Osteoporosis-Related Health Behaviors in Men With Prostate Cancer and Survivors

Abstract

This descriptive study aimed to (a) determine the extent of osteoporosis knowledge, perceived health beliefs, and self-efficacy with bone healthy behaviors in men with prostate cancer and survivors and (b) identify how dietary bone healthy behaviors are associated with these psychobehavioral and psychosocial factors. Three different questionnaires were used to measure osteoporosis knowledge, health beliefs, and self-efficacy in a group of men with prostate cancer and survivors. Bone health was assessed via dual-energy X-ray absorptiometry and calcium intake using a diet history. The prevalence of osteoporosis and low bone mass was high at over 70%. Participants had inadequate osteoporosis knowledge with a mean score of 43.3% (SD = 18%) on the Facts on Osteoporosis Quiz. Participants scored low on the subscale measuring barriers to exercise (median = 11; interquartile range [IQR] = 6.5), indicating minimal barriers to exercise participation, and the subscale measuring the benefits of exercise scored the highest (median = 24; IQR = 3.5) compared with the other subscales. Men with prostate cancer and survivors were highly confident in their exercise and calcium self-efficacy (83.0%, IQR = 24.0% and 85.7%, IQR = 27.0%, respectively). Participants did not meet their calcium requirements or consume enough dairy products for optimum bone health. Men with prostate cancer and survivors have poor osteoporosis knowledge, but are confident in their self-efficacy of undertaking bone healthy behaviors. This confidence did not translate to specific dietary behaviors as they did not meet their calcium or dairy intake requirements. Implications for cancer survivors is that there is a need for bone health education programs among prostate cancer survivors. These programs should go beyond education and empowerment to provide practical guidance to maximize uptake of bone healthy behaviors.

Keywords

prostatic neoplasm survivors osteoporosis knowledge self-efficacy health behavior

Introduction

Men with prostate cancer are increasingly being treated with androgen deprivation therapy (ADT; Grossmann et al., 2011), which has resulted in better survival rates than a decade ago (National Cancer Institute, 2014). ADT is associated with increased bone loss leading to a high prevalence of osteoporosis, with up to 53% of men with prostate cancer being diagnosed with the disease (Lassemillante, Doi, Hooper, Prins, & Wright, 2014). Since osteoporosis has traditionally been seen as a “women’s disease,” there has been less emphasis on educating men about this bone condition resulting in men feeling less susceptible to this disease (McLeod & Johnson, 2011). A meta-analysis (Laliberté, Perreault, Jouini, Shea, & Lalonde, 2011) on osteoporosis interventions in primary care identified six published studies conducted in men and women (Ashe et al., 2004; Majumdar et al., 2004; Majumdar et al., 2007; Solomon, Katz, et al., 2007; Solomon, Polinski, et al., 2007; Yuksel, Majumdar, Biggs, & Tsuyuki, 2010), five in women only (Bessette et al., 2008; Cranney et al., 2008; Feldstein et al., 2006; Lafata et al., 2007; Pencille et al., 2009), and none in men only. Evidence suggests that bone health monitoring and interventions, such as promotion of bone healthy behaviors, are poorly implemented at initiation of ADT (Pradhan et al., 2012) despite the presence of other osteoporosis risk factors and comorbidities (Tanvetyanon, 2005). Calcium is essential to bone health and vitamin D is needed for calcium homeostasis in the body (Lips & van Schoor, 2011). Calcium supplementation, alone or in conjunction with vitamin D supplements, is associated with decreased risk of osteoporotic fractures (Tang, Eslick, Nowson, Smith, & Bensoussan, 2007). The exact role of vitamin D in bone physiology is currently under debate (Peterlik, Kállay, & Cross, 2013; Takahashi, Udagawa, & Suda, 2014), but cross-sectional studies and clinical trials have demonstrated the protective effects of this vitamin on fracture risk (Lips & van Schoor, 2011). Exercise, notably resistance-training and high-impact loading activities, has a positive impact on hip and/or spine bone mineral density (BMD) of middle-aged and older men (Bolam, van Uffelen, & Taaffe, 2013). Because the epidemiological evidence suggests that diet and exercise have a positive role, many osteoporosis intervention/education studies have included various lifestyle modifications in their protocol (Gaines & Marx, 2011; Ryan, Schlidt, & Ryan, 2013). Clinical guidelines for the management of prostate cancer from the National Comprehensive Cancer Network (2015), make comprehensive recommendations for screening, preventing, and managing poor bone health in men with prostate cancer (especially ADT candidates and users). These include fracture risk assessment, bone density monitoring, calcium and vitamin supplementation, exercise regimes, and allied health involvement (National Comprehensive Cancer Network, 2015).

Nadler et al. (2013) have reported that men with prostate cancer on ADT lack basic osteoporosis knowledge and do not engage in bone healthy behaviors such as participating in exercise, or consuming adequate calcium and vitamin D. A framework that has been used to investigate such phenomenon in osteoporosis research is the health belief model (Janz & Becker, 1984). The health belief model encompasses several primary concepts that predict why individuals will take action to prevent, to screen for, or to control illness conditions. Knowledge, health beliefs, and self-efficacy are modifiable, hence are ideal targets when planning interventions and education programs. This framework has been extensively used in osteoporosis descriptive and intervention studies focusing on women (McLeod & Johnson, 2011). Nadler et al. (2013) are one of the few who have used this framework in men with prostate cancer, and have identified that men with prostate cancer on ADT lack basic osteoporosis knowledge and do not engage in bone healthy behaviors. This study therefore aims to add to this small body of literature by determining the extent of osteoporosis knowledge, and perceived health beliefs and self-efficacy related to osteoporosis in men with prostate cancer and survivors. Men with prostate cancer undergoing treatment (ADT or other treatment) and survivors are included in this study to replicate the heterogeneity of this population as encountered in clinical practice, thus contributing to the clinical relevance of the findings presented here. The secondary aim of this study is to identify whether dietary bone healthy behaviors and biological markers of bone loss are associated with health beliefs, osteoporosis knowledge, and self-efficacy. The dietary methodology used in the current study is more robust than the one used by Nadler et al. (2013), therefore providing more accurate information about dietary bone healthy behaviors (Kristal, Peters, & Potter, 2005). Based on the osteoporosis literature and lack of bone health education in men with prostate cancer, it was hypothesized that this population has inadequate osteoporosis knowledge and poor dietary behaviors.

Method

Study Participants

This cross-sectional study was conducted at The University of Queensland, Australia, (September 2013-June 2014) and included men with prostate cancer and prostate cancer survivors, who attended an exercise clinic. Men with prostate cancer, either undergoing active treatment or not, and prostate cancer survivors (regardless of treatment) were included in this study. Eligible participants were prostate cancer survivors or men with a current prostate cancer diagnosis; aged older than 60 years; either currently undergoing active treatment or not treated; not diagnosed with a bone-related disease; free of cardiovascular, musculoskeletal, or metabolic disorders that would have prevented safe participation in exercise; and a body mass less than 150 kg. The study was approved by The University of Queensland Medical Research Ethics Committee (2013001160). Informed consent was obtained from all individual participants included in the study.

Instruments

The extent of osteoporosis knowledge was measured using the 26-item Facts on Osteoporosis Questionnaire (FOOQ) that has been validated in men. It comprises 20 items from the FOOQ (Ailinger, Harper, & Lasus, 1998), and six items from the Men’s Osteoporosis Knowledge Quiz (Gaines et al., 2011). The psychometric properties of this 26-item tool were determined in an elderly male population (validity r = 0.076 and Cronbach’s α = 0.9; Gaines et al., 2011). This tool was scored based on the percentage of correct answers, with adequate osteoporosis knowledge defined as a total score of 80% or more (Ailinger, Braun, Lasus, & Whitt, 2005). The Osteoporosis Self-Efficacy Scale (OSES) was used to measure osteoporosis-specific self-efficacy and consists of 21 items in a visual analogue format (possible score range 0% to 100%). This tool comprises two subscales measuring confidence for initiating and maintaining calcium intake (OSE-Calcium); and initiating and maintaining exercise habits (OSE-Exercise; Horan, Kim, Gendler, Froman, & Patel, 1998). The internal consistency of each subscales were r = 0.93 (for OSE-Calcium) and r = 0.94 (OSE-Exercise; Horan et al., 1998), and the reliability coefficient of the OSES was α = .9 (Sedlak, Doheny, & Estok, 2000). The Osteoporosis Health Belief Scale (OHBS) was designed with the health belief model as a framework and measures perceived seriousness, perceived susceptibility, perceived benefits, and perceived barriers related to healthy bone behaviors. This 42-item tool is divided into six subscales measuring the aforementioned constructs (benefits and barriers measured separately for calcium and exercise) and general health motivation. The responses were recorded on a 5-point scale, from strongly disagree to strongly agree, that were awarded a numerical score in increasing order (from 1 to 5). The possible range for each subscale was 6 to 30, with higher scores meaning higher perceived susceptibility, seriousness, benefits from exercise, benefits from calcium, barriers to exercise, barriers to calcium, and health motivation. The reliability of the OHBS has been tested in a wide range of gender and groups, revealing acceptable levels of reliability (Cronbach’s α = 0.7-0.9; Shanthi Johnson, McLeod, Kennedy, & McLeod, 2008).

Health and Anthropometric Measures

The outcomes of interest for the present study were collected during a structured interview and included prostate cancer characteristics and treatments, bone health and osteoporosis status, health behaviors such as smoking, and FRAX^® score. Participants with a FRAX 10-year probability risk < 20% for major fractures were classified as needing osteoporosis intervention (Kanis, Johnell, Odén, Johansson, & McCloskey, 2008). Height was measured to the nearest 0.1 cm using a stadiometer and body mass was measured to the nearest 0.1 kg using electronic stand-on scales (A&D Mercury Load Cell Digitizer; A&D Weighting, Melbourne, Australia). Body mass index was calculated as weight in kilograms divided by the square of height in meters (kg/m²). Calcium and dairy intake were derived from detailed structured diet histories (using the validated Wollongong diet history form [Martin, 2004]) collected by an Accredited Practicing Dietitian. Calcium intake was compared with Australian recommended dietary intakes (1,000 mg for men <70 years and 1,200 mg for men >70 years; National Health and Medical Research Council, 2005). Blood tests were taken after an overnight fast in two 10 mL Vacutainers via vein phlebotomy. One Vacutainer was left to clot at room temperature for 20 to 30 minutes, while the remaining Vacutainer containing anticoagulants was placed on ice. Both samples were centrifuged at 2,500 rpm for 10 minutes at 4 °C, then plasma and serum were separated into aliquots and stored at −80 °C until further processing. Samples were thawed on ice and the following biomarkers were tested using the automated Elecsys^® 2010/cobas e411 analyzer (Roche Diagnostics): serum C-terminal telopeptide of type 1 collagen (coefficient of variation [CV] of assay 3.0%), plasma osteocalcin (CV of assay 1.5%), plasma procollagen type 1 N propeptide (P1NP; CV of assay 1.5%), plasma free testosterone (CV of assay 2.0%), and plasma total prostate-specific antigen (CV of assay 3.6%). Serum bone-specific alkaline phosphatase was measured by immunoenzymetric assay (CV assay 4.2%, Immunodiagnostic System Ltd.). Plasma vitamin D was measured, with <50 nmol/L considered as vitamin D insufficiency, and <25 nmol/L as vitamin D deficiency (World Health Organization Group on the Prevention and Management of Osteoporosis, 2003). Body composition and BMD (at lumbar spine and right femoral neck) were measured using dual-energy X-ray absorptiometry (Hologic QDR 4500w) and the World Health Organization definition for osteoporosis was used (low bone mass: −1 < T-score > −2.5; osteoporosis: T-score ≤ −2.5; Kanis & Kanis, 1994).

Statistical Analyses

All normally distributed outcome variables were reported using means and standard deviations, while nonparametric outcome variables were reported using medians and IQR. Because of the primarily nonparametric nature of the outcome variables, Kendall’s tau was used to identify correlations between the psychobehavioral and psychosocial constructs and the health behaviors. Statistical significance was defined as p value ≤ 0.05. All analyses were completed using the Statistical Package for the Social Sciences software (version 22, IBM).

Results

Disease and Participants Characteristics

Fifty-four men with prostate cancer and survivors were invited to participate in this study. A total of 41 men with prostate cancer and survivors were included, as 12 declined to participate and one was ineligible to participate. The characteristics of the study participants are presented in Table 1. Most of the participants were classified as prostate cancer survivors no longer undergoing active treatment, while 26.8% (n = 11) were currently undergoing ADT. All men reported on their prostate cancer stage (past or present), with 58.5% (n = 24) reporting localized disease, 17% (n = 7) reporting advanced disease, and 24.4% (n = 10) could not recall their disease stage. Radical prostatectomy (48.8%, n = 20) and radiation therapy (34.1%, n = 14) were the most common forms of interventions reported by the study participants, and the other treatment modalities are presented in Table 1. The majority of the men in this study were overweight or obese based on their body mass index. Body composition assessment, via dual-energy X-ray absorptiometry, revealed that this sample was obese because the mean fat mass percentage exceeded 30% (30.8 ± 5.1%; Gallagher et al., 2000). Total prostate-specific antigen was negatively skewed, because two participants had active advanced disease. Testosterone levels varied, as expected, from castrate levels to normal levels (0.03-14.4 ng/mL) as this sample comprises a mix of hormone-naïve men and men on ADT.

Table 1.

Characteristics of Men With Prostate Cancer (N = 41).

Variable	M (SD) or median (IQR)	95% CI for M or median
Age, years^a	70 (7.5)	[67, 71]
Weight, kg	82.6 (10.3)	[79.4, 85.9]
Height, cm	176.5 (5.6)	[174.8, 178.3]
BMI, kg/m²	26.5 (3.2)	[25.5, 27.6]
BMI classification
Normal weight (<25 kg/m²), % (n)	26.6 (15)
Overweight (25-30 kg/m²), % (n)	46.3 (19)
Obese (> 30 kg/m²), % (n)	17.1 (7)
Current smoker, % (n)	2.4 (1)
Gleason score (possible range 2-10)^a	7 (1)	[7, 8]
Total PSA, ng/mL^a	0.04 (0.6)	[0.06, 0.2]
Time since prostate cancer diagnosis, years^a	4 (6.5)	[3, 7]
Previous or current ADT^b, % (n)	41.5 (17)
Prostate cancer treatment
Radical prostatectomy, % (n)	48.8 (20)
Radiation therapy, % (n)	34.1 (14)
Brachytherapy, % (n)	2.4 (1)
Radical prostatectomy and radiation therapy, % (n)	9.8 (4)
Chemotherapy and radiation therapy, % (n)	2.4 (1)
Body fat percentage	30.8 (5.1)	[29.2, 32.4]
Lean body mass percentage	66.1 (5.0)	[64.6, 67.7]
Lumbar spine
BMD, g/cm^2a	1.1 (0.2)	[1.0, 1.2]
T-score^a	−0.1 (2.0)	[0.3, 2.8]
Z score^a	0.8 (2.1)	[0.8, 2.1]
Right femoral neck
BMD, g/cm²	0. 8 (0.1)	[0.7, 0.8]
T-score	−1.1 (0.6)	[−1.3, –0.9]
Z score	0.1 (0.7)	[−0.2, 0.3]
Bone health (at any ROI)
Normal bone mass, % (n)	26.8 (11)
Osteopenia, % (n)	63.4 (26)
Osteoporosis, % (n)	9.8 (4)
Previous fracture, % (n)	61 (25)
FRAX^® score for major fractures^a	4.30 (3.2)	[3.7, 5.1]
Calcium supplementation, % (n)	19.5 (8)
Calcium intake, mg/day^a,c	875 (495)	[768, 995]
Meeting calcium requirements, % (n)^d	22 (9)
Vitamin D supplementation, % (n)	26.8 (11)
Plasma vitamin D, nmol/L	101.3 (25.6)	[93.2, 109.5]
Free testosterone, ng/mL^a	3.6 (5.3)	[0.3, 4.9]
SHBG, nmol/L^a	47.9 (41.8)	[40.8, 68.5]
CTx, ng/mL	0.4 (0.2)	[0.3, 0.5]
Osteocalcin, µg/L	21.6 (8.1)	[19.0, 24.3]
P1NP, ng/mL^a	44.5 (29.4)	[38.7, 54.6]
Bone ALP, µg/L^a	17.2 (8.0)	[13.8, 19.7]

Note. IQR = interquartile range; BMI = body mass index; PSA = prostate-specific antigen; ADT = androgen deprivation therapy; BMD = bone mineral density; ROI = region of interest; SHBG = sex hormone binding globulin; CTx = C-terminal telopeptide; P1NP = procollagen type 1 N-telopeptide; ALP = alkaline phosphatase. All values for continuous variable are presented as mean (SD), unless otherwise specified.

The values for these variables are presented as median (IQR). ^bCurrent and past treatment with ADT. ^cIntake including calcium from supplementation. ^dBased on calcium intake including calcium supplementation and the age appropriate recommended dietary intake.

Bone Health and Related Health Behaviors

Over 80% (n = 37) of men with prostate cancer and survivors in this study had low bone mass or osteoporosis. According to the FRAX intervention threshold for major fractures, only 2.5% (n = 1) needed reassessment of bone health in 5 years and the rest of the sample did not need any intervention. Applying the FRAX intervention threshold for hip fractures revealed that 2.5% (n = 1) of the sample met the criteria for pharmacological intervention, 7.5% (n = 3) needed reassessment of bone health in 5 years, and the rest of the sample did not need further interventions (Kanis, McCloskey, et al., 2008). Bisphosphonates were prescribed to 9.8% (n = 4) of the sample, of which only one participant met the FRAX threshold for osteoporosis intervention. All of the study participants currently taking bisphosphonates were either currently treated with or previously treated with ADT.

Intake of dairy products, such as milk, cheese, and yoghurt, ranged from 0 to 840 g per day (median 225 g/day, IQR 210 g/day). While intake of dairy products varied greatly only 2.4% (n = 1) consumed no dairy at all and 9.8 % (n = 4) consumed more than 500 g of dairy products. The rest of the sample consumed between 20 g and 450 g of dairy products per day (56.1% consumed 0-250 g dairy products per day and 31.7% consumed 250-500 g dairy products per day). Milk was the main type of dairy products consumed (50.7% of dairy intake) followed by yoghurt (23.2% of dairy intake) and cheese (10.4% of dairy intake). Calcium supplementation was reported by 19.5% (n = 8) of men and vitamin D supplementation in 26.8% (n = 11). Calcium intake from dietary sources ranged from 440 to 1645 mg per day (median 865 mg/day, IQR 310 mg/day). Total calcium intake, which is the sum of dietary calcium intake and calcium from supplements, ranged from 550 to 1970 mg per day (median 875 mg/day, IQR 495 mg/day). Because of these wide ranges of calcium intake and the varying calcium requirements based on different ages, men in this sample were stratified based on whether they met their calcium requirements or not. It was identified that less than a quarter of men with prostate cancer and survivors (22%, n = 9) met their calcium requirements, even when taking calcium supplementation into account. Three men (7%) exceeded their calcium requirements through diet alone and seven men (17%) exceeded their calcium requirements when taking supplemental calcium into account.

Osteoporosis Knowledge, Health Beliefs, and Self-Efficacy

FOOQ

The majority of men with prostate cancer and survivors had inadequate osteoporosis knowledge, with an average FOOQ score of 43.3% (SD = 18%). Most of the participants knew that “osteoporosis affects men and women” as 95.2% (n = 40) answered this question correctly. About three quarters (71.4%, n = 30) of men with prostate cancer and survivors correctly recognized “bone loss increases in men after the age of 70 years” but scored poorly (7.2-33.3%) on the remaining osteoporosis questions specific to men. For instance, only 33.3% were aware of their calcium requirement and 28.6% were aware that low testosterone level (as seen during ADT) is a risk factor for osteoporosis. More worryingly, only 7.1% (n = 3) of participants knew that hormone treatment for prostate cancer increases the risks of osteoporosis. Few men (7.1%, n = 3) were aware that body weight affects bone health and 11.9% (n = 5) knew of the elevated risks (notably fractures) associated with osteoporosis.

OHBS and OSES

The OHBS subscale measuring barriers to exercise scored the lowest (median = 11; IQR = 6.5), indicating these participants reported minimal barriers to exercise; and the subscale measuring the benefits of exercise scored the highest (median = 24; IQR = 3.5) compared with the other subscales. A similar pattern was observed for the perceived barriers and benefits of calcium intake. Participants had a moderate perception of the seriousness of and their susceptibility to osteoporosis (Figure 1). Men with prostate cancer and survivors scored highly on the perceived health motivation subscale (median = 24; IQR = 5). Men with prostate cancer and survivors reported high confidence on the OSE-Exercise and OSE-Calcium (83.0%, IQR = 24.0% and 85.7%, IQR = 27.0%, respectively). On further examination of these subscales, it was identified that 26.8% (n = 11) and 36.6% (n = 15) of the study participants were poorly to moderately confident (score < 75% on the visual analogue scale) in their exercise and calcium intake self-efficacy.

Figure 1.

Median score on the seven subscales of the Osteoporosis Health Belief Score. Minimum possible score = 6 and maximum possible score = 30.

The correlations between the bone healthy behaviors and the psychobehavioral and psychosocial factors are summarized in Table 2. The correlations were small to moderate (Cohen, 2003) but similar to previous studies investigating osteoporosis knowledge and health beliefs (Gaines & Marx, 2011; McLeod & Johnson, 2011). Osteoporosis knowledge was positively correlated with general health motivation (τ = 0.26, p = 0.05) and perceived benefits of exercise (τ = 0.26, p = 0.05), and negatively correlated with perceived barriers to calcium intake (τ = −0.47, p < 0.001). There was a positive correlation between the perceived benefits of exercise and the perceived benefits of calcium intake (τ = 0.41, p = 0.003). Testosterone levels were also negatively correlated with perceived seriousness of osteoporosis (τ = −0.37, p = 0.01).

Table 2.

Correlation coefficient (τ) for associations between osteoporosis knowledge, health beliefs and self-efficacy, and osteoporosis behaviors and biomarkers of bone health.

n = 33, τ (p Value)	Osteoporosis knowledge	Perceived susceptibility	Perceived seriousness	Perceived benefits exercise	Perceived benefits calcium	Perceived barriers exercise	Perceived barriers calcium	Health motivation	Exercise self-efficacy	Calcium self-efficacy
Calcium intake	0.03 (0.80)	−0.11 (0.39)	0.21 (0.09)	0.28 (0.03)	0.14 (0.27)	−0.21 (0.10)	−0.06 (0.62)	0.05 (0.72)	0.13 (0.29)	0.02 (0.90)
Dairy intake	0 (1.00)	0.12 (0.34)	0.11 (0.38)	0.01 (0.91)	0.04 (0.74)	−0.09 (0.50)	0.04 (0.78)	0.13 (0.31)	0.05 (0.71)	−0.04 (0.73)
Number of alcoholic drinks/week	−0.14 (0.27)	0.16 (0.23)	0.09 (0.77)	−0.05 (0.70)	−0.18 (0.18)	0.03 (0.84)	0.27 (0.04)	0.04 (0.75)	−0.01 (0.96)	−0.04 (0.74)
BMI	−0.13 (0.31)	0.03 (0.79)	0.02 (0.88)	−0.05 (0.69)	0.07 (0.60)	0.24 (0.06)	0.26 (0.04)	−0.20 (0.11)	−0.30 (0.02)	−0.17 (0.16)
FRAX^® for major fractures	−0.02 (0.90)	0.17 (0.18)	−0.23 (0.06)	−0.18 (0.16)	−0.28 (0.03)	−0.12 (0.36)	0.05 (0.67)	−0.06 (0.65)	0.03 (0.83)	0.04 (0.76)
Right femoral neck T-score	−0.04 (0.76)	−0.05 (0.68)	0.34 (0.01)	0.02 (0.86)	0.03 (0.80)	0.10 (0.46)	0.15 (0.23)	0.13 (0.31)	−0.21 (0.09)	−0.15 (0.24)
Right femoral neck Z score	−0.11 (0.37)	−0.03 (0.84)	0.30 (0.02)	−0.08 (0.55)	0.13 (0.34)	0.11 (0.42)	0.17 (0.18)	0.21 (0.11)	−0.17 (0.18)	−0.13 (0.31)
Lumbar spine T-score	0.07 (0.55)	−0.11 (0.41)	0.23 (0.07)	0.01 (0.96)	0.10 (0.43)	−0.05 (0.68)	−0.18 (0.17)	−0.02 (0.88)	−0.22 (0.07)	0.06 (0.61)
Lumbar spine T-score	0.06 (0.64)	−0.11 (0.37)	0.18 (0.16)	−0.06 (0.65)	0.05 (0.68)	0.01 (0.96)	−0.11 (0.38)	−0.03 (0.83)	−0.25 (0.04)	0 (1.00)
CTx (ng/mL)	0.05 (0.70)	0.17 (0.19)	−0.17 (0.17)	−0.29 (0.02)	−0.14 (0.29)	0.10 (0.44)	−0.04 (0.73)	−0.01 (0.95)	0.04 (0.72)	0.06 (0.61)
OC (ng/mL)	0.08 (0.51)	0.07 (0.61)	−0.15 (0.23)	−0.23 (0.07)	−0.08 (0.56)	0.14 (0.25)	−0.09 (0.49)	0.03 (0.84)	0 (1.00)	0.09 (0.46)
Testosterone (ng/mL)	0 (1.00)	−0.37 (0.01)	0.22 (0.09)	0.26 (0.05)	0.21 (0.11)	−0.01 (0.91)	0 (0.98)	−0.03 (0.81)	0.10 (0.42)	−0.07 (0.60)
P1NP (ng/mL)	0.07 (0.56)	0.13 (0.31)	−0.17 (0.18)	−0.29 (0.03)	−0.12 (0.35)	0.15 (0.24)	−0.06 (0.64)	<0.00 (0.99)	−0.01 (0.95)	0.10 (0.44)

Note. BMI = body mass index; CTx = C-terminal telopeptide; OC = osteocalcin; P1NP = procollagen type 1 N-telopeptide. Correlation coefficients are highlighted if p ≤ 0.05, with p values presented in parentheses.

Discussion

Osteoporosis Knowledge, Health Beliefs, and Prostate Cancer

This is the first study to investigate the extent of osteoporosis knowledge, health beliefs, and self-efficacy, and their associations with health behaviors in a well-characterized group of prostate cancer survivors. The present sample, although small, was well characterized in terms of disease profile, bone biomarkers, health behaviors, and behavioral and psychosocial constructs. The significant conclusions for men with prostate cancer and survivors were (a) over 70% had poor bone health; (b) they did not consume enough calcium, which is essential to bone health; (c) they have inadequate osteoporosis knowledge; and (d) did not perceive they were susceptible to osteoporosis or that it was a serious disease. According to the health belief model (Rosenstock, Strecher, & Becker, 1988), these constructs are important to initiate and maintain health behavior change. Such change, especially healthy bone behaviors, are important in men with prostate cancer, who have a high prevalence of poor bone health regardless of ADT status (Lassemillante, Doi, Hooper, Prins, & Wright, 2015; Lassemillante et al., 2014). The findings presented in this article support the idea that osteoporosis is a “silent disease” (Nguyen, Center, & Eisman, 2004) and reflects a lack of osteoporosis education, patient empowerment, and bone health monitoring for men with prostate cancer (Alibhai et al., 2006; Alibhai, Yun, Cheung, & Paszat, 2012; Tanvetyanon, 2005). On the other hand, the negative statistically significant correlation between testosterone levels and perceived susceptibility to osteoporosis reported here, suggests that hypogonadism may be associated with this health belief. These results therefore indicate that only those at greatest risk of fractures and osteoporosis are informed of this bone condition, hence neglecting prostate cancer survivors and those not on ADT. Clinical guidelines recommend bone health monitoring and calcium supplementation in prostate cancer patients on ADT (National Comprehensive Cancer Network, 2014), therefore explaining why they may feel more susceptible to poor bone health. Prostate cancer survivors and those not on ADT should not be neglected in terms of bone health education and empowerment as they experience a higher prevalence of osteoporosis and low bone mass than healthy older men (Lassemillante et al., 2015), likely due to other cancer treatments such as radiation therapy (Daniell et al., 2001). Osteoporosis education focusing on health beliefs alone does not incite behavior change (Rizzoli, Abraham, & Brandi, 2014), therefore justifying the need for practical information on how to tackle barriers to behavior change.

Osteoporosis Knowledge, Health Beliefs, and Dietary Behaviors

This study is in line with previous research (Lassemillante et al., 2015) whereby men with prostate cancer do not meet their calcium requirements despite being at risk of osteoporosis, for example, secondary to ADT. This can be in part due to poor osteoporosis knowledge that led to greater barriers to calcium intake, as reported in this study. In this study, trends in the results indicated calcium intake was not associated with calcium intake self-efficacy or knowledge, but was more closely linked with perceived exercise benefits. This finding demonstrates the importance of multiple disciplines in osteoporosis education and specific bone healthy behaviors. Despite the positive association between perceived exercise benefits and perceived calcium benefits, this did not equate to adequate calcium intake, highlighting the gap between health beliefs and actual health behaviors. The results reported in this study outline that men with prostate cancer and survivors are not aware of their calcium requirements and this may also help in explaining why they do not meet their requirements. The majority of this sample consumed less than the equivalent of 1 cup of milk per day, which is short of the current recommendations of 2.5 cups per day (National Health and Medical Research Council, 2013). Even those taking calcium supplements fell short of their requirements, likely due to low dietary calcium intake. On the other hand, a small percentage of men were consuming high levels of calcium, mainly in the form of supplements. The cardiovascular risk associated from such intakes (supplemental calcium and dietary calcium) are currently under debate (Waldman, Sarbaziha, Merz, & Shufelt, 2015) with more randomized controlled trials needed to ascertain this risk. In the meantime, it is important to ensure that men do not consume calcium in excess of their requirements. The positive correlation between confidence in calcium intake self-efficacy and calcium intake suggests that promoting awareness on how to implement healthy bone strategies is more likely to contribute to calcium intake. A recent review of osteoporosis interventions (Ryan et al., 2013) reported that programs that included skills training were more successful at increasing calcium intake than those that did not include such component. Similarly, self-efficacy remained unchanged after many osteoporosis intervention programs (Francis, Matthews, Van Mechelen, Bennell, & Osborne, 2009; Sedlak, Doheny, & Jones, 1998; Tung & Lee, 2006) despite improvements in osteoporosis knowledge and health beliefs. Therefore, to affect behavior change, it is imperative to go beyond traditional osteoporosis education programs by incorporating other behavioral change theoretical models in the interventions, including social contact, providing longer interventions (over a few months), and be multidimensional (Ryan et al., 2013). Rizzoli et al. (2014) discuss such novel approaches that are being implemented in new osteoporosis intervention programs (Gianoudis et al., 2012) but it is also believed that dietary interventions need to include individualized care. As a result, the dietary preventative behaviors will be tailored to one’s social, family, and financial circumstances. Because men often survive many years after being diagnosed and treated for prostate cancer, it is important to educate them about other aspects of their health, especially those that could have been affected by cancer treatments. More osteoporosis intervention studies are needed but these also need to be sex specific. They also need to take into consideration the health behaviors of men with prostate cancer and survivors in providing practical guidance on how to maintain healthy bone behaviors.

Some of the results reported here, such as osteoporosis knowledge, are in accordance with those reported in a cross-sectional study by Nadler et al. (2013); however, some results such as calcium adequacy are markedly different. This can be explained by a lower proportion of men taking calcium supplements in the present study (19.5% vs. 60%; Nadler et al., 2013). While both studies present different findings, attributed to differences between the study populations, they both outline the gaps in the bone health (related behaviors and determinants) of men with prostate cancer. A recent intervention by Nadler, Alibhai, Catton, Catton, and Jones (2014), providing written education material on osteoporosis, resulted in increased calcium intake among men with prostate cancer who did not meet their requirements. While this intervention was simple, it was based on established behavior change models; therefore, it suggests that more comprehensive interventions, incorporating allied health professionals and behavior changed models, may lead to additional health behavior changes.

Strengths and Limitations

Because the study participants were recruited from an exercise clinic, their confidence in exercise self-efficacy and other related exercise health beliefs measured here will differ from the rest of the prostate cancer population. While exercise and physical activity were reported by the study participants, these data were not objectively measured therefore have not been included in this article. This study reports on men with prostate cancer on ADT as well as underrepresented groups in prostate cancer research, that is, prostate cancer survivors and hormone-naïve men with prostate cancer. Given the association between hypogonadism and perceived susceptibility to osteoporosis, it is anticipated that investigating hormone-naïve men with prostate cancer and survivors will result in poorer psychobehavioral and psychosocial scores than reported here. The tool measuring calcium and exercise self-efficacy specifies what exercise means (“activities such as walking, swimming, golfing, biking, aerobic dancing”) but does not give examples of the terminology “calcium-rich foods,” which is used throughout this questionnaire. As a result, the calcium intake self-efficacy responses might be biased for men who may not be aware of examples of calcium-rich foods; unfortunately, such bias could not be controlled for as data were not collected on their osteoporosis-specific dietary knowledge. Although the sample is small, hence associated statistical analyses problematic, it offers a glimpse of the gap in osteoporosis education in men with prostate cancer and survivors while raising further questions for research and practice. A strength of this study is the robust dietary methodology used to measure calcium intake that is more accurate than the methods used in similar studies (Nadler et al., 2013; Ryan et al., 2013) that have used calcium questionnaires or food frequency questionnaire to gather such data. Overall, the finding reported here are innovative as few studies on the determinants of behavior change also report on anthropometric measures (BMD) and health measures (bone health biomarkers).

Conclusion

Men with prostate cancer and survivors have inadequate osteoporosis knowledge, regardless of ADT status. It is concerning as one would expect that men on ADT would have better knowledge of this bone condition and subsequent bone healthy behaviors since it is a side effect of their treatment. Intervention program designed for men with prostate cancer and survivors are required to address treatment-related bone loss and health behaviors that they may have adopted to manage their cancer. The findings presented here support the complementary role of the multidiscipline approach to the management of bone health postprostate cancer treatment. This approach should also be individually tailored, while being innovative to effect sustainable behavior change.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Ailinger

R. L.

Braun

M. A.

Lasus

Whitt

(2005). Factors influencing osteoporosis knowledge: A community study. Journal of Community Health Nursing, 22, 135-142.

Ailinger

R. L.

Harper

D. C.

Lasus

H. A.

(1998). Bone up on osteoporosis: Development of the Facts on Osteoporosis Quiz. Orthopaedic Nursing, 17, 66-73.

Alibhai

S. M.

Rahman

Warde

P. R.

Jewett

M. A.

Jaffer

Cheung

A. M.

(2006). Prevention and management of osteoporosis in men receiving androgen deprivation therapy: A survey of urologists and radiation oncologists. Urology, 68, 126-131. doi:10.1016/j.urology.2006.01.054

Alibhai

S. M.

Yun

Cheung

A. M.

Paszat

(2012). Screening for osteoporosis in men receiving androgen deprivation therapy. Journal of the American Medical Association, 307, 255-256. doi:10.1001/jama.2011.2022

Ashe

Khan

Guy

Kruse

Hughes

O’Brien

. . . McKay

(2004). Wristwatch—distal radial fracture as a marker for osteoporosis investigation: A controlled trial of patient education and a physician alerting system. Journal of Hand Therapy, 17, 324-328. doi:10.1197/j.jht.2004.04.001

Bessette

Ste-Marie

L.-G.

Jean

Shawn Davison

Beaulieu

Baranci

. . . Brown

J. P.

(2008). Recognizing osteoporosis and its consequences in Quebec (ROCQ): Background, rationale, and methods of an anti-fracture patient health-management programme. Contemporary Clinical Trials, 29, 194-210. doi:10.1016/j.cct.2007.07.007

Bolam

K. A.

van Uffelen

J. G. Z.

Taaffe

D. R.

(2013). The effect of physical exercise on bone density in middle-aged and older men: A systematic review. Osteoporosis International, 24, 2749-2762. doi:10.1007/s00198-013-2346-1

Cohen

(2003). Applied multiple regression/correlation analysis for the behavioral sciences. Mahwah, NJ: Lawrence Erlbaum.

Cranney

Lam

Ruhland

Brison

Godwin

Harrison

M. M.

. . . Graham

I. D.

(2008). A multifaceted intervention to improve treatment of osteoporosis in postmenopausal women with wrist fractures: A cluster randomized trial. Osteoporosis International, 19, 1733-1740. doi:10.1007/s00198-008-0669-0

10.

Daniell

H. W.

Clark

J. C.

Pereira

S. E.

Niazi

Z. A.

Ferguson

D. W.

Dunn

S. R.

. . . Stratte

P. T.

(2001). Hypogonadism following prostate-bed radiation therapy for prostate carcinoma. Cancer, 91, 1889-1895. doi:10.1002/1097-0142(20010515)91:10<1889::AID-CNCR1211>3.0.CO;2-U

11.

Feldstein

Elmer

P. J.

Smith

D. H.

Herson

Orwoll

Chen

. . . Swain

M. C.

(2006). Electronic medical record reminder improves osteoporosis management after a fracture: A randomized, controlled trial. Journal of the American Geriatrics Society, 54, 450-457. doi:10.1111/j.1532-5415.2005.00618.x

12.

Francis

K. L.

Matthews

B. L.

Van Mechelen

Bennell

K. L.

Osborne

R. H.

(2009). Effectiveness of a community-based osteoporosis education and self-management course: A wait list controlled trial. Osteoporosis International, 20, 1563-1570. doi:10.1007/s00198-009-0834-0

13.

Gaines

J. M.

Marx

K. A.

(2011). Older men’s knowledge about osteoporosis and educational interventions to increase osteoporosis knowledge in older men: A systematic review. Maturitas, 68, 5-12. doi:10.1016/j.maturitas.2010.08.013

14.

Gaines

J. M.

Marx

K. A.

Narrett

Caudill

Landsman

Parrish

J. M.

(2011). Validation of the male Osteoporosis Knowledge Quiz. American Journal of Men’s Health, 5, 78-83.

15.

Gallagher

Heymsfield

S. B.

Heo

Jebb

S. A.

Murgatroyd

P. R.

Sakamoto

(2000). Healthy percentage body fat ranges: An approach for developing guidelines based on body mass index. American Journal of Clinical Nutrition, 72, 694-701.

16.

Gianoudis

Bailey

Sanders

Nowson

Hill

Ebeling

Daly

(2012). Osteo-cise: Strong bones for life: Protocol for a community-based randomised controlled trial of a multi-modal exercise and osteoporosis education program for older adults at risk of falls and fractures. BMC Musculoskeletal Disorders, 13(1), 78. Retrieved from http://www.biomedcentral.com/1471-2474/13/78

17.

Grossmann

Hamilton

E. J.

Gilfillan

Bolton

Joon

D. L.

Zajac

J. D.

(2011). Bone and metabolic health in patients with non-metastatic prostate cancer who are receiving androgen deprivation therapy. Medical Journal of Australia, 194, 301-306.

18.

Horan

M. L.

Kim

K. K.

Gendler

Froman

R. D.

Patel

M. D.

(1998). Development and evaluation of the osteoporosis self-efficacy scale. Research in Nursing & Health, 21, 395-403.

19.

Janz

N. K.

Becker

M. H.

(1984). The Health Belief Model: A decade later. Health Education Quarterly, 11(1), 1-47.

20.

Kanis

J. A.

Johnell

Odén

Johansson

McCloskey

(2008). FRAX^® and the assessment of fracture probability in men and women from the UK. Osteoporosis International, 19, 385-397. doi:10.1007/s00198-007-0543-5

21.

Kanis

J. A.

Kanis

J. A.

(1994). Assessment of fracture risk and its application to screening for postmenopausal osteoporosis: Synopsis of a WHO report. Osteoporosis International, 4, 368-381. doi:10.1007/bf01622200

22.

Kanis

J. A.

McCloskey

E. V.

Johansson

Strom

Borgstrom

Oden

(2008). Case finding for the management of osteoporosis with FRAX^®—Assessment and intervention thresholds for the UK. Osteoporosis International, 19, 1395-1408. doi:10.1007/s00198-008-0712-1

23.

Kristal

A. R.

Peters

Potter

J. D.

(2005). Is it time to abandon the food frequency questionnaire? Cancer Epidemiology, Biomarkers & Prevention, 14, 2826-2828. doi:10.1158/1055-9965.epi-12-ed1

24.

Lafata

Kolk

Peterson

McCarthy

Weiss

Chen

Y.-T.

Muma

(2007). Improving osteoporosis screening: Results from a randomized cluster trial. Journal of General Internal Medicine, 22, 346-351. doi:10.1007/s11606-006-0060-9

25.

Laliberté

M. C.

Perreault

Jouini

Shea

B. J.

Lalonde

(2011). Effectiveness of interventions to improve the detection and treatment of osteoporosis in primary care settings: A systematic review and meta-analysis. Osteoporosis International, 22, 2743-2768. doi:10.1007/s00198-011-1557-6

26.

Lassemillante

A.-C. M.

Doi

S. A.

Hooper

J. D.

Prins

J. B.

Wright

O. R.

(2014). Prevalence of osteoporosis in prostate cancer survivors: A meta-analysis. Endocrine, 45, 370-381. doi:10.1007/s12020-013-0083-z

27.

Lassemillante

A.-C. M.

Doi

S. A.

Hooper

J. D.

Prins

J. B.

Wright

O. R.

(2015). Prevalence of osteoporosis in prostate cancer survivors II: A meta-analysis of men not on androgen deprivation therapy. Endocrine. doi:10.1007/s12020-015-0536-7

28.

Lips

van Schoor

N. M.

(2011). The effect of vitamin D on bone and osteoporosis. Best Practice & Research Clinical Endocrinology & Metabolism, 25, 585-591. doi:10.1016/j.beem.2011.05.002

29.

Majumdar

S. R.

Beaupre

L. A.

Harley

C. H.

Hanley

D. A.

Lier

D. A.

. . . Morrish

D. W.

(2007). Use of a case manager to improve osteoporosis treatment after hip fracture: Results of a randomized controlled trial. Archives of Internal Medicine, 167, 2110-2115. doi:10.1001/archinte.167.19.2110

30.

Majumdar

S. R.

Rowe

B. H.

Folk

Johnson

J. A.

Holroyd

B. H.

Morrish

D. W.

. . . Russell

A. S.

(2004). A controlled trial to increase detection and treatment of osteoporosis in older patients with a wrist fracture. Annals of Internal Medicine, 141, 366-373.

31.

Martin

G. S.

(2004). The interviewer-administered, open-ended diet history method for assessing usual dietary intakes in clinical research: Relative and criterion validation studies (Doctoral dissertation). Retrieved from http://ro.uow.edu.au/theses/204/

32.

McLeod

K. M.

Johnson

C. S.

(2011). A systematic review of osteoporosis health beliefs in adult men and women. Journal of Osteoporosis, 2011, 197454. Retrieved from http://www.hindawi.com/journals/jos/2011/197454/

33.

Nadler

Alibhai

Catton

Jones

(2014). The impact of bone mineral density testing, fracture assessment, and osteoporosis education in men treated by androgen deprivation for prostate cancer: A pilot study. Supportive Care in Cancer, 22, 2409-2415. doi:10.1007/s00520-014-2183-6

34.

Nadler

Alibhai

Catton

M. J.

Jones

J. M.

(2013). Osteoporosis knowledge, health beliefs, and healthy bone behaviours in patients on androgen-deprivation therapy (ADT) for prostate cancer. BJU International, 111, 1301-1309. doi:10.1111/j.1464-410X.2012.11777.x

35.

National Cancer Institute. (2014). SEER Cancer Statistics Review 1975-2011: Prostate—Annual Death Rates. Retrieved from http://seer.cancer.gov/csr/1975_2011/browse_csr.php?sectionSEL=23&pageSEL=sect_23_table.06.html

36.

National Comprehensive Cancer Network. (2014). NCCN guidelines for prostate cancer (Version 2.2014). Fort Washington, PA: Author.

37.

National Comprehensive Cancer Network. (2015). NCCN guidelines for prostate cancer (Version 1.2015). Fort Washington, PA: Author.

38.

National Health and Medical Research Council. (2005). Nutrient reference values for Australia and New Zealand. Retrieved from https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/n35.pdf

39.

National Health and Medical Research Council. (2013). Australian dietary guidelines. Retrieved from https://www.nhmrc.gov.au/_files_nhmrc/publications/attachments/n55a_australian_dietary_guidelines_summary_131014.pdf

40.

Nguyen

T. V.

Center

J. R.

Eisman

J. A.

(2004). Osteoporosis: Underrated, underdiagnosed and undertreated. Medical Journal of Australia, 180(5 Suppl.), S18-S22.

41.

Pencille

Campbell

Van Houten

Shah

Mullan

Swiglo

. . . Montori

(2009). Protocol for the osteoporosis choice trial: A pilot randomized trial of a decision aid in primary care practice. Trials, 10(1), 113. Retrieved from http://www.trialsjournal.com/content/10/1/113

42.

Peterlik

Kállay

Cross

H. S.

(2013). Calcium nutrition and extracellular calcium sensing: Relevance for the pathogenesis of osteoporosis, cancer and cardiovascular diseases. Nutrients, 5, 302-327.

43.

Pradhan

M. R.

Mandhani

Chipde

S. S.

Srivastava

Singh

Kapoor

(2012). Bone densitometric assessment and management of fracture risk in Indian men of prostate cancer on androgen deprivation therapy: Does practice pattern match the guidelines? Indian Journal of Urology, 28, 399-404. doi:10.4103/0970-1591.105750

44.

Rizzoli

Abraham

Brandi

M.-L.

(2014). Nutrition and bone health: Turning knowledge and beliefs into healthy behaviour. Current Medical Research and Opinion, 30, 131-141. doi:10.1185/03007995.2013.847410

45.

Rosenstock

I. M.

Strecher

V. J.

Becker

M. H.

(1988). Social learning theory and the Health Belief Model. Health Education & Behavior, 15, 175-183. doi:10.1177/109019818801500203

46.

Ryan

Schlidt

Ryan

(2013). The impact of osteoporosis prevention programs on calcium intake: A systematic review. Osteoporosis International, 24, 1791-1801. doi:10.1007/s00198-012-2259-4

47.

Sedlak

C. A.

Doheny

M. O.

Estok

P. J.

(2000). Osteoporosis in older men: Knowledge and health beliefs. Orthopaedic Nursing, 19, 38-42, 44-46.

48.

Sedlak

C. A.

Doheny

M. O.

Jones

S. L.

(1998). Osteoporosis prevention in young women. Orthopaedic Nursing, 17, 53-60. Retrieved from http://journals.lww.com/orthopaedicnursing/Fulltext/1998/05000/Osteoporosis_Prevention_In_Young_Women_.9.aspx

49.

Shanthi Johnson

McLeod

Kennedy

McLeod

. (2008). Osteoporosis health beliefs among younger and older men and women. Health Education & Behavior, 35, 721-733. doi:10.1177/1090198107301331

50.

Solomon

D. H.

Katz

J. N.

Finkelstein

J. S.

Polinski

J. M.

Stedman

Brookhart

M. A.

. . . Avorn

(2007). Osteoporosis improvement: A large-scale randomized controlled trial of patient and primary care physician education. Journal of Bone and Mineral Research, 22, 1808-1815. doi:10.1359/jbmr.070717

51.

Solomon

D. H.

Polinski

Stedman

Truppo

Breiner

Egan

. . . Brookhart

M. A.

(2007). Improving care of patients at-risk for osteoporosis: A randomized controlled trial. Journal of General Internal Medicine, 22, 362-367. doi:10.1007/s11606-006-0099-7

52.

Takahashi

Udagawa

Suda

(2014). Vitamin D endocrine system and osteoclasts. BoneKEy Reports, 3. doi:10.1038/bonekey.2013.229

53.

Tang

B. M.

Eslick

G. D.

Nowson

Smith

Bensoussan

(2007). Use of calcium or calcium in combination with vitamin D supplementation to prevent fractures and bone loss in people aged 50 years and older: A meta-analysis. Lancet, 370, 657-666. doi:10.1016/s0140-6736(07)61342-7

54.

Tanvetyanon

(2005). Physician practices of bone density testing and drug prescribing to prevent or treat osteoporosis during androgen deprivation therapy. Cancer, 103, 237-241. doi:10.1002/cncr.20766

55.

Tung

W. C.

Lee

I. F. K.

(2006). Effects of an osteoporosis educational programme for men. Journal of Advanced Nursing, 56, 26-34. doi:10.1111/j.1365-2648.2006.03976.x

56.

Waldman

Sarbaziha

Merz

C. N. B.

Shufelt

(2015). Calcium supplements and cardiovascular disease: A review. American Journal of Lifestyle Medicine, 9, 298-307. doi:10.1177/1559827613512593

57.

World Health Organization Group on the Prevention and Management of Osteoporosis. (2003). Prevention and management of osteoporosis: Report of a WHO scientific group. Geneva, Switzerland: Author.

58.

Yuksel

Majumdar

S. R.

Biggs

Tsuyuki

R. T.

(2010). Community pharmacist-initiated screening program for osteoporosis: Randomized controlled trial. Osteoporosis International, 21, 391-398. doi:10.1007/s00198-009-0977-z