When security becomes a barrier: Digital caregivers and rating hijacking in a national mHealth application
International Journal of Medical Informatics, cilt.224, sa.106790, ss.1-9, 2027 (SSCI, Scopus)
- Yayın Türü: Makale / Tam Makale
- Cilt numarası: 224 Sayı: 106790
- Basım Tarihi: 2027
- Doi Numarası: 10.1016/j.ijmedinf.2026.106790
- Dergi Adı: International Journal of Medical Informatics
- Derginin Tarandığı İndeksler: Scopus, Social Sciences Citation Index (SSCI)
- Sayfa Sayıları: ss.1-9
- Isparta Uygulamalı Bilimler Üniversitesi Adresli: Evet
Özet
Background
National mobile health (mHealth) applications are updated for security and usability, yet little is known about how updates affect the users who depend on them; where such platforms are the sole booking channel, access failures are immediate.
Objective
To evaluate how mandatory e-Government authentication and an interface redesign in Türkiye’s compulsory national appointment application (MHRS) affected user sentiment, rating behaviour and institutional communication, and whether digital caregivers were disproportionately affected.
Methods
All 2,943 Google Play reviews retained for 1 January–1 April 2026 were analysed in an explanatory sequential mixed-methods design: a quantitative phase (researcher-reviewed sentiment classification, change-point and interrupted time series analyses, logistic regression), then qualitative thematic analysis. A coverage audit found sub-five-star ratings retrieved only within two windows; rating analyses use those windows, longitudinal analyses the five-star stream observed throughout.
Results
Among five-star reviews, negative text rose from 2.7 % at baseline to 44–52 % weekly from 5 February, the day the authentication build first appeared, two weeks before the announced window; the March redesign showed no sustained effect. Reviews disclosing caregiving were negative about three times as often (period-adjusted OR 3.05, 95 % CI 1.87–4.96), an association entangled with review length. Qualitative coding verified 228 cases of “rating hijacking” — five stars attached to complaints to defeat feed sorting. Replies covered 80.0 % of reviews but were 96.9 % templated; after the redesign only 18.6 % of reviews were answered within 48 h versus 69–76 % earlier.
Conclusions
The access shock coincided with the rollout build, not the announced date. Updates deployed without proxy access, staged rollouts and context-aware crisis communication risk converting security into exclusion. Evidence comes from a self-selected review stream with incomplete rating coverage and reported rather than observed access failures; within those limits, the study offers, to our knowledge, the first verified account of rating hijacking in health.