This paper, published in the Open Access Library Journal, investigates why users stick with (or abandon) mobile health (mHealth) applications. By merging two psychological frameworks—the Theory of Consumption Value (TCV) and the Expectation Confirmation Model (ECM)—the authors provide a roadmap for developers to move beyond the "first download" and build long-term user loyalty.
TL;DR
While mHealth apps are popular, nearly half of users abandon them shortly after download. This study reveals that long-term retention is driven by functional and emotional value rather than novelty, with "habit" and "satisfaction" acting as the ultimate anchors for continued use.
What Problem Does This Paper Address?
The "mHealth paradox" is a significant hurdle in digital medicine: adoption is high, but persistence is low. Statistics cited in the paper suggest that roughly 45.7% of users who download mHealth apps eventually stop using them.
Most existing research focuses on initial adoption (why people download the app). This paper addresses the gap in continuance intention (why people keep using it). To do this, the authors integrate two theories:
- Expectation Confirmation Theory (ECT): Focuses on how a user’s post-purchase satisfaction is determined by whether the product lived up to their initial expectations.
- Theory of Consumption Value (TCV): Breaks down "value" into five dimensions: functional, social, emotional, epistemic (novelty), and conditional (context-specific).
Methods
The researchers conducted a cross-sectional survey in China between July and August 2022. They used a self-designed questionnaire based on established scales, measured on a 7-point Likert scale.
To analyze the complex relationships between these psychological variables, they employed Partial Least Squares Structural Equation Modeling (PLS-SEM) using SmartPLS 3.0. This approach allowed them to test 10 distinct hypotheses simultaneously, looking at both direct effects and mediating paths (e.g., how "Perceived Value" leads to "Satisfaction," which then leads to "Continuance Intention").
Datasets
The study relied on a primary dataset of 317 valid survey responses collected from major social and medical platforms in China.
- Demographics: The sample was slightly more female (53.6%) and skewed young, with nearly 30% of respondents in the 19-25 age group.
- Education: Over 57% held a bachelor’s degree.
- Usage: The most common services used were online consultations (39.4%) and appointment registration (34.4%).
Key Insights and Ideas
- Utility Over Novelty: Perhaps the most striking finding was that Epistemic Value (novelty/curiosity) had no significant correlation with Perceived Value. In the context of health, users don't care if an app is "new and shiny"; they care if it works. Once the initial curiosity fades, only utility remains.
- The Power of Habit: Habit was found to be a direct and powerful predictor of continued use. The authors argue that for mHealth to succeed, it must transition from a "choice" to a "spontaneous behavior" integrated into daily routines.
- Value is Multi-Dimensional: Functional value (efficiency), social value (professional/peer acceptance), emotional value (feeling relaxed/pleased), and conditional value (convenience in specific situations like avoiding hospital queues) all significantly boosted the user's overall perception of the app's worth.
- Satisfaction as the Gatekeeper: Satisfaction acts as a critical mediator. If an app meets expectations (Confirmation) and provides value, it creates satisfaction, which is the strongest driver of the intent to keep using the service.
Key Results
The integrated model was highly effective, explaining 73% of the variance in Continuance Intention (R² = 0.730).
- Supported Hypotheses: 9 out of 10 hypotheses were confirmed.
- Strongest Drivers: Conditional value had the strongest impact on Perceived Value ($\beta$ = 0.378), while Satisfaction had the strongest direct impact on Continuance Intention ($\beta$ = 0.385).
- Perceived Value: The antecedents (functional, social, emotional, and conditional values) explained 78.2% of the variance in Perceived Value.
Limitations and Future Work
- The "Risk" Factor: The authors acknowledge that they did not include "Perceived Risk" (privacy concerns, data security) in their model. In healthcare, where data is sensitive, this is a major variable that could obstruct continued use.
- Demographic Bias: The study was limited to a Chinese population, largely urban and young. Results might differ in regions with different healthcare infrastructures or among older populations with lower digital literacy.
- Cross-Sectional Design: Because the data was collected at a single point in time, it captures a snapshot of intention rather than tracking actual behavior over months or years. Future longitudinal studies could provide deeper insights into how these values shift over the user lifecycle.
Summary updated