Background: Shared decision making is preferred by many patients but it is often viewed from the lens of the patient and physician dyad only. Caregivers often support cancer decision-making, although their decision control preferences are not typically considered. We explored decisional control preferences of older patients with cancer and their caregivers, and examined concordance in preferences of the dyad, as well as patient, caregiver, physician factors associated with concordance. Methods: We utilized data from a national geriatric assessment (GA) cluster-randomized trial (NCT 02054741; PI: Mohile; funding NCI UG1CA189961) that recruited patients aged ≥70 with incurable cancer, their caregivers, and oncologists. Controlled Preferences Scale (CPS) was used to assess shared decision making preferences. Both patients and caregivers were asked about the role of patients (Patient-role CPS) and caregiver (Caregiver-role CPS) in the decision making process. Patient-role CPS options were 1) doctor makes the decisions, 2) doctor makes the decisions with patient’s input, 3) equally shared decision, 4) patient makes the decisions with doctor’s input, 5) patient makes the decisions. Caregiver-role CPS options were 1) doctor makes the decisions, 2) patient makes the decision with doctor, 3) patient and caregiver make the decisions with doctor, 4) caregiver makes the decisions with doctor. Matching responses of the dyads for each CPS question were considered concordant. We used descriptive statistics to summarize CPS for patients and caregivers and the concordance between them. We utilized logistic regression to assess dyads' sociodemographic information, patients' GA and physicians’ practice characteristics associated with patient-caregiver CPS concordance. Results: Enrolled dyads with completed CPS were included (N=332); mean age (SD) of patients and caregivers was 76.6 (5.3) and 66.6 (12.2). Women constituted 39% of patients and 76% of caregivers. Patient-role CPS for patients’ (and caregivers’) were 11% (6%) for doctor makes the decisions, 33% (28%) for doctor makes the decisions with patient’s input, 45% (49%) for equally shared, 9% (14%) for patient makes the decisions with doctor’s input, 2% (3%) for patient makes the decision. Patients’ (and caregivers’) answers on caregiver-role CPS were 20% (19%) for doctor makes the decisions, 18% (28%) for patient makes the decisions with doctor, 61% (52%) for patient and caregiver make the decisions with doctor and 1% (1%) for caregiver makes the decisions with doctor. Dyads concordance occurred in 46% of patient-role CPS and 54% of caregiver-role CPS, indicating moderate concordance. No baseline factors were associated in dyad concordance for either the patient-role or caregiver-role CPS. Conclusions: Clinicians should inquire about decisional control preferences of patients and caregivers in order to optimally support shared decision making. Clinical trial information: NCT02054741.