Original Article

Racial/Ethnic and Socioeconomic Differences in Bone Loss Among Men† Andre B. Araujo, PhD1,2, May Yang, MS1, Elizabeth A. Suarez, MPH1, Nicholas Dagincourt, MS1, Jonathan R. Abraham, MPH1, Gretchen Chiu, MS3, Michael F. Holick, PhD, MD4, Mary L. Bouxsein, PhD5, Joseph M. Zmuda, PhD6 1

Department of Epidemiology, New England Research Institutes, Inc., Watertown, MA Eli Lilly and Company, Indianapolis, IN 3 Trinity Partners, LLC, Waltham, MA 4 Department of Medicine, Section of Endocrinology, Nutrition and Diabetes at Boston University Medical Center, Boston, MA 5 Beth Israel Deaconess Medical Center and Massachusetts General Hospital, Boston, MA 6 University of Pittsburgh, Graduate School of Public Health, Pittsburgh, PA 2

Running Head: Race, SES, and bone loss in men Corresponding Author: Andre B. Araujo, PhD, [email protected] Requests for Reprints: [email protected]



This article has been accepted for publication and undergone full peer review but has not been through the copyediting, typesetting, pagination and proofreading process, which may lead to differences between this version and the Version of Record. Please cite this article as doi: [10.1002/jbmr.2305]

Additional Supporting Information may be found in the online version of this article.

Initial Date Submitted December 23, 2013; Date Revision Submitted May 22, 2014; Date Final Disposition Set May 29, 2014

Journal of Bone and Mineral Research © 2014 American Society for Bone and Mineral Research DOI 10.1002/jbmr.2305 1   

ABSTRACT As men age, they lose bone and are susceptible to fracture. Despite having lower fracture rates than women, men have worse fractures than women do. Racial/ethnic and socioeconomic status (SES) disparities in fracture rates exist, yet data on rates of bone loss by race/ethnicity and SES among men are limited. We examined annualized percentage change in bone mineral density (%∆BMD) at the hip (N=681), spine (N=663) and forearm (N=636) during 7 years of follow-up among men aged 30-79y at baseline. Multivariable models tested whether race/ethnicity, income, genetic ancestry predicted annualized %∆BMD after controlling for an extensive set of covariates. Annualized %∆BMD ranged from -0.65(0.04)% (femoral neck) to +0.26(0.03)% (1/3 distal radius), and changes were consistent across age groups with the exception of the ultradistal radius, where annualized declines increased with age. Neither self-identified race/ethnicity nor genetic ancestry were associated with annualized %∆BMD. In contrast, income was strongly associated (dose-response) with annualized %∆BMD at total hip (independent of confounders, self-identified race/ethnicity, and genetic ancestry). Fully-adjusted least-square mean change in annualized %∆BMD at the total hip were -0.24(0.12)% and -0.16(0.06)% steeper among men with low and moderate incomes, respectively, than among men with higher incomes (overall p=.0293). Results show a linear decline in bone that begins relatively early in life among men, that rates of bone loss do not vary with race/ethnicity (self-identified or ‘objectively’ measured), and that income plays an important role in relation to bone loss at the hip. These data suggest that fracture risk in men may be driven in part by income-related differences in bone loss, but also, that the known higher fracture risk among white men is not due to by racial/ethnic differences in bone loss, but rather by early life exposures that lead to attainment of higher peak bone mass among minorities. Keywords: Aging, DXA, Genetic Studies, Osteoporosis, General Population Studies

2   

INTRODUCTION As men age, they lose bone and are susceptible to fracture. While the majority of osteoporotic fractures occur in women, fracture outcomes are worse in men compared with women, resulting in higher rates of morbidity and mortality.(1) White men have the highest fracture rates, followed by Hispanic, Black, and Asian men.(1,2) Crosssectional data from large epidemiologic studies of bone mineral density (BMD) among men are congruent.(3-6) Three community-based studies have compared longitudinal rates of bone loss among black and white men. Data from these studies are in conflict with studies showing that bone loss is higher(7), lower(8), or no different(9) between black and white men. Rates of bone loss in Asian men tend to be lower,(9) and to our knowledge, estimates of rates of bone loss in Hispanic men are non-existent. Most studies on racial/ethnic differences in bone have relied on self-identified race/ethnicity. This multidimensional construct embeds numerous facets of an individual’s biological, cultural, economic, and social context, yet studies to date have failed to unpack these components. Measures of genetic ancestry, on the other hand, offer the possibility of characterizing the influence of admixture on health. Studies have shown negative correlations between degree of European ancestry and BMD, (10-14) but none have examined bone loss among men. In light of potential future increases in fractures and associated costs in US male and minority populations,(15) we attempt to separate out the biological influence of race/ethnicity (genetic ancestry) from the cultural, economic, and social components of self-identified race/ethnicity on skeletal health,(16) while also seeking to understand how socioeconomic status (SES) fits with racial/ethnic differences in longitudinal rates of bone loss in men. This is especially of interest since the literature on SES and bone health outcomes has a weak evidence base (17,18) with conflicting results. Studies have shown higher SES as having positive, (19-23) negative, (24) no, (25-30) or more complex (e.g., dependent on age (31,32)) associations with poor bone health outcomes, such as reduced BMD and fractures.

3   

METHODS Study Sample The Boston Area Community Health/Bone (BACH/Bone) Survey(3) is a population-based longitudinal survey of musculoskeletal health in men. Participants were originally recruited as part of the BACH Survey, a crosssectional observational study of 5,506 randomly selected (via two-stage stratified cluster design) Black, Hispanic, and White male and female Boston, MA residents ages 30 to 79 years, which examined the prevalence of several urological and gynecological conditions. Further details on the BACH Survey have been published. (33) Male participants from the BACH Survey were invited to participant in the BACH/Bone Survey. Data were collected at baseline (Nov 2002-Jul 2005) and follow-up (Jun 2010-Oct 2012) examinations. Protocols were approved by IRBs at New England Research Institutes, Inc. (NERI) and Boston University School of Medicine (BUSM). All participants provided written informed consent and received $75 and $100 remunerations for their participation in the baseline and follow-up examinations, respectively. A total of 1,219 men (65% of those eligible) completed the baseline examination, and 692 (72% of those eligible) completed the follow-up examination. The study flow chart is displayed in Error! Reference source not found.. Data Collection Examinations, including blood sampling, in-person interview, questionnaire, and anthropometry, and densitometry were conducted at the BUSM General Clinical Research Unit. Measurements Self-Identified Race/Ethnicity Self-identified race/ethnicity was determined according to modifications to the federal standard. It involved a two-step self-identification process, first asking whether the subject considered himself to be of Hispanic origin and then separately about racial group. ‘Objectively-Defined’ Race/Ethnicity: Genetic Ancestry Autosomal ancestry informative markers (AIMs) are set of single nucleotide polymorphisms (SNPs) used to assess an individual’s genetic ancestry. A panel of 62 SNPs with differences in allele frequency (delta, ) between 4   

Native American, African, and European ancestral populations (14,34,35) were genotyped at Broad Institute (Cambridge, MA) using the Sequenom MassArray iPLEX assay (San Diego, CA). Ancestry proportions were estimated for individual participants using ADMIXTURE Software (version 1.12), and percentages of West African and Native American ancestry were analyzed. Socioeconomic Status Household income per capita was determined from reported household income and number of people in the home, and was categorized as $6,000 or less, $6,001 to $30,000, or more than $30,000 per year. Information on education level and employment status were also collected, but preliminary analyses suggested that income was a more important determinant of BMD. Bone Mineral Density (BMD) Hip, anteroposterior lumbar spine, and forearm areal BMD were measured by dual energy x-ray absorptiometry (DXA) using a Hologic QDR 4500 W densitometer (Hologic, Inc., Waltham, MA) at baseline and follow-up. The left hip and forearm were scanned for all participants except those with artifacts or previous fracture, in which case the opposite hip was scanned or the measured forearm was deleted. Regions of interest (ROI) were identified using the ‘compare’ function of the densitometer to map the follow-up onto baseline scan. Scans were visually inspected for proper positioning, analysis of correct ROI, and motion or other artifacts. The DXA system was calibrated daily with a vertebral phantom. Standard site-specific exclusions were made (e.g., participants >300 pounds excluded from whole body/hip/spine scans; participants with bilateral hip replacement excluded from hip scans; participants with metal implants/fractures excluded for the affected site; participants excluded if they received tests with radioactive materials in past 10 days). Change in BMD was calculated for participants who had baseline and follow-up scans appropriate for comparison, including 681 participants at the hip, 663 participants at the spine, and 636 participants at the forearm. Covariates A large set of covariates measured at baseline were considered as potential confounders. Preliminary analyses (see below) reduced the set to a final group of thirteen covariates, including age, smoking status (current/former/never), self-reported history of diabetes, cardiovascular disease, or hypertension (yes/no), self-rated 5   

health (excellent/very good/good/fair-poor), DXA total body lean mass (minus head and BMC), height, alcohol use (0 drinks per day/

ethnic and socioeconomic differences in bone loss among men.

As men age, they lose bone and are susceptible to fracture. Despite having lower fracture rates than women, men have worse fractures than women do. Ra...
457KB Sizes 2 Downloads 4 Views