Archives of Osteoporosis

, 8:155

Concern and risk perception of osteoporosis and fracture among post-menopausal Australian women: results from the Global Longitudinal Study of Osteoporosis in Women (GLOW) cohort

Authors

    • Institute of Bone and Joint Research, Kolling Institute of Medical ResearchUniversity of Sydney
  • J. S. Chen
    • Institute of Bone and Joint Research, Kolling Institute of Medical ResearchUniversity of Sydney
  • L. M. March
    • Institute of Bone and Joint Research, Kolling Institute of Medical ResearchUniversity of Sydney
Original Article

DOI: 10.1007/s11657-013-0155-y

Cite this article as:
Barcenilla-Wong, A.L., Chen, J.S. & March, L.M. Arch Osteoporos (2013) 8: 155. doi:10.1007/s11657-013-0155-y

Abstract

Purpose

The purpose of this study is to identify factors associated with concern and perception of risks of osteoporosis and osteoporotic fractures and determine whether bone mineral density (BMD) testing influenced concern and risk perception.

Methods

Study subjects (n = 1,082, age 55–94 years) were female Australian participants of the Global Longitudinal Study of Osteoporosis in Women (GLOW). Self-administered questionnaires were sent annually from 2007 to 2010. Study outcomes included ‘concern about osteoporosis’, ‘perception of getting osteoporosis’ and ‘perception of fracture risk’ compared to similar aged women. The closest post-BMD testing or baseline questionnaires were used for women with and without BMD testing, respectively. Multinomial logistic regression was used for the analysis.

Results

BMD testing, prior fracture after age 45, younger age and lower self-reported general health were significantly associated with being ‘very’ or ‘somewhat concerned’ about osteoporosis and having a ‘much higher’ or ‘little higher’ risk perception of osteoporosis and fractures. A poorer BMD result was associated with higher concern and higher risk perceptions. The presence of comorbidities, having ≥2 falls in the preceding year and maternal osteoporosis were associated with higher concern. Maternal osteoporosis, presence of comorbidities, weight loss of ≥5 kg in the preceding year and low body mass index were associated with higher perceptions of osteoporosis risk.

Conclusion

Women's concern and risk perception of osteoporosis and osteoporotic fractures were reasonably well founded. However, increasing age, height loss, smoking and drinking were not associated with concern and perception despite being known osteoporosis risk factors. These factors should be considered in planning for education and awareness raising programmes.

Keywords

OsteoporosisPost-menopausal womenRisk perceptionConcern

Introduction

Osteoporosis has commonly been termed as a ‘silent’ disease in that most people are unaware that they have the condition until it manifests as a fracture. The financial and personal burden of osteoporosis-related fractures worldwide as well as in the ageing Australian society has been well documented [14]. In 2005, there were over two million osteoporotic fractures in the USA costing almost US$17 billion in associated costs [1]. In the 2000–2001 financial year, Australian $1.9 billion was spent on direct costs associated with fractures in Australia [2]. In 2007, it was estimated that one Australian was hospitalised with an osteoporosis-related fracture every 5–6 min. Data from the 2011 National Health Survey revealed that 725,500 Australians (600,300 women ≥55 years) self-reported to having osteoporosis with the number increasing from 692,300 Australians (581,800 women ≥55 years) suffering from the disease in 2006 [3]. A recent report indicated that 22.8 % of Australian women aged ≥50 years had osteoporosis [4]. Another publication estimated 3 million Australian who will have the disease by 2021 [2].

As osteoporosis is largely preventable, appropriate and targeted education is paramount in reducing the burden of this disease. The gaps in knowledge of osteoporosis in both men and women in varied age groups have been previously explored [510]. In these studies, many concluded that more was needed in educating both patients and healthcare providers especially in key risk factors for osteoporosis. Despite these educational recommendations, a recent publication from a large multinational study of post-menopausal women has highlighted the underappreciation of actual risk of osteoporosis and fracture despite the presence of personal risk factors [11].

Although some studies have focused on factors associated with knowledge of osteoporosis [10], factors contributing to women's concern and perception of risks of osteoporosis and osteoporotic fractures have not been extensively studied. As awareness to one's susceptibility to a disease may lead to an individual taking positive action to reduce their risk, it is vital to look at factors associated with concern and perception of risks of osteoporosis and osteoporotic fractures. In doing so, key areas of deficiency can be identified to emphasise in future planning for osteoporosis education.

The purpose of this study is to identify factors associated with concern and perception of risks of osteoporosis and osteoporotic fractures and determine whether bone mineral density (BMD) testing influenced concern and risk perception.

Methods

GLOW participants and recruitment

The Global Longitudinal Study of Osteoporosis in Women (GLOW) is a longitudinal, observational cohort study of over 60,000 women ≥55 years from 17 study sites in ten countries, including Australia. Details of the study were described previously [12]. In Australia, a convenience sample of 8,029 eligible female patients was identified through 51 general practitioners from 14 practices around Sydney, between January 2007 and February 2008. All eligible female patients were mailed a GLOW information packet containing study information, a participant consent form and a reply-paid envelope through their general practitioners, inviting them to participate in the study. Written consent was received from 3,011 (38 %) patients who were then mailed the self-administered baseline GLOW questionnaire and a reply-paid envelope. Completed questionnaires were received from 2,904 (96 %) patients (age ranging between 55 and 96 years) who formed the Australian baseline GLOW study sample. The study was approved by the Northern Sydney Human Research Ethics Committee.

Study subjects

Using this subject pool, analyses were conducted on a subgroup of participants (n = 1,082, age 55–94 years) to explore factors associated with concern for osteoporosis and perception of risks of osteoporosis and fracture. Subjects with a confirmed BMD (n = 717) were compared to those who did not have BMD (i.e. answered ‘no’ to having a bone density test in any relevant subsequent survey(s) and also had no results found during the BMD retrieval stage) (n = 365). Women whose BMD status could not be confirmed were excluded from analysis (n = 1,822). BMD reports from 717 participants from 2002 to 2010 were obtained. T-scores were extracted from the reports. Participants were then categorised using their T-scores as either having normal bone density (T-score >−1), osteopenic (T-score between −1 and >−2.5) or osteoporotic (T-score ≤−2.5). The closest post-BMD testing questionnaires or baseline questionnaires were used for women with and without BMD testing, respectively.

Questionnaires

Self-administered questionnaires exploring various aspects of bone health including patient characteristics, risk factors, use of medications, perception of risks and health care use were sent annually from 2007 to 2010. Details regarding the questionnaire have been previously described [12]. For the purposes of this study, outcomes analysed included ‘concern about osteoporosis’, ‘perception of risk of getting osteoporosis’ and ‘perception of risk of having a fracture’ compared to other women of similar age. ‘Concern about osteoporosis’ was assessed by asking participants to rate their level of concern about osteoporosis (i.e. very concerned, a little concerned or not at all concerned). Perception of risk of getting osteoporosis and having a fracture were both assessed by asking participants to rate their own risk of getting osteoporosis and also their own risk of having a fracture (i.e. a lot lower, a little lower, about the same, a little higher or much higher compared to other women their age). Other study outcomes used for descriptive purposes as well as in the regression models included known risk factors for osteoporosis such as age, current smoking, heavy drinking (≥7 drinks/week), body mass index (BMI), maternal osteoporosis and maternal fracture, previous fracture after age 45 and number of falls in the 12 months preceding the survey. The presence of pre-existing conditions (including asthma, emphysema, ulcerative colitis, Parkinson's disease, cancer, diabetes, osteoarthritis, rheumatoid arthritis), self-reported general health score, level of education, time since the last menstrual period, height loss (>3 cm) and weight loss (≥5 kg) in the previous year were also studied.

Data analysis

Descriptive statistics were used to describe characteristics for Australian GLOW women with and without being included in this study and for study subjects with and without BMD testing. The chi-square test or ANOVA test was used to determine difference in characteristics where appropriate. For the purposes of the regression analyses, the ratings for both risk perception of osteoporosis and fracture were reclassified by combining together the lower ratings on the scale (i.e. ‘about the same risk’, ‘a bit lower’ and ‘much lower’). In doing so, the regression models for perception of risks were run using three groups (i.e. ‘much higher’, ‘a little higher’ and ‘low risk’). As the proportional odds assumption did not hold for the data, multinomial logistic regression was used to determine odds ratios (ORs) and 95 % confidence intervals (CIs) for the effects of outcome predictors. In the multivariate regression analyses, backward elimination approach was used to determine independent risk factors. We progressively eliminated variables that were least significant, retaining only those with P < 0.05. Those factors not selected were added back into the model one at a time to assess whether they were confounders of the effect of BMD results (i.e. changed its effect by >10 %).

Results

Table 1 shows the baseline and indexed characteristics of the Australian GLOW participants. Those included in the current analysis were significantly younger than those excluded (66 vs. 67 years; P < 0.05). Those excluded from the analysis were significantly more likely to have been told that they had osteoporosis (23 % vs. 17 %; P < 0.05) and were more likely to have a poorer general health compared to those included (χ2 = 21.3; df = 4; P < 0.05). Those included in the analysis were significantly more likely to have lost more than 3 cm in height from age 25 years compared to those not included in the study (χ2 = 184.7; df = 1; P < 0.001). A significant difference was also noted in the years since the last menstrual period between both groups (χ2 = 29.2; df = 3; P < 0.05). Those included in the analysis were also significantly more likely to have a higher level of education (χ2 = 16.3; df = 6; P < 0.05), with 79 % completing at least a higher school certificate compared to 77 % in those excluded from the analysis.
Table 1

Characteristics of the Australian GLOW participants at baseline and the 1,082 study subjects at baseline or index

 

Women excludeda

Women included in the current study

  
   

Total

Women without BMD

Women with BMD

P valueb

P valuec

No.

Baseline (n = 1,822)

No.

(n = 1,082)

Baseline (n = 365)

Index (n = 717)

Age (years), mean (standard deviation)

1,822

67 (9.1)

1,082

66 (8.3)

65 (8.7)

68 (8.0)

<0.05

<0.001

Years since the last menstrual period, no. (%)

1,804

 

1,072

   

<0.001

<0.001

 <10 years ago

 

421 (23)

 

308 (29)

133 (37)

275 (25)

  

 10–19 years ago

 

613 (34)

 

350 (33)

121 (33)

229 (32)

  

 20–29 years ago

 

376 (21)

 

257 (24)

60 (17)

197 (28)

  

 30 or more years ago

 

394 (22)

 

157 (15)

49 (14)

108 (15)

  

Body mass index (kg/m2), no. (%)

1,635

 

1,010

   

0.13

<0.001

 <18.5

 

39 (2)

 

23 (2)

3 (1)

20 (3)

  

 18.5–24.9

 

753 (46)

 

512 (51)

143 (42)

369 (55)

  

 25.0–29.9

 

533 (33)

 

308 (30)

118 (34)

190 (29)

  

 ≥30

 

310 (19)

 

167 (17)

79 (23)

88 (13)

  

Prior year weight loss (≥5 kg)

1,794

125 (7)

1,077

65 (6)

24 (7)

46 (7)

0.90

0.90

Current smoker, no. (%)

1,808

92 (5)

1,079

46 (4)

25 (7)

16 (2)

0.12

<0.001

Height loss (>3 cm)

1,478

6 (0.4)

907

120 (13)

24 (8)

96 (16)

<0.001

<0.05

Drinking (≥7 drinks/week), no. (%)

1,811

618 (34)

1,079

372 (35)

135 (37)

230 (32)

0.90

0.12

GH score

1,807

 

1,079

   

<0.001

<0.05

 Excellent

 

217 (12)

 

153 (14)

71 (20)

90 (13)

  

 Very good

 

672 (37)

 

452 (42)

142 (39)

315 (44)

  

 Good

 

633 (35)

 

346 (32)

106 (29)

233 (33)

  

 Fair

 

244 (14)

 

116 (11)

38 (10)

71 (10)

  

 Poor

 

41 (2)

 

14 (1)

7 (2)

7 (1)

  

Maternal osteoporosis, no. (%)

1,254

350 (28)

744

214 (29)

47 (17)

167 (36)

0.68

<0.001

Maternal hip fracture, no. (%)

1,709

241 (14)

1,037

150 (15)

45 (13)

105 (15)

0.82

0.35

Prior fracture after 45 years, no. (%)

1,773

465 (26)

1,064

248 (23)

47 (13)

201 (28)

0.09

<0.001

Self-reported ‘ever diagnosed with’, no. (%)

        

 Asthma

1,810

296 (16)

1,076

176 (16)

52 (14)

129 (19)

0.76

0.12

 Chronic bronchitis or emphysema

1,786

110 (6)

1,064

57 (5)

18 (5)

47 (7)

1.00

0.34

 High cholesterol

1,787

953 (53)

1,069

559 (52)

187 (52)

372 (53)

0.62

0.85

 Hypertension

1,792

863 (48)

1,074

491 (46)

175 (48)

316 (44)

0.22

2.19

 Osteoporosis

1,769

408 (23)

1,049

174 (17)

11 (3)

182 (27)

<0.01

<0.001

 Osteoarthritis/degenerative joint disease

1,788

736 (41)

1,072

402 (38)

105 (29)

330 (46)

0.75

<0.001

 Rheumatoid arthritis

1,772

142 (8)

1,068

73 (2)

24 (7)

56 (8)

0.67

0.54

Education level, no. (%)

1,787

 

1,072

   

<0.05

0.55

 School certificate (year 11 or less)

 

593 (33)

 

335 (31)

120 (33)

215 (30)

  

 Higher school certificate

 

297 (17)

 

133 (12)

48 (13)

85 (12)

  

 Trade certificate I, II, III or IV

 

98 (6)

 

59 (6)

23 (6)

36 (5)

  

 Diploma or advanced diploma

 

289 (16)

 

182 (17)

50 (14)

132 (19)

  

 Bachelor degree

 

189 (11)

 

129 (12)

43 (12)

86 (12)

  

 Graduate certificate/graduate degree

 

195 (11)

 

131 (12)

43 (12)

88 (12)

  

 Higher degree (master's or doctorate)

 

128 (7)

 

103 (10)

34 (9)

96 (10)

  

GH general health

aBMD status could not be confirmed

bComparison between the Australian GLOW women who were excluded and those who were included in this study

cComparison between study women with and without BMD testing

No significant difference in mean BMI was found between the two groups (χ2 = 5.8; df = 3; P > 0.05). Likewise no significant differences were found in the presence of common osteoporosis risk factors (such as smoking and drinking status, maternal osteoporosis and maternal fracture and prior fracture after age 45). There were also no significant differences found in the presence of pre-existing conditions (such as asthma, emphysema, high cholesterol, hypertension, osteoarthritis and rheumatoid arthritis) between those included and those excluded from the analysis (Table 1).

When comparing those with a BMD to those without a BMD, significant differences were seen in age, BMI and general health (Table 1). Those with a BMD were older (68 vs. 65 years; P < 0.001), were more likely to have a healthy weight range (χ2 = 28.7; df = 3; P < 0.001) and were more likely to report a more positive general health (χ2 = 11.8; df = 4; P < 0.05) compared to those without a BMD. A significant difference was also noted in the years since the last menstrual period between both groups (χ2 = 25.2; df = 3; P < 0.001). Significant differences were also found in the presence of common osteoporosis risk factors (such as height loss, smoking status, maternal osteoporosis and prior fracture after age 45) (Table 1). Those with a BMD were more likely to have been told that they had osteoporosis (27 % vs. 3 %; P < 0.001) and osteoarthritis compared to those without a BMD (46 % vs. 29 %; P < 0.001).

No significant differences were seen in the presence of pre-existing diseases (including asthma, emphysema, high cholesterol, hypertension and rheumatoid arthritis) between the two groups (Table 1). No significant differences were found in the presence of other common osteoporosis risk factors (such as drinking and maternal hip fracture), prior weight loss of more than 5 kg in the preceding year and in educational levels (χ2 = 5.0; df = 6; P = 0.55) between the two groups.

Factors associated with concern for osteoporosis

Table 2 shows factors associated with concern for osteoporosis. BMD testing, prior fracture after age 45, younger age and lower self-reported general health were significantly associated with being ‘very’ or ‘somewhat concerned’ about osteoporosis. Other factors associated with concern about osteoporosis were the presence of comorbidities for those ‘very concerned’ (OR 1.75; 95 % CI 1.11, 2.78), having ≥2 falls in the preceding year for those ‘somewhat concerned’ (OR 2.31; 95 % CI 1.24, 4.28) and maternal osteoporosis for both responses. A poorer BMD result was also associated with a higher level of concern for osteoporosis.
Table 2

Factors associated with concern for osteoporosis

 

Total no.

Very concerned

Somewhat concerned

No. (%)

Adjusted ORa [95 % CI]

No. (%)

Adjusted ORa [95 % CI]

BMD

     

 Osteoporosis

102

33 (32)

12.6 [5.23, 30.6]

58 (57)

3.45 [1.64, 7.27]

 Osteopenia

386

101 (26)

9.80 [5.46, 17.6]

244 (63)

3.65 [2.38, 5.62]

 Normal BMD

223

42 (19)

2.21 [1.20, 4.04]

119 (53)

1.02 [0.68, 1.54]

 Non-BMD testing

365

31 (9)

1.00

208 (57)

1.00

Age (years)

     

 55–<65

544

101 (19)

4.62 [2.47, 8.65]

332 (61)

3.34 [2.08, 5.35]

 65–<75

331

72 (22)

3.09 [1.66, 5.76]

189 (57)

2.25 [1.39, 3.63]

 ≥75

201

34 (17)

1.00

108 (54)

1.00

Mother with osteoporosis

212

68 (32)

7.95 [3.81, 16.6]

134 (63)

5.31 [2.69, 10.5]

Prior fracture after 45 years of age

246

63 (26)

2.24 [1.30, 3.86]

151 (64)

1.87 [1.18, 2.97]

GH score (fair or poor)

121

38 (31)

2.11 [1.64, 2.73]

69 (57)

1.64 [1.34, 2.00]

Comorbidityb

628

148 (24)

1.75 [1.11, 2.78]

365 (58)

1.09 [0.77, 1.54]

Number of falls in the last year

     

 ≥2

673

26 (22)

1.82 [0.86, 3.84]

77 (65)

2.31 [1.24, 4.28]

 1

278

45 (16)

0.91 [0.55, 1.51]

174 (63)

1.35 [0.92, 1.98]

 0

119

135 (20)

1.00

375 (56)

1.00

aAdjusted for all other variables in the column by multinomial logistic regression

bDefined as the presence of any of the following: asthma, emphysema, ulcerative colitis, Parkinson's disease, cancer, diabetes, osteoarthritis, rheumatoid arthritis

Factors associated with perception of risks of osteoporosis and osteoporotic fractures

Tables 3 and 4 show factors associated with perception of risks of osteoporosis and osteoporotic fractures. BMD testing, younger age and lower self-reported general health were significantly associated with having a ‘much higher’ or ‘little higher’ risk perception of getting osteoporosis and fractures. A poorer BMD result was associated with higher risk perceptions. Of those who were osteoporotic or osteopenic, 47 % had a ‘much higher’ or ‘little higher’ perceived risk of osteoporosis and 34 % had a ‘much higher’ or ‘little higher’ perceived risk of fracture.
Table 3

Factors associated with perception of osteoporosis risk

 

Total no.

Much higher

A little higher

No. (%)

Adjusted ORa [95 % CI]

No. (%)

Adjusted ORa [95 % CI]

BMD

     

 Osteoporosis

99

19 (19)

42.4 [8.57, 210]

32 (32)

9.48 [4.72, 19.0]

 Osteopenia

384

56 (15)

35.9 [8.23, 157]

120 (31)

7.52 [4.47, 12.6]

 Normal BMD

224

7 (3)

5.50 [1.10, 27.6]

34 (15)

1.69 [0.92, 3.09]

 Non-BMD testing

359

2 (1)

1.00

29 (8)

1.00

Age (years)

     

 55–<65

541

42 (8)

2.80 [1.22, 6.42]

115 (21)

2.33 [1.34, 4.04]

 65–<75

327

26 (8)

1.46 [0.64, 3.34]

67 (21)

1.46 [0.84, 2.53]

 ≥75

198

16 (8)

1.00

33 (17)

1.00

Mother with osteoporosis

210

33 (16)

4.60 [2.46, 8.62]

83 (40)

3.99 [2.62, 6.06]

Prior fracture after 45 years of age

241

41 (17)

2.10 [1.16, 3.78]

62 (26)

1.25 [0.81, 1.91]

GH score (fair or poor)

122

20 (31)

1.80 [1.30, 2.48]

27 (44)

1.45 [1.17, 1.80]

Comorbidityb

619

62 (10)

2.38 [1.26, 4.50]

157 (25)

2.30 [1.54, 3.45]

Prior year weight lost (≥5 kg)

69

9 (13)

3.38 [1.31, 8.69]

21 (30)

2.45 [1.22, 4.85]

Body mass index (<25 kg/m2)

529

52 (10)

0.92 [0.86, 0.99]

119 (22)

0.99 [0.95, 1.02]

aAdjusted for all other variables in the column by multinomial logistic regression

bDefined as the presence of any of the following: asthma, emphysema, ulcerative colitis, Parkinson's disease, cancer, diabetes, osteoarthritis, rheumatoid arthritis

Table 4

Factors associated with perception of fracture risk

 

Total no.

Much higher

A little higher

No. (%)

Adjusted ORa [95 % CI]

No. (%)

Adjusted ORa [95 % CI]

BMD

     

 Osteoporosis

103

16 (16)

97.5 [11.9, 796]

31 (30)

21.9 [9.95, 48.0]

 Osteopenia

387

31 (8)

35.8 [4.71, 272]

89 (23)

11.4 [5.88, 22.3]

 Normal BMD

224

4 (2)

8.04 [0.87, 74.0]

18 (8)

3.15 [1.45, 6.86]

 Non-BMD testing

361

1 (0.3)

1.00

11 (3)

1.00

Age (years)

     

 55–<65

540

25 (5)

5.74 [2.23, 14.8]

79 (15)

2.56 [1.49, 4.39]

 65–<75

333

18 (5)

3.17 [1.24, 8.09]

40 (12)

1.24 [0.71, 2.16]

 ≥75

202

9 (5)

1.00

30 (15)

1.00

Prior fracture after 45 years of age

246

37 (15)

8.56 [4.30, 17.0]

53 (22)

1.90 [1.25, 2.90]

GH score (fair or poor)

122

16 (27)

2.11 [1.47, 3.03]

26 (37)

1.45 [1.16, 1.80]

aAdjusted for all other variables in the column by multinomial logistic regression

Factors associated with perception of osteoporosis risk also included maternal osteoporosis, presence of comorbidities and weight loss of ≥5 kg in the preceding year for ‘much higher’ and ‘a little higher’ responses. Low body mass index (OR 0.92; 95 % CI 0.86, 0.99) and prior fracture after age 45 (OR 2.10; 95 % CI 1.16, 3.78) were associated with ‘much higher’ perception of osteoporosis risk (Table 3) Prior fracture after age 45 was also associated with a ‘much higher’ (OR 8.56; 95 % CI 4.30, 17.0) and ‘a little higher’ (OR 1.90; 95 % CI 1.25, 2.90) perception of fracture risk (Table 4)

Current smoking, excessive drinking and height loss (>3 cm) since age 25 years were not associated with concern for osteoporosis nor were they associated with self-perceived risk of osteoporosis or fracture.

Discussion

The results of this study illustrate that women's concern and risk perception of fracture are reasonably well founded in this Australian population. The association between concern and perception of common osteoporosis risk factors such as maternal osteoporosis and prior fracture after age 45 concurs with previous findings [13, 14]. In Chang, first-degree relatives of women suffering from osteoporosis had a higher level of susceptibility to osteoporosis compared to those non-first degree relatives [13]. Those with a higher perception of risk of osteoporosis attributed this risk to family history [14]. Siris et al. also found prior fracture after age 45 years to be associated with subjects' increased perception of risk [11].

A poor BMD result was associated with a higher level of concern for osteoporosis as well as higher risk perceptions. This is consistent with past studies that explored the effect of knowledge of BMD results on concern and perception. Rimes and Shipman found that women with a poor BMD were more anxious about their bone density result and more worried about osteoporosis compared to those with a normal or above average BMD result. Those with a low bone density result also had a higher rating of perceived risk of osteoporosis in the near future compared to those with a normal or high BMD result [15].

One interesting finding that has come out of this current study is that merely having a BMD regardless of the results was also associated with higher concern and risk perception. This is a topic that has not been extensively studied previously. Sedlak et al. found that women who had a DXA screening had a higher perceived susceptibility to osteoporosis that increased over time compared to control women who did not have a DXA scan. An increase in osteoporotic preventive behaviours such as calcium and anti-osteoporosis medication intake also increased in these women [16]. It may be expected that those with a higher concern and perception of risk of disease may only be those who have poor BMD results. It may also be expected that those with a poor general health rating may also have a higher level of concern and perception of risk compared to those with higher general health scores. This is not so with the current sample as the majority of those with a normal BMD in the current study also reported higher general health scores (89 % reporting at least being in ‘good’ general health or above) which raises the question whether those in the current analysis are actually a sample of ‘worried well’. Future studies may need to be conducted to further explore this phenomenon.

The association found between common risk factors and concern and perceived risk may be a consequence of the educational levels of the GLOW participants. The participants in the current study were sampled from differing areas throughout Sydney; however, the majority were sourced from the northern Sydney area which can be classed as possessing a higher socioeconomic background and have a higher level of education. In this current sample, 69 % of the population had attained at least a higher school certificate (Year 12 or equivalent). This proportion is slightly higher than those who had the same level of educational attainment in greater Sydney (65 %) in the 2011 Australian Census [17, 18]. The positive correlation between educational levels and knowledge of and attitudes towards osteoporosis has also been previously explored [5 10 19]. Ali and Bennett reported that post-menopausal American women with less than a high school education had lower mean scores in knowledge of osteoporosis and osteoporotic preventive behaviours compared to those with a high school diploma or higher [19], while Alexandraki et al. found that higher education was associated with greater knowledge of osteoporosis in Greek women [5]. Previous studies have also found an inverse relationship between BMD and educational level. Those with a higher level of education have also been found to have a greater understanding of osteoporosis and associated risk factors due to increased access to health information [10]. Future studies in a more educationally diverse population may need to be conducted in Australia to also allow for tailored osteoporosis awareness programmes.

Other findings in this current study highlight the need for more targeted education in regard to osteoporosis risk factors. Increasing age was not seen as a factor associated with concern and perception despite being a known osteoporotic risk factor [10]. The current findings are similar to those found in a pilot study of 60 women (aged 40–95 years) conducted by Hsieh et al. who reported that these women were more concerned about other diseases such as cancer, CVD and neurological disorders than osteoporosis [9]. Although osteoporosis was seen as a serious disease by 89 % of women, only 29 % perceived to be personally susceptible and consequently engaged in activities preventing osteoporosis. The underestimation of osteoporosis severity was also found by Nayak et al. where older respondents were less likely to consider osteoporosis as serious compared to younger respondents [20]. Phillipov et al. found that older women were less likely to have a high perception of risk of osteoporosis compared to younger participants and that age below 65 years was a factor that significantly increased women's self-perceived risk of osteoporosis [21].

Height loss, excessive drinking and smoking were also not associated with concern and perception, despite being important risk factors for osteoporosis and, in the case of excessive drinking and smoking, risk factors amenable to change. This knowledge gap of risk factors for osteoporosis concurs with previous studies. Ribeiro et al. found that female responders (48 % of which were aged 25–84 years) were more likely to name certain uncontrollable factors such as small body frame and race as major risk factors for osteoporosis [22]; however, no mention was made regarding height loss. Only a small number of women named lifestyle factors such as calcium intake (10 %) and exercise (15 %) as risk factors for osteoporosis. Giangregorio et al. also found a deficiency in knowledge in key osteoporosis risk factors including but not limited to excessive alcohol consumption [23]. In the current analysis, smoking was not associated with concern and perception; however, this may be explained by the small number of participants in the sample (only 4 %) who were current smokers. Also, people who lost height were more likely to have a BMD test compared to those who did not in this study, and this might explain why height loss was not a significant predictor.

Previous osteoporosis education programmes have included information on risk factors for osteoporosis. However, these focused primarily on controllable factors such as calcium and vitamin D levels and/or exercise made evident through specific post-intervention outcome measures exploring changes to these behaviours [2428]. Although these are important risk factors for osteoporosis, other key modifiable risk factors such as current smoking and excessive alcohol intake should also be incorporated and equally emphasised in future education programmes and behaviour changes in these risk factors should also be noted. People who are much older or have lost height with resulting kyphosis may have accepted that osteoporosis is an inevitable result of ageing, and education programmes for these groups may need to place more emphasis on the benefit of disease monitoring and efficacy of osteoporosis drug therapy for reducing future fractures.

Information about uncontrollable osteoporotic risk factors such as age and height loss should also be incorporated in future education programmes for physicians. General practitioners in particular form the frontline to osteoporosis diagnosis and management and also have the responsibility to effectively communicate the impact of these risk factors to their patients. Previous studies have also highlighted the greater need for more tailored education programmes for physicians [29, 30]. A study by Jaglal et al. reported that family physicians had a ‘lack of rationale’ for ordering diagnostic tests for osteoporosis [29]. Nayak et al. also found that in the USA, older adults were less likely to be recommended for screening compared to young adults [30]. The same study found that height loss of >2.54 cm was also not a significant predictor of screening recommendations. Future education programmes for physicians should emphasise the impact that height and increasing age have on osteoporosis risk and highlight the importance of communicating these risks to their patients.

This study has limitations that need to be considered when interpreting the findings. The results presented are those from the Australian cohort of a large population-based multinational longitudinal study of osteoporosis in women and may not be generalisable to women in other countries. This current analysis was further limited to a subset of women who were selected based on their answers as to whether or not they had a BMD in subsequent questionnaires followed by confirmation with acquiring the actual test results directly from the participant or through their general practitioner. Those who answered ‘no’ to having a BMD test in subsequent GLOW questionnaires and where no BMD result was found during the BMD collection phase were allocated to the ‘NO BMD’ group (n = 365). Participants were allocated to the ‘BMD’ group if any BMD result was obtained regardless of whether they indicated in the questionnaire that they had a BMD test done or not (n = 717). Although every effort was made by one of the authors (AWB) to obtain BMD results directly from participants or in failing that, to obtain written consent to access the results directly from their general practitioners, it is possible that not all BMD results were collected. As a result, there is a large proportion of the GLOW sample (n = 1,822) who were omitted from this current analysis as BMD results could not be verified.

In summary, women's concern and risk perception of osteoporosis and osteoporotic fractures in Australia were reasonably well founded. However, increasing age, height loss, current smoking and heavy drinking were not factors associated with concern and perception despite being known osteoporosis risk factors. These factors should be considered in planning for education and awareness raising programmes.

Acknowledgments

We would like to thank the GLOW Advisory Board and the physicians involved in the Global Longitudinal Study of Osteoporosis in Women. The GLOW study is supported by a grant from The Alliance for Better Bone Health (Procter & Gamble Pharmaceuticals and Sanofi-Aventis) to The Center for Outcomes Research, University of Massachusetts Medical School.

Conflicts of interest

None.

Copyright information

© International Osteoporosis Foundation and National Osteoporosis Foundation 2013