The findings of the present study show that people with Type D personality had a twofold increased risk of having metabolic syndrome, independent of established metabolic syndrome risk factors such as age, gender, having a partner, education level, presence of CVD, family history of CVD, smoking, alcohol use, diet and exercise. The twice as large prevalence of metabolic syndrome in people with Type D may be attributed to the metabolic syndrome components dyslipidemia and hypertension, but not to waist-circumference or diabetes. Post-hoc analyses showed that higher levels of both NA and SI were associated with increased prevalence of metabolic syndrome. With respect to the individual metabolic risk markers, lipid abnormalities were associated with both NA and SI subcomponents, hypertension was associated with increased NA levels. This finding stresses the significance of high scores on both Type D subcomponents for the Type D personality construct.
There are no studies to date exploring the relation between Type D personality and metabolic syndrome. Recent studies on cardiac risk factors show a decreased prevalence of dyslipidemia in men with Type D personality, increased hypertension prevalence in women with Type personality, and no differences in cholesterol/HDL quotient, obesity, or diabetes prevalence . Einvik and colleagues observed an increased BMI in persons with Type D personality, and higher triglyceride levels, but not other differences in cholesterol, blood pressure, waist/hip ratio, or fasting serum glucose . Several psychological constructs related to negative affect, such as depression, hostility, and anger, have been found to be associated with metabolic syndrome [38–42]. These studies present cross-sectional evidence for a relationship in cardiac patients, e.g., a study by Vaccarino and colleagues showed an independent association of metabolic syndrome with depression in women with suspected coronary artery disease , while the Heart & Soul study showed significant univariate associations between depression, hostility and optimism-pessimism and metabolic syndrome prevalence in stable coronary artery disease patients . An Australian study examined the association between depression and metabolic syndrome in the general population and found that metabolic syndrome was related to higher depression scores, but not anxiety or psychological distress . In addition to the cross-sectional evidence, there is also some prospective evidence that depression, hostility, and anger predict increased risk for developing the metabolic syndrome . Negative findings have been published as well, as Herva and colleagues, in a 31 year old cohort in Northern Finland, reported that no relation was observed between depression and the metabolic syndrome after controlling for covariates , and recently, a study in 1,212 elderly participants from the Longitudinal Aging Study Amsterdam reported the absence of a relation between major depression and the metabolic syndrome . The significant relation between depression, hostility, and pessimism-optimism with metabolic syndrome in cardiac patients from the Heart and Soul study was no longer significant after controlling for socioeconomic status and health behaviors like physical activity, smoking, regular alcohol use, and BMI .
The current study observed lifestyle differences for Type D personality as well, i.e. persons with a Type D personality adhered less to the Dutch norm for healthy physical activity and the Dutch fit norm, but for the subgroup that did exercise, Type D personality was not related to the average metabolic energy rates or energy expenditure. Further, Type D persons less often ate a varied diet, and restricted the intake of fat to a lesser extent. There were no differences in salt restriction or diet category. However, exercise and dietary lifestyle factors did not affect the increased risk of metabolic syndrome associated with Type D personality. Similar findings have been observed in other studies: people with Type D personality had a somewhat less healthy diet, and were less physically active, or spend less time outdoors [19, 36, 37].
Primary intervention for metabolic syndrome involves lifestyle modification, weight management, diet and exercise changes. In healthy individuals previous studies have shown that poor dietary habits (less prudent food choices)  and low exercise are associated with some individual metabolic risk markers [45, 46], as well as an increased risk of developing the metabolic syndrome . Our results are not concurrent with these previous findings, as dietary habits and exercise did not explain additional variance in the logistic regression model. One explanation for this might be that many of the other variables in the model, i.e. Type D personality, age, BMI, educational level, smoking, presence of cardiovascular disease and eating a varied diet were all significantly associated with adherence to the Dutch norm for healthy physical activity in univariate correlations. A similar correlation pattern was present for dietary habits, serving as an explanation as to why these two lifestyle factors are not associated with increased risk of metabolic syndrome in our study.
Modification of cardiovascular risk by adjusting lifestyle habits involving diet, exercise and smoking by self-management programs are not always effective [48–50]. One approach of modifying the cardiovascular risk associated with Type D personality might be modification of health behaviors. However, a complicating factor in implementing self-management programs to address unhealthy lifestyles may be that cardiovascular patients with Type D personality are less likely to consult their physician , which makes both physician and the person with Type D personality unaware of their increased risk. Hence investigating the risk of Type D patients in terms of metabolic syndrome and unhealthy lifestyle in a community sample, in which consultation behavior is not an issue, is a first step in unraveling the potential effectiveness of behavioral interventions for cardiovascular disease prevention.
One limitation to our study is that we used self-report measures to assess the components of the metabolic syndrome. It is therefore imperative to establish whether the prevalence of the metabolic syndrome and subcomponents are representative for the Dutch population. The prevalence of the metabolic syndrome in our sample was significantly lower than in others [51, 52]. This may largely be due to the use of a proxy self-report measure of metabolic syndrome. We used a more strict three out of four criteria to define metabolic syndrome in order not to overestimate metabolic syndrome prevalence, whereas the IDF definition uses three out of five. However, this reduces the chance of receiving metabolic syndrome diagnosis, e.g. if 25% of a sample has metabolic syndrome, based on 3/5 criteria, we could only detect 20% of this group, based on our 3/4 criteria. The cut-off scores for an increased waist-circumference according to IDF-criteria (55.6%) were comparable with another Dutch sample (57.6% , and being overweight (BMI 25-30: 34.7%) or having obesity (BMI > 30: 10.5%) were comparable with the Dutch population (35.7% and 11.1% respectively, http://statline.cbs.nl).
Another limitation of the current study is the cross-sectional nature of the study sample, precluding any causal statements on the progression of metabolic syndrome in people with Type D personality. Strengths include the large sample size and the inclusion of multiple covariates.