Linking Infectious and Narcology Care (LINC) in Russia: design, intervention and implementation protocol

Participants were recruited from inpatient wards at the City Addiction Hospital in St. Petersburg, Russia from July 2012 through May 2014. The St. Petersburg City Addiction Hospital is a government-funded 500-bed hospital, providing free addiction care to residents of the city of St. Petersburg, who are registered as having a substance use disorder (drug or alcohol). The hospital provides detoxification, early stabilization, including the treatment of co-morbid psychiatric and somatic disorders, and inpatient rehabilitation. Typical length of stay for hospitalized patients is one to 3 weeks.

The study enrolled 349 participants who met the following inclusion criteria: (1) age 18–70 years; (2) HIV-infected; (3) hospitalized at the narcology hospital; (4) history of injection drug use; (5) agree to CD4 cell count testing; (6) have two contacts to assist with follow-up; (7) live within 100 km of St. Petersburg; (8) have a telephone; (9) willing to receive HIV care at Botkin Infectious Disease Hospital. The following served as exclusion criteria for study enrollment: (1) currently on ART; (2) not fluent in Russian; (3) cognitive impairment precluding informed consent.

All HIV-infected patients admitted to the intensive care and detoxification/rehabilitation departments of the City Addiction Hospital were eligible for screening. At the City Addiction Hospital, information about a patient’s HIV status is routinely collected in the admission department and noted on the front page of the patient’s medical chart. Three times a week, a nurse identified medical charts of HIV-infected patients who had not been previously screened. Patients were screened 1–5 days after admission to the narcology hospital and after treatment for most severe initial withdrawal symptoms. Screening was conducted by Research Assessors (RAs) who were City Addiction Hospital physicians with narcology subspecialty training (i.e., narcologists). They had access to medical charts and were trained on the research protocol by the Russian and US study investigators (EB, CC). Narcologists who were not involved with the patient’s medical care approached the patient to assess study eligibility. The RA initiated pre-screening procedures of potentially appropriate patients by reviewing the medical chart to assess exclusion criteria (e.g., significant cognitive impairment, currently on ART).

Once a patient was identified as being HIV-infected based on chart review, but not having evident significant cognitive impairment, not on ART, and not previously approached about the LINC study, the RA met with the patient in a private location (e.g., hospital room or exam room) to briefly describe the study and conduct the screening to confirm the presence of inclusion criteria and absence of exclusion criteria. The RA offered eligible participants enrollment into the study and as appropriate, administered and documented the informed consent. Out of 382 participants, whose charts were reviewed as part of the pre-screening process, 370 (96.9 %) were eligible and agreed to be screened. Out of those, 359 (97.0 %) met eligibility criteria to participate in the study. The main reason for ineligibility was having taken ART in the past 30 days (Table 1). Out of those who were eligible, 349 (97.2 %) were enrolled and randomized (Fig. 1).

After enrollment into the study, the RAs asked participants if they were newly diagnosed with HIV and if so, whether or not they had received post-test counseling. The RA administered HIV post-test counseling to any newly diagnosed patients who had not yet received it. Over the course of the study, six patients were newly diagnosed and received post-test counseling from the study RA. The post-test counseling typically lasted 10 min and was conducted according to the guidelines of the Russian Ministry of Health, which were consistent with the US Centers for Disease Control and Prevention (CDC) recommendations. Following post-test counseling, if indicated, the RA collected locator information for the participant and alternate contacts. The RA then administered the baseline assessment and randomized the participant into the intervention or control group. The RA accompanied the participant to the laboratory for CD4 testing and provided the participant with a resource card containing harm reduction information and contact information for the local HIV clinic (Fig. 2).

Randomization was stratified on two factors: (1) whether an outpatient appointment with an infectious disease physician occurred in the 12 months prior to enrollment, and (2) whether the participant reported ever having been hospitalized for his or her HIV infection. Stratified randomization was used in order to ensure balance with respect to these potential confounding factors. Blocked randomization using random block sizes was used within each stratum. A computer-generated randomization table was created allowing randomization to occur via a custom web-application. Due to the nature of the intervention, participants and case managers could not be blinded to group assignment. However, the study was designed to minimize measurement bias by having the baseline assessment administered prior to randomization and by concealing randomization assignment from the follow-up assessors.

The LINC intervention was a behavioral and structural intervention designed to support and motivate HIV-infected PWID to engage in (i.e., initiate and retain) HIV medical care and ultimately improve their HIV outcomes. This intervention involved coordination between the narcology and HIV systems of care, utilizing HIV strengths-based case management delivered via five one-on-one sessions by a peer case manager (i.e., HIV-infected men and women in recovery from addiction) to help motivate and reduce barriers to HIV care. The initial session was designed to be delivered in the narcology hospital and included provision of CD4 test results by the case manager (CM) in a timely fashion, to increase engagement in HIV medical care. If the participant departed the City Addiction Hospital prior to having their initial intervention session, then this encounter was pursued outside of the hospital. Subsequent sessions were conducted upon discharge from the narcology hospital over a 6-month period in community (e.g., parks, coffee shops) or clinic locations, agreed upon by the case manager and participant.

The LINC intervention was developed via adaptation of the Antiretroviral Treatment Access Study (ARTAS) intervention for use in the Russian setting and specifically with PWID [22]. The ARTAS model was developed in the United States for a spectrum of HIV-infected persons and not specifically for PWID. Consistent with the ARTAS model, the LINC intervention used a strengths-based case management approach in which a trained CM met individually with clients to motivate them to engage in HIV medical care by supporting the recognition of their own strengths to make positive changes in their lives. This approach was grounded in a Social Cognitive Theory (SCT) and Psychological Empowerment Theory (PET) framework [23, 24].

SCT posits that behavioral goals (e.g., HIV care initiation and maintenance) are achieved when individuals (1) perceive their capacity to engage in the behavior (self-efficacy), (2) have the skills to manage “triggers,” such as substance use that can impede the behavior (self-management) and (3) believe the behavior will be both beneficial and socially supported (outcome expectancies) [23]. Additionally, SCT recognizes that such factors can only impact behavior, such as engagement in HIV care, in an environment where access to services exists [25]. Hence, the LINC intervention offered strengths-based case management in the addiction treatment setting and in transition to the HIV care setting in order to (1) build self-efficacy, (2) enable self-management, and (3) increase outcome expectancies regarding engaging in HIV care via education and social support.

Use of SCT is predicated on individuals having some degree of personal empowerment to create change in their own lives. Given the marginalization of HIV-infected PWID in Russia, this approach benefits from use of PET to bolster the effectiveness of the SCT-derived intervention [24]. According to PET, empowerment, the process by which individuals gain power over their own lives, occurs with opportunities to (1) influence the system (e.g., HIV clinical care) affecting them (intrapersonal), (2) have control over their role in transactions (interactional), and (3) contribute to changing the system and its interactions with them to better meet their health needs respectfully (behavioral). The LINC strengths-based HIV case management was designed to promote patient empowerment by providing education on how to control the HIV and addiction care they receive (intrapersonal), their right to receive and how to advocate for quality HIV and addiction care (interactional), and peer HIV case management by HIV-infected PWID in recovery to model for patients how to pursue healthy and productive lives (behavioral).

Six peer case managers (CMs), who were HIV-infected and in recovery (i.e., history of substance use disorder), were hired and trained to conduct the LINC intervention. All CMs had experience providing case management to substance using and/or HIV-infected adults in St. Petersburg. The LINC intervention was developed, and training was organized and led by the team of US and Russian investigators. Researchers (AR, JS, EB) were trained by Richard Rapp, MSW in a two-day training in Boston, which focused on the Antiretroviral Treatment Access Study (ARTAS) Linkage Case Management (ALCM) method [22]. Over the course of 3 days, researchers (AR, JS, EB) trained the CMs in St. Petersburg, providing them with an overview of the theoretical framework, assessment techniques and the LINC intervention. Researchers modeled and role-played the intervention delivery for the CMs to assure the use of a strengths-based approach. A simultaneous interpreter was used to allow the role-playing sessions to be critiqued. The team covered a number of topics with the CMs: HIV in PWID populations, addiction treatment options, HIV care, and antiretroviral therapy adherence. Booster trainings were conducted annually and as necessary, based on findings from quality assurance efforts. These intensive trainings, as well as the monitoring and observation for quality assurance were designed to limit potential variability due to individual CMs.

Each participant randomized to the intervention was assigned a specific CM, to ensure that the participants would be distributed evenly among the CMs. Upon randomization to the intervention, the RA notified the assigned CM via text message or phone call to meet with the participant at the City Addiction Hospital within 3 days. Reassignment to an alternate CM did occasionally occur due to CM illness, but efforts were made to maintain the same CM for each participant for the duration of the intervention.

The first case management session was held at the City Addiction Hospital with sessions 2–5 occurring anywhere in the community, or via phone if necessary. A session could also occur in the HIV clinic, to facilitate HIV care acquisition. The first session consisted of ascertaining the client’s strengths and developing goals related to obtaining HIV care, as well as discussing with the client their most recent CD4 cell count and benefits of HIV care (see Table 2 for a complete list of first session components). Follow-up sessions were based on a similar curriculum to the first session, but were intended to be shorter and more focused on meeting previously set goals and creating new goals based on the client’s previously self-identified strengths. Emphasis was placed on CMs guiding the client towards receiving HIV care without coercion, and supporting the client’s strengths and capacities to achieve his/her goals.

The CM contacted participants between case management sessions by phone, text message, and/or email to remind them of upcoming appointments and to check-in. CMs used their clinical judgment to determine how frequently to provide check-in calls. At a minimum, a reminder text message and/or email using a standard script was sent to participants within a week prior to their next case management session. CMs maintained detailed clinical notes and contact logs, as well as brief tracking forms to record the location and duration of sessions and check-ins.

Participants randomized to the control condition received the narcology hospital’s standard of care. With regard to linkage to HIV medical care, RAs provided all participants with a resource card containing harm reduction information and contact information for the local HIV clinic (Fig. 2).

Several implementation assessments were utilized to determine the adherence to and the fidelity of the LINC intervention.

Study interviews were conducted face-to-face by trained research staff at baseline (pre-randomization) at the narcology hospital, and at 6 and 12-months post-enrollment at First St. Petersburg Pavlov State Medical University. In the event of readmission, follow-up research interviews were occasionally conducted at the City Addiction Hospital. Participants were compensated for completion of each study visit (Table 3).

The components of the baseline and follow-up assessments and study activities are shown in Tables 3, 4 and 5. Assessment components were selected to maximize harmonization efforts with other investigators supported to pursue research projects in the National Institute on Drug Abuse (NIDA) STTR portfolio [26, 27]. Most sections of the assessment were interviewer-administered, while sections containing potentially sensitive questions (e.g., questions regarding sexual partners, HIV disclosure, HIV stigma, depressive symptoms, partner violence and sexual assault) were self-administered. All assessment data were entered directly into study computers. During a reminder call that took place 3 months post-enrollment, participants were asked about the number of times they had overdosed since the baseline interview.

In addition to study questionnaires, at each study visit RAs measured and recorded participants’ blood pressure, height, and weight. Blood was collected for CD4 count testing at baseline and 12-months.

The study aimed to test the hypotheses that participants randomized to the LINC intervention would have better: (1) initiation of HIV care; (2) retention in HIV care; (3) receipt of appropriate HIV care; and (4) HIV health outcomes, as compared to the control group. The primary outcome for hypothesis 1 (initiation of HIV care) was defined as at least one HIV physician appointment within 6 months of enrollment, assessed by medical record review (Fig. 3). A secondary outcome was self-reported initiation of HIV care. The primary outcome for hypothesis 2 (retention in HIV care) was defined as at least one HIV physician appointment in each of the two consecutive 6 month periods within 12 months of enrollment, assessed by medical record review. A secondary outcome was self-reported retention in HIV care. The primary outcome for hypothesis 3 was receipt of appropriate HIV care, defined as a CD4 cell count assessed and ART prescribed if CD4 is <350, or if >350 then another CD4 cell count obtained within 12 months. The primary outcome for hypothesis 4 was CD4 count obtained during the 12-month research visit. A secondary outcome was any self-reported hospitalizations.

Data were collected from patient medical charts to assess the study’s primary outcomes of participant initiation of and retention in HIV care, as well as the appropriateness of HIV care. Chart reviews were conducted at two central HIV care centers, where study participants were expected to seek HIV medical care: Botkin Infectious Disease Hospital and City AIDS Center. In Russia, government-funded care is sought and provided based on the patient’s address. Patients who seek addiction care at the City Addiction Hospital, the LINC recruitment site, would receive HIV care at either the City AIDS Center or Botkin Infectious Disease Hospital. Although eligibility criteria stipulated that participants should be willing to receive care from Botkin Infectious Disease Hospital, they were not required to do so. After initial care is established at one of these main hospitals, patients are able to receive follow-up care at local clinics. At Botkin Hospital, every 3–4 months a study staff member conducted chart reviews on all participants, who were 6 or 12 months post-baseline, and entered data directly into a web-based chart review form. At the City AIDS Center chart reviews were conducted annually for all participants who were 12 months post-baseline by local hospital staff trained by a Russian co-investigator (EB) and the study staff member who conducted the chart review at Botkin Hospital. Chart review information was entered onto paper forms and subsequently transferred to the main research site at First St. Petersburg Pavlov State Medical University and entered into the online chart review form by a study staff member. For a small number (n < 20) of participants, whose City AIDS Center charts indicated the receipt of HIV care from a district clinic, or if participants reported receiving HIV care elsewhere, medical charts were sought for review from those locations. Data obtained from patient medical charts included clinic visit dates, CD4 testing dates and results, and ART information.

Loss to follow-up was a concern in this difficult-to-track population, which included not only those difficult to contact (e.g., actively using substances), but also deaths and incarcerations. However, loss to follow-up was minimized by using medical record review to obtain data for three of the four primary outcomes. The research team had previous experience with long-term follow-up of similar participants, analysis of incomplete longitudinal data, and medical record review of clinical data using a standard protocol, which minimized incomplete observations.

Descriptive statistics will be calculated for all variables at baseline, 6-, and 12-months. All baseline variables will be assessed to ascertain important differences across treatment arms. Spearman correlation analyses will be performed to identify pairs of covariates that are correlated (r > 0.4) and would therefore not be included together in regression analyses. All analyses will use the intention-to-treat approach and analyze participants according to randomized group.

To assess initiation, retention and appropriateness of HIV medical care, initial analyses will be performed comparing these binary outcomes between groups using a Chi square test. The primary analysis will use multivariable logistic regression analyses to control for stratification factors (i.e., outpatient appointment with an HIV doctor in the past 12 months, ever having been hospitalized for HIV infection) in order to improve efficiency. Exploratory analyses will be conducted to examine whether or not there were effects due to HIV care site. The primary outcome for HIV health (i.e., CD4 cell count at 12 months) is continuous and will be analyzed using multiple linear regression models controlling for stratification factors and any baseline characteristics that differ.

Power calculations assumed a 2-sided test, with a significance level of 0.05. It was expected that 350 patients would be enrolled into the study. Because the primary outcome for the overall study, initiation of HIV care, was assessed using medical records, we expected loss to follow-up to be minimal. Based on a prior study from this research team (i.e., the HERMITAGE study), we estimated a 10 % loss to follow-up due to death and participant withdrawals [28]. According to data collected at another Russian narcology hospital, the Leningrad Regional Center for Addiction, by our Russian collaborators (EK, EB), we expected 20 % of controls to attend one or more appointments in 6 months. Based on these assumptions and with 315 evaluable participants, the study has 80 % power to detect an absolute difference of 15 % (i.e., 35 vs. 20 % in the intervention and control groups, respectively) in the proportions initiating HIV care using a Chi square test with continuity correction.

The LINC study was approved by the Institutional Review Boards of Boston University Medical Campus and First St. Petersburg Pavlov State Medical University. All study participants completed the informed consent process and provided written informed consent. Serious Adverse Events (i.e., hospitalizations and deaths) were reported by the study team to the study Principal Investigator (JS), as the team became aware of their occurrence. Events were reported to the Institutional Review Board annually.

With the exception of medical record review at the City AIDS Center, which was collected on paper and entered later, all study data were captured electronically on study computers via a secure, web-based data capture system. Access to the system was protected via secure logins and all data transmissions were encrypted using secure socket layering. Study data were housed on secure servers maintained by the Boston University Office of Information Technology, behind the Boston University firewalls. Identifiers needed to track participants were kept separate from research data in a secure database to which access was restricted.