Patients with cancer experience considerable symptom burden, psychological morbidity, and unmet psychosocial needs. Research suggests that feedback of patient-reported outcomes to clinicians or caseworkers, alongside management strategies, may result in improved patient functioning. Two intervention models were developed to test this effect in a randomized, controlled trial against usual care (UC): a telephone caseworker (TCW) model and an oncologist/general practitioner (O/GP) model. Primary end points included anxiety, depression, physical/emotional functioning, and unmet supportive care needs. Participants with nonlocalized breast or colorectal cancers were surveyed by computer-assisted telephone interview (CATI) at three time points: baseline, 3 months, and 6 months. Data collected from participant CATIs in the supportive care models were used to generate feedback to either each participant's designated TCW, or their nominated O/GPs. Data obtained from participants in the UC model were used only to assess the impact of supportive care models. In total, 356 participants consented to study participation, completed the baseline CATI, and were randomly assigned to the UC, TCW, or O/GP groups. No overall intervention effect was observed. Physical functioning was significantly improved at the third CATI for participants in the TCW model (P = .01), and there was a trend toward fewer participants with unmet needs (P = .07). TCW group participants also were more likely to have the following: identified issues of need discussed (P < .0001); referrals made (P < .0001); and strong agreement that the intervention improved communication with their health care team (P = .0005). The TCW model holds some promise; however, additional work in at-risk populations is required before we recommend implementation.