JMIR Publications

ABSTRACT

Background:

A growing number of cancer and hematopoietic stem cell transplant (HSCT) survivors require long-term follow-up with optimal communication schemes and patients' compliance is crucial. Adolescents and young adults (AYA, age 12-39 years) have various unmet needs. Regarding self-report of symptoms/health status mobile applications (mApp) showed enhanced compliance. Currently, HSCT-aftercare in AYAs is based on handwritten diaries, carrying various disadvantages. Recently, we developed the prototype of a web-based self-monitoring gamified mApp (INTERACCT-App) tailored for AYAs.

Objective:

The prospective study evaluated the INTERACCT-App for tracking real-time self-reported symptoms/health status data in AYA HSCT-patients and a healthy matched control group. We hypothesized that the mApp provides superior medical information for the clinicians than the handwritten diaries.

Methods:

Health data was reported via paper diary and mApp for 5 days, respectively. The quality of medical information was rated (5-point scale) independently and blinded by two HSCT-clinicians and duration of use was evaluated. In 52 participants questionnaires for assessment of gaming patterns and device preferences, self-efficacy, users’ satisfaction, acceptability and suggestions for improvement of the mApp were applied. Interrater-reliability was calculated by the intraclass correlation coefficient (ICC), based on a two-way mixed model; one-way repeated measures ANOVA and t-tests were conducted post-hoc. Descriptive methods were used for correlation with participants’ demographics. For users’ satisfaction and acceptability of the mApp the median and the interquartile range (IQR) were calculated.

Results:

Data of 42 participants (15 patients, 27 healthy students) with comparable demographics were evaluated. We could prove significantly the hypothesis, that the informative content of health data submitted by the INTERACCT-App is superior to the paper (mApp 4.14 vs paper 3.77, p = .023). The mApp outperformed paper and pen mainly within the patients, in particular in patients with treatment-associated complications (mApp 4.43 vs paper diary 3.73, p = .011). The mApp was significantly longer used by AYAs (14years 4,57 days vs 13years 3,14 days, p = .025) and females (4,76 vs 2,95 days, p = .004). This corresponds with a longer duration in impaired patients with comorbidities. User satisfaction and acceptability rating for the mApp was high across all groups, but adherence entering a large amount of data decreased over time. Based on our results we developed a case vignette of the target group.

Conclusions:

We could prove significantly that the patient-reported medical information submitted in the INTERACCT-App embedded in a serious game, is superior to the handwritten diary. In light of the results, a refinement of the mApp supported by a machine learning approach is planned within an international research project.