Table 6.
Overview of the identified technological barriers and facilitators.
| Topic | ||
|---|---|---|
| Barriers | Facilitators | |
| Interoperability and integration (n = 17 studies) | Inadequate data connectivity between IT-systems (including different versions of one system) (45, 47, 50, 53, 55, 58, 61, 66, 67, 69, 70) | Develop health technologies that seamlessly integrate with existing workflows and generate interpretable data for clinicians (40, 58, 67) |
| Lack of integration between health technology and electronic medical records (50, 53, 64, 68) | Establish an eHealth infrastructure rather than standalone health technologies (60, 67, 68) | |
| Challenges integrating health technology with healthcare professional work practices (45, 53, 67) | Create regulatory environments that encourage integration across data sources without stifling innovation (40) | |
| Complexities in ensuring that health technology interoperability complies with legal system and protection legislation (67, 68) | Incorporate a connection between electronic patient files and health technologies (67) | |
| Non-transferable data across countries (70) | Enable seamless information exchange among healthcare providers within and between healthcare facilities (45) | |
| Difficulty achieving interoperability within complex healthcare organizations (60) | Align and link data from various disparate sources of origins (59) | |
| Utilize the potential of AI to automate data capture, distribution and communication (45) | ||
| Data reliability (n = 11 studies) | Discrepancies in how clinicians record or interpret data (52, 56, 59, 61, 70) | Ensure honest, accurate, and conscientious data entry (39, 45) |
| Bias in the (trained) dataset (45, 47, 62, 63, 70) | Utilize quality assessment tools and adhere to quality standards (70) | |
| Possibility of data loss or delay (52) | Improve data reporting by using standardized reporting guidelines (70) | |
| Data deletion complexities in AI contexts; algorithms do not forget like humans (63) | Prioritize the reliability of data communications (52) | |
| AI's limited ability to differentiate causation from correlation (63) | Managing missing and unstructured data (70) | |
| Health technology's narrow focus might be unsuitable for defining the total health status of a patient (51) | ||
| Uncertainty about the appropriate incorporation of patient-initiated digital health data into clinical decision-making (62) | ||
| Malfunction and errors (n = 7 studies) | External factors may impact the health technology's performance (62) | Offer support and assistance from IT-staff (36, 62) |
| Risks associated with introducing new interfaces or features that could break application functionality (40) | Ensure patient awareness for potential errors, prevention measures, and response/reporting procedures (62) | |
| Software errors (48) | Prohibit the addition of hardware or modification of systems software (36) | |
| Data acquisition might adversely affect performance (39) | Routinely update software and systems (47) | |
| Implement a system quality control process (45) | ||
| Accessibility (n = 4 studies | Health technologies inundated with excessive unsolicited data can overwhelm (clinical) users (40) | Ensure health technology interfaces are accessible for users (36) |
| Offer user manuals and technical support services (45, 64) | ||
| Minimize additional activities, time and user workload associated with the use of the health technology (36) | ||
| Schedule data extraction from health technologies outside peak office hours (36) | ||
| Automate data extraction from patient health records (36) | ||
| Ensure data export functions within the organization's local network (36) | ||
| Periodically present summarized data routine management channels (40) | ||
| Layout of eHealth (n = 2 studies) | Lack of clarity of the language usage (43) | Supply data and information flexibly, catering to individual contexts and roles (40) |
| Adopt a user-centered design approach with close stakeholder collaboration (40) | ||