Consumers' continuance intention to use fitness and health apps: an integration of the expectation–confirmation model and investment model

Abstract

<jats:sec><jats:title content-type="abstract-subheading">Purpose</jats:title><jats:p>This study aimed to explore consumers' continuance intention to use a fitness and health app by applying two theoretical models: the expectation–confirmation model (ECM) and the investment model (IM).</jats:p></jats:sec><jats:sec><jats:title content-type="abstract-subheading">Design/methodology/approach</jats:title><jats:p>An online survey was administered to consumers who are currently using fitness and health apps (<jats:italic>N</jats:italic> = 342). Structural equation modeling (SEM) was conducted using Amos 22.0 software to examine the hypothesized relationships in the research model.</jats:p></jats:sec><jats:sec><jats:title content-type="abstract-subheading">Findings</jats:title><jats:p>Results revealed that users' continuance intention was significantly predicted by the two models. Within the ECM, the positive and significant relationships among variables were found. Moreover, users' satisfaction and investment size had positive impacts on their commitment, which, in turn, positively affected the intention. Also, confirmation of expectations had a positive impact on investment size.</jats:p></jats:sec><jats:sec><jats:title content-type="abstract-subheading">Originality/value</jats:title><jats:p>The integrated model helps better understand fitness and health app users' decision-making process from the perspective of relationship commitment and suggests practical implications for health and fitness app providers.</jats:p></jats:sec>

The Intention to Use Fitness and Physical Activity Apps: A Systematic Review

“…Analyzing the sample used in the different studies, there is a total of 16,025 subjects with an average sample of 843.42 subjects per study, with the Ndayizigamiye; Kante, and Shingwenyana study [54] having the smallest sample (n = 139) and Wei, Vinnikova, Lu and Xu study [55] having the largest sample with a total of 8840 subjects. Approximately a half of the studies (n = 8) used university students as a sample, followed by studies that considered users of sports applications (n = 4) and other studies took as their general population [54][55][56], a population of sports consumers [45], employees of a sports organization [57] and members of a fitness community [44,58]. Most studies had a higher proportion of females than males (n = 9), followed by studies that had parity in the sample (n = 5), four studies had a higher proportion of males while one study did not indicate the gender distribution of the sample [34].…”

“…Li, Liu, Ma and Zhang [46] sampled subjects over 25 years of age, while Huang and Ren [60] and Mohammadi and Isanejad [57] were at least 30 years old. Regarding the type of App evaluated, six studies evaluated the intention to use diet and fitness applications [30,44,51,52,55,59], another five studies evaluated sports information Apps [42,43,53,57,62], two studies measured the intention of fans to use the sports team app [1,45], health and fitness app [56,61], or fitness [58,60] and one study evaluated a social fitness-tracking app [63] and mHealth related to promote physical activity [54]. The most widely used theory for the design and use of the mobile sports app intent of use assessment instrument was TAM (n = 10).…”

“…The most widely used theory for the design and use of the mobile sports app intent of use assessment instrument was TAM (n = 10). Chen and Lin [44] and Chiu and Cho [61] used a variant of the TAM, Theory of Readiness and Acceptance Model (TRAM), three studies used UTAUT [54,55,62] and UTAUT2 [30,58,63], and an article with the Expectation-Confirmation Model (ECM) [56]. The most common method of analysis used was a content analysis by structural equations using the AMOS statistical package (n = 9) and the rest of the Partial Least Square studies.…”

“…The most common method of analysis used was a content analysis by structural equations using the AMOS statistical package (n = 9) and the rest of the Partial Least Square studies. Seven studies also performed a correlation analysis of the data [54][55][56][57][58][59]62], and eight studies performed descriptive analysis in addition to content analysis [1,42,51,[55][56][57][58]61].…”

“…Analyzing the quantitative data on the relationships between the most common variables associated with TAM (Table 3), the six studies that had a different theory such as UTAUT [54,55,62] or UTATUT2 [30,58,63] were excluded; however, the study that used ECM was included because there was a relationship between variables "perception usefulness" and "intention to use" [56]. The sample was very heterogeneous in terms of the results of the existing relationships and sample size in each study.…”