Investigating health information systems-induced errors

Evaluating health information systems-related errors using the human, organization, process, technology-fit (HOPT-fit) framework

Complex socio-technical health information systems (HIS) issues can create new error risks. Therefore, we evaluated the management of HIS-related errors using the proposed human, organization, process, and technology-fit framework to identify the lessons learned. Qualitative case study methodology through observation, interview, and document analysis was conducted at a 1000-bed Japanese specialist teaching hospital. Effective management of HIS-related errors was attributable to many socio-technical factors including continuous improvement, safety culture, strong management and leadership, effective communication, preventive and corrective mechanisms, an incident reporting system, and closed feedback loops. Enablers of medication errors include system sophistication and process factors like workarounds, variance, clinical workload, slips and mistakes, and miscommunication. The case management effectiveness in handling the HIS-related errors can guide other clinical settings. The potential of HIS to minimize errors can be achieved through continual, systematic, and structured evaluation. The case study validated the applicability of the proposed evaluation framework that can be applied flexibly according to study contexts to inform HIS stakeholders in decision-making. The comprehensive and specific measures of the proposed framework and approach can be a useful guide for evaluating complex HIS-related errors. Leaner and fitter socio-technical components of HIS can yield safer system use.

Best Practices for Technology in Clinical Social Work and Mental Health Professions to Promote Well-being and Prevent Fatigue

The shift to communication technologies during the pandemic has had positive and negative effects on clinical social worker practice. Best practices are identified for clinical social workers to maintain emotional well-being, prevent fatigue, and avoid burnout when using technology. A scoping review from 2000 to 21 of 15 databases focused on communication technologies for mental health care within four areas: (1) behavioral, cognitive, emotional, and physical impact; (2) individual, clinic, hospital, and system/organizational levels; (3) well-being, burnout, and stress; and (4) clinician technology perceptions. Out of 4795 potential literature references, full text review of 201 papers revealed 37 were related to technology impact on engagement, therapeutic alliance, fatigue and well-being. Studies assessed behavioral (67.5%), emotional (43.2%), cognitive (57.8%), and physical (10.8%) impact at the individual (78.4%), clinic (54.1%), hospital (37.8%) and system/organizational (45.9%) levels. Participants were clinicians, social workers, psychologists, and other providers. Clinicians can build a therapeutic alliance via video, but this requires additional skill, effort, and monitoring. Use of video and electronic health records were associated with clinician physical and emotional problems due to barriers, effort, cognitive demands, and additional workflow steps. Studies also found high user ratings on data quality, accuracy, and processing, but low satisfaction with clerical tasks, effort required and interruptions. Studies have overlooked the impact of justice, equity, diversity and inclusion related to technology, fatigue and well-being, for the populations served and the clinicians providing care. Clinical social workers and health care systems must evaluate the impact of technology in order to support well-being and prevent workload burden, fatigue, and burnout. Multi-level evaluation and clinical, human factor, training/professional development and administrative best practices are suggested.

Health technology, quality and safety in a learning health system

Health technology quality and safety is an important issue for health informatics (i.e. digital health) professionals. Health technologies have been used to (1) collect data that can be analyzed to improve the quality and safety of healthcare activities and (2) re-engineer and/or automate error-prone processes. Health technologies are also able to introduce new types of errors (i.e. technology-induced errors) and have been implicated in propagating errors across digital health ecosystems. To develop a learning health system, health technologies need to be considered in terms of how they can improve the quality and safety of health activities traditionally carried out by humans (patients and health professionals) and also how the technology's quality and safety can be improved. This article outlines how this can be done by integrating evidence from health informatics research into practice using a learning health systems approach.

Unintended consequences of pharmacy information systems: A case study

BACKGROUND

Pharmacy information systems (PhIS) can cause medication errors that pharmacists may overlook due to their increased workload and lack of understanding of maintaining information quality. This study seeks to identify factors influencing unintended consequences of PhIS and how they affect the information quality, which can pose a risk to patient safety.

MATERIALS AND METHODS

This qualitative, explanatory case study evaluated PhIS in ambulatory pharmacies in a hospital and a clinic. Data were collected through observations, interviews, and document analysis. We applied the socio-technical interactive analysis (ISTA) framework to investigate the socio-technical interactions of pharmacy information systems that lead to unintended consequences. We then adopted the human-organization-process-technology-fit (HOPT-fit) framework to identify their contributing and dominant factors, misfits, and mitigation measures.

RESULTS

We identified 28 unintended consequences of PhIS, their key contributing factors, and their interrelations with the systems. The primary causes of unintended consequences include system rigidity and complexity, unclear knowledge, understanding, skills, and purpose of using the system, use of hybrid paper and electronic documentation, unclear and confusing transitions, additions and duplication of tasks and roles in the workflow, and time pressure, causing cognitive overload and workarounds. Recommended mitigating mechanisms include human factor principles in system design, data quality improvement for PhIS in terms of effective use of workspace, training, PhIS master data management, and communication by standardizing workarounds.

CONCLUSION

Threats to information quality emerge in PhIS because of its poor design, a failure to coordinate its functions and clinical tasks, and pharmacists' lack of understanding of the system use. Therefore, safe system design, fostering awareness in maintaining the information quality of PhIS and cultivating its safe use in organizations is essential to ensure patient safety. The proposed evaluation approach facilitates the evaluator to identify complex socio-technical interactions and unintended consequences factors, impact, and mitigation mechanisms.

Efficiency and Safety of Electronic Health Records in Switzerland-A Comparative Analysis of 2 Commercial Systems in Hospitals

OBJECTIVES

Differences in efficiency and safety between 2 electronic health record (systems A and B) in Swiss hospitals were investigated.

METHODS

In a scenario-based usability test under experimental conditions, a total of 100 physicians at 4 hospitals were asked to complete typical routine tasks, like medication or imaging orders. Differences in number of mouse clicks and time-on-task as indicators of efficiency and error type, error count, and rate as indicators of patient safety between hospital sites were analyzed. Time-on-task and clicks were correlated with error count.

RESULTS

There were differences in efficiency and safety between hospitals. Overall, physicians working with system B required less clicks (A: 511, B: 442, P = 0.001) and time (A: 2055 seconds, B: 1713 seconds, P = 0.055) and made fewer errors (A: 40%, B: 27%, P < 0.001). No participant completed all tasks correctly. The most frequent error in medication and radiology ordering was a wrong dose and a wrong level, respectively. Time errors were particularly prevalent in laboratory orders. Higher error counts coincided with longer time-on-task (r = 0.50, P < 0.001) and more clicks (r = 0.47, P < 0.001).

CONCLUSIONS

The variations in clicks, time, and errors are likely due to naive functionality and design of the systems and differences in their implementation. The high error rates coincide with inefficiency and jeopardize patient safety and produce economic costs and burden on physicians. The results raise usability concerns with potential for severe patient harm. A deeper understanding of differences as well as regulative guidelines and policy making are needed.