JMIR Publications

ABSTRACT

Background:

Non-communicable diseases represent a major global health burden, causing 74% of all deaths worldwide. Effective management of these diseases requires ongoing coordination and monitoring, often facilitated by e-health technologies such as mobile health (mHealth) applications. The end-user Mobile Application Rating Scale (uMARS) is a valuable tool for assessing the quality of mHealth apps from a user perspective, but it is not currently available in French despite it is the5th most spoken language worldwide. This lack of a French version limits the tool's use in French-speaking countries, where mHealth app evaluation and regulation remain insufficiently addressed although the growing number of applications and a high level of interest in their importance for healthcare.

Objective:

This study aims to translate, culturally adapt a French version of the uMARS (uMARS-F) and to test its overall and internal reliability.

Methods:

Cross-cultural adaptation and translation were carried out with the universalist approach. The uMARS-F evaluation was performed using the MonSherpa application. One hundred sixty-seven participants assessed the app using the uMARS-F twice (T1 and T2) one week apart. Summary scores for each section and overall were assessed for normal distribution using the Shapiro-Wilk test, and means, standard deviations and potential floor or ceiling effects were calculated accordingly. Overall reliability was evaluated using Intraclass Correlation Coefficients (ICCs) and internal reliability using Cronbach’s Alpha. Concordance between the 3 subscales (objective quality, subjective quality, and perceived Impact), 4 sections and 26 items at T1 and T2 was evaluated using the Paired t-test and Pearson correlation coefficient.

Results:

The uMARS-F ICCs were excellent for objective quality (0.959), excellent for subjective quality (0.993) and moderate for perceived Impact (0.624). Cronbach’s alpha was good for objective quality (0.881), acceptable for subjective quality (0.701) and excellent for perceived impact (0.936). The paired t-tests demonstrated similar scores between the two assessments (P>0.05) and the Pearson Correlation coefficient indicated high consistency in each subscale, section and item (r>0.76 and P<0.001). Reliability and validity of the measures were similar to those found in the original English version, as well as in the Spanish, Japanese, Italian, Greek, and Turkish versions that have already been translated and validated.

Conclusions:

The uMARS-F is a valid tool for end-users to assess the quality of mHealth apps in French-speaking countries. French app developers and researchers could use the uMARS-F to evaluate the end-user perspective and improve the quality of mHealth apps before their market launch or after introduction of new functionalities. The uMARS-F used in combination with the MARS-F could now enable healthcare professionals and public health authorities to identify reliable, high quality and valid apps for the patients, and should be part of French healthcare education programs in the future.