Enhanced inpatient rounds,appointment reminders,and patient education improved HIV care engagement following hospital discharge

Abstract

Human immunodeficiency virus (HIV) care engagement post hospital discharge is often suboptimal. Strategies to improve follow-up are needed. A quasi-experimental study was conducted among hospitalized HIV-infected patients between the period from 1 January 2013 to 30 June 2014 (preintervention period) and 1 July 2014 to 31 December 2015 (intervention period). During the intervention period, an HIV care team consisting of an Infectious Diseases physician, a nurse, a pharmacist, a social worker, and an HIV-infected volunteer made daily inpatient rounds. Prior to discharge, patients received a structured HIV education session and an outpatient appointment was scheduled for them with two telephone reminder calls following discharge. There were 240 HIV-infected patients enrolled (120 in each study period), of which the median age was 37 years (interquartile range [IQR] 28–44 years), 58% were male, 39% were newly diagnosed with HIV infection, 46% were hospitalized because of AIDS-related conditions, and the median CD4 cell count on admission was 158 cells/µl (IQR 72–382 cells/µl). The rate of HIV care engagement within 30 days after discharge was significantly higher in the intervention period compared to the preintervention period (95% versus 69%; P < 0.001). Independent factors associated with no care engagement within 30 days were patients in the preintervention period (adjusted odds ratio [aOR] 6.36; P < 0.001) and new diagnosis of HIV infection (aOR 2.77; P = 0.009). The study findings suggest that enhanced inpatient rounds, appointment reminders, and patient education were shown to be associated with improved HIV care engagement after hospital discharge. Patients with a new diagnosis of HIV infection benefit from more intense outreach.

ClinicalTrials.gov Identifier: NCT02578654

Keywords

Inpatient rounds appointment reminders patient education care engagement human immunodeficiency virus

Introduction

Linkage to care after diagnosis, retention in care, and adherence to antiretroviral therapy (ART) among people living with human immunodeficiency virus (HIV) are essential components for long-term HIV management and transmission prevention. Rates of linkage to HIV care have been reported to be 50–84% depending on clinical settings.^1–6 Several factors including being female, minority populations, and being uninsured for medical care were associated with poor linkage to HIV care.^2,3 Interventions that focus on at-risk populations, socioeconomic status, psychological support, and education about the disease and ART adherence are recommended to improve HIV care engagement.

Hospitalization is considered a critical life event that can impact HIV care engagement. For those newly diagnosed with HIV infection during hospitalization, severity of HIV-related illnesses, being in shock or denial about the HIV diagnosis, and having inadequate knowledge and understanding of the disease can all affect linkage to care following hospital discharge. Among those with known HIV infection regardless of ART use, severity of additional illnesses and postillness changes in functional status and related socioeconomic problems can disrupt HIV care engagement. Previous published studies of interventions to improve linkage to HIV care have focused on general clinic populations undergoing HIV testing or participants recruited from clinical trials.^4–13 None have specifically been conducted among hospitalized HIV-infected patients. This study aims to evaluate a set of interventions selected to improve linkage to care following hospital discharge among HIV-infected patients.

Methods

Study design and setting

A quasi-experimental study was conducted among hospitalized HIV-infected adult patients (aged ≥18 years) between the period from 1 January 2013 and 31 December 2015 at Thammasat University Hospital (TUH), a tertiary care center in central Thailand. The study was conducted in accordance with the amended Declaration of Helsinki and approved by Faculty of Medicine, Thammasat University Ethics Committee.

Study protocol, definitions, and outcome measurement

The list of all HIV-infected patients hospitalized in all inpatient units of TUH was compiled each day and sent to the HIV care team, which consisted of an Infectious Diseases physician, a nurse, a pharmacist, a social worker, and an HIV-infected volunteer. All of the HIV-infected patients hospitalized during the study period were screened for eligibility and informed consent was obtained prior to study participation. All participating patients were followed throughout hospital stay and until 30 days after hospital discharge to assess for care engagement. Exclusion criteria were death or readmission within 30 days after hospital discharge and patients who could not be contacted for outcome assessment.

The study period was divided into the preintervention period (1 January 2013 to 30 June 2014) and the intervention period (1 July 2014 to 31 December 2015). During the preintervention period, the study patients received standard of care which consisted of discussion about the disease at their primary physicians’ discretion and regular discharge procedures including a 5-min session to advise about the disease, home medications, and how to make a follow-up appointment with their physician(s) within 30 days of hospital discharge. Additional interventions were implemented during the intervention period. The first intervention consisted of daily inpatient rounds by the HIV care team with detailed assessment of the patients’ chief complaints, comorbidities, HIV clinical and laboratory status, socioeconomic status, support system, perception of stigmatization, medical coverage, and plan for continuity of HIV care. The second intervention was a 20-min education session provided to each patient at the bedside by the HIV care team when the patient became clinically stable and communicable. The education topics included epidemiology of HIV infection, routes of HIV transmission, HIV risk behaviors, natural history of HIV infection, management of HIV infection including the benefits of ART to improving outcome, importance of long-term continuity of care and adherence to ART, transmission prevention including safer sex practices and management of socioeconomic issues. The third intervention involved additional discussion to select an HIV care facility based on the patients’ medical coverage and preferences and contacting the chosen facility for them to schedule a follow-up appointment within 30 days of hospital discharge. The last interventions were two telephone calls to remind patients of their upcoming appointment on day 7 after discharge and one day prior to the appointment date.

Demographic, clinical, and HIV-related characteristics were obtained by interview and medical record review. Current alcohol use was defined as having used alcohol every day for at least one year. AIDS-related diseases were defined as opportunistic infections or malignancies that were associated with AIDS. Latest CD4 cell count was the CD4 cell count within three months of enrollment. The outcome of interest was the rate of care engagement within 30 days of hospital discharge, which was assessed by telephone call to the patients, medical record review (for patients whose appointments were at TUH), and calling to the patients’ HIV care facilities (if they chose to follow-up at facilities other than TUH).

Statistical analysis

Given there were no previous reports of effectiveness of any intervention on the rate of care engagement after hospital discharge, we estimated that the rate would improve from 69¹⁴ to 85% after the intervention. To detect this difference in rates of care engagement with the level of significance of 0.05 and power of 0.8, the sample size required for each period was 120. Data analysis was performed using SPSS version 15 (Chicago, IL, USA). Pearson’s Chi square or Fisher’s exact test was used to compare categorical data, as appropriate. Continuous variables were compared using the Mann–Whitney U-test. Adjusted odds ratio (aOR) and 95% confidence interval (CI) were calculated in multivariable logistic regression analysis to determine factors associated with no care engagement. All P values were two-tailed; P < 0.05 was considered statistically significant.

Results

Baseline characteristics of the study patients

A total of 290 patients were approached, eight (3%) refused to participate in the study (four patients each in the preintervention and intervention periods). There were 282 patients enrolled: 140 in the preintervention period and 142 in the intervention period. Twenty and 22 patients in the preintervention and intervention period, respectively, were excluded due to death within 30 days (N = 10 and 11), could not be contacted for outcome assessment (N = 6 and 6), and readmission within 30 days (N = 4 and 5). The final cohorts included 120 patients each in the preintervention and the intervention periods. Baseline characteristics of the study patients are shown in Table 1. Comparing the two cohorts, the preintervention cohort had a higher number of patients with a new HIV diagnosis while the intervention cohort had higher number of patients with current alcohol use and acquiring HIV from homosexual sexual contact.

Table 1.

Demographic and clinical characteristics of the study patients.

Characteristics	All (N = 240)	Preintervention cohort (N = 120)	Intervention cohort (N = 120)	P a
Demographic
Age (years, median, IQR)	37 (28–44)	38 (30–45)	36 (28–44)	0.28
Male gender	138 (58)	76 (63)	62 (52)	0.07
Nationality				0.16
Thai	220 (92)	107 (89)	113 (94)
Non-Thaib	20 (8)	13 (11)	7 (6)
Occupation				0.22
Company employee	115 (48)	56 (47)	59 (49)
Merchant	38 (16)	22 (18)	16 (13)
Government officer	24 (10)	14 (12)	10 (8)
Student	21 (9)	14 (12)	7 (6)
Housewife	12 (5)	3 (3)	9 (8)
Farmer	4 (2)	2 (2)	2 (2)
Unemployed	26 (11)	9 (8)	17 (14)
Marital status				0.68
Single	74 (31)	41 (34)	33 (28)
Married	97 (40)	46 (38)	51 (43)
Divorced	14 (6)	7 (6)	7 (6)
Widowed	13 (5)	4 (3)	9 (8)
Living with domestic partner	42 (18)	22 (19)	20 (17)
Highest education				0.5
Primary school or less	58 (24)	25 (21)	33 (28)
Secondary school	118 (49)	62 (52)	56 (47)
Bachelor degree and higher	64 (27)	33 (28)	31 (26)
Monthly household income				1.00
$US 1500 or less	226 (94)	113 (94)	113 (94)
More than $US 1500	14 (6)	7 (6)	7 (6)
Medical coverage				0.07
National or Social Security Health Plan	171 (71)	78 (65)	93 (78)
Self-pay	35 (15)	24 (19)	11 (8)
Government Officer Health Plan	34 (14)	18 (15)	16 (13)
Comorbidities
Medical comorbidities	66 (28)	39 (33)	27 (23)	0.08
Psychiatric comorbidities	10 (4)	6 (3)	4 (2)	0.75
Current alcohol usec	9 (4)	1 (1)	8 (7)	0.04
HIV data
HIV infection				0.008
Newly diagnosed	94 (39)	57 (48)	37 (31)
Known	146 (61)	63 (52)	83 (69)
Prior HIV care among patients with known HIV infection	117/146 (80)	50/63 (79)	67/83 (81)	0.95
Time since diagnosis (months, median, IQR)	36 (6–91)	24 (6–84)	36 (5–92)	0.98
Route of HIV acquisition				0.03
Heterosexual sexual contact	182 (76)	92 (77)	90 (75)
Homosexual sexual contact	38 (16)	13 (11)	25 (21)
Injection drug use	4 (2)	2 (2)	2 (2)
Othersd	16 (7)	13 (11)	3 (1)
Latest CD4 cell counte (cells/µl, median, IQR)	158 (72–382)	157 (57–347)	204 (92–396)	0.1
Receiving antiretroviral therapy (ART)	103 (43)	50 (42)	53 (44)	0.7
Duration of ART (months, median, IQR)	11 (3–48)	10 (1–36)	12 (4–62)	0.08
Having a family member who was a care giver	216 (90)	106 (88)	110 (92)	0.39

HIV: human immunodeficiency virus; IQR: interquartile range.

Note: Data are in number (%), unless otherwise indicated.Bold values represent significant P values (<0.05).

^aComparing between the preintervention and intervention cohorts.

^bIncluded Burmese, Laotian, and Cambodian.

^cDefined as having used the substance every day for at least one year.

^dIncluded tattooing, piercing, blood transfusion, and unknown.

^eWithin three months prior to enrollment.

Care engagement after hospital discharge

The proportions of patients who followed up at TUH were not statistically different between the preintervention and intervention cohorts (75% versus 70%). The rate of care engagement within 30 days after hospital discharge was significantly higher in the intervention cohort than in the preintervention cohort (95% versus 69%; P < 0.001). Among 43 patients who did not show up for care engagement, the reasons were work burden (37%), need to raise a child (16%), feeling well (14%), being afraid that others will know HIV status (7%), being busy (5%), going back to own countries (for immigrants) (5%), do not like taking drugs (5%), financial problem (5%), no care giver (5%), and feeling sick (3%). These reasons for no care engagement were not different between the preintervention and intervention cohorts. Among the 94 participants newly diagnosed with HIV infection during hospitalization, the rate of care engagement was significantly higher in the intervention cohort than in the preintervention cohort (92% versus 56%; P < 0.001).

Factors associated with no care engagement

Compared to patients who engaged in care within 30 days after discharge, patients who had no care engagement were more likely to be in the preintervention cohort (86% versus 42%), newly diagnosed with HIV infection (65% versus 34%), male (72% versus 54%) and immigrants (21% versus 6%), received no formal education (19% versus 3%) and be hospitalized for AIDS-related diseases (61% versus 43%). Age, occupation, marital status, monthly household income, medical coverage, comorbidities, alcohol use, route of HIV acquisition, and prior receipt of ART and the latest CD4 cell count were not associated with no care engagement. In multivariable logistic regression analysis, after adjusted for current alcohol use, type of HIV diagnosis and route of HIV acquisition, independent factors associated with no care engagement within 30 days were patients in the preintervention period (aOR 6.36, 95% CI 2.48–16.31; P < 0.001) and new diagnosis of HIV infection (aOR 2.77, 95% CI 1.29–5.96; P = 0.009) (Table 2).

Table 2.

Factors associated with no care engagement within 30 days after discharge.

Risk factors	Univariable analysis		Multivariable analysis
Risk factors	OR (95% CI)	P	aOR (95% CI)	P
Preintervention cohort	8.47 (3.42–21.00)	<0.001	6.36 (2.48–16.31)	<0.001
Newly-diagnosed HIV infection	3.71 (1.85–7.41)	<0.001	2.77 (1.29–5.96)	0.009
Male gender	2.17 (1.06–4.48)	0.04	2.14 (0.91–5.04)	0.08
Receipt of no formal education	8.78 (2.71–28.39)	<0.001	5.65 (0.61–52.37)	0.13
Non-Thai patient (immigrant)	4.48 (1.73–11.62)	0.002	1.02 (0.15–7.08)	0.99
Hospitalization due to AIDS-related diseases	2.06 (1.05–4.03)	0.04	1.61 (0.71–3.63)	0.26
Latest CD4 cell count levela	1.00 (0.99–1.01)	0.07	1.00 (0.98–1.02)	0.84

AIDS: acquired immune deficiency syndrome; aOR: adjusted odds ratio; CI: confidence interval; HIV: human immunodeficiency virus; OR: odds ratio.

^aDefined as CD4 cell count within three months of enrollment.Bold values represent significant P values (<0.05).

Discussion

Our study demonstrated that the intervention bundle which included enhanced inpatient rounds by the HIV care team, a 20-min bedside education session, scheduling an outpatient appointment within 30 days after discharge by the HIV care team and telephone calls to remind the patient of the upcoming appointment, significantly improved the rate of HIV care engagement within 30 days of hospital discharge to 95%. These interventions are not complex, can be done in resource-limited settings, and did not incur additional cost as the healthcare personnel were existing hospital employees and were uniformly willing to improve and promote comprehensive and holistic care for HIV-infected patients. However, the additional time spent for the intervention (approximately one person-hour per week) is to be considered.

Several studies have evaluated various interventions to improve linkage to HIV care in outpatient settings (Table 3). Most of the studies demonstrated significant improvement in rates of linkage to care with interventions including case management, peer support, orientation visit and field service unit for HIV-infected individuals,^4–6,11 and routine opt-out HIV testing for the general population.^7,8 However, provider-initiated HIV testing or oral-based point-of-care service testing did not improve linkage to care.^11,12 These and our study findings suggest that the interventions that focus on individual assessment of patients’ problems, psychological support, reminder of care appointment, and methods to overcome barriers of entering care including solving socioeconomic and uninsured problems are essential for improving HIV care engagement. The differences between our study and the other studies are that (1) our study was conducted among hospitalized patients while the others were conducted in outpatient settings; (2) ours reported important characteristics that may affect care engagement including education level, income, comorbid medical and psychological conditions and medical coverage, while others reported some or did not report all of these characteristics; and (3) barriers/obstacles to linkage to care were reported among our HIV-infected patients who did not show up for care within 30 days while the other studies did not report these.

Table 3.

Summary of studies evaluating interventions to improve linkage to HIV care.

Type of study/year	Study population/country	N	Intervention	Comparator	Outcome	Ref
Randomized controlled trial/2001–2003Randomized controlled trial/2011–2013Quasi experimental study/2004–2006 versus 2007–2008	HIV-infected patients from non-HIV care settings (STI clinics, local health centers, community-based organization, jails, and prisons)/USAPre-ART HIV-infected participants/UgandaHIV-infected outpatients/USA	146 (I) versus 149 (C)194 (I) versus 199 (C)364 (I) versus no data (Pre)	Case management Time-limited interview by a trained interventionist case manager (helping patients overcome barriers to entering care) Five in-person contacts with the case manager within 90 days after randomization Peer support Structured home visit to promote clinic attendance Assessment of participant’s status and sexual behaviors Provision of psychological support, encouragement, and reminders of clinic appointment Counseling on risk reduction Scheduled orientation visit within five days of initial call to clinic Semistructured interview Psychosocial questionnaire Baseline laboratory testing Social worker meeting for uninsured patients	A simple passive referral following posttest counselingStandard of carePreintervention cohort (historical control)	Linkage to care within the appointment date72% (I) versus 54% (C) P <0.0001HIV clinic attendance within one year92% (I) versus 84% (C)P = 0.039Linkage to care within six months81% (I) versus 69% (Pre) P <0.01	Gardner et al.⁴Chang et al.⁵Mugavero⁶
Quasi experimental study/2004–2006 versus 2008–2010Quasi experimental study/2004–2006 versus 2008–2010Quasi experimental study/2013–2014 versus 2014–2015	HIV-infected participants/USAHIV-infected participants/USAHospitalized HIV-infected patients/Thailand	2118 (I) versus 2456 (Pre) 755 (I) versus 1129 (Pre) 120 (I) versus 120 (Pre)	Routine opt-out HIV testing for adults 13–64-year-oldsCity-wide routine opt-out HIV testing and immediate linkage to care with confirmatory testingIntervention bundles Daily inpatient round by the HIV care team Detailed assessment of the patients’ clinical and laboratory status, socioeconomic status, supporting system, perception of stigmatization, medical coverage, and plan for HIV continuity care A 20-min bedside education session by the HIV care team Arranging outpatient appointment within 30 days after discharge Two telephone calls to remind the upcoming appointment	Preintervention cohort (no routine opt-out testing) Preintervention cohort (no routine opt-out testing) Preintervention cohort (standard of care)	Linkage to care more than 12 months aOR 0.42 (0.34–0.51) for intervention cohort compared to preintervention cohortLinkage to care within three months76% (I) versus 51% (Pre) P <0.0001Linkage to care within 30 days after discharge95% (I) versus 69% (Pre) P <0.001	Onyeajam et al.⁷Castel et al.⁸Our study
Prospective cohort study/2005–2006Nested cohort of a controlled trial/2007Retrospective cohort/2007–2011Retrospective cohort/2008	HIV-infected participants in The Antiretroviral Treatment Access Study-II (ARTAS-II)/USAPatients with STIs and HIV infection/South AfricaHIV-infected participants/USAGeneral population/USA	626326 (I) versus 604 (C) 4108 (I) versus 5987 (C) 2265 (I) versus 2836 (C) (no data on number of HIV-positive participants in each cohort)	Case management Up to five sessions with a case manager over 90-day period Intensive short-term assistance to facilitate linkage to care Strengths-based approach calls on clients to identify internal abilities and strengths to overcome personal or system barriers to healthcare and to develop personal plan to acquire needed resources Provider-initiated HIV testingField service unit (HIV partner service) Assessment of participant’s understanding of the follow-up care plan Facilitating appointment scheduling Provision of transportation Oral-based point-of-care service testing	NoneVoluntary counseling and testingHIV surveillance registry (standard of care) Conventional HIV testing	Linkage to care within six monthsaOR 2.95 (1.88–4.62) associated with the interventionLinkage to care defined by having CD4 test record70% (I) versus 65% (C) P = 0.53Linkage to care within 91 days79% (I) versus 66% (C) P <0.0001Linkage to care within six months52% (I) versus 42% (C) P = 0.58	Craw et al.⁹Leon et al.¹⁰Bocour et al.¹¹Keller et al.¹²
Retrospective cohort/2009–2010	HIV-infected participants/USA	3192 (I) versus 2439 (C)	Case management Client-centered services linking clients with healthcare psychological support and other services Assisting clients in obtaining medical, social, community, legal, financial, or other necessary services	Standard of care	Linkage to care within three months72% (I) versus 80% (C) P = 0.58	Willis et al.¹³

aOR: adjusted odds ratio; ART: antiretroviral therapy; C: comparator; I: intervention; Pre: preintervention cohort; Ref: reference; STI: sexually transmitted infection.

Barriers to linkage to care identified in this study were work burden, need to take care of a child, HIV stigma, and financial problems. These are consistent with those reported previously.^3,15,16 Some beliefs and attitudes, including not believing or being unsure about test results, feeling well thus no treatment needed, do not like taking drugs, believing that ART can make one sick, and placing low value on health, were among reasons for no care engagement.^3,15,16 In addition, previous studies reported that difficulty accessing care providers, mental illnesses, substance abuse, and being incarcerated were important barriers to HIV care.^3,15,16 Thus, implementing interventions to improve care engagement needs to be tailored according to individual unmet needs and made appropriate for each setting. In our study, patients who are newly diagnosed with HIV infection during hospitalization were identified as an at-risk population for no care engagement after discharge. Given their HIV-related illnesses, feeling frustration and denial about the HIV diagnosis, and being not ready to accept long-term care, these patients require appropriate education about HIV disease, multifaceted support, and close monitoring for care engagement.

The limitations of this study include that data on previous hospitalizations for any causes, sick leave availability, access to transportation, and detailed support structures for care follow-up were not collected and compared between the preintervention and intervention cohorts. This study is subject to time bias and differences in characteristics between patients in preintervention and intervention cohorts given the quasi-experimental design. However, multivariable logistic regression analysis was used and confirmed the significant association between the interventions and improvement of care engagement.

In conclusion, the study interventions were shown to be associated with improved care engagement after discharge among hospitalized HIV-infected patients. Patients newly diagnosed with HIV infection should be closely monitored for care engagement after discharge. Further randomized controlled studies are needed to confirm the study findings and their validity in other settings.

Footnotes

Declaration of conflicting interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Hull

Z and

Montaner

JS.

Optimizing the engagement of care cascade: a critical step to maximize the impact of HIV treatment as prevention. Curr Opin HIV AIDS 2012; 7: 579–586.

Mugavero

Lin

Allison

et al . Failure to establish HIV care: characterizing the “no show” phenomenon. Clin Infect Dis 2007; 45: 127–130.

Khawcharoenporn

Zawitz

Young

et al . Continuity of care in a cohort of HIV-infected former jail detainees. J Correct Health Care 2013; 19: 36–42.

Gardner

Marks

Strathdee

et al . Faster entry into HIV care among HIV-infected drug users who had been in drug-use treatment programs. Drug Alcohol Depend 2016; 165: 15–21.

Chang

Nakigozi

Billioux

et al . Effectiveness of peer support on care engagement and preventive care intervention utilization among pre-antiretroviral therapy, HIV-infected adults in Rakai, Uganda: a randomized trial. AIDS Behav 2015; 19: 1742–1751.

Mugavero

MJ.

Improving engagement in HIV care: what can we do?

Top HIV Med 2008; 16: 156–161.

Onyeajam

Eke

Stephens

et al . Time to linkage to care and viro-immunologic parameters of individuals diagnosed before and after the 2006 HIV testing recommendations. South Med J 2013; 106: 257–266.

Castel

Greenberg

Befus

et al . Temporal association between expanded HIV testing and improvements in population-based HIV/AIDS clinical outcomes, District of Columbia. AIDS Care 2014; 26: 785–789.

Craw

Gardner

Marks

et al . Brief strengths-based case management promotes entry into HIV medical care: results of the antiretroviral treatment access study-II. J Acquir Immune Defic Syndr 2008; 47: 597–606.

10.

Leon

Mathews

Lewin

et al . A comparison of linkage to HIV care after provider-initiated HIV testing and counseling (PITC) versus voluntary HIV counselling and testing (VCT) for patients with sexually transmitted infections in Cape Town, South Africa. BMC Health Serv Res 2014; 14: 350.

11.

Bocour

Renaud

Udeagu

et al . HIV partner services are associated with timely linkage to HIV medical care. AIDS 2013; 27: 2961–2963.

12.

Keller

Jones

J and

Erbelding

Choice of Rapid HIV testing and entrance into care in Baltimore City sexually transmitted infections clinics. AIDS Patient Care STDS 2011; 25: 237–243.

13.

Willis

Castel

Ahmed

et al . Linkage, engagement, and viral suppression rates among HIV-infected persons receiving care at medical case management programs in Washington, DC. J Acquir Immune Defic Syndr 2013; 64: S33–S41.

14.

Ayudhya

DP and

Khawcharoenporn

HIV care engagement within 30 days after hospital discharge among patients from a Thai tertiary-care centre. Int J STD AIDS 2015; 26: 467–469.

15.

Kuznetsova

Meylakhs

Amirkhanian

et al . Barriers and facilitators of HIV care engagement: results of a qualitative study in St. Petersburg, Russia. AIDS Behav 2016; 20: 2433–2443.

16.

Remien

Bauman

Mantell

et al . Barriers and facilitators to engagement of vulnerable populations in HIV primary care in New York City. J Acquir Immune Defic Syndr 2015; 69: S16–S24.