Psychosocial Variables Are Associated With Atherosclerosis Risk Factors Among Women With Chest Pain: The WISE Study THOMAS RUTLEDGE,PHD, STEVEN E. REIS, MD, MARIAN OLSON, MS, JANE OWENS,PHD, SHERYL F. KELSEY,PHD, CARL J. PEPINE, MD, NATHANIEL REICHEK, MD, WILLIAM J. ROGERS, MD, C. NOEL BAIREY MERZ, MD, GEORGE SOPKO, MD, CAROL E. CORNELL,PHD, AND KAREN A. MATTHEWS,PHD Objective: We investigated associations between atherosclerosis risk factors (smoking behavior, serum cholesterol, hypertension, body mass index, and functional capacity) and psychological characteristics with suspected linkages to coronary disease (depression, hostility, and anger expression) in an exclusively female cohort. Methods: Six hundred eighty-eight middle-aged women with chest pain warranting clinical investigation completed a compre- hensive diagnostic protocol that included quantitative coronary angiography to assess coronary artery disease (CAD). Primary analyses controlled for menopausal status, age, and socioeconomic status variables (income and education). Results: High depression scores were associated with a nearly three-fold risk of smoking (odds ratio (OR) = 2.8, 95% confidence interval (CI) = 1.4 –5.7) after covariate adjustment, and women reporting higher depression symptoms were approximately four times more likely to describe themselves in the lowest category of functional capacity (OR = 3.7, 95% CI = 1.7–7.8). High anger-out scores were associated with a four-fold or greater risk of low high-density lipoprotein cholesterol concentration (50 mg/dl; OR = 4.0, 95% CI = 1.4 –11.1) and high low-density lipoprotein cholesterol concentration (160 mg/dl; OR = 4.8, 95% CI = 1.5–15.7) and a larger body mass index (OR = 3.5, 95% CI = 1.1–10.8) after covariate adjustment. Conclusions: These results demonstrate consistent and clinically relevant relationships between psychosocial factors and atherosclerosis risk factors among women and may aid our understanding of the increased mortality risk among women reporting high levels of psychological distress. Key words: coronary artery disease, women, psychosocial risk factors. BDI = Beck Depression Inventory; BMI = body mass index; CAD = coronary artery disease; CI = confidence interval; HDL-C = high-density lipoprotein cholester- ol; LDL-C = low-density lipoprotein cholesterol; OR = odds ratio; WISE = Women’s Ischemia Syndrome Evaluation. INTRODUCTION There is compelling literature to support the asso- ciations between specific psychosocial characteristics and the incidence and prognosis of CAD (1– 4). Re- views of this research (5–7) cite particularly strong evidence for four psychological factors: depression, hostility, and two types of anger expression (ie, an- ger-in and anger-out). The predictive value of these four factors has been demonstrated across both cross- sectional and prospective studies of CAD, in cohorts of healthy subjects as well as subjects with disease, and the effects have proven robust after controlling for more established risk factors, such as age, family his- tory, and disease severity (8 –10). The rationale for the relationship between psycho- social variables and CAD commonly takes one of two forms. First, a series of recent findings support adverse psychophysiological effects relevant to CAD, such as increased responsiveness of the sympathetic nervous system (ie, increased blood pressure and heart rate activity) and hypothalamic-pituitary axis (eg, hyper- cortisolemia) under conditions of mental stress (11, 12). Second, psychological variables may impact the course of coronary disease through behavioral mecha- nisms, for example, by fostering negative health behav- iors such as smoking, decreased fitness or physical activity, poor diet, and reduced adherence to treatment (13). Although many previous psychosocial studies of coronary disease have included measures of behav- ioral and physiological risk factors (ie, cholesterol lev- els and BMI) as covariates, with the possible exception of hypertension (14, 15), less is known about direct associations between psychosocial variables and major coronary risk factors among traditionally understudied female groups (16 –18). In light of evidence from male patient groups suggesting that psychosocial effects on cardiovascular disease may be in part mediated by atherosclerosis risk factors (19), it is important to ex- amine this issue in women. To evaluate the relationship between psychosocial traits and atherosclerosis risk factors, we studied women undergoing evaluation for suspected myocar- dial ischemia enrolled in the National Heart, Lung, From the University of Pittsburgh (T.R., S.E.R., M.O., J.O., K.A.M.), Pittsburgh, Pennsylvania; University of Florida at Gaines- ville (S.F.K., C.J.P.), Gainesville, Florida; Allegheny General Hospi- tal (N.R.), Pittsburgh, Pennsylvania; University of Alabama at Bir- mingham (W.J.R., C.E.C.), Birmingham, Alabama; Cedars-Sinai Medical Center (C.N.B.M.), Los Angeles, California; and National Heart, Lung, and Blood Institute (G.S.), Bethesda, Maryland. Address reprint requests to: Thomas Rutledge, PhD, c/o WISE Coordinating Center, 127 Parran Hall, 130 DeSoto St., Pittsburgh, PA 15261. Email: dr.tom@musclemail.com Received for publication April 3, 2000; revision received August 10, 2000. 282 Psychosomatic Medicine 63:282–288 (2001) 0033-3174/01/6302-0282 Copyright © 2001 by the American Psychosomatic Society