The usefulness of HIV partner services in the age of treatment as prevention: a registry-based study.

Background: Partner services are effective tools to identify new cases among sex or needle-sharing partners of people with a new HIV diagnosis. Little is known about partners previously diagnosed with HIV who are not in care or are in care with unsuppressed HIV viral load. We aimed to quantify the previously diagnosed partners of people with a new HIV infection and examine their HIV care status and viral suppression in the 12 months before elicitation.

Methods: We did a registry-based study. We used the New York City HIV Surveillance Registry to determine HIV care status and viral load of partners elicited from newly diagnosed people between Jan 1, 2007, and Dec 31, 2018. Previously diagnosed partners with no report of CD4 count or viral load in the preceding 12 months were presumed not to be in care, viral load suppression (<200 copies per mL) was based on the last viral load in the year preceding elicitation, and viraemia was defined as a viral load of 200 copies per mL or more. We used multinomial logistic regression to generate covariates of care and viral load status and their marginal effects.

Findings: 11 964 partners were elicited; 2603 (33%) were previously diagnosed and 485 (20%) were not in care. 1153 (49%) of 2343 with a viral load report were in care and viraemic at elicitation. The odds of being not in care were higher in non-Hispanic black than non-Hispanic white or other partners (adjusted odds ratio 1·89, 95% CI 1·09-3·27) and lower in partners with male-to-male sex transmission risk (0·37, 0·26-0·51) and country of birth other than the USA (0·57, 0·39-0·85). The odds of being viraemic were higher in partners younger than 30 years than in those aged 30 years or older (1·68, 1·35-2·09) and lower among people with male-to-male sex transmission risk (0·36, 0·29-0·44) and country of birth other than the USA (0·78, 0·66-0·97).

Interpretation: People with HIV should receive ongoing HIV prevention counselling and partner services data should inform engagement in care for previously diagnosed partners.

Funding: None.