Tap 20 — TAPs <<<Documents<<<Home

Drug Use, Sexually Transmitted Diseases, and Sex-Related Risk Behaviors in Alaska

David M. Paschane, B.S.
Henry H. Cagle, B.S.
Andrea M. Fenaughty, Ph.D.
Dennis G. Fisher, Ph.D.
Department of Psychology
University of Alaska Anchorage
Anchorage, Alaska

Abstract The association between illicit drug use and sexually transmitted diseases (STDs) is well established in the literature; however, little is known about the networks of disease transmission among rural drug using populations. This paper explores issues related to STD risk among hidden, drug using populations. A structured interview was administered to 1,088 out-of-treatment, drug using adults. The sample included a large proportion of Alaska Natives and American Indians. Several descriptive statistics illustrate associations between sex-related risk behaviors, drug use, and disease transmission. Treating self-reported history of gonorrhea infection as a possible indirect indicator of STD/HIV risk, predictors of infection were identified through logistic regression modeling. The characteristics of drug users most likely to have reported gonorrhea infection were (a) a history of snorting or injecting cocaine, (b) income from prostitution, (c) being black as compared with being non-black, (d) a history of using nonheroin opiates, and (e) being in a younger age group. The model also included an interaction between prostitution and age. This paper includes a discussion of issues related to barriers to treatment and rural-urban mobility.

Alaska, the northernmost territory in the United States, with a population density of one person per square mile, has characteristics that are common to many other rural States (Cordes 1989). Alaska has a stressed primary economic base, underdeveloped infrastructure for accessing health care, and many isolated communities. Hence, delivery of health services and the operation of research in Alaska is challenging. Anchorage, Alaska's largest metropolitan area, is a centralized source of health care and other human services, as well as free shelter and food for a large disenfranchised population. As a result, Anchorage attracts populations migrating from rural Alaska. Seasonal employment opportunities and the centralization of service-oriented jobs also contribute to the high levels of migration. In addition, Anchorage attracts those populations seeking to purchase drugs or participate in drug-related activities (e.g., prostitution). These conditions, plus the necessary resources required for effective research, suggest Anchorage as an opportune and cost effective place to sample Alaskan drug users.

Drug abuse and its contribution to diseases is a growing concern in rural communities throughout the United States. Diseases associated with illicit drug use, such as AIDS and other sexually transmitted diseases (STDs), have stimulated health professionals' interest in implementing prevention models among drug using populations. Injection drug users (IDUs) have been commonly recognized for contributing to the spread of hepatitis and HIV. In a collection of ethnographic studies, Ratner (1993) describes the relationship between trading sex for drugs or money and smoking cocaine (i.e., crack; Ouellet et al. in press), and illustrates the potential risk of disease transmission where drug use and sex behaviors are combined. Other factors contributing to HIV/ STD risk, besides an increase in the number of sex partners, are genital ulcer disease caused by an earlier STD (Chirgwin et al. 1991), genital tissue damage (e.g., penile abrasion), and ulcers in the mouth from cocaine smoking burns (Ratner). The levels of risk for STDs at locations where cocaine smokers trade sex have been equated with risk in the gay bathhouses of the past (Goldsmith 1988).

STDs are among the most common infectious diseases (CDC 1994), affecting more than 13 million adults in the United States (NIAID 1992). These contribute to a sizable morbidity and mortality, place a significant burden on medical services, and have an estimated annual cost of $5 billion (NIAID). The decreases of screening resources in the United States (Yankauer 1994) may further complicate the burden of STDs for rural populations because of barriers to health care access (Steel and Haverkos 1992). Additionally, drug users may be at a greater disadvantage because of individual resistance to access services if they fear disclosure and consequences for their drug using behaviors (Haverkos 1991). These conditions mean that better targeting schemes are necessary when attempting to control the prevalence of diseases among high-risk populations. Watters (1993) recommends targeted sampling of the noninstitutionalized hidden populations in order to provide information leading to indicators of infection rates and behavioral risks associated with STDs.

Alaska has a unique history of STDs. They were an important cause of illness and sterility as early as the 1700s, when they were first introduced to the Native populations by Europeans (Fortuine 1989). The earliest reports of a behavioral association with STDs are alcohol consumption and sex-related risk behaviors (Fortuine). Historically, gonorrhea (GC) rates have been an important surrogate indicator of HIV risk and other STDs in Alaska. Because chlamydia is not a reportable disease and syphilis rates (1.34 per 100,000) are too low to be reliable indicators, GC rates are the most reliable long-term indicators of unsafe sexual behavior. Twenty years ago, Alaska's GC rates were the highest in the Nation (Eisenberg and Wiesner 1976). These rates have since declined and are now similar to other rural States where GC is below the national objective of the Centers for Disease Control and Prevention (CDC 1994). However, the incidence of GC is high among some groups in Alaska. In 1993, a total of 676 cases of GC were reported with an overall rate of 115 per 100,000; highest rates were among 15- to 19-year-old women, 834 per 100,000; and blacks, 894 per 100,000 (State of Alaska HIV Prevention Planning Group 1995).

The potential for reinfection makes GC fundamentally different from most other bacterial STDs. Some factors contributing to the prevalence of GC are (a) the host's lack of acquired immunity, (b) the potential for asymptomatic infection, and (c) its unique biological makeup (Bignell 1994). Asymptomatic-infected persons are believed to contribute disproportionately to the perpetuation of GC (Upchurch et al. 1990). Moreover, the dramatic increases of penicillin-resistant strains occurring in many regions of the United States may increase the rates of GC prevalence (CDC 1994; Gorwitz et al. 1993; Handsfield et al. 1989). Beller et al. (1992) found that nearly 34 percent of the multiple GC infections in Alaska were among a core group of infected individuals. The existence of a core group may suggest a network of disease transmission among a specific population not easily recognized by traditional surveillance methods.

Behavioral characteristics have been reported to be associated with GC (Beller et al. 1992; Handsfield, et al. 1989; Schwarcz et al. 1992; Upchurch et al. 1990) and other STDs (Booth et al. 1993; Chirgwin et al. 1991; Kim et al. 1993; Marx et al. 1991; Richert et al. 1993). The behaviors associated with STD acquisition can be both direct (e.g., deliberate unprotected sexual contact) and indirect (e.g., drug use leading to unprotected sexual contact). The relationship between drug use and sexual behavior is often due to the context in which drugs are obtained and the extent of the drug user's perceived need; that is, a compulsive urgency for those drugs and willingness to take greater sexual risk (Ratner 1993; Zinberg 1984). At this time, there is little known about these behaviors and their relationship to diseases in Alaska. Even though Alaskans have long had a reputation for high alcohol consumption rates, the high rates of drug use have been underreported (Fisher and Booker 1990), and even less is known about the networks of disease transmission among these drug users.

Haverkos (1991) argues that the integration of drug abuse and STD treatments would improve the effectiveness of public health interventions directed at controlling STDs. Support for this argument has been tested by clinical trials among intravenous drug users (Umbricht-Schneiter et al. 1994). The overall purpose of this study was to describe those factors that may better explain the networks of disease transmission among rural drug using populations. This required an illustration of associations between sex-related risk behaviors, drug use, and disease transmission. Even though clinical and surveillance data are normally a primary source of STD information, neither source of data reflects the correlates of STDs as they are found among specific high-risk populations (Anderson et al. 1994). In addition to the associations, a risk profile for a possible indirect indicator of high-risk behaviors (i.e., GC) will be modeled for the purpose of better describing the subgroup of the population most responsible for the disease network (Yorke et al. 1978) in a transitional rural population. Because of GC's epidemiological nature, it is an appropriate indicator and allows for such exploratory modeling and targeting of a high-risk hidden population.

Method

This research is part of a longitudinal, multisite study of out-of-treatment cocaine smokers and injection drug users at risk for HIV acquisition and transmission. The National Institute on Drug Abuse Cooperative Agreement for AIDS Community-Based Outreach/Intervention Research Program is designed to assess the efficacy of a locally developed enhanced intervention compared with a standardized intervention for HIV risk reduction. Participant recruitment for this study was guided by a targeted sampling plan based on Watters and Biernacki (1989).

All research activities occurred in an office-based setting, the Drug Abuse Research Field Station. Participants provided informed consent under a Federal Certificate of Confidentiality and received monetary compensation for their time in research. Individuals eligible for research participation were at least 18 years old and self-reported (a) no drug treatment in the preceding 30 days, (b) injecting heroin, non-heroin opiates, cocaine, or amphetamines, and presented needle track marks indicative of recent injection drug use, or (c) cocaine smoking and produced positive urinalysis for cocaine metabolites. Participants routinely received urinalysis screening for cocaine metabolites, morphine, and amphetamines (Abusescreen ONTRAK; Roche Diagnostic Systems, Montclaire, NJ).

Data in this study are cross-sectional, with participant recruitment and data collection beginning in November 1991 and ending in August 1995. Dependent and independent (predictor) variables were drawn from the Risk Behavior Assessment (RBA) (National Institute on Drug Abuse 1991). The RBA is a structured interview that elicits demographic, alcohol and illicit drug use, drug treatment, sexual behavior, health, criminal activity, and income information. Most of the RBA questions are phrased to use a time reference of the last 30 days, followed by lifetime, the last 48 hours, and the last year. History of sexually transmitted disease is assessed by responses to two items: (a) the number of times participants report being told by a doctor or nurse that they had the specific STD, and (b) the year they report last being treated. The RBA has good test-retest reliability on the drug use and sexual behavior variables (Fisher et al. 1993b; Needle et al. 1995) and high validity coefficients on the drug use variables (Dowling-Guyer et al. 1994; Weatherby et al. 1994).

All scientifically relevant RBA variables were considered for analysis. For the purpose of these analyses, some recoding of the variables was necessary. Categorical variables were either dummy coded or coded dichotomously; continuous variables were maintained, but those with skewed distributions were coded dichotomously. Statistical tests that were applied to the data included: (a) Student's t test, (b) Pearson chi-square, (c) categorical modeling analysis, (d) ordinal logistic regression analysis, and (e) a binary response exploratory logistic regression analysis. All analyses were performed using the SAS System (SAS Institute Inc. 1990). Logistic regression model building and regression diagnostics were performed using techniques developed by Hosmer and Lemeshow (1989).

Barriers

There are a number of common criticisms of self-reported survey data: (a) underreporting due to asymptomatic infections, (b) unwillingness to discuss sensitive subject matter, and (c) inability to recall disease information provided by a medical provider (Anderson et al. 1994). However, clinical studies and surveillance data rarely include assessments of risk behaviors. The outcome variable in this study is self-reported history of STDs, and this may introduce undesirable measurement error to the model. Reliability of self-report is believed to be high when the RBA instrument is utilized (see Method). The validity of self-reported hepatitis infection has been investigated (Fisher et al. 1996), and findings suggest underreporting. The same problem may be present for other infections.

In order to minimize the effect of misreporting specific STDs, responses to number of times participants reported being told by a doctor or nurse they had an STD (i.e., gonorrhea, syphilis, genital warts, chlamydia, and genital herpes), including hepatitis B, were aggregated. The data were recoded because of the nonnormality of the distributions. The resulting variable had three categories: (a) no history of STDs, (b) history of one STD, and (c) history of multiple STDs. Drug- and sex-related risk behaviors having occurred 30 days preceding the interview were defined as recent behaviors. Drug use was categorized as those who only injected drugs, those who only smoked cocaine, and those who did both. Earlier studies have demonstrated high validity on drug use variables (see Method). The sex-related risk behavior variable consisted of four categories: (a) traded sex for drugs or money, (b) traded drugs or money for sex, (c) participated in both items a and b, and (d) did not participate in either item a or b. Utilizing the Dowling-Guyer et al. (1994) data, separate reliability analyses were performed on the GC variable. Test-retest reliability coefficients for number of times (r = .94; n = 222) and year treated (r = .93; n = 64) were both substantial. Test-retest reliability analyses were also conducted by gender. Among women, reliability coefficients for number of times (r = .95; n = 57) and year treated (r = .99; n = 15) were only slightly greater than those for number of times (r = .94; n = 164) and year treated (r = .91; n = 48) for men. For logistic regression modeling, GC was recoded as "ever" or "never" because of similar use of the variable in previous research (Kim et al. 1993; Schwarcz et al. 1992; Upchurch et al. 1990).

Many of the methodological recommendations identified by Marx et al. (1991) in their review of studies reporting associations between sex, drugs, and STDs risk (e.g., comparison to uninfected group, specification of drugs used, nonminorities, and rural populations) are addressed in this study. Due to overall sample size, model replication could not be attempted within this sample; therefore, it is recommended that similar modeling be attempted using other samples of drug users.

Findings

The study sample (N = 1,088) consisted of 740 men (68 percent) and 348 women (32 percent). The mean age was 35.1 years (SD = 7.6) for men and 32.8 years (SD = 7.4) for women, t(1,086) = 4.71, p < .0001. A summary of selected characteristics of the study sample is included in table 1. More whites (43 percent) participated than other race groups; however, a greater proportion of blacks (31 percent) and American Indians/Alaska Natives (AI/ANs; 20 percent) participated than are represented in the Municipality of Anchorage, 6 percent each (Municipality of Anchorage, Community Planning and Development Department 1993). A majority of participants were not homeless (82 percent) and described themselves as heterosexual (93 percent). Levels of education are distributed almost evenly over these categories: (a) less than high school (36 percent), (b) high school or its equivalent (35 percent), and (c) greater than high school (29 percent). The most frequently reported STD was GC (36 percent), followed by hepatitis B (15 percent), chlamydia (14 percent), genital warts (10 percent), syphilis (5 percent) and genital herpes (5 percent). Perceived risk for HIV infection (n = 1,046) was skewed toward none to some chance, with only 27 percent perceiving themselves having half or a high chance of infection.

Table 2 contains a summary of drug use and sex-related risk behaviors by history of total number of STDs. Results of the multivariate categorical modeling utilizing the weighted-least-squares analyses indicated that the main effects were significant for both the trading variable, c2 (6, N = 1,088) = 55.89, p < .001, and the drug using variable, c2 (4, N = 1,088) = 16.54, p < .01; however, the interaction parameter was not significant. The reduced model excluded the interaction term.

Among the 12 possible risk categories (i.e., the interaction between trading behaviors and drug using behaviors), the largest proportions were among those with a history of no STDs and history of multiple STDs. Fifty-six percent of those who did not trade and reported smoking cocaine only (n = 408), and 48 percent of those who traded only money/drugs for sex and only smoked cocaine (n = 194), reported histories of no STDs. Fifty-seven percent of participants who traded sex, smoked cocaine, and injected drugs (n = 38), and 49 percent of those who traded sex, traded money/ drugs for sex, and only smoked cocaine (n = 39), and 47 percent of those who traded sex and only smoked cocaine (n = 122) reported histories of multiple STDs.

The most parsimonious ordinal logistic regression model of risk factors, where all three levels of the response variable (history of STDs) are represented, retained two significant risk categories (i.e., trading behavior by drug use) and one protective factor: (a) traded sex, smoked cocaine, and injected drugs (OR = 2.67; CI = 1.41, 5.05), (b) traded sex, and only smoked cocaine (OR = 1.73; CI = 1.20, 2.50), and (c) did not trade, and reported smoking cocaine only (OR = 0.53; CI = 0.41, 0.68). Results of the analysis are reported in table 3. The effects of the combined behaviors multiply the odds ratio for either of the comparisons of the combined response variables represented by Constant A (multiple STDs versus one and no STD) and Constant B (multiple and one STDs versus no STDs).

Table 3. Ordinal logistic regression model for predicting sexually transmitted diseases among Alaskan drug users
(N=1,088)

Factor	ß	SE(ß)	OR	95% CI
Constant A	-0.73***	0.09
Constant B	0.35***	0.08
Trade sex, inject, smoke cocaine	0.98**	0.32	2.67	1.41, 5.05
Trade sex, smoke cocaine	0.55**	0.19	1.73	1.20, 2.50
No trade, smoke cocaine	-0.64***	0.13	0.53	0.41, 0.68
	**p<.01. ***p<.001 SE=standard error; OR=odds ratio; CI=confidence interval.

Results of the exploratory logistic regression analysis modeling predictors of GC infection are presented in table 4. Risk factors for GC were (a) a history of snorting or injecting cocaine (OR = 2.31; CI = 1.20, 4.43), (b) income from prostitution, (c) being black as compared with being non-black (OR = 1.79; CI = 1.34, 2.40), (d) a history of using other opiates (OR = 1.55; CI = 1.18, 2.03), and (e) being in a younger age group. Table 5 shows the interaction of age with income from prostitution; presented are the odds ratios for having income from prostitution. The Hosmer-Lemeshow goodness-of-fit tests (Hosmer and Lemeshow 1989), c2 (8) = 8.16, p = .42, demonstrated adequate model fit.

Table 4. Logistic regression model for predicting gonorrhea among Alaskan drug users
(N=1,083)

Factor	ß	SE(ß)	OR	95% CI
Constant	-3.19	0.45
Cocaine (snort, inject)	0.84	0.33	2.31	1.20, 4.43
Blacks	0.58	0.15	1.79	1.34, 2.40
Opiates (non-heroin)	0.44	0.14	1.55	1.18, 2.03
Prostitution	4.45	1.47
Age	0.05	0.01
Prostitution x Age	-.011	0.05

Note: Change in N due to missing observations. See Table 5 for estimated odds ratios and 95 percent confidence intervals for prostitution, controlling for age.

OR=odds ratio; CI=confidence interval.

Note. Odds are the ratio of the odds of gonorrhea infection among those reporting income from prostitution, to the odds of gonorrhea infection among those reporting not receiving income from prostitution (age is set to four levels). Change in N is due to missing observations.

OR=odds ratio; CI=confidence interval.

Conclusions

Results confirmed prior hypotheses that associations exist between sex-related risk behaviors, drug use, and STDs in rural populations (Forney et al. 1992; Steel and Haverkos 1992; Thomas et al. 1995). In this case, a history of multiple STDs was best predicted by those who engaged in trading sex for drugs or money and smoked cocaine. Cocaine smokers who did not trade sex for money or drugs were more likely than any other group to have reported no STDs. Almost half of those who were trading money/drugs for sex and smoking cocaine had no STDs. These findings suggest that although drug users are commonly considered at high risk for STDs, not all drug users represent the core group perpetuating STDs in a given rural population. Accurate identification of those most likely to represent the source of STDs and diseases spread through injection drug use can make targeted interventions most effective. The associations described herein illustrate the several possible profiles for STDs within drug using populations. Trading sex for money or for drugs was a component of both groups with significant odds ratios for multiple STDs. The two groups each smoked cocaine; one also injected drugs. These findings support Ratner (1993), in that trading sex and smoking cocaine is the most significant risk combination for STDs. Even though half of those who smoked cocaine, traded sex, and purchased sex with money or drugs (n = 39) reported multiple STDs, this group was not a significant predictor in ordinal modeling. This would suggest that at least half of this group, who have resources to both sell and buy sex, also have characteristics that reduce risk status.

Prostitution, especially among AI/ANs, has received little attention in Alaska. This may be due to the belief that high-risk sexual behaviors (i.e., prostitution) are primarily a characteristic of urban populations (Forney et al. 1992). In addition, rural populations are stereotypically thought of as members of isolated communities where trading sex is not relied upon for economic survival. Anchorage attracts many AI/ANs seeking treatment and assistance resources, and they may be at risk of separation from their traditional community norms that prevent further risk behaviors. The AI/AN women in this cohort often report multiple sex partners, unsafe sex practices, and high-risk drug using behaviors (Fenaughty et al. 1994; Fisher et al. 1993a). Recent reports found that among a sample of drug using women, AI/ANs were more than two and a half times likely to have had GC (Paschane et al. in press) and nearly two times more likely to have had chlamydia (Orr et al. 1995). Conway et al. (1992) describe a potential diffusion of HIV and STDs into rural populations of AI/ANs and suggest this may be due to regular migration between rural and urban areas.

The predictors identified in the GC model better define the hidden, high-risk population for STDs. The model includes two factors often reported in the literature, being black compared with other race groups (Kim et al. 1993; Upchurch et al. 1990), and a history of cocaine use (Marx et al. 1991; Schwarcz et al. 1992). Being black, as a risk factor for GC, agrees with surveillance data in Alaska where GC infection rates among blacks are highest (see the Introduction). Gershman and Rolfs (1991) suggest that race may be a surrogate marker for high-risk behavior, and if race is better defined (i.e., cofactors are identified), it may further describe the core group. A history of using non-heroin opiates has been reported to be associated with ethnicity (whites compared to blacks and AI/ANs; Cagle et al. 1996), infection with hepatitis B (Kuhrt-Hunstiger and Fisher 1994), and a history of chlamydia infection (Orr et al. 1995). The presence of non-heroin opiate use in the model may account for those non-blacks with a history of GC. More research is needed to explain the association between non-heroin opiate use, and risk for GC.

The interaction between age and income from prostitution is an interesting risk factor and may further explain the findings presented in this paper. If individuals who are trading sex for drugs or money and smoking cocaine are most likely to have had multiple STDs, then the age distribution of this population may further explain the disease network. The odds ratios illustrate the interaction by considering risk from prostitution at four age levels. The youngest age group of 20 years is more than nine times as likely to have had GC. Such a dramatic ratio of the odds between those reporting income from prostitution, and those who did not, suggest that this sex-related risk behavior may best define the core risk group in a rural population. These findings are further supported by the surveillance data reporting 15- to 19-year-old women as having one of the highest rates of GC in Alaska (see the Introduction).

Surprisingly, the pattern is contrary to the expected condition in which older age would be associated with history of GC because of the greater opportunity to contract the disease; however, this may also illustrate risk behaviors unique to younger sex-workers. Koester and Schwartz (1993) report that condom use was least among women trading sex directly for smokable cocaine versus women who traded sex for money. This same group may experience a number of conditions that increase their risk for multiple STDs, such as decreased power for negotiating safe sex practices (Worth 1989), being homeless (Zhao et al. 1995; Paschane et al. in press), and being poorly educated (Zhao et al.). Because asymptomatic-infected persons are believed to contribute disproportionately to the perpetuation of GC (Upchurch et al. 1990), it is possible that the non-drug users who purchase sex from the younger sex-workers may become infected, remain asymptomatic, and unknowingly transmit STDs to other members of the non-drug using population. Whether being older is a protective factor among prostitutes (OR = 0.34) is unclear and may require further investigation.

The models reported here better describe risk factors for GC and multiple STDs in a sample commonly believed to be at high risk for HIV and other STDs. Having unique risk profiles for GC may benefit public health professionals developing HIV/STD interventions in Alaska and other rural States. Future studies should investigate the strength of these associations in describing the populations of other rural areas. Such research may provide additional insights into the prevention of disease transmission where characteristics of populations vary. Two issues relevant to the control of STDs in rural areas are rural-urban mobility as it applies to disease prevalence and barriers to treatment services that control disease.

Haverkos (1991) contends that the integration of drug and STD services will best serve the public health need to control the ever-increasing rates of STDs and drug use. Rural locations, where resources are even more limited than their urban counterparts, can benefit from effective integration of services. The resistance that individuals may have to access services where they fear disclosure and consequences for their drug using behaviors (Haverkos 1991) may worsen the effects of these public health burdens. Barriers to treatment services (Steel and Haverkos 1992) are a reality for drug users in Alaska (Johnson et al. 1995). Common treatment barriers reported by Johnson et al. are excessive cost, lack of availability, inaccessible location, nonculturally relevant programs, and lack of child services. Service integration may help to better overcome some of these barriers experienced by drug users. A subgroup described at even more risk is those with psychiatric illness other than drug or alcohol use (Johnson et al.). In cases in which individuals are less able to make decisions regarding their welfare, mental health agencies, in cooperation with drug treatment centers, may help overcome this barrier to treatment.

As mentioned earlier, migration from rural communities to urban centers for accessing treatment, is, for some individuals, the only choice when services are centralized and local services are inadequate. Recent reports have recognized a significant amount of migration to Anchorage by AI/AN women (Fisher et al. 1993a; Fogel-Chance 1993; Hamilton and Seyfrit 1994). It is unclear what effect migration has on STD/HIV risk; however, other data suggest that behavioral trends among AI/ANs are leading to destructive outcomes. For example, the highest rates of suicide are among AI/AN men (Berman and Leask 1994), and a majority of those are intoxicated at time of death (Soule 1994). Fisher et al. (1995) also report a number of significant high-risk behaviors among AI/AN women that may be due to the effects of migration. Williams et al. (1993) suggest that travel patterns among drug users may increase the HIV risk to other populations (e.g., rural, non-drug users). Because rural areas contain seasonal employment and subsistence opportunities and most health and human services are centralized, migration is likely to continue in this population.

Recommendations

This study illustrates findings that are useful for developing targeting schemes for STD/HIV interventions. As budget reductions continue (Yankauer 1994), effective management of diseases requires accurate identification of core risk groups. For example, interventions are often designed for the purpose of serving those individuals most likely to seek treatment rather than the populations practicing high-risk behaviors. As a result, members of core groups responsible for the prevalence of STDs may not receive the treatment and counseling necessary for controlling further transmissions of the disease. One way to improve the likelihood of treatment among these high-risk populations is to plan coordinated referrals for treatment among public service agencies. If agencies are better equipped to make active referrals and have the opportunity to recognize the high-risk individuals, they can improve the likelihood of STD control. Future alterations to intervention programs should target the high-risk rural populations and find means of improving their access to testing and treatment.

A number of changes have been made to local STD/HIV screening and treatment services in an effort to reduce the STDs in this study population. A public health nurse (M.A. Lee, personal communication, February 8, 1996) and an HIV outreach worker (M.R. Covone, personal communication, February 8 1996) have attended social gatherings where members of high-risk populations are known to congregate (i.e., bars, massage parlors, the bus station) and performed HIV testing and risk reduction education. Peer outreach appears to be most effective; however, language barriers exist and may be addressed by including outreach workers with appropriate language skills. The most difficult group of drug users to screen for STDs, and at greatest risk as illustrated by these data, are those trading sex for drugs or money. This group may participate in sex-related risk behaviors during hours that cause them to sleep during the day and reduce the likelihood of accessing services during the same hours of operation. A mobile testing unit (e.g., van) operating during an evening shift may best improve access to testing for this high-risk group. Again, effective peer outreach is necessary because of the possible negative effect drug use may have in facilitating cooperation with these clients. Women trading sex are easier than men to target because they are more likely to walk the streets or work at massage parlors; men trading sex are more difficult to target because of lack of visibility. Further studies are necessary for identifying means of improving outreach to this population.

References

Anderson, J.E.; McCormick, L.; and Fichtner, R. Factors associated with self-reported STDs: Data from a national survey. Sexually Transmitted Diseases 21:303-308, 1994.

Beller, M.; Middaugh, J.; Gellin, B.; and Ingle, D. The contribution of reinfection to gonorrhea incidence in Alaska, 1983 to 1987. Sexually Transmitted Diseases 19:41-46, 1992.

Berman, M., and Leask, L. Violent death in Alaska: Who is most likely to die? Alaska Review of Social and Economic Conditions 29:1-12, 1994.

Bignell C. The eradication of gonorrhoeae. The British Medical Journal 309:1103-1104, 1994.

Booth, R.E.; Watters, J.K.; and Chitwood, D.D. HIV risk-related sex behaviors among injection drug users, crack smokers, and injection drug users who smoke crack. American Journal of Public Health 83:1144-1148, 1993.

Cagle, H.H.; Fisher, D.G.; Fenaughty, A.M.; and Paschane, D.M. "Predicting Recent Injection Drug Use Among Out-of-Treatment Injection Drug Users." Unpublished manuscript, 1996.

Centers for Disease Control and Prevention. Sexually Transmitted Disease Surveillance, 1993. Atlanta, GA: Centers for Disease Control and Prevention, 1994.

Chirgwin, K.; DeHovitz, J.A.; Dillon, S.; and McCormack, W.M. HIV infection, genital ulcer disease, and crack cocaine use among patients attending a clinic for sexually transmitted diseases. American Journal of Public Health 81:1576-1579, 1991.

Conway, G.A.; Ambrose, T.J.; Chase, E.; Hooper, E.Y.; Helgerson, S.D.; Johannes, P.; Epstein, M.R.; McRae, B.A.; Munn, V.P.; Keevama, L.; Raymond, S.A.; Schable, C.A.; Satten, G.A.; Petersen, L.R.; and Dondero, T.J. HIV infection in American Indians and Alaska Natives: Surveys in the Indian Health Service. Journal of Acquired Immune Deficiency Syndromes 5:803-809, 1992.

Cords, S.M. The changing rural environment and the relationship between health services and rural development. Health Services Research 23:757-784, 1989.

Dowling-Guyer, S.; Johnson, M.E.; Fisher, D.G.; Needle, R.; Watters, J.; Anderson, M.; Williams, M.; Kotranski, L.; Booth, R.; Rhodes, F.; Weatherby, N.; Estada, A.L.; Fleming, D.; Deren, S.; and Tortu, S. Reliability of drug users' self-reported HIV risk behaviors and validity of self-reported recent drug use. Assessment 1:383-392, 1994.

Eisenberg, M., and Wiesner, P.J. Reporting and treating gonorrhea: Results of a statewide survey in Alaska. Journal of the American Venereal Disease Association 3:79-83, 1976.

Fenaughty, A.M.; Fisher, D.G.; MacKinnon, D.P.; Wilson, P.J.; and Cagle, H.H. Predictors of condom use among Alaskan Native, white, and black drug users in Alaska. Arctic Medical Research 53:704-711, 1994.

Fisher, D.G.; and Booker, J.M. Drug abuse in Alaska: Myths versus reality. Psychology of Addictive Behaviors 4(1):2-5, 1990.

Fisher, D.G.; Cagle, H.H.; Davis, D.C.; Fenaughty, A.M.; Kuhrt-Hunstiger, T.I.; and Fison, S.R. Health consequences of illicit drug use in rural areas: Questions without answers. In: Robertson, E., ed. Rural Substance Abuse: State of Knowledge and Issues. Washington, DC: U.S. Govt. Print. Off., 1996. pp. 1-37.

Fisher, D.G.; Cagle, H.H.; and Wilson, P.J. Drug use and HIV risk in Alaska Natives. Drugs and Society 7:107-117, 1993a.

Fisher, D.G.; Fenaughty, A.M.; and Paschane, D.M. Alaska Native drug users: Results of a five-year study. Abstracts of the 123rd Annual Meeting and Exhibition of the American Public Health Association, San Diego, CA (2005-112), 1995.

Fisher, D.G.; Needle, R.; Weatherby, N.; Brown, B.; Booth, R.; Williams, M.; et al. Reliability of drug user self-report [Abstract]. Proceedings of the IXth International Conference on AIDS 9(2):776, 1993b.

Fogel-Chance, N. Living in both worlds: "Modernity" and "tradition" among North Slope Inupiaq women in Anchorage. Arctic Anthropology 30:94-108, 1993.

Forney, M.A.; Inciardi, J.A.; and Lockwood, D. Exchanging sex for crack-cocaine: A comparison of women from rural and urban communities. Journal of Community Health 17:73-85, 1992.

Fortuine, R. Chills and fever: Health and diseases in the early history of Alaska. Anchorage: University of Alaska Press, 1989.

Gershman, K.A., and Rolfs, R.T. Diverging gonorrhea and syphilis trends in the 1980s: Are they real? American Journal of Public Health 81:1263-1267, 1991.

Goldsmith, M.F. Sex tied to drugs = STD spread. Journal of the American Medical Association 260:2009, 1988.

Gorwitz, R.J.; Nakashima, A.K.; Moran, J.S.; and Knapp, J.S. Sentinel surveillance for antimicrobial resistance in Neisseria gonorrhoeae—United States, 1988-1991. Morbidity and Mortality Weekly Report 42:29-39, 1993.

Hamilton, L.C., and Seyfrit, C.L. Female flight? Gender balance and outmigration by Native Alaskan villagers. Arctic Medical Research 53:189-193, 1994.

Handsfield, H.H.; Rice, R.J.; Roberts, M.C.; and Holmes, K.K. Localized outbreak of penicillinase-producing Neisseria gonorrhoeae. Journal of the American Medical Association 261:2357-2361, 1989.

Haverkos, H.W. Infectious diseases and drug abuse. Journal of Substance Abuse Treatment 8:269-275, 1991.

Hosmer, D.W., Jr.; and Lemeshow, S. Applied Logistic Regression. New York: John Wiley and Sons, 1989.

Johnson, M.E.; Fisher, D.G.; and Brems, C. "Unmet Treatment Needs of Drug Users in Alaska: Correlates and Societal Costs." Unpublished manuscript, University of Alaska Anchorage, 1995.

Kim, M.Y.; Marmor, M.; Dubin, N.; and Wolfe, H. HIV risk-related sexual behaviors among heterosexuals in New York City: Associations with race, sex, and intravenous drug use. AIDS 7(3):409-414, 1993.

Koester, S.; and Schwartz, J. Crack, gangs, sex, and powerlessness: A view from Denver. In: Ratner, M., ed. Crack Pipe as Pimp: An Ethnographic Investigation of Sex-for-Crack Exchanges. New York: Lexington Books, 1993.

Kuhrt-Hunstiger, T.I., and Fisher, D.G. Hepatitis B: How unique is Alaska? Abstracts of the 122nd Annual Meeting and Exhibition of the American Public Health Association, Washington, DC (3106-310), 1994.

Marx, R.; Aral, S.O.; Rolfs, R.T.; Sterk, C.E.; and Kahn, J.G. Crack, sex, and STD. Sexually Transmitted Diseases 18:92-101, 1991.

Municipality of Anchorage, Community Planning and Development Department. Anchorage Indicators. Anchorage, AK: Municipality of Anchorage, 1993.

National Institute of Allergy and Infectious Diseases. Sexually Transmitted Diseases. Bethesda, MD: National Institute of Allergy and Infectious Diseases, 1992.

National Institute on Drug Abuse. Risk Behavior Assessment. Rockville, MD: National Institute on Drug Abuse (Community Research Branch), 1991.

Needle, R.; Fisher, D.G.; Weatherby, N.; Chitwood, D.; Brown, B.; Cesari, H.; Booth, R.; Williams, M.L.; Watters, J.; Andersen, M.; and Braunstein, M. Reliability of self-reported HIV risk behaviors of drug users. Psychology of Addictive Behaviors 9:242-250, 1995.

Orr, S.M.; Fenaughty, A.M.; and Fisher, D.G. Predictors of Chlamydia trachomatis infection in Alaskan drug users. Abstracts of the 123rd Annual Meeting and Exhibition of the American Public Health Association, San Diego, CA (3004-291), 1995.

Ouellet, L.; Cagle, H.H.; and Fisher, D.G. "Crack" versus "rock" cocaine: The importance of local drug nomenclature in research and education. Contemporary Drug Problems, in press.

Paschane, D.M.; Cagle, H.H.; Fenaughty, A.M.; and Fisher, D.G. Gender differences in risk factors for gonorrhea among Alaskan drug users. Women and Health, in press.

Ratner, M. Sex, drugs, and public policy: Studying and understanding the sex-for-crack phenomenon. In: Ratner, M., ed. Crack Pipe as Pimp: An Ethnographic Investigation of Sex-for-Crack Exchanges. New York: Lexington Books, 1993.

Richert, C.A.; Peterman, T.A.; Zaidi, A.A.; Ransom, R.L.; Wroten, J.E.; and Witte, J.J. A method for identifying persons at high risk for sexually transmitted infections: Opportunity for targeting intervention. American Journal of Public Health 83:520-524, 1993.

SAS Institute Inc. SAS/STAT User's Guide, Version 6. 4th ed. Cary, NC: SAS Institute, 1990.

Schwarcz, S.K.; Bolan, G.A.; Fullilove, M.; McCright, J.; Fullilove, R.; Kohn, R.; and Rolfs, R.T. Crack cocaine and the exchange of sex for money or drugs. Sexually Transmitted Diseases 19:7-13, 1992.

Soule, S. Individual, Community, and Agency Responsibility for the Prevention of Suicide. Division of Mental Health and Developmental Disabilities, Department of Health and Social Services, State of Alaska, 1994.

State of Alaska HIV Prevention Planning Group. HIV Prevention Plan. Anchorage, AK: State of Alaska HIV Prevention Planning Group, 1995.

Steel, E., and Haverkos, H. AIDS and drug abuse in rural America. The Journal of Rural Health 8:70-73, 1992.

Thomas, J.C.; Kulik, A.L.; and Schoenbach, V.J. Syphilis in the South: Rural rates surpass urban rates in North Carolina. American Journal of Public Health 85:1119-1122, 1995.

Umbricht-Schneiter, A.; Ginn, D.H.; Pabst, K.M.; and Bigelow, G.E. Providing medical care to methadone clinic patients: Referral vs. on-site care. American Journal of Public Health 84:207-210, 1994.

Upchurch, D.M.; Brady, W.E.; Reichart, C.A.; and Hook, E.W., III. Behavioral contributions to acquisition of gonorrhea in patients attending an inner city sexually transmitted disease clinic. Journal of Infectious Diseases 161:938-941, 1990.

Watters, J.K. The significance of sampling and understanding hidden populations. Drugs and Society 7:13-21, 1993.

Watters, J., and Biernacki, P. Targeted sampling: Options for the study of hidden populations. Social Problems 36:416-430, 1989.

Weatherby, N.L.; Needle, R.; Cesari, H.; Booth, R.; McCoy, C.B.; Watters, J.K.; Williams, M.; and Chitwood, D.D. Validity of self-reported drug use among injection drug users and crack cocaine users recruited through street outreach. Evaluation and Program Planning 17:347-355, 1994.

Williams, M.L.; McCoy, C.B.; Menon, R.; and Khoury, E.L. Mobility as a factor in the spread of HIV among intravenous drug abusers. In: Brown, B.S., and Beschner, G.M., eds. Handbook on Risk of AIDS. Westport, CN: Greenwood Press, 1993.

Worth, D. Sexual decision-making and AIDS: Why condom promotion among vulnerable women is likely to fail. Studies in Family Planning 20:297-307, 1989.

Yankauer, A. Sexually transmitted diseases: A neglected public health priority. American Journal of Public Health 84:1894-1897, 1994.

Yorke, J.A.; Hethcote, H.W.; and Nold, A. Dynamics and control of the transmission of gonorrhea. Sexually Transmitted Diseases 5:51-56, 1978.

Zhao, Z.; Williams, M.; Rusek, R.; Wild, J.; and Freeman, R. Sexual risk profiles of crack using female sexual partners of injection drug users. Abstracts of the 123rd Annual Meeting and Exhibition of the American Public Health Association, San Diego, CA (3003-290), 1995.

Zinberg, N.E. Drug, Set, and Setting: The Basis for Controlled Intoxicant Use. New Haven: Yale University Press, 1995.

Last Updated 11-7-02