Adherence to highly active antiretroviral therapy among people living with HIV and associated high-risk behaviours and clinical characteristics: A cross-sectional survey in Vietnam

Abstract

Although Vietnam has promoted the utilisation of highly active antiretroviral therapy (HAART) towards HIV elimination targets, adherence to treatment has remained under-investigated. We aimed to describe high-risk behaviours and clinical characteristics by adherence status and to identify the factors associated with non-adherence. We included 426 people living with HIV (PLWH) currently or previously involved in HAART. Most participants were men (75.4%), young (33.6 years), with low income and low education levels. Non-adherent PLWH (11.5%) were more likely to have a larger number of sex partners (p-value = 0.053), sex without condom use (p-value = 0.007) and not receive result at hospital or voluntary test centre (p-value = 0.001). Multiple logistic regression analysis showed that demographic (education levels), sexual risk behaviours (multiple sex partners and sex without using condom) and clinical characteristics (time and facility at first time received HIV-positive result) were associated with HAART non-adherence. There are differences in associated factors between women (education levels and place of HIV testing) and men (multiple sex partners). Gender-specific programs, changing risky behaviours and reducing harms among PLWH may benefit adherence. We highlight the need to improve the quantity and quality of HIV/AIDS services in Vietnam, especially in pre- and post-test counselling, to achieve better HAART adherence, working towards ending AIDS in 2030.

Keywords

Non-adherence highly active antiretroviral therapy people living with HIV associated factors Vietnam

Introduction

HIV/AIDS is a global public health problem that affects people in the most productive stage of life. Highly active antiretroviral therapy (HAART) has contributed to prolonging life expectancy and improving quality of life of people living with HIV (PLWH) and has become a key component of the HIV response.¹ The Joint United Nations Programme on HIV/AIDS (UNAIDS) introduced the 90-90-90 targets for HAART programs in 2014. These global targets call for 90% of PLWH to know their HIV status, 81% to be on HAART and 72% to be virally suppressed by 2020.² The World Health Organization (WHO) considers adherence to HAART to be a central part of the treatment as prevention (TasP) strategy, a global approach to control the HIV epidemic.^3,4 Nevertheless, UNAIDS projects that the global 2020 target will not be achieved, and the HIV elimination goal will be delayed by 10 years or more due to unequal progress within and between countries.^5,6

Vietnam has a stable HIV epidemic concentrated in some high-risk groups of people who inject drugs, sex workers and men who have sex with men (MSM).^7,8 There were approximately 230,000 PLWH in 2018, with 5000 people dying annually due to AIDS and 5200 newly infected cases.⁹ Committing to achieving the UNAIDS targets, Vietnam had promoted WHO’s initiative Treatment 2.0 and issued new guidelines for diagnosis and treatment of HIV/AIDS to increase the uptake of HAART.¹⁰ By 2019, 160,000 PLWH were on HAART, which accounted for 70% of the total number of PLWH.⁹ In the context of resource-restricted settings, adherence is not only crucial for improving HAART coverage and ensuring treatment effectiveness but is also needed for preventing drug resistance and avoiding the cost of moving to second- and third-line HAART. However, a recent study showed that 25% of people receiving ART reported non-optimal adherence and missed 25% of doses in the last 7 days.¹¹ HIV/AIDS prevention and control in Vietnam is facing several challenges with inequalities in receiving health services^12,13; thus, more endeavours are needed in progress towards HIV elimination in this low- and middle-income country.¹⁴

Since HAART non-adherence is one of the most serious challenges to the success of HIV treatment, much has been done to identify its associated factors. Prior studies revealed that depression, stigma, drug side effects, forgetfulness, marital status and inequalities in socio-economic conditions were associated with HAART non-adherence in Vietnam.¹⁵ However, quality of the decentralised HAART delivery services needs to be routinely evaluated for its impact on HAART utilisation. Additionally, context-specific knowledge regarding high-risk behaviours and clinical and health service characteristics is still lacking in Vietnam. This scientific-based information is of great interest to policymakers and program managers to make appropriate public health plans and practice decisions. In this study, we determined the high-risk behaviours and clinical and healthcare service characteristics of PLWH on HAART by their adherence status and identified the factors associated with HAART non-adherence in Vietnam.

Methods

Study design

We used data from a cross-sectional study of PLWH in Vietnam in 2017. Hanoi and Nghe An were selected as the capital city and the largest province of Vietnam’s central coast regions, as they are among the top provinces with the highest number of PLWH.¹⁶ This survey used a convenience sampling method to achieve a rapid, readily available and cost-effective sample.^17,18

Participant recruitment

We recruited participants using HIV/AIDS case reporting procedures of the Vietnam Authority of HIV/AIDS Control (VAAC). In Vietnam, HIV screening tests are currently performed in three main domains, including hospital systems, voluntary counselling and testing (VCT) centres and other settings such as private clinics or prison facilities. In contrast, the confirmation tests are conducted only at provincial centres of HIV/AIDS prevention and control. They summarise all HIV-positive results and report them to VAAC as provincial monthly reports. The research team used those reports to selected facilities in Hanoi and Nghe An in 2017. A total of 440 adult clients (older than 15 years) with confirmed HIV-positive results were approached for this study. Finally, information of 426 PLWH (response rate = 97%), who were currently or previously involved in HAART, was used for analysis. This sample size is considered as appropriate for multiple regression analyses with less than 30 covariates according to Cohen et al.¹⁹

Measured information

We measured HAART adherence using participants’ self-reported compliance with prescribed medications in the 30 days before the interview.²⁰ The adherence questions were coded as a dichotomised variable with ‘adherence to HAART’ defined as intake of greater than or equal to 90% prescribed doses, versus ‘non-adherence to HAART’ for less than 90%. Participants’ demographic information (i.e. gender, age, education level, income, marital situation and HIV transmission mode) was collected. We asked clients for their sexual risk behaviours encompassing the number of partners over the last 1 year, sexual history (having sex with or without a condom) and history of ever having been forced to have sex. We asked participants for information about their history of using drugs or stimulants, the age at first time use, injection of any drug, needle sharing and the number of shared injecting partners over the last 6 months. We finally asked participants for their clinical characteristics (i.e. history of sexually transmitted diseases, recently successful treatments, time and the facility since receiving first HIV-positive results, recent treatment status and regimen and the results of the initial CD4 test).

Statistical analysis

We performed univariate analysis including descriptive and inferential methods for demographic information, high-risk behaviours and clinical characteristics between HAART non-adherent and adherent groups. We applied the Student’s t-test for continuous variables and Pearson’s chi-squared and Fisher’s exact test for categorical variables. To examine the key determinants of HAART non-adherence behaviours, we used multivariable logistic regression models. Since there are differences in high-risk behaviours between genders, we conducted models separately for men and women along with an overall model. Associations were measured by the estimation of adjusted odds ratios and 95% confidence intervals (95% CIs). We selected predictors for the multivariate model based on our prior knowledge about relationships between variables and outcomes, results from univariate analysis and our specific research interests. Finally, reduced multivariable models were developed by using forward stepwise variable selection based on the smallest Akaike information criterion. All analyses were conducted using R version 4.0.2.

Ethics statement

This study used data from the survey of PLWH in Vietnam of the Center for Community Health Research and Development. The United States Agency for International Development approved this survey’s ethics with IRB approval number IRB00006556. All the participants were provided verbal consent, information on the study’s objectives and methods and anonymity and opt-out rights. Additionally, identifying information of participants was not collected.

Results

Table 1 presents the demographic characteristics of 426 PLWH separately for the non-adherent (11.5%) and adherent groups (88.5%). The majority of participants were men (75.4%), Kinh ethnicity (85.2%), quite young (average of 33.6 years old) and with education levels of secondary school or lower (71.9%). They mostly were living with spouses (50.0%), had unstable jobs and casual employment (65.3%), with an average income of 3.9 million dong (approximately US$170). The HIV transmission modes were reported mainly as same- or opposite-sex sexual activity. Table 1 shows a statistically significant difference only in monthly income (p-value = 0.078) between non-adherent and adherent groups.

Table 1.

Demographic characteristics of people living with HIV by adherence status (n = 426).

Characteristic	Overall	Non-adherence	Adherence	p-value
N (%)	426 (100)	49 (11.5)	377 (88.5)	p-value
Age (years), mean (SD)	33.63 (9.45)	35.22 (8.91)	33.41 (9.52)	0.208
Gender, n (%)
Female	105 (24.6)	15 (30.6)	90 (23.9)	0.393
Male	321 (75.4)	34 (69.4)	287 (76.1)
Ethnicity, n (%)
Kinh (Vietnamese)	363 (85.2)	43 (87.8)	320 (84.9)	0.749
Others (minority ethnicities)	63 (14.8)	6 (12.2)	57 (15.1)
Education level, n (%)
Complete primary school or below	114 (26.8)	8 (16.3)	106 (28.1)	0.129
Complete secondary or high school	192 (45.1)	28 (57.1)	164 (43.5)
Complete undergraduate or higher	120 (28.2)	13 (26.5)	107 (28.4)
Occupation, n (%)
Students	23 (5.4)	0 (0.0)	23 (6.1)	0.281*
Farmers	35 (8.2)	4 (8.2)	31 (8.2)
Workers	66 (15.5)	11 (22.4)	55 (14.6)
Unstable casual employment	278 (65.3)	32 (65.3)	246 (65.3)
Unemployed	24 (5.6)	2 (4.1)	22 (5.8)
Monthly income (millions dong), mean (SD)	3.92 (2.90)	4.61 (2.41)	3.83 (2.95)	0.078
Marital status, n (%)
Single	132 (31.0)	12 (24.5)	120 (31.8)	0.179*
Divorced	22 (5.2)	4 (8.2)	18 (4.8)
Widowed	29 (6.8)	7 (14.3)	22 (5.8)
Living with spouse	213 (50.0)	23 (46.9)	190 (50.4)
Live with sex partner	30 (7.0)	3 (6.1)	27 (7.2)
Mode of transmission, n (%)
Drug injection	139 (32.6)	14 (28.6)	125 (33.2)	0.159*
Homosexual transmission	82 (19.2)	8 (16.3)	74 (19.6)
Heterosexual transmission	183 (43.0)	27 (55.1)	156 (41.4)
Others (blood transfusion, occupational accidents, etc.)	22 (5.2)	0 (0.0)	22 (5.8)

Note: SD: standard deviation; *: exact test.

The high-risk behaviours (including sexual and drug injection–related risk behaviours) of participants by HAART adherence status are shown in Table 2. There was a significant difference in having multiple sex partners between adherence status (p-value = 0.053), with 26.1% of 426 PLWH reporting multiple sex partners with two or more partners in the last 12 months. Among those having at least one sex partner (n = 383), 85.1% reported unsafe sexual behaviour without using a condom with a significant difference (p-value = 0.007) between adherence status. Drug injection behaviours were reported in 157 participants (36.9%), and the majority of those who had ever used drugs or stimulants were men (156/157) starting at the average age of 21.9 years old. Among them, nearly half (49.4%) had injected drugs with 48.1% recently sharing needles.

Table 2.

High-risk behaviours of PLWH by HAART adherence status (n = 426).

Characteristic	Overall	Adherence status
Characteristic	Overall	Non-adherence	Adherence	p-value
Sexual risk behaviours (n = 426)
Number of sex partners (last 12 months), n (%)	426 (100)	49 (100)	377 (100)
None	43 (10.1)	4 (8.2)	39 (10.3)	0.053
One sex partner	272 (63.8)	25 (51.0)	247 (65.5)
More than 2 sex partners	111 (26.1)	20 (40.8)	91 (24.1)
Having sex without condom use (last 6 months), n (%)	383 (100)	45 (100)	338 (100)
No	57 (14.9)	1 (2.2)	56 (16.6)	0.007
Yes	326 (85.1)	44 (97.8)	282 (83.4)
Being forced to have sex by partners (last 6 months), n (%)	383 (100)	45 (100)	338 (100)
Yes	8 (2.1)	1 (2.2)	7 (2.1)	0.321
No	362 (94.5)	41 (91.1)	321 (95.0)
Do not answer/unsure of the definition of forced	13 (3.4)	3 (6.7)	10 (3.0)
Injection risk behaviours (n = 426)
Ever used drug/stimulants, n (%)	426 (100)	49 (100)	377 (100)
No	269 (63.1)	27 (55.1)	242 (64.2)	0.27
Yes	157 (36.9)	22 (44.9)	135 (35.8)
Age at first used drug, mean (SD)	21.94 (4.92)	20.41 (5.53)	22.19 (4.79)	0.115
Injected drug (last 6 months), n (%)	157 (100)	22 (100)	135 (100)
Yes	77 (49.0)	7 (31.8)	70 (51.9)	0.108
No	80 (51.0)	15 (68.2)	65 (48.1)
Shared needle (last 6 months), n (%)	77 (100)	7 (100)	70 (100)
Yes	37 (48.1)	2 (28.6)	35 (50.0)	0.433
No	40 (51.9)	5 (71.4)	35 (50.0)
Number of needle-sharing partners (last 6 months), mean (SD)	2.11 (1.15)	1.50 (0.71)	2.14 (1.17)	0.45

Note: PLWH: people living with HIV; HAART: highly active antiretroviral therapy; SD: standard deviation; *: exact test.

The clinical characteristics of PLWH by adherence status are described in Table 3. Among 383 sexually active participants, 15.1% reported having a history of sexually transmitted infections (STIs), and only 87.9% of those 51 PLWH stated that the treatment was successful. Additionally, 42.6% of those 383 people knew their partner’s HIV status with 71.8% reported as serodiscordant couples, with no difference by adherence status. Nearly half (46.2%) of 426 PLWH reported the time since their last HIV test of 3–12 months, and most of the respondents (78.4%) had taken voluntary HIV testing at that time, with statistically significant differences between adherence status.

Table 3.

Clinical characteristics of PLWH by HAART adherence (n = 426).

Characteristic	Overall	Adherence status
Characteristic	Overall	Non-adherence	Adherence	p-value
Experienced STIs (last 6 months), n (%)	383 (100)	45 (100)	338 (100)
Yes	58 (15.1)	11 (24.4)	47 (13.9)	0.092
No	311 (81.2)	34 (75.6)	277 (82.0)
No response	14 (3.7)	0 (0.0)	14 (4.1)
Have cured STIs, n (%)	58 (100)	11 (100)	47 (100)
Yes	51 (87.9)	10 (90.9)	41 (87.2)	1
No	4 (6.9)	1 (9.1)	3 (6.4)
No response	3 (5.2)	0 (0.0)	3 (6.4)
Know whether the sex partner took an HIV test (last 6 months), n (%)	383 (100)	45 (100)	338 (100)
No	220 (57.4)	29 (64.4)	191 (56.5)	0.339
Yes	163 (42.6)	16 (35.6)	147 (43.5)
HIV test results of the sex partner, n (%)	163 (100)	16 (100)	147 (100)
Negative	117 (71.8)	8 (50.0)	109 (74.1)	0.075
Positive	46 (28.2)	8 (50.0)	38 (25.9)
Time since received first HIV-positive result, n (%)	426 (100)	49 (100)	377 (100)
Less than 3 months	187 (43.9)	2 (4.1)	185 (49.1)	<0.001
From 3 to 12 months	197 (46.2)	29 (59.2)	168 (44.6)
More than 1 year	42 (9.9)	18 (36.7)	24 (6.4)
Facility received first HIV-positive result, n (%)	426 (100)	49 (100)	377 (100)
General hospital	67 (15.7)	4 (8.2)	63 (16.7)	0.001
Voluntary HIV testing	334 (78.4)	36 (73.5)	298 (79.0)
Others (prison and projects)	25 (5.9)	9 (18.4)	16 (4.2)
Currently on ART, n (%)	426 (100)	49 (100)	377 (100)
No	22 (5.2)	22 (44.9)	0 (0.0)	<0.001
Yes	404 (94.8)	27 (55.1)	377 (100.0)
Time to involve in ART after receiving last HIV-positive result, n (%)	404 (100)	27 (100)	377 (100)
Less than 1 month	392 (97.0)	25 (92.6)	367 (97.3)	0.188
More than 1 month	12 (3.0)	2 (7.4)	10 (2.7)
Current HAART regimen, n (%)
First-line regimen	404 (100.0)	27 (100.0)	377 (100.0)	NA
Initial CD4 count, mean (SD)	344.22 (280.15)	281.05 (124.25)	348.46 (287.21)	0.286

Note: PLWH: people living with HIV; HAART: highly active antiretroviral therapy; STIs: sexually transmitted infections; SD: standard deviation; *: exact test; NA: not applicable.

Table 4 presents multivariable logistic regression models for factors associated with HAART non-adherence. The overall model shows that demographic (education levels), sexual risk behaviours (multiple sex partners and sex without using condom) and clinical characteristics (time and facility since receiving the first HIV-positive result) were associated with non-adherence to HAART. Factors associated with non-adherence in women include education levels and place of HIV testing. In men, multiple sex partners were associated with non-adherence. Time since the first HIV-positive result was significant in both women (p-values of 0.046 and 0.002) and men (p-values of 0.003 and <0.001).

Table 4.

Multivariable logistic regression models for HAART non-adherence behaviours in overall and separately for men and women.

Factor	Overall model			Women model			Men model
Factor	aOR	95% CIs	p-value	aOR	95% CIs	p-value	aOR	95% CIs	p-value
Education levels
Primary and lower	1			1
Secondary to high school	3.47	1.13 to 10.60	0.029	11.41	1.43 to 91.40	0.022
Undergraduate and higher	2.93	0.86 to 10.04	0.087	4.6	0.49 to 42.83	0.18
Having multiple sex partners
No	1						1
Yes	2.34	1.08 to 5.07	0.032				4.69	1.87 to 11.78	0.001
Having sex without condom use (last 6 months)
No	1
Yes	13	1.53 to 110.45	0.019
Time since received first HIV-positive result
From 3 to 12 months	1			1			1
Less than 3 months	0.07	0.01 to 0.29	<0.001	0.1	0.01 to 0.96	0.046	0.05	0.01 to 0.35	0.003
More than 1 year	5.02	1.93 to 13.05	0.001	6.94	1.07 to 45.05	0.042	8.94	3.08 to 25.91	<0.001
Facility received first HIV-positive result
Voluntary HIV testing	1			1
Others (prison and projects)	3.47	1.00 to 12.00	0.049	6.49	0.77 to 54.91	0.086
General hospital	0.53	0.16 to 1.69	0.281	0.22	0.02 to 2.06	0.183
Had ever had STI in the last 6 months
No	1						1
Yes	2.4	0.96 to 5.99	0.062				2.62	0.88 to 7.81	0.084

Note: HAART: highly active antiretroviral therapy; STI: sexually transmitted infection; aOR: adjusted odd ratio; CIs: confidence intervals.

Discussion

The present study examined the high-risk behaviours and clinical characteristics of PLWH on HAART and the factors associated with non-adherence to HAART. This study’s participants are comparable with previous studies of PLWH in Vietnam in some demographic characteristics (e.g. age and gender distribution) and socio-economic status (e.g. educational level and income).²¹ The distribution of HIV transmission mode in this study is highly similar to the current HIV transmission pattern in Vietnam.²² Overall, 11.5% of our sample reported non-adherence, which is quite high. If this result was generalisable to the current number of PLWH in Vietnam,²² nearly 24,380 patients can be expected to be non-adherent. These figures should deserve close attention among AIDS control policymakers, programmers and practitioners. More importantly, we found that high-risk behaviours are more common among non-adherent PLWH. If these trends were to continue without policy adaptation, the transmission of HIV/AIDS would be expected to continue.

Our results highlight the critical role of clinical factors and health services, especially the time and facility of the first HIV test at which participants received a positive result, in non-adherence to HAART among PLWH in Vietnam. In particular, time since the first HIV-positive test is a similarly associated factor in both men and women. There are several possible explanations for this, such as a decline over time in awareness of HAART’s effectiveness and adherence, drug resistance and other difficulties in taking medicine regularly.²³ Furthermore, HIV testing context and facilities are the notable factor in HAART non-adherence among PLWH and particularly women. This is understandable as voluntary testing is recognised as a cost-effective strategy in reducing high-risk behaviours of HIV transmission and ensuring HAART adherence in low- and middle-income countries such as Vietnam.²⁴ Our result indicates that women, who received HIV-positive result outside of mainstream health services (i.e. hospitals and VCT), were more likely to be non-adherent to HAART. This finding suggests the vital role of HIV counsellors in providing HIV service packages, including testing, counselling, referral to treatment and adherence support.²⁵ Knowing that adherence counselling is essential in increasing treatment effectiveness and reducing drug resistance, findings from our study emphasise the need to strengthen the counselling procedure at the HIV testing facilities.²⁶ Vietnam has gained specific achievements in health care in general and HIV and AIDS prevention and control in particular by reforming health systems and policies.²⁷ Given this, shortages in both quantity and quality of staff may have a significant impact on the quality of HIV services and may deserve more attention.²⁸

Results from multivariable logistic regression models suggest differences between men and women in factors associated with HAART non-adherence, which are comparable with previous studies in other countries.²⁹ While socio-economic factors (education levels) had a considerable impact on HAART non-adherence among women, sexual risk behaviours (number of sex partners) substantially affected men. These findings suggest that future HIV/AIDS interventions and programs should be explicitly and separately designed for women and men to achieve higher efficiency. Education has been shown to have an association with non-adherence to HAART among women and is recommended by UNAIDS as a critical target for improving women’s HIV/AIDS outcomes.³⁰ Our results confirm this and suggest overconfidence and insufficient knowledge of HIV/AIDS even among highly educated women. More efforts are still needed in both the short term by implementations of health education strategies (e.g. integrating HIV/AIDS education into school and HIV/AIDS communication campaigns via the Internet and social networks) and long term with a focus on women’s empowerment and gender equality.³¹ On the other hand, our multivariable analysis shows a positive association between having multiple sex partners and HAART non-adherence among men, which is consistent with previous studies investigating sexual risk behaviours.³² Hence, our results suggest the coexistence of non-adherence and other high-risk behaviours among men, which may amplify the risk of HIV transmission and present an immense challenge for future HIV and AIDS programs and interventions targeting men.³³ The lack of association between having ever experienced an STI and non-adherence among men who living with HIV is also consistent with other studies, in that adherence was not associated with the presence or absence of coinfections.³⁴

Among our sample, the PLWH who participated in treatment reported a higher prevalence (85.1%) of having sex without using a condom in the last sexual intercourse than previous results (71%) in other regions.³⁵ Also, the majority of those knowing the HIV status of sex partners (71.8%) had HIV-negative partners. Those results suggest a high transmission risk among those serodiscordant couples. This may partially reflect the inefficiency of behaviour change communication for harm reduction in HIV/AIDS facilities in Vietnam. The prevalence of 6.9% untreated STIs, which can heighten the risk of HIV transmission, indicates that WHO’s recommendation of screening STIs for all PLWH is still not fully achieved in Vietnam.³⁶ As such, we argue the need for appropriate interventions for STIs and more effective counselling in harm reduction programs.

To the best of our knowledge, this is the first study to examine the association between high-risk behaviours and clinical characteristics with HAART non-adherence among PLWH in Vietnam. We recognise several limitations in the present study. Firstly, as this study was cross-sectional, we may not ascribe causality to the associated factors in our results. Secondly, we collected and analysed the information on high-risk behaviours and HAART non-adherence based on self-reporting, which may lead to misclassification, recall bias and intentional concealment. Thirdly, this study may have potential bias and limitations in its generalisability due to the convenience sampling methods applied. Fourthly, other factors that might impact on HAART non-adherence among PLWH, such as stigma, discrimination and depression, have not been investigated in this study. Forthcoming research that applies implementation science can improve the HAART adherence in both genders most effectively.

Conclusions

This study identified differences in high-risk behaviours and clinical and health service factors by adherence status among PLWH and suggested gender differences in factors associated with HAART non-adherence. Future HIV programs that are gender-specific and tailored-made interventions that increase socio-economic status, change risky behaviours and reduce harms among PLWH can enhance adherence. The present study indicates the crucial importance of clinical and health service factors (the time and facility of the first HIV testing and receiving positive results) for adherence. We, therefore, call for maintaining and improving the quantity and quality of HIV services in Vietnam, especially in pre- and post-test counselling, to better achieve HAART adherence. By promoting HAART adherence programs as the centrality of WHO’s treatment as prevention strategy,³⁷ we believe Vietnam will have better progress towards the global targets of HIV and AIDS elimination for its people.

Footnotes

Acknowledgements

We would like to express our sincere gratitude and appreciation to colleagues from the Center for Community Health Research and Development, who supported in interviewing participants and collecting data.

Declaration of conflicting interests

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

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

ORCID iD

Phuong T Nguyen

References

The Antiretroviral Therapy Cohort Collaboration . Life expectancy of individuals on combination antiretroviral therapy in high-income countries: a collaborative analysis of 14 cohort studies. Lancet 2008; 372: 293–299.

The Joint United Nations Programme on HIV/AIDS (UNAIDS). 90-90-90: An Ambitious Treatment Target to Help End the AIDS Epidemic. Geneva, Switzerland: UNAIDS, October 2014.

Gilmour

Zhang

, et al. The epidemiological impact and cost-effectiveness of HIV testing, antiretroviral treatment and harm reduction programs. AIDS 2012; 26: 2069–2078.

The World Health Organization(WHO) . Progress Report 2006: Prevent HIV, Test and Treat All - WHO Support for Country Impact. Geneva, Switzerland: WHO Press, 2016.

The Joint United Nations Programme on HIV/AIDS (UNAIDS) . Global AIDS Update: Seizing the Moment - Tackling Entrenched Inequalities to End Epidemics. Geneva, Switzerland: UNAIDS, 2020.

Gilmour

Mai

Nguyen

, et al. Progress towards health for all: time to end discrimination and marginalization. Int J Environ Res Public Health 2020; 17: 1696.

Tran

Fleming

Nguyen

TMT

, et al. Changes in substance abuse and HIV risk behaviors over 12-month methadone maintenance treatment among vietnamese patients in mountainous provinces. Int J Environ Res Public Health 2019; 16: 2422. DOI: 10.3390/ijerph16132422

Nguyen

Gilmour

, et al. Factors associated with high-risk behaviors of people newly diagnosed with HIV/AIDS: results from a cross-sectional study in Vietnam. In: AIDS Care - Psychol Socio-Medical Asp AIDS/HIV. London, UK: Taylor & Francis, Ltd 2020. DOI: 10.1080/09540121.2020.1761519

Joint United Nations Programme on HIV/AIDS (UNAIDS) . Overview of HIV in Vietnam. http://www.unaids.org/en/regionscountries/countries/vietnam (2017, accessed 11 May 2019).

10.

Ministry of health of Vietnam. Decision No. 3047/QD-BYT of the Ministry of Health Issued Guidelines for the HIV/AIDS Treatment and Care 2015; 53: 1689–1699.

11.

Tran

Nguyen

, et al. Determinants of antiretroviral treatment adherence among HIV/AIDS patients: a multisite study. Glob Health Action 2013; 6: 19570. DOI: 10.3402/gha.v6i0.19570

12.

Chu

D-T

Vo Truong Nhu

Tao

, et al. Achievements and challenges in HIV/AIDS control in Vietnam. HIV Med 2018; 19: e75–e76.

13.

Chu

D-T

H-L

Tran

D-K

, et al. Socioeconomic inequalities in the HIV testing during antenatal care in vietnamese women. Int J Environ Res Public Health 2019; 16: 3240. DOI: 10.3390/ijerph16183240

14.

Frank

Carter

Jahagirdar

, et al. Global, regional, and national incidence, prevalence, and mortality of HIV, 1980-2017, and forecasts to 2030, for 195 countries and territories: a systematic analysis for the global burden of diseases, injuries, and risk factors study 2017. Lancet HIV 2019; 6: e831–e859.

15.

Tran

Hwang

Nguyen

, et al. Impact of socioeconomic inequality on access, adherence, and outcomes of antiretroviral treatment services for people living with HIV/AIDS in Vietnam. PLoS One 2016; 11, e0168687. DOI: 10.1371/journal.pone.0168687

16.

The Vietnam Administration for HIV/AIDS Control . Report on HIV/AIDS Prevention and Control in the First 6 Months of 2017 and Key Tasks for the Last 6 Months of 2017. Hanoi, Vietnam: VAAC, 2017.

17.

Gary

. Practical Sampling. Newbury Park: SAGE Publications, Inc., 1990.

18.

Given

. Convenience sample. In: The SAGE Encyclopedia of Qualitative Research Methods. Thousand Oaks, CA: SAGE Publications, Inc., 2012. DOI: 10.4135/9781412963909.n68

19.

Cohen

West

, et al. Applied Multiple Regression/Correlation Analysis for the Behavioral Sciences. 3rd ed. London: Lawrence Erlbaum Associates, 2003.

20.

Simoni

Kurth

Pearson

, et al. Self-report measures of antiretroviral therapy adherence: a review with recommendations for HIV research and clinical management. AIDS Behav 2006; 10: 227–245.

21.

Dunne

Kato

, et al. Factors associated with suboptimal adherence to antiretroviral therapy in Viet Nam: A cross-sectional study using audio computer-assisted self-interview (ACASI). BMC Infect Dis 2013; 13: 154. DOI: 10.1186/1471-2334-13-154

22.

Ministry of Health of Vietnam . Report on HIV/AIDS Prevention and Control in 2019 and Key Tasks in 2020. Hanoi, Vietnam: Ministry of Health, 2020.

23.

Liao

Zhang

Wang

, et al. Analysis of factors associated with dropping-out from HIV antiretroviral therapy in Kunming City, China. BMC Infect Dis 2019; 19: 1043.

24.

Sweat

Gregorich

Sangiwa

, et al. Cost-effectiveness of voluntary HIV-1 counselling and testing in reducing sexual transmission of HIV-1 in Kenya and Tanzania. The Lancet 2000; 356: 113–121.

25.

Bemelmans

Baert

Negussie

, et al. Sustaining the future of HIV counselling to reach 90-90-90: a regional country analysis. J Int AIDS Soc 2016; 19: 20751. DOI: 10.7448/IAS.19.1.20751

26.

Uusküla

Laisaar

K-T

, et al. Effects of counselling on adherence to antiretroviral treatment among people with HIV in Estonia: a randomized controlled trial. AIDS Behav 2018; 22: 224–233.

27.

Van Nguyen

Van Hoang

Dao

ATM

, et al. An adaptive model of health system organization and responses helped Vietnam to successfully halt the Covid‐19 pandemic: what lessons can be learned from a resource-constrained country. Int J Health Plann Manage 2020; 35: 988–992.

28.

Nguyen The

Nguyen Van

Phan Thi Thu

, et al. Situation of human resources and staff training of four provincial HIV/AIDS control centers in Northern Vietnam and some related factors in 2013. Vietnam J Prev Med 2015; 25: 75.

29.

Gebremichael

Hadush

Kebede

, et al. Gender difference in health related quality of life and associated factors among people living with HIV/AIDS attending anti-retroviral therapy at public health facilities, western Ethiopia: comparative cross sectional study. BMC Public Health 2018; 18: 537.

30.

The global coalition of women and AIDS (GCWA). Educate girls: Fight AIDS. Geneva, Switzerland: UNAIDS, 2005.

31.

The United Nations Development Programme (UNDP) . Making the Law Work for Women and Girls in the Context of HIV. New York: UNDP, 2020.

32.

Remien

Dolezal

Wagner

, et al. The association between poor antiretroviral adherence and unsafe sex: differences by gender and sexual orientation and implications for scale-up of treatment as prevention. AIDS Behav 2014; 18: 1541–1547.

33.

Kalichman

Cherry

Amaral

, et al. Adherence to antiretroviral therapy and HIV transmission risks: implications for test-and-treat approaches to HIV prevention. AIDS Patient Care and STDs 2010; 24: 271–277.

34.

Champredon

Bellan

Delva

, et al. The effect of sexually transmitted co-infections on HIV viral load amongst individuals on antiretroviral therapy: a systematic review and meta-analysis. BMC Infect Dis 2015; 15: 249.

35.

Ssewanyana

Mwangala

Van Baar

, et al. Health risk behaviour among adolescents living with HIV in sub-Saharan Africa: a systematic review and meta-analysis, Biomed Res Int 2018; 2018, 1. DOI: 10.1155/2018/7375831

36.

The World Health Organization (WHO) . Essential Prevention and Care Interventions for Adults and Adolescents Living with HIV in Resource-Limited Settings. Geneva, Switzerland: WHO Press, 2009.

37.

The World Health Organization (WHO) . Antiretroviral Treatment as Prevention (TASP) of HIV and TB. Geneva, Switzerland: WHO Press, 2012.