JMIR Publications

ABSTRACT

Background:

As the number of people diagnosed with cancer continues to rise each year, technology has come to the forefront in enhancing the efficacy of self-management for patients recovering from cancer surgery or undergoing chemotherapy. Despite the growing availability of digital technologies, there is still a lack of high-level evidence to support the effectiveness of digital interventions for cancer patients.

Objective:

This study aimed to investigate the effect of long-term physical activity on cancer patients by categorizing them into active and inactive groups based on step count time-series data, using the mHealth intervention (Walkon app).

Methods:

Cancer patients who had previously used the Walkon app in a prior randomized controlled trial were chosen for this study. Individuals without any recorded step count data were excluded from the analysis. The Walkon app provided step count data from the patients. Biometric measurements, including BMI, waist circumference, blood sugar levels, and body composition, along with Quality of Life (QOL) questionnaire responses (EQ-5D-5L and HINT-8), were collected during both the baseline and 6-month follow-up at an outpatient clinic. To analyze step count patterns over time, the Sample Entropy module in Python was used for patient clustering, distinguishing between the Active Walking Group (AWG) and Inactive Walking Group (IWG). Statistical analysis was performed using the Shapiro–Wilk test for normality, with paired t-tests for parametric data, Wilcoxon signed-rank tests for non-parametric data, and chi-squared tests for categorical variables.

Results:

The results validated our methodology by successfully categorizing patients based on step count trends. The Inactive Walking Group (IWG) and the Active Walking Group (AWG) had significant disparities in daily, weekly, and monthly step counts between them. Higher physical activity levels were observed in breast cancer patients and younger individuals. However, there were no significant differences in biometric data between the baseline and the six-month follow-up for both the AWG (n=137) and IWG (n= 75) groups. In contrast, in terms of QOL assessments, pain (p=0.01) and the Visual Analog Scale (VAS) (p=0.03) exhibited significant differences. Nonetheless, within the realm of biometric measurements, only waist circumference (p=0.01) and visceral fat (p<0.01) demonstrated a significant improvement exclusively within the AWG. Furthermore, regarding the QOL questionnaire, aspects such as energy (p=0.01), work (p<0.01), depression (p=0.02), memory (p=0.01), and happiness (p=0.05) displayed significant improvements solely in the AWG, while other factors did not exhibit statistical significance.

Conclusions:

This study introduces a novel methodology for categorizing cancer patients based on physical activity using step count data. While significant improvements were noted in the AWG, particularly in QOL and specific physical metrics, differences between the AWG and IWG were statistically insignificant. These findings highlight the potential of digital interventions in improving cancer patient outcomes, contributing valuable insights to cancer care and self-management.