Preplanned Studies: An Index–Contact Paired Data Analysis on Sexual Contact Tracing Outcomes of HIV-Infected Individuals — Yunnan Province, China, 2022–2024

Despite numerous innovative interventions, achieving the second 95 target [95% of people living with human immunodeficiency virus (PLWH) know their human immunodeficiency virus (HIV) status] set by the Joint United Nations Programme on HIV and AIDS (UNAIDS) remains challenging (1). In China, approximately 20% of PLWH were estimated to be unaware of their infection status in 2020 (2), with more than 40% of HIV-infected individuals receiving diagnoses at advanced stages of infection (3). This late diagnosis pattern leads to poorer clinical outcomes and impedes efforts to control HIV transmission (4). Contact tracing has emerged as a promising strategy for controlling the spread of infectious diseases (5–7). While previous research has primarily focused on examining associations between contact characteristics and tracing outcomes, the intimate nature of HIV transmission suggests that incorporating index case characteristics could provide crucial insights into understanding transmission networks within populations (8). To enhance and optimize contact tracing strategies, we conducted an egocentric contact tracing study from 2022 to 2024 in Yunnan, China. Our findings demonstrate that both index case and sexual contact characteristics significantly influence sexual contact tracing outcomes. These results suggest that interventions leveraging these facilitating factors offer promising pathways toward achieving the UNAIDS targets.

This egocentric contact tracing survey was conducted between January 2022 and June 2024 in Honghe Hani and Yi Autonomous Prefecture (Honghe Prefecture) of Yunnan Province, China. Newly diagnosed HIV-infected individuals aged 18 years or older who could provide sexual contact information were eligible and invited to participate as index cases. After providing informed consent, participants were interviewed by trained local health specialists using an anonymous questionnaire with unique identification numbers linking index cases. Interviews were conducted in private rooms where participants provided detailed information about their individual characteristics and all sexual partners with whom they had engaged in sexual intercourse. Index cases were subsequently encouraged to notify their sexual partners and refer them for HIV counseling and testing. Sexual contacts who tested HIV-positive were invited to participate as new index cases in subsequent rounds of contact tracing. This iterative process continued until either no additional HIV-positive contacts were identified or no further sexual contacts could be traced. The study protocol was approved by the Ethics Committee of Yunnan Center for Disease Control and Prevention (Permit Number: YNCDC/QR-KJB-2021-003).

Contact tracing outcomes were classified as “successful” when at least one HIV-positive sexual partner was identified from an index HIV case. Testing modalities were categorized as either active, defined as HIV testing at a voluntary counseling and testing (VCT) clinic, or passive, which included all other identification pathways [e.g., provider-initiated testing and counseling (PITC)]. Data analysis was conducted at both index case and contact levels. At the index case level, bivariate and multivariate logistic regression analyses evaluated associations between index case characteristics and tracing outcomes. At the contact level, index–contact paired data were analyzed using multilevel logit models to assess relationships between both index case and contact factors related to HIV testing uptake and infection status among sexual contacts. All statistical analyses were performed using SAS (version 9.4; SAS Institute Inc., Cary, NC, USA) with a significance level of 0.05.

The contact tracing process and identification of HIV infections are detailed in Supplementary Table S1 and Figure S1. Among 1,981 enrolled index cases (1,925 newly reported during the study period, 54 identified from first-round contact tracing, and 2 from second-round), a total of 2,171 sexual contacts were reported, with a mean of 1.1 partners per index case (range: 1–4) and a median of 1 partner. Of these contacts, 69.5% (1,509/2,171) underwent HIV testing, with 21.0% (317/1,509) testing positive. The sexual network index (9) (contacts recruited per HIV-infected index case) was 0.8 (1,509/1,981). In the first round, 1,925 index cases reported 2,097 contacts, of whom 1,474 were tested and 308 were positive. Of these positive contacts, 54 participated as second-round index cases, reporting 69 contacts, with 32 tested and 9 positive. In the third round, 2 of the 9 positive contacts participated as index cases, reporting 5 contacts, of whom 3 were tested with no positive results.

The demographic profile of index cases showed a predominance of male individuals, aged 36–55 years, from ethnic minority backgrounds, with primary school education or less, who were married and employed. Notably, 85% were identified through passive testing, and approximately 35% presented with CD4 counts ≤200 cells per microliter at baseline. Sexual contacts were predominantly female (>50%), aged 36–55 years, from ethnic minorities, with primary school education or less, and employed. Most contacts were married, with approximately half engaged in non-marital, non-commercial sexual partnerships with their index case. The majority of relationships (71.9%) were of 1 year or less duration, with nearly 65% reporting sexual contact once or twice weekly in the previous 6 months. Condom use was notably low, with 65.8% reporting never using condoms during sexual encounters with their index case. Significant differences between index cases and sexual contacts were observed across gender, age, ethnicity, education, and marital status (Supplementary Table S2).

The success rate of tracing HIV-positive sexual contacts varied significantly by index case characteristics, with notably higher rates among female and married index cases. Additionally, index cases identified through active testing methods demonstrated superior contact tracing outcomes (Table 1).

Multivariate analysis revealed that sexual contacts of ethnic minorities, those with higher educational attainment, and employed individuals demonstrated increased likelihood of HIV testing uptake. Testing rates were notably higher among spouses and long-term partners, those in established relationships, individuals reporting frequent sexual activity, and those practicing consistent condom use. The probability of HIV infection was significantly elevated among contacts whose index case was female and identified through active testing protocols. Additionally, established relationships, particularly among spouses and long-term partners, exhibited higher infection rates compared to temporary or commercial relationships. HIV positivity was markedly increased among individuals reporting inconsistent or no condom use versus those maintaining consistent condom use practices (Table 2).