Individualised care for patients with breast or prostate cancer aided by an interactive app : a frame of process evaluation

Abstract: Background: Patients receiving outpatient cancer treatment often experience distressing symptoms and unmet needs. Collecting patient-reported outcomes via apps (ePROs) facilitates patient-clinician communication regarding symptoms and is recommended in clinical guidelines. Previous studies of an interactive app (Interaktor) for individualised symptom management show reduced symptom burden for patients undergoing breast and prostate cancer treatment. Aim: To contribute to the knowledge of the value of implementing ePRO in clinical practice by studies framed as a process evaluation of an intervention for individualised symptom management assisted by Interaktor. Methods: Following the Medical Research Council framework for process evaluation of complex interventions, qualitative and quantitative data were collected along two randomised controlled trials (RCTs). Patients receiving neoadjuvant chemotherapy for breast cancer (N=149), and radiotherapy for prostate cancer (N=150) were randomised to standard care with or without intervention. Intervention group patients reported symptoms and concerns daily by questionnaire and free text. The app included selfcare advice and symptom history graphs. Oncology nurses responded to alerts triggered by severe symptoms. Study I investigates which and how patients engaged, by analysing adherence and usage predictors from logged data and telephone interviews with patients. Study II analyses the effects on patients' perceptions of individualised care and health literacy by questionnaires. Study III assesses if the intervention is cost-effective according to the Swedish National Board of Health and Welfare. Cost-effectiveness analyses (CEA) estimate gains in Quality-adjusted lifeyears (QALYs), intervention costs, and the patient's healthcare utilisation as obtained from the Stockholm Council database. Acute healthcare use is also explored. Results: Study I shows that adherence to daily symptom reporting was 83 %; most patients used the self-care advice and free text. Patients regarded the app easy to use and helpful for self-management. Marital status, age, education level, and comorbidity were associated with usage variations. Study II shows no between-group differences in individualised care or health literacy among patients with breast cancer. Intervention group patients with prostate cancer rated their support for decision control as more individualised than their control group, and their ability to seek, understand and communicate health information improved. Study III shows the intervention produced significantly more QALYs, although the effect was small. The weekly intervention cost per patient was low. The cost-effectiveness depended on the type of healthcare costs studied. The intervention was cost-effective for patients with breast cancer if non-acute healthcare costs were excluded, and for patients with prostate cancer, considering all healthcare costs. Healthcare costs varied greatly. Patients with breast cancer who used the app had more acute visits for fever. Patients with prostate cancer who used the app had fewer acute visits for urinary problems. Conclusions: Patients used and valued Interaktor as promoting assurance and participation in care. Using the app can positively affect care individualisation and health literacy for patients with prostate cancer during radiotherapy. It may be beneficial to increase the individualisation of features and settings for patients with breast cancer. The intervention may be cost-effective, but to show if healthcare savings can be achieved requires a larger study.