Factors associated with linkage to care for HIV patients in Tajikistan

Abstract

There is a growing human immunodeficiency virus (HIV) epidemic in Tajikistan. This paper presents factors associated with linkage to HIV care among people aged 15 years and older in Tajikistan. This retrospective cross-sectional study used the Tajikistan Ministry of Health HIV registry data from patients diagnosed with HIV at age 15 years or older from 2000 to 2016. Chi squared tests and logistic regression models tested factors associated with linkage to care. A multivariable logistic regression model examined effect modifications. While linkage to care had an overall increase from 2000 to 2016, the odds of linkage were lower among certain sub-groups including among people in Dushanbe, men, people engaging in sex work, injection drug users, and older people. Regional differences exist with linkage to care, occurring least frequently in Dushanbe. While access to care and quality of care have increased significantly over time, findings suggest that linkage to care is low, especially in the capital city where many services are provided. Evaluation focusing on acceptability of HIV services should be undertaken to understand why certain people do not link with services. Additional research about the types of barriers to linking with HIV care is needed to increase linkage to HIV care.

Keywords

Human immunodeficiency virus Tajikistan care gender

Introduction

Antiretroviral therapy (ART) averts acquired immune deficiency syndrome (AIDS)-related deaths, prevents human immunodeficiency virus (HIV)-related illness, and reduces the economic burden of AIDS.¹ In 2014, UNAIDS established its 90-90-90 goals to eradicate AIDS including through accomplishing the following ART-specific objective: by 2020, 90% of all people with diagnosed HIV infection will receive sustained ART.¹ Because of this, linkage to ART care is important for the eradication of AIDS.

The quickest growing HIV epidemic is in Central Asia and Eastern Europe.² In Tajikistan, HIV has become a great concern since 2000 because of the high rate of injection drug use (IDU) and unprotected sex, which are two high-risk behaviors for HIV infection.^3–5 National surveillance data from 2017 demonstrate that 8892 people were living with HIV in Tajikistan. Because only 41% of people living with HIV (PLWH) are aware of their status according to the Tajikistan Ministry of Health, it is estimated that over 16,000 people were living with HIV in Tajikistan in 2017.^4,6

Tajikistan employs a universal health care system and HIV services are rendered free of charge. In 2005, the Republic of Tajikistan established 130 HIV testing and management sites throughout the country and 53,423 people were served that year.⁷ By 2013, 158 testing facilities were functional, and the number of people tested increased nearly 10-fold (517,376 people tested).⁴ By 2017, ART care was made available at 45 AIDS treatment centers. However, only 70% of PLWH who were aware of their status were in HIV-related care, and 59% were receiving ART, according to the Tajikistan Ministry of Health. The group of people lost to follow-up are at significant risk of complications and death from AIDS. As a result, AIDS is a substantial public health challenge because of the high risk of transmission when treatment is not followed.¹

Because preliminary data suggest that a large percentage of PLWH are not receiving treatment for HIV in Tajikistan, this study aims to identify factors associated with linkage to care among PLWH in Tajikistan from years 2000 to 2016.

Methods

Study population

This retrospective cross-sectional study uses HIV/AIDS health records from the HIV registry database of the Republic of Tajikistan Ministry of Health. Diagnosis of HIV occurred at 45 AIDS care centers, 12 mobile HIV testing sites, and 40 outreach sites. No other testing or care centers reported positive cases. Upon discovery of HIV-positive status, patients were informed about the standardized HIV registry surveillance questionnaire, the intention of gathering information, and assurance of confidentiality. Patients were instructed that they were not required to participate, and that if they agreed to participate, they could still choose not to answer single questions. Patients who agreed to participate were asked to provide informed consent verbally. All HIV-positive patients agreed to participate. As a result, all people aged 15 years or older who were officially diagnosed with HIV between 1 January 2000 and 31 December 2016 are included in this analysis.

This study analyzed all survey data available related to sociodemographic and risky behavior variables. Survey data were self-reported on gender (male or female); age, region of residence (Dushanbe, Democratic Republican Subordination, Gorno-Badakhshan, Sughd, Khatlon); urbanness of living area (urban and rural); engagement in commercial sex (paying for commercial sex and/or selling commercial sex [yes or no]), and IDU experience (yes or no). The questionnaire data were collected through in-person interviews with health care professionals and epidemiologists who were trained on HIV stigma and discrimination. Results from the questionnaires were linked with confidential electronic HIV medical records. All personal identifiers were removed from the dataset and each patient was given a unique code identifier.

The HIV registry collected information about the primary diagnosis visit as well as if the patient returned for a second AIDS-related care visit. Because of accessible data, linkage to care is defined as a binary variable where subjects who returned for a second AIDS-related care visit after diagnosis were considered to have linked with care.

Statistical methods

Bivariate associations of linkage to care with sociodemographic and risky behavior factors were assessed using simple logistic regression models. A multivariable logistic regression model was used to examine the associations after adjustment. A second multivariable logistic regression model with interaction effects examined potential effect modifications. Within the second multivariable model, interaction terms were analyzed individually. Interaction was assessed for gender, region and all other variables. Gender was chosen for the interaction analysis because of the documented impact that gender has on linkage to care⁸ and region was chosen because of the opportunity to identify region-specific policies to improve linkage to care based on identification of key populations within regions. Bar plots of predicted probabilities of linkage are used to visualize the interactions.

Ethical considerations

This study was undertaken with permission from the Tajikistan Ministry of Health. The University at Albany Institutional Review Board determined that this study was exempt from review because the study used de-identified secondary data.

Results

Of the 7686 total patients diagnosed with HIV, 71% linked to care. The majority (67.7%) were male and 76.7% were diagnosed at ages 25–44 years. Only 3.5% of patients were diagnosed in 2000–2004, 14.2% were diagnosed in 2005–2008, 39.1% were diagnosed in 2009–2012, 43.2% were diagnosed in 2013–2016. While 11.8% patients reported engaging in commercial sex, 45.5% had history of IDU.

The descriptive characteristics of the cohort are presented in Table 1. The majority were male (67.7%), most were between the ages of 25 and 44 years (76.6%), and slightly more than half (56.3%) lived in urban areas. Most cases were diagnosed after 2009 (82.2%). The fewest amount of cases occurred in Gorno-Badakhstan region (6.9%), most had never engaged in sex work (88.3%) and slightly more than half (54.5%) had used injection drugs.

Table 1.

Characteristics of PLWH aged 15 years and older in Tajikistan from 2000 to 2016 (n = 7686).

Gender
Male	5206 (67.73%)
Female	2480 (32.36%)
Age
15–24 years	817 (10.62%)
25–34 years	3320 (43.19%)
35–44 years	2575 (33.50%)
45+ years	974 (12.67%)
Region
Dushanbe (capital)	2162 (28.12%)
Republican Subordination	1332 (17.33%)
Gorno-Badakhshan Region	532 (6.92%)
Khatlon Region	1978 (25.74%)
Sughd Region	1682 (21.88%)
Urbanness of living location
Urban	4328 (56.31%)
Rural	3358 (43.68%)
Year HIV diagnosed
2000–2004	271 (3.52%)
2005–2008	1094 (14.23%)
2009–2012	3004 (39.08%)
2013–2016	3317 (43.15%)
Engagement in commercial sex
Yes	903 (11.74%)
No	6783 (88.25%)
IDU experience
Yes	3500 (45.54%)
No	4186 (54.46%)

Bivariate and multivariable analysis

The unadjusted and adjusted odds ratios are shown in Table 2. Gender, age, region, year of HIV confirmation, engagement in commercial sex status, and IDU experience are all significantly correlated with linkage to care. In adjusted analysis, all variables were significantly associated with linkage to care, controlling for covariates.

Table 2.

Factors associated with linkage to care among PLWH in Tajikistan 2000–2016.

	Unadjusted OR (95% CI)	Chi square p-value	Adjusted OR (95% CI)	Chi square p-value
Gender		p < 0.0001		p < 0.0001
Male	1 (Ref.)		1 (Ref.)
Female	3.13 (2.76–3.54)		1.84 (1.58–2.15)
Age		p < 0.0001		p = 0.0016
15–24 years	1 (Ref.)		1 (Ref.)
25–34 years	0.69 (0.58–0.83)		0.83 (0.68–1.02)
35–44 years	0.56 (0.47–0.68)		0.74 (0.60–0.91)
45+ years	0.60 (0.48–0.74)		0.64 (0.50–0.82)
Region		p < 0.0001		p < 0.0001
Dushanbe (capital)	1 (Ref.)		1 (Ref.)
Republican Subordination	2.97 (2.56–3.45)		2.72 (2.25–3.28)
Gorno-Badakhshan Region	1.63 (1.34–1.97)		2.08 (1.67–2.59)
Khatlon Region	5.27 (4.55–6.11)		4.86 (4.08–5.78)
Sughd Region	4.58 (3.94–5.33)		4.48 (3.76–5.35)
Urbanness of living location		p < 0.0001		p = 0.0005
Urban	1 (Ref.)		1 (Ref.)
Rural	2.27 (2.04–2.52)		0.77 (0.66–0.89)
Year of HIV diagnosis		p < 0.0001		p < 0.0001
2000–2004	1 (Ref.)		1 (Ref.)
2005–2008	1.01 (0.77–1.31)		1.16 (0.86–1.55)
2009–2012	1.43 (1.11–1.84)		1.58 (1.20–2.10)
2013–2016	3.45 (2.67–4.45)		2.71 (2.03–3.62)
Engagement in commercial sex		p < 0.0001		p < 0.0001
Yes	1 (Ref.)		1 (Ref.)
No	1.45 (1.25–1.68)		1.51 (1.28–1.78)
IDU experience		p < 0.0001		p < 0.0001
Yes	1 (Ref.)		1 (Ref.)
No	3.43 (3.09–3.80)		1.89 (1.64–2.17)

Ref: reference value.

The odds of linkage to care in Tajikistan significantly increased from 2000 to 2016. Compared to subjects diagnosed in the 2000–2004 period, the odds of linkage in the 2005–2008, 2009–2012, and 2013–2016 periods increased by 16%, 58%, and 171%, respectively. Females had 84% higher odds of linkage than males. The odds of linkage to care decreased with increase in age. The odds of linkage to care in District of Republican Subordination, Gorno-Badakhshan Region, Khatlon Region, and Sughd Region were 2.72 times, 2.08 times, 4.86 times, and 4.48 times, respectively, of the odds of linking in Dushanbe. The odds of linkage for people in rural areas were 2.27 times the odds for those living in urban areas. Furthermore, the odds of linkage to care among people who had not engaged in commercial sex were 51% higher than the odds for people who did, and the odds of linkage among patients with no IDU experience were 89% higher than the odds for patients with no IDU experience.

Interaction analysis

Table 3 displays the p-values for significance tests for interaction effects by gender and region, after controlling for all other variables. Significant effects were observed for gender by age, gender by engagement in commercial sex, gender by IDU experience, gender by region, region by age, region by year, and region by IDU experience.

Table 3.

Significance tests for effects of interactions on linkage to care using logistic regression model.

Variable	DF	Wald Chi square statistic	p
Gender by age	3	10.18	0.0171
Gender by year of diagnosis	3	1.26	0.7390
Gender by engagement in commercial sex	1	7.43	0.0064
Gender by IDU experience	1	8.28	0.0040
Gender by region	4	12.89	0.0118
Region by age	12	29.44	0.0034
Region by year of diagnosis	12	109.76	<.0001
Region by IDU experience	4	12.19	0.0160
Region by engagement in commercial sex	4	6.31	0.1772

More specifically, while mean predicted linkage probability (PLP) increased over 2000–2016 across all regions, the increase was greater in the District of Republican Subordination, Gorno-Badakhshan Region, and Dushanbe compared to Khatlon and Sughd Regions (Table 3). Females were more likely to link to care across all regions; however, the difference was much higher in Dushanbe (0.74 vs. 0.41) compared to other regions (Table 3). The gender by age interaction was significant, where the mean PLPs for females aged 15–24, 25–34, 35–44, and 45 or older were 0.90, 0.85, 0.83, and 0.79, respectively. However, for males the corresponding mean PLPs were 0.65, 0.65, 0.64, and 0.66, respectively (Figure 1 and Table 3). Gender was also associated with IDU experience (p = 0.004), with men engaging in IDU more often than women. Year of HIV infection was significantly associated with region (p < 0.001), with PLP increasing for all regions for all years, except Dushanbe. With regard to age by region, the District of Republican Subordination, Khatlon, and Sughd Regions exhibited a declining mean PLP with increase in age (Table 3 and Figure 2). A greater proportion of patients with IDU experience were living in Dushanbe (0.69 vs. 0.38) than other regions. This corresponding difference is greater in Gorno-Badakhshan (0.75 vs. 0.55) and District or Republican Subordination (0.83 vs. 0.61) compared to Khatlon (0.89 vs. 0.76) and Sughd (0.86 vs. 0.76) (Table 3). Finally, the interaction between gender and engagement in commercial sex was significant, where the difference in mean PLP for women who engaged in commercial sex compared to those who did not was higher than the corresponding difference in their male counterparts (Table 3). Comprehensive results from the interaction analysis are displayed in the supplementary file.

Figure 1.

Linkage trends by gender and age of diagnosis.

Figure 2.

Linkage trends by age of diagnosis and region of residence.

Discussion

This paper presents findings related to linkage to HIV care for PLWH in Tajikistan, as indicated through attending a second HIV/AIDS care visit after primary diagnosis. Our analysis suggests that women, older people, people living in places other than Dushanbe, people living in rural areas, people diagnosed later, people who had not engaged in commercial sex, and people with IDU experience were significantly more likely to link to care.

Over time, as the Republican AIDS Center has improved its HIV programming since 2000, linkage has exponentially increased, with the odds of linkage to care 171% greater from 2013–2016 compared to 2000–2004. In the Sughd region, however, linkage to care was significantly higher from 2000 to 2004 compared to all other regions. Sughd remained a high linking region from 2013 to 2016, with Khatlon and the Democratic Republican Subordination regions also presenting high linkage to care rates. In fact, since 2009, all regions except for Dushanbe had significant increases in linkage to care. While access to AIDS Treatment Centers in Dushanbe is comparable to access in other regions, other regions are performing better at linkage to HIV care. The single most significant difference between the regions is that Dushanbe is the most urban of areas compared to the other regions included. The fact that linkage increased after 2009 indicates a functioning health system since funding for the national HIV program in Tajikistan nearly doubled from 2009 to 2014, resulting in more opportunities for treatment.⁴ Our research suggests that men are less likely to link with HIV care compared to women. Prior research also suggests that men have higher rates of loss to follow-up within three months and six months in South Africa, the United States, and Brazil.^9–11 In general, women participate in health care more often than men, and it is suggested that this dynamic occurs with regard to linkage to HIV care as well. El-Bassel and Wechsberg¹² have explored the efficacy of couple-based therapy HIV interventions to improve use of HIV services. Because of the gender interaction with IDU, perhaps Tajikistan may consider exploring inclusion of this form of therapy to reduce the disparity of linkage to care by gender.

While women link to care more often than men overall, as women age, they are less likely to seek care. The most recent Tajikistan Demographic Health Survey¹³ estimated that while only 62% of women had a baseline understanding of AIDS, older women had the most knowledge about the disease. While these data capture women’s knowledge at one point in time, it suggests that older women know more about HIV/AIDS. While knowledge of AIDS is an important factor in linking to HIV care, it may not be the only important factor. For example, King et al.⁸ have elaborated upon the importance of de-stigmatization of HIV and of reducing barriers to seeking care as key factors related to seeking HIV care. Other factors may play a role including having an efficient and effective health system where services are accessible.^1,8 Providing flexible and convenient hours at HIV treatment clinics, reducing stigma especially for key populations, and providing incentives for participation in management are strategies for linkage to care.¹⁴ Just as our findings related differences in linkage to care by age and gender, perhaps greater awareness of HIV/AIDS does not always result in more linkage to care.

The effect of engagement in commercial sex on linkage to care varied significantly by gender. In general, women who had experience engaging in commercial sex work were more likely to have linked with care than men who engaged in commercial sex work. Women who have never engaged in sex work are more likely to link with care compared to women who have engaged in sex work. Among men, we do not see this same pattern. The probability of linking with care among men, regardless of if they have ever sought services from sex workers, is relatively equal. Prior studies have indicated that shame and stigma associated with sex work impacts female sex workers and results in their decision not to, or inability to, seek care.^15,16 Among men, engagement in commercial sex work does not appear to impact probability of linkage to HIV care. In this region, men who are labor migrants are documented to engage in commercial sex more often than men who are not migrants, and this serves as a significant risk factor for HIV.¹⁷ These results suggest that de-stigmatization of commercial sex work to increase linkage to care is a priority among women engaging in commercial sex work.

Additionally, the significant interaction between gender and IDU experience and region and IDU experience suggests that IDU is a key factor impacting linkage to care. Similar findings were found in research from Kazakhstan where IDU served as an important factor related to HIV infection⁵ and even serves as a driver of the HIV epidemic.¹⁸

Limitations

While this study provides critical information regarding factors associated with linkage to care among PLWH in Tajikistan, there are several limitations. This study relied on routine surveillance data collected by the Ministry of Health of Tajikistan. As such, data points were limited to the surveillance forms. Because of this, information regarding men who have sex with men, experience of having unprotected sex, and multiple sex partners was not included.

Conclusion

This paper presents the first findings related to linkage to HIV care by gender, region, age, IDU, and engagement in commercial sex work among PLWH in Tajikistan. We found that women, older people, and people with no history of IDU were more likely to link with care overall. Regional differences existed, and linkage to care occurred least frequently in the capital Dushanbe. Interaction effects for linkage to care were found by gender and engagement in commercial sex work where women were less likely to link with care and there was no significant impact on linkage to care among men. While access to care and quality of care have increased significantly over the past 20 years, our findings suggest that linkage to care continues to remain low, especially in the capital city where most services are provided. Evaluation focusing on acceptability of HIV services should be undertaken to understand why certain people do not continue to seek services. Additional research about the types of barriers to link with HIV care is needed to increase linkage to HIV care.

Supplemental Material

Supplemental material for Factors associated with linkage to care for HIV patients in Tajikistan

Supplemental material for Factors associated with linkage to care for HIV patients in Tajikistan by Arash Alaei, Nisheet Nautiyal, Kathryn Mishkin, D Saifuddin Karimov, Dilshod Saidi and Kamiar Alaei in International Journal of STD & AIDS

Footnotes

Acknowledgments

The authors thank the Ministry of Health of Tajikistan for their support, with a special thank you to Minister N Olimzoda, the Republican AIDS Center staff especially Dr Muloasad, M Nazarov, and Shahzoda F Negmatova.

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

UNAIDS. 9090-90: an ambitious treatment target to help end the AIDS epidemic. Geneva: UNAIDS, 2014.

Thorne

Ferencic

Malyuta

et al . Central Asia: hotspot in the worldwide HIV epidemic. Lancet Infect Dis 2010; 10: 479–488.

WHO. World health statistics 2017. Geneva: World Health Organization, 2017.

Mansfeld

Ristola

Klinte

et al.

HIV programme review in Tajikistan. Geneva: World Health Organization, 2015.

El-Bassel

Gilbert

Terlikbayeva

et al . HIV among injection drug users and their intimate partners in Almaty, Kazakhstan. AIDS Behav 2013; 17: 2490–2500.

UNAIDS. Tajikistan HIV and AIDS estimates, www.unaids.org/en/regionscountries/countries/tajikistan (2015, accessed 12 May 2018).

Laukamm-Josten U, Khotenashvili L, Akkazieva B, et al. HIV/AIDS in Tajikistan: Mid-term review of the National AIDS Programme 2011–2015. Denmark: World Health Organization, 2014.

King

Maksymenko

Almodovar-Diaz

et al . ‘ If she is a good woman …’ and ‘to be a real man …’: gender, risk and access to HIV services among key populations in Tajikistan. Cult Health Sex 2016; 18: 422–434.

Arnesen

Moll

AP and

Shenoi

SV.

Predictors of loss to follow-up among patients on ART at a rural hospital in KwaZulu-Natal, South Africa. PLoS One 2017; 12: e0177168.

10.

Fleishman

Yehia

Moore

et al . Establishment, retention, and loss to follow-up in outpatient HIV care. J Acquir Immune Defic Syndr 2012; 60: 249–259.

11.

de Almeida

de Jesus Pedroso

do Socorro Lina van Keulen

et al . Loss to follow-up in a cohort of HIV-infected patients in a regional referral outpatient clinic in Brazil. AIDS Behav 2014; 18: 2387–2396.

12.

El-Bassel

N and

Wechsberg

WM.

Couple-based behavioral HIV interventions: placing HIV risk-reduction responsibility and agency on the female and male dyad. Couple Family Psychol 2012; 1: 94.

13.

Statistical Agency of Republic of Tajikistan. Tajikistan demographic and health survey 2012.

Dushanbe

: Tajikistan Ministry of Health, 2013.

14.

Govindasamy

Meghij

Negussi

et al . Interventions to improve or facilitate linkage to or retention in pre-ART (HIV) care and initiation of ART in low- and middle-income settings – a systematic review. J Int AIDS Soc 2014; 17: 19032.

15.

Wood

Kerr

Tyndall

et al . A review of barriers and facilitators of HIV treatment among injection drug users. AIDS (London, England) 2008; 22: 1247–1256.

16.

Rhodes

T and

Simic

Transition and the HIV risk environment. BMJ 2005; 331: 220–223.

17.

Weine

Bahromov

Loue

et al . HIV sexual risk behaviors and multi-level determinants among male labor migrants from Tajikistan. J Immigr Minor Health 2013; 15: 700–710.

18.

El-Bassel

Shaw

Dasgupta

et al . Drug use as a driver of HIV risks: re-emerging and emerging issues. Curr Opin HIV AIDS 2014; 9: 150.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.12 MB