Influence of descriptive network norms on injection behaviors among people who inject drugs during the COVID-19 pandemic: a latent profile analysis


BACKGROUND: Descriptive norms within networks influence HIV risk (i.e., sharing injection equipment vs using a new syringe). People who inject drugs (PWID) along the U.S.-Mexico border are at high risk of HIV. The closure of the U.S.-Mexican border during the COVID-19 pandemic posed structural HIV risk, as it disrupted cross-border mobility, harm reduction services, and established networks. We assessed the impact of descriptive network norms on HIV risk and harm reduction behaviors during the COVID-19 pandemic.
METHODS: Participants were PWID aged '¥18 from 3 groups: PWID who engage in 'drug tourism' in Tijuana but live in San Diego (drug tourists (DT)), PWID who have never used illicit drugs across the border and live in San Diego (SD NDT) or Tijuana (TJ NDT). From 10/2020-10/2021, participants were administered a behavioral and egocentric social network questionnaire. We used Latent Profile Analysis (LPA) to categorize PWID into network risk profiles based on proportions of their network members who used injection and non-injection drugs, were DT, lived in Mexico, shared a syringe with the participant, offered the participant drugs, and either doubled daily use or mixed drugs. Multinomial logistic regression was used to assess the influence of network descriptive norms on individual-level HIV risk and harm reduction behaviors in the last 6 months.
RESULTS: Of 399 PWID (n=150 DT, n=90 SD NDT, n=159 TJ NDT), mean age was 43 years and 26% were female. Fit indices indicated a 4-latent profile solution. Network risk groups were classified as 1) very low risk (n=128), 2) low risk (n=94), 3) high risk (n=34), and 4) very high risk (n=93). In the past 6 months, relative to participants in the very low risk group, participants in other groups had higher odds of giving/lending a syringe they used (p<0.001), using a used syringe (p<0.001), using a cooker/cotton after someone else used it (p<0.001); and lower odds of consistently using sterile syringes (p<0.01).
CONCLUSIONS: PWID's gradient of HIV risk behaviors mirrored that of their networks', suggesting that intervening on high risk networks rather than individuals is warranted when services are limited and networks are PWID's main source of influence.