Page 1
Singapore Med J 2009; 50 (1) : 20O r i g i n a l A r t i c l e
Department of
Obstetrics and
Gynaecology,
Bolan Medical
College,
8-13/36 Kasi Road,
Quetta,
Balochistan 87300,
Pakistan
Fatima M, MBBS,
FCPS, MCPS
Assistant Professor
Nawaz H, MBBS, MD
Research Assistant
Kassi M, MBBS,
MPhil
Assistant Professor
Rehman R, MBBS,
MD
Research Assistant
Kasi PM, MBBS, MD
Research Assistant
Kassi M
Student Research
Assistant
Afghan AK
Student Research
Assistant
Baloch SN, MBBS,
FCPS
Professor and Head
Correspondence to:
Mr Pashtoon Murtaza
Kasi
Tel: (92) 306 373 7347
Fax: (92) 81 283 3794
Email: pashtoon.kasi@
gmail.com
Determining the risk factors and
prevalence of osteoporosis using
quantitative ultrasonography in
Pakistani adult women
Fatima M, Nawaz H, Kassi M, Rehman R, Kasi P M, Kassi M, Afghan A K, Baloch S N
ABSTRACT
Introduction : Osteoporosis-related bone
fractures are a significant public health problem.
The aim of this study was to determine the
prevalence of osteoporosis among Pakistani
women and identify modifiable risk factors.
Methods: A cross-sectional study was conducted
in an obstetrics /gynaecology setting during
March–April 2007 in Quetta, Pakistan. A
total of 334 women older than 20 years of age
underwent quantitative ultrasonography and
were interviewed to find out the risk factors for
osteoporosis. Bone mineral density was assessed by
the speed of sound using a quantitative ultrasound
device. The sociodemographic characteristics
of normal, osteopenic and osteoporotic women
were compared using the chi-square test for
categorical variables and ANOVA for continuous
variables. Binary logistic regression was used to
determine the independent predictors of being
osteopenic or osteoporotic.
Results: 146 (43.7 percent) women were reported
to be normal, 145 (43.4 percent) were osteopenic
and 43 (12.9 percent) were osteoporotic. The mean
age and standard deviation of the participants
were 36.7 years +/- 13.0 years, with a body mass
index (BMI) of 25.81 (standard deviation 5.10)
kg per square metre. In the univariate analysis,
factors that were associated with osteoporosis/
osteopenia included age, parity, BMI, smoking
(pack years), consumption of calcium-rich food/
week, personal and family history of osteoporosis,
education and socioeconomic status (p-value is
less than 0.05). Using binary logistic regression
with osteoporosis /osteopenia as an outcome
compared to normal individuals, BMI, smoking
pack years, a family history of osteoporosis /
fracture and house ownership were found to be
independent predictors of the outcome.
Conclusion: The prevalence of osteoporosis
and osteopenia is high, especially among
young Pakistani women, and is associated with
modifiable risk factors.
Keywords: osteoporosis prevalence, osteoporosis
risk factors, quantitative ultrasonography,
women’s health
Singapore Med J 2009; 50(1): 20-28
INTRODUCTION
Osteoporosis and related fractures are a major public
health problem and become more important with an ageing
population.(1) 30%–50% of women and 15%–30% of men
suffer from osteoporosis-related fractures in their lifetime.(2)
Fractures are associated with increased morbidity and
mortality, and impose a considerable financial burden on
the community.(3) The total cost of osteoporotic fractures,
mostly hip fractures, has been estimated to be USD 10–20
billion per year in the United States alone.(4) Considering the
overall burden of hip, spinal and other fractures, including
hospitalisation and time off work, the human and financial
costs of osteoporotic fractures are enormous.(5)
In Pakistan, life expectancy at birth has increased from
41 years in 1950 to 61.9 years in 1998, and is expected
to be 72.4 years in 2023.(6) The proportion of elderly and
post-menopausal women is on the rise. In the future,
more Pakistani women will suffer from osteoporosis-
related fractures that lead to a poor quality of life. Studies
indicate a high prevalence of risk factors associated with
osteoporosis in the community.(7,8) According to one survey,
72% of people lead a sedentary lifestyle, and vitamin D
deficiency among Pakistani women has been reported to
be as high as 83%.(7,8) Moreover, the Pakistani diet has
been found to be deficient in calcium.(9) The prevalence
of smoking has been reported to be 22%–40% in most
Page 2
Singapore Med J 2009; 50 (1) : 21
recent population-based studies.(10,11) To our knowledge,
the prevalence of osteoporosis and its risk factors among
Pakistani women are hitherto unknown. It is hypothesised
that a large proportion of young Pakistani women are at
risk of developing osteoporosis in the future. Through
cross-sectional research, this study aimed to determine the
prevalence of osteoporosis in a representative sample of
Pakistani women and uncover its associated risk factors.
MeTHODS
Ethical approval for the study was obtained from the
Department of Obstetrics and Gynaecology, Bolan
Medical College, Quetta, Pakistan as well as the local
ethical committee for research, and the research conducted
was performed according to the Declaration of Helsinki.
The study was carried out during March–April 2007 in
the city of Quetta in Balochistan, the largest province in
Pakistan. Quetta is a metropolitan city and the capital of
the province. People belonging to different castes live here
along with many refugees who were from the adjacent war-
torn country of Afghanistan and migrated during the early
1980s and 1990s. Patients were enrolled into the study from
the outpatient department of the Department of Obstetrics
and Gynaecology, Bolan Medical College. This is one of the
major teaching/tertiary care centres in the province.
In March 2007, all female patients, who were older
than 20 years of age and presented to the clinic with a
variety of obstetrical and gynaecological complaints, along
with their female attendants above the age of 20 years, were
informed about the purpose of the study, its risks as well as
the long-term benefits for the subjects and the community
as a whole. Posters inviting participants were also put up
in the department and in various other leading healthcare
centres in the city. Participants were ensured of complete
confidentiality and informed about the dissemination
of the research findings. Formal consent was obtained
from women who agreed to participate in the study. All
participants were requested to visit Fatima Clinic, Quetta,
where they were interviewed and where they underwent
Variable No. (%)
Bone mineral density
Normal 146 (43.7)
Osteopenic 145 (43.4)
Osteoporotic 43 (12.9)
Total 334 (100)
Mother tongue
Pashto 221 (66.17)
Urdu 45 (13.47)
Balochi/Brahvi 25 (7.49)
Persian 23 (6.89)
Punjabi 17 (5.09)
Other 3 (0.9)
City of permanent residence
Quetta 311 (93.11)
Other 23 (6.89)
Marital status
Married 269 (80.5)
Single 45 (13.5)
Widowed 16 (4.8)
Other 4 (1.2)
Occupation
Housewife 298 (89.22)
Working woman 36 (10.78)
Literate
Yes 103 (30.84)
No 231 (69.16)
House ownership
Yes 228 (68.26)
No 106 (31.74)
Car ownership
Yes 63 (18.86)
No 271 (81.14)
Monthly income (rupees)
< 5,000 139 (41.6)
5,000–10,000 115 (34.4)
10,000–20,000 27 (8.1)
> 20,000 14 (4.2)
Not available 39 (11.7)
Familiar with osteoporosis
Yes 39 (11.7)
No 295 (88.3)
Ever smoked?
Yes 66 (19.76)
No 268 (80.24)
Present smoking status
Daily 48 (14.37)
Occasionally 14 (4.19)
Not smoking at present/never smoked 272 (81.44)
Atraumatic fracture before 25 years of age
Yes 16 (4.79)
No 318 (95.21)
Family history of osteoporosis
Yes 134 (40.12)
No 200 (59.88)
Presence of any comorbidity
Yes 54 (16.17)
No 280 (83.83)
Steroid use
Yes 16 (4.79)
No 318 (95.21)
Use of homoeopathic medicines
Yes 68 (20.36)
No 266 (79.64)
Use of calcium supplements
All/most of the time 12 (3.59)
Sometimes 83 (24.85)
Little/never 239 (71.56)
Type of clothing
Whole body covered 238 (71.26)
Hands/face visible 82 (24.55)
Head/arms visible 14 (4.19)
Table I. Basic sociodemographic characteristics and
other variables.
Exercise/walk
Yes 23 (6.89)
No 311 (93.11)
Amenorrhoea > 6 months
Yes 44 (13.17)
No 290 (86.83)
Postmenopausal
Yes 85 (25.45)
No 249 (74.55)
Premature menopause (< 45 years)
Yes 34 (10.2)
No 51 (15.3)
Premenopausal 249 (74.6)
Page 3
Singapore Med J 2009; 50 (1) : 22
quantitative ultrasonography to determine their bone
mineral density (BMD). Women older than 20 years of age
were included in our study to analyse preventive behaviour
for osteoporosis among young women in Pakistan.
The questionnaire was designed in English and
then translated into both Urdu (the national language of
Pakistan) and Pashto (the local language). An independent
back translation was done, compared with the original
questionnaire and the discrepancies were corrected. An
initial pilot questionnaire with 30 items was tested on a
representative sample to check if the interviewers or patients
faced any difficulties with any part of it. The final version
was prepared according to the feedback received from the
pilot study. The research team comprised six doctors and four
research assistants. The research assistants selected were
females who were trained to administer the questionnaire
in both Urdu and Pashto. The interview also included
an assessment of the subjects’ basic sociodemographic
characteristics, medical, smoking and menstrual history.
Their history of low-trauma fracture was obtained. Low-
trauma fracture was defined as a fracture occurring from a
trivial/minor injury. Interviewers first determined if patients
had any first-degree relative who had suffered a fracture.
They then enquired as to how it had occurred. It was only
included if it occurred from low-trauma injuries. Fractures
such as those sustained as result of a car accident or fall
from a major height were excluded. The family history was
elucidated by asking for the occurrence of osteoporosis
or low-trauma fracture among the patients’ first-degree
female relatives. Steroid use was also noted. The use of
homoeopathic medicines was also recorded because it has
been reported that many homoeopathic medicines that are
sold in Pakistan contain steroids. Surrogate markers of
socioeconomic status reported in the National Survey of
Pakistan (reference) such as “house ownership” and “car
ownership” were also included.(12)
Height and weight of all the subjects were recorded
while they were wearing light clothes and no shoes. Care
was taken to ensure that the upper border of the external
auditory meatus was in line with the lower border of the
orbit for the height measurement.(13) Body mass index
(BMI) was calculated as the weight divided by the square
of height (kg/m2).
BMD was assessed by the speed of sound (m/sec) using
a quantitative ultrasound device manufactured by Sahara
Clinical Bone Sonometer (Hologic Inc, Bedford, MA,
USA). The sonometer measures the broadband ultrasound
attenuation (BUA, in dB/MHz) and speed of sound (SOS,
in m/sec) of an ultrasound beam that is passed through the
heel. The BUA and SOS are combined to yield an index
(quantitative ultrasound or “stiffness” index [QUI]), which
is then used to estimate the calcaneal BMD (in g/cm2).(14)
The latter is inferred from a linear combination of BUA
and SOS, and is not an actual measurement of calcaneal
BMD.(15) This device is small and portable, with a gel
coupled (dry) system that can measure the SOS on the
calcaneus. A phantom supplied by the manufacturer was
used to calibrate the machine before each screening session
and then standardised to the Asian women before usage.(16)
For all subjects, the measurements were done on the right
calcaneus. If the subject had a history of fracture or any
bone disorder of the right foot, the left heel was evaluated.
The measurement was taken in a temperature-controlled
environment (20°C), and was performed only by a trained
technician.
Several techniques to measure BMD have been
introduced in the last two decades.(17) Although dual
energy X-ray absorptiometry (DXA) is considered the gold
standard for the assessment of BMD, the last decade has
seen the advent of quantitative ultrasonography (QUS) for
the indirect assessment of bone quality. It is especially suited
to developing countries and for screening purposes; and
since there is no ionising exposure, the devices are portable
and the costs are considerably lower.(18) The combination
of ultrasonography with risk factor inquiry identified 90%
of women with osteoporosis.(19) T-score measurement
was used to determine the BMD level and presence and
risk of osteoporosis. A T-score is the standard deviation of
a patient’s BMD compared to a healthy young reference
population.(16) According to the World Health Organisation
criteria, individuals with a T-score of less than −2.5 are
diagnosed as osteoporotic, those with a T-score of between
−2.5 and −1 as having a low bone density and at risk of
osteoporosis (osteopenia) and a T-score of greater than −1.0
as normal.(15,16,20) Although the gold standard for evaluating
BMD is via a DXA scan, measurements were done using
Variable Mean ± SD Range
Age (years) 36.7 ± 13.0 20–60
No. of children 5.24 ± 3.11 0–15
Education (years) 2.87 ± 4.66 0–18
Education of spouse (years) 3.90 ± 5.80 0–18
Weight (kg) 62.35 ± 12.79 33–98
Height (cm) 155.40 ± 6.40 139–175
BMI (kg/m2) 25.81 ± 5.10 16.6–39.1
Smoking (pack years) 1.59 ± 4.51 0–30
Intake of calcium-rich food 4.10 ± 2.65 0–7
(times/week)
Exercise/walk (mins/week) 5.39 ± 31.99 0–420
Table II. Basic sociodemographic characteristics
(continuous variables).
Page 4
Singapore Med J 2009; 50 (1) : 23
QUS due to portability and affordability.
Data was entered into Microsoft Access 2000 and
analysed using the Statistical Package for Social Sciences
version 14.0 (SPSS Inc, Chicago, IL, USA). Participants
were divided into three groups: normal, osteopenic and
osteoporotic. The sociodemographic characteristics of
the groups were compared using the chi-square test for
categorical variables and ANOVA for continuous variables.
The primary analysis used binary logistic regression to
determine independent predictors of being osteopenic/
Variable Normal Osteopenic Osteoporotic p-value
Mean age (years) 34.7 33.7 54.0 < 0.001***
Mean parity 4.72 4.92 7.70 < 0.001***
Mean BMI (kg/m2) 26.9 25.7 22.6 < 0.001***
Mother tongue
Pashtoon 89 100 32 0.167
Others 57 45 11
Occupation
Housewife 129 128 41 0.382
Working woman 17 17 2
Familiar with osteoporosis
Yes 17 19 3 0.547
No 151 107 37
House ownership
Yes 112 94 22 0.003**
No 34 51 21
Car ownership
Yes 31 28 4 0.210
No 115 117 39
Have smoked
Yes 19 27 20 < 0.001***
No 127 118 23
Smoking (pack years) 0.53 1.47 5.36 < 0.001***
Calcium-rich food/week 4.78 3.94 2.31 < 0.001***
Postmenopausal
Yes 23 28 34 < 0.001***
No 123 117 9
Amenorrhoea > 6 months
Yes 14 19 11 0.024*
No 132 126 32
Premature menopause
Yes 12 10 12 0.377
No 11 8 22
Literate
Yes 54 45 4 0.002**
No 92 100 39
Education (years) 3.42 3.01 0.58 0.002**
Education of husband (years) 4.48 4.19 1.11 0.005**
Family history
Yes 46 63 25 0.004**
No 100 82 18
Atraumatic fracture
Yes 1 5 10 < 0.001***
No 145 140 33
Steroid use
Yes 2 11 3 0.035*
No 144 134 40
Use of homoeopathic medicine
Yes 33 21 14 0.024*
No 113 124 29
Calcium supplements
All/most of the time 5 6 1 0.479
Sometimes 42 34 7
Never 99 105 35
Clothing
Whole body covered 100 103 35 0.511
Hands/face visible 38 37 7
Head/arms visible 8 5 1
Exercise/walk 0.175
Yes 14 8 1
No 132 137 42
Exercise (mins/week) 10 1.39 3.26 0.064
Comorbidity
Yes 18 24 12 0.051
No 128 121 31
* statistically significant at < 0.05; ** statistically significant at < 0.01; *** statistically significant at < 0.001.
Table III. Comparison of risk factors among normal, osteopenic and osteoporotic individuals.
Page 5
Singapore Med J 2009; 50 (1) : 24
osteoporotic. Variables reaching statistical significance (p
< 0.05) were included in this final model.
ReSUlTS
The basic sociodemographic characteristics of the
individuals are shown in Tables I and II. A total of 334
individuals consented to be screened and participated in
the interviews. Only 39 (11.7%) participants were familiar
with the term/disease “osteoporosis”. The mean age and
standard deviation of the participants was 36.7 ± 13.0
years. 221 (66.2%) participants spoke the native language
Pashto, followed by 45 (13.5%) who spoke Urdu, 25
(7.5%) Balochi/Brahvi, 23 (6.9%) Persian and 17 (5.1%)
Punjabi. Almost all (n = 311, 93.1%) of the participants
were residents of Quetta. 269 (80.5%) were married, 45
(13.5%) were single and 16 (4.8%) were widowed. Most of
the women were housewives (n = 298, 89.2%). The literacy
rate, defined as whether or not an individual could read,
was low (n = 103, 30.8%). 228 (68.3%) individuals owned
a house, while 63 (18.9%) owned a car. 139 (41.6%) had a
monthly income of less than Rs 5,000 (USD 84), and 115
(34.4%) had a monthly income of between Rs 5,000 (USD
84) and Rs 10,000 (USD 167).
Based on the World Health Organisation’s criteria
for BMD, 146 (43.7%) participants were reported as
normal, 145 (43.4%) as osteopenic and 43 (12.9%) as
osteoporotic. The mean T-scores were −0.29, −1.68 and
−2.95, respectively. The QUI /stiffness scores were 109.6,
85.4 and 63.5, respectively. Similarly, the estimated BMD
values for normal, osteopenic and osteoporotic individuals
were 0.61, 0.46 and 0.33 g/cm2, respectively.
The comparison of various risk factors among normal,
osteopenic and osteoporotic individuals is shown in Table
III. In the univariate analysis, factors reaching statistical
significance (p < 0.05) included age, parity, BMI, smoking
(pack years), calcium-rich food/week, years of schooling,
years of schooling of husband, menopausal status, history
of low-trauma fracture, amenorrhoea > 6 months, house
ownership, having a first-degree relative with osteoporosis/
history of low-trauma fracture, a history of steroid use and a
history of homoeopathic medicine use. Using binary logistic
regression with osteoporosis/osteopenia as the outcome
compared to normal individuals, BMI (p = 0.014), smoking
pack years (p = 0.036), family history of osteoporosis/
fracture (p = 0.040) and house ownership (p = 0.029) were
found to be independent predictors of outcome (Table IV).
DISCUSSION
The prevalence of osteoporosis and osteopenia was found
to be high in this study. 43 (12.9%) women with a mean
age of 54 years were osteoporotic and 145 (43.4%) were
osteopenic. Assuming that the status of risk factors for
these women does not change, a large proportion of young
Pakistani women will develop osteoporosis-related fractures
in the future. This will be a huge burden on the health
resources of this country. A number of important risk factors
for osteoporosis and osteopenia were identified, some of
which are modifiable. Firstly, smoking was identified as an
important risk factor with a prevalence as high as 19.8%.
This is alarming as smoking among women is considered
to be a taboo in Pakistani society and previous studies have
reported much lower rates.(21,22) This report along with the
most recent ones suggest an increase in the prevalence of
women who smoke, and this is a serious risk factor not
only for osteoporosis but also for lung cancer, coronary
and cerebrovascular diseases.(23) Women may also be using
alternative forms of tobacco, such as “Huqqa” (hubble
bubble). One study reported a prevalence of tobacco use of
52% in a low socioeconomic status urban community.(21)
A smoking history of 5.4 pack years was significantly
associated with osteoporosis, while a smoking history of
1.5 years was significantly associated with osteopenia (p <
0.001). The high prevalence of smoking among Pakistani
women signifies the acceptance of a hitherto “shameful act”
and highlights the changing attitudes of women towards
smoking. This also suggests that the smoking cessation
programme has failed to deliver results in Pakistan. The
entire focus and media attention has been on male smoking,
while female smoking was considered to be less of a problem
and thus, neglected completely. It has been observed that
Pakistani women smoke cigarettes as well as smokeless
tobacco. An association between smokeless tobacco and
osteoporosis, however, has not yet been established.
Secondly, this study found that women with
osteoporosis and osteopenia had a lower BMI compared
to normal individuals. However, the idea of gaining
Table IV. Predictors of osteoporosis/osteopenia in Pakistani women.
Variable Odds ratio 95% confidence interval p-value
BMI (kg/m2) 0.92 0.87–0.98 0.014
House ownership 0.47 0.24–0.92 0.029
Smoking (pack years) 1.12 1.01–1.24 0.036
Family history of osteoporosis/fracture 1.91 1.03–13.53 0.040
End of preview.
