Symptom Monitoring with Patient-Reported Outcomes during Definitive Radiation Treatment

Abstract

Purpose/Objective(s)

While patient symptom self-reporting improves clinical outcomes in patients with metastatic cancer receiving outpatient chemotherapy, the effects are not well established in the setting of patients undergoing radiation therapy (RT). In this phase II multi-institutional study, we assessed the impact of patient symptom self-reporting using the mPROS mobile application during definitive chemoradiation on health-related quality of life (HRQOL).

Materials/Methods

Patients initiating definitive radiation therapy with concurrent chemotherapy for GI, GYN, lung, CNS, and H&N cancers were randomly assigned to symptom self-reporting with mPROS (experimental arm; EXP) or usual care (control arm, CON). In the EXP arm, participants were asked to report symptoms at least weekly; severe or worsening symptoms triggered alerts to the clinical team. Participants remained on study until 3 months after completion of RT. PROMIS-29, a validated questionnaire assessing HRQOL, was administered at baseline, completion of RT (EOT), and 3 months after RT completion. The primary endpoint was HRQOL at EOT. Patient assessment of the usefulness of mPROS was measured via a 24 question exit survey using a 7-point Likert scale.

Results

Fifty-nine (n = 30 experimental, and n = 29 control) patients (mean age 61.9 years. 51% male) were included, of which 46 (n = 22 experimental and n = 24 control) completed both baseline and EOT PROMIS-29 questionnaires. On average, EXP participants logged symptoms twice per week, with a mean of 15 severe attributes and 11 unique symptoms reported. There were no significant differences between mean changes to PROMIS physical (EXP, 52.31 to 46.12, ∆ -6.03 vs. CON 50.72 to 45.77, ∆ -4.04, P = 0.45) or mental (EXP 49.25 to 45.80, ∆ -2.82 vs. CON 50.82 to 47.52, ∆ -2.17, P = 0.78) function scores from baseline to EOT. Likewise, there were no significant differences between median changes to PROMIS physical (EXP 56.12 to 44.92, ∆ -3.87 vs. CON 49.14 to 44.87, ∆ -2.02, P = 0.29) or mental (EXP 49.79 to 43.82, ∆ -1.74 vs. CON 52.47 to 48.22, ∆ -1.39, P = 0.77) function scores from baseline to EOT. On multivariate linear regression adjusting for age, sex, race, ethnicity, and education, treatment arm was not a statistically significant predictor of changes in physical (2.11 (95% CI = -3.65 to -7.87), P value = 0.48) or mental scores 0.12 (95% CI = -5.14 to -5.38, P = 0.96) between baseline and EOT. On exit survey, mean scores for selected questions were as follows: care team understanding symptoms (6.27), feeling more involved in care (6.36), help with symptom tracking (6.18), and recommending mPROS to other patients (6.09).

Conclusion

Use of a mobile application to monitor symptoms, compared with usual care, did not result in statistically significant differences in physical or mental HRQOL. However, patients in the experimental group expressed high satisfaction with self-reporting, feeling more engaged in their cancer care and helping them keep track of symptoms.