Abstract
We utilized the Protection Motivation Theory (PMT) to assess predictors of intention and behavior of consistent condom use among Chinese female sex workers (FSWs). A self-administered questionnaire was used in a cross-sectional survey among 700 FSWs in Guangxi, China. Multivariate logistic regression analysis indicated that extrinsic and intrinsic rewards, self-efficacy and response costs predicted consistent condom use intention and behavior among FSWs. STI/HIV prevention programs need to reduce FSWs’ perceptions of positive extrinsic rewards and intrinsic rewards for engaging in consistent condom use, reduce FSWs’ perception of response costs for using a condom, and increase condom use self-efficacy among FSWs.
Similar to those in other countries, female sex workers (FSWs) in China experience significant abuse and harm and are a marginalized and stigmatized group (Boittin, 2013). FSWs face not only psychosocial stress but also public health problems such as sexually transmitted infection (STI) including human immunodeficiency virus (HIV) infection due to their vulnerability in society (Zhou et al., 2014). Because of the high risk for HIV/STI of FSWs, a better understanding of risk factors will inform HIV prevention programs among FSWs in China as well as other countries, especially Asian countries with high HIV prevalence.
Commercial sex in China has grown dramatically over the past decades (Tucker, Ren, & Sapio, 2010). The estimated number of FSWs has increased from 25,000 in 1985 to 2.8-4.5 million in 2006 (Lu et al., 2006). In 1949 when the Communist Party took over China, sex work was banned and reportedly eliminated (Boittin, 2013). During the 1980s, China's open door policy and economic reforms have been attributed to a remarkable resurgence in commercial sex (Cai et al., 2010). Most Chinese FSWs are rural-to-urban migrants (Kaufman, 2011). Sex work is a criminal offense in China. Law enforcement drives sex workers underground, affecting both their safety and access to health services (Kaufman, 2011). The presence of condoms or condom wrappers is sufficient evidence to prosecute FSWs by some law enforcement agencies (Zou, Xue, Wang, & Lu, 2012).
FSWs are at high risk for STI including HIV infection in China. In a study in five cities of China, researchers reported that the prevalence of gonorrhea and chlamydia among FSWs was 26% and 41%, respectively (Rou et al., 2007). Others reported that the prevalence of syphilis among FSWs was 11% (Lu et al., 2009). In China, most FSWs work in the entertainment or personal service establishments such as karaoke, night clubs, dance halls, discos, bars, saunas, hair salons, massage parlors, barbershops, roadside restaurants, and mini-hotels (Yang & Xia, 2006; Zhang et al., 2011). In China, most of the FSWs have both casual (commercial or non-commercial) and stable sexual partners (Rou et al., 2007). The rate of condom use among FSWs with their casual sexual partners has been reportedly low in China (Lau, Tsui, Siah, & Zhang, 2002). Behavioral surveillance data show that 60% of FSWs in China use condoms inconsistently (Wang et al., 2009). FSWs have less power to negotiate condom use since condom use depends on male cooperation and behavior, particularly in Chinese society with patriarchal norm (To, Tam, & Chu, 2012). Consistent condom use may also differ by the type of sexual partners (e.g., casual or stable) (Lofberg, Sharma, Keli, Bukusi, & Cohen, 2004; Macaluso, Demand, Artz, & Hook, 2000). To better understand factors predicting the intention and behavior of condom use among FSWs in China, a survey was conducted among 700 FSWs in two cities in Southwest China. We intend to explore the individual cognitive factors related to consistent condom use among FSWs in an area with high HIV prevalence.
Among individual factors, individual's perceptions and beliefs related to STI/HIV risk and self-efficacy of condom use have impact on FSWs’ decision to use a condom (Zhang et al., 2004). We employ a social cognitive theory, Protection Motivation Theory (PMT) developed by Rogers (Rogers, 1983). PMT has been widely used to predict a range of health related intentions and behaviors, including STI/HIV risk reduction (Milne, Sheeran, & Orbell, 2000; Pack, Li, Stanton, & Cottrell, 2011; Stanton, Black, Kaljee, & Ricardo, 1993; Stanton et al., 1996). PMT postulates that facing a health threat evokes two cognitive pathways: threat appraisal and coping appraisal (Rippetoe & Rogers, 1987). Threat appraisal pathway evaluates the seriousness of the perceived threat and likelihood of suffering from the threat while the coping appraisal pathway assesses the effectiveness of the response to the threat and the personal ability to take the response. These two appraisals include seven cognitive factors: intrinsic rewards, extrinsic rewards, vulnerability, severity, self-efficacy, response efficacy, and response costs (Rogers, 1983). In spite of the wide use of PMT in sexual risk reduction and condom use research among adolescents in different cultures (Chen et al., 2009; Kaljee et al., 2005; Li, Zhang, Mao, Zhao, & Stanton, 2011; Stanton et al., 2006; Zhang et al., 2004), there has been limited research assessing predictive value of PMT for predicting intention of consistent condom use and condom use behaviors among FSWs, a most-at-risk population of HIV infection in China. Our purpose in this study was to assess whether PMT factors predict intention and behavior of consistent condom use with different types of sexual partners (i.e., stable or casual) among Chinese FSWs.
Methods
Study site
This study was conducted in two touristic cities in Guangxi Zhuang Autonomous Region (Guangxi) in China. Guangxi is located in the southern part of China with a population of 46 million (National Bureau of Statistics of China, 2012). Guangxi has the second largest number of reported HIV cases among 31 provinces in China (China Ministry of Health, 2012). There is a high demand and a big market for commercial sex in Guangxi due to the prosperous economic development and tourism with a large number of FSWs providing sexual service primarily through commercial sex venues (Fang et al., 2007).
Participant recruitment and data collection
There were an estimated 2,000 FSWs in about 150 entertainment establishments in each of the two participating cities. In the current study, commercial sex venues in each city were identified by the research team through ethnographic mapping and 76 commercial sex venues were selected and participants were recruited from these commercial sex venues. Gatekeepers of these commercial sex venues were contacted for permission to invite FSWs in the venues to participate in the study. Women who worked in these venues and had provided sex for money or living materials were eligible to participate. A total of 700 women were recruited in two cities and a self-administered questionnaire was completed by each FSW in a private room of participating commercial sex venues. Each questionnaire took about 45 minutes to complete. Trained interviewers provided necessary assistance to participants during the survey. For those FSWs who had limited literary or had reading difficulty (less than 5%), interviewers read the questions and response options to the participants who selected and recorded their answers in the questionnaire. The study protocol was approved by the Institutional Review Boards at Wayne State University in the United States and Guangxi Center for Disease Control and Prevention (CDC) in China.
Measures
Demographic factors
Participants’ age, ethnicity (Han vs. minority), original place of residence (urban vs. rural), education (not more than elementary school, middle school, at least high school), and monthly income were included in data analyses. Stable sexual partners were defined in this study as male sexual partners with whom FSWs had a long-term (e.g., more than six months) sexual relationship (e.g., husband, boyfriend or lover). Casual sexual partners were defined as male sexual partners with whom FSWs had commercial or a short term sexual relationship (e.g., one-night stand).
PMT constructs
Items and Cronbach's alpha for seven PMT factors related to condom use are listed in Table 1. Extrinsic rewards: the perception of extrinsic rewards was assessed using four items related to peer approval or peer engagement of no-condom use (e.g., perceived proportion of FSWs who do not use condoms). Intrinsic rewards: the perception of intrinsic rewards was assessed with six items by asking participants about the psychological or physical pleasure of not using a condom (e.g., “I feel good without using a condom when having sex”). Vulnerability: the perception of vulnerability to be infected with STI/HIV was assessed using two items regarding the likelihood of acquisition of STI/HIV (e.g., “It is possible to be infected with STI/HIV without using a condom”). Severity: the perception of severity to be infected with HIV was assessed using five items (e.g., “When someone acquires HIV, his/her life is over”). Response efficacy: two items were used to assess perceptions of the effectiveness of using a condom (e.g., “Using a condom can reduce the likelihood of being infected with STI/HIV”). Self-efficacy: self-efficacy towards using a condom was measured using 11 items (e.g., “Even though my sexual partner does not want to use a condom, I can persuade him to use”). Response costs: ten items were used to assess possible barriers or negative consequences of using a condom (e.g., “If I insist on using a condom with my sexual partner, he may think that I have STI”). All items to assess the PMT constructs (except extrinsic rewards) had a 4-point response option (strongly disagree, disagree, agree, strongly agree). The items measuring extrinsic rewards had a 5-point response option (none, few, some, most, all).
Table 1.
List of items and Cronbach alpha of seven PMT constructs regarding condom use
| Items | Alpha |
|---|---|
| Factors affecting maladaptive response probability | |
| Extrinsic rewards (4 items) | 0.85 |
| 1.How many of FSWs use condom with stable sexual partners | |
| 2.How many of FSWs use condom with casual sexual partners | |
| 3.How many of FSW think that FSWs should use condom consistently | |
| 4.How many of FSWs think that FSWs should use condom when having sex with a person known well | |
| Intrinsic rewards (6 items) | 0.66 |
| 1.I feel better without using a condom when having a sexual intercourse | |
| 2.It is easy to have organism without use a condom when having a sexual intercourse | |
| 3.I feel more close to sexual partner without use a condom | |
| 4.I need to use a condom when having a sexual intercourse (R) | |
| 5.I feel safe for using a condom when having a sexual intercourse (R) | |
| 6.I need to use a condom when having sex with a person known well (R) | |
| Vulnerability (2 items) | 0.90 |
| 1.One can acquire STI when having a sexual intercourse without using a condom | |
| 2. One can acquire HIV when having a sexual intercourse without using a condom | |
| Severity (5 items) | 0.88 |
| 1.If a person has HIV, his/her whole life is over | |
| 2.If a person has STI, he/she is difficult to face family members | |
| 3.If a person has HIV, he/she will lose job | |
| 4.If a person has HIV, he/she will lose friends | |
| 5.If a person has HIV, he/she is difficult to face family members | |
| Factors affecting adaptive response probability | |
| Response efficacy (2 items) | 0.93 |
| 1.Using a condom can reduce the risk of contracting STI | |
| 2. Using a condom can reduce the risk of contracting HIV | |
| Self-efficacy (11 items) | 0.92 |
| 1.It is my responsibility to use a condom when having a sexual intercourse | |
| 2.I can use condoms consistently | |
| 3.I believe that I can talk about condom use with my sexual partners | |
| 4.I believe that I can put the condom on the penis of my sexual partner | |
| 5.My sexual partner and myself know where to obtain the condoms | |
| 6.If my sexual partner does not want to use a condom, I can persuade him to use | |
| 7.If my sexual partner refuse to use a condom, I will not have sex with him | |
| 8.I know how to use a condom correctly | |
| 9.Even though I drink alcohol or use drugs, I still can insist on use a condom | |
| 10.Even though my sexual partner drinks alcohol or use drugs, I still can insist on use a condom | |
| 11.Whether using a condom or not is men's issue, women cannot make the decision (R) | |
| Response costs (10 items) | 0.88 |
| 1.If I insist on using a condom, the atmosphere of having sex could be interrupted | |
| 2.If I insist on using a condom, my sexual partner's mood could be interrupted | |
| 3.Using a condom can reducing sexual pleasure | |
| 4.Using a condom when having a sexual intercourse, a condom is often broken, leaked, or slipped off | |
| 5.If a police finds me carrying a condom, I could have trouble | |
| 6.If my family members find me carrying a condom, I could have trouble | |
| 7.If I insist on using a condom, my sexual partner could think that I have STI | |
| 8.Buying condoms in a store is embarrassed. | |
| 9.I feel uncomfortable for carrying a condom with me | |
| 10.It is troublesome to use a condom during sexual intercourse |
(R): item with reverse coding.
Intention of consistent condom use
We used two items to assess participants’ intention to use condoms consistently in the future with stable or casual sexual partners (“How often will you use condoms with your stable [casual] sexual partners in the future?”). Each of these two items had a 5-point response option from “not use at all” to “always use a condom”. These two items were dichotomized and used as dependent variables in the logistic regression analyses. If a participant answered that she will always use condoms with her sexual partners in the future, she was considered having intention of consistent condom use; otherwise, she was considered having no intention of consistent condom use.
Consistent condom use
Participants were asked about the frequency of condom use during the most recent three sexual episodes with stable sexual partners and casual sexual partners. These two questions had a 4-point response option (none, once, twice, three times). These items were dichotomized and used as dependent variables in the logistic regression analyses. If a participant answered that she used condoms three times during the last three sexual episodes, she was considered using condoms consistently; otherwise, she was considered using condoms inconsistently.
Data analyses
First, descriptive statistics are calculated to show the frequency distribution of demographic characteristics, condom use intentions and condom use. Second, Pearson Product Moment Correlations were employed to assess the associations between PMT factors and consistent condom use intention and consistent condom use. Third, four multivariate logistic regression models were performed to further examine the association of PMT constructs with consistent condom use intention and behavior separately with stable sexual partners (Models 1 and 2) and with casual sexual partners (Models 3 and 4), while controlling for key demographic characteristics (age, education, residence, and income). Adjusted odds ratio (aOR) and their 95% confidence intervals (95% CI) were calculated for each predictor in the regression models.
Results
The mean age of FSWs was 26 years (SD = 7.02). About 78.6% of FSWs reported having stable sexual partners. FSWs who were Han ethnicity accounted for 79.4% of the sample (Table 2). FSWs who were rural residents accounted for 65.5%. The percentage of FSWs who completed at least high school education was 32.3%. The average monthly income of FSWs was about 3,772 Yuan (approximately US$ 600, with an exchange rate one U.S. $ = 6.3 Yuan at the time of survey).
Table 2.
Sociodemographic characteristics, intention of condom use, and condoms use in the last three sexual episodes, n (%)
| Total sample | FSWs had stable | |
|---|---|---|
| N | 700 | 550 |
| Age, mean (SD) | 25.7 (7.02) | 26.2 (7.28) |
| 16~19 | 123 (17.6) | 92 (16.7) |
| 20~29 | 403 (57.6) | 309 (56.2) |
| 30~39 | 132 (18.9) | 110 (20.0) |
| ≥ 40 | 42 (6.0) | 39 (7.1) |
| Ethnicity | ||
| Han | 556 (79.4) | 435 (79.1) |
| Minority | 144 (20.6) | 115 (20.9) |
| Residence | ||
| Urban | 241 (34.5) | 178 (32.5) |
| Rural | 457 (65.5) | 370 (67.5) |
| Education | ||
| ≤ Elementary school | 82 (11.7) | 69 (12.6) |
| Middle school | 391 (55.9) | 315 (57.4) |
| ≥ High school | 226 (32.3) | 165 (30.1) |
| Monthly income (Yuan), mean (SD) | 3772.4 (3204.5) | 3673.5 (3330.6) |
| ≤ 1000 | 51 (7.4) | 42 (7.8) |
| 1001~2000 | 176 (25.6) | 149 (27.6) |
| 2001~3000 | 196 (28.5) | 150 (27.8) |
| 3001~4000 | 86 (12.5) | 73 (13.5) |
| 4001~5000 | 92 (13.4) | 65 (12.1) |
| ≥ 5001 | 87 (12.6) | 60 (11.1) |
| Intention of condom use | ||
| Not use at all | 19 (2.8) | 84 (15.3) |
| Occasionally | 22 (3.3) | 118 (21.5) |
| Sometimes | 28 (4.1) | 115 (21.0) |
| Often | 64 (9.5) | 99 (18.1) |
| Always | 542 (80.3) | 132 (24.1) |
| Condoms use in the last three sexual episodes | ||
| None | 68 (10.1) | 203 (37.2) |
| Once | 52 (7.7) | 75 (13.8) |
| Twice | 73 (10.8) | 85 (15.6) |
| Three times | 480 (71.3) | 182 (33.4) |
SD stands for standard deviation.
About 80.3% of FSWs reported intention of consistent condom use. Rate of intention of consistent condom use (24.1%) with stable sexual partners was lower than that with casual sexual partners (80.3%). Rate of consistent condom use during the last three sexual episodes was 33.4% with stable sexual partners and 71.3% with casual sexual partners (Table 2).
The means of seven PMT constructs in terms of intention of consistent condom use and consistent condom use behavior with stable sexual partners or casual sexual partners are shown in Table 3. Four of the constructs (extrinsic rewards, intrinsic rewards, vulnerability, and self-efficacy) were significantly associated with intention of consistent condom use with stable sexual partners. Compared with those who had no intention of consistent condom use, FSWs who reported an intention of consistent condom use had lower extrinsic rewards (7.04 vs. 9.07, p < .01), lower intrinsic rewards (11.49, vs. 12.92, p < .01), higher perceived vulnerability (6.19 vs. 5.82, p < .05), and higher condom use self-efficacy (35.09 vs. 32.85, p < .05).
Table 3.
Association analyses of PTM constructs to intention of consistent condom use and consistent condom use in the last three sexual episodes with stable sexual partners or casual sexual partners, mean (SD)
| PMT factors | Total | With stable sexual partners |
With casual sexual partners |
||||||
|---|---|---|---|---|---|---|---|---|---|
| Intention of consistent condom use | Consistent condom use | Intention of consistent condom use | Consistent condom use | ||||||
| no | yes | no | yes | no | yes | no | yes | ||
| Extrinsic rewards | 8.47 (4.00) | 9.07 (3.86) | 7.04 (3.68)** | 9.17 (3.83) | 7.35 (3.83)** | 11.02 (4.05) | 7.77 (3.74)** | 10.51 (4.25) | 7.60 (3.60)** |
| Intrinsic rewards | 12.49 (2.79) | 12.92 (2.60) | 11.49 (2.88)** | 12.99 (2.56) | 11.72 (2.92)** | 13.44 (2.54) | 12.21 (2.79)** | 13.21 (2.52) | 12.14 (2.84)** |
| Vulnerability | 5.93 (1.48) | 5.82 (1.51) | 6.19 (1.45)* | 5.84 (1.50) | 6.06 (1.51) | 5.44 (1.49) | 6.07 (1.46)** | 5.61 (1.53) | 6.08 (1.44)** |
| Severity | 13.73 (3.23) | 13.56 (3.32) | 13.87 (3.08) | 13.52 (3.24) | 13.89 (3.36) | 13.19 (3.29) | 13.91 (3.20)* | 13.15 (3.16) | 14.03 (3.23)** |
| Response efficacy | 6.15 (1.37) | 6.09 (1.35) | 6.21 (1.42) | 6.07 (1.36) | 6.24 (1.42) | 5.89 (1.35) | 6.23 (1.35)* | 5.95 (1.44) | 6.26 (1.31)** |
| Self-efficacy | 33.54 (5.75) | 32.85 (5.76) | 35.09 (4.61)** | 32.72 (5.58) | 34.79 (5.43)** | 30.62 (5.40) | 34.32 (5.60)** | 31.62 (5.63) | 34.37 (5.61)** |
| Response costs | 22.20 (5.09) | 22.28 (5.09) | 22.29 (4.73) | 22.54 (5.09) | 21.73 (4.89) | 23.35 (4.79) | 21.91 (5.15)** | 23.03 (4.96) | 21.85 (5.15)** |
p < .05
p < .01.
SD stands for standard deviation.
Three of the constructs (extrinsic rewards, intrinsic rewards, and self-efficacy) were significantly associated with consistent condom use in the last three sexual episodes with stable sexual partners. Compared with those who reported using condoms inconsistently in the last three sexual episodes, FSWs with stable sexual partners who reported using condoms consistently in the last three sexual episodes reported having lower extrinsic rewards (7.35 vs. 9.17, p < .01), lower intrinsic rewards (11.72 vs. 12.99, p < .01), higher self-efficacy (34.79 vs. 32.72, p < .01).
All seven PMT constructs were significantly associated with intention of consistent condom use with casual sexual partners among FSWs. Compared with those who had no intention of consistent condom use with casual sexual partners, FSWs who reported an intention of consistent condom use with casual sexual partners had lower extrinsic rewards (7.77 vs. 11.02, p < .01), lower intrinsic rewards (12.21, vs. 13.44, p < .01), higher perceived vulnerability (6.07 vs. 5.44, p < .01), higher perceived severity (13.91 vs. 13.19, p < .05), higher response efficacy (6.23 vs. 5.89, p < .05), higher self-efficacy (34.32 vs. 30.62, p < .01), and lower response costs (21.91 vs. 23.35, p < .05). All seven PMT constructs were significantly associated with consistent condom use with casual sexual partners. Compared with those who reported using condoms inconsistently with casual sexual partners in the last three sexual episodes, FSWs who reported using condoms consistently in the last three sexual episodes reported lower extrinsic rewards (7.60 vs. 10.51, p < .01), lower intrinsic rewards (12.14, vs. 13.21, p < .01), higher perceived vulnerability (6.08 vs. 5.61, p < .01), higher perceived severity (14.03 vs. 13.15, p < .01), higher response efficacy (6.26 vs. 5.95, p < .01), higher self-efficacy (34.37 vs. 31.62, p < .01), and lower response costs (21.85 vs. 23.03, p < .01).
Multivariate logistic regression analyses results in Table 4 showed that while controlling for age, ethnicity, residence, education and income, extrinsic rewards (aOR = 0.892 [95% CI: 0.833~0.955]) and intrinsic rewards (aOR = 0.862 [95% CI: 0.786~0.945]) remained being predictive of intention of consistent condom use with stable sexual partners, (Model 1).
Table 4.
Multivariate logistic regression analyses of consistent condom use intention and behavior
| Factors | With stable sexual partners |
With casual sexual partners |
||
|---|---|---|---|---|
| Model 1: Intention of consistent condom use | Model 2: Consistent condom use | Model 3: Intention of consistent condom use | Model 4: Consistent condom use | |
| aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | aOR (95% CI) | |
| Age | 1.015(0.984~1.047) | 1.022 (0.993~1.051) | 1.001(0.970~1.033) | 1.016 (0.987~1.045) |
| Ethnicity | 0.834 (0.498~1.399) | 0.487 (0.307~0.774)** | 0.662 (0.370~1.184) | 0.979 (0.615~1.561) |
| Residence | 0.828(0.522~1.311) | 0.795 (0.519~1.215) | 1.007 (0.630~1.610) | 0.945 (0.633~1.410) |
| Education | 1.083 (0.809~1.448) | 1.089 (0.836~1.419) | 1.088 (0.804~1.472) | 0.834 (0.643~1.082) |
| Income | 0.923 (0.796~1.072) | 0.985 (0.859~1.129) | 1.118 (0.955~1.309) | 1.122 (0.982~1.282) |
| Extrinsic rewards | 0.892 (0.833~0.955)** | 0.906 (0.854~0.961)** | 0.860 (0.815~0.909)** | 0.866 (0.823~0.910)** |
| Intrinsic rewards | 0.862 (0.786~0.945)** | 0.903 (0.830~0.983)* | 0.972 (0.879~1.074) | 0.973 (0.895~1.058) |
| Vulnerability | 1.089 (0.910~1.303) | 0.942 (0.800~1.108) | 1.148 (0.959~1.374) | 1.068 (0.912~1.252) |
| Severity | 0.950 (0.876~1.029) | 0.998 (0.927~1.075) | 0.989 (0.900~1.086) | 1.042 (0.964~1.127) |
| Response efficacy | 0.898 (0.734~1.097) | 0.947 (0.789~1.136) | 0.836 (0.668~1.048) | 0.868 (0.717~1.051) |
| Self-efficacy | 1.040 (0.985~1.097) | 1.050 (1.000~1.103)* | 1.099 (1.041~1.161)** | 1.049 (1.002~1.099)* |
| Response costs | 1.036 (0.992~1.083) | 0.992 (0.953~1.033) | 0.936 (0.892~0.981)** | 0.955 (0.918~0.994)* |
p < .05
p < .01.
Results of regression Model 2 showed that extrinsic rewards (aOR = 0.906 [95% CI: 0.854~0.961]), intrinsic rewards (aOR = 0.903 [95% CI: 0.830~0.983]), and condom use self-efficacy (aOR = 1.050 [95% CI: 1.000~1.103]) remained being predictive of consistent condom use in the last three sexual episodes with stable sexual partners. In Model 2, ethnicity was significantly associated with consistent condom use with stable sexual partners (aOR = 0.487, [95% CI: 0.307~0.774]).
Results of regression Model 3 in Table 4 indicated that extrinsic rewards (aOR = 0.860 [95% CI: 0.815~0.909]), condom use self-efficacy (aOR = 1.099 [95% CI: 1.041~1.161]), and response costs (aOR = 0.936 [95% CI: 0.892~0.981]) remained being predictive of intention of consistent condom use with casual sexual partners.
Likewise, extrinsic rewards (aOR = 0.866 [95% CI: 0.823~0.910]), condom use self-efficacy (aOR =1.049 [95% CI: 1.002~1.099]), and response costs (aOR =0.955 [95% CI: 0.918~0.994]) remained being predictive of consistent condom use in the last three sexual episodes with casual sexual partners (Model 4).
In summary, multivariate logistic regression analyses indicated that the perception of extrinsic rewards was negatively associated with consistent condom use intention and behavior with both stable sexual partners and casual sexual partners. The perception of intrinsic rewards was negatively associated with consistent condom use intention and behavior with stable sexual partners but not with casual sexual partners. Condom use self-efficacy was positively associated with both consistent condom use intention and behavior with casual sexual partners and positively associated with only condom use behavior with stable sexual partners. Perception of response costs was negatively associated with consistent condom use intention and behavior with casual sexual partners but not with stable sexual partners.
Discussion
Results in the current study showed that FSWs had a high level of intention of consistent condom use and they were more likely to use condoms consistently with casual sexual partners than with stable sexual partners. This result is consistent with previous studies among FSWs (Wu et al., 2012). The rate of consistent condom use during the last three sexual episodes with casual sexual partners was higher in the current study (71%) than in other studies. For instance, researchers in India reported that the rate of consistent condom use during every sexual act with regular and occasional sexual clients was 9% and 16% (Medhi et al., 2012). Others who studied male clients of FSWs in the Philippines indicated the rate of consistent condom use was 22% (Regan & Morisky, 2012). These differences may attribute to some methodological differences between different studies. Because we measured the consistent condom use during the last three sexual episodes, recall bias may be low. The study of Regan and Morisky measured the consistent condom use based on the report from clients of FSWs. There may be differences in reporting of consistent condom use by men compared to women.
All seven PMT constructs assessed in this study had adequate internal consistency estimates (e.g., Cronbach alpha ranged from 0.66 to 0.93). Bivariate statistical analyses indicated that all seven PMT constructs were significantly associated with intention of consistent condom use and consistent condom use with casual sexual partners.
Results of regression analyses in this study indicated that self-efficacy is a critical predictor for consistent condom use with both stable sexual partners and casual sexual partners. The result of the positive association between condom use self-efficacy and intention and consistent condom use is consistent with other studies conducted among early adolescents, men who have sex with men (MSM), FSWs (Chen et al., 2012; Teng & Mak, 2011; Zhao, Wang, Fang, Li, & Stanton, 2008). Condom use self-efficacy includes two components: communication skills and emotional control (Baele, Dusseldorp, & Maes, 2001). FSWs who have high level of condom use self-efficacy may have good communication skills with their sexual partners regarding condom use. FSWs with high level of condom use self-efficacy may be able to control their emotions during sex. FSWs who have high level of condom use self-efficacy may be more confident in persuading their partners (both stable sexual partners and casual sexual partners) to use a condom.
Results in the current study indicated that the perception of extrinsic rewards is predictive of FSWs’ intention of consistent condom use and behavior of consistent condom use with both stable sexual partners and casual sexual partners. The formation of extrinsic rewards perception of consistent condom use is based on the social norm and peer influence. Venue-based interventions to establish the atmosphere of health benefits of consistent condom use and negative health consequences of inconsistent condom use among this vulnerable group need to be strengthened.
The perception of intrinsic rewards is predictive for consistent condom use intention and behavior with stable sexual partners. When having sex with stable sexual partners, FSWs may be concerned more about the intrinsic rewards of not using a condom – advantage of not using a condom (e.g., not using a condom may be good for the sexual relationship and emotional support, good for improving sexual pleasure). These kinds of intrinsic rewards may not often be produced when FSWs have sex with casual sexual partners since FSWs’ purpose of having sex with casual sexual partners is not seeking or building a relationship.
Another factor that may affect intention of consistent condom use and consistent condom use is the response costs. Response costs played an important role on intention and behavior of consistent condom use with casual sexual partners. When having sex with casual sexual partners, FSWs may be concerned more about the costs and negative consequences of using or insisting to use a condom (e.g., reduced income, possible partner violence). When having sex with stable sexual partners, FSWs may also be concerned about the costs and negative consequences of using a condom. The costs may be possible partner violence, but not reduced income. Therefore, it can be surmised that FSWs may be more concerned about income rather than partner violence since economic factors play an important role in shaping FSWs’ sexual decision-making in China (Wu et al., 2012). Economic factor also shows the powerful influence in other Asian countries. For instance, despite the prevalent partner violence against FSWs in Thai society, more than 86% of FSWs in Thai society would engage in unsafe sex (Nemoto, Iwamoto, Sakata, Perngparn, & Areesantichai, 2013).
PMT constructs of vulnerability, severity and response efficacy did not show significant associations with condom use with either casual or stable partners in the current study. It is possible that many FSWs may be unaware their susceptibility to infections and they may therefore be less concerned about the severity of the infections. Sexual and reproductive health education programs and STI/HIV prevention programs have been implemented for many years in China (Lou, Wang, Shen, & Gao, 2004; Wang, 2007). About 52% of the FSWs in this study were never married. Compared with married people, unmarried people were less likely to have access to contraceptive service (Zhang et al., 2004). In addition, some FSWs did not perceive the adverse outcomes of unprotected sex. Moreover, the high prevalence of STI among FSWs may result in the misconception that STI was common and acceptable among FSWs. This misconception may lead FSWs a low perceived severity of STI/HIV. Moreover, the underestimate of condoms’ efficacy in preventing STI/HIV among FSWs may be because of the lack of knowledge about the role of condom in preventing STI/HIV or may be because of their previous experiences of condom failure.
Many researchers have examined the utilities of various PMT constructs in predicting condom use intention and behavior in different culture and different populations (Boer & Mashamba, 2005; Lwin, Stanaland, & Chan, 2010; Stanton et al., 1996). In a study among black adolescents in South Africa (Boer & Mashamba, 2005), only the perception of response efficacy to use condoms was found to be related to condom use intention. In a sample of urban adolescent African-Americans from a large eastern metropolis in the US, perceptions of response efficacy and self-efficacy predicted actual and intended condom use while perceptions of severity and vulnerability were not significant predictors (Stanton et al., 1996). Researchers who studied adolescents in Vietnam where the culture is similar to the culture of China did not find the significant differences for the constructs of extrinsic and intrinsic rewards (Kaljee et al., 2005). The different results from the study of Kaljee et al. may be because the population in their study had an extremely low rate of engagement in sex (Kaljee et al., 2005). One study conducted among male clients of FSWs in the Philippines indicted that perceptions of HIV/AIDS severity and response efficacy toward condoms were significant factors of consistent condom use (Regan & Morisky, 2012). Chinese society is a patriarchal society and condom use is dominated by male sexual partners. To examine this particular culture has impact on PMT constructs in the population of FSWs, cross-cultural research is needed in future.
There are some potential limitations in this study. First, the sample in this study was a convenient sample. Subject-selection bias may exist since it is possible that some FSWs who had STDs or HIV infection might not be willing or not be able to participate in the survey. Second, data used in this study were collected through self-report, therefore, recall bias may exist. Third, this sample was selected in the commercial sex venues of two cities and data from this sample may not be extrapolated to FSWs in other areas. In addition, because this sample was selected from a province with a high HIV prevalence, the sample of the current study may not be representative of FSWs who are from areas with low HIV prevalence. Fourth, as this study is based on cross-sectional data, all the relationships between perceptions and intentions and behaviors were associational. The causal relationships need to be explored using longitudinal data.
Despite these potential limitations, the current study is one of the first efforts to examine the predictability of PMT constructs on consistent condom use among FSWs in China. The findings of the current study have several implications for the future HIV prevention programs among FSWs in China and other low- and middle-income countries. First, PMT may be an appropriate theoretical model to guide the assessment and development of intervention aiming to promote condom use among FSWs in China and other resource-poor countries. Second, interventions need to reduce FSWs’ perceptions of rewards for non-consistent condom use, to reduce FSWs’ perception of response costs for using a condom, and to increase condom use self-efficacy among FSWs. FSWs’ condom use self-efficacy needs to be increased through improving FSWs’ communication skills and emotional control skills with both stable sexual partners and casual sexual partners. Third, interventions promoting condom use should reduce response costs of condom use through correcting misconception regarding the association between condom use or STD/HIV infection. Fourth, it is necessary to consider the social norm and peers’ influence in conducting a condom promotion intervention. The intervention needs to engage not only FSWs, but also other parties in their life and work, such as partners, clients, peers, gatekeepers, health care providers, and law reinforcement agencies.
Figure 1.
Hypothesized model of condom use based on PMT
Acknowledgements
We wish to thank all of the participants and the local researchers and staff for their contributions for this study. We are grateful for Dr Iqbal Shah's valuable comments for our manuscript. We thank Joanne Zwemer for assistance with manuscript preparation.
Funding
The study is supported by NIH research grant (R01AA018090) from the National Institute for Alcohol Abuse and Alcoholism (NIAAA). The content of paper is solely the responsibility of the authors and does not necessarily represent the official views of NIH.
Contributor Information
Liying Zhang, Prevention Research Center, Department of Pediatrics, Wayne State University School of Medicine, Detroit, Michigan, USA.
Xiaoming Li, Prevention Research Center, Department of Pediatrics, Wayne State University School of Medicine, Detroit, Michigan, USA.
Yuejiao Zhou, Guangxi Center for Disease Control and Prevention, Nanning, China.
Danhua Lin, Institute of Developmental Psychology, Beijing Normal University, Beijing, China.
Shaobing Su, Institute of Developmental Psychology, Beijing Normal University, Beijing, China.
Chen Zhang, Department of Epidemiology, Vanderbilt University, Nashville, Tennessee, USA.
Bonita Stanton, Prevention Research Center, Department of Pediatrics, Wayne State University School of Medicine, Detroit, Michigan, USA.
References
- Baele J, Dusseldorp E, Maes S. Condom use self-efficacy: Effect on intended and actual condom use in adolescents. Journal of Adolescent Health. 2001;28(5):421–431. doi: 10.1016/s1054-139x(00)00215-9. [DOI] [PubMed] [Google Scholar]
- Boer H, Mashamba MT. Psychosocial correlates of HIV protection motivation among black adolescents in Venda, South Africa. AIDS Education and Prevention. 2005;17(6):590–602. doi: 10.1521/aeap.2005.17.6.590. [DOI] [PubMed] [Google Scholar]
- Boittin ML. New Perspectives from the Oldest Profession: Abuse and the Legal Consciousness of Sex Workers in China. Law & Society Review. 2013;47(2):245–278. [Google Scholar]
- Cai Y, Shi R, Shen TA, Pei B, Jiang XQ, Ye XX, Xu G, Li SH, Huang H, Shang ML. A study of HIV/AIDS related knowledge, attitude and behaviors among female sex workers in Shanghai China. Bmc Public Health. 2010;10:377. doi: 10.1186/1471-2458-10-377. doi: 10.1186/1471-2458-10-377. [DOI] [PMC free article] [PubMed] [Google Scholar]
- China Ministry of Health . 2012 China AIDS Response Progress Report. China Ministry of Health; Beijing: 2012. [Google Scholar]
- Fang X, Li X, Yang H, Hong Y, Zhao R, Dong B, Liu W, Zhou Y, Liang S, Stanton B. Profile of female sex workers in a Chinese county: does it differ by where they came from and where they work? World health & population. 2007;9(1):46–64. doi: 10.12927/whp.2007.18695. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Kaljee LM, Genberg B, Riel R, Cole M, Tho le H, Thoa le TK, Stanton B, Li X, Minh TT. Effectiveness of a theory-based risk reduction HIV prevention program for rural Vietnamese adolescents. AIDS Education and Prevention. 2005;17(3):185–199. doi: 10.1521/aeap.17.4.185.66534. [DOI] [PubMed] [Google Scholar]
- Kaufman J. HIV, Sex Work, and Civil Society in China. Journal of Infectious Diseases. 2011;204(Suppl 5):S1218–S1222. doi: 10.1093/infdis/jir538. [DOI] [PubMed] [Google Scholar]
- Lau JTF, Tsui HY, Siah PC, Zhang KL. A study on female sex workers in southern China (Shenzhen): HIV-related knowledge, condom use and STD history. AIDS Care. 2002;14(2):219–233. doi: 10.1080/09540120220104721. [DOI] [PubMed] [Google Scholar]
- Lofberg K, Sharma A, Keli F, Bukusi E, Cohen C. Female sex worker condom use patterns by partner type. Journal of Investigative Medicine. 2004;52(1):S128–S128. [Google Scholar]
- Lu F, Jia YJ, Sun XH, Wang L, Liu W, Xiao Y, Zeng G, Li CM, Liu JB, Cassell H, Chen HT, Vermund SH. Prevalence of Hiv Infection and Predictors for Syphilis Infection among Female Sex Workers in Southern China. Southeast Asian Journal of Tropical Medicine and Public Health. 2009;40(2):263–272. [PMC free article] [PubMed] [Google Scholar]
- Lu F, Wang N, Wu Z, Sun X, Rehnstrom J, Poundstone K, Yu W, Pisani E. Estimating the number of people at risk for and living with HIV in China in 2005: methods and results. Sexually Transmitted Infections. 2006;82(Suppl 3):iii87–iii91. doi: 10.1136/sti.2006.020404. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Lwin MO, Stanaland AJS, Chan D. Using protection motivation theory to predict condom usage and assess HIV health communication efficacy in Singapore. Health Communication. 2010;25(1):69–79. doi: 10.1080/10410230903473540. [DOI] [PubMed] [Google Scholar]
- Macaluso M, Demand MJ, Artz LM, Hook EW. Partner type and condom use. AIDS. 2000;14(5):537–546. doi: 10.1097/00002030-200003310-00009. [DOI] [PubMed] [Google Scholar]
- Medhi G, Mahanta J, Paranjape R, Adhikary R, Laskar N, Ngully P. Factors associated with HIV among female sex workers in a high HIV prevalent state of India. AIDS Care. 2012;24(3):369–376. doi: 10.1080/09540121.2011.608787. [DOI] [PubMed] [Google Scholar]
- Milne S, Sheeran P, Orbell S. Prediction and intervention in health-related behavior: A meta-analytic review of protection motivation theory. Journal of Applied Social Psychology. 2000;30(1):106–143. [Google Scholar]
- National Bureau of Statistics of China . Sixth national census data released. National Bureau of Statistics of China; Beijing: 2012. Retrieved from http://www.stats.gov.cn/was40/gjtjj_detail.jsp?searchword=%C8%CB%BF%DA&channelid=6697&record=18. [Google Scholar]
- Nemoto T, Iwamoto M, Sakata M, Perngparn U, Areesantichai C. Social and cultural contexts of HIV risk behaviors among Thai female sex workers in Bangkok, Thailand. AIDS Care. 2013;25(5):613–618. doi: 10.1080/09540121.2012.726336. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Pack RP, Li X, Stanton BF, Cottrell LA. Psychosocial correlates of dual methods for contraception and STI protection in urban adolescents. ISRN obstetrics and gynecology. 2011;2011:469–610. doi: 10.5402/2011/469610. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Regan R, Morisky D. Perceptions about HIV and condoms and consistent condom use among male clients of commercial sex workers in the Philippines. Health Education & Behavior. 2012;40(2):216–222. doi: 10.1177/1090198112446809. [DOI] [PubMed] [Google Scholar]
- Rippetoe PA, Rogers RW. Effects of Components of Protection-Motivation Theory on Adaptive and Maladaptive Coping with a Health Threat. Journal of Personality and Social Psychology. 1987;52(3):596–604. doi: 10.1037//0022-3514.52.3.596. [DOI] [PubMed] [Google Scholar]
- Rogers R. Cognitive and psychological processes in fear appeals and attitude change: A revised theory of protection motivation. In: Cacioppo J, Petty R, editors. Social psychology: A sourcebook. Guilford Press; New York: 1983. pp. 153–176. [Google Scholar]
- Rou KM, Wu ZY, Sullivan SG, Li F, Guan JH, Xu C, Liu W, Liu DH, Ying YP. A five-city trial of a behavioural intervention to reduce sexually transmitted disease/HIV risk among sex workers in China. AIDS. 2007;21(Suppl 8):S95–S101. doi: 10.1097/01.aids.0000304703.77755.c7. [DOI] [PubMed] [Google Scholar]
- Stanton B, Black M, Kaljee L, Ricardo I. Perceptions of sexual behavior among urban early adolescents: translating theory through focus groups. The Journal of Early Adolescence. 1993;13(1):44–66. [Google Scholar]
- Stanton B, Li X, Black MM, Ricardo I, Galbraith J, Feigelman S, Kaljee L. Longitudinal stability and predictability of sexual perceptions, intentions, and behaviors among early adolescent African-Americans. Journal of Adolescent Health. 1996;18(1):10–19. doi: 10.1016/1054-139x(95)00070-9. [DOI] [PubMed] [Google Scholar]
- To S, Tam H, Chu F. A qualitative study of the lived experiences of young Chinese females in condom use negotiations. International Journal of Adolescence and Youth. iFirst article. 2012 Online: doi:10.1080/02673843.2012.692656. [Google Scholar]
- Tucker J, Ren X, Sapio F. Incarcerated sex workers and HIV prevention in China: Social suffering and social justice countermeasures. Social Science & Medicine. 2010;70(1):121–129. doi: 10.1016/j.socscimed.2009.10.010. [DOI] [PubMed] [Google Scholar]
- Wu J, Zhou XL, Lu CY, Moyer E, Wang H, Hong LY, Deng XQ. A qualitative exploration of barriers to condom use among female sex workers in China. Plos One. 2012;7(10):e46786. doi: 10.1371/journal.pone.0046786. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Yang X, Xia G. Gender, work, and HIV risk: determinants of risky sexual behavior among female entertainment workers in China. AIDS Education and Prevention. 2006;18(4):333–347. doi: 10.1521/aeap.2006.18.4.333. [DOI] [PubMed] [Google Scholar]
- Zhang C, Li X, Hong Y, Chen Y, Liu W, Zhou Y. Partner violence and HIV risk among female sex workers in China. AIDS and Behavior. 2011;16(4):1020–1030. doi: 10.1007/s10461-011-9968-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zhou C, Rou K, Dong W, Wang Y, Dong W, Zhou Y, Chen X, Jia M, Liu W, Zheng J, Ma Y, Li Y, Wu Z. High prevalence of HIV and syphilis and associated factors among low-fee female sex workers in mainland China: a cross-sectional study. BMC Infectious Diseases. 2014;14(1):225–233. doi: 10.1186/1471-2334-14-225. [DOI] [PMC free article] [PubMed] [Google Scholar]
- Zou HC, Xue H, Wang XF, Lu DM. Condom use in China: prevalence, policies, issues and barriers. Sexual Health. 2012;9(1):27–33. doi: 10.1071/SH11021. [DOI] [PubMed] [Google Scholar]