What Clinical Information Is Valuable to Doctors Using Mobile Electronic Medical Records and When?

Junetae Kim, Yura Lee, Sanghee Lim, Jeong Hoon Kim, Byungtae Lee, Jae‐Ho LeeJournal of Medical Internet Research 2017 · DOI

TL;DR

This study investigates which clinical information doctors find most valuable on mobile electronic medical records (m-EMRs) and when they access it. By analyzing usage logs from a large hospital, they found that patient investigation status and conditions are crucial during early morning rounds, suggesting m-EMRs should prioritize this information for efficient clinical decision-making.

What Problem Does This Paper Address?

In modern healthcare, doctors often need to access patient information while on the move, whether between wards, during patient consultations, or even outside of regular working hours. Mobile electronic medical records (m-EMRs) were developed to address this need for mobility and timely access to clinical data. However, despite the increasing adoption of m-EMRs, there's been a lack of clear understanding regarding what specific types of clinical information are most valuable to doctors when using these mobile tools, and when during their daily routines this information is most critical. This knowledge gap hinders the optimal design and implementation of m-EMR systems to truly enhance clinical decision-making and workflow efficiency.

Methods

The researchers employed a two-step empirical analysis using usage logs from an m-EMR system at a tertiary hospital in Seoul, South Korea, collected between March and December 2015.

  1. Usage Pattern Analysis: They first compared the hourly usage patterns of the m-EMR with the hospital information system (HIS) by analyzing HIS CPU utilization rates. This comparison helped identify periods when m-EMR usage was particularly high relative to HIS usage, suggesting specific times when m-EMRs offered unique value. They defined "peak usage times" for the m-EMR as continuous hourly intervals where normalized usage values exceeded 0.5.
  2. Clinical Information Value Assessment: To understand what information was valuable during these peak times, they used factor analysis on the m-EMR usage logs. This technique grouped various menu accesses into broader "representational clinical information" categories. Finally, they used a random intercept logistic regression model to determine the association between these identified clinical information factors and the peak m-EMR usage intervals, while controlling for variables like weekday/holiday usage, doctor's position, age, and gender.

Datasets

The study utilized real-world usage log data from an m-EMR system implemented in a tertiary hospital in Seoul, South Korea. This dataset comprised 524,929 usage logs from 653 doctors over a 10-month period (March to December 2015). Additionally, HIS CPU utilization data for one week in November 2016 was used for comparative analysis. No new benchmarks were introduced; the focus was on analyzing existing usage data.

Key Insights and Ideas

  • Morning Rounds are Peak m-EMR Time: The most striking finding is that m-EMR usage peaks early in the morning (6:00 AM to 10:00 AM), coinciding with ward rounds. This suggests doctors are using m-EMRs to quickly access critical patient information before or during their rounds, likely due to the limitations of accessing desktop PCs in these settings.
  • HIS vs. m-EMR Usage Schedules: The HIS is primarily used during standard business hours (9:00 AM to 5:00 PM), while m-EMRs see significant use in the early morning and again from early evening through midnight. This highlights the distinct roles and usage contexts of these systems.
  • Information Type Matters: Not all m-EMR usage is equal. The study identified that specific categories of information, particularly "investigation status" and "patient conditions," are strongly associated with these peak usage times. This implies these are the types of data doctors most urgently need when making decisions during rounds.
  • Location Independence is Key: The study reinforces the idea that m-EMRs provide crucial location independence, allowing doctors to access information anytime, anywhere, which is particularly beneficial when desktop access is impractical.

Key Results

  • Top Information Categories: The most frequently accessed menus during peak m-EMR usage times were the inpatient list, lab results, and investigation list, accounting for 60.1% of peak usage.
  • Valuable Information Factors: Factor analysis identified five key patterns of representational clinical information. Two of these, "investigation status" (F1) and "patient conditions" (F3), were found to be significantly positively associated with peak m-EMR usage intervals (P<.01).
  • Usage Patterns: m-EMR usage was highest among medical residents. The HIS was heavily used during regular business hours, while m-EMR usage peaked from 6:00 AM to 10:00 AM and again from 5:00 PM to midnight.
  • Control Variables: Weekday usage was positively associated with peak intervals. Doctors in general medical departments used the m-EMR more than those in surgical departments.

Limitations and Future Work

  • Single Hospital Study: The research was conducted at a single hospital, and each institution may have unique m-EMR systems and schedules. Future work could explore diverse environments.
  • Data Collection Period Mismatch: The m-EMR usage data and HIS CPU utilization data were collected over different periods. While the authors believe this difference is minor due to consistent HIS usage patterns, collecting data concurrently would strengthen the comparison.
  • Focus on Reading, Not Entry: The study only analyzed information read through m-EMRs, not information entered. Future research could investigate whether patient investigation status and conditions are frequently recorded via m-EMRs during ward rounds.
  • Personal/Organizational Factors: The study did not account for personal or organizational characteristics that can influence technology adoption. Incorporating these could enhance understanding.

The authors suggest that m-EMRs should be designed to prioritize accessibility and presentation of information related to patient investigation status and conditions, especially for morning ward rounds, to maximize their value in clinical decision-making. Further research could explore how information acquisition needs vary across different medical tasks and departments, and how representational clinical information can be optimized for various m-EMR designs.

Summary updated

Abstract

Access to information regarding patient investigation status and patient conditions is crucial for decision making during morning activities, including ward rounds. The m-EMRs allow doctors to maintain the continuity of their clinical information regardless of the time and location constraints. Thus, m-EMRs will best evolve in a manner that enhances the accessibility of clinical information helpful to the decision-making process under such constraints.
BibTeX
@article{kim2017what,
  author = {Junetae Kim and Yura Lee and Sanghee Lim and Jeong Hoon Kim and Byungtae Lee and Jae‐Ho Lee},
  title = {{What Clinical Information Is Valuable to Doctors Using Mobile Electronic Medical Records and When?}},
  journal = {Journal of Medical Internet Research},
  year = {2017},
  doi = {10.2196/jmir.8128},
}
Figure