Infectious Disease · Medical Literature
HIV Care Retention and Cost Analysis in Multi-Month ART Dispensing: A Randomized Controlled Trial in China.
Abstract
Multi-month dispensing (MMD), as a differentiated service delivery model, can reduce the frequency of clinic visits, waiting time and travel costs for clinically stable people living with HIV. This study aimed to evaluate the impact of 6-month MMD of antiretroviral therapy (ART) on retention and conduct a cost analysis in China. We conducted a randomized, non-blind, non-inferiority study from December 2022 to March 2023 at The First Hospital of China Medical University. Eligible participants were randomly assigned to a 3-month dispensing group (n = 789) or a 6-month dispensing group (n = 799) and followed up for 18 months. The proportion of patients continuing ART after 18 months, virological suppression rate ( 0.05). In the intention-to-treat analysis, treatment retention rates at 18 months were 94.9% (749/789) in the 3-month dispensing group and 94.2% (753/799) in the 6-month dispensing group. The risk difference (6-month minus 3-month) was -0.7% (95% CI -2.9% to 1.5%); non-inferiority was demonstrated as the lower bound of the 95% CI (-2.9%) exceeded the pre-specified margin of -5%. Viral suppression rates (<50 copies/mL) were similarly high in both groups in intent-to-treat analysis: 94.9% (3-month) versus 94.2% (6-month), with no statistically significant difference. Per-protocol analysis confirmed these findings (viral suppression 97.59% vs. 97.75%; χ2 = 0.0421, p = 0.8375). In terms of cost, the 6-month dispensing group had two fewer annual outpatient visits (3 vs. 5), with total treatment costs reduced by 27.7% (¥931.82 vs. ¥1288.85) and work value loss decreased by 16.7% (¥174.20 vs. ¥209.04 yearly). Six-month MMD of ART did not reduce treatment retention; instead, it decreased patients' clinic visit costs, thereby meeting cost-effectiveness criteria. ChiCTR2200066438.
Clinical relevance
In clinically stable people living with HIV on ART, does 6-month multi-month dispensing (6-MMD) maintain non-inferior treatment retention and virological suppression compared with standard 3-month dispensing over 18 months, and what are the associated cost differences? __SNAPSHOT__ {"studyDesign":"Randomized, open-label (non-blind), non-inferiority trial; both intention-to-treat (ITT) and per-protocol (PP) analyses performed","setting":"Single academic tertiary-care HIV clinic, The First Hospital of China Medical University, Shenyang, China; enrollment December 2022 – March 2023","sampleSize":"n = 1,588 (3-month dispensing group: n = 789; 6-month dispensing group: n = 799)","population":"Clinically stable adults living with HIV on ART; median age 40.0 years (IQR 34.0–50.0); 94.8% male; 96.9% fully suppressed (HIV RNA 0.05). Cohort was predominantly male and likely MSM-dominant, reflecting the local epidemic profile.","intervention":"6-month ART multi-month dispensing (6-MMD): participants received a 6-month supply of antiretroviral medications per dispensing cycle, reducing required clinic visits to approximately 3 per year","comparator":"3-month ART dispensing (3-MMD): standard of care in China; approximately 5 clinic visits per year","primaryOutcome":"Treatment retention in care at 18 months, defined as no gap of ≥60 days without ART (consistent with the INTERVAL trial definition); non-inferiority margin pre-specified at −5%","followUp":"18 months","fundingSource":"Not reported in the accessible abstract/publisher text; no conflict-of-interest statement identified in the accessible PubMed record"}
Key findings
PRIMARY OUTCOME — RETENTION: In the ITT analysis, 18-month treatment retention was 94.9% (749/789) in the 3-MMD group and 94.2% (753/799) in the 6-MMD group. The risk difference (6-MMD minus 3-MMD) was −0.7% (95% CI: −2.9% to +1.5%). Non-inferiority was demonstrated because the lower bound of the 95% CI (−2.9%) exceeded the pre-specified non-inferiority margin of −5%. No superiority claim was made. LOSS TO FOLLOW-UP: Loss to follow-up was marginally higher in the 6-MMD group (28 patients, 3.5%) versus the 3-MMD group (21 patients, 2.7%); absolute difference 0.8 percentage points (calculated). VIROLOGICAL SUPPRESSION: Viral suppression rates (<50 copies/mL) were similarly high in both groups; Cox proportional hazards regression confirmed no statistically significant difference in the risk of virological failure between arms in either ITT or PP analyses (ITT: HR 95% CI 0.5667–2.056, p = 0.8149; PP: HR 95% CI 0.5700–1.3270, p = 0.5155). Exact between-group suppression percentages at 18 months were not fully reported in the accessible abstract text. VIROLOGIC FAILURE RATES: ITT virologic failure was reported as approximately 5.07% in one arm; the corresponding figure for the other arm was not fully retrievable from the accessible text and is therefore not reported here. COST OUTCOMES: Participants in the 6-MMD group averaged 3 annual outpatient visits versus 5 in the 3-MMD group (a reduction of 2 visits/year). From the healthcare provider perspective, provider-side treatment costs were 27.7% lower in the 6-MMD group (¥931.82 per patient, 3-MMD group; full 6-MMD figure not completely retrievable from accessible text). Patient-borne costs (travel, time, productivity loss, estimated using provincial average wage) were lower in the 6-MMD arm; exact patient-side cost figures were not fully reported in the accessible text. ART medication costs reflected patient co-payment only, as 98.9% of ART is provided free under China's national policy. Cost analysis excluded ART procurement, laboratory monitoring (identical between groups and covered under national free policy), building/infrastructure depreciation, and administrative overhead costs. No NNT/NNH calculable from available data. No safety events or adverse effects differentially reported between arms in the accessible abstract.
Practice implications
This trial established that 6-month ART dispensing is non-inferior to 3-month dispensing for treatment retention and virological suppression in clinically stable PLHIV over 18 months in a Chinese urban academic HIV clinic. The authors describe this as the first individually randomized MMD evidence from Asia, filling a gap previously addressed only by trials from sub-Saharan Africa. The 6-MMD strategy simultaneously reduced provider costs by approximately 27.7% and substantially reduced patient-borne burdens (travel time, transportation costs, work absenteeism). Findings apply most directly to virologically stable adult PLHIV—predominantly male and likely MSM-dominant—managed at an academic HIV clinic in urban northeastern China. Women, patients with recent virological instability or active comorbidities, rural or migrant populations, and those on complex or non-standard regimens were not adequately represented and may face different risk profiles. The marginally higher loss to follow-up in the 6-MMD group (3.5% vs. 2.7%), while small in absolute terms, warrants attention: less frequent contact may disproportionately affect vulnerable patients (e.g., those with migration, stigma-related avoidance, or treatment fatigue), and the study lacked systematic tracing protocols to characterize reasons for loss to follow-up. The RCT design does not capture patient choice or preference, which is central to differentiated service delivery models. The eligibility threshold of HIV RNA <200 copies/mL is less stringent than <50 copies/mL used in some international trials, though 96.9% of enrolled participants were fully suppressed at baseline, supporting the validity of findings. Cost analysis was conducted from a provider and patient perspective using micro-costing, but excluded infrastructure, overhead, and laboratory costs, which limits the completeness of the economic evaluation. The single-center design at a national clinical research center limits generalizability to lower-resource settings, community clinics, or other Asian healthcare systems. Given the consistency of these findings with the INTERVAL trial and prior MMD evidence from sub-Saharan Africa, and the magnitude of cost reductions observed, these results provide meaningful support for 6-MMD policy expansion in China's national HIV program, particularly for clinically stable patients. Confirmation in multi-center and community-based settings and among more diverse patient populations—including women and those with comorbidities—would further strengthen the evidence base before broad programmatic implementation. Only an abstract and publisher-page text were accessible; full-text review may reveal additional subgroup analyses, safety data, and cost figures not captured here. __LIMITATIONS__ Single-center design at a high-volume academic HIV center in urban China limits generalizability to community clinics, rural settings, lower-resource environments, and other Asian health systems.,Open-label (non-blind) design introduces potential performance and detection bias, though blinding is not feasible in this type of pragmatic dispensing trial.,Cohort is 94.8% male, likely predominantly MSM, substantially limiting applicability to women, heterosexual populations, and other key populations.,Eligibility HIV RNA threshold of <200 copies/mL is less stringent than the <50 copies/mL used in some international MMD trials; though most enrollees were fully suppressed at baseline, this may modestly widen the eligible population compared to prior trials.,Loss to follow-up was marginally higher in the 6-MMD arm (3.5% vs. 2.7%); reasons for loss to follow-up (migration, stigma, treatment fatigue) were not systematically traced, representing an important gap.,Cost analysis excluded ART medication procurement, laboratory monitoring, infrastructure, and overhead costs, potentially underestimating total cost differentials and limiting comparability with health technology assessment frameworks.,Productivity loss was estimated using provincial average wage, which may not accurately reflect costs for patients in informal employment.,RCT design does not capture patient preference or choice, which are central to differentiated service delivery model implementation.,Full virological suppression rates by arm at 18 months and complete patient-side cost figures were not fully reportable from the accessible abstract text; clinicians should consult the full publication for complete quantitative data.,Single-site recruitment may introduce selection bias toward more adherent and engaged patients who self-select into academic center care.,18-month follow-up, while meaningful, does not capture very long-term effects of reduced clinical contact on virological outcomes, resistance development, or comorbidity management.,Funding source was not reported in the accessible record, limiting assessment of potential conflicts of interest. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ In clinically stable PLHIV at a Chinese academic HIV clinic, 6-month ART dispensing was non-inferior to 3-month dispensing for treatment retention (risk difference −0.7%, 95% CI −2.9% to +1.5%; non-inferiority margin −5%) and virological suppression at 18 months, while reducing provider-side costs by approximately 27.7% and meaningfully reducing patient visit burden. As the first individually randomized MMD trial from Asia, these findings provide strong evidence to support national policy expansion of 6-MMD for virologically stable adult PLHIV in China, though confirmation in multi-center, community-based, and more diverse (including female and rural) populations is needed before unrestricted programmatic scale-up.