The relation between lifestyle factors and biochemical markers of bone turnover among early postmenopausal women.

We examined the associations of two biochemical markers of bone turnover with lifestyle factors in 340 postmenopausal women in Hawaii, ages 45-59 years, from the Early Postmenopausal Intervention Cohort. Physical activity, calcium supplement use, smoking and alcohol use in the prior 2 weeks were measured and examined as independent variables in multiple regression analyses with bone turnover markers as dependent variables, adjusted for weight, height, whole body bone mass, serum estradiol, years since menopause, and ethnicity. Calcium supplement and alcohol use were significantly associated with reduced levels of urinary type I collagen cross-linked N-telopeptides (NTX). The mean NTX level was 12% lower among women using > or = 250 mg of calcium supplements per day as compared with other women, and 20% lower among alcohol users compared with nonusers. Both calcium supplement use and alcohol intake were associated with lower mean serum osteocalcin (a marker of bone formation) and NTX z-scores. By contrast, smoking was associated with lower osteocalcin levels, without any effect on NTX. The osteocalcin level was 12% lower among smokers compared with nonsmokers. In addition, the z-score difference between NTX and osteocalcin was significantly associated with smoking, with a shift towards more NTX than osteocalcin. Physical activity was not significantly associated with either of the markers. These findings suggest that biochemical markers may help to identify lifestyle factors that affect bone, and provide estimates of the relative magnitude of these effects on bone formation and resorption, independent of each other.

Relation between body composition and biochemical markers of bone turnover among early postmenopausal women.

We studied the associations between body composition and biochemical markers of bone formation and resorption among 1600 postmenopausal women, ages 45-59. Multiple regression analyses were performed to examine the independent associations of fat mass, muscle strength (quadriceps strength), height, and whole body bone mineral content (BMC) with biochemical markers of bone formation (serum osteocalcin) and resorption (urinary type I collagen crosslinked N-telopeptides [NTX]). Per interquartile range (IQR) (the difference between 75th and 25th percentiles) increase in fat mass and whole body BMC, the mean levels of osteocalcin decreased by 3% and 13%, respectively; NTX decreased by 5 and 21%. Fat mass and whole-body BMC were also significantly associated with decreases in the average of osteocalcin and NTX Z-scores. By contrast, the mean levels of serum osteocalcin increased by 2 and 11%, respectively, per IQR increase in muscle strength and height; NTX increased by 4 (not significant) and 14%, respectively. Both muscle strength and height were significantly associated with increases in the average Z-scores. These exploratory analyses suggest that fat mass and whole-body BMC were associated with decreased bone turnover, while muscle strength and height were associated with increased bone turnover.

Breastfeeding duration in a multiethnic population in Hawaii.

BACKGROUND: The increasing ethnic diversity in the United States necessitates a study of variations in infant feeding patterns among ethnic groups. This study was conducted as part of Hawaii's surveillance system to identify infant feeding patterns in Hawaii; specifically, to identify factors influencing duration of breastfeeding among ethnically diverse mothers. METHODS: All women who delivered an infant in Hawaii between January 1 and March 31, 1989, were mailed surveys 14 to 19 months after delivery. Fifty-one percent (n = 2011) of women responded, of whom 1574 (78%) did some breastfeeding and are included in the analysis of prediction of weaning (cessation of breastfeeding). Cox regression (survival) analysis was used to predict weaning. RESULTS: The median duration of breastfeeding was 150 days; 45 percent of infants were still breastfeeding at age 6 months and 16 percent at age 1 year. Factors associated with early weaning were Japanese ethnicity; mother born in a country other than the United States, Japan, or the Philippines; first language other than English, or two languages at home; employed full-time outside the home; introduced formula or fruit before age 6 months; received formula from the WIC program; and stopped breastfeeding for convenience, breast problems, problems getting breastfeeding started, insufficient milk, baby refusing the breast, and a sick baby. Factors associated with late weaning were older maternal age; college education; living on a rural island; previous breastfeeding experience; helpful breastfeeding advice from family or friends; receiving WIC for breastfeeding mothers; introducing the cup before age 6 months; and not giving fruit to the baby. CONCLUSION: In Hawaii, programs that address how and when to introduce foods, use of formula, and management of outside employment and breastfeeding should be made available to those groups of women at risk for early weaning to lengthen their duration of breastfeeding.

A multicenter study of the influence of fat and lean mass on bone mineral content: evidence for differences in their relative influence at major fracture sites. Early Postmenopausal Intervention Cohort (EPIC) Study Group.

We examined the relative influence of fat and lean mass on bone mineral content (BMC) among 1600 early postmenopausal women aged 45-59 y from four geographical locations (Nottingham, United Kingdom; Portland, OR; Honolulu; and Copenhagen). Bone sites investigated included the major fracture sites: hip, spine, and radius. Body weight had strong associations at all skeletal sites examined [BMC differences of 4-6% per interquartile range (IQR) of weight]. Associations with the fat and lean components of weight were more variable. The BMC differences per IQR of lean mass were 5-7% at the hip sites, 3% at the spine, and 2% at the radial sites. The greater differences for lean mass at the hip may reflect the high physical mobility and muscular activity of this site. The BMC differences per IQR of fat mass were 4-6% at the hip sites, 4% at the spine, and 5% at the ultradistal radius. These results suggest that low fat mass or low lean mass, particularly at the extremes, may adversely affect the major fracture sites. The bone sites with the greatest differences for fat mass were the most highly trabecular sites. With only a few exceptions, the associations of BMC with fat mass and lean mass were similar in direction and comparable in magnitude across the four geographic locations. We conclude that both fat and lean mass have independent influences on bone mass, but that their relative influence may vary by bone site depending on the trabecular content, physical mobility, and muscularity of the site.