Development and Testing of a TTDSS for Cancer Patients

This study aimed to understand the use of different educational models (eRAPID-MV) and self-management behaviors of cancer patients in managing treatment-induced side effects at home. To this end, interventions were provided through a decision support system involving telephone triage. Finally, the study explored the effectiveness of the intervention method (eRAPID-MV \& TTDSS) on patient care outcomes and telephone triage service outcomes.

Study design/setting This was a prospective, nonrandomized clinical study.Study procedures and data collection were carried out at the outpatient department of St. Martin De Porres Hospital, Taiwan. The trial followed the Declaration of Helsinki and Good Clinical Practice guidelines for trial conduct. Participants a breast surgeon or colorectal surgeon or hematologist oncologist were invited or referred to participate in the research project. Participants provided their written informed consent prior to taking part in this 24-week program. Participants has received a weekly reminder from the APP or emails to complete an online TRAEs (eRAPID-MV) assessment at home. Participants used an electronic device (e.g., smart phone, tablet, computer) with real-time electronic data transfer to the Health Care Information System (HIS). During the study period, participants were compared to a usual care control group on patient care outcomes and telephone screening service outcomes.

Interventions Two key intervention components were standardized across participating hospitals: Phase I Interventions: eRAPID-MV system-a) TRAEs surveys, b) survey feedback reports, and c) patient self-management and clinical alerts; Phase II Interventions: TTDSS.

The Phase I intervention details are as follows: (1) The patients completed the TRAEs survey-18 items. Based on the survey results, the eRAPID-MV system provided a summary report of the traffic signal and recommendations to the patient for seeking medical advice (red or yellow light), and a link to the TRAEs websites for self-management (green light). The TRAEs websites on health education summarized available information from symptom management guidelines, telephone triage protocols for nurses, and published articles. (2) The survey results were also integrated into the hospital information system (HIS) and case management system (Cancer Patient Registry) to enable the clinician and case managers to access report summaries and review patient-reported outcomes for severe TRAEs. (3) All the patients (experimental group) completed the TRAEs survey every week during their active treatment period, and the eRAPID-MV system's automatic schedule was sent to them. (4) The intervention duration was 24 weeks. (5) Participants were asked to fill out a satisfaction survey on weeks 2, 12, and 24. The system usability scale (SUS) was also used on the day of the final survey. Finally, patients were recruited to attend a single interview that assessed their views on the eRAPID-MV system. Participation was voluntary, and the interview lasted approximately 20-30 minutes.

The Phase II intervention details are as follows: (1) a phone consultation with the oncology case manager was determined based on the patient's TRAEs (red or yellow light-experimental group or the need for a return visit). The oncology case manager referred to the last TRAEs data and inquired about the patient's present condition. Based on the information and the purpose of the patient's call, the oncology case manager evaluated the severity of TRAEs by referring to the guidelines for TTDSS. The patient received the best possible recommendation to help manage TRAEs based on a summary report of the traffic signal. When the light was red or yellow, the oncology case manager arranged for immediate outpatient or emergency medical treatment and assisted in hospitalization if necessary. (2) Following Phase II intervention (TTDSS) for 72 hours, the first post-test was performed to assess the immediate effects on the patient. (3) Long-term effects were also evaluated, including the effectiveness of patient care outcomes and telephone triage service outcomes on cancer patients after 24 weeks of intervention. (4) Finally, the oncology case manager asked to attend a single interview that assessed their views on the TTDSS. Participation was voluntary, and the interview lasted approximately 20-30 minutes.

Data collection and management For the 24-week study period, participants were evaluated at baseline, and every week (experimental group) or every 3 weeks (control group) by TRAEs assessment. Participants were asked to complete 3, brief, self-administered questionnaires: the treatment-induced adverse events (TRAEs) questionnaire, satisfaction questionnaire, system usability scale (SUS), and interview with patients (experimental group). All questionnaires and the consent form were translated into Chinese (Traditional) by a certified translator. The TRAEs data were fully integrated into HIS and the case management system (including the Cancer Patient Registry) of St Martin De Porres Hospital (id: STM-IRB-22B-001-2022-04). Data were processed anonymously; therefore, data for which participants' identities could be identified were omitted. Strict safety and confidentiality measures were established in the archives of clinical trial institutions. Every six months, the Institutional Review Board (IRB) will be responsible for checking the status of the study and the quality of the data collected.

Sample size calculation The multiple linear regression statistical method was followed to estimate the sample size of the cluster, and the number of patients required for the study was determined accordingly. The total sample size required for the computer software calculation was 114. Considering the attrition rate of approximately 20%, the estimated sample size for this study was at least 136, and the average sample size for each group was at least 68.

Statistical analysis Descriptive statistics, including means, standard deviations, frequencies, and percentages, and chi-square tests and independent t-tests, were used to test the homogeneity of context-level factors (e.g. demographic characteristics, disease status, and family caregiver support) between the two groups. Binary logistic regression and polynomial linear regression were used to analyze the factors influencing self-management behaviors. Inferential statistics, including paired sample t-test, 2-group sample t-test, Mc Nemar's test, and analysis of covariance (ANCOVA), were used. Following the eRAPID- MV and TTDSS intervention, analyses were performed to assess changes in patient care outcomes (TRAEs severity, disconnection rate, hospitalization rate, retention rate, and completion rate) and triage service outcomes (ease of use of the system).