Purpose Depression and posttraumatic stress disorder (PTSD) may be linked to the metabolic syndrome (MetS). Consistency of this association across ethnic groups and the influence of comorbidity of depression/PTSD were examined. Methods Cross-sectional baseline data from the HELIUS study were used (4527 Dutch, 2999 South-Asian Surinamese, 4058 African Surinamese, 2251 Ghanaian, 3522 Turkish and 3825 Moroccan participants). The Patient Health Questionnaire-9 (PHQ-9) (score range 0–27) measured depressive symptoms. A 9-item questionnaire (score range 0–9) measured PTSD symptoms. The MetS was defined according to the International Diabetes Federation. The association of a depressed mood (PHQ-9 sum score ≥ 10) and severe PTSD symptoms (sum score ≥ 7) with the MetS was examined using logistic regression. Interaction with ethnicity and between a depressed mood and severe PTSD symptoms was tested. Results A depressed mood was associated with the MetS [OR (95% CI) = 1.37 (1.24–1.51)] in the total sample and consist- ent across ethnic groups (p values for interaction all > 0.05). Severe PTSD symptoms were significantly associated with the MetS in the Dutch [OR (95% CI) = 1.71 (1.07–2.73)]. The South-Asian Surinamese, Turks and Moroccans showed weaker associations than the Dutch (p values for interaction all < 0.05). A depressed mood and severe PTSD symptoms did not interact in the association with the MetS (p values for interaction > 0.05). Conclusions A depressed mood was consistently associated with the MetS across ethnic groups, but the association between severe PTSD symptoms and the MetS maybe ethnicity dependent. The association with the MetS was not different in case of depressed mood/severe PTSD symptoms comorbidity. Keywords Metabolic syndrome · Depression · Posttraumatic stress disorder · Ethnicity · HELIUS study Introduction Electronic supplementary material The online version of this article (https ://doi.org/10.1007/s0012 7-018-1533-y) contains The metabolic syndrome (MetS) is a common cluster of supplementary material, which is available to authorized users. somatic symptoms, including elevated waist circumference, elevated blood pressure, dyslipidemia (reduced HDL and * Marieke J. van Leijden elevated triglyceride levels) and elevated fasting glucose . email@example.com The MetS contributes to the rising prevalence of cardiovas- Department of Public Health, Academic Medical Center, cular disease and type 2 diabetes . Psychopathology may Meibergdreef 15, 1105 AZ Amsterdam, The Netherlands be a risk factor for the MetS; two frequently comorbid men- Department of Psychiatry, Amsterdam Public Health tal disorders, depression and posttraumatic stress disorder Research Institute, VU University Medical Center, Oldenaller (PTSD), have been linked to the MetS [3, 4]. The pathways 1, 1081 HL Amsterdam, The Netherlands that link depression and PTSD with the MetS are not com- Department of Psychiatry, Academic Medical Center, pletely understood, but evidence points towards unhealthy Meibergdreef 5, 1105 AZ Amsterdam, The Netherlands lifestyle, biological stress dysregulation and/or inflammation Department of Health Sciences, Vrije Universiteit [5–9]. Amsterdam, De Boelelaan 1085, 1081 HV Amsterdam, The Netherlands Vol.:(0123456789) 1 3 922 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 The association between depression and the MetS has psychopathology and the MetS. For example, it is possi- been established; several large meta-analyses show this ble that some ethnic groups are more prone towards devel- association [4, 10, 11]. While previous studies mostly used oping the MetS due to a number of risk factors other than cross-sectional data , several large prospective stud- depression and PTSD, such as migration-related stress- ies have demonstrated an association between depression ors, perceived ethnic discrimination, biological disposi- and subsequent MetS development [11, 12]. Moreover, this tion, unhealthy lifestyle, poor social support and/or socio- association has been examined in diverse populations and economic status [30–33]. On the other hand, it has been in several countries . However, the consistency of this argued that ethnic minorities may be more vulnerable to the association across ethnic groups remains unclear. One study effects of psychopathology on somatic health, because these in the US found an association between depression and the groups may experience less adequate care for psychopathol- MetS in white Americans, but not in African-Americans ogy, which can prolong its course and increase the risk of , while another study found no interaction by ethnic- harmful effects . As the Western society is becoming ity in a sample including white, Mexican Americans and increasingly multicultural , multi-ethnic comparisons of African-Americans . Similarly, one study found no the link between psychopathology and MetS are important interaction by ethnicity in the association with metabolic to better understand the nature of this comorbidity and its risk factors (BMI, obesity rate, waist-hip ratio, LDL level, clinical implications in a multicultural society. and triglyceride level) . In contrast, studies on depres- Because depression and PTSD are frequently comorbid sion and cardiovascular mortality and coronary heart disease [35, 36], and may both increase the risk of developing the showed an association only in African-Americans, but not MetS [3, 4], it is important to investigate both disorders in in white Americans [16–18]. However, these studies were relation to the MetS. Depression and PTSD have an over- all conducted in the US and it is unclear how these findings lap in symptoms, in particular negative affect and dysphoric translate to ethnic groups with different migration back - arousal . Depression is seen as a confounder in the asso- grounds and living conditions. ciation between PTSD and the MetS [24, 38]. Additionally, The association between PTSD and the MetS is poorly it has been suggested that comorbidity of depression and established compared to the association between depression PTSD increases the risk of the MetS more than PTSD alone and the MetS. In a meta-analysis including 9 cross-sectional [25, 39]. There is, however, a scarcity of studies that inves- studies, Rosenbaum et al. found a 1.82 times higher risk tigate the independence of depression and PTSD in their of the MetS for persons with PTSD compared to age- and association with the MetS, and the existence of a potential sex-matched controls . Recently, a study showed a lon- interaction between these disorders in their association with gitudinal association between PTSD and the MetS, con- the MetS. It seems plausible that when the two disorders trolling for age, sex, ethnicity and educational level . co-occur, the impact on health may be strengthened. After Evidence on this association is still lacking, however. For all, previous studies have shown that comorbidity in mental example, the association between PTSD and the MetS has health conditions, due to higher symptom burden or more rarely been studied in large samples . Moreover, the diverse pathophysiology, have a negative impact on physi- majority of studies on PTSD and the MetS were conducted cal health [40, 41]. Consequently, there may be a significant in veterans [19–25]. Furthermore, similar to depression, interaction between depression and PTSD in the association there may be ethnic differences in the relationship between with the MetS. PTSD and the MetS. In fact, in one study on the relation- Therefore, the aim in this study was to examine the asso- ship between PTSD and chronic illnesses, the association ciations of both symptoms of depression and PTSD with with diabetes and heart disease was strongest in African- the MetS. We investigated the consistency of these associa- Americans . In contrast, another small study showed tions across ethnic groups living in the Netherlands and the an association between PTSD and metabolic risk factors potential interaction between depression and PTSD in the in white American women, but not in African American association with the MetS. women . Both studies were conducted in the US. Again, uncertainty remains whether these differences between US ethnic groups are applicable to the European context, and Methods how this translates to the association with the MetS. As these studies illustrate, there may be differential asso- The HELIUS study ciations between psychopathology and somatic outcomes between ethnic groups. We know that some ethnic minor- The HEalthy LIfe in an Urban Setting (HELIUS) study is a ity groups have an increased risk of developing depres- multi-ethnic cohort study aiming to unravel the mechanisms sion, PTSD and the MetS [27–29]. One could speculate underlying the impact of ethnicity on communicable and why ethnicity might moderate the association between non-communicable diseases . Full details on this study 1 3 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 923 are described elsewhere . In brief, baseline data col- having a sum score of 10 or higher were considered to have lection took place in 2011–2015 and included people aged a ‘depressed mood’ . 18–70 years from different ethnic groups living in Amster - dam, i.e., those of Dutch, Surinamese, Ghanaian, Moroc- PTSD symptoms can and Turkish origin. Participants were randomly sampled from the municipal register, stratified by ethnicity. Data were PTSD symptoms were measured by a questionnaire. Ques- collected by a questionnaire and a physical examination in tions were derived from the PTSD Symptom Scale-Self- which biological samples were also obtained. Report Version (PSS-SR), an instrument with good validity For the current study, we used baseline data of all par- and high reliability . The questionnaire consisted of nine ticipants in whom questionnaire data as well as data from items: three re-experiencing items, two avoidance items and the physical examination were available (n = 22,165). Java- four hyperarousal items. Cronbach’s alpha was 0.90 in the nese Surinamese participants (n = 233) and those with an whole sample and ranged from 0.85 to 0.92 between the unknown/other Surinamese origin (n = 267) or unknown/ six ethnic groups. Answer categories were yes (1) and no other ethnic origin (n = 48) were excluded because of their (0). We calculated the sum score of the nine PTSD items relatively small sample sizes. We further excluded partici- (ranging from 0 to 9). In case of more than two missing pants with missing values on a depressed mood (n = 240), items, the sum score was not calculated and considered as severe PTSD symptoms (n = 240) and/or on the MetS missing. Based on the non-linear association of the PTSD (n = 123). In the current analysis, the total sample consisted sum score with our main outcome variable (the MetS) (see of 21,182 participants; 4527 Dutch, 2999 South-Asian Suri- online resource Table S1), the PTSD sum score variable namese, 4058 African Surinamese, 2251 Ghanaian, 3522 was dichotomized: participants with a sum score of seven Turkish and 3825 Moroccan origin participants. or higher were considered to have ‘severe PTSD symptoms’. Metabolic syndrome Ethnicity The MetS was determined according to the harmonized Ethnicity was defined according to the country of birth of definition proposed by the International Diabetes Federa- the participant as well as that of his/her parents, which is tion (IDF) . By this definition, the MetS is present if at currently the most widely accepted assessment of ethnicity least three of the following five criteria are met (yes/no): in Netherlands . Participants were considered to be of (a) elevated fasting glucose (≥ 5.6 mM, or glucose-lowering Dutch origin if the person and both parents were born in medication); (b) elevated blood pressure (systolic ≥ 130 and/ the Netherlands. Participants were considered to be of non- or diastolic ≥ 85 mm Hg, or blood pressure–lowering medi- Dutch origin if the person fulfills either of the following cation); (c) reduced HDL-C (< 1.0 mM for men, 1.3 mM for criteria: (1) he or she was born abroad and has at least one women, or lipid-lowering medication); (d) elevated triglyc- parent born abroad (first generation) or (2) he or she was erides (≥ 1.7 mM, or lipid-lowering medication); and (e) born in the Netherlands but both his/her parents were born elevated waist circumference (ethnic specific cutoff values abroad (second generation). Surinamese subgroups (i.e., were used; for all women ≥ 80 cm, South-Asian men ≥ 90 cm African and South-Asian) were classified according to self- and other men ≥ 94 cm) . reported ethnic origin. Biomedical measurements Depressive symptoms Overnight fasting blood samples were drawn and plasma Depressive symptoms were assessed using the Patient Health samples were used to determine the concentration of glucose Questionnaire-9 (PHQ-9) . The PHQ-9 determines the by spectrophotometry, using hexokinase as primary enzyme prevalence of depressive symptoms over the preceding (Roche Diagnostics, Tokyo, Japan). Triglycerides and 2 weeks. Its cross-cultural validity has been demonstrated HDL-C were determined by colorimetric spectrophotom- across the ethnic groups included in the HELIUS study etry (Roche Diagnostics). Blood pressure was measured in . The PHQ-9 consists of nine items, with a response a seated position using a semiautomatic sphygmomanometer scale varying from never (0) to nearly every day (3). Cron- (Microlife WatchBP Home; Microlife AG, Widnau, Swit- bach’s alpha was 0.89 in the whole sample and ranged from zerland). Using appropriate cuff sizes, two readings were 0.84 to 0.90 between the ethnic groups. If one of the items taken on the upper left arm at heart level after being seated was missing, the mean score of the other eight items was for at least 5 min. The mean of the two readings was used used to replace the missing item. If more than one item was for analysis. Waist circumference was measured using a non- missing, the variable was considered missing. Participants elastic, flexible tape measure at the level midway between 1 3 924 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 the lower rib margin and the iliac crest. Waist circumference associations between depression and the MetS , we also was measured in duplicate, and a third measurement was tested for effect-modification by sex and age groups by add- taken if the difference between the first two measurements ing interaction terms to the regression models. However, was > 1 cm. In addition, all participants were asked to bring no significant effect modification by age or sex was found their prescribed medications to the research location, which in our study (data not shown). IBM SPSS Statistics version were identified and categorized using the Anatomical Thera- 24.0 was used for statistical analysis. p values of < 0.05 were peutic Chemical (ATC) classification system . deemed statistically significant. Covariates Results Educational level was used as an indicator of socioeconomic status. It was defined as the highest qualification obtained in Sample characteristics the Netherlands or in the country of origin. The categories were: (1) no education or elementary education only, (2) Table 1 shows the characteristics of the study population, lower vocational or general secondary education, (3) inter- for the total sample (n = 21,182) and stratified by ethnicity. mediate vocational or higher secondary education and (4) In all ethnic groups, the percentage of women was larger higher vocational education or university. Current smoking than men. The mean age was 44.2 (SD = 13.2) years, but the (yes/no) and heavy alcohol use (yes/no) were determined by Turkish and Moroccan groups were slightly younger than the questionnaire. Heavy alcohol use was defined as consum - other ethnic groups. The Dutch were most often highly edu- ing more than two alcoholic beverages on more than two cated. The PHQ-9 and PTSD sum scores show the highest occasions per week. Physical activity (PA) was self-reported mean scores among the South-Asian Surinamese, Turkish using the Short Questionnaire to Assess Health-Enhancing and Moroccan groups. The smoking rate was remarkably Physical Activity (SQUASH) questionnaire . Adequate low among Ghanaians. The Dutch are heavy alcohol users PA (yes/no) is defined as reaching the international goal for most often, and are most often compliant to the PA norm. PA (moderate- to high-intensity PA for at least 30 min per Figure 1 shows the prevalence of the main variables: day on at least five days per week). depressed mood, severe PTSD symptoms and the MetS. Turks, Moroccans and South-Asian Surinamese had the Statistical analyses highest prevalence of a depressed mood, followed by the African Surinamese and Ghanaians. A depressed mood was Descriptive statistics were used to calculate percentages lowest in the Dutch. Severe PTSD symptoms followed the and means with standard deviations for demographic vari- same pattern, with the Turks having the highest prevalence, ables, main variables and covariates. The association of a and the Dutch having the lowest. The MetS was most com- depressed mood and severe PTSD symptoms with the MetS mon in the South-Asian Surinamese, followed by the Turks, was investigated using logistic regression analyses. The the African Surinamese, then the Moroccans and Ghanaians independent variables were a depressed mood (PHQ-9 sum and finally the Dutch. When looking at the individual com- score ≥ 10) or severe PTSD symptoms (sum score ≥ 7), and ponents of the MetS (Table 1), elevated fasting glucose was the dependent variable was the MetS. We adjusted for con- mostly seen in the South-Asian Surinamese, while elevated founding by sex, age, ethnicity and educational level. To blood pressure mostly presented itself in the Ghanaians and account for comorbidity of a depressed mood and severe African Surinamese. South-Asian Surinamese and the Turks PTSD symptoms, both variables were added to the final had the highest prevalence of dyslipidemia. model. The correlation between the continuous PHQ-9 sum score and PTSD sum score was poor (Spearman’s rho = 0.44, Main analyses p < 0.01), thus ruling out multicollinearity. Additionally, we tested for effect modification by ethnicity by adding inter - Table 2 shows the association between a depressed mood and action terms to the regression models (using the Dutch as the MetS in the total sample and by ethnicity. This associa- the reference population). Additionally, interaction analy- tion was consistent across ethnic groups, after adjustment for ses were performed to examine whether a depressed mood age and sex (model 1) and additional adjustment for educa- and severe PTSD symptoms interact in the association with tional level (model 2). The association between a depressed MetS. Finally, we adjusted for lifestyle factors (current mood and the MetS did not reach statistical significance in smoking, heavy alcohol use and compliance to PA norm) the Ghanaians, but it was not significantly different from the to study whether these variables attenuate the associations association in the Dutch population when tested (p-value of a depressed mood and severe PTSD symptoms with the interaction 0.18). Adjusting for severe PTSD symptoms MetS. Because some studies report sex- or age-specific attenuated these results slightly (model 3). After adjusting 1 3 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 925 Table 1 Characteristics of the study population, in the total sample and by ethnicity Total sample Dutch (n = 4527) South-Asian African Ghanaian Turkish Moroccan (n = 21,182) Surinamese Surinamese (n = 2251) (n = 3522) (n = 3825) (n = 2999) (n = 4058) Female (%) 57.7 54.1 54.6 61.0 61.2 54.9 61.2 Age in years, 44.21 (13.22) 46.18 (14.04) 45.44 (13.41) 47.87 (12.55) 44.64 (11.21) 40.27 (12.16) 40.41 (12.93) mean (SD) Education (%) 1 (lowest) 17.6 3.3 14.2 5.5 28.6 31.3 31.0 2 26.2 14.2 33.3 35.7 40.0 25.0 17.8 3 29.2 21.9 29.3 35.8 25.1 28.7 33.6 4 (highest) 27.0 60.7 23.2 23.1 6.2 15.0 17.6 PHQ-9 sum 4.76 (5.20) 3.57 (3.75) 5.41 (5.84) 3.92 (4.61) 3.37 (4.31) 6.42 (6.01) 5.86 (5.68) score, Mean (SD) PTSD sum score, 1.40 (2.40) 0.90 (1.79) 1.72 (2.64) 1.22 (2.20) 1.03 (1.99) 1.92 (2.79) 1.67 (2.69) Mean (SD) Lifestyle factors (%) Current smok- 24.0 24.7 28.4 31.6 4.4 34.6 13.4 ing Heavy alcohol 11.3 33.4 8.5 9.2 3.8 3.2 1.5 use Compliant to 56.7 75.7 53.4 61.4 53.4 42.1 47.0 PA norm MetS components (%) Reduced 31.6 19.6 46.0 26.9 20.1 42.4 36.0 HDL-C Elevated tri- 19.3 17.2 32.0 14.6 10.4 26.5 15.3 glycerides Elevated fast- 30.5 25.5 41.9 30.4 28.5 28.5 30.7 ing glucose Elevated blood 46.5 38.8 51.6 60.4 66.7 38.6 32.1 pressure Elevated waist 64.9 53.6 70.1 65.2 69.1 69.6 67.1 circumference (1) no education or elementary education only, (2) lower vocational or general secondary education, (3) intermediate vocational or higher sec- ondary education and (4) higher vocational education or university for current smoking, heavy alcohol use and compliance to attenuated by additional adjustment for a depressed mood, the PA norm, the association did not change [adjusted odds and only remained significant among the Dutch (model 3). ratio (95% confidence interval) = 1.36 (1.23–1.50) in the After additional adjustment for current smoking, heavy total sample]. When taking the continuous PHQ-9 scale as alcohol use and compliance to the PA norm, the associa- a determinant, instead of the dichotomous depressed mood tion did not change [adjusted odds ratio (95% confidence variable, similar positive associations consistent across eth- interval) = 1.73 (1.07–2.78) among the Dutch]. nic groups were shown (see online resource table S2). Table 2 also shows the association between severe PTSD symptoms and the MetS in the total sample and Interaction between a depressed mood and severe by ethnicity. The South-Asian Surinamese, Turks and PTSD symptoms Moroccans showed weaker associations compared to the Dutch population (p values for interaction < 0.01; p = 0.01; A depressed mood and severe PTSD symptoms did not p < 0.01, respectively). In the Dutch and the African Suri- have a significant interaction in the association with the namese only, severe PTSD symptoms were significantly MetS, neither in the total sample nor in the individual associated with the MetS after adjustment for age, sex and ethnic groups (see online resource table S3). educational level (models 1 and 2). This association was 1 3 926 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 Fig. 1 Prevalence of depressed 50 mood (PHQ-9 sum score ≥ 10), severe PTSD symptoms (PTSD sum score ≥ 7) and the MetS, by ethnicity. Asterisk: Prevalence is significantly higher than in the Dutch, after adjustment for age, sex and educational status (p value < 0.05) Dutch * * South-Asian Surinamese African Surinamese Ghanaian Turkish * Moroccan Depressed mood Severe PTSD Metabolic syndrome symptoms Table 2 Association (ORs with 95% CI) of a depressed mood (PHQ-9 sum score ≥ 10) and severe PTSD symptoms (PTSD sum score ≥ 7) with the MetS, in the total sample and by ethnicity Total sample Dutch (n = 4527) South-Asian African Ghanaian Turkish Moroccan (n = 21,182) Surinamese Surinamese (n = 2251) (n = 3522) (n = 3825) (n = 2999) (n = 4058) Depressed mood Model 1 1.24 (1.13–1.36) 1.80 (1.35–2.40) 1.48 (1.20–1.82) 1.73 (1.38–2.18) 1.13 (0.81–1.57) 1.28 (1.07– 1.26 (1.04–1.53) 1.54)* Model 2 1.17 (1.06–1.28) 1.52 (1.14–2.04) 1.43 (1.15–1.76) 1.67 (1.33–2.10) 1.12 (0.80–1.55) 1.22 (1.02–1.46) 1.24 (1.03–1.51) Model 3 1.37 (1.24–1.51) 1.37 (1.01–1.87) 1.57 (1.24–1.98) 1.64 (1.29–2.10) 1.09 (0.77–1.54) 1.22 (1.01–1.49) 1.32 (1.07–1.63) Severe PTSD symptoms Model 1 1.16 (1.02–1.31) 2.33 (1.51–3.61) 0.98 (0.75– 1.43 (1.04–1.96) 1.23 (0.73–2.05) 1.12 (0.89– 0.96 (0.74–1.24)* 1.29)* 1.42)* Model 2 1.11 (0.98–1.25) 1.97 (1.26–3.09) 0.95 (0.72– 1.37 (1.00-1.88) 1.20 (0.72–2.01) 1.09 (0.86– 0.94 (0.72–1.21)* 1.25)* 1.38)* Model 3 0.94 (0.82–1.08) 1.71 (1.07–2.73) 0.74 (0.55– 1.07 (0.76–1.50) 1.15 (0.67–1.98) 0.98 (0.76– 0.81 (0.61–1.07)* 1.01)* 1.27)* Model 1: adjusted for age and sex (and for ethnicity in the total sample only) Model 2: adjusted for age, sex and educational level (and for ethnicity in the total sample only) Model 3: adjusted for age, sex, educational level and comorbid depressed mood or severe PTSD symptoms (and for ethnicity in the total sample only) p value < 0.05 are in bold *Association is significantly different compared to the association in the Dutch population (p value interaction < 0.05) mood had increased odds of the MetS, consistent across Discussion ethnic groups. Participants with severe PTSD symptoms had increased odds of the MetS in the Dutch host popula- This study is unique in that it examines the association of tion, but this association was not observed in the ethnic both symptoms of depression and PTSD with the MetS minority groups. Comorbidity of a depressed mood and across large samples of six diverse ethnic groups living in severe PTSD symptoms did not moderate the association Amsterdam, the Netherlands. Participants with a depressed between the two mental disorders and the MetS. 1 3 Prevalence (%) Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 927 This study extends the evidence for the association caution given certain limitations. First, the PTSD question- between depressive symptoms and the MetS , by demon- naire is not yet validated. While the questions were derived strating consistency of this association across ethnic groups. from another validated questionnaire , the (transcultural) While some previous studies in the US suggest that ethnic validity of this selection of questions is unclear and should minorities (African-Americans) have stronger associations be investigated. Second, a limitation of this study is its cross- between depression and cardiovascular disease [13, 16–18, sectional nature, which implies we should be careful making 50], this was not observed in the association with the MetS causal inferences. There may be a bidirectional association in our population. This discrepancy may be caused by the between depression and the MetS . However, a large differences in migration background and living conditions longitudinal study showed depressive symptoms predicting between ethnic groups in the US and Europe . MetS development, but MetS components did not predict This study adds to previous studies on the association depressive symptoms . Similarly, a large longitudinal between PTSD and the MetS, by investigating this associa- study showed that PTSD symptoms predict subsequent MetS tion in a diverse population, in terms of sex and ethnicity. development, but the MetS did not predict development of Evidence derived from previous studies, with populations PTSD symptoms . The lack of an association between consisting mostly of male, white American veterans [19–25], the MetS and PTSD development is also shown in another suggests a strong association between PTSD and the MetS, large study . However, it has been hypothesized that independent of depression . However, we only observed inflammation (i.e., higher C-reactive protein levels), which an association in the Dutch host population, but not in the often coincides with the MetS, predicts PTSD symptoms in ethnic minority groups. This is in line with a previous study veterans , so there may be shared vulnerability underly- that showed an association between PTSD and metabolic ing the link between PTSD and the MetS. risk factors (BMI, obesity rates, abdominal obesity and tri- Future studies should replicate the findings of this study glycerides) in white Americans, but not in African-Ameri- longitudinally and investigate the pathways in the link cans . The results from this study lead us to suggest that between mental health and the MetS. Because increased in ethnic minority groups, the risk of developing the MetS MetS risk is seen in several psychiatric patient populations, may be more strongly influenced by other risk factors, such general mechanisms may link poor mental health to the as depressive symptoms, disadvantageous biological factors MetS . Our findings suggest that an increased risk of that originate in early life  or ethnicity-specific dietary the MetS is not explained by an unhealthy lifestyle that may patterns  andexercise beliefs . be a result of suffering from a mental disorder. However, we Because we investigated both a depressed mood and did not assess all lifestyle factors (e.g., nutrition and seden- severe PTSD symptoms in the association with the MetS, we tary behavior) in much detail. Another pathway is biologi- were able to test the interaction between these mental disor- cal stress dysregulation through disruption of the hypotha- ders. We found no evidence that the association between a lamic–pituitary–adrenal (HPA) axis [9, 59]. The disruption depressed mood and severe PTSD symptoms and the MetS of the HPA axis has been linked to metabolic and cardiovas- is moderated by comorbidity of the two disorders. This n fi d - cular disorders [60, 61]. Interestingly, persons with PTSD ing is at odds with a small previous study that reported an generally show lower cortisol levels, whereas in depressed increased risk of the MetS when PTSD is comorbid with persons these levels are generally higher [59, 62]. Another depression .This study, however, did not use statistical possible pathway is through inflammatory processes. Several analysis to test this interaction. Another study examining studies show that persons suffering from PTSD or depres- PTSD symptoms and onset of cardiovascular events did sion show higher level of inflammatory markers [8 , 63], but not find an interaction with depression . In contrast, a this has also been observed preceding trauma and PTSD study examining the association between PTSD and diabetes development . Similarly, there is evidence that suggests among asylum seekers in the Netherlands, found an interac- shared vulnerability of depression and the MetS . Evi- tion with comorbid depression: they only found an associa- dently, the link between mental health and the MetS remains tion between PTSD and diabetes in the group that was not complex, with multiple pathways acting synergistically . depressed . This study has several strengths. This study is the first European study to describe both depression and PTSD Conclusions symptomatology in relation to the MetS among several eth- nic groups, including large sample sizes and detailed infor- In conclusion, this study shows a consistent association mation on psychological, biological and anthropometric between a depressed mood and the MetS across ethnic measurements. With over 20,000 participants, this study is groups. In contrast, the evidence for the association between well powered, even after stratifying by ethnicity. Neverthe- PTSD symptoms and the MetS was only observed in the less, the results of this study need to be interpreted with Dutch participants and not among ethnic minority groups, 1 3 928 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 5. Conway KP, Green VR, Kasza KA et al (2017) Co-occurrence of suggesting this association may be ethnicity dependent. tobacco product use, substance use, and mental health problems When a depressed mood and severe PTSD symptoms were among adults: findings from Wave 1 (2013–2014) of the Popu- comorbid, the association with the MetS was not different. lation Assessment of Tobacco and Health (PATH) Study. Drug Future studies should replicate these findings and include Alcohol Depend 177:104–111. https ://doi.org/10.1016/j.dr ug a lcdep .2017.03.032 more factors that may shed light on possible mechanisms 6. Penninx BWJH. (2017) Depression and cardiovascular disease: underlying the link with the MetS. epidemiological evidence on their linking mechanisms. Neuro- sci Biobehav Rev 74:277–286. https: //doi.org/10.1016/j.neubi Acknowledgements The HELIUS study is conducted by the Aca- orev.2016.07.003 demic Medical Center Amsterdam and the Public Health Service of 7. Steudte-Schmiedgen S, Kirschbaum C, Alexander N, Stalder Amsterdam. Both organizations provided core support for HELIUS. T (2016) An integrative model linking traumatization, cortisol The HELIUS study is also funded by the Dutch Heart Foundation dysregulation and posttraumatic stress disorder: insight from (2010T084), the Netherlands Organization for Health Research and recent hair cortisol findings. Neurosci Biobehav Rev 69:124– Development (ZonMw) (200500003), the European Union (FP- 135. https ://doi.org/10.1016/j.neubi orev.2016.07.015 7) (278901), and the European Fund for the Integration of non-EU 8. Lamers F, Milaneschi Y, de Jonge P et al (2017) Metabolic and immigrants (EIF) (2013EIF013). We acknowledge the AMC Biobank inflammatory markers: associations with individual depressive for their support in biobank management and high-quality storage symptoms. Psychol Med. https ://doi.org/10.1017/S0033 29171 of collected samples. We are most grateful to the participants of the 70024 83 HELIUS study and the management team, research nurses, interview- 9. Olff M, van Zuiden M (2017) Neuroendocrine and neuroimmune ers, research assistants and other staff who have taken part in gathering markers in PTSD: pre-, peri- and post-trauma glucocorticoid the data of this study. and inflammatory dysregulation. Curr Opin Psychol 14:132–137 10. Vancampfort D, Correll CU, Wampers M et al (2014) Meta- bolic syndrome and metabolic abnormalities in patients with Compliance with ethical standards major depressive disorder: a meta-analysis of prevalences and moderating variables. Psychol Med 44:2017–2028. https ://doi. Conflict of interest The authors declare that they have no conflict of org/10.1017/S0033 29171 30027 78 interest. 11. Pan A, Keum N, Okereke OI et al (2012) Bidirectional associa- tion between depression and metabolic syndrome: a systematic Ethical approval The HELIUS study is conducted in accordance with review and meta-analysis of epidemiological studies. Diabetes the Declaration of Helsinki and has been approved by the AMC Ethical Care 35:1171–1180. https ://doi.org/10.2337/dc11-2055 Review Board. All participants provided written informed consent. 12. Hiles SA, Révész D, Lamers F et al (2016) Bidirectional pro- spective associations of metabolic syndrome components with depression, anxiety, and antidepressant use. Depress Anxiety Open Access This article is distributed under the terms of the Crea- 33:754–764. https ://doi.org/10.1002/da.22512 tive Commons Attribution 4.0 International License (http://creat iveco 13. Vogelzangs N, Beekman ATF, Kritchevsky SB et al (2007) Psy- mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- chosocial risk factors and the metabolic syndrome in elderly tion, and reproduction in any medium, provided you give appropriate persons: findings from the Health, Aging and Body Composi- credit to the original author(s) and the source, provide a link to the tion Study. J Gerontol Ser A Biol Sci Med Sci 62:563–569. https Creative Commons license, and indicate if changes were made. ://doi.org/10.1093/geron a/62.5.563 14. Kinder LS, Carnethon MR, Palaniappan LP et al (2004) Depres- sion and the metabolic syndrome in young adults: findings from the Third National Health and Nutrition Examination Survey. Psychosom Med 66:316–322 References 15. Dedert EA, Harper LA, Calhoun PS et al (2013) The impact of race on metabolic disease risk factors in women with and with- 1. Alberti KGMM, Eckel RH, Grundy SM et al (2009) Harmonizing out posttraumatic stress disorder. J Clin Psychol Med Settings the metabolic syndrome: a joint interim statement of the Inter- 20:46–55. https ://doi.org/10.1007/s1088 0-012-9305-4 national Diabetes Federation Task Force on Epidemiology and 16. Sims M, Redmond N, Khodneva Y et al (2015) Depressive Prevention; National Heart, Lung, and Blood Institute; American symptoms are associated with incident coronary heart disease Heart Association; World Heart Federation. Int Circ 120:1640– or revascularization among Blacks but not among Whites in 1645. https ://doi.org/10.1161/CIRCU LATIO NAHA.109.19264 4 the Reasons for Geographical and Racial Differences in Stroke 2. DeBoer MD, Gurka MJ, Morrison JA, Woo JG (2016) Inter-rela- (REGARDS) Study. Ann Epidemiol 25:426–432. https ://doi. tionships between the severity of metabolic syndrome, insulin org/10.1016/j.annep idem.2015.03.014 and adiponectin and their relationship to future type 2 diabetes 17. Lewis TT, Guo H, Lunos S et al (2011) Depressive symptoms and cardiovascular disease. Int J Obes 40:1353–1359. https://doi. and cardiovascular mortality in older African-American and org/10.1038/ijo.2016.81 white adults: evidence for a differential association by race. Circ 3. Rosenbaum S, Stubbs B, Ward PB et al (2015) The prevalence Cardiovasc Qual Outcomes 4:293–299. https://doi.or g/10.1161/ and risk of metabolic syndrome and its components among people CIRCO UTCOM ES.110.95754 8 with posttraumatic stress disorder: a systematic review and meta- 18. Read JG, Gorman BK (2007) Racial/ethnic differences in analysis. Metabolism 64:926–933. https://doi.or g/10.1016/j.metab hypertension and depression among US adult women. Ethn Dis ol.2015.04.009 17:389–396 4. Ghanei Gheshlagh R, Parizad N, Sayehmiri K (2016) The rela- 19. Wolf EJ, Bovin MJ, Green JD et al (2016) Longitudinal asso- tionship between depression and metabolic syndrome: system- ciations between post-traumatic stress disorder and metabolic atic review and meta-analysis study. Iran Red Crescent Med J syndrome severity. Psychol Med 46:2215–2226. https ://do i. 18:e26523. https ://doi.org/10.5812/ircmj .26523 org/10.1017/S0033 29171 60008 17 1 3 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 929 20. Babić D, Jakovljević M, Martinac M et al (2007) Metabolic and depression in a sample of adolescents exposed to an explo- syndrome and combat post-traumatic stress disorder intensity: sion accident. Eur J Psychotraumatol 8:1272789. https ://doi. preliminary findings. Psychiatr Danub 19:68–75org/10.1080/20008 198.2016.12727 89 21. Jakovljević M, Sarić M, Nad S et al (2006) Metabolic syndrome, 38. Weiss T, Skelton K, Phifer J et al (2011) Posttraumatic stress dis- somatic and psychiatric comorbidity in war veterans with post- order is a risk factor for metabolic syndrome in an impoverished traumatic stress disorder: Preliminary findings. Psychiatr Danub urban population. Gen Hosp Psychiatry 33:135–142. https ://doi. 18:169–176org/10.1016/j.genho sppsy ch.2011.01.002 22. Maslov B, Jakovljević M, Crncević Z et al (2008) Metabolic 39. Bartoli F, Crocamo C, Clerici M, Carra G (2015) The associa- syndrome and schizophrenia from integrative medicine perspec- tion between PTSD and metabolic syndrome: a role for comorbid tive. Psychiatr Danub 20:384–389 depression? Metabolism 64:1373–1375 23. Linnville S, Hoyt RE, Moore JL et al (2011) Posttraumatic stress 40. Phillips AC, Batty GD, Gale CR et al (2009) Generalized anxi- disorder and metabolic syndrome: retrospective study of repatri- ety disorder, major depressive disorder, and their comorbidity as ated prisoners of war. Mil Med 176:369–374 predictors of all-cause and cardiovascular mortality: the Viet- 24. Heppner PS, Crawford EF, Haji UA et al (2009) The associa- nam experience study. Psychosom Med 71:395–403. https ://doi. tion of posttraumatic stress disorder and metabolic syndrome: a org/10.1097/PSY.0b013 e3181 9e670 6 study of increased health risk in veterans. BMC Med 7:1. https 41. Carroll D, Phillips AC, Gale CR, Batty GD (2010) Generalized ://doi.org/10.1186/1741-7015-7-1 anxiety and major depressive disorders, their comorbidity and 25. Jakovljević M, Babić D, Crncević Z et al (2008) Metabolic syn- hypertension in middle-aged men. Psychosom Med 72:16–19. drome and depression in war veterans with post-traumatic stress https ://doi.org/10.1097/PSY.0b013 e3181 c4fca 1 disorder. Psychiatr Danub 20:406–410 42. Stronks K, Snijder MB, Peters RJ et al (2013) Unravelling the 26. Nobles CJ, Valentine SE, Borba CPC et al (2016) Black-white impact of ethnicity on health in Europe: the HELIUS study. BMC disparities in the association between posttraumatic stress dis- Public Health 13:402. https: //doi.org/10.1186/1471-2458-13-402 order and chronic illness. J Psychosom Res 85:19–25. https :// 43. Stronks K, Kulu-Glasgow I, Agyemang C (2009) The utility of doi.org/10.1016/j.jpsyc hores .2016.03.126 “country of birth” for the classification of ethnic groups in health 27. Alegría M, Fortuna LR, Lin JY et al (2013) Prevalence, risk, research: the Dutch experience. Ethn Health 14:255–269. https:// and correlates of posttraumatic stress disorder across ethnic and doi.org/10.1080/13557 85080 25092 06 racial minority groups in the United States. Med Care 51:1114– 44. Kroenke K, Spitzer RL, Williams JBW, Löwe B (2010) The 1123. https ://doi.org/10.1097/MLR.00000 00000 00000 7 Patient Health Questionnaire Somatic, Anxiety, and Depres- 28. Fassaert T, Peen J, van Straten A et al (2010) Ethnic differ - sive Symptom Scales: a systematic review. Gen Hosp Psychiatry ences and similarities in outpatient treatment for depression 32:345–359. https://doi.or g/10.1016/j.genhosppsy c h.2010.03.006 in the Netherlands. Psychiatr Serv 61:690–697. https ://doi. 45. Galenkamp H, Stronks K, Snijder MB, Derks EM (2017) Meas- org/10.1176/ps.2010.61.7.690 urement invariance testing of the PHQ-9 in a multi-ethnic popula- 29. Ikram UZ, Snijder MB, Fassaert TJL et al (2015) The contri- tion in Europe: the HELIUS study. BMC Psychiatry 17:349. https bution of perceived ethnic discrimination to the prevalence ://doi.org/10.1186/s1288 8-017-1506-9 of depression. Eur J Public Health 25:243–248. https ://doi. 46. Foa EB, Riggs DS, Dancu CV, Rothbaum BO (1993) Reliabil- org/10.1093/eurpu b/cku18 0 ity and validity of a brief instrument for assessing post-trau- 30. Ikram UZ, Snijder MB, Agyemang C et al (2017) Perceived matic stress disorder. J Trauma Stress 6:459–473. https ://doi. Ethnic Discrimination and the Metabolic Syndrome in Eth-org/10.1002/jts.24900 60405 nic Minority Groups: The Healthy Life in an Urban Setting 47. WHO Collaborating Centre for Drug Statistics Methodology Study. Psychosom Med 79:101–111. https ://doi.org/10.1097/ (2013) Guidelines for ATC Classification and DDD Assignment PSY.00000 00000 00035 0 2012. Oslo 31. Williams ED, Nazroo JY, Kooner JS, Steptoe A (2010) Sub- 48. Wendel-Vos GCW, Schuit AJ, Saris WHM, Kromhout D (2003) group differences in psychosocial factors relating to coronary Reproducibility and relative validity of the short questionnaire heart disease in the UK South Asian population. J Psycho- to assess health-enhancing physical activity. J Clin Epidemiol som Res 69:379–387. https ://doi.or g/10.1016/j.jpsy c hor es 56:1163–1169. https ://doi.org/10.1016/S0895 -4356(03)00220 -8 .2010.03.015 49. Block A, Schipf S, Van der Auwera S et al (2016) Sex- and age- 32. Gualdi-Russo E, Zironi A, Dallari GV, Toselli S (2009) Migra- specific associations between major depressive disorder and meta- tion and health in Italy: a multiethnic adult sample. J Travel Med bolic syndrome in two general population samples in Germany. 16:88–95. https ://doi.org/10.1111/j.1708-8305.2008.00280 .x Nord J Psychiatry 70:611–620. https ://doi.or g/10.1080/08039 33. Skogberg N, Laatikainen T, Jula A et al (2017) Contribution of 488.2016.11915 35 sociodemographic and lifestyle-related factors to the differences 50. Assari S, Sonnega A (2016) Racial differences in the predictive in metabolic syndrome among Russian, Somali and Kurdish role of high depressive symptoms on incident heart disease over migrants compared with Finns. Int J Cardiol 232:63–69. https :// 18 years: results from the Health and Retirement Study. Res Car- doi.org/10.1016/j.ijcar d.2017.01.051 diovasc Med. https ://doi.org/10.5812/cardi ovasc med.34767 34. Marmot M (2016) Society and health of migrants. Eur J Epidemiol 51. Gishti O, Gaillard R, Felix JF et al (2015) Early origins of eth- 31:639–641. https ://doi.org/10.1007/s1065 4-016-0174-2 nic disparities in cardiovascular risk factors. Prev Med (Baltim) 35. Farhood LF, Fares S, Sabbagh R, Hamady C (2016) PTSD and 76:84–91. https ://doi.org/10.1016/j.ypmed .2015.03.027 depression construct: prevalence and predictors of co-occurrence 52. Dekker LH, van Dam RM, Snijder MB et al (2015) Comparable in a South Lebanese civilian sample. Eur J Psychotraumatol dietary patterns describe dietary behavior across Ethnic Groups in 7:31509 the Netherlands, but different elements in the diet are associated 36. Horesh D, Lowe SR, Galea S et al (2017) An in-depth look into with glycated hemoglobin and fasting glucose concentrations. J PTSD-depression comorbidity: A longitudinal study of chroni- Nutr 145:1884–1891. https ://doi.org/10.3945/jn.114.20747 2 cally-exposed Detroit residents. J Affect Disord 208:653–661. 53. Romeike K, Abidi L, Lechner L et al (2016) Similarities and dif- https ://doi.org/10.1016/j.jad.2016.08.053 ferences in underlying beliefs of socio-cognitive factors related to diet and physical activity in lower-educated Dutch, Turkish, and 37. Yang H, Wang L, Cao C et al (2017) The underlying dimen- sions of DSM-5 PTSD symptoms and their relations with anxiety 1 3 930 Social Psychiatry and Psychiatric Epidemiology (2018) 53:921–930 Moroccan adults in the Netherlands: a focus group study. BMC 59. Penninx BW, Milaneschi Y, Lamers F, Vogelzangs N (2013) Public Health 16:813. https://doi.or g/10.1186/s12889-016-3480-4 Understanding the somatic consequences of depression: biologi- 54. Sumner JA, Kubzansky LD, Elkind MSV et al (2015) Trauma cal mechanisms and the role of depression symptom profile. BMC exposure and posttraumatic stress disorder symptoms predict Med 11:129. https ://doi.org/10.1186/1741-7015-11-129 onset of cardiovascular events in women. Circulation 132:251– 60. Martins CS, Elias D, Colli LM et al (2016) HPA axis dysregu- 259. https ://doi.org/10.1161/CIRCU LATIO NAHA.114.01449 2 lation, NR3C1 polymorphisms and glucocorticoid receptor iso- 55. Agyemang C, Goosen S, Anujuo K, Ogedegbe G (2012) Relation- forms imbalance in metabolic syndrome. Diabetes Metab Res Rev ship between post-traumatic stress disorder and diabetes among 33:e2842. https ://doi.org/10.1002/dmrr.2842 105 180 asylum seekers in the Netherlands. Eur J Public Health 61. Kinlein SA, Wilson CD, Karatsoreos IN (2015) Dysregulated 22:658–662. https ://doi.org/10.1093/eurpu b/ckr13 8 hypothalamic–pituitary–adrenal axis function contributes to 56. Chang J-C, Yen AM-F, Chen H-H et al (2017) Comorbid diseases altered endocrine and neurobehavioral responses to acute stress. as risk factors for incident posttraumatic stress disorder (PTSD) Front Psychiatry 6:31. https ://doi.org/10.3389/fpsyt .2015.00031 in a large community cohort (KCIS no.PSY4). Sci Rep 7:41276 62. Farr OM, Sloan DM, Keane TM, Mantzoros CS (2014) Stress- 57. Eraly SA, Nievergelt CM, Maihofer AX et al (2014) Assess- and PTSD-associated obesity and metabolic dysfunction: a grow- ment of plasma C-reactive protein as a biomarker of PTSD Risk. ing problem requiring further research and novel treatments. JAMA psychiatry 71:423–431. https ://doi.or g/10.1001/jamap sy c hi Metabolism 63:1463–1468. https ://doi.or g/10.1016/j.me t ab atry.2013.4374 ol.2014.08.009 58. Penninx BW, Lange SM (2018) Metabolic syndrome in psychiatric 63. O’Donovan A (2016) PTSD is associated with elevated inflam- patients: overview, mechanisms and implications. Dialogues Clin mation. Any impact on clinical practice? Evid Based Ment Health Neurosci 20:75–84 19:120. https ://doi.org/10.1136/eb-2016-10237 6 1 3
Social Psychiatry and Psychiatric Epidemiology – Springer Journals
Published: May 23, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera