Associations of caregiver-oncologist discordance in prognostic understanding with caregiver-reported therapeutic alliance and anxiety

15 Background: Discordance in prognostic understanding between caregivers of adults with cancer and the patient’s oncologist is common. However, the relationship between caregiver-oncologist discordance and caregiver bereavement outcomes is unknown. We evaluated the associations of caregiver-oncologist discordance in beliefs about the patient’s curability and life expectancy with caregiver-reported therapeutic alliance and anxiety. Methods: This is a secondary analysis of a multicenter study that assessed the effect of a communication intervention among patients with advanced cancer and their caregivers. Prior to intervention exposure, caregivers and oncologists were asked about their belief in the patient’s chances for cure and living ≥2 years: 100%, about 90%, about 75%, about 50/50, about 25%, about 10%, and 0%. Discordance was defined as a difference by 2 response levels on each prognostic understanding item. Outcomes at 7 months after patient death included caregiver-reported therapeutic alliance [modified 5-item Human Connection (THC) scale] and anxiety (Generalized Anxiety Disorder-7). We used multivariable linear regression models to assess the independent associations of discordance with therapeutic alliance and anxiety. Results: We included 97 caregivers (mean age 63, range 22-83). Approximately 40% of caregiver-oncologist dyads had discordant beliefs about curability (caregivers were more optimistic in 100% of dyads) and 63% had discordant beliefs about life expectancy (caregivers were more optimistic in 94% of dyads). On multivariate analysis, discordance in beliefs about prognostic estimates was associated with lower THC score (b = -6.94, SE 3.17, p = 0.03). Discordance in beliefs about curability was associated with lower anxiety levels (b = -1.79, SE 0.90, p = 0.05). Conclusions: Caregiver-oncologist discordance may decrease caregiver-reported therapeutic alliance and anxiety, both of which may shape how caregivers interact with the healthcare system. A better understanding the role of caregivers’ prognostic understanding will guide interventions to improve caregiver-oncologist therapeutic alliance and caregiver anxiety. Clinical trial information: NCT01485627.