Peer-reviewed article

Loneliness, connection, and personal networks among perinatal individuals with opioid use disorder: A convergent parallel mixed methods study.

Alicia M Allen, Linnea B Linde-Krieger, Arushi M Chalke, Mariana Felix, Jendar Deschenes et al. · 25 Aug 2026

Abstract

BACKGROUND: Loneliness is highly pervasive among women with opioid use disorder (OUD). Emerging evidence indicates that loneliness, as well as social connection and social support, may influence OUD recovery outcomes. Given that both loneliness and risk for return to opioid misuse are heightened during the postpartum period, a better understanding of loneliness within perinatal OUD is warranted. METHODS: We conducted a convergent parallel mixed methods study with participants who were either pregnant (≥20 weeks gestation) or postpartum (≤25 weeks post-delivery) and had OUD. Data collection included self-administered surveys to measure loneliness, connection, and social support followed by a semi-structured interview that included a personal network analysis (PNA). Survey and PNA data were analyzed with bivariate correlations. Interviews were transcribed and coded. A joint display matrix was used to align qualitative and quantitative observations. RESULTS: Participants (n = 12) identified 135 total network members. Numerous associations were identified. Of highlight, networks with more professional relationships (e.g., healthcare providers) were associated with lower loneliness (r = -0.34, p = 0.028). Type of familial relationships in networks had variable results. Networks with siblings were associated with greater instrumental and emotional support (r = 0.43, p < 0.001; r = 0.46, p = 0.003; respectively). Networks with significant others were associated with less loneliness (r = -0.37, p = 0.041), lower connection (r = -0.30, p = 0.027), and more instrumental support (r = 0.28, p < 0.001). Networks with children had greater emotional support (r = 0.43, p < 0.001). Both contact frequency and length of relationship were inversely associated with emotional support (r = -0.27, p = 0.002; r = -0.19, p = 0.025; respectively). Network density was inversely associated with instrumental and emotional support (r = -0.40, p < 0.001; r = -0.55, p = 0.003; respectively). Generally, qualitative data aligned with quantitative data observations. CONCLUSIONS: Our observations suggest that relationship type (e.g., professional versus personal), interaction characteristics (e.g., contact frequency), and network characteristics (e.g., density) are each uniquely related to feelings of loneliness, social connection, and social support. Future research is warranted to explore how these observations may support OUD recovery in the postpartum period.