Jan-09.pdf
Abedin, Khan, Taz Uddin and Rahman
Health Hazards Faced by Floating Commercial Sex Workers…
Health Hazards Faced by Floating
Commercial Sex Workers
A Study in Rajshahi City, Bangladesh
SUMAIYA ABEDIN, MOHAMMAD ATIKUR RAHMAN KHAN, TAZ UDDIN AND
J.A.M. SHOPQUILUR RAHMAN

The clandestine nature of floating commercial sex workers (CSWs) in Bangladesh has
made healthcare difficult, especially since a large number of CSWs work outside
brothels — non-brothel commercial sex is considered illegal in Bangladesh. The aim of
this study is to investigate the stark reality of commercial sex work of floating sex
workers, their socioeconomic levels and the hazardous health condition they face.
Information on socioeconomic, demographic and health-related data was collected from
176 floating CSWs of Rajshahi city in Bangladesh. The study revealed that most CSWs
resorted to commercial sex for their daily survival and that they suffered from various
physical problems — vaginal pain (64.8%), blood discharge (4.5%), vaginal pus
discharge (61.4%) itching and stitch problem (43.7%). To identify a set of influential
factors that cause such problems, we fit logistic regression and linear probability models
by forward selection method. The number of years in commercial sex, number of
children born before resorting to commercial sex work, body fat and pulse pressure
were found to be significant factors responsible for most health problems.
Sumaiya Abedin and Md. Atikur Rahman Khan are both Assistant Professors; Taz
Uddin is a Research Fellow; and J.A.M. Shopquilur Rahman is Associate Professor at
the Department of Population Sciences, University of Rajshahi, Bangladesh.

INTRODUCTION
Com mer cial sex work ers (CSWs) are most vul ner a ble to sex u ally trans mit -
ted dis eases (STDs) and HIV due to un safe sex and a high-risk life style.
Hong and Li (2008), who con ducted be hav ioural stud ies in China, ob served
that the CSWs were young, mo
bile, had both com
mer cial and
non-commercial sex part ners, a low rate of con sis tent con dom use and high
rate of STD in fec tion. Other stud ies have emphasised the risks as so ci ated
with brothel-based sex, low lev els of in for ma tion and knowl edge about safe
sex and of their own sex ual health as well (Huang, Henderson, Pan and
Cohen , 2004). An other study con ducted by Yang and Xia (2006) on Chi nese
en ter tain ment work ers ex plored the high-risk sex ual be hav iour and lack of
knowl edge for a healthy sex ual part ner ship. In fact, most re search was car -
ried on brothel or ho tel-based sex work ers. Apart from this, Wong, Holroyd,

104
Abedin, Khan, Taz Uddin and Rahman
Gray and Ling (2006) stud ied the health com pli ca tions and un healthy sex
trade of street sex work ers in Hong Kong. This study has been in spired by the
work of Wong and others (2006), and fo cus ses on the health haz ards of float -
ing CSWs in Rajshahi City, Ban gla desh.
In Ban gla desh, with the ex cep tion of 15 broth els, sex trade is il le gal.
While the exact num ber of CSWs is not known, some es ti mate it at 50,000
to 100,000 (Nessa, Waris, Sul tan, Monira, Hossain, Nahar, Rahman, Alam,
Baatsen and Rahman, 2004; Nessa, Waris, Alam, Huq, Nahar, Faisal,
Hasan, Monira, Munir, Sul tana, Kazi, Das, Dipak, Sul tan, Hossain and
Rahman , 2005). The CSWs work from broth els, ho tels, street and res i -
dences. How ever, in re cent years, there have been re mark able changes in
the sex trade and pri mary health care of CSWs with better ac cess to re ha bil -
i ta tion cen tres where they can avail of some basic healthcare. Re ha bil i ta -
tion cen
tres that work closely with the As so ci a tion for Com mu nity
De vel op ment (ACD) es ti mate that 900–1500 float ing CSWs are pres ent in
Rajshahi city. In the past, some non-government or gani sa tions (NGOs)
stud ied the lit er acy lev els and cli ent groups of CSWs. It was ob served that
most CSWs enter the pro fes sion due to pov erty, mar i tal break-up or are co -
erced into it (Nessa and oth ers, 2005). Most are un aware of the haz ards of
high risk sex ual be hav iour to their health. As a re sult, they suf fer from a
num ber of sex-related health com pli ca tions like diz zi ness, nau sea, heart
dis ease, chest pain, joint aches, back pain, waist pain, pres sure, hy per ten -
sion, and so on.
While blood and pulse pres sure, body mass index (BMI) and body fat
per cent age sig nif i cantly af fect health (Millar-Craig, Hawes and Whit ting -
ton, 1978), other fac tors like obe sity is as so ci ated with many dis ease and
health com pli ca tions (Must, Spadane, Coakley, Field, Colditz and Spitz ,
1999). This study at tempts to ana lyse these health haz ards which may be
as so ci ated with many fac tors re lated to high-risk sex ual be hav iour. Since
no such re search has so far been con ducted on float ing CSWs in Ban gla -
desh, the aim of this paper is to iden tify the core fac tors that sig nif i cantly
lead to com mon health prob lems.
METHOD
Ban gla desh has few healthcare op tions for float ing CSWs; be sides sex
trade is largely il le gal and hence clan des tine — this made ac cess for in ter -
views and data col lec tion chal leng ing. Rajshahi is the third larg est city in
Ban gla desh with nearly 388,000 in
hab
it
ants (Ban gla desh: Min is try of
Planning, 2006) and has an es ti mated pop u la tion of 900–1500 float ing

Health Hazards Faced by Floating Commercial Sex Workers… 105
CSWs. How ever, the sam ple of 176 float ing CSWs was ob tained from an
ear lier sur vey con ducted in the months of June and July 2007 by the De -
part ment of Pop u la tion Sci ence and Human Re source De vel op ment, Uni -
ver sity of Rajshahi. All nec es sary in for ma tion were gath ered by the di rect
in ter view method and con ducted with the ready co op er a tion of of fi cers at
the drop-in-centre of the NGO-run, Cen ter for Com mu nity De vel op ment,
where float ing CSWs vis ited to avail of basic healthcare sup port. Each
CSW was in ter viewed for 90 min utes and in for ma tion on their so cio eco -
nomic and health sta tus gath ered. While 200 CSWs were in ter viewed,
com plete in for ma tion was avail able for only 176 CSWs. Thus, the num ber
of sex work ers under study was 176. To col lect data for some health-related
in di ca tors, the fat scale (or weigh ing ma chine) and blood and pulse pres -
sure me ters were used.
To in ves ti gate the bi o log i cal fac tors, lo gis tic re gres sion and lin ear prob -
a bil ity mod els were fit ted along with de mo graphic and health-related in di -
ca tors. Using the for ward se lec tion tech nique of vari able se lec tion, a set of
sig nif i cant fac tors re spon si ble for such health haz ards was se lected. Com -
paring the re sults from both fit ted mod els, some con sis tent com mon fac tors
were iden ti fied.
To fit the model, all vari ables are con sid ered di chot o mous. The 20 in de -
pend ent vari ables are: length of ser vice in years (PY), total num ber of chil -
dren (TC), num ber of chil dren be fore en ter ing the pro fes sion (BFC), use of
con dom, num ber of cli ents per night (NCP), num ber of nights per form ing
per week (NPPW), sys tolic blood pres sure low, nor mal and high (SLOW,
SNOR and SHIGH), di
a
stolic low, nor
mal and high blood pres
sure
(DLOW, DNOR and DHIGH), drug use (DU), fat level under body fat,
nor mal, over and obese (BFUN, BFN, BFOV and BFO) and pulse pres sure
low, nor mal and high (PLOW, PNOR, PHIGH). More over, some in di ces
have been com puted and mea sures of some vari ables have been clas si fied
at dif fer ent lev els.
MEASUREMENT OF VARIABLES
Body Mass Index
Body Mass Index (BMI) is cal cu lated by di vid ing a per son’s weight in ki -
lo grams by the square of his/her height in metres. This is an in di ca tor of
body fit ness that in di cates if a per son is un der weight, of nor mal weight,
over weight or obese. Growth spec i fi ca tion through BMI (Must and others ,
1999) can be ex pressed as fol lows:

106
Abedin, Khan, Taz Uddin and Rahman
TABLE 1: Measure of BMI
Category
BMI
Underweight
<18.5
Nor mal weight
18.5–24.9
Overweight
25–29.9
Obe sity 1
30–34.9
Obe sity 2
35–39.5
Ex treme Obe sity
40.0
Body Fat
This is the most im por tant mea sure of over all phys i cal fit ness. The human
body is made up of water, fat, pro tein, car bo hy drates, var i ous vi ta mins and
min er als. Per sons with very high or very low body fat and body water per -
cent age are at higher risk of health prob lems. Fit ness sta tus clas si fi ca tion of
the body fat is ex pressed in Table 2.
TABLE 2: Fitness Status and Body Fat Percentage
Fitness Status
Women
Men
(%)
(%)
Nor mal
15–25
10–20
Over weight
25.1–29.9
20.1–24.4
Obese
Over 30
Over 25
Source: http://www.annecollins.com/weight-loss/ideal-weight-men.htm.
Blood Pressure and Hypertension
A blood pres sure read ing below 120/80mmHg is con sid ered low but good;
90/50mmHg and less is low blood pres sure and a read ing of 140/90 mmHg and
more is high blood pres sure (Millar-Craig, and oth ers, 1978; Urbina, Alpert,
Flynn, Hay man, Harshfield, Ja cob son, Mahoney, McCrindle, Mietus-Snyder,
Steinberger, Daniels and Harshfield, 2008). For young peo ple, the av er age
blood pres sure is 120/80 mmHg and for old peo ple it is 140/90 mmHg. The lev -
els of se ver ity of hy per ten sion are dis played in Table 3.
TABLE 3: Measure of Hypertension
Level of Severity
Systolic Blood Pressure
Diastolic Blood Pressure
Mild Hy per ten sion
140–160
90–100
Mod er ate Hy per ten sion
160–200
100–120
Se vere Hy per ten sion
Above 200
Above 200
Source: http://medindia.net/patients/calculators/bp_chart.asp> May 26, 2006.

Health Hazards Faced by Floating Commercial Sex Workers… 107
STATISTICAL MODELS
A multivariate lo gis tic re gres sion model was used to study the ef fect and
in ter ac tion of the cate gor ised vari ables (Hosmer and Lemeshow, 2000).
An other multivariate model, sim
i
lar to lo gis tic re gres sion is the lin
ear
prob a bil ity model (LPM) where the es ti ma tion tech nique is quite dif fer ent
from lo gis tic trans for ma tion. The gen eral ex pres sion of the LPM is of the
form (Guja rati, 1995) given below:
1.
Y = β X + β X + … + β X + ε
i
1
1i
2
2i
k
ki
i
Here, X k ’s are independent variables and βk is the coefficient of Xk ,
β0 is constants and ε I is the error term. A dependent variable can be
expressed as
1, if the ith sex worker claims vaginal pain problem
Yi =
0, otherwise.
2.
To perform certain analysis like logistic regression and linear probability
model, an informative dichotomous variable is built. For example, four
categories of body fat are used — each of these categories are variables,
that is, for the variable underweight we provide ‘1’ for an underweight
sex worker; and ‘0’ otherwise. One major caveat is that if we include all
four variables of body fat status (underweight, normal, overweight and
obese), then we cannot use the intercept term.
How ever, as soon as any of the vari ables of a cer tain mea sure (body fat
or hy per ten sion) is ex cluded dur ing vari able se lec tion pro cess, we can in -
clude the in ter cept term. An other very im por tant issue is the vari able se lec -
tion pro ce dure. If we use the back ward elim i na tion pro ce dure then the
mo ment ma trix may be sin gu lar. This is due to the in clu sion of all the vari -
ables for body fat, pulse pres sure. So, we are bound to fol low the for ward
se lec tion method in which the most sig nif i cant vari able (cat e gor i cal vari -
able for body fat, BMI or other) is in cluded at the first stage and then in -
clude an other vari able that seems to be the most sig nif i cant among the
re main ing vari ables. Finally, we get a set of sig nif i cant vari ables.
Many vari ables may cause health com pli ca tions; how ever, the strength
of caus ing health com pli ca tions by all vari ables is not the same. Even the
di rec tion or in flu ence of those vari ables may be dif fer ent, based on the type
of the fit ted model, that is, a vari able may be found sig nif i cant while fit ting
a lo gis tic re gres sion but may be in sig nif i cant while fit ting a lin ear prob a bil -
ity model. But if the vari able bears a strong in flu ence on the de pend ent
vari
able, it is most likely to be sig
nif
i
cant in both mod
els (Lo gis tic

108
Abedin, Khan, Taz Uddin and Rahman
Re gres sion or LPM) with the same sign of co ef fi cient, that is, the vari able
is model in vari ant. Ob vi ously, model in vari ant vari ables are stron ger com -
pared to other vari ables. Thus, we are in ter ested in some model in vari ant
vari ables which are as sumed to have a strong in flu ence on the de pend ent
vari able.
Phys i cal health char ac ter is tics pre sented in Table 5 in di cate that 42%
of sex work ers ex pe ri enced low sys tolic blood pres sure, 7.4% ex pe ri-
enced high and the rest were nor mal. Also for di a stolic blood pres sure,
77% were nor mal, 17% low and 4.57% high. In hy per ten sion cat e gory for
sys tolic pres sure, 4.0% ex pe ri enced mild hy per ten sion — how ever, di a -
stolic mild hy per ten sion was ex pe ri enced by 4% and mod er ate hy per ten -
sion was ob tained for 1.1% sex work
ers. Body fat per
cent
age is an
in di ca tor of body fit ness that mea sures the fat level in the body — 50%
had nor mal body fat, 13.1% were un der weight, 17.0% were over weight
and 13.6% obese. Pulse pres sure gives an idea about the reg u lar ity of
blood cir cu la tion in the body and is also an in di ca tor of hy per ten sion. Our
study re veals that 55% had high pulse rate and 29% had low; 54.5% did
not con sult a phy si cian for reg u lar check-ups. It is also ob served that
64.8% suf fered from vag i nal pain, 4.5% from blood and pus dis charge,
43.7% from itch ing and stitch prob lem around the va gina; and 61.4%
from vag i nal pus dis charge.
DISCUSSION
The pop u la tion under study, work clan des tinely as most are il le gal CSWs.
So cial char ac ter is tics of the study pop u la tion are as fol lows: 76% are il lit -
er ate; 50% from urban slums; 97.2% are Mus lims; max i mum fam ily size is
five; 92% have be come CSWs due to pov erty; av er age monthly in come is a
mea gre 2,999 taka (less than US$ 50); av er age num ber of cli ents per night
are four; 58.5% charge 50 taka (less than US$ 1) per cli ent; 90% CSWs
work an av er age of four nights per week and 63.1% use a con dom (see
Table 4).
De mo graphic char ac ter is tics of the study pop
u la tion show that
two-fifths of CSWs are in the age group 25–34 years and av er age num ber
of chil dren per CSW is four; 27.22% have en tered the pro fes sion be tween
age group of 15–19 years (as teen ag ers have a high de mand); and 23.85%
en tered when they were 20–24 years. Three-fourths had at least one child
be fore en ter ing the pro fes sion; 17.1% had one child, 26.1% had two chil -
dren, 20.5% had three chil dren, 8.5% had four chil dren, 26.1% had no child
and 1.1% had five chil dren (see Table 4).

Health Hazards Faced by Floating Commercial Sex Workers… 109
TABLE 4: Demographic and Socioeconomic Characteristics
Characteristics
Percentage
Characteristics
Percentage
(Numbers)
(Numbers)
Age (Years)
Marital Status
15 – 24
26.1 (46)
Un mar ried
1.13 (2)
25 – 34
42.0 (74)
Married
73.30 (129)
35 – 44
28.4 (50)
Sep a ra tion
18.18 (32)
45 – 54
3.4 (6)
Widowed
2.85 (5)
(Mean Age 29 Years)
Di vor ced
4.54 (8)
Re li gion
Fam ily Num ber
Islam
97.2 (171)
1 to 5
82.95 (146)
Hindu
2.8 (5)
5 to 10
17.05 (30)
Place of Birth
Place of Res i dence
Urban (Slum)
51.7 (91)
Urban (Slum)
100 (176)
Rural
48.3 (85)
Rural
0
Ed u ca tion
Condom Use
Il lit er ate
76.1 (134)
No
36.9 (-)
Pri mary
18.2 (32)
Yes
63.1 (111)
Sec ond ary
5.7 (10)
Rea son for Being (CSW)
Clients per Night
Lin eage
4.5 (8)
1 or 2 per sons
8.25 (11)
Eco nomic
92.0 (162)
3 or 4 per sons
84.1 (148)
Com pelled
2.8 (5)
5 or 6 per sons
8.52 (15)
Oth ers
0.6 (1)
7 or 8 per sons
1.13 (2)
Fee per Cli ent (tk)
Age of Entry as CSW
<50
58.5 (103)
15 – 19
27.22 (48)
50 – 100
38.6 (68)
20 – 24
23.85 (42)
100 – 150
2.9 (5)
25 – 29
24.42 (13)
30 – 34
18.74 (33)
35 – 39
3.95 (7)
Average Cli ent per Night: 4
40 – 44
0.6 (1)
Average Nights per Week (NPPW): 4
45 – 49
0.6 (1)
50+
0.6 (1)
Total Num ber of Children
Number of Children
before becoming a CSW

1
14.2 (25)
Child less
26.1 (46)
2
19.3 (34)
1
17.6 (31)
3
19.9 (35)
2
26.1 (46)
4
10.2 (18)
3
20.5 (36)
5
5.7 (10)
4
8.5 (15)
6
1.9 (2)
5
1.1 (2)
Reg u lar Body Check-up
Drug Use
No
21.0 (37)
Yes
89.2 (157)
Yes
45.5 (80)
No
8.5 (15)
Oc ca sionally
33.5 (59)
Occasionally
2.3 (4)

110
Abedin, Khan, Taz Uddin and Rahman
TABLE 5: Selected Health and Health Hazard-Related Variable
Characteristics
Percentage
Characteristics
Percentage
(Numbers)
(Numbers)
Blood Pressure Sys tolic
Blood Pres sure Di a stolic
(mm/Hg)
(mm/Hg)
<100 = very low
3.4 (6)
<50 = very low
0.6 (1)
100 – 110 = low
38.6 (68)
50 – 60 = low
17.0 (30)
110 – 140 = nor mal
50.6 (89)
60 – 90 = nor mal
77.8 (137)
140 – 160 = crit i cal
4.0 (7)
91 – 95 = crit i cal
1.7 (5)
>160 = high
3.4 (6)
>95 = high
2.8 (5)
Av er age BP Sys tolic
Av er age BP Di a stolic
(mm/Hg)
(mm/Hg)
140 – 160 = mild
4.0 (7)
90– 100 = mild
4.0 (7)
hy per ten sion
hy per ten sion
160 – 200 = mod er ate
3.4 (6)
100 – 120 = mod er ate
1.1 (2)
hy per ten sion
hy per ten sion
200+ = se vere hy per ten sion
200+ = se vere hy per ten sion
Pulse Pres sure
Body Fat (%)
<76 = low
29 (51)
<15 = under weight
13.1 (23)
76 – 80 = nor mal
15.3 (27)
15 –2 5 = nor mal weight
56.3 (99)
80+ = high
55.7 (98)
25.1 – 29.9 = over weight
17.0 (30)
30+ = obese
13.6 (24)
Vag i nal Pain
Blood and Pus Dis charge
No
35.2 (62)
No
95.5 (168)
Yes
64.8 (114)
Yes
45. (8)
Itching and Stitch Problem
Vag i nal Pus Dis charge
No
56.3 (99)
No
38.6 (68)
Yes
43.7 (77)
Yes
61.4 (108)
To de tect the fac tors lead ing to the above haz ard ous health con di tions,
the MLR and LPM were used to iden tify con sis tent fac tors. Tables 6 and
Table 7 show the es ti mated pa ram e ters as sum ing lo gis tic and lin ear prob a -
bil ity mod els.
For vag i nal pain, six com mon fac tors are sta tis ti cally sig nif i cant in both
mod els. Four fac tors — BFC, BFN, PNOR and PHIGH — had a pos i tive
im pact but two fac tors (PY and TC) had a neg a tive im pact. Thus, vag i nal
pain in creases with in crease in body fat level and pulse rate. Thus as much
as the fat level de vi ates from its nor mal level, higher the risk of vag i nal
pain. Fur ther, the num ber of years in the sex trade and num ber of chil dren
have a neg a tive im pact on vag i nal pain which may be due to neg a tive im -
pact as lon ger years in sex trade means that the CSW’s phys i cal con di tion
has be come more ad justed to do the haz ard ous job. Also hav ing chil dren

Health Hazards Faced by Floating Commercial Sex Workers… 111
means there has been some ma ter nal health care, which may re sult in less
vag i nal pain.
TABLE 6: Estimate of Parameters Assuming Logistic Approach
Variable
Beta
S.E.
Wald
Df
P
Odd
Value
Ratio
Vag i nal Pain
PY
–0.062
0.028
5.016
1
0.025
0.940
TC
–0.414
0.192
4.618
1
0.032
0.661
BFC
0.457
0.221
4.293
1
0.038
1.580
SNOR
0.548
0.318
2.964
1
0.085
1.730
BFN
0.618
0.309
4.003
1
0.0454
1.856
PNOR
1.158
0.550
4.484
1
0.035
3.185
PHIGH
0.542
0.318
2.904
1
0.088
1.719
Blood and Pus Discharge
BFC
0.895
0.322
7.717
1
0.005
2.448
PHIGH
2.374
1.117
4.513
1
0.034
10.734
Constant
6.786
1.465
21.466
1
0.000
0.001
Vag i nal Pus Discharge
PY
–0.048
0.027
3.225
1
0.073
0.953
BFC
0.342
0.113
9.210
1
0.002
1.408
DNOR
0.224
0.257
0.671
1
0.383
1.251
Itching and Stitch Problem
PY
–0.086
0.303
8.342
1
0.004
0.917
TC
–0.504
0.222
5.140
1
0.023
0.604
BFC
0.738
0.250
8.704
1
0.003
2.092
DU
1.645
0.694
5.612
1
0.018
5.180
BFOV
–0.786
0.464
2.873
1
0.090
0.456
BFC

birth before commencing into profession; BFN

body fat normal,
BFOV — body fat over; DNOR — diastolic normal blood pressure; DU — drug use;
PHIGH — pulse pressure high; PNOR — pulse pressure normal; PY — length of service
in years; SNOR — systolic normal blood pressure; and TC — total number of children.
For blood and pus dis charge, two fac tors are com mon in both MLR and
LPM and these are con sis tent to the model. Two fac tors, BFC and PHIGH
bear pos i tive im pact on blood and pus dis charge. There fore, blood and
pus dis charge are more fre quent for those who have given birth be fore en -
ter ing the sex trade and for those with high pulse rate. In other words, sex
work ers with high pulse rate may cause ir reg u lar blood cir cu la tion over
the body and so blood and pus dis charge may be more prev a lent.
For itch ing and stitch prob lem, four fac tors are found con sis tent to the
model. Here, birth be fore com menc ing to the pro fes sion (BFC) and drug
use (DU) are pos i tively as so ci ated, but years of ser vice (PY) and total num -
ber of chil dren (TC) are neg a tively as so ci ated.

112
Abedin, Khan, Taz Uddin and Rahman
TABLE 7: Estimate of Parameters in LPM
Variables
Beta
Std. Error
T
P-Value
Vaginal Pain
PY
–0.130
0.006
–2.222
0.028
TC
–0.393
0.040
–2.941
0.004
BFC
0.356
0.044
2.962
0.004
BFN
0.112
0.070
1.714
0.088
PLOW
0.388
0.105
5.528
0.00
PNOR
0.394
0.117
6.782
0.000
PHIGH
0.642
0.084
8.225
0.000
Blood and Pus Discharge
BFM
0.372
0.011
3.275
0.001
NPPW
–0.308
0.008
–2.008
0.038
PHIGH
0.295
0.031
2.709
0.007
Vag i nal Pus Discharge
PY
–0.123
0.006
–1.668
0.097
BFC
0.211
0.027
2.848
0.005
Constant
0.065
8.200
0.000
Itching and Stitch around Vagina
PY
–0.192
0.006
–2.686
0.008
TC
–0.267
0.039
–2.206
0.029
BFC
0.394
0.045
3.280
0.001
DU
0.179
0.131
2.494
0.014
PLOW (pulse pressure low) and NPPW (Number of nights performing per week)
TABLE 8: Common Factors for Hazard Variables
Hazards Variables
MLR
LPM
Common
Factors

Vag i nal Pain
+ve: BFC, SNOR,
+ve: BFC, BFN,
BFC, BFN,
BFN, PNOR,
PLOW, PNOR,
PNOR
PHIGH
PHIGH
–ve: PY, TC
–ve: PY, TC
PHIGH, PY,
TC
Ab dom i nal and Pus
+ve: BFC, PHIGH
+ve: BFC, PHIGH BFC, PHIGH
Discharge
–ve: NPPW
Ab dom i nal Pain
–ve: NCP, DU
–ve: NCP, DU
NCP, DU
Itching and Stitch Prob lem
+ve: BFC, DU
+ve: BFC, DU
BFC, DU,
PY, TC
–ve: PY, TC, BFOV –ve: PY, TC
Vag i nal Pus Dis charge
+ve: BFC, DNOR
+ve: BFC
BFC, PY
–ve: PY
–ve: PY
Note:
The notations here are same as in Tables 6 and 7. Common factors are model
invariant.

Health Hazards Faced by Floating Commercial Sex Workers… 113
Two com mon fac tors — length of ser vice in the sex trade (PY) and num ber
of chil dren be fore com menc ing to the pro fes sion (BFC) sig nif i cantly af fect the
risk of vag i nal pus dis charge. Al though length of ser vice is neg a tively as so ci -
ated, giv ing birth be fore en ter ing the pro fes sion is pos i tively as so ci ated with
the risk of vag i nal pus dis charge. Thus, with the ser vice length the phys i cal ad -
just ment as well as bi o log i cal ad just ment in creases and so the risk of this prob -
lem de creases. Again, in the logit model, di a stolic nor mal blood pres
sure
shows pos i tive im pact on vag i nal pus dis charge.
RECOMMENDATIONS AND CONCLUSION
Floating CSWs are de prived from al most all health op por tu ni ties. Be sides,
they are il lit er ate and ig no rant about their phys i cal fit ness, nu tri tional sta -
tus and var i ous STDs. To im prove their over all health prob lems and al le vi -
ate the health risks, the fol low ing steps can be adopted:
The existing level of awareness on health status and correct information
about transmission as well as prevention of particular diseases should be
updated so that all CSWs can avail of this opportunity.
Both governmental and non-governmental organisations should take
up measures to increase information and awareness on family laws. A
large number of women, unaware of marital rights, ultimately choose
this profession finding no other means of subsistence economy.
Care and support programmes, regular body check-ups, voluntary
counselling and testing should be strengthened by both government
and non-government programmes.
Sex education, knowledge on sexual behaviour, correct use of
condom (to include condom demonstration among clients of CSWs)
can reduce the risk of sexually transmitted diseases. Also special
training of sex workers to use female condoms may substantially
reduce the risk of health complications as well as spreading STDs to a
large extent.
REFERENCES
Bangladesh: Ministry
: Statistical Year Book, Dhaka: Bangladesh Bureau of Statistics.
of Planning
2006
Gujarati, D.N.
: Basic Econometrics (Edition 3), New York: McGraw-Hill.
1995

114
Abedin, Khan, Taz Uddin and Rahman
Hong, Y. and Li, X.
: Behavioral Studies of Female Sex Workers in China: A Literature
2008
Review and Recommendation for Future Research, AIDS and
Behavior
, 12(4), 623–636.
Hosmer, D.W. and
: Applied Logistic Regression (Edition 2), John Wiley & Sons,
Lemeshow, S.
Inc.
2000
Huang, Y.,
: HIV/AIDS Risk Among Brothel-Based Female Sex Workers in
Henderson, G.E.,
China: Assessing the Terms, Content, and Knowledge of Sex
Pan, S. and
Work, Sexually Transmitted Diseases, 31(11), 695.
Cohen, M.S.
2004
Millar-Craig M.W.,
: New System for Recording Ambulatory Blood Pressure in Man,
Hawes, D. and
Medical & Biological Engineering & Computing, 16, 727–731.
Whittington, J.
1978
Must, A., Spadano, J.,
: The Disease Burden Associated with Overweight and Obesity,
Coakley, E.H.,
JAMA, 282, 1523–1529.
Field, A.E., Colditz, G.
and Dietz, W.H.
1999
Nessa, K., Waris, S.A.,
: Sexually Transmitted Infections Among Brothel-based Sex
Alam, A., Huq, M.,
Workers in Bangladesh: High Prevalence of Asymptomatic
Nahar, S. Faisal, A.,
Infection, Sexually Transmitted Diseases, 32(1), 13–19.
Hasan, C., Monira, S.,
Monir, U.B., Sultana, J.,
Kazi, F.M., Das, J.,
Dipak, K.M., Sultan, Z.,
Hossain, N. and Rahman, M.
2005
Nessa, K., Waris, S.,
: Epidemiology and Etiology of Sexually Transmitted Infection
Sultan, Z., Monira, S.,
among Hotel-Based Sex Workers in Dhaka, Bangladesh,
Hossain, M., Nahar, S.,
Journal of Clinical Microbiology , 42(2), 618–621.
Rahman, H., Alam, M.
Baatsen, P. and Rahman, M.
2004
Urbina, E., Alpert, B.,
: Ambulatory Blood Pressure Monitoring in Children and
Flynn, J., Hayman, L.,
Adolescents: Recommendations for Standard Assessment. A
Harshfield, G.A.,
Scientific Statement From the American Heart Association
Jacobson, M.,
Atherosclerosis, Hypertension, and Obesity in Youth
Mahoney, L.,
Committee of the Council on Cardiovascular Disease in the
McCrindle, B.,
Young and the Council for High Blood Pressure Research,
Mietus-Snyder, M.,
Hypertension, 52, 433–451.
Steinberger, J.,
Daniels, S.
and Harshfield, A.
2008

Health Hazards Faced by Floating Commercial Sex Workers… 115
Wong, W.C.J.,
: Female Street Sex Workers in Hong Kong: Moving Beyond
Holroyd, E.A., Gray, A.
Sexual Health, Journal of Women’s Health, 15(4), 390–399.
and Ling, D.C.
2006
Yang, X. and Xia, G.
: Gender, Work, and HIV Risk: Determinants of Risky Sexual
2006
Behavior Among Female Entertainment Workers in China,
AIDS Education and Prevention, 18(4), 33.
THE INDIANJ OURNAL OF SOCIAL WORK, Vol ume 70, Issue 1, Jan u ary 2009