• cervical cancer;
  • health delivery models;
  • Hong Kong;
  • prostitution;
  • screening


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

The purpose of this study was to see if an outreach model could be a feasible option for early detections of preinvasive cervical cancer in women with double identities of female sex workers (FSW) and illegal migrant workers in Hong Kong who most needed cervical screening but were often deprived of such a service. High turnover rates, acceptability, and compliance for follow-ups could potentially render such a clinical model unsuccessful. A total of 245 FSW were screened at the outreach clinic from January 2004 to December 2005, which was set up in a nongovernmental organization in a red light district. A questionnaire regarding their lifestyles and demographic details was used before a gynecological history, Papanicolaou (PAP) smear, and other health checkup were conducted. Chi-square test and multinomial logistic regression were used to analyze the results. Of 235 women tests, 9.8% of them had CIN I–III, and places of origin were found to be important risk factors for abnormal PAP smears. The nonlocal workers were significantly more likely to have abnormal PAP smears (χ2= 10.55, P= 0.04). Among the women, 88.1% of them who had the tests returned for follow-up with poorer compliance among those with an abnormal result. We conclude that an outreach well-women clinic seems to be an acceptable option for these women and an effective way for the early detection of cervical cancer.

The sex trade in Hong Kong has been an integral and ever-expanding component of the city’s socio-cultural and economic structure for many years, with current estimates suggesting that there are 200,000 female sex workers (FSW) within a total population of 6.8 million people(1). FSW are recommended to have more frequent cervical screening due to their higher risks of developing cervical cancer(2). Paradoxically, this is often the group of women that are much less likely to participate in a cervical screening program(3). In the survey conducted in 2004, we found that only 27.5% of the street FSW had ever had a cervical smear(4) whereas the uptake rate of such a preventive measure by the female population of Hong Kong exceeded 60%(5).

With the closer integration and high volume of cross-border travel between China and Hong Kong, Hong Kong’s FSW are increasingly drawn from mainland Chinese women with lower education and little negotiating power compared to the general public and indeed other Hong Kong FSW(6). In fact, one survey showed that over 88.8% of street FSW in Hong Kong were nonlocal residents(7). Fundamentally, it is the disparities in poverty and wealth between Hong Kong and some mainland cities which creates a strategic opportunism that encourages these women to risk their health and safety to work in Hong Kong. This situation is further exacerbated since June 2003 when China permitted independent travelers, a move which is welcomed by many people as it is seen to be beneficial to the local economy. On the other hand, entering Hong Kong on a visitor’s or fraudulent visa, these women are unable to get protection from the Hong Kong police or to report crimes without risking criminal charges or deportation(8). They are unfamiliar with the local medical system and the public venereal service charges HK$1400 (US$1 = HK$7.8) per time for all non-Hong Kong residents (compared to HK$200 per time for the residents). The high turnover and secretive nature of sex work combined deter many FSW from seeking medical assistance and treatment for acute illnesses, let alone preventive measures such as cervical screening.

Outreach clinics are shown to be successful to control sexually transmitted infections (STI) in FSW(9), yet we were unable to identify studies on the effectiveness of outreach clinics for cervical cancer control for this group of women. Using a decision analysis, You et al.(10) showed that an outreach clinic in preventing transmission of gonorrhea and chlamydia between FSW and their clients in Hong Kong could potentially be less costly and more effective. Nonetheless, there are possibly other problems with an outreach approach for these mobile migrant workers including acceptability, compliance for follow-ups, or affordability for further investigation and management. In January 2004, we started on experimental basis an outreach well-women clinic for FSW in which basic screening of blood pressure, weight and height, as well as breast examination and Papanicolaou (PAP) smear were provided. We also offered contraceptive advice and opportunistic health education on their occupational risks. The cervical smear was donated by Cytyc Corporation and a private laboratory, and thus offered free to the workers, with an optional high vaginal sample microscopy and culture or any other laboratory investigation charged at cost.

In this paper, we report the clinical activities and the preliminary results of the outreach clinic to see if it was a feasible model. Comparing local and nonlocal FSW, we examined the potential benefits and difficulties of the outreach clinic for different groups of women. In the official statistics, it is estimated that in China the migrant population now constitutes more than 100 million, and the government expects this number will grow by 46 million over the next 5 years(11). Given the huge number of migrant workers in and from China, it is anticipated that the experience of such a health delivery model is invaluable for policymakers and health professionals to improve the health of these women.

Materials and methods

  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

An outreach clinic was held at a nongovernmental organization, Ziteng, established in 1996 to help FSW in Hong Kong. They operated an integrated community drop-in center free to all FSW irrespective of their places of origin or legal status. They maintained a very good network, relationship, and trust among their clients who were mainly street sex workers, single-household brothels, and some organized brothels accounting for the major population of the visible commercial sex workforce.

From January 2004 to December 2005, the outreach workers from Ziteng invited FSW to the twice-monthly well-women clinic. In the clinic, they were asked to fill in a simple questionnaire regarding their lifestyles (e.g. smoking and drinking habits) and demographic details including age, place of origin, marital status, and educational level. Unidentified names could be used and unlike government clinics official documents were not checked. Nevertheless they were asked to leave the means of contact if they wished to be informed when the results were available. A volunteer doctor would conduct a gynecological history including urogenital and sexual risk assessments before a physical examination that included a breast examination, measuring vital signs, a gynecological examination with cervical smear as well as relevant examinations addressing an individual’s concern. Health education was given during the session and referrals if appropriate would be arranged at the end or in a subsequent visit.

Liquid-based PAP technique with CytoBrush as the sampling device was used. Samples were sent to a private laboratory accredited by National Association of Testing Authorities, Australia and Hong Kong Accreditation Services. The results of cervical smear were reported in CIN system and Bethesda system (2003). High vaginal swabs were requested if there was the clinical evidence of vaginitis or at the sex worker’s request. The specimen taken from cervix by high vaginal swab (invasive sterile Eurotubo collection swab [Deltalab, Rubi, Spain]) was kept in Amies transport medium and cultured for microorganisms such as gonococci.

The results were explained to the FSW in a follow-up visit 1–4 weeks later. On some occasions when the FSW could not make a follow-up visit or had left Hong Kong, results were also explained by the doctor through telephone. In the event of CIN II or III results, a referral letter was given to the FSW to attend a gynecologist of choice. For other abnormal smear results, the FSW was offered the option of repeating another smear in 3–6 months or to be referred to a specialist.

For practical reasons, we grouped those FSW holding a visitor’s visa as nonlocal workers. New immigrants were grouped as local workers. SPSS version 13.0 was used for data management and analysis. Because there was only one CIN II, we grouped it with CIN III for statistical analysis. We used descriptive statistics for the FSW personal and social data, chi-square test to compare the results, and multinomial logistic regression for associations of risk factors. The P≤ 0.05 was taken as statistically significant.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Among FSW, 245 were recruited through the outreach workers and by word of mouth. Their age ranged from 20–57 years, of whom 75 (30.6%) were illegal migrant workers on a temporary (visitor’s) visa. Table 1 shows their demographic and family characteristics. In this study, there were 199 women (81.9%) with children in the family and 76 (38.2%) of them with two or more children. Of these, 87 (43.7%) were bringing up their children without a partner, 24.9% of them smoked while only 22.4% did regular exercise. Comparing to nonlocal FSWs, the local FSWs were more likely to be older (t= 3.14, P= 0.002), smoker (χ2= 4.57, P= 0.032), alcohol drinker (χ2= 4.29, P= 0.038), and single-mother (χ2= 14.32, P= 0.006).

Table 1.  Demographic and family characteristics of FSWs in the study
CharacteristicLocal FSW (n= 170)Nonlocal FSW (n= 75)Total (N= 245)Independent t-testP values
Mean age (in years) with Standard Deviation (SD)36.6 (6.00)34.0 (5.78)35.8 (6.04)3.140.002
CharacteristicsLocal FSW (N= 170) (%)Nonlocal FSW (N= 75) (%)Total (N= 245) (%)Chi-squareP values
Smoking49 (28.8)12 (16.2)61 (24.9)4.580.032
Regular exercise41 (24.5)13 (17.6)54 (22.4)1.440.23
Sexual partners(n= 164)(n= 69)(n= 233)15.78<0.01
 Married/cohabitated56 (34.1)43 (62.3)99 (42.5) 
 Single/divorced/widowed108 (65.9)26 (37.7)134 (57.5) 
Education(n= 162)(n= 75)(n= 237)0.440.80
 Primary or below28 (17.3)13 (17.3)41 (17.3)
 Low secondary91 (56.2)45 (60.0)136 (57.4)
 High secondary or above43 (26.5)17 (22.7)60 (25.3)
Number of children in family(n= 74)(n= 74)(n= 243)3.660.60
 018 (24.3)18 (24.3)44 (18.1)
 133 (44.6)33 (44.6)123 (50.6)
 216 (21.6)16 (21.6)55 (22.6)
 33 (4.1)3 (4.1)11 (4.5)
 43 (4.1)3 (4.1)7 (2.9)
 51 (1.3)1 (1.3)3 (1.3)
Marital status of FSW who has child(ren)(n= 143)(n= 51)(n= 194)14.320.006
 Married54 (37.8)33 (64.7)87 (44.8)
 Cohabited2 (1.4)1 (2.0)3 (1.5)
 Single14 (9.8)1 (2.0)15 (7.7)
 Divorced66 (46.1)15 (29.5)81 (41.8)
 Widowed7 (4.9)1 (2.0)8 (4.2)

Sexual behavior and gynecological history is presented in Table 2. Among them, 67.4% performed regular vaginal douche with over-the-counter medicine. Among FSW, 69.7% of them reported having gynecological examinations in the past but 35.5% never had a cervical smear. Of the 186 FSW who answered this question, 55 (29.6%) had no regular sex partners and 115 (61.8%) had one regular partner; 16 (8.6%) had two or more. Eighty percent of FSW always used condom with clients, but only 35.1% insisted on condoms when they had sex with their own sexual partners.

Table 2.  Sexual behavior and gynecological screening history of FSW
BehaviorLocal FSW (n= 170) (%)Nonlocal FSW (n= 75) (%)Total (N= 245) (%)χ2P value
Noncommercial sexual partner(n= 138)(n= 48)(n= 186)3.780.29
 045 (32.6)10 (20.8)55 (29.6)
 180 (58.0)35 (72.9)115 (61.8)
 2 or more13 (9.4)3 (6.3)16 (8.6)
Condom use with client(n= 167)(n= 73)(n= 240)7.980.09
 Always138 (82.6)54 (74.0)192 (80.0)
 Not always29 (17.4)19 (26.0)48 (20.0)
Condom use with partner(n= 123)(n= 68)(n= 191)0.970.97
 Always41 (33.3)26 (38.2)67 (35.1)
 Not always82 (66.7)42 (61.8)124 (64.9)
Vaginal douching(n= 167)(n= 72)(n= 239)0.670.41
 Yes110 (65.9)51 (70.8)161 (67.4)
 No57 (34.1)21 (29.2)78 (32.6)
Previous STI(n= 167)(n= 72)(n= 239)0.010.94
 Yes45 (26.9)19 (26.4)64 (26.8)
 No122 (73.1)53 (73.6)175(73.2)
Previous gynecological check(n= 168)(n= 73)(n= 241)4.120.04
 Yes124 (73.8)44 (60.3)168 (69.7)
 No44 (26.2)29 (39.7)73 (30.3)
Previous PAP smear(n= 169)(n= 73)(n= 242)11.170.001
 Yes72 (42.6)14 (19.2)86 (35.5)
 No97 (57.4)59 (80.8)156 (64.5)

In our sample, the local FSW were significantly more likely to have PAP smear done (χ2= 11.17, P= 0.001) and gynecological checks (χ2= 4.12, P= 0.04) (Table 2). There were no differences between the two groups of FSW in sexual risk behavior such as condom use, vaginal douche (χ2= 0.67, P= 0.41), and the history of STI (χ2= 0.01, P= 0.98).

A total of 236 PAP smears (96.3% of the recruits) were performed, of which 207 (87.7%) were reported as “normal” (Table 3). The nonlocal workers were significantly more likely to have abnormal PAP smears (χ2= 10.55, P= 0.04) (Table 3). Multinomial logistic regression of PAP smear results against independent variables (including age, smoking, drinking, exercise, pregnancy, previous STI, previous gynecological examinations, previous PAP smears, and places of origin) show that local workers were much less likely to have CIN II and CIN III (P= 0.016, OR = 17.13, 95% CI: 1.71, 171.68) (Table 4).

Table 3.  PAP smear results of FSWs attended the outreach service
ResultsLocal FSW n (%)Nonlocal FSW n (%)Total N (%)χ2P value
Normal147 (90.7)60 (81.1)207 (87.7)10.550.04
Reactive changes in squamous cells4 (2.5)2 (2.7)6 (2.5)
CIN I10 (6.2)6 (8.1)16 (6.8)
CIN II–III1 (0.6)6 (8.1)7 (3.0)
Table 4.  Risk factors of abnormal PAP smears using multinomial logistic regression modela
Independent variablesP valueOR (95% CI)
  • a

    The reference category is “first category”.

 Age0.460.66 (0.22–2.00)
  Gravidity0.641.76 (0.16–19.22)
  Abortion0.142.51 (0.75–8.44)
 Smoking0.460.59 (0.15–2.41)
 Drinking0.850.85 (0.14–4.97)
 Exercise0.650.74 (0.21–2.66)
 Previous STI0.120.40 (0.12–1.27)
 Previous PAP smear0.202.37 (0.63–8.84)
 Place of origin
  Non-HK resident0.531.44 (0.46–4.48)
  HK resident 
 Age0.570.61 (0.11–3.35)
  Gravidity9.79E-008 (9.79E-008–9.79E-008)
  Abortion0.250.25 (0.02–2.60)
 Smoking0.931.11 (.109–11.16)
 Drinking1.004662213.26 (0.00)
 Exercise0.710.71 (0.12–4.28)
 Previous STI0.080.18 (0.03–1.19)
 Previous PAP smear0.244.30 (0.38–48.80)
 Place of origin
  Non-HK resident0.0217.132 (1.71–171.68)
  HK resident

Table 5 shows the follow-up pattern of the patients. Of these 236 FSW, 113 (47.9%) returned for results in 2 weeks, 71 (30.1%) in 3–4 weeks, 24 (10.2%) in more than 1 month, and 28 (11.9%) defaulted follow-up visits and could not be traced. Of the 29 abnormal PAP smears, only 3 had previous cervical smears (2 workers 2 years ago and 1 worker 5 years ago). We repeated the cervical smear in seven (24.1%) workers (four CIN I and one CIN-III reverted normal, one CIN I became CIN III, and one CIN II reported reactive change in squamous cells). Nine workers were referred for further management (two with reactive change in squamous cell, two with CIN II, and five with CIN III). We failed to contact 13 workers (44.8%) with abnormal PAP results: 4 with reactive changes and 9 with CIN I, of whom 8 were local workers.

Table 5.  Follow-up results of the outreach clinic
CharacteristicsCases N (%)
Return for results(n= 236)
 In 2 weeks113 (47.9)
 In 3–4 weeks71 (30.1)
 More than 1 month24 (10.1)
 Failed to turn up for appointments28 (11.9)
Follow-up of abnormal smear(n= 29)
 Repeat smear7 (24.1)
 Referred9 (31.1)
 Failed to contact13 (44.8)


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

Most services and research on FSW so far have focused on their high risk of STI/HIV(12) with little interest in cervical cancer prevention and looking at their health in a holistic manner. Previous study of FSW in Hong Kong showed that these women scored significantly lower in physical, psychological, and environmental health in World Health Organization Quality of Life score when compared to nonsex workers of the same age group and sex(4). Since April 2003, non-Hong Kong residents are subject to a fee when utilizing medical services in Hong Kong and this has deterred many mainland Chinese FSW from seeking medical investigations and treatment when needed(7). Arguably, access to health care is a fundamental human right and that right is often deprived because of the legal, financial, and social status of the sex workers, in particular of migrant FSW. Sex workers are a subsector of the population who have common health problems as well as more specific health risks and require greater sensitivity. With these fundamental beliefs, we tested this service-delivering model in form of an outreach well-women clinic. Due to limited resources, we were only able to offer two clinical sessions a month and hence explained a modest sample size. Nonetheless, we have shown that such an outreach clinic is not only feasible but also worthwhile in identifying abnormal early changes in cervical cells and referring most of these women for further management and care.

Despite 69.7% of FSW reported having had gynecological examinations previously, only 35.5% realized that they should have had a cervical smear as well. As predicted in the “Inverse Care Law”(13), those in need are less likely to receive the care: 9.8% (23/235) FSW had preinvasive lesions (CIN I–CIN III) in contrast to 5.5% (395/3601) of the Hong Kong general population(14). Comparing to other studies of abnormal smears in sex workers, it is much higher than those found in Singapore (5.6%)(15) and Austria (6.3%)(16), but more comparable to a startlingly high prevalence in Venezuela (13.9%)(17). Moreover, the regression analysis showed that the place of origin (local or nonlocal FSW) was the single most important risk factor for CIN II–III—this independent variable was unlikely a confounding factor for whether cervical smear had ever been done as this has been controlled.

This study also provides healthcare providers and policymakers the essential information about the FSW and their work that are important in organizing future health services and interventions. As we can see, majority of our sampled women had dependents and half were single mothers. Traditional public genitourinary clinic of 9 am–5 pm opening hours may not accommodate their needs. It is also of concern that 16 FSW (8.6%) admitted having more than one noncommercial sex partners, of whom only two reported “always” using condoms. As compared to commercial sex activity, 80% of the FSW “always” used condoms, an improvement over the 71–75% reported in Hong Kong for the years of 1999–2000(1). Many FSW had undesirable health practices and thus required intensive and proactive intervention. Opportunistic education on safe sex at the time when they seek medical help, whether it is resulted from increased perceived susceptibility or severity, will act as a strong motivating factor for a change of behavior(18–19). We conducted considerable education on vaginal douching in the outreach clinic setting as most of them douched regularly and frequently.

The headquarters of Ziteng was close to an area of concealed commercial sex and they have a very experienced outreach team. Ziteng approached their clients from their old record, those they met on the street and through snowballing method. Thus, nonresponse rate and characteristics of the nonrespondents were not available and arguably irrelevant. Given the sensitive and underground nature of their work, this method of sample collection is common, as seen in a recently published article(20). Nearly 80% of FSW returned for the results within one month with 11.9% lost to follow-up. Considering the mobile and implicit nature of this population and their work, such compliance is regarded as satisfactory and it shows, with basic education, these women can appreciate the importance and needs of such tests. It will be useful to see how many (at least in local FSW) return for repeated cervical smear at one year and to conduct a cohort study to see if such intervention in long term can reduce morbidity and mortality in these women. A previous study of human papillomavirus (HPV) serotypes revealed large variation of geographic and epidemiologic determinants in Asia(21) and this will have significant implication in view of the introduction of HPV vaccine in the near future. A separate study of the serotyping of HPV infection among FSW in Hong Kong is now underway.

In summary, an outreach well-women clinic for FSW is not only feasible but very valuable in preventing cervical cancer in FSW. The follow-up rate was high. The clinic does not only facilitate the early detection of cervical cancer but also provides an opportunity for health education in the prevention of STI, and the proper care of the urogenital system among Chinese FSW in Hong Kong.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References

We owe a great debt to Ziteng and Cytyc Corporation who have kindly agreed to take part in this study. Were it not for their support and active participation, completion of the study would have never been accomplished. We hope this piece of work will address some of the sex workers’ health problems they have to go through and make a small step toward health equality.


  1. Top of page
  2. Abstract
  3. Materials and methods
  4. Results
  5. Discussion
  6. Acknowledgments
  7. References
  • 1
    Chan MK, Ho KM, Lo KK. A behavior sentinels surveillance for female sex workers in the social hygiene service in Hong Kong (1999-2000). Int J STD AIDS 2000:13:81520.
  • 2
    Mak R, Van Renterghem L, Cuveller C. Cervical smears and human papillomavirus typing in sex workers. Sex Transm Infect 2004;80:11820.
  • 3
    Tideman RL, Marks C, Knight VM. Opting off the New South Wales PAP test register in a sexual health setting. Aust N Z J Public Health 2001;25:5014.
  • 4
    Wong WCW, Gray A, Ling DC. Female street sex workers in Hong Kong: moving beyond sexual health. J Womens Health 2006;15:3907.
  • 5
    The Behavioural Risk Factor Surveillance System of Department of Health of Hong Kong website. Available at: ( (Accessed Apr. 2006).
  • 6
    Abdullah ASM. Cross-border travel and HIV/AIDS risk among travelers in Hong Kong. Travel Medicine News Share 1996;4.
  • 7
    Wong WCW, Gray A, Ling DC et al. Patterns of health care utilization and health behaviours among street sex workers in Hong Kong. Health Policy 2006;77:1408.
  • 8
    Holroyd EA, Wong WCW, Gray A et al. Hong Kong’s female sex workers: occupational dangers and health perils in the workplace. J Adv Nurs (forthcoming).
  • 9
    Adab P, McGhee SM, Yanova J et al. Effectiveness and efficiency of opportunistic cervical cancer screening. Med Care 2004;42:6009.
  • 10
    You J, Wong WCW, Sin CW, Woo J. The cost-effectiveness of an outreach clinical model in the management and prevention of gonorrhea and chlamydia among Chinese female sex workers in Hong Kong. Sex Transm Dis 2005;33:2207.
  • 11
    Zhan SK, Sun ZW, Blas E. Economic transition and maternal health care for internal migrants in Shanghai, China. Health Policy Plan 2005;17:4755.
  • 12
    Vanwesenbeeck I. Another decade of social scientific work on sex work: a review of research 1990-2000. Annual Review of Sex Research 2001;12:24289.
  • 13
    Seema Parikh, Paul Brennan, Paolo Boffetta. Meta-analysis of social inequality and the risk of cervical cancer. Int J Cancer 2003;105:68791.
  • 14
    Chueng AN, Szeto EF, Ng KM. Atypical squamous cells of undetermined significance on cervical smears: follow-up study of an Asian screening population. Cancer 2004;102:7480.
  • 15
    Chan R, Khoo L, Ho TH et al. A comparative study of cervical cytology, colposcopy and PCR for HPV in female sex workers in Singapore. Int J STD AIDS 2001;12:15963.
  • 16
    Gitsch G, Kainz Ch, Reinthaller A. Cervical neoplasia and human Papilloma virus infection in prostitutes. Genitourin Med 1991;67:47880.
  • 17
    Nunez JT, Delgado M, Giron H. Prostitution and other cofactors in preinvasive and invasive lesions of the cervix. Aust N Z J Obstet Gynaecol 2005;44:23943.
  • 18
    Rosenstock I. Why people use health services. Milbank Mem Fund Q 1996;44:94121.
  • 19
    Becker MH. The health belief model and personal health behavior. Thorofare, NJ: Charles B. Slack, Inc., 1974.
  • 20
    Jeal N, Salisbury C. Self-reported experiences of health services among female street-based prostitutes: a cross-sectional survey. Br J Gen Pract 2004;54:5159.
  • 21
    Chan PKS. Epidemiology of human Papillomavirus in Asia: do HPV-52 and HPV-58 play a role? Papillomavirus Report 2005;16:26571.