Do Maternal Psychosocial Factors Predict Adolescent Weight? Sandra Jody MarksDepartment of Exercise Sciences, BYUMaster of SciencePurpose: This study investigated the possible relationship between maternal psychosocial factors, mainly maternal stress and maternal depression, and adolescent weight status. Also, this study examined the predictive effect of these maternal psychosocial factors on adolescent weight loss during a health education intervention as well as the months following the health education. Methods: Study design was a longitudinal pretest posttest with a health education intervention. We assessed 40 adolescents and their mothers on four occasions over a 1-year period. At each occasion, the Stress Index for Parents of Adolescents (SIPA) was used to measure maternal stress and the Beck Depression Inventory (BDI-II) was used to measure maternal depression. Also, at the four occasions, adolescent anthropometric data were obtained by research assistants using a digital scale for weight and a portable stadiometer for measuring height. Body Mass Index scores (BMI = [weight (kg)]/[height (m)]2) were calculated and converted into a percentile score (zBMI), adjusting for age and gender, using the standard Center for Disease Control and Prevention calculator. At the onset of the study, the adolescent participants and their mothers all received 12 weeks of health education, which included group behavioral therapy, family-based intervention, motivational interviewing and electronic intervention. Results: Hierarchical regression analysis revealed that no significant relationships existed between maternal stress and adolescent zBMI or between maternal depression and adolescent zBMI at baseline (Time 1). Nor did the study find that maternal stress and/or depression scores at Time 1 significantly predicted a greater amount of adolescent weight loss. Lower stress and/or depression also did not significantly predict adolescent weight maintenance after the 12-week intervention (Time 2). However, results did indicate that the adolescent component of the maternal stress domain (AD) from Time 1 to Time 2 was a significant predictor of adolescent zBMI from Time 1 to Time 2, (R2 = 0.238, F (1,21) = 6.571, p = 0.018). This means that 23.8% of the variability in overall zBMI change from Time 1 to Time 2 is being accounted for by change in the maternal AD stress domain from Time 1 to Time 2. Conclusion: Adolescent zBMI decreased concurrently with maternal stress during the health education intervention stage. Although the correlational nature of this study prevents causal claims, this result suggests that decreasing maternal stress may strengthen the ability of obese adolescents to effectively lose weight. This study encourages further research to examine the effects that maternal psychosocial factors may have on adolescent weight status, weight loss, and weight maintenance.

Submitted by Fabio Sobreira Campos da Costa (fabio.sobreira@ufpe.br) on 2016-08-04T15:04:06Z No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DISSERTAÇÃO VERSÃO ONLINE CD.pdf: 3278151 bytes, checksum: db4fb019af69b7f6dea1042d768d89d1 (MD5) / Made available in DSpace on 2016-08-04T15:04:06Z (GMT). No. of bitstreams: 2 license_rdf: 1232 bytes, checksum: 66e71c371cc565284e70f40736c94386 (MD5) DISSERTAÇÃO VERSÃO ONLINE CD.pdf: 3278151 bytes, checksum: db4fb019af69b7f6dea1042d768d89d1 (MD5) Previous issue date: 2016-01-18 / O ganho de peso excessivo durante as duas primeiras décadas de vida, parece resultar em um efeito adverso no perfil lipídico e a trajetória ascendente do índice de massa corporal/idade (IMC/I) durante a fase de crescimento e desenvolvimento pode estar associada ao surgimento de fatores de risco cardiovasculares mesmo que o aumento em escore z seja pequeno. A dissertação teve como objetivo avaliar a influência do ganho de peso sobre o perfil lipídico; verificar o perfil lipídico, estado nutricional, estilo de vida e situação socioeconômica dos adolescentes no baseline e no follow-up; comparar a variação do perfil lipídico entre os sexos; correlacionar a variação do escore z do IMC/I com a variação do perfil lipídico e investigar a influência do ganho de peso excessivo no perfil lipídico. Foi realizado um estudo de coorte com 135 adolescentes de ambos os sexos, com idade entre 10 a 14 anos no baseline e 15 a 19 anos no follow-up matriculados na rede pública de ensino do Recife-PE. Foram avaliadas variáveis antropométricas, bioquímicas, consumo alimentar, estilo de vida e aspectos socioeconômico e demográficos. O ganho de peso foi avaliado pela diferença entre o escore z do IMC/I no follow-up e no baseline e a variação do perfil lipídico foi determinada pela diferença entre as frações lipídicas no follow-up e no baseline. A análise estatística foi realizada pelo programa statistical package for the social sciences for windows student (SPSS) version 13.0. Os resultados mostraram que o aumento do escore z do IMC/I se correlacionou positivamente com os níveis de triglicerídeo (TG, p=0,006) e com a razão triglicerídeo/lipoproteína de alta densidade (TG/HDL-c, p=0,003) no sexo masculino e com colesterol total (CT, p=0,024), lipoproteína de baixa densidade (LDL-c, p=0,003) e colesterol não-HDL (p=0,002) no sexo feminino, além de uma correlação inversa com o HDL-c (p=0,037). Para cada unidade de aumento no escore z houve um aumento de 14,7 mg/dL no TG e 0,4 na razão TG/HDL-c no sexo masculino e um aumento de 9,4 mg/dL no CT, 11,6 mg/dL no LDL-c, 11,8 mg/dL no colesterol não-HDL e uma redução de 2,3 mg/dL no HDL-c no sexo feminino. No sexo masculino, o terceiro tercil de ganho de peso apresentou maiores elevações nos níveis de TG (p=0,003) e na razão TG/HDL-c (p=0,002) quando comparado ao primeiro tercil. Já no sexo feminino, o terceiro tercil apresentou maiores elevações nos níveis de CT (p=0,004), LDL-c (p=0,005) e no colesterol não-HDL (p=0,004) quando comparado ao primeiro tercil. Quanto ao ganho de peso excessivo, encontramos resultados similares aos observados para o sexo masculino e para o sexo feminino surge um maior aumento da razão TG/HDL-c (p=0,031), do colesterol não-HDL (p=0,035) e menor aumento do HDL-c (p=0,006) no grupo com ganho de peso excessivo. É preciso mais estudos para a elaboração de uma classificação para o ganho de peso ideal durante a adolescência, visando a promoção do crescimento adequado e restrição do ganho de peso excessivo diante dos indícios de que os adolescentes com maior aumento do escore z do IMC/I durante o crescimento pode ter um perfil lipídico mais desfavorável. / Excessive weight gain during the first two decades of life seems to have an adverse effect on lipid profile, and the ascending course of body mass index-for-age (BMI/A) during growth and development may be associated with the onset of cardiovascular risk factors, even if the z-score increase is small. This dissertation aimed to assess the influence of weight gain on lipid profile; verify lipid profile, nutritional status, lifestyle, and socioeconomic characteristics of adolescents at baseline and follow-up; compare lipid profile variation between genders; correlate BMI/A z-score variation with lipid profile variation; and investigate the influence of excessive weight gain on lipid profile. A cohort study was done with 135 male and female adolescents aged 10 to 14 years at baseline and 15 to 19 years at follow-up. They were enrolled in public schools of Recife-PE. Anthropometric, biochemical, food intake, lifestyle, socioeconomic, and demographic variables were assessed. Weight gain was assessed by the difference between the BMI/A z-scores at follow-up and baseline, and lipid profile variation was determined by the difference between the lipid fractions at follow-up and baseline. The statistical analyses were performed by the software Statistical Package for the Social Sciences (SPSS) version 13.0. An increase in BMI/A z-score correlated positively with triglycerides (TG, p=0.006) and the ratio triglyceride-to-high-density lipoprotein cholesterol (TG/HDL-c, p=0.003) in males, and with total cholesterol (TC, p=0.024), low-density lipoprotein cholesterol (LDL-c, p=0.003), and non-HDL cholesterol (p=0.002) in females, and correlated inversely with HDL-c (p=0.037). For each unit increase in z-score, TG increased by 14.7 mg/dL and the TG/HDL-c ratio increased by 0.4 in males, and TC increased by 9.4 mg/dL, LDL-c increased by 11.6 mg/dL, non-HDL cholesterol increased by 11.8 mg/dL, and HDL-c decreased by 2.3 mg/dL in females. In males the third weight gain tertile presented higher increases in TG (p=0.003) and in the TG/HDL-c ratio (p=0.002) than the first tertile. In females the third tertile presented higher increases in TC (p=0.004), LDL-c (p=0.005), and non-HDL cholesterol (p=0.004) than the first tertile. As for excessive weight gain, the results are similar as those seen for males; for females, there is a higher increase in the TG/HDL-c ratio (p=0.031) and non-HDL cholesterol (p=0.035), and a smaller increase in HDL-c (p=0.006). More studies are needed for the development of a classification for ideal weight gain during adolescence, enough to promote adequate growth and prevent excessive weight gain, given that adolescents with higher increases in BMI/A z-score during growth may have worse lipid profiles.