Abstract
Background: The Russian human immunodeficiency virus (HIV) epidemic among people who inject drugs (PWID) originated in Kaliningrad, but research into risk behaviours among PWID has been lacking. The potential for heterosexual spread has not been analysed. Methods: A sample of PWID was accrued using two methods. A questionnaire was administered to assess HIV-related risk behaviours for parenteral and sexual transmission, sociodemographic factors, HIV knowledge and attitudes about sexual risks. Data were analysed focusing on the role of imprisonment, factors associated with awareness of being HIV infected and condom use. Results: More than a quarter of the sample reported having been diagnosed with HIV infection, with higher prevalence among women and those with a history of incarceration. More than half reported having been diagnosed with hepatitis C virus infection. Those reporting being HIV positive were less likely to distribute used syringes to other PWID and more likely to have used a condom the last time they had sex. A history of incarceration was associated with higher rates of receptive syringe sharing among those not having ever received an HIV-positive diagnosis and a lower likelihood of believing that condoms are needed when having sex with a casual partner. Conclusion: Although extensive HIV testing has alerted many PWID to their HIV-positive status, which is associated with less distributive syringe sharing and higher likelihood of condom use, substantial risk for parenteral and especially sexual HIV transmission remains. More active prevention programs will be required to control the heterosexual spread of HIV.
Introduction
The epidemic of human immunodeficiency virus (HIV) in Kaliningrad—the Russian exclave between Poland and Lithuania—emerged in 1995–96, earlier than elsewhere in the country.1 Over the intervening years, the prevalence has continued to rise. By late 2013, there were more than 8400 registered cases (561 per 100 000 inhabitants). As elsewhere in Russia,2,3 those most affected by the epidemic have been people who inject drugs (PWID); in Kaliningrad, PWID constituted 67% of registered HIV cases in 2009.4 According to the Kaliningrad AIDS Centre, the need for HAART has been rapidly growing. In 2009, according to international treatment standards, HAART was indicated for 27.3% of all registered cases; by 2013, this number increased to 52% of registered cases.5 The urgent need for expanding treatment in Kaliningrad (and in Russia in general) is attributable to both late diagnoses and delays in accessing care among PWID.6,7 However, even after PWID enter care, low retention and highly active antiretroviral treatment (HAART) adherence have caused great concern.8
Although the Kaliningrad region has played an important role in both the Russian HIV epidemic as a primary entry point and hotspot for spread among PWID, very little is known about the current epidemic’s social characteristics and dynamics. There have been few published reports on this topic since 2006,9–12 and the few reports that do exist have focused on the molecular epidemiology of HIV strains. A rapid policy assessment and response report conducted in 2006 by a team of local and US researchers documented substantial stigmatization, absence of targeted HIV prevention or appropriate substance abuse treatment services and high rates of overdose death and incarceration among Kaliningrad’s community of PWID.13 One outcome of the rapid policy assessment and response project was the resumption of harm reduction activities that had been halted in 2000 as a result of local and federal opposition to the principles of harm reduction.14,15 A second outcome was initiation of a project, ‘Promoting HIV and TB testing among vulnerable groups in Kaliningrad Region (Russia)’, designed to obtain a better understanding of the HIV epidemic in Kaliningrad. The project, a collaboration between the Kaliningrad Ministry of Health and a local non-governmental organization (NGO) ‘Young Leaders’ Army’ (YLA), was funded by the Nordic Council of Ministers in July 2009. The principal objective of the project was to promote TB and HIV testing and treatment among members of vulnerable and overlapping groups—PWID, former inmates released from prisons and sex workers—through enhanced cooperation among medical services and NGOs. The project included a surveillance component that directed its attention to these high-risk populations. In this article, we will be presenting the data from this project. This is the first report in over a decade that addresses the gaps in knowledge about the nature and extent of the HIV epidemic in Kaliningrad. In our analyses, we pay particular attention to the role of incarceration as a factor that has influenced PWIDs’ health behaviours and outcomes.
Methods
The data were collected from members of vulnerable groups in Kaliningrad in 2009–11 using a heterogeneous sampling methodology that combined program-based convenience and respondent-driven sampling strategies. In total, 968 individuals were recruited as study participants of whom 871 (90%) were PWID. Institutional review board (IRB) approval was not obtained due to the absence of such mechanisms in the region and because Russian social-behavioural studies when the data were collected did not require IRB approval. Analysis of the data by the American researchers used completely de-identified data and was considered exempt from human subjects review by the Yale University Human Investigation Committee. Nonetheless, the study adhered to international standards to ensure human subjects’ protection.
Data collection
Participants were recruited by ‘YLA’ and the other participating NGOs relying on peer recruitment mechanisms from vulnerable groups attracted to the health counselling, HIV and TB testing, and referrals to other medical services. Individuals visiting NGO sites who met the inclusion criteria—age 18 years or older, reporting drug injection in the prior 30 days and willing and deemed able to provide informed consent—were enrolled and if they expressed an interest in becoming a recruiter, they received three coupons to refer their peers to initiate respondent-driven sampling.16,17 Each coupon had a unique number registered under the respective client in the project’s database. For each successful recruitment, the recruiter received a bonus in the form of telephone card.
All study participants were tested for HIV, STDs and TB and then interviewed using a behavioural questionnaire (see below). Data from the questionnaire and patient test results were not linked.
Measures and analytical approach
The questionnaire was used to obtain self-reported data on socio-demographic variables, incarceration experiences, current and past drug use and related risk behaviours, sex work and sexual risk behaviours, registration at the HIV Centre and/or TB dispensary and HIV/TB testing and care experiences. HIV knowledge was measured adopting the UNGASS Guidelines on Construction of Core Indicators and consisted of five statements measuring basic knowledge about HIV transmission.18 Responses were scored using a four-point Likert scale format (1 = strongly disagree, 4 = strongly agree). In our analysis, we dichotomized between well-informed respondents (those who correctly answered all five items) and respondents with insufficient awareness (those who were not able to answer all five items correctly).
Condom use questions were divided into several categories: self-efficacy, attitudes and expected stigma. The same four-point Likert scale was used. ‘Self-efficacy in condom-use’ was measured with the question ‘I am confident that I am able to convince my partner to use a condom during sex’. Answers to this question were abnormally distributed and contained missing data for 187 respondents. Therefore, we excluded it from further analysis. Questions about ‘attitudes towards condom-use’ were yes/no forced choice questions phrased in the negative (will not use a condom even if under alcohol; will not use a condom even if under the influence of drugs, will not use a condom the next time) and used the same four-point Likert scale response format described earlier. Answers to these three items were positively correlated (alpha = 0.79).19 Therefore, we created a generalized index [−1; 0; 1] for all negative (agreeing with the statements), mixed or all positive (disagreeing with the statement) attitudes. ‘Expected stigma’ regarding condom use was measured with the following questions: ‘If I were to suggest condom use to a partner, he/she would become suspicious’ and ‘Condom use compromises trust between partners’. Responses used the same four-point Likert scale response format. We combined positive and negative responses to create a generalized index [−1; 0; 1] for all negative, mixed or all positive responses.
Perceived barriers to condom use were considered separately using five questions of which only two had well distributed answers: that condoms decrease pleasure and that condom use does not make sense if one is sharing syringes. The responses used the same Likert scale, and the answers to both questions were pooled to create a single score (range 2–8).
HIV and TB infection status were ascertained by self-report and by cross-referencing disease registration records at the Kaliningrad AIDS and TB Centres, whereas knowledge of infection status for hepatitis C virus (HCV) was based only on self-report. Individuals reporting that they did not have HCV or TB were classified as negative for those diseases, while those who self-reported infection with HCV and/or TB were categorized as positive for those infections. Individuals who indicated that they were not infected or who did not know their HIV status were categorized as HIV negative. Individuals were categorized as HIV positive if they reported having been diagnosed with HIV, being registered at the local AIDS centre or if they responded to questions pertaining to living with HIV/AIDS. On the basis of preliminary analyses to account for overlap in the two HIV-positive populations, we determined that the total number of PLWH in our sample was 230 (26.4% of the PWID sample).
The primary dependent variable for the analysis was any report of having been incarcerated. The dependent variable in some secondary analyses was self-reported HIV status. All statistical analyses were conducted using R software.20 In addition to descriptive statistics, bivariate statistics and logistic regression modelling were used to identify social, behavioural and health-related correlates to the variables of interest: high health-risk behaviours—syringe sharing, insufficient condom use. We considered only main effects and not interactions. Variables were removed iteratively to produce a best fit model based on Akaike Information Criterion.20 The alpha Cronbach coefficient was used, as described above to identify correlated items on condom use questions.19
Results
The data presented in Supplementary table S1 divide the sample of 871 PWID into those reporting a history of incarceration (ex-prisoners) and those without such a history (non-prisoners). The majority of both groups was male and 60 PWID reported being not of Russian ethnicity. One quarter (n = 230, 26.4%) reported being HIV positive and 55.5% (n = 483) reported having received a diagnosis of HCV infection. Among those aware of their HIV-positive status, 79.1% (n = 182) reported having HCV, of those who were not aware the proportion was significantly lower 47.0% (n = 301) (P < 0.001). The ex-prisoner group was older, had a higher proportion of males, had a lower educational level and was more likely to be unemployed. Ex-prisoners were almost twice as likely to report having received a HIV+ diagnosis (40.7% vs. 23.0%); however, the difference in receiving an HCV+ diagnosis between prisoners and non-prisoners was not significant.
In terms of drug use and risky drug use practices, ex-prisoners had longer durations of injecting, were less likely to report having experienced a drug overdose and, among those who reported being HIV negative, were more likely to engage in receptive syringe sharing. There was also a difference in injection frequency with between ex-prisoners and non-prisoners: the former group had a significantly higher proportion of individuals who either injected daily or had abstained from injection in the 30 days prior to the interview (Supplementary table S1).
With the use of binomial logistic regression, we modelled factors associated with awareness of HIV infection in the population by including the following variables: age, gender, education, employment, main drug of choice heroin, frequency and duration of drug-use, shared syringes, condom use, HIV knowledge, drug-dependency treatment and social-care experiences (table 1). The model revealed that women were more likely to be infected than men. We analysed the data to determine whether the higher prevalence could be the result of recruitment of commercial sex work, but only 11 of 200 women reported both engaging in transactional and injecting drugs. We further determined whether there was any association between the number of partners the women reported and HIV-positive status, but this was not significant (Mann–Whitney U test, P = 0.14). Other factors associated with HIV-positive status in Kaliningrad were lower educational achievement, heroin as a drug of choice, having had prison experience, positive HCV status, perfect HIV knowledge, experiences of receiving drug treatment, condom use during the last sex, injection abstinence in the 30 days prior to interview and a lower rate of distributing used syringes to others.
Table 1.
Sociodemographic, behavioral and health-related factors associated with HIV infection among PWID in Kaliningrad
| Factors | Odds Ratio | 95% CI | P |
|---|---|---|---|
| Sex (males as reference) | 2.602 | [1.651, 4.118] | <0.001 |
| Age (per year) | 1.032 | [1.000, 1.066] | 0.051 |
| Education | |||
| School (reference) | 1 | Reference | |
| Tech school | 0.592 | [0.231, 0.627] | 0.013 |
| University | 0.449 | [0.263, 0.854] | 0.027 |
| Heroin is drug of choose | 1.528 | [0.894, 2.689] | 0.129 |
| Prisoner | 2.221 | [1.384, 3.567] | <0.001 |
| HCV status (self-report) | 2.942 | [1.930, 4.554] | <0.001 |
| Perfect HIV knowledge | 3.917 | [2.431, 6.522] | <0.001 |
| Drug-dependency treatment experiences | 1.557 | [1.046, 2.321] | 0.029 |
| Condom used on last sex | 1.713 | [1.168, 2.517] | 0.005 |
| Inject frequency | |||
| No in last 30 days (reference) | 1 | Reference | |
| Less than daily | 0.381 | [0.231, 0.627] | <0.001 |
| Daily | 0.476 | [0.263, 0.854] | 0.013 |
| Distributing syringes | 0.259 | [0.151, 0.432] | <0.001 |
| Using others’ syringes | 1.407 | [0.895, 2.216] | 0.139 |
CI, confidence interval.
Given that contaminated syringe sharing is among the main drivers of the HIV epidemic in Russia and in Kaliningrad in particular, we examined the association between specific injection drug use practices and self-reported disease status. Overall, 371 participants (42.6%) reported some form of syringe sharing. Receptive sharing (n = 305, 35.0%) was more common than distributive sharing (n = 270, 31.0%). We divided the sample into those who reported being HIV+ (n = 230, 26.4%), for whom the risk is transmitting HIV to others by distributing their used syringes to others, and those who reported last testing negative or had not been tested (n = 641, 73.6%) for whom the risk is receiving used syringes from others. For distributive sharing, we found that 34.7% of HIV-negative PWID and 20.7% of HIV-positive PWID gave their unclean syringes to others (P < 0.0001). In contrast, there was no difference in the proportion of PWID accepting used syringes (34.8% of HIV-negative and 35.5% of HIV-positive participants).
We employed logistical regression to identify predictors of syringe sharing distinguishing between PWID who were aware of being HIV+ and all others based on previous studies of PWID in other Russian cities.21,22 Models included the following variables of interest: age, gender, incarceration experiences, frequency and duration of drug use, main drug of choice heroin, HIV status, TB status, HCV status, IDU treatment experiences, social care experiences, HIV knowledge, education, employment status and any reported overdose experience. The relevant outcome variable was selected based on risk of transmission: for PWID aware of being HIV+, distribution of used syringe poses a threat to others and for those who are uninfected or unaware of their status, receipt of used syringe poses an infection risk. We found that PWID who were HIV negative or unaware of their status were more likely to accept used syringes if they had a history of incarceration or had experience with drug treatment. They were less likely to accept used syringes if they reported heroin as their drug of choice, had been diagnosed with TB and reported condom use during last sex (table 2). HIV-positive PWID were more likely to distribute used syringes if they had experience with drug treatment, were younger and had insufficient HIV knowledge.
Table 2.
Factors associated with risky use of syringes
| HIV-negative IDUs receiving used syringes |
HIV-positive IDUs distributing used syringes |
|||||
|---|---|---|---|---|---|---|
| Adjusted Odd Ratio | 95% CI | P | Adjusted Odds Ratio | 95% CI | P | |
| Drug-dependency treatment ever | 1.571 | [1.092, 2.264] | 0.011 | 2.415 | [1.142, 5.439] | 0.026 |
| Prisoner | 1.506 | [0.927, 2.437] | 0.058 | |||
| Drug of choice is heroin | 0.626 | [0.406, 0.965] | 0.004 | |||
| TB | 0.331 | [0.199, 0.534] | <0.001 | |||
| Condom used on last sex | 0.651 | [0.444, 0.949] | 0.022 | |||
| Age (per year) | 0.900 | [0.837, 0.960] | 0.002 | |||
| Perfect HIV knowledge | 0.306 | [0.128, 0.742] | 0.008 | |||
CI, confidence interval.
Sexual risks behaviours, attitudes and beliefs were examined to detect any significant differences between the groups of ex- and non-prisoners. A total of 816 individuals reported being sexually active and, among these, 307 reported using condom during the last intercourse (table 3). Few differences in sexual risk behaviours were found between PWID with and without incarceration histories. Both groups had a median of two partners within the previous year. Unprotected sex was especially common with regular partners although about 30% of PWID engaged in unprotected sex with commercial or casual partners. Compared with PWID with incarceration histories, PWID without incarceration histories were more likely to have had sex with other drug users (49.3% vs. 57.5%; P < 0.05) and under the influence of alcohol (67.2% vs. 74.6%; P < 0.05).
Table 3.
Sexual risk behaviors, attitude and beliefs among sexually active PWID
| Total N | Non-prisoners, N (%) | Ex-prisoners, N (%) | P | |
|---|---|---|---|---|
| N of sexual partners (12 months) | 816 | Mean = 5.24, median = 2 | Mean = 3.75, median = 2 | 0.131 |
| 667 | 149 | |||
| Sexual behaviors | ||||
| Condom use (last episode) | 809 | 247 (37.4%) | 60 (40.5%) | 0.472 |
| 661 | 148 | |||
| Condom use with the regular partner (last episode) | 592 | 138 (28.0%) | 27 (27.0%) | 0.831 |
| 492 | 100 | |||
| Condom use with the commercial partner (last episode) | 109 | 65 (73.9%) | 15 (71.4%) | 0.821 |
| 88 | 21 | |||
| Condom use with the casual partner (last episode) | 402 | 164 (49.7%) | 40 (55.6%) | 0.368 |
| 330 | 72 | |||
| Had sex with a drug user | 816 | 384 (57.5%) | 73 (49.3%) | 0.085 |
| 668 | 148 | |||
| Had sex under the influence of drugs | 812 | 506 (76.3%) | 104 (70.2%) | 0.129 |
| 663 | 149 | |||
| Had sex under the influence of alcohol | 812 | 495 (74.7%) | 100 (67.2%) | 0.060 |
| 663 | 149 | |||
| Attitudes and beliefs | ||||
| Condom use is needed with the regular partner | 682 | 186 (33.2%) | 39 (32.2%) | 0.845 |
| 561 | 121 | |||
| Condom use is needed with the commercial partner only | 676 | 183 (33.2%) | 37 (30.6%) | 0.585 |
| 552 | 121 | |||
| Condom use is needed with the random partner only | 685 | 196 (34.8%) | 29 (24.0%) | 0.022 |
| 564 | 121 | |||
| ‘I will not use condom in next sexual intercourse’ | 682 | 245 (43.7%) | 47 (38.8%) | 0.330 |
| 561 | 121 | |||
| ‘I will likely not use a condom in sexual intercourse under the influence of alcohol’ | 672 | 276(49.5%) | 53(45.3%) | 0.413 |
| 558 | 117 | |||
| ‘I will likely not use a condom in sexual intercourse under the influence of drugs’ | 670 | 224(40.5%) | 51(43.6%) | 0.538 |
| 553 | 117 | |||
| If I suggest using a condom my partner will become suspicious | 165 (28.8%) | 43 (35.8%) | 0.129 | |
| 572 | 120 | |||
| Condom use compromises trust between partners | 157 (27.9%) | 43 (36.1%) | 0.073 | |
| 563 | 119 | |||
The high levels of unprotected sex were consistent with reported attitudes and beliefs about condom use (table 3). These negative attitudes and beliefs were strongly associated in multivariate analysis with not using a condom during the last sexual intercourse (table 4). Agreeing that suggesting condom use creates suspicion reduced the likelihood of condom use by 38%. Two variables combining highly correlated attitudes each significantly reduced the likelihood. Conversely, two items were associated with a greater likelihood: sex with a casual partner and reporting being HIV positive.
Table 4.
Social-psychological and behavioral determinants of condom usage
| Adjusted Odds Ratio [95% CI] | P | |
|---|---|---|
| Having sex with commercial partner in last year | 1.625 [0.909, 2.924] | 0.103 |
| Having regular sex partner | 0.635 [0.391, 1.029] | 0.065 |
| Having sex with occasional partner in last year | 2.323 [1.482, 3.676] | <0.001 |
| HIV | 2.112 [1.341, 3.347] | 0.001 |
| Positive answer to suggesting condom creates suspicion | 0.621 [0.386, 0.988] | 0.046 |
| Positive answers to ‘Condoms reduce pleasure’ | 0.575 [0.365, 0.904] | 0.002 |
| ‘Will not use condom…’ per positive answer | 0.478 [0.394, 0.574] | <0.001 |
| Receiving social care | 0.534 [0.261, 1.049] | 0.076 |
| Positive answers ‘Condoms are not useful if you share syringe’ | 0.737 [0.488, 1.114] | 0.146 |
CI, confidence interval. Bolded p-values indicate those variables that remained statistically significant in the logistic regression model.
Discussion
Data from our sample of 871 PWID accrued in Kaliningrad between 2009 and 2011 were analysed using several different approaches to dichotomize the sample: (i) those with and without a history of incarceration, (ii) those who have received an HIV-positive diagnosis and those who have not and (iii) condom use at last intercourse. Our primary dichotomy was incarceration based on the hypothesis that imprisonment produces serious social dislocations. Those who had been incarcerated were older, more likely to be male, less well educated, less likely to be employed, more likely to have received an HIV-positive diagnosis, but also more likely to report no recent injections and less likely to have had an overdose experience. However, we detected few differences in either injection or sexual risk behaviours between PWID with and without incarceration experiences.
To refine our analysis of risk behaviours, we distinguished between those with and without an HIV-positive diagnosis because the direction of syringe sharing matters for HIV transmission. We did not know the true status of individuals who reported that they were negative or who did not know their HIV status, but 91.5% of all participants reported having been tested for HIV and individuals who were negative at last test or had not been tested were both assumed to behave as if they were uninfected. Incarceration experience was associated with risky syringe sharing only for the negative group, whereas drug treatment experience was associated with risky syringe sharing for both groups. This represented a failure of drug treatment programs to provide adequate harm reduction education for their patients even though they are aware that their programs are effective less than 10% of the time in preventing relapse to opioid abuse and injection.13,23 Overall, our analysis suggested that neither incarceration nor drug treatment experiences exerted much of a protective role; on the contrary, both experiences heighten health risks and vulnerability among PWID and drug treatment may be more detrimental than imprisonment. The mechanism by which this occurred needs further study. However, we can speculate that either experience is a marker of larger structural and psychosocial problems among PWID that have made this portion of the population especially vulnerable. Although these issues are not addressed in this study, our findings provide a clear signal to Russian healthcare providers and officials that the current strategies for addressing the needs of PWID are insufficient and that alternative approaches, including those that focus on harm reduction and opioid substitution therapy, are needed.
An intriguing finding was the 2.6-fold higher self-reported prevalence of HIV among the women in the sample. Given the high rate of testing among both men and women, this cannot be an artefact of inadequate knowledge of HIV status. Nor can the higher HIV prevalence among women be attributed to commercial sex work or higher numbers of male partners. Only 5% of the women in the sample reported engaging in commercial sex at the time of interview although it is possible that women who had previously engaged commercial work had stopped or that women who were then actively engaged in transactional sex chose not to report it to the interviewer. And there was no difference in the number of sexual partners between HIV-positive and -negative women. Since the difference cannot be attributed to the additional risk associated with commercial sex work or a larger number of partners, further field study will be needed to identify the factors that put women at substantial additional risk for acquiring HIV.
Analysis of data on HIV prevalence revealed that individuals who were aware of their HIV status were likely to engage in lower levels of HIV risk behaviour. They were less likely to pass on their used syringes and more likely to have used a condom the last time they had sex. Awareness of HIV+ status was also associated with awareness of having been infected with HCV. In fact, the self-reported prevalence of HCV+ was higher than that for HIV, which is to be expected since serological testing in Russia has consistently revealed higher prevalence of HCV compared with HIV among PWID.2,3,21,24,25 Although HCV prevalence based on self-reports will yield lower numbers than those obtained through epidemiological surveillance, the relatively high levels of self-reported HCV testing allows us to estimate the sample-wide prevalence. If 80.3% of the sample reported having been testing and, of these, 55.5% received a positive HCV diagnosis and assuming the same distribution of HCV in the untested fraction, the prevalence estimate for the entire sample would be 69%. This estimate is quite comparable to our findings from PWID samples accrued in eight other Russian cities where testing was conducted in 2009 and 2010. In these samples, HCV prevalence ranged from 49% to 90%, with an overall prevalence of 70.8%.27
Although the high percentage of participants in our population that indicated an experience of overdose is of great concern, our data were consistent with overdose experiences observed internationally and in the limited data available from studies conducted in Russia. In a range of international studies, between 29% and 66% of PWID reported overdose experiences.26–30 In Russia, the proportion ranged from 64% in Togliatti to 82% in Yekaterinburg.31–33 Factors associated with fatal and non-fatal overdose experiences in Kaliningrad need to be more thoroughly studied to determine whether there are parallels to findings in other Russia cities that include concurrent alcohol consumption, poorer health and family problems. It is quite clear that making naloxone available to the PWID community needs to be considered as has been done successfully in other locations to reduce overdose mortality.
In line with other studies that demonstrate that HIV testing is an important prevention strategy, our findings suggest that PWID who know they are HIV positive actively change their behaviours to reduce risk of transmission to others. Although our study was cross-sectional and therefore cannot speak to causality between HIV status and engaging in risk or protective behaviours, the association we found can be used to support HIV activists’ efforts to expand prevention and treatment programs for HIV-affected populations and increase their availability for PWID in Russia. It is certainly clear that lack of effective preventive measures has resulted in the loss of the control over epidemic spread, with an increasing role of heterosexual transmissions. By 2011, the number of newly diagnosed infections attributed to parenteral (IDU) transmission decreased to 31.1%.34 The trend continued in 2013, when only 16.9% of the newly diagnosed cases were attributed to parenteral transmission and the majority (64.3%) to heterosexual contacts. Given the high rates of unprotected sex, heterosexual transmission, especially from HIV-infected PWID to their non-injecting sex partners, may be contributing to the changing demographics of HIV infection. These data suggest a generalization of the HIV epidemic may be occurring in Kaliningrad, although absence of routinely implemented sentinel surveillance in Russia limits the ability to draw final conclusions.
Our study has a number of limitations. First and foremost, the sample cannot be proven representative of the entire population of PWID in Kaliningrad although the two sampling methods yielded similar samples (data not shown). Second, all information was obtained by self-report, so data on HIV, HCV and TB status are limited by potential response bias and gaps in testing and communication of testing results. Some of the individuals who we have categorized as being unaware of their HIV status may well have been HIV infected. However, without verifiable knowledge of their status, we felt it was reasonable to classify them as at risk for acquiring infection rather than transmitting it since this is likely how the lack of awareness would influence their behaviour. Finally, the social circumstances that stigmatize and criminalize injection drug use may have led to under-reporting of risks through socially desirable responses on the part of study participants. Despite these limitations, our findings indicate a need to improve the effectiveness of the health care system in reaching PWID with HIV, hepatitis and overdose prevention programs. More work is needed to explore barriers specific to political and social circumstances that PWID in Kaliningrad face and to identify and implement possible actions needed to better serve this population.
Supplementary data
Supplementary data are available at EURPUB online.
Acknowledgements
The authors would like to thank the Director of the Information office of the Nordic Council of Ministers in Kaliningrad Arne Grove and program manager Olga Kovaleva for their support and important advises for the project ‘Promoting HIV and TB testing among vulnerable groups in Kaliningrad Region (Russia)’. The analysis of the data collected was supported by an AIDS International Training and Research Program award from NIH/Fogarty International Center (5D43TW001028; PI: Heimer R.). The data were collected with the financial support of Nordic Council of Ministers.
Conflicts of interest: None declared.
Key Points
This is the first report on human immunodeficiency virus (HIV) awareness, risks, knowledge, attitudes and beliefs among people who inject drugs (PWID) in Kaliningrad since the HIV epidemic first emerged there nearly 2 decades ago.
Self-reported HIV prevalence was higher among women and those with a history of incarceration even though more men than women had been in prison.
Those aware of their HIV status were less likely to give away their used syringes and more likely to have used a condom when last having sex
Nonetheless, substantial risk exists for HIV transmission among PWID through unsafe injection and from PWID to their sexual partners through unprotected intercourse.
References
- 1.Bobkov A, Kazennova E, Selimova L, et al. A sudden epidemic of HIV type 1 among injecting drug users in the former Soviet Union: identification of subtype A, subtype B, and novel gagA/envB recombinants. AIDS Res Hum Retroviruses 1998;14:669–676. [DOI] [PubMed] [Google Scholar]
- 2.Eritsyan K, Heimer R, Barbour R, et al. Individual-, network-, and city-level factors associated with HIV prevalence among people who inject drugs in eight Russian cities. BMJ Open 2013;3:6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Rhodes T, Platt L, Maximova S, et al. Prevalence of HIV, hepatitis C and syphilis among injecting drug users in Russia: a multi-city study. Addiction 2006;101:252–266. [DOI] [PubMed] [Google Scholar]
- 4.Sergeev BP. Promoting HIV and TB Testing among Vulnerable Groups in Kaliningrad Region (Russia). Report on the Project’s Activities. Nordic. Copenhagen, Denmark: Nordic Council of Ministers, 2010. [Google Scholar]
- 5.Cherkes N. Taking Up the Challenge: Developing Services to Contain the Spread of HIV and TB among Injecting Drug Users in Kaliningrad Oblast . Final Project Conference Kaliningrad, RF: Kaliningrad AIDS Center, 2014. [Google Scholar]
- 6.Mathers B, Degenhardt L, Ali H, et al. HIV prevention, treatment, and care services for people who inject drugs: a systematic review of global, regional, and national coverage. Lancet 2010;375:1014–1028. [DOI] [PubMed] [Google Scholar]
- 7.Foundation for Intersectoral Partnership. Ways to Optimize and Improve the Quality of Medical Services for People with HIV in the AIDS Centres of the Russian Federation . Moscow, RF, 2009. [Google Scholar]
- 8.UNAIDS AIDS Epidemic Regional Update, Eastern Europe and Central Asia, 2007. Report No.: UNAIDS/08.11E/JC1529E. Geneva, Switzerland: UNAIDS, 2008. [Google Scholar]
- 9.Hamers FF, Downs AM. HIV in central and eastern Europe. Lancet 2003;361:1035–1044. [DOI] [PubMed] [Google Scholar]
- 10.Liitsola K, Tashkinova I, Laukkanen T, et al. HIV-1 genetic subtype A/B recombinant strain causing an explosive epidemic in injecting drug users in Kaliningrad. AIDS 1998;12:1907–19. [DOI] [PubMed] [Google Scholar]
- 11.Mashkilleyson N, Leinikki P. Evolution of HIV epidemic in Kaliningrad, Russia. J Clin Virol 1999;12:37–42. [DOI] [PubMed] [Google Scholar]
- 12.Smolskaya T, Liitsola K, Zetterberg V, et al. HIV epidemiology in the Northwestern Federal District of Russia: dominance of HIV type 1 subtype A. AIDS Res Hum Retroviruses 2006;22:1074–1080. [DOI] [PubMed] [Google Scholar]
- 13.Vyshemirskaya I, Lazzarini Z, Osipenko V, et al. HIV and Drug Policy in Kaliningrad: Risk, Silence and the Gap between Human Needs and Health Services . Rapid Policy Assessment and Response. Kaliningrad, 2008 Philadelphia, PA: Temple University, 2008. [Google Scholar]
- 14.Burrows D, Sarang A. Mixed signals: responses to HIV/AIDS in the Russian Federation. Outreach Dev 2004;6:15–17. [Google Scholar]
- 15.Csete J. Lessons not learned: human rights abuses and HIV/AIDS in the Russian Federation. Hum Rights Watch 2004;16:1–62. [Google Scholar]
- 16.Heckathorn DD. Respondent-driven sampling; a new approach to the study of hidden populations. Soc Probl 1997;44:174–199. [Google Scholar]
- 17.Heckathorn DD, Semaan S, Broadhead RS, Hughes JJ. Extensions of respondent-driven sampling: a new approach to the study of injection drug users aged 18-25. AIDS Behav 2002;6:55–67. [Google Scholar]
- 18.UNAIDS Guidelines on the Construction of Core Indicators: 2010 Reporting. Geneva, Switzerland, 2009. http://www.unaids.org/sites/default/files/sub_landing/files/jc1676_core_indicators_2009_en.pdf. [Google Scholar]
- 19.Cronbach LJ. Coefficient alpha and the internal structure of tests. Psychometrika 1951;16:297–334. [Google Scholar]
- 20.Akaike H. A new look at the statistical model identification. IEEE Trans Automatic Contr 1974;19:716–23. [Google Scholar]
- 21.R Core Team. R: A Language and Environment for Statistical Computing. Vienna, Austria: R Foundation for Statistical Computing, 2013. [Google Scholar]
- 22.Eritsyan KU, Levina OS, White E, et al. HIV-prevalence and risk behavior among injection drug users and their sex-partners in two Russian cities. AIDS Res Hum Retroviruses 2013;29:687–690. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Grau LE, White E, Niccolai LM, et al. HIV disclosure, condom use, and awareness of HIV infection among HIV-positive, heterosexual drug injectors in St. Petersburg, Russian Federation. AIDS Behav 2011;15:45–57. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Torban MN, Heimer R, Ilyuk RD, Krupitsky EM. Practices and attitudes of addiction treatment providers in the Russian Federation. J Addiction Res Ther 2011;2:104. [Google Scholar]
- 25.Paintsil E, Verevochkin SV, Dukhovlinova E, et al. Hepatitis C virus infection among drug injectors in St. Petersburg, Russia: social and molecular epidemiology of an endemic infection. Addiction 2009;104:1881–90. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Rhodes T, Platt L, Judd A, et al. Hepatitis C virus infection, HIV co-infection, and associated risk among injecting drug users in Togliatti, Russia. Int J STD AIDS 2005;16:749–54. [DOI] [PubMed] [Google Scholar]
- 27.Heimer R, Eritsyan K, Barbour R, Levina OS. Hepatitis C virus seroprevalence among people who inject drugs and factors associated with infection in eight Russian cities. BMC Infect Dis 2014;14(Suppl 6):S12. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Coffin PO, Tracy M, Bucciarell A, et al. Identifying injection drug users at risk of nonfatal overdose. Acad Emerg Med 2007;14:616–623. [DOI] [PubMed] [Google Scholar]
- 29.Ochoa KC, Hahn JA, Seal KH, Moss AR. Overdosing among young injection drug users in San Francisco. Addict Behav 2001;26:453–460. [DOI] [PubMed] [Google Scholar]
- 30.Talu A, Rajaleid K, Abel-Ollo K, et al. HIV infection and risk behaviour of primary fentanyl and amphetamine injectors in Tallinn, Estonia: implications for intervention. Int J Drug Policy 2010;21:56–63. [DOI] [PubMed] [Google Scholar]
- 31.Neira-León M, Barrio G, Bravo MJ, et al. Infrequent opioid overdose risk reduction behaviours among young adult heroin users in cities with wide coverage of HIV prevention programmes. Int J Drug Policy 2011;22:16–25. [DOI] [PubMed] [Google Scholar]
- 32.Sherman SG, Cheng Y, Kral AH. Prevalence and correlates of opiate overdose among young injection drug users in a large U.S. city. Drug Alcohol Depend 2007;88:182–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Grau LE, Green TC, Torban M, et al. Psychosocial and contextual correlates of opioid overdose risk among drug users in St. Petersburg, Russia. Harm Reduct J 2009;6:e17. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Sergeev B, Karpets A, Sarang A, Tikhonov M. Prevalence and circumstances of opiate overdose among injection drug users in the Russian Federation. J Urban Health 2003;80:212–219. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Walley AY, Cheng DM, Coleman SM, et al. Risk factors for recent nonfatal overdose among HIV-infected Russians who inject drugs. AIDS Care 2014;26:1013-1018. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.Northern Dimension. Partnership in Public Health and Social Well-being. EG on HIV/AIDS and Associated Infections Third Meeting, Kaliningrad, Russia, 2011. [Google Scholar]