DO DEPRESSIVE SYMPTOMS MEDIATE OR MODERATE THE RELATIONSHIP OF GENDER TO METABOLIC SYNDROME?

Jennifer Saylor, PhD, RN, ANCS-BC

University of Delaware

Metabolic Syndrome (MetS) A cluster of medical disorders that characteristically

occur together Obesity Hypertension Insulin Resistance Dyslipidemia (low HDL & triglycerides)

Researchers continue to study its many influencing factors

Biomedical Factors

Health Status Psychosocial Factors

Biosocial Factors

Age Pathophysiologic

Gender Race Education Income Marital status

Depressive symptoms Physical activity Diet intake Smoking Sleep

Metabolic Syndrome

“The Biopsychosocial Model of MetS in U.S. Adults” based on The Concentric Biopsychosocial Model of Health Status1

Pathophysiologic factors includes hypothalamic pituitary adrenal, sympathetic adrenal medullay, and C-reactive protein are factors not measured in the study

Depression: Background

Depressive symptoms differ than clinical depression Individual can have depressive symptoms [low score on

Patient Health Questionnaire-9 (PHQ-9)], but not be diagnosed with clinical depression

Roughly 6.7% of the population of the US is diagnosed with major depressive disorder- more than 14M Americans in any given year2 Only 39% of those with severe depression sought the help of a medical professional3

Depressive symptoms may be overlooked by the individuals themselves and healthcare providers Not routinely screen for mental health conditions or

depressive symptoms Symptoms (appetite changes, weight changes, and altered

sleep) mimic many other medical disorders

Relationship of Depression and Gender to Metabolic Syndrome: Background

A significantly greater prevalence of depression in females than males 4-8

A meta-analysis study showed a significantly higher rate of major depression among females than in males9

Using the Tel Aviv Medical Center Inflammation Survey, women had significantly higher levels of depression as compared to men (mean = 1.19 vs. 1.35, p = <.001)8

Lack of association between gender and MetS6, 10-11

Gender is not a significant predictor of MetS when examining a wide range of ages such a population based studies12-13

When gender is crossed with age, patterns emerge

Methods and Data Source

A secondary data analysis using the National Health and Nutrition Examination Survey (NHANES) 2007-2008 Conducted by the National Center for Health Statistics Combination if interviews and physical examinations

Data collected between January 07-December 08 Complex samples regression models were utilized

to examine the moderating and mediating effects of depressive symptoms on the relationship between gender and MetS

Sample

NHANES survey uses a stratified, multistage probability sampling. Representative sample of groups

by age, sex, & income level Each NHANES participant

represents approximately 50,000 other U.S. residents

NHANES 2007-2008 over-samples persons over 60 years of age African Americans Low income population Entire Hispanic population, not just

Mexican Americans14

The study sample of 2,583 adults represented 212 million non-institutionalized civilian Americans in the U.S.

N

• Metabolic Syndrome Study Inclusion Criteria

12,943

• Selected for 2007-2008 Survey

10,149

• Completed Interview (78.4%)

9,762

• Completed physical exam & interview (75.4%)

6,587

• Participants > 20 yrs. old & not pregnant

2,583

• Participants who completed AM labs

NHANES Sample Metabolic Syndrome Study Sample

Variable: Metabolic Syndrome

Multiple definitions yielding various prevalence rates

Diagnosis: individual has any 3 of the 5 criteria

Factors Defining Metabolic Syndrome

National Cholesterol Education Program’s Adult Treatment Panel III (NCEP ATP III-2001)

Obesity WC ≥ 102 cm in men or ≥88 cm in women

Blood Pressure ≥ 130 systolic or ≥ 85 diastolic or antihypertensive medication

Glucose >100 mg/dL or drug TX++ for elevated glucose

Triglycerides ≥ 150 mg/dL or medication for elevated triglycerides

High-density Lipoprotein

<40 mg/dl in men or <50 in women or TX of reduced HDL-C

Variable: Depression Symptoms

Over the last 2 weeks, how often have you been bothered by any of the following problems?

Not at all

Several

days

More than half the

days

Nearly

every day

1. Little interest pleasure in doing things 0 1 2 34. Feeling tired or having little energy 0 1 2 35. Poor appetite or overeating 0 1 2 37. Trouble concentrating

on things, such as reading the newspaper or watching television 0 1 2 3

9. Thoughts that you would be better off dead of hurting yourself in some way 0 1 2 3

10. If you checked off any problems, how difficult have these problems made it for you to do work, take care of things at home, or get along with other people?

 

Not difficult at all _____ Somewhat difficult _____

Very difficult _____ Extremely difficult _____

 

Depressive symptoms: level of severity of depressive symptoms as scored on PHQ-9

Scores ranged from 0-27; high the score, higher the depressive symptoms

Derived from the Primary Care Evaluation of Mental Disorders Brief Patient Health Questionnaire (PRIME-MD) Evaluated 18 mental disorders

divided into four groups (mood, anxiety, somatoform, and alcohol)

Questions are based on nine DSM-IV signs and symptoms from depression

Patient Health Questionnaire (PHQ-9)

Analysis: Moderation

Moderator: qualitative or quantitative variable that affects the direction and/or strength of the relationship between an independent variable or predictor variable (X) & a dependent variable or outcome variable (Y)15

Moderator hypotheses is supported- significant interaction (path c) Significant main effects for the predictor (X) and the moderator (Y),

but these are not directly relevant conceptually to testing the moderator hypothesis 15

Model for testing moderation

a

b

Predictor (X) Gender

Moderator (M) Depressive Symptoms

X x M

Outcome Variable (Y) Metabolic Syndrome

Analysis: Mediation Mediation: “the generative mechanism through which the focal

independent variable is able to influence the dependent variable of interest.”15

Predictor (X): related to mediator (M) & (M): related to the outcome (Y) Mediated effect may exist regardless of statistical significance of the

effect of the predictor on the outcome Relationship from X to Y does not go through M and is a direct effect of the

predictor variable to the outcome variable. MacKinnon 16 Mediation: If the b parameter is significant Total mediation: when the a and b parameter are significant17

Product of ab quantifies how much a one unit change in X affects Y indirectly through M Model for testing mediation

a b

c’

Predictor (X) Gender

Mediator (M) Depressive Symptoms

Outcome (Y) Metabolic Syndrome

Descriptive Results

  Population Size

FrequencyEstimate (%)

Range

Mean Estimate

(Std. Error)

95% CI

Age 2.121x108   20-80 46.91 (.60)

45.64-48.17

Family Income (PIR) 2.121x108   0-5 3.07 (.09) 2.89-3.25

Depressive Symptoms

1.994x108   0-27 3.00 (.16) 2.67-3.33

Gender 2.121x108        

Male   1.036 x 108 (48.8)

     

Female   1.085 x 108 (51.2)

     

Metabolic Syndrome   1.145x108 (54.1)      

Met the criteria   6.101x107 (28.8)      

Did not meet the criteria

  1.511x 108 (71.2)

     

Description of variables for the population represented by the NHANES 2007-2008 participants over 20 years of age, not pregnant and completed fasting labs

29% met criteria for metabolic syndrome Mean age: 47 years old and 51% were female

Note. Complex samples analysis used; PIR= poverty income ratio

Moderation Results

Using complex samples logistic regression analysis to examine moderating effects of depressive symptoms on the association between gender & MetS

Parameter B SE 95% CI Hypothesis Test ORLower Upp

erT Df Sig

.Depressive symptoms and gender

             

Intercept -1.301

.169 -1.659 -.944 -7.720 16 .000

.272

Gender (P) .373 .172 .009 .738 2.170 16 .045

1.453

Depressive Ssymptomsa (M)

.300 .114 .059 .541 2.636 16 .018

1.349

Gender*Depressive symptomsa (P*M)

-.228 .125 -.494 .038 -1.818 16 .088

.796

Note. Table includes only analyses resulting in interactions with p < .20; Dependent variable: dichotomized MetS (reference category is 0 for no MetS); a log transformed; P = predictor; M = moderator; P*M = interaction; * for p < .05 for moderating effect

Gender (OR=1.453, p=.045) and depression symptoms (p<.001, OR=1.19) After controlling for gender, those with elevated depressive symptoms

were 19% more likely to have MetS After controlling for depressive symptoms, males were 45% more likely

to have MetS Females with elevated depressive symptoms score were 1.5 times more

likely to have MetS than males with high depressive symptoms score

Conclusion

More than 95% of the sample had at least one criterion of MetS!

61 million people in the U.S met the criteria for MetS according to the NCEP-ATP III, 2001

Depressive symptoms but not mediate the relationship between gender and MetS 

The relationship between gender and MetS is strengthened by the moderating effects of depressive symptoms

Females with elevated depressive symptoms score were 1.5 times more likely to have MetS than males

A one point increase on the PHQ-9 resulted in a 28% increase in the odds of the presence of MetS

Reducing depression may prevent/reverse MetS and reduce associated morbidity and mortality

For further information, please contact me at

Jsaylor@udel.edu.

Thank you

Questions or Comments?

References 1 Hoffman, M. A., & Driscoll, J. M. (2000). Health promotion and

disease prevention: A concentric biopsychosocial model of health status. In S.D.Brown & & R.W.Lent (Ed.), Handbook of counseling psychology (3rd ed., pp. 532-567). New York: John Wiley & Sons, Inc.

2 National Institute of Mental Health. (2009). The numbers count: Mental disorders in Americaa. Retrieved August 10, 2012, from http://www.nimh.nih.gov/health/publications/the-numbers-count-mental-disorders-in-america/index.shtml#MajorDepressive

3 Pratt, L. A., & Brody, D. J. (2009). Depression in the united states household population, 2005-2006. Retrieved August 10, 2012, from http://www.cdc.gov/nchs/data/databriefs/db07.htm

4 Akbaraly, T., N., Kivimaki, M., Brunner, E.,J., Chandola, T., Marmot, M., G., Singh-Manoux, A., et al. (2009). Association between metabolic syndrome and depressive symptoms in middle-aged adults. Diabetes Care, 32, 499.

5 Gary, T. L., Crum, R. M., Cooper-Patrick, L., Ford, D., & Brancati, F. L. (2000). Depressive symptoms and metabolic control in african-americans with type 2 diabetes. Diabetes Care, 23(1), 23-29.

References

6 Kinder, L. S., Carnethon, M. R., Palaniappan, L. P., King, A. C., & Fortmann, S. P. (2004). Depression and the metabolic syndrome in young adults: Findings from the third national health and nutrition examination survey. Psychosomatic Medicine, 66(3), 316-322.

7 Skilton, M. R., Moulin, P., Terra, J. L., & Bonnet, F. (2007). Associations between anxiety, depression, and the metabolic syndrome. Biological Psychiatry, 62(11), 1251-1257.

8 Toker, S., Shirom, A., & Melamed, S. (2008). Depression and the metabolic syndrome: Gender-dependent associations. Depression & Anxiety (1091-4269), 25(8), 661-669.

9 Goldbacher, E. M., & Matthews, K. A. (2007). Are psychological characteristics related to risk of the metabolic syndrome? A review of the literature. Annals of Behavioral Medicine, 34(3), 240-252.

10 Dunbar, J. A., Reddy, P., Davis-Lameloise, N., Philpot, B., Laatikainen, T., Kilkkinen, A., et al. (2008). Depression: An important comorbidity with metabolic syndrome in a general population. Diabetes Care, 31(12), 2368-2373.

11 Remsberg, K. E., Rogers, N. L., Demerath, E. W., Czerwinski, S. A., Choh, A. C., Lee, M., et al. (2007). Sex differences in young adulthood metabolic syndrome and physical activity: The fels longitudinal study. American Journal of Human Biology, 19(4), 544-550.

References

12 Ervin, R. B. (2009). Prevalence of metabolic syndrome among adults 20 years of age and over, by sex, age, race and ethnicity, and body mass index: United states, 2003ΓÇô2006. Retrieved September/28, 2009, from http://www.cdc.gov/nchs/data/nhsr/nhsr013.pdf

13 Marquezine, G. F., Oliveira, C. M., Pereira, A. C., Krieger, J. E., & Mill, J. G. (2008). Metabolic syndrome determinants in an urban population from brazil: Social class and gender-specific interaction. International Journal of Cardiology, 129(2), 259-265.

14 National Center for Health Statistics. (2009). Key concepts about NHANES survey design. Retrieved September 23, 2009, from http://www.cdc.gov/nchs/tutorials/nhanes/surveydesign/SampleDesign/intro.htm

15 Baron, R. M., & Kenny, D. A. (1986). The moderator-mediator variable distinction in social psychological research: Conceptual, strategic and statistical considerations. . Journal of Personality and Social Psychology, 51, 1173-1182.

16 MacKinnon, D. P., & Dwyer, J. H. (1993). Estimating mediated effects in prevention studies. Evaluation Review, 17, 144-158.

17 MacKinnon, D. P. (2008). Introduction to statisitical mediation analysis. New York: Taylor & Francis Group, LLC.