1. Table 1Sample Characteristics Age (in years) M (SD)23.10 (+5.02) Range17 - 38 Total Household Income (in US dollars) < $5,000, n (%)5 (12.5) $5000 - 29,999, n (%)17 (42.5) $30,000 - 49,999, n (%)2 (5) $50,000 - 69,999, n (%)1 (2.5) $ Unknown, n (%)15 (37.5) Highest Level of Education (grade level) High School (9-12), n (%)33 (82.5) College (13 +), n (%)7 (17.5) Number of pregnancies M (SD)2.43 (+1.88) Range1 – 10 Table 2 Intercorrelations Among Study Variables 12345678910111213 1.Age -.09.31.58**.11.22.17.19-.17-.14-.30.10-.03 2.Income --.31.06.09-.18.03.10.21.12.28.13-.13 3.Education ----.21.07.13.11.32*.21.17.19.13.25 4.Number of Pregnancies ----.07.04.01.58**-.21-.18-.40*.25-.13 5.Depression -----.65**.25.51**.16.21-.04-.09 6.Stress ------.33*.39*.07.08.01.04-.17 7.Weight Concern -------.68**.40*.41**.11-.08-.24 8.Body Image --------.67**.68**.21.10-.01 9.EAT-26 Total ---------.97**.68**.25.13 10.Diet Subscale ----------.55**.05.12 11.Bulimia Factor Subscale ------------.12.02 12.Oral Control Subscale ------------.01 13.Change in BMI ------------- *. p <.05. **. p <.01. Weight Concerns and Body Image as Related to Compensatory Behavior among Rural Pregnant Smokers Kristen Carlosh 1, Suzanne Allen, M.A. 1, William T. Dalton III, Ph.D. 1, & Beth Bailey Ph.D. 2 Department of Psychology 1 & Department of Family Medicine 2 East Tennessee State University, Johnson City, TN Abstract Introduction Methods References DiscussionResults Weight concerns have been identified as a contributing and maintaining factor in smoking behavior. Additionally, smoking has been identified as a weight-related compensatory behavior; therefore, weight concerns among women who smoke may be predictive of other compensatory behaviors. Our objective was to examine the relationship between weight concerns/body image and compensatory behaviors among rural women who smoke during pregnancy. We hypothesized that greater weight concern/body image dissatisfaction at 1 st trimester would predict the following at 3 rd trimester 1) greater compensatory behaviors, and 2) less pregnancy weight gain. As part of the Tennessee Intervention for Pregnant Smokers (TIPS), forty pregnant smokers were recruited at a first trimester prenatal health care visit and assessed throughout pregnancy. Participants completed a battery of measures including the Weight Concern Scale (WCS), Body Image Concern Inventory (BICI), and Eating Attitudes Test-26 (EAT-26), which assesses compensatory behaviors. Weight was assessed using self-report and anthropometric measures gathered at doctors’ visits. WCS and BICI scores were significantly correlated with the EAT-26 scores ( p <.05). In a multiple regression analysis, demographic variables (i.e., age, income, education), WCS scores, and BICI scores were regressed on EAT-26 scores, yielding a significant effect ( R 2 =.59, F (6, 29) = 7.015, p <.01). However, standardized beta coefficients showed that BICI score ( β =.94, p <.01) was the only significant predictor. Thus, greater body image concerns surveyed in 1 st trimester predicted more engagement in eating disordered behaviors by 3 rd trimester among smoking pregnant women. Targeting body image early in pregnancy could counteract the development of potentially harmful weight control behaviors. This may be especially important in rural populations where smoking is highly prevalent and pregnancy outcomes are poor. Available upon request.  Similar to findings in non-pregnant populations (Stice, 2002), the current study found body image concerns to predict engagement in compensatory behaviors among rural pregnant women. Thus, pregnant women who are most concerned about their body are more likely to engage in unhealthy practices such as purging, laxative abuse, and excessive exercise. Although weight and body image concerns did not predict change in weight, we did find that pregnant women with less concern about weight gained more weight over the course of pregnancy than those with greater weight concern. This finding provides some indirect evidence that higher weight concerns may, in fact, influence weight control practice.  The fact that greater weight concern did not significantly predict greater compensatory practices was somewhat surprising. It may be that the transitions in the body shape have a greater impact on mothers than expected weight gain. Recent research suggests that eating disorders are common in individuals with body dysmorphic disorder (BDD) or those with distressing or impairing concerns with imagined or minor defects in physical appearance (Ruffolo et al., 2006). Further, BDD symptoms have been found to predict smoking in college age women (Stickney & Black, 2008). Together, these findings suggest that future research may aim at better understanding the effects of body change during pregnancy and clinician’s efforts to prepare mothers for bodily changes in addition to weight gain may be worthwhile.  This preliminary investigation has some limitations but also provides directions for future research. The small sample size limited statistical power thus emphasizing the need for future research before definitive conclusions can be drawn. The use of a pre-pregnancy self-report of weight and height was an additional limitation as women do not always accurately report their weight (Meyer et al., 2009). We intentionally restricted this study to smokers therefore future studies may examine differences between women who do not smoke. Future studies may also examine differences in more rural versus less rural populations for further generalization of findings. Consideration may also be given to methods for gaining retrospective accounts of weight and body concerns. Finally, future studies may aim to understand the potential differences between weight concern and body image as well as consider utilization of additional measures of compensatory behavior that more narrowly capture DSM-IV TR characteristics of compensatory behaviors.  To our knowledge, the current study was the first to examine whether weight/body image concerns predict higher engagement in compensatory behaviors among rural pregnant women who smoke. The findings hold strong implications for programs aimed at smoking cessation or promotion of optimal health in pregnant women. Specifically, screening for these concerns may allow for identification of individuals that may benefit from techniques aimed at facilitating success with cessation as well as promoting healthy weight control practices over the course of pregnancy.  Intercorrelations among study variables are shown in Table 2. Weight concern [r(34) =.40, p<.05] and body image concern [r(34)=.67, p<.01] were significantly associated with engagement in compensatory behaviors.  To examine Hypothesis One, a regression was conducted to examine the predictive value of weight/body image concern and background factors on compensatory behaviors. Together, weight and body image concerns, education, number of pregnancies, depression, and stress significantly predicted compensatory behaviors at 3 rd trimester, R 2 =.59, F (6, 29) = 7.015, p <.01. Each predictor was examined in terms of its relative influence on compensatory behaviors. Overall, body image had the greatest influence on compensatory behaviors ( β =.94, p <.01) followed by number of pregnancies ( β = -.31, p <.05). Paradoxically, weight concerns did not have a significant impact on compensatory behaviors in this model ( β = -.11, p =.515).  To examine Hypothesis Two, a second regression was conducted to assess the predictive value of the same variables on change in BMI from pre-pregnancy to 3 rd trimester. This regression model was not significant R 2 =.183, F (6, 33) = 1.24, p =.31. Exploratory analyses were conducted to examine weight concerns between those with greater BMI change (i.e., >3) to those with less BMI change (i.e., <3). Independent-sample t-test revealed significantly less weight concerns among those with greater BMI change (M = 4.58, SD = 2.76) compared to those with less BMI change ( M = 6.29, SD = 1.94), t (38) = -2.06, p <.05. I.Participants  40 pregnant women enrolled in the TIPS study  Smoking at first trimester  Primarily Caucasian  20% Married  See Table 1 for other demographic information II.Measures  Weight Concern Scale (WCS; Question 1: How important is losing weight or maintaining your current weight compared with other personal health concerns?)  Body Image Concern Inventory (BICI)  Compensatory Behaviors; Eating Attitudes Test-26 (EAT-26 Total; Diet, Bulimia Factor, Oral Control Subscales)  Change in BMI (kg/m 2 ) [Differences in BMI between pre-pregnancy self- report BMI and third trimester actual BMI]  Demographic Questionnaire (Age, Income, Education, & # of Pregnancies,)  Center for Epidemiologic Studies Short Depression Scale (CESD-10)  Stress; Prenatal Psychosocial Profile (PPP) III. Procedure  Participants were recruited through OB/GYN doctor  During first trimester the following assessments were completed: WCS and BICI (reflect on pre-pregnancy), self -report pre-pregnancy weight and height, demographic questionnaire, CESD-10, PPP  During the third trimester, EAT-26 completed and actual weight and height obtained  Health-risk behaviors are among a number of factors contributing to health disparities within rural populations (Bennett et al., 2008). Health education campaigns have stressed positive health behaviors including attention to healthy weight. Unfortunately, increased focus on weight may lead to excessive concern about weight, resulting in engagement in compensatory behaviors.  Pregnancy is a time of significant body change and how pre-existing weight and body concerns might be associated with compensatory behaviors during pregnancy is unclear. This is important as eating disorder behaviors are not only harmful to the mother but may be especially detrimental to the child (Conti et al., 1998).  Weight/body concerns also contribute to the development/maintenance of smoking behavior (Borrelli & Mermelstein, 1998). Research has shown that women who quit smoking prior to or during pregnancy gain more pregnancy weight than their smoking counterparts (Groff et al., 1997). Consequently, smoking during pregnancy is associated with significant complications for both the mother (Roth & Taylor, 2001) and child (Hofhuis et al., 2003).  A salient health-risk behavior in rural Appalachia is use of tobacco, especially among pregnant women with smoking rates as high as 50% in some counties (Bailey, 2006). Little is known about weight/body concerns related to compensatory behaviors in rural pregnant women who smoke. Greater understanding of these factors may aid in the development of more effective prevention/intervention programs for improving maternal and child health.  The purpose of the current study was to examine the relation between pre-pregnancy weight/body concerns and compensatory behaviors/weight gain among women who smoke during pregnancy. Greater pre-pregnancy weight/body concern assessed retrospectively at entry to prenatal care were predicted to yield 1) greater engagement in compensatory behaviors at 3 rd trimester, and 2) less weight gain from pre- pregnancy to 3 rd trimester.