Implementation of Patient-Reported Outcomes in a Medical Oncology Setting (the iPROMOS Study):Type II Hybrid Implementation Study

Background:<bold> </bold>Clinical trials have demonstrated that patient-reported outcome measures (PROMs) can improve mortality and morbidity outcomes when used in clinical practice. Objective:<bold> </bold>This study aimed to prospectively investigate the implementation of PROMs in routine oncology. Outcomes measured included improved symptom detection, clinical response to symptom information, and health service outcomes. Methods:<bold> </bold>Two of 12 eligible clinics were randomized to implement symptom PROMs in a medical oncology outpatient department in Australia. Randomization was carried out at the clinic level. Patients in control clinics continued with usual care; those in intervention clinics completed a symptom PROM at presentation. This was a pilot study investigating symptom detection, using binary logistic models, and clinical response to PROMs investigated using multiple regression models. Results: A total of 461 patient encounters were included, consisting of 242 encounters in the control and 222 in the intervention condition. Patients in these clinics most commonly had head and neck, lung, prostate, breast, or colorectal cancer and were seen in the clinic for surveillance and oral or systemic treatments for curative, metastatic, or palliative cancer care pathways. Compared with control encounters, the proportion of symptoms detected increased in intervention encounters (odds ratio 1.05, 95% CI 0.99-1.11; P=.08). The odds of receiving supportive care, demonstrated by nonroutine allied health review, increased in the intervention compared with control encounters (odds ratio 3.54, 95% CI 1.26-9.90; P=.02). Conclusions:<bold> </bold>Implementation of PROMs in routine care did not significantly improve symptom detection but increased the likelihood of nonroutine allied health reviews for supportive care. Larger studies are needed to investigate health service outcomes.