Methods of Estimation in Multiple Linear Regression: Application to Clinical Data

Nesse artigo, tem-se o interesse em avaliar diferentes estratégias de estimação de parâmetros para um modelo de regressão linear múltipla. Para a estimação dos parâmetros do modelo foram utilizados dados de um ensaio clínico em que o interesse foi verificar se o ensaio mecânico da propriedade de força máxima (EM-FM) está associada com a massa femoral, com o diâmetro femoral e com o grupo experimental de ratas ovariectomizadas da raça Rattus norvegicus albinus, variedade Wistar. Para a estimação dos parâmetros do modelo serão comparadas três metodologias: a metodologia clássica, baseada no método dos mínimos quadrados; a metodologia Bayesiana, baseada no teorema de Bayes; e o método Bootstrap, baseado em processos de reamostragem.

Influence of body composition on bone mass in postmenopausal osteoporotic women

We aimed at evaluating the relationship of lean and fat mass to bone mass in osteoporotic postmenopausal women. We invited 65 women who were being treated at the Sao Paulo Hospital osteoporosis outpatients` clinic to participate. Body composition and bone mineral density (BMD) measurements were performed using Dual-energy X-ray absorptiometry methodology (DXA). The mean age and weight were 69.7 +/- 6.4 years and 56.3 +/- 7.6 kg, respectively. Accordingly to the body mass index (BMI), 52.8% were of normal weight and 47.1% of the patients were overweight. Overweight women had significantly higher bone mass. Similarly, skeletal muscle index (SMI) showed a positive effect on BMD measurements and women with sarcopenia had significantly lower BMD measurements in total femur and femoral neck. In multiple regression analysis only lean mass and age, after adjustments to fat mass and BMI, were able to predict total body bone mineral content (BMC) (R(2) = 28%). Also lean mass adjusted to age and BMI were able to predict femoral neck BMD (R(2) = 14%). On the other hand, none of the components of the body composition (lean mass or fat mass) contributed significantly to explaining total femur BMD and neither body composition measurements were associated with spine BMD. These findings suggest that lean mass has a relevant role in BMC and BMD measurements. In addition, lower BMI and lean mass loss (sarcopenia) is associated to lower BMC and BMD of femoral neck and total femur and possible higher risk of osteoporotic fracture. (C) 2010 Elsevier Ireland Ltd. All rights reserved.