LifeLines provide a
general visualization environment for personal histories. We
explored its use for medical patient records. A one screen overview
of the record using timelines provides direct access to the data.
Problems, hospitalization and medications can be represented
as horizontal lines, while icons represent discrete events such
as physician consultations (and progress notes) or tests. Line
color and thickness can illustrate relationships or significance.
Techniques are described to display large records. Rescaling
tools and filters allow users to focus on part of the information,
revealing more details.
Computerized medical records
pose tremendous problems to system developers. Infrastructure
and privacy issues need to be resolved before physicians can even
start using the records. Non-intrusive hardware is required for
physicians to do their work (i.e. interview patients) away from
their desk and cumbersome workstations. But all the efforts to
solve those problems will only succeed if appropriate attention
is also given to the user interface design [1][8]. Long lists
to scroll, clumsy search, endless menus and lengthy dialogs will
lead to user rejection. But techniques are being developed to
summarize, filter and present large amount of information, leading
us to believe that rapid access to needed data is possible with
careful design.
While more attention is
now put on developing standards for gathering medical records
we found that very little effort had been made to design appropriate
visualization and navigation techniques to present and explore
personal history records. An intuitive approach to visualizing
histories is to use graphical time series. The consistent, linear
time scale allows comparisons and relations between the quantities
displayed. Data can be graphed on the timeline to show time series
of quantitative data. Highly interactive interfaces turn the
display into a meaningfully structured menu with direct access
to the data needed to review a problem or conduct the diagnosis.
The use of overviews or
summaries has been well studied: the Summary Time-Oriented Record,
a pure text flowchart, was found helpful in an arthritis clinic
[10]. Another text summary record uses scaled values to indicate
the severity of a symptom and also notes the duration of the symptoms,
allowing the timing of clinical events to become apparent [11].
Intensive care summary visualization software includes notions
of filtering, severity scaling, and details on demand [2].
Tufte [9] describes timelines as a frequent and powerful form of graphic design and presents many examples. Timelines have been used for presenting specific cases in reports and studies [5][7]. A design using timelines for medical records was proposed by Powsner and Tufte [6] who developed a graphical summary using a table of individual plots of test results and treatment data. But little has been proposed to present interactive personal histories. Our work on the visualization of personal history started with a project with the Maryland Department of Juvenile Justice (DJJ). We developed LifeLines to represent youth records (including, cases, placements, worker assignments and reviews) [11]. LifeLines were designed to 1) present a youth record overview on a single screen, 2) provide direct access to all detailed information from the overview with one or two clicks of the mouse, 3) promote critical information or alerts to the overview level. An experiment in home automation showed that timelines could be quickly understood by novice users and used for data entry when scheduling devices in the home [4].
As a first step we used
a simple medical record found on the Internet and prepared a screen
mockup of the LifeLine overview of the record [Figure 1]. This
mockup was reviewed by several physicians and their feedback encouraged
us to explore further. We then worked with a cardiologist who
reviewed one of his patient record and designed the interface
with us. We are now in the process of rapid prototyping the
design. Figures 2 to 4 are screen prints from the prototype.
An idealized scenario
This medical record (Figure
1) spans from 1990 to 1995 and shows on a series of timelines
consultations and tests, a list of medical conditions, documents,
hospitalizations and medications. Each physician is assigned
a different color (not seen on the B&W figure) tying documents
and medications to the originating physician. Icons are used
for discrete events, lines for continuous items such as conditions
or hospital stays. The size of the line is used to indicate changes
in the severity of conditions or the dosage of medications. The
right most side of the timelines give the current status showing
the active conditions. The area at the top of the screen gives
general information about the patient. By default alerts such
as known allergies are shown on the top right of the LifeLines
(In Figure 1 a user already selected another tab to read the contact
information).
Each element of the display
acts as a "giant" menu to access the detailed information
about the item selected. A click on a consultation brings the
progress note, a click on a lab test displays the results of the
test. A click on a doctor's name brings up the contact information.
The detail information may cover part of the display but if a
large screen is available it will appear in side windows. Tiled
windows allow parallel review (e.g. a lab test and a letter, or
a recent lab test and 2 older ones). The respective size of the
tiled windows can be adjusted by dragging the sides of the windows.
Dealing with complex records
In our previous work for the Maryland
department of Juvenile Justice [3] we explored mechanisms to handle
complex records. Crowding problems arise when the time span,
the number of categories, the number of lines or the number of
events increases significantly. Scroll bars are the common answer
to pixel shortage but scroll bars are inadequate - if not harmful
- when presenting overviews as users often forget to browse the
complete image. We believe that LifeLines should always begin
with an overview of the entire record in one screen even if some
details are lost in this first view.
In the case of DJJ, some juvenile
records contain dozens of cases whose lines cannot be shown on
the overview screen. We proposed to present the silhouette of
the record (compacted lines with no labels, but color and size
coding for severity and deepness of penetration in the legal system).
Those silhouettes were found to be useful to estimate the volume
and type of information available and to guide users to the most
important parts of the records (e.g. a few serious cases lost
in myriads of less important offenses). Users can explore the
silhouette dynamically, revealing details about the cases as they
move the cursor above the lines. When users have appreciated the
overview they can access the relevant information directly, zoom
on a time period of interest, or filter the display to show only
relevant information.
In the case of the medical record,
the silhouette of a record might be useful when a physician sees
the record for the first time, then a simple click zooms the silhouette
to show the current time period and possibly filter on a subset
of categories chosen in the physician's preference control panel.
The record would appear in the zoomed state during following
visits. Another technique is to use a logarithmic scale to emphasize
recent events.
Categories of data can be organized
in a hierarchy (as shown in Figure 1 - left side) which can be
expanded or contracted to filter out some of the data.
Finally smaller icons can be used
to display a larger number of events information . In the silhouette
a single pixel can represent a visit or a test, giving a sense
of the periodicity of the events; and color can indicate type
or importance.
Cardiology record design
In this design (Figure 2) the
specialist who worked with us elected to display only a subset
of the information in the overview screen (problems and interventions)
giving more space for the progress notes and current medications.
The exact choice of what information is presented on the LifeLines
in the overview is selected by the user. Tests are available
on top of the timelines. A click on the EKG tab brings the last
EKG (with a "previous" button to review previous EKGs),
a click on lab brings the last lab test. A click on a PTCA of
the interventions line brings the report and an hand-drawn diagram
associated with the intervention (Figure 3). The LifeLine of
the medication history appears when clicking on "med. history"
next to the current medication. In certain cases users might
choose to display specific quantitative data e.g. cholesterol
levels over the whole span of the overview to see changes over
the years. But more likely the physician will decide to zoom
in onto a period of interest and see more details. In Figure
4 the physician zoomed to the 1990 period. The zoomed display
now reveals the tests available (a click on any one brings the
data) and the medications lines. Here the user also asked to
show the three cholesterol levels plotted on one line in order
to verify that the medications used at that time had been effective
in lowering the cholesterol levels. The LifeLines can be scrolled
to later time periods if needed.
Many events only require to be
time stamped on the timeline. Test results can be received by
an office clerk, scanned and marked as test results of a given
type, if not transferred automatically. Those events require
little or no input, merely leaving their footprints on the LifeLines
for easy retrieval of the information. Major summary items (problems,
treatments) are found in the progress notes which are often dictated
by the physician and typed by a trained secretary. Codes can
be added to the progress notes or lines can be entered by the
trained secretary for review along with the notes by the physician.
Personal histories contain
information that plays an important role in critical decision
making. However, problems are associated with current methods
of presenting the information to physicians. The proposed solution,
LifeLines, charts the course of a personal history across time
using lines to mark periods and icons to denote events of that
history.
We want to thank Dr. Warren Levy
and Rina Levy for their collaboration on the cardiology case study,
Steve Karrassik and Ara Kotchian for their help producing screen
prototypes and Brett Milash and Seth Widoff for their early designs
of medical record LifeLines.
1. Dewey, J.D., Manning, P., Brandt, S., Acceptance of direct physician access to a computer based patient record in a managed care setting. Proc. 17th Symposium on Computer Applications in Medical Care. (1993) 79-83, AMIA, Washington DC.
2. Factor, M., Gelernter,
D.H., Kolb, C.E., Miller, P.L., Sittig, D.F. Real-time dasta fusion
in the intensive care unit. IEEE Computer, November 24
(1991) 45-54.
3. Plaisant, C., Milash,
B., Rose, A., Widoff, S., Shneiderman, B., LifeLines: Visualizing
Personal Histories. To appear in Proc. of CHI 96, ACM,
New York.
4. Plaisant, C., Shneiderman,
B. Scheduling home control devices: design issues and usability
evaluation of four touchscreen interfaces. Int. J Man-Machine
Studies, 36 (1992) 375-393
5. Post R.M. et al., Am.
J. Psychiatry, 1988:145 844-848
6. Powsner, S.M., Tufte,
E.R., Graphical summary of patient status. The Lancet,
344 (Aug 6, 1994) 386-389
7. R Stiphout, R. Schiffman,
R., Christer, M., Ward, R., Purves, T., Medical Information management
system (MIMS) Carewindows. Proc. 15th Symposium on Computer
Applications in Medical Care. (1991) 929-931, AMIA, Washington
DC.
8. R Tang, P., Annevelink,
J., Fafchamps, D., Stanton, W., Young, C., Physicians' workstations:
Integrated information management for clinicians. Proc. 15th
Symposium on Computer Applications in Medical Care. (1991)
569-573
9. Tufte, E.R., The
Visual Display of Quantitative Information. Graphics Press.
Cheshire, Connecticut. 1983.
10. Whiting-O'Keefe, Q.E.,
Simbork, D.W., Epstein, W.V., Warger, A. A Computerized summary
medical record system can provide more information than the standard
medical record. JAMA, 254 (1985) 1185-1192
11. Wyatt J.C., Clinical
data systems, part 1: data and medical records. The Lancet,
344 (Dec 3 1994) 1543-1547