More than twenty-five years into the pandemic, HIV infection continues to spread worldwide. The most recent epidemic has emerged in Eastern Europe. In 2001, the Russian Federation reported 60.24 new HIV cases per 100,000 population; Estonia reported 49.17 and Latvia 33.29 1. Although the number of new cases in 2006 had declined to 27 per 100,000 in Russia and 13 in Latvia, the incidence rate in Estonia is still notably high at 50 per 100,000 1. Injection drug use is the main cause of this prolonged HIV outbreak and according to the Joint United Nations Programme on HIV/AIDS the sharing of needles and syringes (referred to collectively, for the purposes of this study, as "syringes") are the key factors in transmitting around 80% of HIV infections in this region 2. Studies in other locations have shown that HIV infection among injection drug users (IDUs) may lead to self-sustained heterosexual transmission of the virus 3.

Although syringe exchange programs (SEP) have been shown to be an effective response to injection-related HIV risks, there are too few SEPs, with limited hours of operation, to meet the needs of the growing IDU population 345. Previous studies have shown that pharmacies can be an alternative source for sterile syringes and a venue for providing other services to IDUs 678910. Pharmacies are already involved in providing treatment for addiction, dispensing methadone, supervising methadone consumption and providing information on drug misuse and HIV prevention 679. Pharmacies have also participated in syringe exchange and distributing or selling low price kits containing injecting equipment 6810. However, data on the use of pharmacies as a source of sterile syringes in Eastern European countries and Russia is lacking.

Syringe exchange began in Estonia in 1997 11. At the end of 2006 there were 26 SEPs, located in high drug use areas (in Tallinn and in the North-Eastern part of Estonia) 12. The SEPs provide additional services besides syringe exchange (health education, social welfare advice, referral for blood-borne virus testing, medical and drug treatment). Over-the-counter sterile syringes are available in Estonian pharmacies without prescription.

In a location with high HIV incidence and prevalence among IDUs, limited resources, and evidence of high risk injection practices among IDUs, we need to understand the factors associated with getting injection equipment from different sources. Previous studies of HIV among IDUs in Estonia have examined prevalence and risk behavior among the samples either as a whole or by demographic subgroups 13141516. Our aims were to examine the levels of risk behavior and the levels of HIV infection and to assess the prevention needs among IDUs who primarily use pharmacies as their source of sterile syringes and to compare them with IDUs who primarily use SEPs in Tallinn.

Respondent-driven sampling (RDS) 1718 was used to recruit 350 current IDUs for a cross-sectional risk behavior survey and biological sample collection for HIV testing. The study was conducted in spring 2007, in Tallinn. Inclusion criteria were being 18 years or older, Russian or Estonian language speakers, use of injection drugs in the previous two months and ability to provide informed consent. The inclusion criterion of drug use within two months was used with the aim of recruiting current IDUs.

Recruitment began with the non-random selection of five "seeds" representing diverse IDU types (by gender, ethnicity, main type of drug used, engaging in sex for money and HIV status). Eligible participants were provided with coupons for recruiting up to three of their peers. Coupons were uniquely coded to link participants to their survey responses and biological specimens and for monitoring who recruited whom. Participants who completed the study received a primary incentive (a food coupon worth 6.4 EUR) for participation in the study and a secondary incentive (food coupons worth 3.2 EUR for each eligible person they recruited to the study). The RDS technique uses participants' social networks to access individuals who may not appear in public venues and are not in contact with service providers. Data collected using RDS can be generalized to the sampled population when information about recruitment patterns (who recruited whom), network connections and social network sizes are gathered and incorporated into the analysis of estimates and confidence intervals 1920.

We used an interviewer-administered questionnaire, in a face-to-face interview setting, based on the WHO Drug Injecting Study Phase II survey (version 2b (rev.2)) 21. In order to assess local conditions and to subsequently adjust the instrument to best fit Tallinn's IDUs, we gathered information during a rapid assessment in October, 2006. Questions were selected that would elicit data on demographics, drug use history, HIV risk behavior, HIV testing, access and utilization of harm reduction services. Interviews were held in confidence, in a room of the SEP, between the IDU participant and the interviewer. Recruitment was conducted and the survey administrated by a team of trained fieldworkers. The study protocol included pre- and post-HIV test counseling for study participants.

Venous blood was collected from participants and tested with commercially available kits for HIV antibodies (using Abbott IMx HIV-1/HIV-2 III Plus from Abbott Laboratories, Abbott Park, Illinois, USA) and hepatitis C virus (HCV) antibodies (using ETI-AB-HCVK-3 from DiaSorin S.p.A, Via Crescentino, 13040 Saluggia, Italy). HIV test kits have proved to have high sensitivity and specificity (> 99%) 2223. The testing was conducted at the HIV/AIDS reference laboratory of the Tallinn Merimetsa Hospital.

IDUs were categorized into two groups, pharmacy or SEP (which included IDUs who got their syringes from SEP outreach workers), based on their self-reported main source for syringes within the previous six months. Risk behaviours and characteristics were compared between the two groups.

Descriptive statistics, including mean, standard deviation (SD) and range were used for continuous variables. For categorical variables, percentages and absolute (n) frequencies are presented. Student's t-test was used for continuous variables and chi-square test for categorical variables to explore differences. Odds ratios (OR) and 95% confidence intervals (95% CI) together with p-values were used to compare characteristics and risk factors between groups. Multivariate analysis based on conceptual hierarchical framework 24 was conducted to explore factors associated with using pharmacies as a main source of syringes. We also included factors that reached a statistically significant p-value (p < 0.05). Adjusted odds ratios (AOR) were calculated using gender, age, employment status, duration of injection career and frequency of injecting per day as control variables in a logistic regression model. Analyses were carried out using Stata 9 software 25.

The Ethics Review Board at the University of Tartu approved the study procedures.

Three hundred and fifty IDUs completed the questionnaire. A total of 99% answered that they had received new and unused syringes during the previous six months. The sources for new and unused syringes in that six month period were: pharmacies, 80%; SEPs, 72%; SEP outreach workers, 37%; friends, 23%; other drug users, 6%; drug workers and drug agencies, 4%; sexual partners, 2%; and street vendors, 1% (multiple responses allowed).

Of the 350 current IDUs recruited in Tallinn 328 (94%) were retained for the further analysis: 59% (n = 195) reported using SEP/SEP outreach and 41% (n = 133) reported using a pharmacy as the main source for sterile needles. We excluded 22 participants who reported alternative main source for sterile syringes from the two under consideration.

Twenty eight (21%) of the IDUs in the pharmacy group used also SEP outreach workers as an additional source for new syringes, and 68 (51%) used the SEP site as an additional source of syringes. However, 59 (45%) of the IDUs in the pharmacy group reported that they had never used SEPs. 132 (68%) of the IDUs in the SEP category had also used a pharmacy as an additional source of syringes.

Table 1 presents the univariate comparisons of IDUs in the pharmacy and SEP groups. The majority of the participants were male. The ages of the two groups were similar (mean 26.3 years, range 17 to 54 years, SD = 5.6 for pharmacy workers and mean 26.9 years, range 17 to 50 years, SD = 5.7 for SEP users). There were no differences in ethnicity, marital status or educational level. There were modest statistical differences between the factors of employment and health insurance. Pharmacy users were more likely to have regular or temporary employment than SEP users (61% versus 48%, p = 0.024), and more likely to have health insurance (50% vs. 38%, p = 0.040).

There were no differences in mean age at IDU initiation between the SEP (18.4 years, range 10 to 42 years, SD = 4.8) and pharmacy users (19.1 years, range 9 to 39 years, SD = 4.4). However there were significant differences in terms of the proportion of new injectors (IDUs with 0 to 2 years of injecting) and frequency of injecting daily between the two groups (Table 1). Pharmacy users were more likely to be new injectors (16% vs. 6%, p = 0.002) and less likely to inject daily (62% vs. 76%, p = 0.009). They also reported lower injecting frequency on the last day they injected, with 75% of pharmacy users and 89% of SEP users reporting more than one injection per day (p = 0.002). There were fewer fentanyl users among pharmacy users (74% vs 85%, p = 0.015), but no differences in terms of either injecting amphetamine or experiencing drug related overdoses. There was no difference in sharing practices between the groups, except that fewer pharmacy users reported sharing syringes with HIV positive individuals (71% vs. 82%, p = 0.024). However, pharmacy users reported riskier sexual behaviors – with a higher proportion reporting more than one sexual partner (53% vs. 40%, p = 0.020) within the preceding year and more reporting casual sexual partners (47% vs. 32%, p = 0.007) within previous six months.

There were important differences in HIV/HCV serostatus between the groups: fewer IDUs in the pharmacy group were HIV seropositive (46% vs. 64.0%, p = 0.001) or HCV seropositive (88% vs. 99%, p < 0.001). There were no differences in HIV testing prior to the study, having received drug abuse treatment or having been in prison.

We used the SEP users as a reference group to calculate the AORs and 95% CI for the pharmacy group for injection risk behavior, HIV and HCV serostatus, sexual behaviour and contacts with harm reduction and health services, see Table 2. Pharmacy users had close to twice the odds for having, in the previous year, either more than one sexual partner (AOR 1.88 95% CI 1.17–3.04) or engaging in casual sexual relationships (AOR 2.09 95% CI 1.24–3.53) in the last six months. The pharmacy group had lower odds for being infected with either HIV or HCV (AOR 0.54 95% CI 0.33–0.87 and 0.10 95% CI 0.02–0.50, accordingly).

The low numbers of participants reporting exclusive pharmacy or SEP/SEP outreach use prohibited a meaningful analysis. Nevertheless, while comparing only SEP/SEP outreach (n = 66) to only pharmacy (n = 68) users we found similar results to those obtained in the analysis presented above in terms of employment, health insurance statuses, but shorter injecting careers, higher numbers of sexual partners and less frequent injections and lower HIV status of exclusively pharmacy users in comparison to exclusively SEP/SEP outreach (data not shown).

We found that the great majority of IDUs in Tallinn are using either SEPs or pharmacies as their primary source for sterile syringes. Over half the study participants (59%) mentioned using SEP, whereas 41% reported pharmacy as their main source of sterile syringes. A similar study conduced in Tallinn two years earlier, found lower proportions using SEP, with 46% reporting SEPs including outreach workers and 50% of the participants reporting pharmacies as their main sources for sterile syringes 26.

HIV prevalence was high among both those who used SEPs and those who used pharmacies as their primary source of sterile syringes. In addition, both groups reported an unacceptably high rate of sharing syringes, especially sharing with people who are known to be HIV positive. Additional measures to reduce sharing, particularly between people who are HIV positive and HIV negative are urgently needed in Estonia.

There were significant differences between IDUs primarily using pharmacies and those using SEPs. Also the comparison of IDUs who used only SEPs or only pharmacies showed the same results suggesting that there are two distinct groups of IDUs with different risk profiles using pharmacies and SEPs. Pharmacy users were more likely to report multiple sexual partners and casual sexual partners. Condom use with casual and main partners was, however, equally low in both groups. This finding suggests that interventions which increase awareness of the risks associated with sexual transmission of HIV are also needed.

Our findings suggest that IDUs in the initial stages of their injection careers use pharmacies. According to the univariate analysis, pharmacy users included a higher proportion of new injectors (those reporting injecting two years or less) and reported lower injection frequencies. Previous studies from other locations have reported similar results: pharmacy users tend to be younger 272829, with shorter injecting careers 2930 and lower injecting frequencies 2930. New injectors create special problems for HIV prevention. Studies have found higher rates of injecting risk behavior and a higher incidence of blood-borne infections among new injectors 16313233. Recent cohort studies confirm the same, recent initiates to injecting have a higher incidence of HIV and hepatitis C 3435. New injectors may increase the size of the local IDU population, increasing the need for prevention and treatment services. Also they may not self identify as IDUs, may not fully appreciate the need to protect themselves against HIV and other blood-borne diseases, and may find HIV prevention and drug services difficult to access.

In our study, nearly half the pharmacy users had never attended needle exchange, similar to a US study where pharmacy users were less likely to report recent SEP use 28. This finding suggests that there could be IDUs who are beyond the reach of harm reduction services. One possible solution could be if pharmacies can take a role in linking IDUs with SEP services, especially for recent initiates. There are several examples in Europe, Australia, New Zealand, and the United States where pharmacies have been involved in providing services for drug users 891027. Despite some concerns for safety and about improperly discarded syringes, as well as undesired effects that drug users might have on the sensitivities of other business customers 8283637 it has been feasible to recruit pharmacists to provide services to IDUs 3839 and to cultivate a public health perspective among pharmacists 364041.

Our study has some limitations. The cross-sectional study design does not allow us to establish a causal relationship or a direction of causality. Second, we used a non-probability sample that may have implications for the representativeness of the study. However, we used RDS for recruitment in order to overcome some of the limitations of convenience sampling 1718. The statistical theory upon which RDS is based suggests that if peer recruitment proceeds through a sufficiently large number of waves, the composition of the sample will stabilize, becoming independent of the seeds from which recruitment began, and thereby overcoming any bias the nonrandom choice of seeds may have introduced 1718. Based on our results, the equilibrium state was achieved. Also, there might be residual (non-differential) misclassification due to the way we defined our study groups (pharmacy or SEP/SEP outreach users) leading to bias. Given that there were still statistically significant behavioral and HIV status differences between the groups, however, this supports rather than detracts from the contention that these two groups are significantly different.

Studies conducted in Russia and Eastern Europe have stressed the need for additional sources of syringes besides SEPs 4243. This study indicates the different profiles between IDUs who mainly use pharmacies for getting sterile syringes and those who mainly use SEPs. Data on the risk profiles of different groups of IDUs may be useful for developing targeted interventions. Encouraging pharmacies not only to sell sterile injection equipment to IDUs, as a regulated alternative to SEPs, but also to provide linkages to other services may be widely applicable in those areas where injecting drug use is a major driving force in HIV transmission.

Our results show that pharmacy users were at a less "advanced" stage of their injection career than SEP users. This result reinforces the need for a comprehensive approach and the need for additional sources for acquiring syringes besides SEPs. Strategies to expand syringe access should be combined with other harm reduction services to make both sources more effective and easily utilized.

AOR: Adjusted Odds Ratio; HIV: Human Immunodeficiency Virus; HCV: Hepatitis C Virus; IDU: Injection Drug User; OR: Odds Ratio; SD: Standard Deviation; SEP: Syringe Exchange Program; WHO: World Health Organisation.

The authors declare that they have no competing interests.

AU, KAO, AT, KR designed the study. KAO, AT, KR supervised the data collection. SV, AU, KAO, DDJ planned the analysis of the manuscript. SV conducted the statistical analysis and wrote the first draft of the manuscript. All of the authors contributed to the final version of the manuscript.