Voice dictation to improve the inpatient nurse workflow


Project Overview

Nuance had previously ran a study with a previous research group from Bentley University from this course that focused on the physician/patient experience and interaction. They now wanted to look at nurses and how they collected data. They had some basic knowledge on nurses and their use of paper and flowsheets informed by preliminary research but they wanted to examine their current assumptions by seeing what was happening in the field.

The major milestones we hit included goal setting and alignment, nurse interviews, coding and insight generation, as well as a workshop we ran with stakeholders.

Challenge

Working with our stakeholders at Nuance, our team outlined 4 goals for the project that we would try to attain with our research:

  1. Understand nursing workflow for patient data collection
  2. Capture nurses' goals and motivations
  3. Identify challenges to those goals and motivations
  4. Understand which nurse segments may be adopters of voice technology

Role and Team

5 researchers from Bentley University, including myself. Our team partnered with Nuance Communications, a healthcare technology company. From their team, we worked directly with the Principal UX designer and Senior UX researcher. I was responsible for running interviews, qualitative coding, thematic analysis, and developing several sections of the report.

Methods

We used various methodologies including semi-structured remote interviews, questionnaires, and a round-robin ideation session. We conducted semi-structured interviews with ten nurses working in high-acuity inpatient settings, to learn more about nurses' motivations, patient data collection workflows, and associated challenges.

Due to constraints due to the COVID-19 pandemic, the logistic hurdles of direct observation in hospitals, and time, our team agreed that semi-structured remote interviews would be an appropriate method for this research.

Multiple rounds of thematic analysis led to several high-priority insights:

  1. When inpatient nurses collect critical details like patient vitals, they typically document them on paper. These papers are commonly referred to as “brains,” and their usage is deeply ingrained in nurse workflows.
  2. Busy days with many interruptions are the norm. While nurses would understandably prefer to find larger chunks of uninterrupted time to document patient data, this isn't always possible.
  3. The primary motivator to work in this field is patient care and nurses feel that documentation can be excessive and/or irrelevant. When this happens, it becomes a hindrance to providing quality care.
  4. Nurses use telecommunication devices to collaborate with care teams via voice and text messaging. However, these devices were not integrated with EHRs, leading to duplicative efforts and wasted time.
  5. Doctors and inpatient nurses have different patient data collection needs. Doctors take longer narrative-like notes, whereas nurses typically capture short and discrete pieces of information.

After understanding current data collection workflows through the interviews, our goal was to explore whether these workflows might be improved through voice technology. We developed three artifacts to aid this effort:

  1. Spectrum map of nurse participants: This was created to better identify correlations among nine participant demographic and behavioral categories. The map attempted to pinpoint whether anyone in our participant group might be an adopter of voice technology.
  2. Experience map: This consolidates the most salient interview data into a detailed journey map format. It showcases important quotes, events, motivations, and challenges throughout a nurse’s shift. It additionally gathers attitudes about the use of documenting patient data through voice dictation technology.
  3. Ideation session: Two interview insights that were conducive to technological solutions were selected and reframed as “How might we?” questions. Performed in Miro, these served as the foundation for a brainstorming activity to generate potential solutions to challenges nurses face while collecting patient data.

Impact and Opportunities

Our research showed that despite the time-consuming demands of patient documentation and the use of telecommunication devices, a majority of nurses did not favor the idea of using voice dictation when it came to documenting patient data. However, voice technology is more than just voice dictation, and our final recommendations on the use of voice technology for nurses in inpatient settings are as follows:

  1. Explore ways to replace paper as a stopgap solution for quickly adding highly structured information to an EHR system.
  2. Investigate a potential correlation surfaced through the spectrum map: Nurses who work at hospitals which require structured, templated note-taking also reported that documentation has a “very high impact” on their work-life balance.
  3. Research the potential intersection of voice dictation technology with communication devices already used in these settings, if not for charting, then at least for note-taking.
  4. Investigate clinics that don't allow paper into the patient's room due to COVID-19. If this practice is enforced in more inpatient settings than what was reported, it could be an opportunity for voice-enabled note-taking and/or charting.

Reflection

  • COVID-19 resulted in all interviews being conducted remotely, which limited our ability to observe, film, or photograph workflows in situ. This caused us to think deeper about ways we would collect and interpret the data from the study.
  • Having defined roles on the research team helps with organization, planning, and delineation of work.
  • Even though our study does not detect enough meaningful signals that point toward participants desiring this type of technology, we presently feel it is worth exploring the narrower usage of using dictation for note-taking purposes.

← Return to Selected Work

→ Next Project