CROI 2018 Abstract eBook

Abstract eBook

Poster Abstracts

of 80 days (IQR: 45-102). Given the limited number of early gaps in UG (n=21), we explored associated findings in an adjusted model in SA (n=71) (see Table 1). In the adjusted model, PLWH in SA who initiated ART later had double the odds of early gaps in treatment (aOR=2.0, 95%CI: 1.0-4.0). Those who used denial to cope were also at higher odds of early losses (aOR=1.2, 95%CI: 1.0-1.4). Education provided a protective effect against early losses (aOR=0.4, 95%CI: 0.2-0.8). There was no significant difference across age, gender, marital status or employment. Those with early gaps were more likely to have detectable viremia across sites (OR:6.3, p=0.001 in UG vs. OR:2.6, p=0.008 in SA). Conclusion: Despite global efforts to promote early and enduring treatment, early gaps in ART persist, resulting in higher odds of detectable viremia. These gaps remain significant for key vulnerable populations, specifically those who present late to care, who lack educational opportunities, and who use denial to cope.

and 245 at control sites (44%> 40, 67% female, median CD4 count 212 cells/ μl3). We found no difference between groups in resuppression. During the study period, control sites incorporated rapid tracing into standard care, however, potentially masking intervention effects. Conclusion: EAC appears to increase viral re-suppression modestly for patients who return to the clinic for a three-month viral load but as most did not return, the overall effect was small. Implementation of the tracing intervention under the new guidelines did not differ from standard care. Interventions that aim to return unstable patients to care should incorporate active monitoring to determine if the interventions are effective.

Poster Abstracts

1141 SUSTAINABLE VIRAL LOAD MONITORING SCALE-UP: GEOSPATIAL OPTIMIZATION MODEL FOR ZAMBIA Brooke E. Nichols 1 , Sarah Girdwood 2 , Thomas Crompton 3 , Lynsey E. Stewart- Isherwood 4 , Leigh Berrie 4 , Dorman Chimhamhiwa 3 , Crispin Moyo 5 , John Kuehnle 6 , Sydney Rosen 1 1 Boston University, Boston, MA, USA, 2 Health Economics and Epidemiology Research Office, Johannesburg, South Africa, 3 Right to Care, Johannesburg, South Africa, 4 National Health Laboratory Service, Johannesburg, South Africa, 5 EQUIP Health Zambia, Lusaka, Zambia, 6 United States Agency for International Development, Washington, DC, USA Background: WHO recommends viral load monitoring at 6 and 12 months, then annually, after initiating ART. Expansion of viral load testing has been slow in many countries due to lack of an efficient system for blood sample transportation. An estimated 1.2 million people are infected with HIV in Zambia. Methods: The model optimizes an STN in Zambia for the anticipated 1.5 million viral load tests that will be needed in 2020, taking into account the country’s geography, infrastructure, and district political boundaries. Data incorporated into the model included the location of all 2,500 Zambian health facilities and laboratories, lab and hub infrastructure and capacity, driving distances and driving times for different types of vehicles, and expected future viral load demand by health facility. Under the status quo, each district independently provides sample transport for facilities within its borders. We evaluated the all-inclusive STN costs of 2 alternative scenarios: 1) an optimized status quo where each district provides its own weekly or daily sample transport for the anticipated viral load volume; and 2) an optimized borderless STN that ignores district boundaries, provides weekly or daily sample transport, and reaches the same facilities/viral load volumes as scenario 1. Results: Under both scenarios, coverage of viral load testing would increase from 10% in 2016 to 89% in 2020. Mean transport cost per viral load in scenario 2 was $1.86 per test (SD $0.27), 55% less than the mean cost/test in scenario 1 of $4.14 (SD $0.70). When fully scaled-up to the anticipated 2020 volumes, the borderless systemwould save the government of Zambia $3,537,000 annually (SD $660,000) compared to the district-based system. This saving is primarily due to a reduction in the number of vehicles and drivers needed, along with more efficient routes enabled by intra-district routing. Conclusion: We found that an efficient STN that optimizes sample transport on the basis of geography and test volume, rather than political boundaries, can cut the cost of sample transport by more than half. This model, which can be used in other countries and for other types of samples, has the potential to increase the sustainability of ART programs throughout Africa. To assist Zambia in scaling up testing capacity, we designed a geospatial optimization model to minimize the cost of a national viral load sample transportation network (STN).

1140 VIRAL SUPPRESSION EFFECTS OF INTERVENTIONS FOR UNSTABLE ART PATIENTS IN SOUTH AFRICA Matthew P. Fox 1 , Sophie Pascoe 2 , Amy Huber 2 , Joshua Murphy 2 , Mokgadi Phokojoe 3 , Marelize Gorgens 4 , Sydney Rosen 1 , David Wilson 4 , Yogan Pillay 3 , Nicole Fraser 4 1 Boston University, Boston, MA, USA, 2 Health Economics and Epidemiology Research Office, Johannesburg, South Africa, 3 South African National Department of Health, Pretoria, South Africa, 4 World Bank, Washington, DC, USA Background: As loss from HIV care is an ongoing challenge globally, interventions are needed for patients who don’t achieve or maintain ART stability. The 2015 South African National Adherence Guidelines (AGL) for Chronic Diseases include two interventions targeted at unstable patients: rapid tracing of patients who miss visits (TRIC) and enhanced adherence counselling (EAC). Methods: As part of a cluster-randomized evaluation at 12 intervention and 12 control clinics in 4 provinces, intervention sites implemented the AGL interventions, while control sites retained standard care. We report early outcomes of EAC for patients with an elevated viral load (>400 copies/ml3) and on TRIC who missed a visit by >5 days. We included patients meeting these criteria from 20 June 2016 and 16 December 2016 and followed them through record review. We estimated risk differences (RD) of 3 month viral resuppression (<400 copies/ml3) with cluster adjustment using generalized estimating equations and controlled for imbalances using difference in differences compared to all eligible for these strategies in 2015, prior to intervention roll out. Results: For EAC, we had 358 intervention site and 505 control site patients (61% female, median ART initiation CD4 count 157 cells/μl3). Few in either group had evidence of resuppression by 3 months (4.2% EAC vs 4.7% control) but few had a three-month repeat viral load recorded (71/358 intervention, 68/505 control). Among all eligible for EAC with a repeat viral load in the intervention- period (n=934), EAC showed a small increase in resuppression (28% vs 25%, RD 3.0%; 95% CI-2.7% to 8.8%)(Table). Adjusting for baseline differences increased the RD to 8.1% (-0.1% to 17.2%). For TRIC, we enrolled 155 at intervention sites

CROI 2018 441