Cohort Profile: Geelong Osteoporosis Study

The Geelong Osteoporosis Study (GOS) began as a population-based study designed to investigate the epidemiology of osteoporosis in Australia. Osteoporosis is a systemic skeletal disease characterized by low bone mineral density (BMD) and micro-architectural deterioration of bone tissue, with a consequent increase in susceptibility to fracture. 1 Fragility fractures at the spine, hip, wrist and other sites are a major public health problem in both sexes because these fractures are responsible for considerable morbidity and mortality. 2, 3 Fractures cost the Australian community an estimated 7 billion annually4 and this is expected to increase in absolute terms because of the ageing population. 5 The cohort was recruited to address the need for definitive data on the prevalence of osteoporosis, to describe age-related changes in BMD and to characterize the risks for osteoporosis and fracture. Initially the GOS comprised only women; men were recruited later. In addition to the cohort study, the GOS has developed a comprehensive fracture register for the study region and has also conducted case–control studies to investigate risk factors for fracture. Only the cohort study will be described here. The major objectives of this study are to describe the epidemiology of osteoporosis and to generate evidence for defining osteoporosis risk and fracture risk in Australian men and women. We aimed to: compare BMD and rates of bone loss among various groups of the population; compare and combine BMD measurement at different sites to predict fracture risk; and generate environmental and genetic evidence to define the risks for osteoporosis and fracture. The study region is described by the Australian Bureau of Statistics (ABS) as the Barwon Statistical Division (BSD), situated in South-Eastern Australia. The BSD comprises the Australian Electoral Commission (AEC) Divisions of Corio, Corangamite (part) and Lalor (part). Voting is compulsory in Australia for adults aged ≥18 years, so the electoral roll provides a comprehensive listing of residents registered with the AEC. A listing on the Commonwealth electoral roll as a resident of the BSD fulfilled inclusion criterion for the study. Nationwide census data reported the population of the BSD as 221 687 (108 606 male, 113 072 female) in 1996 (ABS catalogue 2020. 0), 241 446 (118 207 male, 123 239 female) in 2001 (ABS 2001. 0) and 259 013 (126 889 male, 132 124 female) in 2006 (ABS 2001. 0). Persons resident for <6 months and those unable to provide written informed consent were excluded from the study. Population characteristics of the BSD were comparable with national levels for each census; differences did not exceed 1. 1% for age, 9. 5% for country of birth, 7. 5% for school leavers’ age, 2. 6% for marital status and 2. 1% for weekly income (Table 1). Maximum differences for population characteristics between the BSD and national levels according to census data collected in 1996, 2001 and 2006 aAge profiles expressed in 5-year age groups from birth to 79 years and ≥80 years. bCountry of birth categorized as Australia vs Canada/Ireland/NewZealand/South Africa/UK/USA vs others. cSchool leavers’ age grouped as ≤14 years vs 15 years vs 16 years vs 17 years vs 18 years vs ≥19 years vs still at school vs never attended school. dSchool leavers’ age grouped as ≤8 years vs 9 years vs 10 years vs 11 years vs 12 years vs still at school vs did not go to school). eMarital status expressed as married vs separated vs divorced vs widowed vs never married. fWeekly income grouped as negative/nil vs 1–39 vs 40–79 vs 80–119 vs 200–299 vs 300–399 vs 400–499 vs 500–599 vs 600–699 vs 700–799 vs 800–999 vs 1000–1499 vs ≥1500. gWeekly income grouped as negative/nil vs 1–149 vs 150–249 vs 250–399 vs 400–599 vs 600–799 vs 800–999 vs 1000–1299 vs 1300–1599 vs 1600–1999 vs ≥2000). Maximum differences for population characteristics between the BSD and national levels according to census data collected in 1996, 2001 and 2006 aAge profiles expressed in 5-year age groups from birth to 79 years and ≥80 years. bCountry of birth categorized as Australia vs Canada/Ireland/NewZealand/South Africa/UK/USA vs others. cSchool leavers’ age grouped as ≤14 years vs 15 years vs 16 years vs 17 years vs 18 years vs ≥19 years vs still at school vs never attended school. dSchool leavers’ age grouped as ≤8 years vs 9 years vs 10 years vs 11 years vs 12 years vs still at school vs did not go to school). eMarital status expressed as married vs separated vs divorced vs widowed vs never married. fWeekly income grouped as negative/nil vs 1–39 vs 40–79 vs 80–119 vs 200–299 vs 300–399 vs 400–499 vs 500–599 vs 600–699 vs 700–799 vs 800–999 vs 1000–1499 vs ≥1500. gWeekly income grouped as negative/nil vs 1–149 vs 150–249 vs 250–399 vs 400–599 vs 600–799 vs 800–999 vs 1000–1299 vs 1300–1599 vs 1600–1999 vs ≥2000). An age-stratified sampling method was utilized, involving 12 strata for each sex. Individuals selected at random from the electoral roll were mailed a letter of invitation, with a request to return a response slip or telephone the research centre at Barwon Health (The Geelong Hospital). Follow-up letters were dispatched to non-responders. At least 100 women and 100 men were recruited in each 5-year age group from 20 to 69 years and 200 of each sex for both the age groups of 70–79 years and ≥80 years. Reasons for non-participation were documented. No indigenous Australians participated. All participants provided written, informed consent at each assessment. During the years 1993–97, 2390 women were invited to participate, of whom 432 lapsed (Figure 1) and 444 declined to participate, citing personal reasons (53. 2%), old age (18%), illness (12. 6%), time constraints (10. 4%), distance (2. 0%), language barrier (1. 6%), failure to keep appointments (1. 4%) and unknown reasons (0. 9%). Thus, there were 1494 female participants, representing a participation of 77%. 6 Socio-economic status (SES) for participants and non-participants were ascertained using Socio-Economic Index for Areas scores based on census data from the ABS (1996). These data, used to derive an Index of Relative Socio-Economic Disadvantage (IRSD) and categorized into groups according to quintiles of IRSD for the study region, revealed that there were no differences in SES between participants and non-participants (χ2, P = 0. 5). Flow diagram showing number of female participants at each phase of the study and reasons for loss to follow-up A new sample of 246 women listed as aged 20–29 years on the 2005 electoral roll was recruited in 2006–08. As this sample has not yet been recalled for follow-up, their details have not been presented. During the years 2001–06, 3273 men were invited to participate, of whom 167 had died, 311 had left the region, 482 were unable to be contacted and 17 were not able to give informed consent. Of the 2296 remaining, 756 declined to participate due to personal reasons (44%), old age (11%), illness (8%), time constraints (26%), distance (2%), language barrier (2%), failure to keep appointments (2%) and unknown reasons (5%). Thus, there were 1540 male participants representing a participation of 67%. 7 Comparison of participants and non-participants by quintiles of SES (IRSD, ABS 2001) revealed that there were more than expected non-participants with low SES (χ2, P < 0. 001). The 1494 women recruited at the baseline years 1993–97 were followed by re-assessment phases commencing 1995, 1998, 2000, 2001, 2002 and 2004, referred to as the 2-year, 4-year, 6-year, 7-year, 8-year and 10-year follow-up assessments. Clinical measures were obtained at all phases except the 7-year follow-up. Participation and reasons for loss to follow-up are detailed in Figure 1. Of the eligible women enrolled at baseline, 85% attended the 2-year follow-up. Subsequent phases retained 70% at the 4-year, 86% at the 6-year and 73% at the 7-year follow-up. A subgroup of 325 women was assessed at the 8-year follow-up; 83% of eligible women returned for assessment at the 10-year follow-up. The 1540 men recruited at the baseline years 2001–06 were followed by re-assessment phases commencing 2006 and 2007, referred to as the 5-year and 6-year follow-up assessments; baseline and 5-year follow-up involved clinical assessments. Of the 1540 men enrolled at baseline, 141 had died before the 5-year follow-up, 41 had left the region, 16 were unable to provide informed consent, 139 were not able to be contacted and the remaining 225 declined; the 978 participants represented 81% of eligible men. At the time of writing, the 6-year follow-up was still in progress, so participation details in this phase have not been finalized. A listing of clinical measures, biochemical and questionnaire data collected at each phase are presented in Table 2. Measurements for each follow-up phase, indicated by year of commencement for women and men enrolled in the GOS cohorts Fasting blood samples taken, DNA extracted, fasting glucose, albumin, calcium, phosphate assayed, serum/plasma aliquots stored at −80°C. Subsequent analyses for 25-hydroxyvitamin D, parathyroid hormone, β-isomerized C-terminal telopeptide, bone-specific alkaline phosphatase and procollagen type 1 N-terminal peptide, leptin, cholesterol, vitamin B12 and high sensitivity C-reactive protein 2-h timed morning urine collection; urinary calcium, phosphate, creatinine assayed, aliquots stored at −20°C 24-h urine collection (age <30 years) Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs (coded according to the structured drug codes developed by the Society of Hospital Pharmacists, Victorian Branch, 1986) and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth history, exposure (current and past of type and to Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Blood pressure Bone densitometry (DXA) —multi-site hip and of Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and of history, of Anthropometric measures: weight, Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, exposure of of and of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at analyses for β-isomerized C-terminal telopeptide, procollagen type 1 N-terminal peptide, and vitamin Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth exposure (current and past of type and to of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at not collected at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, of of of of and and of Fasting blood samples taken, DNA extracted, fasting glucose, albumin, calcium, phosphate assayed, serum/plasma aliquots stored at −80°C. Subsequent analyses for 25-hydroxyvitamin D, parathyroid hormone, β-isomerized C-terminal telopeptide, bone-specific alkaline phosphatase and procollagen type 1 N-terminal peptide, leptin, cholesterol, vitamin B12 and high sensitivity C-reactive protein 2-h timed morning urine collection; urinary calcium, phosphate, creatinine assayed, aliquots stored at −20°C 24-h urine collection (age <30 years) Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs (coded according to the structured drug codes developed by the Society of Hospital Pharmacists, Victorian Branch, 1986) and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth history, exposure (current and past of type and to Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Blood pressure Bone densitometry (DXA) —multi-site hip and of Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and of history, of Anthropometric measures: weight, Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, exposure of of and of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at analyses for β-isomerized C-terminal telopeptide, procollagen type 1 N-terminal peptide, and vitamin Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth exposure (current and past of type and to of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at not collected at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, of of of of and and of Measurements for each follow-up phase, indicated by year of commencement for women and men enrolled in the GOS cohorts Fasting blood samples taken, DNA extracted, fasting glucose, albumin, calcium, phosphate assayed, serum/plasma aliquots stored at −80°C. Subsequent analyses for 25-hydroxyvitamin D, parathyroid hormone, β-isomerized C-terminal telopeptide, bone-specific alkaline phosphatase and procollagen type 1 N-terminal peptide, leptin, cholesterol, vitamin B12 and high sensitivity C-reactive protein 2-h timed morning urine collection; urinary calcium, phosphate, creatinine assayed, aliquots stored at −20°C 24-h urine collection (age <30 years) Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs (coded according to the structured drug codes developed by the Society of Hospital Pharmacists, Victorian Branch, 1986) and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth history, exposure (current and past of type and to Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Blood pressure Bone densitometry (DXA) —multi-site hip and of Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and of history, of Anthropometric measures: weight, Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, exposure of of and of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at analyses for β-isomerized C-terminal telopeptide, procollagen type 1 N-terminal peptide, and vitamin Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth exposure (current and past of type and to of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at not collected at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, of of of of and and of Fasting blood samples taken, DNA extracted, fasting glucose, albumin, calcium, phosphate assayed, serum/plasma aliquots stored at −80°C. Subsequent analyses for 25-hydroxyvitamin D, parathyroid hormone, β-isomerized C-terminal telopeptide, bone-specific alkaline phosphatase and procollagen type 1 N-terminal peptide, leptin, cholesterol, vitamin B12 and high sensitivity C-reactive protein 2-h timed morning urine collection; urinary calcium, phosphate, creatinine assayed, aliquots stored at −20°C 24-h urine collection (age <30 years) Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs (coded according to the structured drug codes developed by the Society of Hospital Pharmacists, Victorian Branch, 1986) and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth history, exposure (current and past of type and to Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Bone densitometry (DXA) —multi-site Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Anthropometric measures: weight, Blood pressure Bone densitometry (DXA) —multi-site hip and of Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and of history, of Anthropometric measures: weight, Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, exposure of of and of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at analyses for β-isomerized C-terminal telopeptide, procollagen type 1 N-terminal peptide, and vitamin Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and fracture history, family fracture history, smoking (current and past patterns of tobacco smoking included manufactured and ‘hand-rolled’ cigarettes, cigars and pipes), and of marital Australian of based on and status and birth exposure (current and past of type and to of Fasting blood samples taken, DNA extracted, serum/plasma aliquots stored at not collected at Anthropometric measures: weight, height, waist and hip circumferences, arm span Blood pressure Bone densitometry (DXA) —multi-site hip Socio-economic status Questionnaire: self-reported exposure to drugs and diseases, falls and family history, of of of of and and of provided measures of bone mineral and BMD of the and and were at baseline for women; was at the 10-year follow-up for women and 5-year follow-up for men. and measures of and were from A was used for the and the men at the male were using a No differences were in or BMD was on aged years. All data were obtained with the of an of both was using a that provided measures of of and was using and using a arm span was using a and waist and hip with a Blood pressure and were obtained using a and assessed using a of were obtained with a using a The and timed assessed and A exposure the in a of and with a as with a Blood was collected an for women at baseline and 10-year follow-up, and for men at baseline or 5-year follow-up. aliquots of and were stored at −80°C. DNA was from At baseline, women provided a morning urine collection an and women aged 20–29 years also provided a 24-h urine A of of and clinical risk factors exposure to of and falls and and tobacco and family of fractures and diseases, of and (Table address for each was to the ABS and ABS to derive the Socio-economic Index for Areas a of of was reported in from to to (Table of were based on of (Table scores were also for at and other a questionnaire for the was used for self-reported fractures were from were recalled the A of and falls were also details of age of exposure to and were for women (Table the in and in addition to and were using the Clinical for and Statistical of of presented as in of presented as in from the GOS cohort has in and at We have reported on prevalence of osteoporosis and using Health based on BMD of using a random sample as the BMD for osteoporosis and for BMD developed by the GOS and by the Australian and Bone and Society for on in population of fractures that from women with not between BMD with and prevalence of fractures among men and the on and of The of a fracture risk was based on fracture and The is risk expressed as a be to absolute risk for of BMD that fracture risk in women to those with osteoporosis and to Health Commission for BMD and for in Australia. GOS data were the to that of C-reactive protein are with fracture that the be used as a An between levels and BMD in that the between and to A for the of was with fracture an not by An between and fracture risk the that bone is by the a between and a of bone in women this is with low and the that exposure to used in the for is with BMD was with a for the in bone We have reported that is with fracture exposure to is with and is We that women with and hip fractures had different hip to be with age of from both of the were to have low this was also for the was for An was between SES and hip We described in of 25-hydroxyvitamin with parathyroid and bone in An study that this was with an of falls in fracture and an risk of wrist and hip fractures in These informed the of a of high vitamin for the of falls and We also reported BMD among women with and among men with the of vitamin and have been described at a population with and characteristics with vitamin and an between vitamin and of bone We have on the patterns of the in and in and the of SES with and We estimated the increase in the number of women in Australia with the of the Australian to the fasting blood for the of from to We the of the Australian a developed by the for an fracture risk with was as a of In with the Barwon reported the prevalence of the in and tobacco and systemic as with risk for In with the DNA was to DNA from the GOS was included in the Australian has other BMD and with revealed that and in women increase the risk for of the The major of the study is that participants were selected at random from the The BSD provides an for as the population is and to the Australian The of the of clinical measures and of a with of health and factors for of study participants has been both and yet have recruited and loss to follow-up. low participation in cohort studies has on of baseline participation of for women and for men provided for the disease prevalence and rates 86% at 6-year follow-up and 83% at 10-year follow-up; 81% at 5-year are for the of the study. to Commonwealth electoral provided a comprehensive sampling from random samples be as a of and changes in the the cohort the of to the 1996 there was a of in the BSD with national profiles and this to 9. 5% by the by a of BSD residents have from or the the of cohort are These factors the of the study Clinical were using fractures ascertained from and major from on self-reported data to and The GOS has and data that the to risk factors for the and of disease and health in addition to major is the GOS and details and be the Barwon Health from the cohort are by the Barwon and for to the data and of are with by an of the group and the Barwon Health The was in the with from the Victorian Health was from the Health and the Geelong Barwon the and the of the the Society for Bone and and The study was by the Barwon and The of the GOS is to describe the of osteoporosis in the population and to risk factors for fracture. The cohort study was in 1494 women and 1540 selected from the electoral return to the study centre at Barwon Health years to have their bone mineral density and their health We and past for their in the cohorts and and the are also to study of