This chapter presents a new rational-objective (R-O) model of e-health use that accounts for effects of facilitating conditions as well as patients’ behavioral intention. An online questionnaire measured patients’ behavioral intention to use a new e-health application as well as proxy measures of facilitating conditions that assess prior use of and structural need for health services. A second questionnaire administered three months later collected patients’ self-reported use of e-health during the intervening period. The new model increased predictions of patients’ e-health use (measured in R2) by more than 300% over predictions based upon behavioral intention alone, and all measured factors contributed significantly to prediction of use during the three-month assessment period.
Wilson and Lankton (2004) studied factors that contribute to initial acceptance of e-health among new registrants to a prototype e-health application. That study found patients’ behavioral intention (BI) toward e-health use is predicted well by three prominent models of IT acceptance: the technology acceptance model (TAM) (Davis, Bagozzi, & Warshaw, 1989), the motivational model (Davis, Bagozzi, & Warshaw, 1992), and the integrated model (Venkatesh, Speier, & Morris, 2002). All are examples of rational models (Ajzen, 2002; Kim & Malhotra, 2005), so named because predictions are based upon individuals’ beliefs regarding such factors as ease of use and usefulness of the IT. Within these models, effects of beliefs upon IT use behaviors are theorized to be fully mediated by BI that individuals form through rational processes. Wilson and Lankton (2004) also report that belief factors in the models are significantly predicted by three patient characteristics that are developed prior to use of e-health: satisfaction with the provider, information-seeking preference, and Internet dependence. This finding is important, as it implies that patients’ BI toward e-health use can be predicted early in design stages of application development.
Rational models of behavior have performed well in predicting individual behaviors across a wide range of research domains. In the preponderance of published studies, a positive association is reported between BI and behavior (see reviews by Ouellette & Wood, 1998; Sheppard, Hartwick, & Warshaw, 1988). Based on this substantial literature most IT acceptance studies do not assess actual use, under the assumption that IT use is normatively predicted by BI (Lee, Kozar, & Larsen, 2003). Indeed, a recent review of 277 published IT acceptance studies conducted by one of the authors finds just 13 that evaluate effects of BI on self-reported IT use. Results from these 13 studies are summarized in Table 1.Table 1.
Review of associations between BI and self-reported IT use
|Reference||IT Type||Variance in IT Use Explained by BI|
Davis et al. (1989)
Dishaw & Strong (1999)
||Software maintenance tool||36%, including direct effect of perceived usefulness|
Hartwick & Barki (1994)
||Business IS application||35-74%|
Horton, Buck, Waterson, & Clegg (2001)
||Short message services||15%|
Limayem & Hirt (2003)
||Communication application||47%, including direct effects of habit and facilitating conditions|
Moon & Kim (2001)
||World wide Web||38%|
Morris & Dillon (1997)
||Netscape Web browser||19%|
Shih & Fang (2004)
Stoel & Lee (2003)
Suh & Han (2002, 2003)||Internet banking||3%|