gms | German Medical Science

Background/research question: The electronic health record (EHR) promises more effective, safer, and faster treatment, but whether these benefits can be realized, largely depends on individuals actually using it. Models such as the privacy calculus (PC) try to predict people’s intention to use EHRs based on perceived risk and benefits of the technology. To increase the validity of these models, we suggest that (1) its technology focus should be complemented with a focus on the risks and benefits related to disease characteristics [1], [2] and that (2) both intention to use and actual behavior should be considered as outcome criteria. To do so, we investigated whether the influence of disease related stigma and the time course of disease influence the predictions of the PC model [3] and whether this model is a reliable predictor of actual behavior.

Methods: The study (N=241) consisted of two parts: a survey and a user study. In the first part, participants explored an EHR app and read a randomly assigned case vignette describing a medical record, which was varied between-subjects with respect to disease-related stigma (high vs. low) and time course (acute vs. chronic). They then answered a PC questionnaire. In the second part, participants were asked to interact with the medical record mentioned in Part 1 and decide whether they would upload it into the EHR (behavioral measure).

Results: The intention to use was not influenced by the disease-related stigma but by the time course. Chronic diseases increased the influence of perceived benefits, privacy concerns, and trust in the provider on intention to use the EHR when compared to acute diseases, while the influence of social norms was decreased. This impact was reversed when examining uploading behavior. The disease-related stigma and intention to use influenced uploading behavior, whereas time course did not.

Conclusion: Whereas intention to use is increased for chronic diseases despite an increase in privacy concerns, uploading behavior is reduced by disease-related stigma. That is, people are aware that using the EHR is beneficial for managing chronic diseases, but they are concerned about data safety when it comes to actually uploading stigmatized diseases. Conceptually, this suggests ways to extend the PC model to more reliably predict uploading behavior. Practically, EHR advocates may use these insights to improve public knowledge about data security and, ultimately, EHR uptake.

Competing interests: None declared