We employed the WHOQOL-HIV BREF instrument in measuring HRQOL among HIV/AIDS patients in three epicenters of Vietnam. The findings contribute to the cumulative evidence on measurement properties of WHOQOL-HIV BREF in the Vietnamese settings. Using this measure, we found that HRQOL of HIV/AIDS patients were moderate across six measure domains; it was the lowest in Social and Performance and the highest in Environment. There were significant decreases in HRQOL of patients at lower levels of the health system. We determined various factors that influenced the HRQOL of patients with HIV/AIDS, including socioeconomic status (gender, education, employment, and income), HIV/AIDS stages, CD4 cell counts, and the duration of taking ART.
This is the first assessment of HRQOL of HIV/AIDS patients taking ART across levels of the health system in Vietnam. It provides update information regarding the HRQOL outcome of ART services and its associated factors that is helpful for identifying additional interventions needed for patients during ART. In 2008, Tran et al.
measured HRQOL in a convenient sample of 155 HIV/AIDS patients in one of the very first ART clinic in the northern Vietnam 
. We compared findings of this present study with previous work in , and found indifferent domain scores in Physical and Performance. However, there were significant increments in Morbidity and Environment, and decrements in Social and Spirituality in the 2012 assessment versus the 2008 study 
. This could be explained by the remarkable efforts of the Government of Vietnam towards universal access to HIV/AIDS care and treatment for HIV/AIDS patients over the last few years. From 20% ART coverage for those patients in need of treatment in 2008, it has reached 50% by 2012 
. Availability of free-of-charge ART services might encourage patients to seek health care service earlier and improve their compliance with treatment that in turn improve the HIV-related morbidity of patients. This study also found similar factors associated with HRQOL to previous works, for instance, employment was positively associated with Physical 
. However, we have not found that patients living with spouse or partner had poorer HRQOL than those who were single as did the 2008 study 
. As for the duration of ART, there was a consistent finding that patients might experience HRQOL reduction during the first year of treatment 
. This might be due to the negative impact of side effects during early ART which was observed in Vietnam as well as other countries 
. Besides, we have found significant higher HRQOL among those patients who had taken ART for 2 years or more. Finally, the range of domains scores for asymptomatic (11.7; 14.8), symptomatic (11.4; 13.8) and AIDS (10.7; 13.5) patients in this sample were more varied and lower than that of the multi-center pilot study of the WHOQOL-HIV BREF, which were (12.9; 15;3), (12.1; 13.1), and (11.1; 12.0), accordingly 
Comparison of means and 95% confident intervals of HRQOL domain scores in 2012 versus 2008 assessment.
The findings of measurement properties of the WHOQOL-HIV BREF may inform the selection of HRQOL measures in HIV/AIDS population. First, the range of internal consistency reliability of this measurement was [0.67; 0.89], similar to other studies 
, suggesting that the WHOQOL-HIV BREF could be used for measuring HRQOL at the population level. Second, our measurement shows that it has discriminative validity in different HIV/AIDS stages and immunological statuses. Moreover, this instrument was able to detect differences not only in Physical and Morbidity, but also in broader dimensions of HRQOL, such as Social, Environment, Spirituality, and Performance. Finally, we found the differences of HRQOL over ART periods that implies the potential use of WHOQOL-HIV BREF in evaluating the longitudinal impact of ART on HIV/AIDS patients. In fact, our previous work showed responsiveness of the generic WHOQOL-BREF in detecting clinically important changes in HIV/AIDS patients taking ART and methadone maintenance treatment 
. Since this is a cross-sectional study, we are not able to confirm the responsiveness of the WHOQOL-HIV BREF; further longitudinal study would be necessary.
There are some policy implications derived from the study findings could be good references for the further expansion of ART services towards universal access targets in Vietnam. The decrements in HRQOL outcomes among patients at lower level of the health system raised concerns about the quality and efficiency of ART services at the grassroots level. In the rapid expansion of ART, services decentralization should go along with health care system strengthening in capacity, facilities, and cases management. In addition, gender-specific impact mitigation and support interventions should be in place. We found lower Spirituality and Social status in women that is similar to previous studies 
. In Vietnam, women’s traditional gender roles as mothers and wives did not only made them not vulnerable to HIV/AIDS, but also confined them in some ways to living positively with HIV/AIDS 
. Many women with HIV/AIDS were burdened by the responsibility of child rising. In addition, they were the ones who were taking care of their husbands among whom a significant proportion were drug users with complicated needs 
. Peer’s support, vocational training, job referrals, microfinance are potential interventions that have been proved effective in improving HRQOL and health status of women with HIV/AIDS 
. Furthermore, it has been known in the Vietnamese setting that side effects and adherence difficulties as well as social stigma significantly influenced patients’ responses to ART 
. The physical and social deterioration in the first year taking ART suggests additional attention and supports for patients when they begin ART.
The strengths of this study include a sufficient number of patients in three epicenters across levels of health systems and different regions of Vietnam. In addition, we employed an international instrument for HRQOL measurement which has been validated in the Vietnamese settings that enhanced the comparability of the study findings. Besides, there are some limitations. Fist, causal inference might be limited in a cross-sectional study design, particularly, for assessing the responsiveness of the WHOQOL-HIV BREF. In addition, patients at clinics were selected conveniently making the sample not representative for the population of HIV/AIDS patients. Therefore, generalizability of the results findings should be cautious.
In conclusion, the HRQOL outcomes of ART for HIV/AIDS patients were moderate across 6 measure domains. However, the efficiency of HIV/AIDS services at the grassroots level of health system should be improved. Gender-specific interventions, early treatment supports with focus on the first year of ART are recommended as integral interventions to improve the outcomes of ART services. Over the last 4 years, we observed significant improvements in morbidity but deterioration in social functioning of HIV/AIDS patients. The study highlights the important of regular assessment of HRQOL among HIV/AIDS patients in ART monitoring and evaluation.