This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR Medical Informatics, is properly cited. The complete bibliographic information, a link to the original publication on http://medinform.jmir.org/, as well as this copyright and license information must be included.
Nurses are the largest group of health information technology (HIT) users. As such, nurses’ adaptations are critical for HIT implementation success. However, longitudinal approaches to understanding nurses’ perceptions of HIT remain underexplored. Previous studies of nurses’ perceptions demonstrate that the progress and timing for acceptance of and adaptation to HIT varies.
This study aimed to explore nurses’ experience regarding implementation of HIT over time.
A phenomenological approach was used for this longitudinal qualitative study to explore nurses’ perceptions of HIT implementation over time, focusing on three time points (rounds) at 3, 9, and 18 months after implementation of electronic health records and bar code medication administration. The purposive sample was comprised of clinical nurses who worked on a medical-surgical unit in an academic center.
Major findings were categorized into 7 main themes with 54 subthemes. Nurses reported personal-level and organizational-level factors that facilitated HIT adaptation. We also generated network graphs to illustrate the occurrence of themes. Thematic interconnectivity differed due to nurses’ concerns and satisfaction at different time points. Equipment and workflow were the most frequent themes across all three rounds. Nurses were the most dissatisfied approximately 9 months after HIT implementation. Eighteen months after HIT implementation, nurses’ perceptions appeared more balanced.
It is recommended that organizations invest in equipment (ie, wireless barcode scanners), refine policies to reflect nursing practice, and improve systems to focus on patient safety. Future research is necessary to confirm patterns of nurses’ adaptation to HIT in other samples.
Health information technology (HIT or Health IT) is a broad concept that includes a variety of technologies, including computer equipment, system software, and infrastructure that records, stores, protects, and retrieves clinical, administrative, or financial information [
A myriad of individual and organizational factors influence HIT, contributing to its complexity and multi-dimensional nature [
Longitudinal perceptions of EHRs have also been studied in nurses. Intensive care unit (ICU) nurses completed two cross-sectional survey questionnaires at 3 months and 12 months after EHR implementation and reported greater acceptance of the EHR at 12 months compared to 3 months [
In our prior study, we found that nurses’ expectations regarding HIT can be stratified from personal-level (human-computer interaction) to organizational-level (quality of care) [
We investigated HIT adaptation, which is, “a process of modifying existing conditions in an effort to achieve alignment” [
We conducted a phenomenological qualitative study at a large Midwestern academic medical center. The medical center implemented a customized commercialized EHR system (EPIC platform [
We used purposive sampling to recruit staff nurses who worked on a medical-surgical unit that used EHRs and BCMA and had a minimum of two years of working experience in the organization. We approached participants either face-to-face or via email and hosted private face-to-face interviews in a location away from the clinical area to ensure privacy and avoid disruption.
We used a semi-structured interview guide with additional probes (
Three researchers (IZ, JGS, and PY) read the transcripts independently and located relevant statements in the transcripts that expressed units of meaning. The researchers generated common themes by synthesizing the meaning units. Themes reflected a general description of the nurse participants’ experience with EHRs and BCMA. The structure for the first five themes was derived from previous work that delineated confirmed nurses’ expectations ranging from personal-level to organizational-level [
We assessed and established our intercoder reliability using Cohen’s Kappa. Cohen’s Kappa has been commonly used to assess intercoder reliability, with recommendations greater than 0.7 for semi-structured interviews, especially with multiple complex codes [
Nineteen nurses participated in the study, with some nurses participating in multiple rounds. We conducted a total of 30 interviews: 9 from R1, 11 from R2, and 10 from R3. We were unable to interview all nurses across all 3 rounds because some nurses were unavailable (ie, schedule conflicts, transferred to new positions, declined participation). Among the 9 nurses that participated in R1, 7 participated in R2, and 3 participated in R3. Eleven nurses participated in only one interview, and 8 nurses participated in more than one interview. Among the 19 nurses, 17 were female. The age of the nurses ranged from 22 to 52 years old, with 3 to 25 years of working experience, and 58% (11/19) worked the day shift. Most nurses (15/19, 79%) were Bachelor of Science in Nursing-prepared and all worked as staff nurses. All nurses owned home computers and 79% (15/19) owned smartphones. Across all rounds, nurses rated themselves an average of 4 out of 5 in computer competency, with 1 meaning
We assembled themes into a table format to review nurse perceptions over time (
The E1 subthemes involved
Nurses viewed documentation as time-consuming and arduous, yet thorough. Nurses valued feedback on performance from visual indicators (green and red dots) for complete or incomplete documentation. In R1 and R2, nurses thought documentation was inefficient and not streamlined because it contained elements irrelevant to their population, required too much scrolling, appeared chaotic, and took longer than paper charting. However, by R3 some nurses expressed that documentation was streamlined and efficient since the IT department had updated the EHRs to be more compact. Nonetheless, documenting rare events caused nurses stress and confusion. For instance, blood administration involved scanning multiple barcodes in a strange pattern and emergency documentation had complicated screen layouts.
Since all nurses experienced glitches with EHRs, they relied on nursing unit collaboration for assistance. Collaboration impacted patient safety because nurses appreciated the ability to view all records when administering medicine to other nurses’ patients. Teamwork was also associated with improved nurses’ satisfaction. Due to strong teamwork, nurses relied on one another more than IT staff for resolving system concerns.
E4 themes concerned communication between departments and disciplines. Interprofessional notes helped nurses understand the care plan promoting integrated and better care. Nurses reviewed patient transfer information before arrival. Unexpectedly, some departments stopped telephone handoffs, leading to potential missed information (ie, last pain medication). Nurses identified unequal standards between departments, such as not scanning all medications or omitting parts of admission documentation. Documentation standards varied and some prescribers did not enter orders immediately after patient assessment. Unequal standards existed in the context of shifting responsibilities, such as doctors asking nurses to input their orders. This factor frustrated nurses, and they proposed standardized documentation classes and policies to improve interprofessional communication.
E5 themes related to quality of care and nurse satisfaction. Although BCMA reduced some medication errors, the potential for error remained because BCMA did not verify multi-dose medication containers (like insulin) and prescribers could still input orders incorrectly. Additional potential for errors during physical assessment documentation could occur due to the repetitive nature of electronic documentation (mouse clicking), leading to distraction and loss of attention. Patient-nurse interactions were also altered because nurses had to position their backs to patients to document on WOWs. In addition, occasionally WOWs logged nurses off or malfunctioned, leading to pain medication delays that negatively impacted care. During R1 and R2, nurses were so frustrated with learning the system that they did not appreciate potential improvements in patient care quality. However, by R3, after nurses adapted to EHRs, they frequently mentioned better quality of patient care through access to patient history, notes from all disciplines, task/documentation reminders, and improved patient safety. Perceptions of care quality improved in R3 compared to R1 and R2.
When the EHR system was first implemented, some nurses felt scared or intimidated, although eventually it met expectations or appeared that it would in the future. The most common expectation was reduced documentation that would allow for more time with patients, which did not happen, leading to disappointment. Nurses’ dissatisfaction and satisfaction were mentioned with similar frequencies in R1 and R3, although in R2 nurses were more dissatisfied than satisfied. The greatest sources of nurse dissatisfaction were equipment, system functionalities, inefficient documentation, and lengthy logins. Conversely, nurses were satisfied with BCMA error reduction, workflow simplification, patient protection, better care, and documentation thoroughness and reminders. By R3, after nurses adapted, some nurses expressed that the EHR system offered more time for higher-quality patient care.
Adaptation affected nurses’ acclimatization to new technology over time. Nurses discussed self-learning through personal motivation, practice, and long-term use. Self-learning occurred through the EHR playground, where nurses could explore the EHR layout. However, training was a major concern with rushed, fast, and overwhelming classes that were provided too far in advance of implementation. Nurses commented that training classes did not reflect nurses’ workflow; they only showed system design and navigation.
Organizational factors, such as policies, requirements, and decisions made by leadership, were frequently mentioned with nurses’ dissatisfaction. Clinicians expected leaders to explain rationale for HIT decisions that would impact clinical practice. For example, nurses were not aware of the rationale for policies not allowing a copy/paste function, or real-time documentation which expected nurses to chart assessments immediately after care. Nurses expected hospital leadership to be more aware of bedside nursing workflow and resolve issues quickly. Leadership added requirements or responsibilities but did not retire old/unnecessary requirements, which added to nursing work and complicated workflow. Nurses expected leadership to advocate for system features to improve nursing workflow, so that they could spend more time at the bedside.
Nurses’ suggestions for improvement grew with each round and were often related to the system and equipment. Some suggestions were incorporated by R3, such as an exact time stamp and wireless scanners. Nurses were thankful that nursing management filtered information regarding important EHR updates that affected nursing work. Nurses valued leadership’s feedback on their performance regarding percent of scanned medications, and advised leadership to be patient, remain supportive, provide resources, and answer questions.
We used Gephi to create a network (relationship) graph of themes for each round. Each quote contained one to four themes, while the edges/lines revealed theme cooccurrence within the same quote (
Nurses’ dissatisfaction was related to equipment and documentation.
There were trends in collaboration, communication, quality of care, and workflow.
The longitudinal approach led to discovery of two additional themes:
Due to the research approach, nearly all quotes could be related to
Visualization of theme and subtheme interrelationships.
This study explored the trajectory of change through a qualitative analysis of nurses’ experiences after EHR and BCMA implementation at three time points: 3, 9, and 18 months. We used the socio-technical model [
Nurses shared opinions regarding HIT equipment. Workstation preferences varied: some nurses liked working with portable workstations so they could store medications in locked drawers, while other nurses desired fixed workstations in patient rooms or portable tablets. Mobile devices may promote nurses’ ability to document at the bedside and point-of-care [
Nurses offered multiple system suggestions: they wanted illuminated new and abnormal laboratory results, parameters for holding medication (eg, blood pressure), and unacknowledged orders to be highlighted in red. Nurses suggested system recognition and display of insidious abnormal trends, such as increasing white blood cell counts or decreasing hemoglobin. Physical assessment customization, such as removal of irrelevant fields or adding fields/drop-down boxes, would improve nurses’ documentation. EHR navigation could be improved with a help sheet or search tab with a glossary or key words. A patient calendar to track scheduled tests would help prepare patients and improve workflow, communication, and possibly patient satisfaction.
Clinician involvement in HIT design is a potential strategy for successful adaptation [
Training and competency are sociotechnical factors that affect HIT adaptation [
In the socio-technical model for studying HIT, the clinical workflow involved with operating HIT systems must be consistent with internal policies and procedures [
Nurses voiced that a reduced patient load (better nurse staffing) was very helpful for adapting to the new system. During the first day of EHR implementation, medical-surgical nurses cared for only one or two patients. By R3, nurses returned to caring for four to five patients, but some nurses continued to struggle with completing documentation requirements. In the future, augmented clinical HIT (where nurse staffing decisions are based on patient volume, acuity levels, etc [
A 2013 integrative review found that strong leadership ensures that the team works toward successful HIT implementation [
Adaptation to HIT over time may be explained in part by the Gartner Hype Cycle [
Previous research indicates that clinician satisfaction with HIT is mixed, which may be related to clinical documentation practice, workload, and productivity [
Physicians and nurses may differ in perceptions of productivity after HIT implementation. While physicians may experience better productivity due to increased charges, improved work relative value units, and less time writing orders [
Generalizability of these findings is limited due to sampling nurses from one unit within an academic medical center. Self-selection bias may have occurred from voluntary participation. The timing of interview rounds was based upon interviewer availability rather than change theory. Although we conducted 30 interviews, some nurses participated over multiple time periods and the group comprised a small sample size (n=19). Despite the small sample size, recurrence of similar themes across multiple individuals established information saturation and data quality.
A longitudinal qualitative approach for studying HIT adaptation facilitated understanding of thematic relationships over time. Although thematic interconnectivity differed due to nurses’ concerns and satisfaction at different time points, some trends were noted. Nurses appeared the most dissatisfied in R2, but many sources of dissatisfaction may be rectified, such as new equipment, refined policies, and improved systems to focus on patient safety. Approximately 18 months after HIT implementation, nurses’ perceptions appeared more balanced, as indicated by more consistent thematic frequencies and weaker cooccurrences in the Gephi chart. Balanced thematic distribution and interconnectivity within Gephi charts may be a visual indicator of HIT adaptation progress. Future research is necessary to confirm if researchers can replicate these findings in other samples.
Semistructured interview guide.
Theme codebook and definitions.
Themes with quote examples.
bar code medication administration
theme
electronic health record
health information technology
intensive care unit
information technology
round (of data collection)
workstation on wheels
The authors would like to thank the computer science student Xiaonan Ji for providing the Gephi graph. We thank Esther Chipps, Laura Droll, and Kelly Torrence at The Ohio State University Wexner Medical Center for coordinating the recruitment, and all participating nurses for their time and effort.
None declared.