary disease is in fact contributed by factors that were not assessed and controlled for (eg, cholesterol, obesity, alcohol, etc.).
Publication Bias
Though publication bias affects all meta-analyses, usually by suppressing negative studies, we are aware of only one example of a relevant abstract that has not been published as a full-length report (46). In an earlier calculation for the current study, when we included this study, the combined overall relative risk was not significantly different from the current value. Our examination of the relationship between relative risks and variances (Figure 2) supports the presence of a publication bias in favor of positive studies. However, using the Rosenthal (49) method for estimating the "file drawer" problem, we calculated that 572 unpublished negative studies would have to exist in the mythical file drawer to negate the reported combined relative risk from these 10 studies. The presence of a publication bias does not undermine the value of these studies.
Limitations
Thus, the limitations of this report include the heterogeneity of methods of the studies, the limitations of the measures of depression, and the possible publication bias. We have also limited our sample to English-language reports, which may have excluded some relevant studies.
Because these studies focus only on community samples, it is not yet clear what magnitude of risk for coronary disease depression may contribute in clinical samples, such as a primary care sample. These limitations justify a cautious interpretation of our findings.
Effect Size
More than 200 risk factors for heart disease have been identified (50), including various forms of stress and distress (23, 51) . None of the 10 studies reviewed here compared the effect of depression to the relative risks of other risk factors for coronary disease. Van de Mheen and Gunning-Schepers (24) reviewed 83 smoking studies and found that smoking conferred relative risks for coronary disease that ranged from 1.2 to 3.0, similar to the range for depression (0.98–3.5). They attributed most of the variation in the risk of smoking to variation in study methods. The best meta-analysis of the risk of smoking on incident coronary disease, the National Cooperative Pooling Project (25), estimated that smoking confers a standard incidence ratio of 2.5 for the development of coronary disease. Using methods similar to ours, He et al. (26) found that passive smoking conferred a risk for developing coronary disease of 1.24 (95% CI = 1.17–1.32). Our finding that depression predicts coronary disease with an effect size greater than passive smoking but less than active smoking also adds an important piece to the growing argument for the role of depression as a risk factor for coronary disease.
Future Research
These studies, all emerging over the last 10 years, point first to the need for more comparative longitudinal studies of the relative risks of multiple cardiac risk factors, including depression, to better estimate the relative importance of depression’s contribution to the onset of coronary disease. In particular, future community studies should assess depression in a repeated-measures design over 5 to 10 years, by diagnostic interview as well as self-report, to better describe the types of depressive disorders and the effects of duration and severity of depressive symptoms on the development of coronary disease.
Although these 10 studies suggest that depression may play a major role as a risk factor for the development of coronary disease, they do not tell us how important this role is relative to other major risk factors, nor how depression might exert its effect. The mechanisms by which depression increases the risk of coronary disease are likely to be both direct and indirect, both independent and interactive (14, 17) . The focus of these 10 studies on depression as an independent exposure helps to isolate the effect of depression, but it may also underestimate the effect of depression by ignoring interaction effects. As Rozanski et al. (18) have pointed out, depression often coexists with other psychosocial factors that contribute to the risk for developing coronary disease, such as anxiety, hostile personality traits, social isolation, and chronic stress. In addition, at the outcome end of the equation, depression has been shown to contribute to cardiovascular disease processes that are related to the onset and progression of coronary disease, such as hypertension (52) and stroke (53). With more rigorous prospective community studies of depression in the context of other psychosocial and coronary disease risk factors, we will understand better what role depression plays in the complex interactions that contribute to the development of coronary disease.
ACKNOWLEDGMENTS
Thanks to Robert Carney, PhD, for his helpful suggestions on revising the manuscript; to Jeff Welge, PhD, for his helpful statistical review; and to William Nienaber and Jeff Welge, PhD, for their design of Figure 1.
NOTES
Presented in part at the Annual Meeting of the American Psychiatric Association, Chicago, Illinois, May 15, 2000.
Received for publication August 7, 2001.
REFERENCES
This article has been cited by other articles:
 |
|
 |
|
  C. Pizzi, L. Manzoli, S. Mancini, and G. M. Costa Analysis of potential predictors of depression among coronary heart disease risk factors including heart rate variability, markers of inflammation, and endothelial function Eur. Heart J., May 1, 2008; 29(9): 1110 - 1117. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. Terre Behavioral Medecine Review: Psychological Risks for Metabolic Syndrome American Journal of Lifestyle Medicine, April 1, 2008; 2(2): 126 - 129. [Abstract] [PDF]
|
|
|
|
 |
|
 P. G. Surtees, N. W.J. Wainwright, R. N. Luben, N. J. Wareham, S. A. Bingham, and K.-T. Khaw Depression and Ischemic Heart Disease Mortality: Evidence From the EPIC-Norfolk United Kingdom Prospective Cohort Study Am J Psychiatry, April 1, 2008; 165(4): 515 - 523. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rutledge, S. E. Linke, M. B. Olson, J. Francis, B. D. Johnson, V. Bittner, K. York, C. McClure, S. F. Kelsey, S. E. Reis, et al. Social Networks and Incident Stroke Among Women With Suspected Myocardial Ischemia Psychosom Med, April 1, 2008; 70(3): 282 - 287. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R Rugulies, B Aust, H Burr, and U Bultmann Job insecurity, chances on the labour market and decline in self-rated health in a representative sample of the Danish workforce J. Epidemiol. Community Health, March 1, 2008; 62(3): 245 - 250. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Ohira, A. V. D. Roux, R. J. Prineas, M. A. Kizilbash, M. R. Carnethon, and A. R. Folsom Associations of Psychosocial Factors With Heart Rate and Its Short-Term Variability: Multi-Ethnic Study of Atherosclerosis Psychosom Med, February 1, 2008; 70(2): 141 - 146. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. C. Stewart, D. Janicki-Deverts, M. F. Muldoon, and T. W. Kamarck Depressive Symptoms Moderate the Influence of Hostility on Serum Interleukin-6 and C-Reactive Protein Psychosom Med, February 1, 2008; 70(2): 197 - 204. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Vaccarino, C. McClure, B. D. Johnson, D. S. Sheps, V. Bittner, T. Rutledge, L. J. Shaw, G. Sopko, M. B. Olson, D. S. Krantz, et al. Depression, the Metabolic Syndrome and Cardiovascular Risk Psychosom Med, January 1, 2008; 70(1): 40 - 48. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Vaccarino, B. D. Johnson, D. S. Sheps, S. E. Reis, S. F. Kelsey, V. Bittner, T. Rutledge, L. J. Shaw, G. Sopko, and C. N. Bairey Merz Depression, Inflammation, and Incident Cardiovascular Disease in Women With Suspected Coronary Ischemia: The National Heart, Lung, and Blood Institute Sponsored WISE Study J. Am. Coll. Cardiol., November 20, 2007; 50(21): 2044 - 2050. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Dickens, L. McGowan, C. Percival, B. Tomenson, L. Cotter, A. Heagerty, and F. Creed Depression Is a Risk Factor for Mortality After Myocardial Infarction: Fact or Artifact? J. Am. Coll. Cardiol., May 8, 2007; 49(18): 1834 - 1840. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. Terre Behavioral Medecine Review: The Lifestyle Factor American Journal of Lifestyle Medicine, May 1, 2007; 1(3): 181 - 184. [Abstract] [PDF]
|
|
|
|
 |
|
 F. N. Jacka, J. A. Pasco, S. McConnell, L. J. Williams, M. A. Kotowicz, G. C. Nicholson, and M. Berk Self-Reported Depression and Cardiovascular Risk Factors in a Community Sample of Women Psychosomatics, February 1, 2007; 48(1): 54 - 59. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. Nicholson, H. Kuper, and H. Hemingway Depression as an aetiologic and prognostic factor in coronary heart disease: a meta-analysis of 6362 events among 146 538 participants in 54 observational studies Eur. Heart J., December 1, 2006; 27(23): 2763 - 2774. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
R. J. Contrada, D. A. Boulifard, E. L. Idler, T. J. Krause, and E. W. Labouvie Course of Depressive Symptoms in Patients Undergoing Heart Surgery: Confirmatory Analysis of the Factor Pattern and Latent Mean Structure of the Center for Epidemiologic Studies Depression Scale Psychosom Med, November 1, 2006; 68(6): 922 - 930. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. Rutledge, V. A. Reis, S. E. Linke, B. H. Greenberg, and P. J. Mills Depression in Heart Failure: A Meta-Analytic Review of Prevalence, Intervention Effects, and Associations With Clinical Outcomes J. Am. Coll. Cardiol., October 17, 2006; 48(8): 1527 - 1537. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 C. M. DICKENS, L. McGOWAN, C. PERCIVAL, B. TOMENSON, L. COTTER, A. HEAGERTY, and F. H. CREED Contribution of depression and anxiety to impaired health-related quality of life following first myocardial infarction The British Journal of Psychiatry, October 1, 2006; 189(4): 367 - 372. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
O. L. Solis, D. A. Khan, and E. S. Brown Age at Onset of Major Depression in Inner-City Adults With Asthma Psychosomatics, August 1, 2006; 47(4): 330 - 332. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 T. Rutledge, S. E. Reis, M. B. Olson, J. Owens, S. F. Kelsey, C. J. Pepine, S. Mankad, W. J. Rogers, C. N. B. Merz, G. Sopko, et al. Depression Symptom Severity and Reported Treatment History in the Prediction of Cardiac Risk in Women With Suspected Myocardial Ischemia: The NHLBI-Sponsored WISE Study. Arch Gen Psychiatry, August 1, 2006; 63(8): 874 - 880. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. V. Diez Roux, N. Ranjit, L. Powell, S. Jackson, T. T. Lewis, S. Shea, and C. Wu Psychosocial factors and coronary calcium in adults without clinical cardiovascular disease. Ann Intern Med, June 6, 2006; 144(11): 822 - 831. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. Rugulies, U. Bultmann, B. Aust, and H. Burr Psychosocial Work Environment and Incidence of Severe Depressive Symptoms: Prospective Findings from a 5-Year Follow-up of the Danish Work Environment Cohort Study Am. J. Epidemiol., May 15, 2006; 163(10): 877 - 887. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Y. Aina and J. L. Susman Understanding Comorbidity With Depression and Anxiety Disorders J Am Osteopath Assoc, May 1, 2006; 106(5_suppl_2): S9 - S14. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. A. Hayden, P. Cote, and C. Bombardier Evaluation of the quality of prognosis studies in systematic reviews. Ann Intern Med, March 21, 2006; 144(6): 427 - 437. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Hilton, G. Parker, S. McDonald, G. A. Heruc, A. Olley, H. Brotchie, C. Friend, and W. F. Walsh A Validation Study of Two Brief Measures of Depression in the Cardiac Population: The DMI-10 and DMI-18. Psychosomatics, March 1, 2006; 47(2): 129 - 135. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. Bunde and R. Martin Depression and Prehospital Delay in the Context of Myocardial Infarction Psychosom Med, January 1, 2006; 68(1): 51 - 57. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Dickens, L. McGowan, C. Percival, J. Douglas, B. Tomenson, L. Cotter, A. Heagerty, and F. Creed Association Between Depressive Episode Before First Myocardial Infarction and Worse Cardiac Failure Following Infarction Psychosomatics, December 1, 2005; 46(6): 523 - 528. [Abstract] [Full Text] [PDF] |
