Predictive Value of Social Inhibition and Negative Affectivity for Cardiovascular Events and Mortality in Patients With Coronary Artery Disease: The Type D Personality Construct JOHAN DENOLLET, PHD, SUSANNE S. PEDERSEN, PHD, CHRISTIAAN J. VRINTS, MD, PHD,

AND

VIVIANE M. CONRAADS, MD, PHD

Objective: Methodological considerations and selected null findings indicate the need to reexamine the Type D construct. We investigated whether associations with cardiac events in patients with coronary artery disease (CAD) involve the specific combination of negative affectivity (NA) and social inhibition (SI), or rather the main effect of either trait alone. Methods: In this 5-year follow-up of 541 patients with CAD, the Type D construct (14-item Type D Scale) was tested by examining a) the interaction of continuous NA and SI z scores and b) a four-group classification defined by low/high trait scores. End points were major adverse cardiac events (MACEs; death, myocardial infarction [MI], coronary revascularization) and cardiac death/MI. Results: At follow-up, 113 patients had a MACE, including 47 patients with cardiac death/MI. After adjustment for disease severity, age, sex, and main trait effects, the interaction of NA and SI z scores was associated with MACE (odds ratio [OR] = 1.36, 95% confidence interval [CI] = 1.11Y1.67). This continuous measure of Type D was also associated with cardiac death/MI (OR = 1.48, 95% CI = 1.11Y1.96) and remained an independent predictor of events after adjustment for depressive symptoms. Using a cutoff of 10 on both NA and SI scales, Type D was associated with an adjusted OR of 1.74 (95% CI = 1.11Y2.73) for MACE and an OR of 2.35 (95% CI = 1.26Y4.38) for death/MI but was unrelated to noncardiac death. Patients with high NA or SI alone were not at increased risk. Conclusions: Continuous (NA  SI interaction) and dichotomized measures of Type D were associated with cardiovascular events in patients with CAD. Research is needed to explore moderating factors that may alter this association. Key words: coronary artery disease, type D personality, prognosis, adverse cardiac events.

BDI = Beck Depression Inventory; CABG = coronary bypass surgery; CAD = coronary artery disease; DS14 = 14-item Type D Scale; LVEF = left ventricular ejection fraction; MACE = major adverse cardiac events; MI = myocardial infarction; NA = negative affectivity; PCI = percutaneous coronary intervention; SI = social inhibition.

INTRODUCTION sychosocial factors have long been known to be associated with cardiovascular outcomes (1), and both depression (2) and anxiety (3) have been related to coronary artery disease (CAD). However, patients may go in and out of depressive or anxiety episodes, and people differ in their propensity to these mood disorders (4,5). The study of personality (4) is a complementary approach that may improve our understanding of individual differences in psychosocial risk. Negative affectivity (the tendency to experience negative emotions) and social inhibition (the tendency to inhibit self-expression) are two personality traits that are relevant in this context (5Y8). The distressed (Type D) personality model refers to the combined effect of social inhibition and negative affectivity and was introduced in 1996 (5) to study the cardiovascular effects of chronic psychosocial distress accumulating across an individual’s life. This model holds that a) negative emotions tend to cluster together within individuals (9) and that b) social inhibition modulates the effect of negative emotions on cardiac prognosis (8).

P

From CoRPS - Center of Research on Psychology in Somatic diseases (J.D., S.S.P.), Tilburg University, Tilburg, the Netherlands; Department of Cardiology (J.D., C.J.V., V.M.C.), University Hospital Antwerp, Antwerp, Belgium; and Department of Cardiology (S.S.P.), Odense University Hospital & Institute of Psychology, Southern University of Denmark, Odense, Denmark. Address correspondence and reprint requests to Johan Denollet, PhD, Center of Research on Psychology in Somatic Diseases, Department of Medical and Clinical Psychology. Tilburg University, PO Box 90153, 5000 LE Tilburg, the Netherlands. E-mail: [email protected]. Received for publication September 27, 2012; revision received August 8, 2013. DOI: 10.1097/PSY.0000000000000001

The threat of negative reactions from others is a main source of stress (10), particularly in inhibited individuals (11). Evidence suggests that Type D may be associated with a more than two-fold increased risk of adverse events in cardiac patients (12Y15) and that dysregulations of the neuroendocrine, autonomic nervous, and immune systems are pathways that may explain this association (12). However, there are also null studies on Type D personality. In 2010, we published a study that found no association between Type D and all-cause mortality in patients with heart failure (16), and these null findings were replicated in another heart failure study (17). Type D was also unrelated to all-cause mortality in a sample of cardiac patients with significant somatic comorbidity (18) and in patients treated with percutaneous coronary intervention (PCI) (19). In addition to these null findings, other criticisms about Type D research include the use of cutoff scores and the preponderance of work by the original research group (20,21). However, most important, research needs to test the conceptually unique prediction (20) that it is the combination of social inhibition and negative affectivity that predicts cardiovascular events and not only the separate effect of one of these traits alone or of other negative emotions. Therefore, we wanted to test the validity of the Type D prediction model of the combined versus separate social inhibition and negative affectivity effects in patients with CAD. In addition, we wanted to compare the predictive value of Type D and depressive symptoms. METHODS Patients In total, 541 patients with CAD (87% men; mean [standard deviation {SD}] = 58.7 [10.5] years) from the Antwerp University Hospital, Belgium, were included between January 1998 and December 2005; data from this cohort have not been published previously. This study included 248 patients with myocardial infarction (MI), 239 patients with coronary bypass surgery (CABG),

Psychosomatic Medicine 75:873Y881 (2013) 0033-3174/13/7509Y0873 Copyright * 2013 by the American Psychosomatic Society

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

873

J. DENOLLET et al. and 233 patients on PCI, with some patients experiencing both an acute MI and subsequently undergoing revascularization. Patients with cancer or other life-threatening medical conditions were excluded. At baseline, all patients had an exercise stress test and completed psychological questionnaires at enrollment in an outpatient rehabilitation program. Patients provided written informed consent, and the study was approved by the Medical Ethics Committee of the University Hospital of Antwerp, Belgium (protocol no. 5/48/193). End Points The follow-up interval was fixed at 5 years. As in previous studies (5Y7), patients and their families were contacted by telephone to determine the end points. Information on mortality, MI, and CABG/PCI were extracted from hospital records, and the patient’s attending physician was involved in the classification of cause of death. The first end point was major adverse cardiac events (MACEs; a composite of death, MI, and CABG or PCI). The second end point was a composite of cardiac death and MI as a more rigorous measure of cardiac prognosis. Clinical Covariates The covariates for statistical analyses were chosen because they have been shown to predict cardiac events (22,23). Left ventricular ejection fraction (LVEF) and exercise tolerance have previously been related to cardiac prognosis (5Y7). Because borderline decreased LVEF is also associated with poor prognosis (22), decreased systolic function was defined as LVEF of 50% or less (6). Cardiorespiratory fitness has been shown to explain the association of depression with poor prognosis in a recent study of patients with CAD (2). As described previously (6), poor exercise tolerance was defined by a median split for peak work load on a symptom-limited exercise test (e140 and e120 W for men aged e55/Q56 years, and e100 and e80 W for women aged e55/ Q56 years). We also controlled for age, sex, MI at baseline, and CABG or PCI at baseline. Type D Personality The 14-item Type D Scale (DS14) was used to assess negative affectivity, social inhibition, and Type D personality (24). Both the negative affectivity (> = .88) and social inhibition (> = .86) subscales of the DS14 are internally consistent and stable over time (24). The robustness of the DS14 measure of negative affectivity and social inhibition was confirmed in studies that used item response theory (25), Mokken scale analysis (26), and confirmatory factor analysis (26,27). In addition, the two-factor structure and reliability of the DS14 was also supported in a large international study among cardiac patients from 21 different countries across the world (27). The negative affectivity subscale includes the facets dysphoria (three items; e.g., ‘‘often feels unhappy’’), anxiety (two items; e.g., ‘‘often worries about something’’), and irritability (two items; e.g., ‘‘is often irritated’’); social inhibition includes social discomfort (three items; e.g., ‘‘often feels inhibited in social interactions’’), reticence (two items; e.g., ‘‘closed kind of person’’), and lack of social poise (two items, reversed; e.g., ‘‘makes contact 874

easily when meeting people’’) (24). At the intermediate level, negative affectivity/social inhibition is assessed as continuous dimensions, and Type D refers to the combination of these traits at a superordinate level. A recent multimethod analysis of the DS14 supported this three-level hierarchical model (26). In the current study, the mean (SD) scores of the negative affectivity (> = .87; skewness = 0.57) and social inhibition (> = .87; skewness = 0.41) subscales were 9.03 (6.32) and 9.78 (6.35), respectively. The correlation between both subscales was 0.35, indicating 12% shared variance. Depressive Symptoms The 10-item Beck Depression Inventory (BDI10) was used to assess depressive symptoms (28), including sadness, pessimism, sense of failure, self-dislike, negative body-image, suicidal ideas, dissatisfaction, indecisiveness, work inhibition, and fatigability. The BDI10 is a reliable (> = .82) and valid (r = 0.89 with standard BDI) self-report measure that has been shown to predict death and MI in patients with CAD, adjusting for disease severity (28). Because depression and Type D may represent different forms of emotional distress in patients with CAD (29), the BDI10 was included to examine the extent to which the Type D construct predicted cardiovascular events after adjustment for depressive symptoms in this study. The mean (SD) score of the BDI10 measure of depression was 3.66 (3.28; range 0Y18). Statistical Analyses To compare findings from the present study with previously published Type D reports (5Y7), multiple logistic regression analyses were used to examine the prognostic validity of this personality construct. On the basis of these reports (5Y7), we a priori selected demographics (age, sex), disease severity (LVEF, exercise tolerance, index MI), and invasive treatment (CABG, PCI) at baseline to be included as covariates that could explain the link between Type D and clinical events in these regression models. We used additional logistic regression analyses to explore whether standard risk factors (hypertension, hyperlipidemia, smoking) or medical treatment (aspirin, A-blocker, angiotensinconverting enzyme inhibitor) were associated with clinical events. To examine the robustness of the Type D model, we applied 2 different analytical approaches. We first analyzed the prediction of events from the interaction of continuous social inhibition and negative affectivity z scores (normalized scores with a mean [SD] = 0 [1]) (16,20), while simultaneously entering the separate main effects of these traits in the regression analysis. Next, we classified patients in personality categories to compare the separate and combined effects of high and low trait levels (8). The previously established cutoff score of 10 higher on the social inhibition and negative affectivity scales of the DS14 (24,25) was used to define four categories: a) ‘‘low negative affectivity/ low inhibition,’’ b) ‘‘low negative affectivity/high inhibition,’’ c) ‘‘high negative affectivity/low inhibition,’’ and d) ‘‘high negative affectivity/high inhibition’’ (Type D). Analyses were performed using the Statistical Package for Social Sciences 19.0 for Windows. Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

TYPE D AND ADVERSE EVENTS IN PATIENTS WITH CAD RESULTS At follow-up, 113 patients (21%) had experienced at least one event (all-cause death: n = 41; MI: n = 25; CABG: n = 8; PCI: n = 50). All deaths were attributable to natural causes, with 22 deaths being related to cardiac causes. Hence, 47 patients (9%) had a cardiac death or MI. Predictive Power of Statistical Covariates We first analyzed the association between covariates (age, sex, disease severity as indicated by LVEF, exercise tolerance and index MI, and CABG/PCI at baseline) and end points. At baseline, patients with MACE had lower levels of LVEF (odds ratio [OR] = 0.98), were more likely to have a poor exercise tolerance (OR = 1.72), and were less likely to have been treated with CABG (OR = 0.43) compared with patients with event-free survival (Table 1). Lower levels of LVEF (OR = 0.96) and less CABG treatment (OR = 0.32) were associated with cardiac death/ MI. These findings confirmed the validity of these covariates that were selected a priori to control for in adjusted analyses. Post hoc analyses showed that other clinical variables (aspirin, A-blockers, angiotensin-converting enzyme inhibitors, hypertension, hyperlipidemia, smoking) were not associated with MACE or death/MI. Synergistic Interaction of Type D Components The unadjusted z scores of negative affectivity (OR = 1.11, 95% confidence interval [CI] = 0.90Y1.36, p = .32) and social inhibition (OR = 1.14, 95% CI = 0.93Y1.40, p = .21) were not associated with MACE. When simultaneously entered in a multivariable regression model, the main effect of the individual negative affectivity (p = .94) or social inhibition (p = .20) z score was not significantly related to MACE (Table 2, Model A). Their main effect was also not related to cardiac death/MI. Consistent with the conceptually unique prediction of the Type D model, the synergistic interaction of continuous negative affectivity and social inhibition z scores was associated with MACE in univariate analysis (OR = 1.36, 95% CI = 1.13Y1.65, p = .001) and also remained significant after adjustment for covariates and the main trait effects (Table 2, Model B). Once again, negative affectivity and social inhibition were not significantly

related to MACE, whereas their interaction was independently associated with MACE (OR = 1.36, p = .003). This interaction of traits also contributed to the prediction of cardiac death/MI (OR = 1.48), but was not significantly associated with all-cause mortality (OR = 1.23, 95% CI = 0.88-1.72, p = .12). Depressive symptoms were significantly associated with MACE (OR = 1.09, 95% CI = 1.03Y1.16, p = .004) and cardiac death/MI (OR = 1.12, 95% CI = 1.03Y1.21, p = .006) in univariate analysis. The continuous Type D score ( p = .004) and depressive symptoms ( p = .045) independently predicted MACE in multivariable analysis (Table 3, left). The continuous Type D score also predicted cardiac death/MI ( p = .008), adjusting for depressive symptoms (Table 3, right). Overall, these analyses of continuous scores indicated the robustness of the Type D model and the usefulness of stratifying patients in distinct negative affectivity/ inhibition categories. Classification in Personality Categories Next, patients were classified in four categories based on the cutoff score of 10 or higher, that is, low on both traits, social inhibition only, negative affectivity only, and Type D (Table 4). This four-group classification was significantly associated with the risk of MACE (Fig. 1, top). Comparison of separate effects of high trait levels regarding the frequency of MACE showed that patients with negative affectivity only (11/82 [13%]; OR = 0.54, 95% CI = 0.28Y1.06) or social inhibition only (18/110 [16%]; OR = 0.69, 95% CI = 0.40Y1.20) did not differ significantly from patients with low levels of both traits (39/192 [20%]). Type D patients had the highest frequency of MACE (45/157 [29%]; OR = 1.87, 95% CI = 1.21-2.88). They also had a higher frequency of cardiac death/MI (23/157 [15%]) than did those with negative affectivity (3/82 [4%]) or social inhibition (5/110 [5%]) only, or with low levels of both traits (16/192 [8%]; p = .007; Fig. 1, bottom). All-cause mortality did not differ significantly between the reference (15/192 [8%]), social inhibition (6/110 [6%]), negative affectivity (2/82 [2%]), and Type D (18/157 [12%]) groups ( p = .066). Noncardiac mortality also did not differ between these personality groups ( p = .56). Cardiac mortality, however, differed

TABLE 1. Model of Statistical Covariates Included in the Studya MACE (n = 113) Covariates

n (%)

Age (m [SD] = 58.6 [10.5] y)

OR

95% CI

Cardiac Death/MI (n = 47) p

OR

95% CI

p

V

0.99

0.98Y1.02

.73

1.00

0.97Y1.03

.91

Male sex

473 (87)

1.08

0.55Y2.13

.82

1.20

0.43Y3.38

.73

Index MI at baseline

248 (46)

0.83

0.50Y1.38

.48

0.64

0.30Y1.33

.23

V

0.98

0.96Y0.99

.005

0.96

0.94Y0.98

.0001

Poor exercise toleranceb

194 (36)

1.72

1.09Y2.72

.020

1.30

0.66Y2.56

.44

CABG at baseline

239 (44)

0.43

0.23Y0.78

.005

0.32

0.13Y0.80

.015

PCI at baseline

233 (43)

0.90

0.52Y1.56

.70

0.73

0.33Y1.62

.44

LVEF (M [SD] = 52.7% [14.5%])

MACE = major adverse cardiac event; MI = myocardial infarction; OR = odds ratio; CI = confidence interval; M = mean; SD = standard deviation; LVEF = left ventricular ejection fraction; CABG = coronary artery bypass surgery; PCI = percutaneous coronary intervention. a Multivariable logistic regression model with all variables entered simultaneously. b For younger/older men, e140/e120 W; for younger/older women, e100/e80 W (6). Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

875

J. DENOLLET et al. TABLE 2. Multivariable Models of MACE and Cardiac Death/MI Including Main NA and SI Effects Only (Model A) and NA  SI Interaction Effect (Model B), Adjusting for Covariatesa MACE (n = 113) Characteristics

Cardiac Death/MI (n = 47)

OR

95% CI

p

OR

95% CI

p

0.99 1.16

0.78Y1.26 0.92Y1.47

.94 .20

1.11 1.02

0.78Y1.58 0.72Y1.45

.56 .91

Model A: NA and SI main effect Type D scale scores NA (z score) SI (z score) Covariates Age

0.99

0.98Y1.02

.74

1.00

0.97Y1.04

.83

Male sex

1.07

0.54Y2.14

.85

1.28

0.44Y3.71

.65

Index MI at baseline

0.82

0.49Y1.36

.44

0.65

0.31Y1.37

.26

Left ventricular ejection fraction

0.98

0.96Y0.99

.006

0.96

0.94Y0.98

.0001

Poor exercise toleranceb

1.75

1.10Y2.76

.017

1.30

0.66Y2.57

.44

CABG at baseline PCI at baseline

0.42 0.89

0.23Y0.77 0.51Y1.56

.005 .69

0.32 0.71

0.13Y0.81 0.32Y1.59

.017 .40

NA  SI interaction (z scores)

1.36

1.11Y1.67

.003

1.48

1.11Y1.96

.007

NA (z score)

0.92

0.71Y1.18

.51

1.04

0.71Y1.51

.86

SI (z score)

1.13

0.89Y1.45

.32

0.93

0.64Y1.36

.71

0.99 0.99

0.97Y1.02 0.50Y2.00

.71 .99

1.00 1.14

0.97Y1.04 0.39Y3.30

.82 .82

Model B: NA  SI interaction Type D scale scores

Covariates Age Male sex Index MI at baseline

0.81

0.48Y1.35

.42

0.64

0.30Y1.37

.25

Left ventricular ejection fraction

0.98

0.96Y0.99

.004

0.96

0.93Y0.98

.0001

Poor exercise toleranceb

1.74

1.09Y2.77

.020

1.26

0.63Y2.52

.51

CABG at baseline

0.42

0.23Y0.78

.006

0.32

0.13Y0.82

.017

PCI at baseline

0.91

0.52Y1.59

.74

0.74

0.33Y1.68

.48

MACE = major adverse cardiac event; MI = myocardial infarction; NA = negative affectivity; SI = social inhibition; OR = odds ratio; CI = confidence interval; CABG = coronary artery bypass surgery; PCI = percutaneous coronary intervention. a Multivariable logistic regression models with all variables entered simultaneously. b For younger/older men, e140/e120 W; for younger/older women, e100/e80 W (6). TABLE 3. Multivariable Models of Cardiovascular Outcome Including Continuous Scores of Type D (NA  SI) and Depressive Symptoms, Adjusting for Main NA and SI Effects and Cardiac Covariatesa MACE (n = 113) Baseline Characteristics

OR

95% CI

Cardiac Death/MI (n = 47) p

OR

95% CI

p

Type D and depression NA  SI interaction (DS14) Depressive symptoms (BDI10) Main NA and SI effects

1.35

1.10Y1.66

.004

1.46

1.11Y1.93

.008

1.09

1.01Y1.18

.045

1.08

0.97Y1.21

.16

Negative affectivity (DS14)

0.78

0.58Y1.05

.10

0.86

0.56Y1.33

.49

Social inhibition (DS14)

1.14

0.90Y1.46

.28

0.95

0.65Y1.39

.78

Cardiac covariates Index MI at baseline

0.76

0.47Y1.23

.27

0.55

0.27Y1.10

.091

Left ventricular ejection fraction

0.98

0.96Y0.99

.006

0.96

0.93Y0.98

.0001

Poor exercise toleranceb

1.67

1.05Y2.64

.029

1.27

0.64Y2.50

.49

CABG at baseline

0.42

0.24Y0.73

.002

0.36

0.15Y0.86

.021

NA = negative affectivity; SI = social inhibition; MACE = major adverse cardiac event; MI = myocardial infarction; OR = odds ratio; CI = confidence interval; DS14 = 14-item Type D Scale (24); BDI10 = 10-item Beck Depression Inventory (28); CABG = coronary artery bypass surgery. a Multivariable logistic regression models with all variables entered simultaneously. b For younger/older men, e140/e120 W; for younger/older women, e100/e80 W (6). 876

Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

TYPE D AND ADVERSE EVENTS IN PATIENTS WITH CAD TABLE 4. Classification in Four Categories Based on the Cutoff Score Q10 on the NA and SI Subscales of the DS14 DS14 Subscales

Reference Group (NAj/SIj), n = 192

SI (NAj/SI+), n = 110

NA (NA+/SIj), n = 82

Type D Personality (NA+/SI+), n = 157

NA, M (SD)

4.09 (2.79)

4.69 (2.45)

14.28 (3.84)

15.37 (4.33)

SI, M (SD)

4.29 (2.77)

14.42 (3.87)

5.10 (2.57)

15.67 (4.16)

NA = negative affectivity; SI = social inhibition; DS14 = 14-item Type D Scale; NAj = negative affectivity e9; SIj = social inhibition e9; NA+ =negative affectivity Q10; SI+ = social inhibition Q10 (as defined by the previously published cutoff scores on the DS14 subscales (24,25)); M = mean for the raw continuous subscale scores; SD = standard deviation.

between the reference (8/192 [4%]), social inhibition (1/110 [1%]), negative affectivity (1/82 [1%]), and Type D (12/157 [8%]) groups (p = .015). Critically, Type D was associated with an OR of 2.59 for MACE (95% CI = 1.26Y5.35) and an OR of 4.52 (95%

CI = 1.32Y15.54) for cardiac death/MI as compared to negative affectivity only. Type D Versus NonYType D Dichotomy Pooling patients with negative affectivity only, social inhibition only or low trait levels in one group of nonYType D patients yielded the predicted association of Type D with higher frequencies of both MACE (p = .004) and cardiac death/MI (p = .002), respectively. In adjusted analyses, Type D was associated with an OR of 1.74 for MACE (Table 5); other independent predictors were markers of disease severity (poor systolic function: OR = 1.62; poor exercise tolerance: OR = 1.86) and no CABG at baseline (OR = 2.19). Type D was also associated with an adjusted OR of 2.35 for cardiac death/MI (Table 5). Finally, Type D and nonYType D patients had similar frequencies of noncardiac death (6/157 [4%] and 13/384 [3%], p = .80), but Type D patients had an increased risk of cardiac death (12/157 [8%] versus 10/384 [3%], p = .007). DISCUSSION The findings of the present study showed that it was the specific combination of being both emotionally distressed and inhibited, as defined by the Type D model, that was associated TABLE 5. Association of Type D Personality With MACE and Cardiac Death/MIa OR

95% CI

p

Type D personalityb Decreased systolic functionc

1.74 1.62

1.11Y2.73 1.02Y2.59

.016 .042

Poor exercise toleranced

1.86

1.19Y2.91

.006

No CABG at baselinee

2.19

1.34Y3.58

.002

Type D personalityb

2.35

1.26Y4.38

.007

Decreased systolic functionc

2.10

1.09Y4.05

.027

Poor exercise toleranced

1.81

0.96Y3.42

.068

No CABG at baselinee

2.85

1.29Y6.29

.010

MACE (n = 113)

Cardiac death/MI (n = 47)

Figure 1. MACE (top) and cardiac death/MI (bottom) as a function of personality categories of patients with social inhibition only, negative affectivity only, and Type D personality. Number of cases and number of patients in the categories are presented on top of each bar. Patients of the ‘‘reference category’’ are low in both NA and SI; those in the ‘‘social inhibition category’’ are high in SI but low in NA; those in the ‘‘negative affectivity category’’ are high in NA but low in SI; and ‘‘Type D patients’’ are high in both NA and SI. MACE = major adverse cardiac event; MI = myocardial infarction; NAj = negative affectivity e9; NA+ = negative affectivity Q10; SIj = social inhibition e9; SI+ = social inhibition Q10 as defined by the previously published cutoff on the 14-item Type D Scale (24,25).

Number of cases appears in parentheses. MACE = major adverse cardiac event; MI = myocardial infarction; OR = odds ratio; CI = confidence interval; CABG = coronary bypass surgery. a All variables entered simultaneously. b Cutoff score Q10 on the 14-item Type D Scale measures of NA and SI (24,25). c Left ventricular ejection fraction e50% (22). d For younger/older men and women, e140/e120 and e100/e80 W (6). e CABG coded as 0 and no bypass as 1.

Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

877

J. DENOLLET et al. with an increased risk for adverse events and not the separate effect of one of these traits. Type D was also associated with the more rigorous end point of MI and cardiac death, but not with other, noncardiac causes of death. Finally, Type D remained a significant predictor of cardiovascular events after adjustment for depressive symptoms and disease severity. The odds associated with Type D were lower than the previously reported odds of 4.1 to 4.8 (5Y7) but were comparable with the risk associated with decreased LVEF and poor exercise tolerance as established markers of disease severity. These findings are important because recent editorial comments discussed the need to further investigate the validity of Type D (20,21). Social Inhibition, Negative Affectivity, and Type D We used two different analytical approaches to test the validity of the predicted Type D pattern in which it is only the high social inhibition/negative affectivity category that is at higher risk. First, we analyzed the interaction of continuous negative affectivity and social inhibition z scores (20). This analysis showed that this interaction predicted adverse events, above and beyond the main effect of the separate traits. Next, we classified patients in discrete categories to compare the separate and combined effects of high and low trait levels (8). The risk of adverse events was higher in Type D patients with the combination of high social inhibition/ negative affectivity than in patients with a high score on only one of these traits. The OR of 2.35 for cardiac death/MI associated with Type D corresponded well with the pooled OR of 2.28 in a recent meta-analysis of Type D (15). Both analytical approaches supported the robustness of the Type D construct and its usefulness in clinical research and practice. Because Type D classification in most previous studies captures both negative affectivity and inhibition, it is impossible to determine if associations with health reflect the effect of one of these traits or their combination (20). Animal stress models show that an inhibited response is associated with a particular susceptibility to deleterious physiological responses (30,31) and that the combination of anxiety and inhibition better defines an animal’s vulnerability to stress than single behavioral traits (31). The effects of this behavioral inhibition profile in nonhuman social species share much in common with the effects of social inhibition in humans. Social avoidance in men (32), as well as the combination of emotional instability with low extraversion in ‘‘overcontrolled’’ children (33) and emotional exhaustion with inhibition in adults (34), has been associated with an increased risk of cardiovascular disease and death. These authors concluded that their findings were consistent with the Type D literature (32Y34). However, there is a need to further examine the proposition of the Type D construct that social inhibition interacts with negative affectivity to predict health outcomes (20). With this report, there are now four studies that examined the statistical interaction of continuous social inhibition and negative affectivity scores in cardiac patients (16Y18). In all studies, this interaction was not associated with all-cause mortality, but depression or anxiety also failed to independently predict mortality in these studies. However, the present study found the synergistic interaction effect to be an independent predictor of fatal and nonfatal cardiac events. 878

In other studies, the interaction of social inhibition and negative affectivity was related to biological pathways of acute cardiac events such as dysfunctional cardiac output (35), acute stunning of the myocardium (36), and higher cortisol levels (37). Although we have no ready explanation for these discrepant findings, they may indicate the involvement of moderating factors that modify the association between Type D and health outcomes. Moderating and Mediating Variables CAD is a complex condition, and understanding its causal associations involves the search for moderators that reveal at what ages or in what subgroups such associations hold (38). Age is of particular interest because age is strongly associated with mortality outcomes (39) caused by dysfunction of the aging heart (40) and age-related comorbidities such as diabetes or kidney disease (41). Psychosocial factors may be important in younger patients with CAD (42), but the major predictors of mortality in patients with heart failure are older age, cardiac dysfunction, and comorbid diseases (41). Two studies in patients with heart failure with a mean age of 66.6 years (16) and 70.7 years (17) found no effect of Type D on mortality. In contrast, the mean age was 58.6 years in this study and 57.8, 59.0, and 57.7 years in three other recent studies that confirmed the link between Type D and cardiac events (43Y45). Hence, the predictive value of Type D may be more pronounced in younger patients (42) but recedes in older patients that already have a high mortality risk because of cardiovascular changes and comorbidities (39Y41). Most null studies of Type D used all-cause mortality as an end point. However, it is also possible that Type D is not associated with noncardiovascular events. Previously, others reported that social avoidance predicted cardiovascular death in men but not death due to other causes (32). Likewise, Type D was unrelated to all-cause mortality or noncardiac death in our study but was associated with an increased risk of fatal and nonfatal cardiac events. Other groups reported that Type D was associated with acute coronary events (46), acute stress cardiomyopathy (36), and cardiovascular events after cardiac surgery (43). A better understanding of mediating mechanisms is needed to explain these findings on outcomes. Several independent research teams have reported previously that biological (36,37,47Y51) and behavioral (52Y58) pathways may mediate the link between Type D and cardiovascular events. Biological mediators related to Type D include heightened cardiovascular reactivity (47), acute myocardial stunning (36), increased risk of ventricular arrhythmias (48), decreased heart rate recovery (49), higher cortisol levels (37,47), and metabolic dysfunction (50,51). We reported that Type D was related to an increase in proinflammatory cytokines (59) and oxidative stress (60) as well as impaired repair of the vascular wall because of a decrease in endothelial progenitor cells (61). Others also showed that Type D was related to behavioral mediators, including smoking and physical inactivity (52,53), nonadherence to treatment (54Y56), and higher stress levels (57,58). However, there still is much to be learned about the conditions under which Type D influences health. Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

TYPE D AND ADVERSE EVENTS IN PATIENTS WITH CAD Limitations and Challenges This study has a number of limitations. Most patients were men, and all of them were diagnosed as having CAD. Therefore, our results may not generalize to women or to other populations, such as patients with heart failure (16). Although we attempted to control for somatic confounding by adjusting for disease severity, the current design does not allow us to rule out the possibility that the diagnosis of CAD may have an effect on Type D scores. Others reported that Type D was related to cardiovascular stress reactivity (47) and increased risk of ventricular arrhythmias (48) in healthy people without CAD diagnosis, but longitudinal research is needed to examine whether Type D also precedes the development of CAD. Finally, it remains possible that the association of Type D with cardiac events may have been influenced by other covariates that were not accounted for in the regression models. Strengths of this study include its long-term follow-up, standard assessment of disease severity, and assessment of Type D as both a dichotomized and a continuous measure. We addressed two primary criticisms concerning Type D in this study (the uncertainty about the synergistic effect of social inhibition and negative affectivity, and the dichotomization of these traits to represent Type D) (20), but other criticisms include the potential overlap with other measures of distress and the preponderance of work by our own research group (20,62). We were the first to publish negative results about Type D in patients with heart failure (16) and reported other null findings (19). Ioannidis (62) notes that this is a ‘‘sign of excellence’’ because investigators may be hesitant to publish negative results about their theories. However, it still is unclear why some studies by independent groups yielded negative results (17,18) and others supported the biological basis of Type D (47Y51) and its association with cardiac events (36,43,46). Factor analytical studies of the DS14 (24) have consistently shown that social inhibition and negative affectivity represent two distinct traits (27,46,63,64). Imaging with functional magnetic resonance imaging (11) and other experimental research (47) confirmed that these two components of Type D activate different brain and cortisol responses to social threat. In addition, epidemiological research indicated that Type D is different from depressive disorder (29) and other forms of psychopathology (46). This may explain why Type D remained a significant predictor of cardiovascular events after adjustment for depressive symptoms in our study. Previously, Type D has also been related to heightened cortisol (37), oxidative stress (60), poor health (65), and adverse events (44,66) in cardiac patients, independent from depression. Type D is associated with the prevalence and persistence of depression (53,57,67), but both forms of distress may have independent cardiovascular effects (12). However, additional research is needed to investigate this proposition, and multiple other issues remain to be resolved in this area of research (20).

cardiac events. If Type D assessment could also improve estimates of patients’ future health risk, this would enable identification of patients who may benefit from more intensive follow-up and care. The Type D concept may add a complementary approach to intervention because it highlights a) the clustering of negative emotions (9) and b) the potential role of social inhibition (69). Type D has also been associated with poor outcome of invasive treatments, including implantable cardioverter defibrillator therapy (45,70), cardiac surgery (43,71), and heart transplantation (72). Type D patients believe that their condition will be less responsive to medical treatment (73) and may be reluctant to consult clinical staff for their cardiovascular symptoms (74). Hence, in addition to reducing their levels of social inhibition and negative affectivity (75), behavioral intervention in Type D patients may be useful to improve coping and self-management (76). Not all coronary patients have an equal risk of future adverse events. In addition to clinical characteristics (77), general psychological distress as represented by the Type D construct may also affect cardiovascular outcomes (12). Previously, some groups (17,18), including our own (16,19), reported that Type D was not related to all-cause mortality in cardiac patients. In this study, the dichotomous Type D classification identified a highrisk group of patients with CAD, and social inhibition interacted with negative affectivity to predict cardiac events. The significant interaction in this study has no implications for the meaning of prior Type D studies that did not test this interaction, and a reanalysis of these prior studies is needed (20). This heterogeneity in findings highlights the need to explore the role of moderating factors that may alter the association between Type D and health outcomes (38), including subgroup analyses that are clearly defined on a prioriYbased knowledge about disease processes (78,79). For example, future research is needed to test our hypothesis (79) that the effect of Type D may differ among younger versus older (39Y42) and CAD versus heart failure (15Y17) subgroups. It is also possible that Type D is related to cardiovascular events but not noncardiac death. Many other challenges remain, including the pathways that mediate associations between Type D and cardiovascular events, but our findings indicate that the prognostic role and modification of emotional distress in Type D patients with CAD merits further investigation. Source of Funding and Conflicts of Interest: Part of this study was supported by the Netherlands Organization for Scientific Research (NWO), the Hague, the Netherlands, with a Vici grant (45304004; to Dr. Denollet) and a Vidi grant (91710393; to Dr. Pedersen), and by a grant from the Research Foundation Flanders, Belgium (to Dr Conraads). No conflict of interest exists for any of the authors.

REFERENCES Implications and Conclusions The experience of social stress reflects the subjective appraisal of threats in daily life (68), and people differ in their sensitivity to social contexts (38). Accounting for these individual differences may improve our understanding of the link between stress and

1. Spertus J. Improving the care for myocardial infarction: editor’s perspective. Circ Cardiovasc Qual Outcomes 2010;3:439Y40. 2. Whooley MA, de Jonge P, Vittinghoff E, Otte C, Moos R, Carney RM, Ali S, Dowray S, Na B, Feldman MD, Schiller NB, Browner WS. Depressive symptoms, health behaviors, and risk of cardiovascular events in patients with coronary heart disease. JAMA 2008;300:2379Y88.

Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

879

J. DENOLLET et al. 3. Roest AM, Martens EJ, de Jonge P, Denollet J. Anxiety and risk of incident coronary heart disease: a meta-analysis. J Am Coll Cardiol 2010;56:38Y46. 4. Chapman BP, Roberts B, Duberstein P. Personality and longevity: knowns, unknowns, and implications for public health and personalized medicine. J Aging Res 2011;2011:759170. 5. Denollet J, Sys SU, Stroobant N, Rombouts H, Gillebert T, Brutsaert DL. Personality as independent predictor of long-term mortality in patients with coronary heart disease. Lancet 1996;347:417Y421. 6. Denollet J, Vaes J, Brutsaert DL. Inadequate response to treatment in coronary heart disease: Adverse effects of Type D personality and younger age on 5-year prognosis and quality of life. Circulation 2000;102:630Y5. 7. Denollet J, Pedersen SS, Vrints CJ, Conraads VM. Usefulness of Type D personality in predicting five-year cardiac events above and beyond concurrent symptoms of stress in patients with coronary heart disease. Am J Cardiol 2006;97:970Y3. 8. Denollet J, Pedersen SS, Ong AT, Erdman RA, Serruys PW, van Domburg RT. Social inhibition modulates the effect of negative emotions on cardiac prognosis following percutaneous coronary intervention in the drug-eluting stent era. Eur Heart J 2006;27:171Y7. 9. Denollet J, Pedersen SS. Anger, depression and anxiety in cardiac patients: the complexity of individual differences in psychological risk. J Am Coll Cardiol 2009;53:947Y9. 10. Smith TW, Birmingham W, Uchino BN. Evaluative threat and ambulatory blood pressure: cardiovascular effects of social stress in daily experience. Health Psychol 2012;31:763Y6. 11. Kret ME, Denollet J, Gre`zes J, de Gelder B. The role of negative affectivity and social inhibition in perceiving social threat: an fMRI study. Neuropsychologia 2011;49:1187Y93. 12. Denollet J, Schiffer AA, Spek V. A general propensity to psychological distress affects cardiovascular outcomes: evidence from research on the Type D (distressed) personality profile. Circ Cardiovasc Qual Outcomes 2010;3:546Y57. 13. Reich J, Schatzberg A. Personality traits and medical outcome of cardiac illness. J Psychiatr Res 2010;44:1017Y20. 14. O’Dell KR, Masters KS, Spielmans GI, Maisto SA. Does Type D personality predict outcomes among patients with cardiovascular disease? A meta-analytic review. J Psychosom Res 2011;71:199Y206. 15. Grande G, Romppel M, Barth J. Association between Type D personality and prognosis in patients with cardiovascular diseases: a systematic review and meta-analysis. Ann Behav Med 2012;43:299Y310. 16. Pelle AJ, Pedersen SS, Schiffer AA, Szabo´ B, Widdershoven JW, Denollet J. Psychological distress and mortality in systolic heart failure. Circ Heart Fail 2010;3:261Y7. 17. Coyne JC, Jaarsma T, Luttik ML, van Sonderen E, van Veldhuisen DJ, Sanderman R. Lack of prognostic value of Type D personality for mortality in a large sample of heart failure patients. Psychosom Med 2011;73:557Y62. 18. Grande G, Romppel M, Vesper JM, Schubmann R, Glaesmer H, Herrmann-Lingen C. Type D personality and all-cause mortality in cardiac patients: data from a German cohort study. Psychosom Med 2011;73:548Y56. 19. Damen NL, Versteeg H, Boersma E, Serruys PW, van Geuns RJ, Denollet J, van Domburg RT, Pedersen SS. Depression is independently associated with 7-year mortality in patients treated with percutaneous coronary intervention. Int J Cardiol 2013;167:2496Y501. 20. Smith TW. Toward a more systematic, cumulative and applicable science of personality and health: lessons from Type D personality. Psychosom Med 2011;73:528Y32. 21. Bacon SL, Moullec G. Type D personality and heart disease: it might be ‘one small step’, but it is still moving forward. Ann Behav Med 2012; 43:280Y1. 22. Gottdiener JS, McClelland RL, Marshall R, Shemanski L, Furberg CD, Kitzman DW, Cushman M, Polak J, Gardin JM, Gersh BJ, Aurigemma GP, Manolio TA. Outcome of congestive heart failure in elderly persons: influence of left ventricular systolic function. The Cardiovascular Health Study. Ann Intern Med 2002;137:631Y9. 23. Laukkanen JA, Ma¨kikallio TH, Rauramaa R, Kiviniemi V, Ronkainen K, Kurl S. Cardiorespiratory fitness is related to the risk of sudden cardiac death: a population-based follow-up study. J Am Coll Cardiol 2010; 56:1476Y83. 24. Denollet J. DS14: standard assessment of negative affectivity, social inhibition, and Type D personality. Psychosom Med 2005;67:89Y97. 25. Emons WH, Meijer RR, Denollet J. Negative affectivity and social inhibition in cardiovascular disease: evaluating Type D personality and its assessment using Item Response Theory. J Psychosom Res 2007;63:27Y39. 880

26. Straat JH, van der Ark LA, Sijtsma K. Multi-method analysis of the internal structure of the Type D Scale-14 (DS14). J Psychosom Res 2012;72:258Y65. 27. Kupper N, Pedersen SS, Ho¨fer S, Saner H, Oldridge N, Denollet J. Crosscultural analysis of Type D (distressed) personality in 6222 patients with ischemic heart disease: a study from the International HeartQoL Project. Int J Cardiol 2013;166:327Y33. 28. Denollet J, Martens EJ, Smith OR, Burg MM. Efficient assessment of depressive symptoms and their prognostic value in myocardial infarction patients. J Affect Disord 2010;120:105Y11. 29. Denollet J, de Jonge P, Kuyper A, Schene AH, van Melle JP, Ormel J, Honig A. Depression and Type D personality represent different forms of distress in the Myocardial INfarction and DepressionYIntervention Trial (MIND-IT). Psychol Med 2009;39:749Y56. 30. Cavigelli SA, McClintock MK. Fear of novelty in infant rats predicts adult corticosterone dynamics and an early death. Proc Natl Acad Sci U S A 2003;100:16131Y6. 31. Castro JE, Diessler S, Varea E, Ma´rquez C, Larsen MH, Cordero MI, Sandi C. Personality traits in rats predict vulnerability and resilience to developing stress-induced depression-like behaviors, HPA axis hyperreactivity and brain changes in pERK1/2 activity. Psychoneuroendocrinology 2012;37:1209Y23. 32. Berry JD, Lloyd-Jones DM, Garside DB, Wang R, Greenland P. Social avoidance and long-term risk for cardiovascular disease death in healthy men. Ann Epidemiol 2007;17:591Y6. 33. Chapman BP, Goldberg LR. Replicability and 40-year predictive power of childhood ARC types. J Pers Soc Psychol 2011;101:593Y606. 34. Appels A, Golombeck B, Gorgels A, de Vreede J, van Breukelen G. Behavioral risk factors of sudden cardiac arrest. J Psychosom Res 2000;48:463Y9. 35. Howard S, Hughes BM. Construct, concurrent and discriminant validity of Type D personality in the general population. Psychol Health 2012; 27:242Y58. 36. Compare A, Bigi R, Silva Orrego P, Proietti R, Grossi E, Steptoe A. Type D personality is associated with the development of stress cardiomyopathy following emotional triggers. Ann Behav Med 2013;45:299Y307. 37. Whitehead DL, Perkins-Porras L, Strike PC, Magid K, Steptoe A. Cortisol awakening response is elevated in acute coronary syndrome patients with Type D personality. J Psychosom Res 2007;62:419Y25. 38. Boyce WT, Sokolowski MB, Robinson GE. Toward a new biology of social adversity. Proc Natl Acad Sci U S A 2012;109(suppl 2):17143Y8. 39. Arnold SV, Alexander KP, Masoudi FA, Ho PM, Xiao L, Spertus JA. The effect of age on functional and mortality outcomes after acute myocardial infarction. J Am Geriatr Soc 2009;57:209Y17. 40. Shih H, Lee B, Lee RJ, Boyle AJ. The aging heart and post-infarction left ventricular remodeling. J Am Coll Cardiol 2011;57:9Y17. 41. Barlera S, Tavazzi L, Franzosi MG, Marchioli R, Raimondi E, Masson S, Urso R, Lucci D, Nicolosi GL, Maggioni AP, Tognoni G. Predictors of mortality in 6975 patients with chronic heart failure in the Gruppo Italiano per lo Studio della Streptochinasi nell’Infarto Miocardico-Heart Failure Trial: proposal for a nomogram. Circ Heart Fail 2013;6:31Y9. 42. Denollet J, Freedland KE, Carney RM, de Jonge P, Roest AM. Cognitiveaffective symptoms of depression after myocardial infarction: different prognostic importance across age groups. Psychosom Med 2013;75:701Y8. 43. Csere´p Z, Balog P, Sze´kely J, Treszl A, Kopp MS, Thayer JF, Sze´kely A. Psychosocial factors and major adverse cardiac and cerebrovascular events after cardiac surgery. Interact Cardiovasc Thorac Surg 2010;11:567Y72. 44. Martens EJ, Mols F, Burg MM, Denollet J. Type D personality predicts clinical events after myocardial infarction, above and beyond disease severity and depression. J Clin Psychiatry 2010;71:778Y83. 45. Pedersen SS, van den Broek KC, Erdman RA, Jordaens L, Theuns DA. Pre-implantation implantable cardioverter defibrillator concerns and Type D personality increase the risk of mortality in patients with an implantable cardioverter defibrillator. Europace 2010;12:1446Y52. 46. Beutel ME, Wiltink J, Till Y, Wild PS, Mu¨nzel T, Ojeda FM, Zeller T, Schnabel RB, Lackner K, Blettner M, Zwiener I, Michal M. Type D personality as a cardiovascular risk marker in the general population: results from the Gutenberg health study. Psychother Psychosom 2012;81:108Y17. 47. Habra ME, Linden W, Anderson JC, Weinberg J. Type D personality is related to cardio-vascular and neuroendocrine reactivity to acute stress. J Psychosom Res 2003;55:235Y45. 48. Einvik G, Dammen T, Namtvedt SK, Hrubos-StrLm H, Randby A, Kristiansen HA, Nordhus IH, Somers VK, Omland T. Type D personality is associated with increased prevalence of ventricular arrhythmias in community-residing persons without coronary heart disease. Eur J Prev Cardiol In press. [Epub: DOI: 10.1177/2047487312462800] Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

TYPE D AND ADVERSE EVENTS IN PATIENTS WITH CAD 49. von Ka¨nel R, Barth J, Kohls S, Saner S, Znoj H, Saner G, Schmid JP. Heart rate recovery after exercise in chronic heart failure: role of vital exhaustion and Type D personality. J Cardiol 2009;53:248Y56. 50. Tziallas D, Kostapanos MS, Skapinakis P, Milionis HJ, Athanasiou T, S Elisaf M, Mavreas V. The association between Type D personality and the metabolic syndrome: a cross-sectional study in a university-based outpatient lipid clinic. BMC Res Notes 2011;4:105. 51. Altmaier E, Emeny RT, Krumsiek J, Lacruz ME, Lukaschek K, Ha¨fner S, Kastenmu¨ller G, Ro¨misch-Margl W, Prehn C, Mohney RP, Evans AM, Milburn MV, Illig T, Adamski J, Theis F, Suhre K, Ladwig KH. Metabolomic profiles in individuals with negative affectivity and social inhibition: a population-based study of Type D personality. Psychoneuroendocrinology 2013;38:1299Y309. 52. Gilmour J, Williams L. Type D personality is associated with maladaptive health-related behaviors. J Health Psychol 2012;17:471Y8. 53. Hausteiner C, Klupsch D, Emeny R, Baumert J, Ladwig KH; for the KORA Investigators. Clustering of negative affectivity and social inhibition in the community: prevalence of Type D personality as a cardiovascular risk marker. Psychosom Med 2010;72:163Y71. 54. Williams L, O’Connor RC, Grubb N, O’Carroll R. Type D personality predicts poor medication adherence in myocardial infarction patients. Psychol Health 2011;26:703Y12. 55. Molloy GJ, Randall G, Wikman A, Perkins-Porras L, Messerli-Bu¨rgy N, Steptoe A. Type D personality, self-efficacy, and medication adherence following an acute coronary syndrome. Psychosom Med 2012;74:100Y106. 56. Brostro¨m A, Stro¨mberg A, Ma˚rtensson J, Ulander M, Harder L, Svanborg E. Association of Type D personality to perceived side effects and adherence in CPAP-treated patients with OSAS. J Sleep Res 2007;16:439Y47. 57. Michal M, Wiltink J, Grande G, Beutel ME, Bra¨hler E. Type D personality is independently associated with major psychosocial stressors and increased health care utilization in the general population. J Affect Disord 2011; 134:396Y403. 58. Williams L, Wingate A. Type D personality, physical symptoms and subjective stress: the mediating effects of coping and social support. Psychol Health 2012;27:1075Y85. 59. Denollet J, Schiffer AA, Kwaijtaal M, Hooijkaas H, Hendriks EH, Widdershoven JW, Kupper N. Usefulness of Type D personality and kidney dysfunction as predictors of interpatient variability in immune activation in chronic heart failure. Am J Cardiol 2009;103:399Y404. 60. Kupper N, Gidron Y, Winter J, Denollet J. The association between Type D personality, depression and oxidative stress in chronic heart failure patients. Psychosom Med 2009;71:973Y80. 61. Van Craenenbroeck EM, Denollet J, Paelinck BP, Beckers P, Possemiers N, Hoymans VY, Vrints CJ, Conraads VM. Circulating CD34+/KDR+ endothelial progenitor cells are reduced in chronic heart failure patients as a function of Type D personality. Clin Sci 2009;117:165Y72. 62. Ioannidis JP. Scientific inbreeding and same-team replication: Type D personality as an example. J Psychosom Res 2012;73:408Y10. 63. Kudielka BM, von Ka¨nel R, Gander ML, Fischer JE. The interrelationship of psychosocial risk factors for coronary artery disease in a working

64. 65. 66. 67. 68. 69. 70.

71.

72. 73. 74. 75. 76.

77.

78. 79.

population: do we measure distinct or overlapping psychological concepts? Behav Med 2004;30:35Y43. Pelle AJ, Denollet J, Zwisler AD, Pedersen SS. Overlap and distinctiveness of psychological risk factors in patients with ischemic heart disease and chronic heart failure. J Affect Disord 2009;113:150Y6. Mols F, Martens EJ, Denollet J. Type D personality and depressive symptoms are independent predictors of impaired health status following acute myocardial infarction. Heart 2010;96:30Y5. Denollet J, Pedersen SS. Prognostic value of Type D personality compared with depressive symptoms. Arch Intern Med 2008;168:431Y2. Romppel M, Herrmann-Lingen C, Vesper JM, Grande G. Type D personality and persistence of depressive symptoms in a German cohort of cardiac patients. J Affect Disord 2012;136:1183Y7. McEwen BS. Brain on stress: how the social environment gets under the skin. Proc Natl Acad Sci U S A 2012;109(suppl 2):17180Y5. Denollet J, Sys SU, Brutsaert DL. Personality and mortality after myocardial infarction. Psychosom Med 1995;57:582Y91. Denollet J, Tekle FB, Pedersen SS, van der Voort PH, Alings M, van den Broek KC. Prognostic importance of distressed (Type D) personality and shocks in patients with an implantable cardioverter defibrillator. Int J Cardiol 2013;167:2705Y9. Al-Ruzzeh S, Athanasiou T, Mangoush O, Wray J, Modine T, George S, Amrani M. Predictors of poor mid-term health related quality of life after primary isolated coronary artery bypass grafting surgery. Heart 2005;91: 1557Y62. Denollet J, Holmes RV, Vrints CJ, Conraads VM. Unfavorable outcome of heart transplantation in recipients with Type D personality. J Heart Lung Transplant 2007;26:152Y8. Williams L, O’Connor RC, Grubb NR, O’Carroll RE. Type D personality and illness perceptions in myocardial infarction patients. J Psychosom Res 2011;70:141Y4. Schiffer AA, Denollet J, Widdershoven JW, Hendriks EH, Smith OR. Failure to consult for symptoms of heart failure in patients with a Type D personality. Heart 2007;93:814Y8. Nyklı´*ek I, van Beugen S, Denollet J. Effects of mindfulness-based stress reduction on distressed (Type D) personality traits: a randomized controlled trial. J Behav Med 2013;36:361Y70. Pelle AJ, van den Broek KC, Denollet J. Interventions in the context of the distressed (Type D) personality. In: Dornelas EA, editor. Stress Proof the Heart: Behavioral Interventions for Cardiac Patients. New York, NY: Springer; 2012:167Y97. Bhatt DL, Eagle KA, Ohman EM, Hirsch AT, Goto S, Mahoney EM, Wilson PW, Alberts MJ, D’Agostino R, Liau CS, Mas JL, Ro¨ther J, Smith SC Jr, Salette G, Contant CF, Massaro JM, Steg PG; REACH Registry Investigators. Comparative determinants of 4-year cardiovascular event rates in stable outpatients at risk of or with atherothrombosis. JAMA 2010;304:1350Y7. Wittes J. On looking at subgroups. Circulation 2009;119:912Y5. Denollet J. Interpersonal sensitivity, social inhibition, and Type D personality: how and when are they associated with health? Comment on Marin & Miller (2013). Psychol Bull 2013;139:991Y7.

Psychosomatic Medicine 75:873Y881 (2013)

Copyright © 2013 by the American Psychosomatic Society. Unauthorized reproduction of this article is prohibited.

881