Deren Puerto Rico Cessation 2010
Content
The American Journal of Drug and Alcohol Abuse, 33: 291–299, 2007 Copyright Q Informa Healthcare ISSN: 0095-2990 print/1097-9891 online DOI: 10.1080/00952990601175037
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Predictors of Injection Drug Use Cessation Among Puerto Rican Drug Injectors in New York and Puerto Rico
Sherry Deren, Ph.D.,1 Sung-Yeon Kang, Ph.D.,1 Hector M. Coln, Ph.D., 2 and Rafaela R. Robles Ed.D.2 o
1
Center for Drug Use and HIV Research, National Development and Research Institutes, Inc., New York, New York, USA 2 Center for Addiction Studies, Universidad Central Del Caribe, Bayamn, Puerto Rico o
Abstract: More than half of all AIDS cases among Puerto Ricans have been attributed to injection drug use. Predictors of injection drug use cessation were examined among Puerto Rican injection drug users (IDUs) in New York and Puerto Rico. Analysis of baseline and 6-month follow-up data from 670 IDUs in NY and 316 in PR showed that 47% NY and 20% in PR reported cessation of injection at follow-up (p < .001). In multivariate analyses, having been in drug treatment since baseline was the only significant predictor of cessation for both sites (NY: AOR ¼ 1.80; PR: AOR ¼ 3.10). Increasing availability of methadone maintenance treatment, especially in PR, was indicated.
Keywords: Injection cessation, injection drug use, HIV=AIDS, Puerto Ricans
INTRODUCTION Injection drug use is the primary risk factor for HIV among Puerto Ricans in New York City and Puerto Rico. In both locations, approximately half of all AIDS cases among Puerto Ricans have been attributed to injection drug use (1, 2). As compared to Puerto Rican injectors in New York, riskier injection-related practices have been identified among
Address correspondence to Sherry Deren, Center for Drug Use and HIV Research, National Development and Research Institutes, Inc., 71 West 23rd St., New York, NY, USA. E-mail: [email protected] 291
292
S. Deren et al.
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
injectors in Puerto Rico, related to a higher frequency of injection (3) and lower access to drug treatment and other health services (4). In addition to HIV transmission, injection drug use is also related to Hepatitis B and C, and other adverse medical and social consequences (5, 6). Few longitudinal studies, however, have been conducted on these high-risk injection drug users to assess predictors of cessation of injection, and none have compared predictors of cessation in these 2 very different risk environments. It is important to identify factors related to cessation of injection, to inform drug and HIV prevention efforts, and to enhance efforts to sustain periods of cessation (7). Drug treatment, particularly methadone maintenance treatment, has been found to be one of the strongest predictors of cessation of injection (8). The significant differences in availability of drug treatment in New York and Puerto Rico (4) provides an opportunity to assess the importance of drug treatment involvement for cessation in environments with different levels of availability of treatment. Network characteristics of injectors have also been found to be related to cessation, in particular, having a smaller proportion of drug users in one’s network was found to be related to cessation of drug use (9) and there is some evidence (10, 11) that the composition of networks differs in the two locations. In addition, most studies of injectors focus on a particular cohort, recruited within a single time period. Examining predictors of cessation for cohorts recruited in different time periods can help increase understanding of the types of interventions which may lead to reductions in injection drug use and thus can be useful in the development or expansion of programs to reduce HIV=AIDS among injectors. This article will examine predictors of cessation of injection over a 6-month period, among Puerto Rican drug injectors in New York and Puerto Rico, recruited during two different time periods.
METHODS Data were collected as part of the ARIBBA study (Alliance for Research in El Barrio and Bayamn), a study of Puerto Rican injection drug users o (IDUs) and crack smokers recruited in East Harlem, NY and Bayamn, o PR. Participants were recruited during two time periods, 1998–1999 and 2002–2003. Recruitment of participants was based on targeted sampling, in locations in the two communities where drug users could be located (12). Criteria for participation were: aged 18 or over, identification as Puerto Rican, and reported injecting drugs or crack smoking in the prior 30 days. Participants were recruited by outreach workers, brought to the local field sites for verification of recent drug user (using ORASURE,
Predictors of Injection Cessation
293
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
only those testing positive for opiates for cocaine were eligible), followed by informed consent procedures and administration of the baseline interview (13). All participants were followed up and re-interviewed 6 months later. Follow-up rates were 74% in New York and 83% in Puerto Rico, and few differences in socio-demographics were found between those followed-up and those who were not followed-up (14). HIV pre-test counseling and testing were offered at each interview, and those who were tested were advised to return within approximately one week, to receive their test results and post-test counseling. This analysis is based only on those who reported injection drug use during the 30 days prior to baseline recruitment and were interviewed at the 6-month follow-up, a total of 986 injection drug users (IDUs). Sample sizes by recruitment date cohort are as follows: 1998–1999 (Cohort I, total n ¼ 670; NY ¼ 417, PR ¼ 253) and 2002–2003 (Cohort II, total n ¼ 316; NY ¼ 169, PR ¼ 147). Baseline (T1) and 6-month (T2) interviews collected data on demographics, injection drug use, and potential predictors of cessation of injecting at T2. Cessation of injection was defined as reporting not injecting during the 30 days prior to the T2 interview. Variables selected as potential predictors of cessation included sociodemographic variables (sex, age, homelessness, Cohort); HIV status; HIV risk and injection-related variables (drug treatment program participation [between T1 and T2], needle exchange program use and injection frequency); drug use-related variables (crack use, drugs injected [injection of heroin only, cocaine only or both together-speedball]), and networkrelated variables (percent of family and of friends who were injectors, and incarceration between T1 and T2). Cohort, i.e., the years of recruitment, was included as a variable to assess temporal influences in predictors of cessation. Incarceration was included as a network variable because of prior research indicating the high prevalence rates of IDUs who have been incarcerated (15) and the relationship found between gang affiliation and injection drug use while incarcerated (16). All behavioral risk and drug use variables were assessed for the 30 days prior to baseline.
ANALYSIS Due to the significant differences found between NY and PR in many of the predictor variables (13), all analyses were conducted by site. Bivariate predictors of injection drug use cessation were identified. Variables which were significant predictors in either site (at p < .10) were included in the site-specific multivariate models.
294
S. Deren et al.
RESULTS All four cohorts, in NY and PR, were predominately male, over 75% (see Table 1). The mean age of the NY samples was older than the PR sample (38 vs. 34) and there was a decline in age across the two NY cohorts (38.9 to 37.1, p < .01). Homelessness showed an increase across the two NY cohorts and HIV prevalence declined (p < .001 for both variables). As has been found in prior analyses of data from this study, the participants in NY were more likely to utilize NEP and MMTP services (and there was little use of other drug treatment modalities) (17, 4). In the PR cohorts, there was an increase from Cohort I to Cohort II in the percent enrolled in methadone maintenance treatment, from 12% to 21% (p < .05). At T2, a total of 47% of the IDUs recruited in NY reported cessation of injecting as did 20% in PR (p < .001). None of the sociodemographic variables examined in multivariate analyses, except for Cohort (i.e., date of recruitment), were significantly related to cessation of injection at T2 (Table 2). Cohort was significant in NY, with those who stopped injecting more likely to be in Cohort II (36% vs. 23%). Bivariate predictors in NY related to risk and injection behaviors included NEP use (46% of those who ceased injecting reported NEP use during the 30 days prior to the baseline interview, compared to 64% of those who continued to inject). Being in any drug treatment program between baseline and follow-up and injection frequency were related to cessation of injection drug use in both sites. Several drug use-related variables were related
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Table 1. Sociodemographic and other baseline characteristics of 2 cohorts of Puerto Rican injection drug users in New York and Puerto Rico New York Cohort I (n ¼ 417) Gender (male) Age (x, SD) Homeless HIVþ NEP usea MMTPa Other drug treatmenta
a
Puerto Rico Cohort I (n ¼ 253) 86% 33.7 (8.2) 19 23 39 12 0 Cohort II (n ¼ 147) 82% 33.5 (7.9) 21 18 48 21Ã 1
Cohort II (n ¼ 169) 81% 37.1 (8.2)ÃÃ 38ÃÃÃ 15ÃÃÃ 57 70 2
76% 38.9 (7.9) 20 29 55 62 1
Ã
During the 30 days prior to baseline interview. p < .05; ÃÃ p < .01; ÃÃÃ p < .001.
Predictors of Injection Cessation
295
Table 2. Bivariate baseline predictors of cessation of injection drug use, by site New York (n ¼ 586)
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Puerto Rico (n ¼ 400) Continued (n ¼ 320) 85 33.8 (8) 21 20 38 44 31 200.5 (159.7) 29 66 62 92 10 43 21 Stopped (n ¼ 80) 83 32.8 (8.4) 16 25 30 35 55ÃÃÃ 147.1 (130.8)ÃÃ 34 64 76Ã 86 10 43 35ÃÃ
Predictors Socio-demographics Gender (% male) Age (mean, SD) Homeless (%) HIV þ Cohort II Risk and injection-related NEP usea (%) Any drug treatmentb (%) Injection frequencya (mean, SD) Drug use-related Crack usea Cocaine injectiona (%) Heroin injectiona (%) Speedball injectiona (%) Network-related IDUs in network of family membersa (%) IDUs in network of friendsa (%) Incarcerationb (%)
a
Continued (n ¼ 308) 76 38 (8.1) 24 24 23 64 76 88.1 (95.9) 40 56 86 60 20 63 33
Stopped (n ¼ 278) 79 38.8 (8) 26 25 36ÃÃÃ 46ÃÃÃ 82y 61.4 (96.5)ÃÃÃ 47y 39ÃÃÃ 85 48ÃÃ 15 46ÃÃÃ 35
During the 30 days prior to baseline interview; bBetween baseline and followup interviews. y p < .10; Ã p < .05; ÃÃ p < .01; ÃÃÃ p < .001.
to cessation: in NY, crack use at baseline was related to cessation of injection at T2, and injection of cocaine alone or speedball injection was inversely related to cessation; in PR, injection of heroin alone was positively related to cessation. In terms of the network variables, for participants in NY, having IDUs in their network of friends was related to continued injection. Finally, being incarcerated was related to cessation of injection for participants from PR. In the multivariate analysis, only one predictor emerged that was significant for both sites: Having been in drug treatment between T1 and T2 (AOR ¼ 1.801 in NY and 3.100 in PR) (see Table 3). Other significant predictors in NY, related to cessation of injection were: Cohort
296
S. Deren et al.
Table 3. Multivariate models: Predictors of cessation of injection drug use, by site New York AOR
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Puerto Rico AOR
ÃÃÃ
95% CI 1.374–3.018 .395–.816ÃÃÃ 1.153–2.815ÃÃ .996–1.000 1.001–2.087Ã .414–.920Ã .355–.731ÃÃÃ .909–1.940
95% CI .377–1.199 .550–1.716 1.808–5.137ÃÃÃ .995–1.000Ã .733–2.324 .505–1.753 .652–1.896 1.336–4.251ÃÃ
Cohort Used NEPa Drug treatmentb Injection frequencya Crack usea Injection cocaine alonea Had IDU friendsa Incarcerationb
a
2.036 .568 1.801 .998 1.446 .617 .509 1.328
.673 .972 3.100 .997 1.305 .941 1.111 2.384
Ã
During the 30 days prior to base line interview, bBetween and follow-up interviews. p < .05; ÃÃ p < .01; ÃÃÃ p < .001.
(AOR ¼ 2.036), NEP use (AOR ¼ .568), and crack use (AOR ¼ 1.446). Predictors of continued injection were: Injected cocaine only (AOR ¼ .617), and had IDU friends (AOR ¼ .509). In PR, the only other variables related to cessation of injection, other than drug treatment, were injection frequency (AOR ¼ .997) and having been incarcerated between T1 and T2 (AOR ¼ 2.384). DISCUSSION About half of Puerto Rican IDUs in NY reported cessation of injecting at a 6-month follow-up, more than twice as many as their counterparts in PR. This higher rate of cessation in NY is likely due to the higher enrollment in methadone treatment, which was a significant predictor of cessation in both sites. In NY, having IDU friends halved the odds of ceasing injection drug use, indicating the importance of network influences on drug use behaviors. In NY, those who used crack and those who injected cocaine (and not heroin) were also less likely to cease injection drug use, showing the continued need for additional treatment efforts for cocaine users. Two variables which were significant in the NY sample require further investigation. The finding that those who reported using an NEP at baseline were less likely to stop injecting may have been related to the fact that those who use NEPs have higher levels of addiction (e.g., inject more frequently) or have a more erratic pattern of drug use (18,19). The Cohort effect, i.e., that those recruited in Cohort II in NY were more likely to stop injecting, may be related to factors associated with the other differences between the two NY Cohorts (e.g., younger
Predictors of Injection Cessation
297
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
age, higher homelessness and lower HIV þ rates in Cohort II), which were not assessed. While being incarcerated was related to cessation of injection at T2, given the prior finding that injectors incarcerated in PR were more likely to continue risky injecting (while incarcerated) than those in NY (16, 20), a longer post-incarceration follow-up period of study may be needed to assess the longer term impact of incarceration on continued injection. There are several limitations to these data, including the limitations inherent in self-report data and in the fact that a random, representative sample of drug users is difficult to obtain. Neighborhood changes from across the two-time periods studied may have impacted recruitment efforts, contributing to reduced comparability across Cohorts. However, efforts to conduct targeted sampling across both time periods, using the same outreach staff, were likely to reduce potential sampling biases. Despite the limitations, several important implications emerge from these findings. Cessation of injection can be facilitated through increasing availability and enrollment in drug treatment, especially methadone maintenance treatment. Although this supports extensive prior literature, increasing availability of methadone treatment in PR was identified as particularly needed. Even for those whose benefit from MMTP is a reduction in injection frequency (8), lower injection frequency was also found to be related to subsequent cessation and lower frequency is related to lower risk of HIV transmission. The impact of Cohort points to the need to examine findings in terms of changes in populations over time. Research on predictors of maintenance of cessation (21) and of relapse to injecting, over longer time periods, is needed. Finally, differences in the significant relationships found in the two locations points to the importance of examining social and structural differences as part of studying and planning methods to enhance HIV prevention methods. ACKNOWLEDGMENT Funding for this study was provided by the National Institute on Drug Abuse, Grant No. R01 DA010425. REFERENCES
1. Centers for Disease Control and Prevention (CDC). Cases of HIV infection and AIDS in the United States, 2002. HIV=AIDS Surveillance Report. Atlanta, GA, 2002. 2. Puerto Rico Department of Health. HIV=AIDS Surveillance Report, April. San Juan, Puerto Rico: Puerto Rico Department of Health, 2003.
298
S. Deren et al.
3. Coln HM, Robles RR, Deren S, Sahai H, Finlinson HA, Andia JF, Cruz o MA, Kang S-Y, Oliver-Velez D. Between-city variation in frequency of injection among Puerto Rican injection drug users: East Harlem, New York, and Bayamn, Puerto Rico. J Acquired Immune Deficiency Syndromes 2002; o 27:405–413. ´ 4. Robles RR, Matos TD, Coln HM, Deren S, Reyes JC, Andıa JF, o Marrero CA, Sahai H. Determinants of health care use among Puerto Rican drug users in Puerto Rico and New York City. Clinical Infectious Diseases 2003; 37:S392–S403. 5. Brown PD, Levine DP. Infections in injection drug users. Infectious Disease Clinics of North America 2002; 16(Special Issue). 6. Hagan H, Thiede H, Weiss NS, Hopkins SG, Duchin JS, Alexander ER. Sharing of drug preparation equipment as a risk factor for hepatitis C. Am J Public Health 2001; 91:42–46. 7. Galai N, Safaeian M, Vlahov D, Bolotin A, Celentano D, the ALIVE Study. Longitudinal patterns of drug injection behavior in the ALIVE Study cohort. 1988–2000: Description and determinants. Amer J Epidemiology 2003; 30:61–83. 8. Sorensen JL, Copeland AL. Drug abuse treatment as an HIV prevention strategy: A review. Drug Alcohol Depend 2000; 59:17–31. 9. Latkin C, Knowlton A, Hoover D, Mandell W. Drug network characteristics as a predictor of cessation of drug use among adult injection drug users: A prospective study. Amer J Drug Alcohol Abuse 1999; 25:463–473. ´ 10. Kang S-Y, Deren S, Coln HM, Andıa JF, Robles RR. (2004). Impact of risk o network change on cessation of injection drug use among Puerto Rican drug injectors in New York. Paper presented at the American Public Health Association Annual Meeting, Washington, DC, November. 11. Kang S-Y, Deren S, Andia JF, Coln HM, Robles RR. Egocentric HIV risk o networks among Puerto Rican crack users in New York and Puerto Rico: Impact on sex risk behaviors over time. AIDS Edu Prev 2005; 17:53–67. ´ 12. Oliver-Velez D, Finlinson HA, Deren S, Robles RR, Coln HM, Andıa JF. o Mapping the air-bridge locations: The application of ethnographic mapping techniques to dual-site study of HIV risk behavior determinants in East Harlem, New York and Bayamn, Puerto Rico. Human Organization o 2002; 61:262–276. 13. Deren S, Oliver-Velez D, Finlinson HA, Robles RR, Andia JF, Coln HM, o Kang S-Y, Shedlin M. Integrating qualitative and quantitative methods: Comparing HIV-related risk behaviors among Puerto Rican drug users in Puerto Rico and New York. Substance Use and Misuse 2003; 38:1–24. ´ 14. Kang S-Y, Deren S, Andıa JF, Coln HM, Robles RR. Effects of changes in o perceived self-efficacy on HIV risk behaviors over time. Addictive Behaviors 2004; 29:567–574. ´ 15. Estebanez P, Zunzunegui MV, Aguilar MD, Russell N, Cifuentes I, Hankins C. The role of prisons in the HIV epidemic among female injecting drug users. AIDS Care 2002; 14:95–104. 16. Andia JF, Deren S, Robles RR, Kang S-Y, Coln HM, Oliver-Velez D, o Finlinson HA. Factors associated with injection and non-injection drug use
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Predictors of Injection Cessation
299
17.
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
18.
19. 20.
21.
during incarceration among Puerto Rican drug injectors in New York and Puerto Rico. The Prison Journal 2005; 85:329–342. ´ Finlinson HA, Oliver-Velez D, Coln HM, Deren S, Robles RR, o ´ Beardsley M, Cant JGH, Andıa JF, Soto-Lpez M. Syringe acquisition o and use of syringe exchange programs by Puerto Rican drug injectors in New York and Puerto Rico: Comparisons based on quantitative and qualitative methods. AIDS Behav 2000; 4:341–351. Schechter M, Strathdee S, Cornelisse P, Currie S, Patrick D, Rekart M, O’Shaughnessy M. Do needle exchange programmes increase the spread of HIV among injection drug users?: An investigation of the Vancouver outbreak. AIDS 1999; 13:45–51. Steensma C, Boivin J-F, Blais L, Roy E. Cessation of injecting drug use among street-based youth. J Urban Health 2005; 82:622–637 Kang S-Y, Deren S, Andia JF, Coln HM, Robles RR, Oliver-Velez D. HIV o transmission behaviors in jail=prison among Puerto Rican drug injectors in New York and Puerto Rico. AIDS Behav 2005; 9:377–338 Bruneau J, Brogly S, Tyndall M, Lamothe F, Franco E. Intensity of drug injection as a determinant of sustained injection cessation among chronic drug users: The interface with social factors and service utilization. Addiction 2004; 99:727–737.
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Predictors of Injection Drug Use Cessation Among Puerto Rican Drug Injectors in New York and Puerto Rico
Sherry Deren, Ph.D.,1 Sung-Yeon Kang, Ph.D.,1 Hector M. Coln, Ph.D., 2 and Rafaela R. Robles Ed.D.2 o
1
Center for Drug Use and HIV Research, National Development and Research Institutes, Inc., New York, New York, USA 2 Center for Addiction Studies, Universidad Central Del Caribe, Bayamn, Puerto Rico o
Abstract: More than half of all AIDS cases among Puerto Ricans have been attributed to injection drug use. Predictors of injection drug use cessation were examined among Puerto Rican injection drug users (IDUs) in New York and Puerto Rico. Analysis of baseline and 6-month follow-up data from 670 IDUs in NY and 316 in PR showed that 47% NY and 20% in PR reported cessation of injection at follow-up (p < .001). In multivariate analyses, having been in drug treatment since baseline was the only significant predictor of cessation for both sites (NY: AOR ¼ 1.80; PR: AOR ¼ 3.10). Increasing availability of methadone maintenance treatment, especially in PR, was indicated.
Keywords: Injection cessation, injection drug use, HIV=AIDS, Puerto Ricans
INTRODUCTION Injection drug use is the primary risk factor for HIV among Puerto Ricans in New York City and Puerto Rico. In both locations, approximately half of all AIDS cases among Puerto Ricans have been attributed to injection drug use (1, 2). As compared to Puerto Rican injectors in New York, riskier injection-related practices have been identified among
Address correspondence to Sherry Deren, Center for Drug Use and HIV Research, National Development and Research Institutes, Inc., 71 West 23rd St., New York, NY, USA. E-mail: [email protected] 291
292
S. Deren et al.
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
injectors in Puerto Rico, related to a higher frequency of injection (3) and lower access to drug treatment and other health services (4). In addition to HIV transmission, injection drug use is also related to Hepatitis B and C, and other adverse medical and social consequences (5, 6). Few longitudinal studies, however, have been conducted on these high-risk injection drug users to assess predictors of cessation of injection, and none have compared predictors of cessation in these 2 very different risk environments. It is important to identify factors related to cessation of injection, to inform drug and HIV prevention efforts, and to enhance efforts to sustain periods of cessation (7). Drug treatment, particularly methadone maintenance treatment, has been found to be one of the strongest predictors of cessation of injection (8). The significant differences in availability of drug treatment in New York and Puerto Rico (4) provides an opportunity to assess the importance of drug treatment involvement for cessation in environments with different levels of availability of treatment. Network characteristics of injectors have also been found to be related to cessation, in particular, having a smaller proportion of drug users in one’s network was found to be related to cessation of drug use (9) and there is some evidence (10, 11) that the composition of networks differs in the two locations. In addition, most studies of injectors focus on a particular cohort, recruited within a single time period. Examining predictors of cessation for cohorts recruited in different time periods can help increase understanding of the types of interventions which may lead to reductions in injection drug use and thus can be useful in the development or expansion of programs to reduce HIV=AIDS among injectors. This article will examine predictors of cessation of injection over a 6-month period, among Puerto Rican drug injectors in New York and Puerto Rico, recruited during two different time periods.
METHODS Data were collected as part of the ARIBBA study (Alliance for Research in El Barrio and Bayamn), a study of Puerto Rican injection drug users o (IDUs) and crack smokers recruited in East Harlem, NY and Bayamn, o PR. Participants were recruited during two time periods, 1998–1999 and 2002–2003. Recruitment of participants was based on targeted sampling, in locations in the two communities where drug users could be located (12). Criteria for participation were: aged 18 or over, identification as Puerto Rican, and reported injecting drugs or crack smoking in the prior 30 days. Participants were recruited by outreach workers, brought to the local field sites for verification of recent drug user (using ORASURE,
Predictors of Injection Cessation
293
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
only those testing positive for opiates for cocaine were eligible), followed by informed consent procedures and administration of the baseline interview (13). All participants were followed up and re-interviewed 6 months later. Follow-up rates were 74% in New York and 83% in Puerto Rico, and few differences in socio-demographics were found between those followed-up and those who were not followed-up (14). HIV pre-test counseling and testing were offered at each interview, and those who were tested were advised to return within approximately one week, to receive their test results and post-test counseling. This analysis is based only on those who reported injection drug use during the 30 days prior to baseline recruitment and were interviewed at the 6-month follow-up, a total of 986 injection drug users (IDUs). Sample sizes by recruitment date cohort are as follows: 1998–1999 (Cohort I, total n ¼ 670; NY ¼ 417, PR ¼ 253) and 2002–2003 (Cohort II, total n ¼ 316; NY ¼ 169, PR ¼ 147). Baseline (T1) and 6-month (T2) interviews collected data on demographics, injection drug use, and potential predictors of cessation of injecting at T2. Cessation of injection was defined as reporting not injecting during the 30 days prior to the T2 interview. Variables selected as potential predictors of cessation included sociodemographic variables (sex, age, homelessness, Cohort); HIV status; HIV risk and injection-related variables (drug treatment program participation [between T1 and T2], needle exchange program use and injection frequency); drug use-related variables (crack use, drugs injected [injection of heroin only, cocaine only or both together-speedball]), and networkrelated variables (percent of family and of friends who were injectors, and incarceration between T1 and T2). Cohort, i.e., the years of recruitment, was included as a variable to assess temporal influences in predictors of cessation. Incarceration was included as a network variable because of prior research indicating the high prevalence rates of IDUs who have been incarcerated (15) and the relationship found between gang affiliation and injection drug use while incarcerated (16). All behavioral risk and drug use variables were assessed for the 30 days prior to baseline.
ANALYSIS Due to the significant differences found between NY and PR in many of the predictor variables (13), all analyses were conducted by site. Bivariate predictors of injection drug use cessation were identified. Variables which were significant predictors in either site (at p < .10) were included in the site-specific multivariate models.
294
S. Deren et al.
RESULTS All four cohorts, in NY and PR, were predominately male, over 75% (see Table 1). The mean age of the NY samples was older than the PR sample (38 vs. 34) and there was a decline in age across the two NY cohorts (38.9 to 37.1, p < .01). Homelessness showed an increase across the two NY cohorts and HIV prevalence declined (p < .001 for both variables). As has been found in prior analyses of data from this study, the participants in NY were more likely to utilize NEP and MMTP services (and there was little use of other drug treatment modalities) (17, 4). In the PR cohorts, there was an increase from Cohort I to Cohort II in the percent enrolled in methadone maintenance treatment, from 12% to 21% (p < .05). At T2, a total of 47% of the IDUs recruited in NY reported cessation of injecting as did 20% in PR (p < .001). None of the sociodemographic variables examined in multivariate analyses, except for Cohort (i.e., date of recruitment), were significantly related to cessation of injection at T2 (Table 2). Cohort was significant in NY, with those who stopped injecting more likely to be in Cohort II (36% vs. 23%). Bivariate predictors in NY related to risk and injection behaviors included NEP use (46% of those who ceased injecting reported NEP use during the 30 days prior to the baseline interview, compared to 64% of those who continued to inject). Being in any drug treatment program between baseline and follow-up and injection frequency were related to cessation of injection drug use in both sites. Several drug use-related variables were related
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Table 1. Sociodemographic and other baseline characteristics of 2 cohorts of Puerto Rican injection drug users in New York and Puerto Rico New York Cohort I (n ¼ 417) Gender (male) Age (x, SD) Homeless HIVþ NEP usea MMTPa Other drug treatmenta
a
Puerto Rico Cohort I (n ¼ 253) 86% 33.7 (8.2) 19 23 39 12 0 Cohort II (n ¼ 147) 82% 33.5 (7.9) 21 18 48 21Ã 1
Cohort II (n ¼ 169) 81% 37.1 (8.2)ÃÃ 38ÃÃÃ 15ÃÃÃ 57 70 2
76% 38.9 (7.9) 20 29 55 62 1
Ã
During the 30 days prior to baseline interview. p < .05; ÃÃ p < .01; ÃÃÃ p < .001.
Predictors of Injection Cessation
295
Table 2. Bivariate baseline predictors of cessation of injection drug use, by site New York (n ¼ 586)
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Puerto Rico (n ¼ 400) Continued (n ¼ 320) 85 33.8 (8) 21 20 38 44 31 200.5 (159.7) 29 66 62 92 10 43 21 Stopped (n ¼ 80) 83 32.8 (8.4) 16 25 30 35 55ÃÃÃ 147.1 (130.8)ÃÃ 34 64 76Ã 86 10 43 35ÃÃ
Predictors Socio-demographics Gender (% male) Age (mean, SD) Homeless (%) HIV þ Cohort II Risk and injection-related NEP usea (%) Any drug treatmentb (%) Injection frequencya (mean, SD) Drug use-related Crack usea Cocaine injectiona (%) Heroin injectiona (%) Speedball injectiona (%) Network-related IDUs in network of family membersa (%) IDUs in network of friendsa (%) Incarcerationb (%)
a
Continued (n ¼ 308) 76 38 (8.1) 24 24 23 64 76 88.1 (95.9) 40 56 86 60 20 63 33
Stopped (n ¼ 278) 79 38.8 (8) 26 25 36ÃÃÃ 46ÃÃÃ 82y 61.4 (96.5)ÃÃÃ 47y 39ÃÃÃ 85 48ÃÃ 15 46ÃÃÃ 35
During the 30 days prior to baseline interview; bBetween baseline and followup interviews. y p < .10; Ã p < .05; ÃÃ p < .01; ÃÃÃ p < .001.
to cessation: in NY, crack use at baseline was related to cessation of injection at T2, and injection of cocaine alone or speedball injection was inversely related to cessation; in PR, injection of heroin alone was positively related to cessation. In terms of the network variables, for participants in NY, having IDUs in their network of friends was related to continued injection. Finally, being incarcerated was related to cessation of injection for participants from PR. In the multivariate analysis, only one predictor emerged that was significant for both sites: Having been in drug treatment between T1 and T2 (AOR ¼ 1.801 in NY and 3.100 in PR) (see Table 3). Other significant predictors in NY, related to cessation of injection were: Cohort
296
S. Deren et al.
Table 3. Multivariate models: Predictors of cessation of injection drug use, by site New York AOR
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Puerto Rico AOR
ÃÃÃ
95% CI 1.374–3.018 .395–.816ÃÃÃ 1.153–2.815ÃÃ .996–1.000 1.001–2.087Ã .414–.920Ã .355–.731ÃÃÃ .909–1.940
95% CI .377–1.199 .550–1.716 1.808–5.137ÃÃÃ .995–1.000Ã .733–2.324 .505–1.753 .652–1.896 1.336–4.251ÃÃ
Cohort Used NEPa Drug treatmentb Injection frequencya Crack usea Injection cocaine alonea Had IDU friendsa Incarcerationb
a
2.036 .568 1.801 .998 1.446 .617 .509 1.328
.673 .972 3.100 .997 1.305 .941 1.111 2.384
Ã
During the 30 days prior to base line interview, bBetween and follow-up interviews. p < .05; ÃÃ p < .01; ÃÃÃ p < .001.
(AOR ¼ 2.036), NEP use (AOR ¼ .568), and crack use (AOR ¼ 1.446). Predictors of continued injection were: Injected cocaine only (AOR ¼ .617), and had IDU friends (AOR ¼ .509). In PR, the only other variables related to cessation of injection, other than drug treatment, were injection frequency (AOR ¼ .997) and having been incarcerated between T1 and T2 (AOR ¼ 2.384). DISCUSSION About half of Puerto Rican IDUs in NY reported cessation of injecting at a 6-month follow-up, more than twice as many as their counterparts in PR. This higher rate of cessation in NY is likely due to the higher enrollment in methadone treatment, which was a significant predictor of cessation in both sites. In NY, having IDU friends halved the odds of ceasing injection drug use, indicating the importance of network influences on drug use behaviors. In NY, those who used crack and those who injected cocaine (and not heroin) were also less likely to cease injection drug use, showing the continued need for additional treatment efforts for cocaine users. Two variables which were significant in the NY sample require further investigation. The finding that those who reported using an NEP at baseline were less likely to stop injecting may have been related to the fact that those who use NEPs have higher levels of addiction (e.g., inject more frequently) or have a more erratic pattern of drug use (18,19). The Cohort effect, i.e., that those recruited in Cohort II in NY were more likely to stop injecting, may be related to factors associated with the other differences between the two NY Cohorts (e.g., younger
Predictors of Injection Cessation
297
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
age, higher homelessness and lower HIV þ rates in Cohort II), which were not assessed. While being incarcerated was related to cessation of injection at T2, given the prior finding that injectors incarcerated in PR were more likely to continue risky injecting (while incarcerated) than those in NY (16, 20), a longer post-incarceration follow-up period of study may be needed to assess the longer term impact of incarceration on continued injection. There are several limitations to these data, including the limitations inherent in self-report data and in the fact that a random, representative sample of drug users is difficult to obtain. Neighborhood changes from across the two-time periods studied may have impacted recruitment efforts, contributing to reduced comparability across Cohorts. However, efforts to conduct targeted sampling across both time periods, using the same outreach staff, were likely to reduce potential sampling biases. Despite the limitations, several important implications emerge from these findings. Cessation of injection can be facilitated through increasing availability and enrollment in drug treatment, especially methadone maintenance treatment. Although this supports extensive prior literature, increasing availability of methadone treatment in PR was identified as particularly needed. Even for those whose benefit from MMTP is a reduction in injection frequency (8), lower injection frequency was also found to be related to subsequent cessation and lower frequency is related to lower risk of HIV transmission. The impact of Cohort points to the need to examine findings in terms of changes in populations over time. Research on predictors of maintenance of cessation (21) and of relapse to injecting, over longer time periods, is needed. Finally, differences in the significant relationships found in the two locations points to the importance of examining social and structural differences as part of studying and planning methods to enhance HIV prevention methods. ACKNOWLEDGMENT Funding for this study was provided by the National Institute on Drug Abuse, Grant No. R01 DA010425. REFERENCES
1. Centers for Disease Control and Prevention (CDC). Cases of HIV infection and AIDS in the United States, 2002. HIV=AIDS Surveillance Report. Atlanta, GA, 2002. 2. Puerto Rico Department of Health. HIV=AIDS Surveillance Report, April. San Juan, Puerto Rico: Puerto Rico Department of Health, 2003.
298
S. Deren et al.
3. Coln HM, Robles RR, Deren S, Sahai H, Finlinson HA, Andia JF, Cruz o MA, Kang S-Y, Oliver-Velez D. Between-city variation in frequency of injection among Puerto Rican injection drug users: East Harlem, New York, and Bayamn, Puerto Rico. J Acquired Immune Deficiency Syndromes 2002; o 27:405–413. ´ 4. Robles RR, Matos TD, Coln HM, Deren S, Reyes JC, Andıa JF, o Marrero CA, Sahai H. Determinants of health care use among Puerto Rican drug users in Puerto Rico and New York City. Clinical Infectious Diseases 2003; 37:S392–S403. 5. Brown PD, Levine DP. Infections in injection drug users. Infectious Disease Clinics of North America 2002; 16(Special Issue). 6. Hagan H, Thiede H, Weiss NS, Hopkins SG, Duchin JS, Alexander ER. Sharing of drug preparation equipment as a risk factor for hepatitis C. Am J Public Health 2001; 91:42–46. 7. Galai N, Safaeian M, Vlahov D, Bolotin A, Celentano D, the ALIVE Study. Longitudinal patterns of drug injection behavior in the ALIVE Study cohort. 1988–2000: Description and determinants. Amer J Epidemiology 2003; 30:61–83. 8. Sorensen JL, Copeland AL. Drug abuse treatment as an HIV prevention strategy: A review. Drug Alcohol Depend 2000; 59:17–31. 9. Latkin C, Knowlton A, Hoover D, Mandell W. Drug network characteristics as a predictor of cessation of drug use among adult injection drug users: A prospective study. Amer J Drug Alcohol Abuse 1999; 25:463–473. ´ 10. Kang S-Y, Deren S, Coln HM, Andıa JF, Robles RR. (2004). Impact of risk o network change on cessation of injection drug use among Puerto Rican drug injectors in New York. Paper presented at the American Public Health Association Annual Meeting, Washington, DC, November. 11. Kang S-Y, Deren S, Andia JF, Coln HM, Robles RR. Egocentric HIV risk o networks among Puerto Rican crack users in New York and Puerto Rico: Impact on sex risk behaviors over time. AIDS Edu Prev 2005; 17:53–67. ´ 12. Oliver-Velez D, Finlinson HA, Deren S, Robles RR, Coln HM, Andıa JF. o Mapping the air-bridge locations: The application of ethnographic mapping techniques to dual-site study of HIV risk behavior determinants in East Harlem, New York and Bayamn, Puerto Rico. Human Organization o 2002; 61:262–276. 13. Deren S, Oliver-Velez D, Finlinson HA, Robles RR, Andia JF, Coln HM, o Kang S-Y, Shedlin M. Integrating qualitative and quantitative methods: Comparing HIV-related risk behaviors among Puerto Rican drug users in Puerto Rico and New York. Substance Use and Misuse 2003; 38:1–24. ´ 14. Kang S-Y, Deren S, Andıa JF, Coln HM, Robles RR. Effects of changes in o perceived self-efficacy on HIV risk behaviors over time. Addictive Behaviors 2004; 29:567–574. ´ 15. Estebanez P, Zunzunegui MV, Aguilar MD, Russell N, Cifuentes I, Hankins C. The role of prisons in the HIV epidemic among female injecting drug users. AIDS Care 2002; 14:95–104. 16. Andia JF, Deren S, Robles RR, Kang S-Y, Coln HM, Oliver-Velez D, o Finlinson HA. Factors associated with injection and non-injection drug use
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
Predictors of Injection Cessation
299
17.
Am J Drug Alcohol Abuse Downloaded from informahealthcare.com by JHU John Hopkins University on 09/16/10 For personal use only.
18.
19. 20.
21.
during incarceration among Puerto Rican drug injectors in New York and Puerto Rico. The Prison Journal 2005; 85:329–342. ´ Finlinson HA, Oliver-Velez D, Coln HM, Deren S, Robles RR, o ´ Beardsley M, Cant JGH, Andıa JF, Soto-Lpez M. Syringe acquisition o and use of syringe exchange programs by Puerto Rican drug injectors in New York and Puerto Rico: Comparisons based on quantitative and qualitative methods. AIDS Behav 2000; 4:341–351. Schechter M, Strathdee S, Cornelisse P, Currie S, Patrick D, Rekart M, O’Shaughnessy M. Do needle exchange programmes increase the spread of HIV among injection drug users?: An investigation of the Vancouver outbreak. AIDS 1999; 13:45–51. Steensma C, Boivin J-F, Blais L, Roy E. Cessation of injecting drug use among street-based youth. J Urban Health 2005; 82:622–637 Kang S-Y, Deren S, Andia JF, Coln HM, Robles RR, Oliver-Velez D. HIV o transmission behaviors in jail=prison among Puerto Rican drug injectors in New York and Puerto Rico. AIDS Behav 2005; 9:377–338 Bruneau J, Brogly S, Tyndall M, Lamothe F, Franco E. Intensity of drug injection as a determinant of sustained injection cessation among chronic drug users: The interface with social factors and service utilization. Addiction 2004; 99:727–737.
