Using the Actor-Partner Interdependence Model to explore the psychological impact of COVID-19 on anxiety in dyads of patients with cancer and caregivers.

Background: Delineating the compound psychological effect of the pandemic on cancer care, and the interdependency across cancer patient-caregiver dyads have yet to be explored. This study examines the levels of psychological impact of COVID-19 on patient-caregiver dyads anxiety, and the interdependent associations between their COVID-19 and cancer concerns, and risk perceptions.

Method: There were 352 patients and caregivers (patient-caregiver dyads, N = 176) included in this study (43.2% spousal dyads). Generalized Anxiety Disorder-7 and questionnaires regarding risk perception, perceived confidence in healthcare system, COVID-19, and cancer-related concerns were administered. Actor-Partner Interdependence Model (APIM) analyses were used to determine the interdependent effects. Indirect effects were tested using mediation pathway analyses.

Results: Patients reported significantly higher levels of risk perceptions and anxiety than their caregivers (p < 0.01). Anxiety rates (GAD-7 ≥ 10) were also significantly higher (26.7% vs 18.2%, p < 0.01). Dyads' anxiety, "general COVID-19 concerns," "cancer-related concerns," and risk perceptions were correlated (ps < 0.01). APIM showed only actor effects of general COVID-19 concerns, cancer-related COVID-19 concerns, and risk perceptions on anxiety (βs = 0.19-0.53, ps < 0.01). No partner effects were observed. Similar results were found in the composite APIM. Indirect effects of the patient/caregiver's variables on their partner's anxiety were observed in the mediation analyses.

Conclusion: Concerns about COVID-19 and cancer care could be indirectly associated in patient-caregiver dyads and need to be proactively addressed. As pandemic evolves into endemicity, engagement with patients and caregivers should strive to be sensitive to their differential needs and messages should be tailored to the informational needs of each.