Patterns of eHealth Website User Engagement Based on Cross-site Clickstream Data: Correlational Study (Preprint)

Abstract

BACKGROUND User engagement is the key performance variable for e-health websites. However, most of the existing studies on user engagement either focus on a single website or depend on survey data. Thus far, we still lack an overview of user engagement on multiple e-health websites that is derived from objective data. Therefore, it is relevant to provide a holistic view of user engagement on multiple e-health websites based on cross-site clickstream data. OBJECTIVE The main objectives of this paper were to (1) describe the patterns of user engagement on e-health websites and (2) investigate how the platform, the channel, gender and income will influence user engagement on e-health websites. METHODS The data used in this study are the clickstream data of 1095 mobile users from a large telecom company in Shanghai, China. The observation period covers 8 months (Jan. 2017 – Aug. 2017). Descriptive statistics, t-tests, and ANOVA were used in the data analysis. RESULTS (1) The medical category accounts for most of the market share of e-health website visits (72.5%), followed by the lifestyle category (25.4%). The e-pharmacy category has the smallest market share, accounting for only 2.14% of the total visits. (2) E-health websites are characterized by very low visit penetration but relatively high user penetration. (3) The distribution of the engagement intensity follows a power law distribution. (4) Visits to e-health websites are highly concentrated. (5) User engagement is generally high on weekdays but low on weekends. Furthermore, user engagement increases gradually from morning to noon. After noon, user engagement declines until it reaches its lowest level at midnight. (6) Lifestyle websites, followed by medical websites, have the highest customer loyalty. E-pharmacy websites have the lowest customer loyalty. (7) Popular e-health websites such as medical websites can effectively provide referral traffic for lifestyle websites and e-pharmacy websites. However, the opposite is not true. (8) Android users are more engaged on e-health websites than are iOS users. (9) The engagement volume of app users is 4.85 times that of browser users, and the engagement intensity of app users is 4.22 times that of browser users. (10) Male users have higher engagement intensity than female users. (11) Income negatively moderates the influence of the platform (Android vs. iOS) on user engagement. Low-income Android users are the users who are the most engaged on e-health websites. Conversely, low-income iOS users are the users who are the least engaged on e-health websites. CONCLUSIONS Clickstream data provide a new way to derive an overview of user engagement patterns on e-health websites and to investigate the influence of factors (e.g., the platform, the channel, gender and income) on engagement behavior. Compared with self-reported data from a questionnaire, cross-site clickstream data are more objective, accurate and appropriate for pattern discovery. Many of the user engagement patterns or findings regarding the influential factors revealed by cross-site clickstream data herein have not been previously reported.