62254 Turkey: HIV and Risky Behavior among Mobile Workers Ethan Yeh 63% of new HIV cases in the last five years. Injecting Key Messages1 drug use, on the other hand, accounted for only 2% of The number of identified HIV cases is on the cases during the same period. Since HIV remains rise in Turkey, mostly due to heterosexual concentrated among most-at-risk populations (MARPs) transmission (63% of new cases in the last five of sex workers, men who have sex with men (MSM), years). Since the epidemic in Turkey remains and injecting drug users (IDUs), these new HIV cases concentrated among most-at-risk populations due to heterosexual transmission are likely through sex (MARPs), these heterosexual infections are likely with sex workers. due to transmission through sex work. While several studies have been conducted in Turkey on Adult mobile workers in Turkey - sailors, sex workers, little evidence exists on their clients, the truck drivers, construction and tourism workers - demand side of the commercial sex market. Research frequently engage in risky sex. On average, 35% from other countries shows that mobile workers (often of mobile workers engage in frequent sex with sex working-age males who spend significant time away workers, and only 45% use condoms. from home) are more likely to have multiple sexual partners, including sex workers, and be HIV positive.2 Factors strongly associated with engaging in For example, a World Bank study in Georgia found that risky sex include: alcohol use and time away from 24% of sailors and 52% of truck drivers reported having home. Frequent alcohol use is associated with a sex with sex workers at least once a week.3 2.5 time increase in the likelihood of engaging in risky sex. Those workers who, on average, spend Figure 1: Identified HIV and AIDS Cases in five months or more away from home are also Turkey, 1985-2009 34% more likely to engage in risky sex. 600 Greater HIV knowledge is not associated with 500 having less risky sex, suggesting that knowledge 400 and information alone are not enough for 300 behavioral change. AIDS 200 HIV+ 100 HIV Cases Are Increasing in Turkey 0 Reported HIV prevalence in Turkey has historically been low - less than 0.2% adult HIV prevalence according to the latest UNAIDS and Turkish Ministry of Health Source: Turkey Ministry of Health, 2009 (MOH) figures. However, new HIV cases have been increasing in recent years (see Figure 1). Most new cases 2 See, for example: World Bank. 2009. “Knowledge, Attitudes, and Behavior Related to HIV/AIDS are due to heterosexual transmission, accounting for Among Transport Sector Workers”; Kulis et al. 2009. “Truck Drivers and Casual Sex: An Inquiry into the Potential Spread of HIV/AIDS in the Baltic Region”; Botros et al. 2009. “HIV Infection and Associated Risk Factors Among Long-Distance Truck Drivers Travelling Through Azerbaijan.” 1 This Knowledge Brief is derived from the report: World Bank; Hacettepe Training, Research and International Journal of STD & AIDS 20:477-482. Services Foundation; UN HIV/AIDS Joint Programme in Turkey; Turkey Ministry of Health. 2010. 3 World Bank. 2008. “Knowledge, Attitudes and Behavior Related to HIV/AIDS among Transport “Risky Business? HIVKnowledge, Attitudes, and Behavior among At-Risk Mobile Workers in Turkey.” Sector Workers: A Case Study of Georgia.” ECA Knowledge Brief Study on HIV among Mobile Workers Knowledge and Awareness In order to understand the behaviors and risks among Given that mobile workers in Turkey engage in frequent mobile workers in Turkey, the World Bank worked in risky sex, what was their current level of HIV collaboration with the Hacettepe Training, Research and knowledge? The study found that 94% of workers had Services Foundation and UN HIV/AIDS Joint heard of HIV/AIDS, indicating high awareness of the Programme in Turkey, at the request of the Turkey virus and disease. Workers were generally aware that Ministry of Health, to conduct the first study in Turkey HIV or AIDS exists and they understood the basic related to HIV among mobile workers. In February- principles of how it is spread (through unprotected sex, March 2010, quantitative and qualitative data were injecting drugs, not using condoms). However, critical collected from 1,239 mobile workers in four of the myths and misconceptions about HIV and AIDS were fastest growing sectors of Turkey’s economy: maritime prevalent among workers. For example, 50% of transportation, road transportation, construction, and respondents believed that HIV/AIDS can be cured, and tourism. These workers were randomly selected from more than half believed that the disease could be four provinces with high estimated HIV prevalence and contracted by sharing food and drinks with infected large estimated populations of mobile workers: Istanbul, people. Only 37% of respondents understood that Izmir, Antalya, and Trabzon. abstinence can protect against HIV infection, and only 39% knew that HIV could not be contracted from Risk Factors mosquito bites. Finally, knowledge of their own HIV status was low among workers, as only 8-13% of the The study found that the primary HIV-risk behavior respondents had been tested for HIV (with the exception which mobile workers engage in is unprotected sex. of sailors where 35% had been tested for HIV, primarily Injecting drug use is not common, with only 1.1% of due to mandatory testing by sailing companies). workers reporting ever injecting drugs. Instead, mobile workers, particularly sailors and truck drivers, engage in Factors Related to Risky Sex frequent sex with multiple partners and with sex workers: 44% of sailors and 42% of truck drivers What are the factors driving mobile workers in Turkey reported having two or more sexual partners in the three to engage in risky sex? Factors associated with risky sex months prior to the study period, compared with 23% of were examined and two factors, in particular, appeared construction workers and 30% of tourism workers. In to be strongly associated with risky sex. addition, 35% of all workers said they engaged in occasional or more frequent sex with sex workers, with First, mobile workers spend considerable time away the highest proportion among sailors (45%) and truck from home, which provides them with opportunities to drivers (38%). Interestingly, the sex with sex workers engage in casual and paid sex. Over 50% of workers did not occur primarily while workers were abroad. The surveyed had spent at least five months away from home frequency of sex with sex workers was similar for in the year prior to the study. Sailors spent the most time workers at home (16%), workers traveling in Turkey away from home on average, whereas truckers and (15%), and workers traveling abroad (17%). construction workers were more likely to spend shorter Unfortunately, only 45% of all workers reported using periods away from home. Tourism workers, on the other condoms during their last sexual encounter with a sex hand, showed a bimodal distribution where they either worker in the period immediately preceding the study. spent very little time away from home or were gone for almost the entire year (see Figure 2). All worker types While the rates of sex by adult mobile workers with sex traveled quite frequently, either within or outside of workers were similar to those in other countries (for Turkey. Sailors and truck drivers were the primary example, 30% in South and Southeast Asia, 21% in worker types that traveled abroad in the previous year, Brazil), the low condom usage rates are a serious with the top three country destinations being Russia concern. Studies in other countries have found much (34.7%), Italy (18.0%) and Romania (16.8%). Overall, higher rates of condom usage--such as, 77% among these workers were extremely mobile, with 97% of them truck drivers and 83% among sailors in Georgia, 85% spending at least one month away, and 87% spending at among sailors and 75% among construction workers in least two months away from home. Croatia, and 75% among construction workers in Vietnam.4 4 World Bank. 2008. “Knowledge, Attitudes and Behavior Related to HIV/AIDS among Transport Workers.” Collegium Antropologicum; Population Council. 2003. “Expanding Workplace HIV/AIDS Sector Workers: A Case Study of Georgia.”; Stulhofer et al. 2006. “HIV/AIDS and Croatian Migrant Prevention Activities for a Highly Mobile Population: Construction Workers in Ho Chi Minh City.” ECA Knowledge Brief Figure 2: Mobile Worker Distribution of Time use condoms, but after drinking, they may forget or Spent Away for Home decide that condoms are not as important. This association has been found among mobile workers - for .15 example, in India, long distance truck drivers who consumed alcohol were found to be 2.71 times more likely to visit sex workers than those who did not drink.6 .1 Density To quantify the relationship of various factors with risky sex, multivariate logit regressions were conducted to .05 control for other background characteristics such as age, income, and education. Once all characteristics were taken into account, only alcohol usage and time away from home remained significantly associated with risky 0 sex (Table 1). The study found that workers who drank 0 3 6 9 12 Time away from home (months) alcohol often were 2.5 times more likely to engage in Sailors Truckers risky sex compared with those who did not. Time away Construction Tourism from home had a much smaller association - workers who spent five additional months away from home were Source: Author’s analysis Note: Kernel density graphs with Epanaechnikov kernel function and 34% more likely to engage in risky sex. Traveling bandwith of 1.2 abroad had little effect, likely because being away from home was the more important factor. Finally, HIV Second, 44% of workers reported drinking alcohol at knowledge had little or no association with risky sex, least occasionally, and among all workers, sailors suggesting that knowledge alone is not enough for HIV reported the highest frequency of alcohol use. Empirical prevention. research suggests that under the influence of alcohol people are more likely to make irrational or risky decisions.5 Prior to drinking, people may fully intend to Table 1: Determinants of Risky Sex (odds ratios) (1) (2) (3) (4) (5) Mobile Workers All Sailors Truck Drivers Construction Tourism Drinks alcohol at least occasionally 2.506*** 2.458*** 2.422*** 3.894*** 2.772** (0.370) (0.726) (0.678) (1.230) (1.347) Months away from home 1.060** 1.114** 1.092* 1.074 1.059 (0.0258) (0.0610) (0.0535) (0.0550) (0.0654) Traveled abroad 1.451* 1.250 1.594 0.616 1.692 (0.312) (0.611) (0.512) (0.323) (1.276) HIV Knowledge Score 0.995 1.006 0.998 0.996 0.974** (0.00406) (0.00812) (0.00759) (0.00995) (0.0124) Province controls Y Y Y Y Y Sector controls Y N N N N Observations 885 263 262 238 117 Source: Author’s analysis Note: Coefficients presented are odds ratios. Robust standard errors in parentheses, *** p<0.01, ** p<0.05, * p<0.1. Regressions include controls for age, marital status, income, and education. 6 Chaturvedi, S., Z. Singh, A. Banerjee, A. Khera, R. Joshi, and D. Dhrubajyoti. 2006. “Sexual Behavior Among Long Distance Truck Drivers,” Indian Journal of Community Medicine 31(3): 153-156. ______________________________ 5 See for example, MacDonald, et al. 2000 “Alcohol Myopia and Condom Use: Can Alcohol Intoxication be Associated with more Prudent Behavior?” Journal of Personality and Social Psychology 78: 605–619; Steele and Josephs. 1990. “Alcohol Myopia: Its Prized and Dangerous Effects,” American Psychologist 45: 921–933. ECA Knowledge Brief Conclusion and Recommendations are also necessary to change behavior. Behavioral change programs that are The study filled a critical gap in the HIV evidence base comprehensive - including HIV education as in Turkey by examining a set of clients on the demand well as other interventions such as counseling, side of the commercial sex market. It is necessary to training, social support programs, and peer recognize, however, that mobile workers are unlikely to groups--are critical. Increasing access to and use be the sole clients of sex workers. Other population of condoms, particularly near the point of sexual groups are also likely to be engaged in risky sex as well. contact, is also important to prevent HIV among A further caveat is that the study collected data only on mobile workers. Condom promotion efforts knowledge, attitudes, and behaviors. Unfortunately, should perhaps target mobile workers at key biomarker data was not collected (for example, HIV locations such as truck stops and ports. testing), so the actual risk of HIV infections and transmission could not be quantified. Although mobile Continue to improve services for voluntary workers engage in risky sex, their actual role in HIV counseling and testing (VCT) and prevention transmission may be lessened by the high proportion of of sexually transmitted infection (STI). Very circumcised workers (99% of workers participating in few workers had been tested for HIV, especially the study were circumcised). voluntarily. VCT services should continue to be expanded to provide opportunities for Still, the evidence presented on knowledge, attitudes, individuals, and their partners, to become aware and behaviors among mobile workers can inform future of their own HIV status. Given that heterosexual policy and research. Three key recommendations can be transmission in Turkey accounts for the majority made: of identified cases in the country, increasing access to and utilization of STI prevention and Better evidence and surveillance. Biomarker treatment services can also be an effective tool data is needed to understand the true risks of in HIV prevention. STI prevention and treatment sexual and drug behaviors, and the findings here services should be expanded and developed for could be combined with HIV testing data to target populations (for example, sex workers) in understand the implications for HIV Turkey. transmission. Integrated bio-behavioral studies should be conducted among at-risk groups, particularly among MSMs and IDUs. Finally, HIV sentinel surveillance (HIV prevalence surveys at regular intervals among at-risk groups - antenatal clinic attendees, STI clinic patients, MARPs, etc.) should be further developed in Turkey. About the Author Ethan Yeh is an Economist in the Human Development Increase behavioral change and condom use. Sector Unit of the Europe and Central Asia Region of Information and knowledge are only necessary the World Bank. conditions for prevention and other interventions “ECA Knowledge Brief” is a regular series of notes highlighting recent analyses, good practices and lessons learned from the development work program of the World Bank’s Europe and Central Asia Region http://www.worldbank.org/eca