CRACK, a smokable and highly addictive form of cocaine, became readily available in the United States in the mid-1980s.1,2 The adverse impact of crack has been most notable in urban areas where economic and physical decay, high unemployment rates, and large concentrations of poor, racial, and ethnic minorities are found.1,3 Crack's pharmacological action produces an intense, short-lived "high" followed by profound depression on withdrawal that may lead to strong craving and compulsive use.4
Ethnographic research suggests that crack addiction forces young women to sell sex directly for crack or money to buy crack.5–7 These data also indicate that "crack for sex" exchanges differ from other types of prostitution because adolescents are disproportionately represented, oral sex is the predominant type of sexual activity, and fees are low (e.g., as little as $5 for a single-use unit "rock" of crack). In addition, ethnographers have observed that the settings for crack for sex exchanges are especially dangerous and degrading (e.g., women working in "crack houses" may be coerced to perform sex publicly with a series of partners).7–8 The crack epidemic has thus resulted in a new, "lower" stratum of sex workers who engage in very high-risk sexual activity and depress the economic value of sex for other sex workers.1,5,7,9
Crack-smoking sex workers have received recent attention because of the alarming prevalence of crack use,2,6 human immunodeficiency virus (HIV) infection,10 and other sexually transmitted diseases (STDs)11,12 in many United States communities. Crack use may increase the risk of HIV and STD on at least three levels. First, crack smokers may begin sex work to finance their addiction.6 Second, crack smokers who have previously engaged in sex work may return to or increase the frequency of sex work to finance their addiction. Third, sex workers under the influence of crack may be less careful when choosing sexual practices or partners.7,8,13–20
Although a few small qualitative studies of street-recruited crack smokers5,7,8,16 and small quantitative studies of crack-smokers seeking health care21,22 have characterized demographic and sexual characteristics of crack smokers, no large quantitative studies have described the sexual practices of actively smoking sex workers who have been recruited from their own neighborhoods. In addition, little is known about the influence of injected drug use on the likelihood of high-risk sexual practices among crack-smoking sex workers. Data was therefore analyzed from the Multicenter Study of HIV Infection and Crack Cocaine9,13 to describe the demographic, sexual, and drug-use characteristics of injecting and noninjecting crack-smoking sex workers recruited from the streets of Miami, New York City, and San Francisco.
The study methods have been detailed elsewhere.9 Briefly, from 1991 through 1992, indigenous outreach workers identified crack smokers and nonsmokers in urban neighborhoods where illicit drug use was common. On the street, they screened participants for eligibility using conversational methods intended to obscure actual criteria. To be eligible, participants had to be 18 to 29 years of age and had to be current, regular smokers of crack cocaine (smoked crack 3 or more days per week in the preceding 30 days) or had to have never smoked crack. Persons who smoked less frequently or had last smoked more than 30 days before interview or were too "high" to interview were not eligible. Outreach workers reviewed key elements of informed consent with eligible participants and directed them to storefront sites where interviewers reviewed eligibility criteria and detailed informed consent. Interviewers questioned study volunteers about demographic characteristics, drug use, sexual activity, STDs, and drug and medical treatment using a 40-minute standardized questionnaire. Counselors provided pretest counseling and antibody testing for HIV (with enzyme immunoassay and Western blot), syphilis (with fluorescent treponemal antibody absorption tests), and herpes simplex type 2 [HSV-2], (with type-specific Western blot for antibody to glycoprotein G), referrals for medical, social and drug-treatment services, appointments for posttest counseling, and $10- to $15-stipends.
We defined crack-smoking sex workers as current, regular smokers of crack cocaine who had sold sex for money or drugs at least once in the 30 days before interview. Crack-smoking sex workers were further classified as current injectors (injected in the 30 days before interview) or noninjectors (never injected drugs). Participants who had last injected more than 30 days before interview were excluded from analyses. To compare proportions, we used chi-square tests and considered a two-tailed test probability of <0.05 statistically significant.23
Of the 2,323 study volunteers, 1,404 were current, regular crack smokers, of whom 419 were current sex workers. This analysis describes these 419 crack-smoking sex workers, of whom 303 women and 66 men were noninjectors and 34 women and 16 men were current injectors.
Of the 419 crack-smoking sex workers, 44% were in Miami, 37% were in New York, and 19% were in San Francisco. Most participants were older than 25 years of age and were African-American (Table 1). Fewer than half had completed high school. Most women but less than half of men were living in a house or an apartment; many were living on the streets or in shelters. In the 12 months before the interview, nearly a third of women and half of men had been incarcerated. More than half of men and women had received health care in the past 12 months and more than a third of men and women had ever received drug treatment (Table 1).
San Francisco participants were significantly more likely than participants from other cities to be female (94% in San Francisco, 81% in Miami, and 73% in New York). Miami participants were significantly less likely than those in New York and San Francisco to be African-American (66% in Miami, 91% in New York, and 95% in San Francisco) and significantly more likely to be Hispanic (30% in Miami, 6% in New York, and 0% in San Francisco). More noninjectors were recruited from New York (42.5%) than from Miami (37.4%) or San Francisco (20.1%).
Compared with noninjectors, injectors were older and were significantly less likely to be African-American (32% vs. 87%) and significantly more likely to be Hispanic (52% vs. 10%) (Table 1). Injectors were more likely to live in shelters or on the streets (40% vs. 27%, p = 0.06) but were similar to noninjectors regarding education, recent medical care, and drug treatment.
High-Risk Sexual Partners
Most female and male crack-smoking sex workers reported having had sex with partners that placed them at high risk for acquiring or transmitting HIV and other STDs, including injectors and persons they believed were HIV infected (Table 2). In addition, 24% of the women reported having been forced to have sex against their will within the 12 months before interview. Most men and women reported having had at least one STD in their lifetimes (Table 2).
Compared with noninjectors, injectors were significantly more likely to have ever had sex with other injectors (74% vs. 37% among women; 69% vs. 21% among men) or with someone they believed was HIV infected (18% vs. 7% among women, p < 0.01 and 31% vs. 16% among men, p = 0.15) (Table 2). However, female injectors were significantly less likely than noninjectors to report ever having had an STD (82% vs. 94%).
Location of Sex Work
Crack-smoking sex workers reported most frequently selling sex in hotels, apartments where people were not using drugs, apartments where people were using drugs (other than crack houses and shooting galleries), cars, and vacant lots (Table 2). Crack houses were among the least commonly reported sites for sex work (3%) (Table 2). Male and female noninjectors were more likely than injectors to most frequently sell sex in apartments where people were not using drugs. Crack-smoking sex workers in Miami were significantly more likely than those in other cities to sell sex in apartments where people were not using drugs (48% in Miami, 29% in New York, and 36% in San Francisco) and in hallways (20% in Miami, 0% in New York, 0% in San Francisco). Sex workers in San Francisco were significantly more likely (40%) than those in New York (27%) or Miami (19%) to sell sex in apartments where people were not using drugs.
Overall, sex workers who most frequently sold sex in crack houses or vacant lots were significantly more likely than sex workers who most frequently sold sex in other sites to have had sex with injectors (crack houses: 75%; lots: 64%; other sites: 38%) or persons they believed were HIV-infected (crack house: 27%; lots: 17%; other sites: 8%). Women who most frequently sold sex in vacant lots were more likely than women who sold sex in other locations to have had more than 30 paying partners in their lifetimes (89% vs. 50%), to have had an STD in the last year (58% vs. 28%), and to practice receptive anal sex (16% vs. 5%), although the latter difference was not statistically significant. Women who most frequently sold sex in vacant lots (52%) and crack houses (63%) were less likely to have used condoms the last time they sold vaginal sex than women who most frequently sold sex in other sites (apartments where people were not using drugs [68%], cars [71%], hotels [70%]), although differences were not statistically significant.
Form of Payment
For both men and women, money was the most common form of payment (Table 2). Fewer than one fifth of both women and men received crack as their most common form of payment. There were no important differences in form of payment by injection status. Sex workers who most frequently sold sex in crack houses were significantly more likely (50%) to receive crack as their most common form of payment than were sex workers who most frequently sold sex in other locations (crack houses or shooting galleries [21%], apartments where people were not using drugs [15%], cars [9%], or hotels [5%]). Compared with women who most frequently sold sex for money, women who most frequently sold sex for crack were significantly more likely to have ever had an STD (89% vs. 78%) and were less likely to have used a condom the last time they sold vaginal sex (27% vs. 59%) or oral sex (48% vs. 78%).
Types of Sex and Condom Use
Because the sexual and condom use practices of injecting and noninjecting crack-smoking sex workers were similar, the two groups were combined for analysis. The majority of women had engaged in vaginal sex both with paying (98%) and nonpaying (57%) partners in the 30 days before interview (Figure 1). Many engaged in receptive oral sex (having a penis in one's mouth) with paying (64%) or nonpaying (31%) partners. Few engaged in receptive anal sex with paying (5%) or nonpaying (5%) partners. Of those who reported engaging in vaginal sex with paying partners, fewer than half reported consistent condom use in the past 30 days (Figure 1). Of those who had vaginal sex with nonpaying partners, only 23% reported consistent condom use. Condom use was less common during receptive oral sex with both paying and nonpaying partners and during anal sex with nonpaying partners. However, condom use was more common during receptive anal sex with paying partners than during other types of sex (Figure 1).
All men had practiced some type of sex with other men. Thirty-six percent engaged in receptive oral-penile sex and 78% engaged in insertive oral-penile sex with paying male partners (Figure 2). Smaller proportions engaged in receptive (18%) or insertive (23%) oral-penile sex with nonpaying male partners. Fewer than one third of men engaged in receptive anal-penile sex with paying male partners (28%) or nonpaying male partners (11%) or insertive anal-penile sex (20%) with nonpaying male partners. Approximately half of men (48%) reported engaging in insertive anal-penile sex with paying male partners. Less than 40% of men reported consistent condom use for sex with paying and nonpaying partners. Condom use was highest for insertive and receptive anal-penile sex with nonpaying partners and lowest for insertive oral-penile sex with paying and nonpaying partners and receptive oral-penile sex with nonpaying partners. Forty-four percent of male sex workers reported having vaginal sex with female nonpaying partners in the 30 days before interview and only 24% reported consistent condom use. Data on sex with paying female partners were not analyzed because data were judged unreliable.
Seroprevalence of HIV, Syphilis, and HSV-2
More than a quarter of sex workers were infected with HIV (women: 25%; men: 39%), syphilis (women: 40%; men: 25%), or HSV-2 (women: 73%; men: 45%). Noninjectors were more likely than injectors to have syphilis among women (42% vs. 24%) and men (29% vs. 13%), although the latter difference was not statistically significant. Rates of HIV and HSV-2 infection did not differ significantly by injection status in women (HIV: injectors 23.5%, noninjectors 25.4%; HSV-2: injectors 65.4%, noninjectors 73.4%) or men (HIV: injectors 31.3%, noninjectors 40.9%; HSV-2: injectors 30.8%, noninjectors 47.6%). HIV infection rates varied by city, among both noninjectors (Miami: 31.9%, New York: 35.7%, San Francisco: 5.4%) and injectors (Miami: 18.8%, New York: 32.1%, San Francisco: 16.7%).
Sequence of Initiation of Crack Use and Sex Work
Overall, sex workers tended to have first engaged in sex work before they first smoked crack. Among women, the median age of first vaginal or anal sex was 19 years with paying partners and 15 years with nonpaying partners. Male sex workers first performed anal sex with paying male partners at a median age of 18 years. The median age of first anal sex with a nonpaying male partner was 16 years. The median year of first sex work was 1984 for men and 1985 for women. For both men and women, the median age of first crack use was 20 years of age and the median year of first crack use was 1986. No significant differences in the age at or year of first sexual activity or crack use by injection status were found.
Female sex workers were significantly more likely than male sex workers (38% vs. 24%) to have first used crack before they engaged in sex work activity or to have started smoking crack and selling sex in the same calendar year (23% vs. 13%). In addition, noninjectors were significantly more likely than injectors to have first smoked crack before selling sex for the first time (37% vs. 20%). There were no important differences in high-risk sexual behaviors (including sex with injectors, number of lifetime sex partners, and consistent condom use) by sequence of crack use and sex work initiation.
In summary, we found that the actively smoking sex workers recruited from the streets of Miami, New York, and San Francisco were largely poor, racial or ethnic minorities who had low education levels, high rates of homelessness and incarceration, and only moderate levels of prior drug treatment, as other studies have shown.1,3 The vast majority os sex workers were at very high risk of acquiring or transmitting HIV and other STDs because of high-risk sexual partners and sexual practices, a finding consistent with small ethnographic and clinic-based studies.5,7,8,13–19,21,22 Given these risk profiles, it is not surprising that these sex workers had very high rates of HIV, syphilis, and HSV-2 infection. Although the crack-smoking sex workers who injected drugs appeared to be at greater risk for HIV infection than noninjectors because they reported more frequent sexual contact with injectors and HIV-infected persons, noninjectors had a similar prevalence of HIV, possibly because they were more likely to be recruited from New York City, the site with highest HIV prevalence. Compared with injectors, noninjectors also had significantly higher rates of syphilis, and among women, higher rates of self-reported prior STD. Thus, the high HIV prevalence among noninjectors may also reflect enhanced susceptibility to HIV among those with ulcerative and nonulcerative STD.24
More than a quarter of sex workers in this study had first used crack before they first sold sex, suggesting that crack addiction may force some to turn to sex work to finance their addiction. However, a sizable proportion of both male and female sex workers had initiated sex work before they first used crack. This may reflect the fact that factors other than crack use are stronger determinants of the initiation of sex work. Alternatively, this may be owing to a cohort effect in our study because the crack form of cocaine was not widely marketed in these neighborhoods until the mid-1980s1,2 by which time many study participants had already first sold sex. Although the sexual practices of the current, heavy crack smokers in this study were very high risk, we could not assess whether crack use resulted in these patterns because we did not collect data on sexual practices before initiation of crack. To evaluate the extent to which crack use leads to sex work or alters sexual practices, large studies of sex workers who entered adolescence after crack became widely available in their communities are needed.
In this study, vaginal sex was the most commonly reported type of sexual activity among female sex workers, in contrast to earlier reports indicating that oral sex was most common.7 Consistent condom use during all types of sex was uncommonly reported by both women and men. Condom use was least common with nonpaying partners of women and men, as other studies have shown.10,25 Nearly half of the male crack-smoking sex workers, all of whom had had sex with other men, also had nonpaying female partners, and consistent condom use was low with such partners. This bisexual activity may place female partners at especially high risk for HIV and STD infections. These findings underscore the importance of addressing vaginal sex and sex with nonpaying partners as critical risk behaviors in HIV/STD prevention programs for crack-smoking sex workers. Prevention specialists should continue to examine barriers to condom use and methods to increase use, especially with nonpaying partners.
We found that crack-smoking sex workers most commonly sell sex for money rather than drugs and that they sell sex in a variety of locations, including hotels, cars, hallways, and other public or semipublic locations where illicit drugs were not being used. These findings contrast with earlier reports showing that "crack whores," unlike "street prostitutes," most commonly exchange sex directly for drugs in crack houses or other locations where illicit drugs were being used.7,8,26 These differences may reflect the recruitment methods necessitated by this large-scale epidemiological study. To ensure personal safety, outreach workers were more likely to recruit participants on the street than in crack houses where direct sex-for-drug exchanges may be more common.8 We found that sex workers who sell sex in crack houses and vacant lots had riskier sexual partners and practices and lower condom use rates than those selling sex in other locations, as ethnographers have observed.8 If HIV/STD prevention programs are to reach the crack-smoking sex workers at greatest risk, they must therefore focus on more inaccessible or elusive populations working in crack houses and public locations like vacant lots. In addition, prevention messages that stress carrying condoms at all times may be most meaningful for sex workers working in impermanent, public settings like cars, vacant lots, and hallways.
This study has some important limitations. First, many survey questions required participants to quantify activity that took place in the recent past, and this may have been difficult, especially for those influenced by illicit drugs. Nevertheless, interviewers considered most participants to be honest (88%) and accurate (76%) "most or all" of the time. Second, some "socially acceptable" responses, such as consistent condom use, may have been overreported. Third, because street-based convenience sampling methods were neither random nor systematic, we cannot generalize our findings to all crack-smoking sex workers in these communities, especially those who do not spend time on the streets. Indeed, many of the observed differences in participant characteristics by city may reflect subtle variations in recruitment methods related to how crack-smoking sex workers operate in a given neighborhood.
Despite these limitations, our findings support the urgent need to target HIV/STD prevention efforts to crack-smoking sex workers in these urban neighborhoods. Prevention among African-Americans is especially critical because they constituted the majority of crack-smoking sex workers in our study and cocaine use during the early 1990s did not decrease in young African-Americans as it did in other racial and ethnic groups.27 Because a high proportion of crack-smoking sex workers live in streets, shelters, and welfare hotels and practice sex work in diverse locations, frontline interventions led by community-based organizations and neighborhood groups are needed to complement prevention efforts in health care, drug treatment, and correctional facilities where many crack-smoking sex workers receive care or live.28–31
At the moment, behavioral interventions that have been demonstrated to reduce high-risk sexual behavior among drug-dependent sex workers are limited.29 Nevertheless, we found that crack smokers were willing to talk about their HIV risks. Moreover, many were already trying to reduce their risk and some were succeeding.32 To reduce the alarming pace of sexual HIV transmission and the burden of STD in the United States, we must devote resources to developing, evaluating, and disseminating effective behavioral interventions tailored to the unique needs of crack-smoking sex workers. Many crack users will not reduce their sexual risk-taking behavior as long as they remain addicted or suffer the psychosocial distress and psychiatric morbidity associated with the initiation of drug use.33,34 Therefore, developing more effective methods to prevent and treat crack addiction and expanding access to treatment, especially for women and youth, will be critical to sustaining success in prevention. In addition, prevention of sex work and crack smoking must address the underlying economic hardships, psychosocial stressors, psychiatric morbidity, and community disintegration that are associated with initiation of sex work, crack smoking, and HIV/STD risk behaviors.33–35 Finally, active diagnosis and antibiotic treatment of STD has been shown to substantially reduce HIV transmission in Tanzanian communities with high rates of HIV and STD.36 If similar community-based biomedical interventions prove to be effective in the United States, crack-smoking sex workers might benefit greatly because of their high rates of HIV and genital ulcer disease and concurrent HIV-STD infections.
1. Bowser BP. Crack and AIDS: An ethnographic impression. JAMA 1989; 81:538-540.
2. Washton AM. Cocaine: Drug epidemic of the 80's. In: Allen DF, ed. The Cocaine Crisis. New York: Plenum, 1987; 45-63.
3. Fullilove R, Fullilove M, Bowser B, Gross S. Risk of sexually transmitted disease among black adolescent crack users in Oakland and San Francisco, California. JAMA 1990; 263:851-855.
4. Hatsukami DK, Fischman MW. Crack cocaine and cocaine hydrochloride. Are the differences myth or reality? JAMA 1996; 276: 1580-1588.
5. Inciardi J, Lockwood D, Pottieger A. Women and Crack Cocaine. New York: Macmillan Publishing Company, 1993.
6. Inciardi J, Pottieger A, Forney M, Chitwood D, McBride D. Prostitution, IV drug use, and sex-for-crack exchanges among serious delinquents: Risks for HIV infection. Criminology 1991; 29:221-235.
7. Ratner M, ed. Crack Pipe as Pimp: An Ethnographic Investigation of Sex for Crack Exchanges. New York: Lexington Books, 1993: 1-35.
8. Inciardi J. Crack, crack house sex, and HIV risk. Arch Sex Behav 1995; 24:249-269.
9. Edlin BR, Irwin KL, Ludwig D, et al. High-risk sex behavior among young street-recruited crack cocaine smokers in three American cities. J Psychoactive Drugs 1992; 24:363-371.
10. Centers for Disease Control. Condom use among male injecting drug users in New York City, 1987-1990. MMWR 1992; 41:17.
11. Schmid GP, Sanders LL, Blount JH, Alexander ER. Chancroid in the U.S., reestablishment of an old disease. JAMA 1987; 258:3265-3268.
12. Marx R, Arai S, Rolfs R, Sterk C, Kahn J. Crack, sex, and STD. Sex Transm Dis 1991; 18:92-101.
13. Edlin BR, Irwin KL, Faruque S, et al. Intersecting epidemics: Crack cocaine use and HIV infection among inner-city young adults. New Engl J Med 1994; 331:1422-1427.
14. Plant M. AIDS, Drugs and Prostitution. New York: Routledge: 1990.
15. Fullilove M, Fullilove R. Intersecting epidemics: Black teen crack use and sexually transmitted disease. J Am Women's Med Assn 1989; 44:146-153.
16. Siegal H, Carlson R, Falck R, Forney M, Wang J, Li L. High risk behaviors for transmission of syphilis and human immunodeficiency virus among crack cocaine-using women. Sex Transm Dis 1992; 19:266-271.
17. Balshem M, Oxman G, van Rooyen D, Girod K. Syphilis, sex and crack cocaine: Images of risk and mortality. Soc Sci Med 1992; 35:147-160.
18. Marshall N, Hendtlass J. Drugs and prostitution. J Drug Issues 1986; 16:237-248.
19. Plant M, Plant M, Peck D, Setters J. The sex industry, alcohol and illicit drugs: implications for the spread of HIV infection. Br J Addict 1989; 84:53-59.
20. Maher L, Curtis R. Women on the edge of crime: crack cocaine and the changing contexts of street level sex work in New York City. Crime, Law and Social Change 1992; 18:221-258.
21. Chiasson MA, Stoneburner RL, Hildebrandt DS, Ewing WE, Telzak EE, Jaffe HW. Heterosexual transmission of HIV-1 associated with the use of smokable freebase cocaine (crack). AIDS 1991; 5:1121-1126.
22. Chirgwin K, DeHovitz JA, Dillon S, McCormack WM. HIV infection, genital ulcer disease, and crack cocaine use among patients attending a clinic for STDs. Am J Public Health 1991; 81:1576-1579.
23. SAS/STAT user's guide: Version 6, 4th ed. Cary, NC: SAS Institute, 1990.
24. Wasserheit, J. Epidemiologic synergy. Interrelationships between human immunodeficiency virus infection and other sexually transmitted diseases. Sex Transm Dis 1992; 2:61-77.
25. Elifson K, Boles J, Sweat M. Risk factors associated with HIV infection among male prostitutes. Am J Public Health 1993; 83:79-83.
26. Ouellet L, Wiebel W, Jimenez A, Johnson W. In: Ratner M, ed. Crack Pipe as Pimp. New York: Lexington Books: 1993:69-96.
27. Braun BL, Murray D, Hannan P, Sidney S, Chap L. Cocaine use and characteristics of young adult users from 1987 to 1992: The CAR-DIA study. Am J Public Health 1996; 86:1736-1741.
28. Dorfman L, Derish P, Cohen J. Hey girlfriend: An evaluation of AIDS prevention among women in the sex industry. Health Educ Q 1992; 19:25-40.
29. Exner TM, Seal DW, Ehrhardt AA. A review of HIV Interventions for at-risk women. AIDS and Behavior 1997; 1:93-123.
30. Jackson L, Highcrest A, Coates R. Varied potential risks of HIV infection among prostitutes. Soc Sci Med 1992; 35:281-286.
31. Sikkema KJ, Kelly J, Heckman T, et al. Effects of community-level behavior change intervention for women in low-income housing developments. In: Proceedings of the XI International Conference on AIDS. Vancouver, 1996; 250 (Abstract Tu.C.453).
32. Edlin BR, Faruque S, McCoy CB, Word CO. Crack cocaine and HIV infection in inner cities [letter]. N Engl J Med 1995; 332:1234.
33. El-Bassel N, Schilling RF, Irwin KL, et al. Sex trading and psychologic distress among women recruited from the streets of Harlem. Am J Public Health 1997; 87:66-70.
34. Word CO, Bowser B. Background to crack cocaine addiction and HIV high risk behavior: The next epidemic. Am J Drug Alcohol Abuse 1997; 23:67-77.
35. Wallace R, Fullilove MT, Flisher AJ. AIDS, violence and behavioral coding: Information theory, risk behavior and dynamic process on coregroup sociogeographic networks. Soc Sci Med 1996; 43:339-352.
36. Grosskurth H, Mosha F, Todd J, et al. Impact of improved treatment of sexually transmitted disease on HIV infection in rural Tanzania: Randomised controlled trial. Lancet 1995; 346:530-536.