Differentiated service delivery for HIV treatment models in Thailand: A cross-sectional assessment of real-world implementation and uptake.

OBJECTIVES: First, to describe the antiretroviral therapy (ART) delivery models available in Thailand to understand differentiated service delivery for further service system optimization and expansion of best practices; second, to determine the client characteristics associated with model uptake. METHODS: Across-sectional assessment using secondary data was conducted to describe ART models implemented as routine services at four public hospitals in three major provinces with a high-HIV burden in Thailand. From April to October 2020, ART clients were screened consecutively according to the inclusion criteria: Thai, aged ≥18 years, and on ART for ≥6 months. HIV treatment models were categorised based on the service type, location, provider, and frequency. Logistic regression was used to identify the associated factors. RESULTS: Seven individual ART delivery models were identified: four were facility-based and three were out-of-facility. No group models were identified. Of 3,366 records of ART clients reviewed, 3,213 (95.5%) met the study criteria and received ART through the following models: conventional (32.6%), nurse-led clinical consultation (10.0%), fast-track refill (29.0%), after-hours clinic (10.6%), pharmacist-led pickup center (3.6%), key population-led community-based organisation (2.7%), and mailing (11.5%). Age, population, duration on ART, and viral load were associated with the uptake of certain alternative service models when compared to the conventional model. CONCLUSIONS: Among the variety of ART delivery approaches available in Thailand, facility-based models remain the most prevalent. Future work should investigate the role of client preference and choice in choosing service models and service utilisation patterns over time, and assess the acceptability and effectiveness of these models.

Acceptability and retention of the key population-led HIV treatment service for men who have sex with men and transgender women living with HIV in Thailand.

INTRODUCTION: In Thailand, where the HIV epidemic is concentrated among key populations (KPs), particularly men who have sex with men (MSM) and transgender women (TGW), an HIV service delivery model tailored to KPs was piloted. This study evaluated the acceptability and retention of clients who accepted and declined the KP-led HIV treatment service. METHODS: A retrospective cohort study was conducted using secondary data from three community-based organizations (CBOs) and three hospitals in Thailand. KP lay providers were trained to lead HIV treatment service in which MSM and TGW living with HIV received counselling and a 3-month antiretroviral therapy (ART) supply at CBOs. Thai MSM and TGW who were at least 18 years, on ART for at least 6-12 months, without co-morbidities/co-infections, and virally suppressed were eligible and offered the service. Those who declined received ART via other service models offered by the hospitals and served as a comparison group. RESULTS: Of 220 clients screened between February 2019 and February 2020, 72% (159/220) were eligible of which 146 were MSM and 13 were TGW. Overall, 45% (72/159) accepted the KP-led service. Of those who declined, 98% (85/87) preferred to see the physician at the hospital. After 12 months of follow-up, among those accepted, 57% were in care at the CBO, 32% were referred back to and in care in other service models offered by the hospital, 10% were successfully transferred out to other hospital and 1% were lost to follow-up (LTFU); among those declined, 92% were in care in any service models offered by the hospital, 5% were successfully transferred out to other hospital, 2% were LTFU and 1% died (p-value<0.001). CONCLUSIONS: Despite moderate acceptability and retention in care at the CBO among the clients accepting the KP-led service, almost all clients were engaged in care overall. Multiple service models that meet the preferences and needs of KPs living with HIV should be available to optimize engagement in care.