Modification and Validation of an mHealth App Quality Assessment Methodology for International Use: Cross-sectional and eDelphi studies (Preprint)

Abstract

BACKGROUND Over 325,000 mobile health (mHealth) applications (apps) are available to download across various app stores. Quality assurance in this field of medicine remains relatively undefined, however. Globally around 84% of the population have access to mobile broadband networks. Given the potential for mHealth app use in health promotion and disease prevention, their role in medicine world-wide is ever apparent. Quality assurance regulations both nationally and internationally will take time to develop. Frameworks such as the Mobile App Rating Scale (MARS) and Enlight Suite have demonstrated potential for use in the interim. These frameworks require adaptation to be suitable for use in Low and Middle-Income Countries (LMIC) however. OBJECTIVE 1) Modify the Enlight Suite, an mHealth app quality assessment methodology, to improve its applicability internationally, and 2) to assess the preliminary validity and reliability of this modified tool in practice. METHODS A two-round Delphi study involving 7 mHealth experts with varied backgrounds in medicine, health and technology was conducted to modify and adapt the Enlight Suite for international use as well as to improve its content validity. The Modified Enlight suite (MES) was then used by 800 healthcare professionals and healthcare students to assess a COVID-19 tracker app in an online survey form. The reliability of the MES was assessed using the Cronbach alpha while the construct validity was evaluated using the confirmatory factor analysis. RESULTS The final version of the MES has 7 sections with 32 evaluating items. Of these items, 5 were novel and based on consensus for inclusion by Delphi panel members. The MES has a satisfactory reliability with an internal consistency Cronbach alpha score of 0.925. The sub-scales also demonstrate acceptable internal consistency. Similarly, the confirmatory factor analysis demonstrates a positive and significant factor loading for all of the 32 items in the MES with a modestly acceptable model fit: thus, indicating the construct validity of the MES. CONCLUSIONS Despite increasing use, access, and reliance on mHealth apps internationally, previous studies have failed to identify a quality assessment methodology which included factors known to hinder the use and uptake of apps in LMICs. This study indicates both the validity and initial reliability of the MES for assessing the quality of mHealth apps internationally. Further reliability assessments are required in LMICs to extrapolate these findings to seek its true potential.