Skip to content

Truncation of Clinical Terms

Truncation of Clinical Terms

Source PDF: truncation.pdf

Design Consultation Document

Prepared for

NHS Connecting for Health

Friday, 18 December 2009

Version 1.0.0.0 Baseline

Prepared by

Clinical Applications and Patient Safety Project

NHS CUI Programme Team

This document was prepared for NHS Connecting for Health which ceased to exist on 31 March 2013. It may contain references to organisations, projects and other initiatives which also no longer exist. If you have any questions relating to any such references, or to any other aspect of the content, please contact cuistakeholder.mailbox@hscic.gov.uk

Executive Summary

Problem

- Truncation of text is already widely used in information rich views in both nonclinical and clinical software
- Truncated clinical terms may be more prone to misinterpretation and misselection than terms displayed in full
- An assessment of the patient safety risks of alternatives to truncation is needed

to inform future design decisions Scope

- This study focuses on truncation of the text of clinical terms displayed in a

selection list, with a brief look at truncation of terms in a summary view Headline Findings

- Truncation of clinical terms should be avoided where possible
- Truncated clinical terms may introduce more patient safety risks than terms

displayed in full

- In most cases, clinical terms should be displayed in full in a list with a vertical

scroll bar

- Truncation may be relevant in specific contexts, types of information, tasks or

views

- It is unlikely that a set of generic rules for truncation can be defined

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 2

  • Executive Summary

  • Table of Contents

  • Problem Statement

  • Scope Statement

  • Process Overview

  • About this design consultation

  • Patient Safety Overview

  • Design Exploration

  • Themes

  • Supporting Materials

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 3

Problem Statement

Customer need

• Effectively display lists of clinical terms when UI space is limited • Review the potential for generic guidance for the truncation of clinical terms in any context

User experience goals

• Support the user task of selecting a clinical term from a list as part of the process of encoding • Suggest mitigations for patient safety risks associated with hidden text and mis-selection in lists

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 4

Scope Statement

In Scope

  • How to truncate coded clinical terms displayed in a selection list
- Exploration and research limited to SNOMED CT®
  • How to truncate coded clinical terms displayed in a list or table
- Limited feedback based on evolving work in the Display of Clinical Statements Workstream
  • How to access the full text of a truncated clinical term
- Limited feedback based on evolving work in the Display of Clinical Statements Workstream

Out of Scope

  • Truncation of coded clinical terms in scenarios not listed above

  • Truncation of user interface elements (such as menu items and window titles)

  • Truncation of other information, such as medications (apart from those displayed as coded clinical terms), patient names and dates

  • Truncation of non-encoded clinical terms

  • Truncation of post coordinated coded clinical terms

  • Using approved acronyms or abbreviations to shorten text

  • Truncation of free text

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 5

Process Overview

Scoping and Research

  • Review of NHS CFH Rules for the Display of Coded Clinical Data

  • Review of truncation in a small number of clinical and non-clinical applications

  • Review of NHS CFH analysis of SNOMED CT

  • Review of UI principles for avoiding truncation

Risk Assessments

  • Preparation of scenario and design alternatives

  • Reporting findings

Design Exploration

  • Design exploration for truncating in selection lists

  • Design exploration for truncating in views that summarise clinical statements

User Research

  • Risk assessment of designs in interviews with six participants

  • Reporting user research findings

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 6

About this design consultation

The purpose of the document is to summarize the current learning that has been achieved in the design and research process to date. This is articulated primarily in terms of the features that comprise the preferred current design. These design features aggregate into seven themes.

For each theme we illustrate a few design points with images of the current design, plus an indication of some of the alternatives we have been considering. We then summarise the key design points, including some that haven’t been illustrated, plus the rationale for choosing these points. We end each theme by outlining the proposed next steps and future exploration.

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 7

Patient Safety Overview

The development lifecycle for this consultation document includes an integrated patient safety assessment and management process.

Known patient safety incidents relevant to this area have been researched and reviewed as part of ongoing development.

The Hazard Log records all the risks that have been identified during development and describes potential mitigatory actions that could be considered alongside future exploratory work in this area. The Hazard Log is a live document that will be updated as this area of work is developed further.

Until this work is progressed and developed to full design guide status it will not be in a position to achieve full Clinical Authority to Release (CATR) from the NHS Connecting for Health (CFH) Clinical Safety Group (CSG).

Please refer to www.cui.nhs.uk for further information on the patient safety process and for the safety status and any relevant accompanying safety documentation for this consultation document.

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 8

DESIGN EXPLORATION

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 9

Design Exploration

All illustrations use current SNOMED CT terms.

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 10

Introduction – what is truncation?

In this document, the word ‘truncation’ is used to refer to the removal of a section of text in order to fit the remainder into a smaller space

Although this document focuses on truncation, the use of truncation must be compared to obscuring

Obscured text

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 11

Introduction – truncation in non-clinical applications

Truncation is widely used in both clinical and non-clinical applications Text is truncated at the end or in the (approximate or exact) ‘middle’

pending on contextCol2Col3
A list of emails truncated at the end

A list of files truncated in the ‘middle’

Truncation of text is generally used when text is limited to single lines in dynamic width spaces so the truncated text changes as the width is changed. Words are usually fragmented to make best use of the available space

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 12

Introduction – truncation in clinical applications

Truncation is a popular solution to solving the need to display long terms in information-rich views, especially when that information is presented in

ular formCol2Truncated Text with an EllipsisCol4Col5Col6
Terms cut short by
the edge of the list
Terms cut short by
the edge of the list
Terms cut short by
the edge of the list
Terms cut short by
the edge of the list
Terms cut short by
the edge of the list
Horizontal and Vertical Scroll Bars

Truncation of medications

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 13

Introduction – lengths of SNOMED CT terms

18,000

16,000

14,000

12,000

10,000

8,000

6,000

4,000

2,000

0

Term Lengths (number of characters)

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 14

Theme 1

AVOIDING the need for TRUNCATION

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 15

the need for

Avoiding truncation – wrapping

Wrapping ensures that all items are displayed in full. If each item in a list can be scrolled into view so that the whole term is visible, then each term can be reviewed easily before a selection is made

the need for

Avoiding truncation – scrolling

Truncation of clinical terms may not be appropriate for some tasks. For example, when the task is browsing (rather than searching for a known term), the list may be presented as a tree

provides more immediate access to full terms

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 17

the need for

Avoiding truncation – grouping

Groups help to break the list down into smaller chunks and reduce the number of list items that need to be reviewed

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 18

the need for

Avoiding truncation

Summary

  1. Wrapping

a) Ensures that all items are displayed in full (with the help of a vertical scroll bar) b) When long words wrap, it can create a lot of unused space on the right of the text c) When wrapping, rules may be needed to keep associated elements such as numbers and

units together 2. If summaries can be created that are meaningful, it may be possible to provide higher level summary views that can then be used to drill down a subset of the detail that can then be displayed in full (without truncation) 3. Approved acronyms and abbreviations can be used to shorten some structured text entries 4. Groups add structure and can be used to hide the text in each list item that is implied by the group it is in 5. Modifying search processes and supporting filtering may help users to work with shorter lists in which truncation is not necessary 6. Showing or hiding certain types of information (or labels in a chart view) may simplify the view and leave more room for other information to be displayed with minimum truncation

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 19

the need for

Avoiding truncation

Emerging RecommendationsEvidence
Use truncation as a last resort and
only when truncated terms are the
exception rather than the rule
• Tasks associated with both summary views and
selection lists are not well supported when truncation
of terms is a common occurrence
• There are many patient safety risks (such as
misinterpretation and missing the presence of
truncated text) associated with the interpretation of
truncated terms
When wrapping encoded text
ensure that elements such as
labels, numbers, units and
separators are kept together
where necessary
• When wrapping separates quantities (100) from units
(milligrams), there is a risk that, in skim reading, the
wrapped text is guessed rather than actively read.
Preserving them together mitigates this risk
The completion of further work in this area may
generate a set of more specific recommendations

Note: Emerging recommendations are at many levels (some high level and generic, others more detailed) and include more than one (mutually exclusive) approach. They need significant further design, research and risk assessment before they can be considered as candidates for future guidance

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 20

the need for

Avoiding truncation

Emerging RecommendationsEvidence
Do not wrap mid-word (thus
creating word fragments with
or without hyphens)
• Hyphenating words that are not usually hyphenated may
create fragments that can be misinterpreted as whole
words
• As above, the wrapped part of a hyphenated word may be
guessed by the user rather than actively read
The completion of further work in this area may identify
the need for a technical solution to prevent applications
from taking hyphens as wrapping points by default
If a list can include items that
are too long to be displayed in
full without scrolling, allow the
list to be resized so that it is
possible to see the longest
term without scrolling
• The task of selecting from a list is not well supported if
rapid scrolling is needed and list items must be held in
short term memory in order to compare between them
and make an appropriate selection

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 21

the need for

Avoiding truncation

Emerging RecommendationsEvidence
Do not provide a horizontal
scroll bar in selection lists
unless the list can be resized to
display all terms in full
• The task of selecting from a list is not well supported if
rapid scrolling is needed and list items must be held in
short term memory in order to compare between them
and make an appropriate selection
• Horizontal scrolling does not allow a list to be scrolled so
that a list item is visible in full
• Comparing list items is more difficult with a horizontal
scroll bar unless the differentiating text is at the end of
each term
• Horizontal scrolling is a more complex mechanical
interaction than vertical scrolling when a mouse with a
scroll wheel is used

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 22

the need for

Avoiding truncation

Areas / questions for further study

  • If truncation is being considered, is the design solution the best way of supporting the user task?

    • For example, if truncation is being considered for selection lists of clinical terms, is

a selection list the best mechanism for making that selection?

  • In which contexts is truncated information sufficiently meaningful to support the tasks of searching and selecting an appropriate term or reviewing a patient using summary views?

  • There may be some contexts in which truncation is appropriate only for expert users or certain roles and should be avoided for novice users or for other roles

  • If groups are used in search results, more work is needed to determine whether the structure of the results should be manuallydefined or influenced by the search criteria

  • Emerging recommendations can be used to inform further work on avoiding the need for truncation

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 23

Theme 2

TRUNCATION POINTS

If truncation cannot be avoided, where should text truncate?

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 24

Truncation points – risks

Truncating in the ‘middle’ may create ‘artificially meaningful’ sentences, cause more than one fragmented word and / or affect the space and

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 25

Truncation points – visibility

Truncating at the end is potentially easier to notice and a truncation point at the end is more consistently placed (visually) than a truncation point in the ‘middle’. However, truncation in the ‘middle’ may be more effective and easier to notice when each item is limited to a single line

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 26

ntial

Truncation points – manually-defined truncation

A set of manually defined truncation points (defined for each term) could ensure that clinical terms are always truncated in acceptable places and that phrases are preserved where necessary

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 27

Truncation points – summary views

A summary view (containing clinical terms) is not effective if the information

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 28

Truncation points – numbers and units

Numbers may be separated from associated units when wrapping or truncation is used

When truncation allows numbers to be fragmented, a number may be missing one or more digits

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 29

Truncation points

Summary

  1. Preserving phrases is difficult to do automatically
  2. Fixed character limit

a) Can create an artificially large space after the point of truncation b) Not likely to make the best use of the available space c) Could be used to ensure consistent truncation in different contexts but likely to introduce problems in a

context where the character limit is less suitable 3. Truncating at the end

a) Is potentially easier to notice b) A truncation point at the end is more consistently placed (visually) than a truncation point in the ‘middle’

(wrapping rules and whole word or fragmented words)

  1. Truncating in the ‘middle’

a) May create ‘artificially meaningful’ sentences b) May create more than one fragmented word and affects the space and wrapping behaviour of the

truncation marker c) May be more effective when limiting to a single line

  1. More than one truncation point (in a single term or text item)

a) May allow phrases or differentiating words to be maintained in view b) Requires sophisticated logic and is more difficult to achieve with a dynamic truncation point 6. A set of approved, manually-defined truncation points could ensure that terms are always truncated in acceptable (and expected) places

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 30

Truncation points

Emerging RecommendationsEvidence
Do not truncate clinical terms
when they appear in a list of
fixed-width
• When a width is fixed, a solution is needed to support
access to the full text for truncated terms
• A fixed-width list cannot be resized (even temporarily) to
reveal the truncated text
• Any solution for accessing the full text of a truncated term
is likely to be even less direct (and effective) than scrolling
Only use truncation in a
selection list of clinical terms if
the choice of text to truncate is
informed by the data structure
or a manually defined means of
selecting acceptable text to
truncate is available
• If truncation can be used in a way that supports accurate
review of a list without accessing the full text, then some
of the patient safety risks associated with truncated text
are mitigated

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 31

Truncation points

Emerging RecommendationsEvidence
Only use ‘middle truncation’
when text is limited to a single
line
• ‘Middle’ truncation is used effectively in non-clinical
contexts and is therefore likely to be effective in some
clinical contexts (depending on the nature of the
information)
• ‘Middle’ truncation may be the best approach for
summary views that display information in which the end
of the text is known to be the likely differentiator and no
actions can be completed on the truncated text that are
unsafe
The completion of further work in this area may generate
exceptions to this rule, such as ‘don’t use ‘middle’
truncation for clinical terms’

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 32

Truncation points

Areas / questions for further study

• Determining when to use truncation and where to truncate may be dependent on an analysis of the specific type of information subject to truncation • Emerging recommendations can be used to inform further work on truncation points

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 33

Theme 3

FRAGMENTING WORDS

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 34

Fragmenting words – dynamic width

Fragmenting words works best with dynamic truncation – since the truncation point can move one letter at a time as a text area is resized

A fragmented word may be more noticeable than truncated text in which whole words are preserved

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 35

Fragmenting words – contexts

Truncation rules may need to be different depending on the context since it may be acceptable or even necessary to fragment words in some contexts

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 36

Fragmenting words

Summary

  1. Fragmenting words works best with dynamic truncation because the truncation point can move one letter at a time as a text area is resized and more of the text is visible than when a whole word is truncated at a time

  2. A fragmented word may be more noticeable than truncated text in which whole words are preserved

  3. A fragmented word might look like a whole word. Similarly, fragmented numbers may be interpreted as whole when it is missing one or more digits

  4. Hyphenation can be used to maximise the quantity of text that can be displayed on each line but creates fragmented words

  5. When words are preserved, there may be greater space at the end of truncated text

  6. In views (such as a timeline, chart or graph) where text cannot wrap onto a new line, it may be more appropriate to truncate and to fragment words in order to preserve as much visible text as possible

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 37

Fragmenting words

Emerging RecommendationsEvidence
Do not truncate such that
characters or digits are missing
from numbers or units
• Removal of digits from numbers creates new numbers
• Separation of numbers from units may lead to the units
being guessed
• Truncation of a unit (such as millilitres) may create or
imply a different unit (such as milli… or…litres)
Do not fragment words when
truncating text that is displayed in
a fixed-width area
• For similar reasons that clinical terms should not be
truncated in fixed-width lists. Ease of access to the full
term is important here

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 38

Fragmenting words

Areas / questions for further study

• Emerging recommendations can be used to inform further work on fragmenting words

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 39

Theme 4

DYNAMIC versus STATIC TRUNCATION

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 40

Dynamic versus static truncation

When truncation is dynamic, the truncation point depends on the space available to display the text. When truncation is static, each term is always truncated at the same point in the text, regardless of the space available

The point of truncation depends
Dynamic – Limited to 2 Creation of valved conduit
on the space available
between right ventricle of…
lines
Col2The point of truncation depends
on the space available
Dynamic– Limited to 2
lines
Creation of valved conduit
between right ventricle of…
Creation of valved conduit
between right ventricle of…

25 characters wide

50 characters wide

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 41

Dynamic versus static truncation

When a space has fixed dimensions, the text will always be truncated at the same point

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 42

Dynamic versus static truncation

If the width of a selection list can be changed, it may be possible to increase the width until none of the items are truncated. When words can be fragmented, resizing may appear smooth as the text is truncated one letter at a time. Truncating whilst preserving whole words causes whole words to disappear when a width is reduced

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 43

Dynamic versus static truncation

Summary

  1. When using static truncation, such as truncating after a fixed number of characters

a) If an area is a set size, the truncation point will always be the same b) Truncation after a set number of characters may leave a lot of white space (thus not making full use of the available space) c) It may be easier to use more sophisticated methods of indicating truncation when the truncation point is static, since the location of the point at which the text is truncated is more predictable

  1. When using dynamic truncation, such as number of lines to wrap in a variable

width space

a) Truncating whole words might create a jerkiness as words are replaced with the truncation marker b) It is more difficult to draw attention to apparent duplicates or text that is open to misinterpretation as a result of the truncation because the truncation point (and thus whether it is duplicated or not) changes c) When truncating within cells, applying truncation to one column may not make best use of space if truncation is not also applied in other columns d) Using a single line limit with a variable width makes it easier to scan a list but it is less effective for selecting from a list when a large number of list items are truncated

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 44

Dynamic versus static truncation

Emerging RecommendationsEvidence
Do not use a set number of characters
as a rule for determining an
appropriate truncation point
• A set number of characters is likely to be
appropriate to a specific context. Since one of the
benefits of using a set truncation point is
consistency, it becomes irrelevant if the chosen
point is not effective in all contexts
Do not provide a horizontal scroll bar
for a view containing terms that are
truncated at a set point
• If it is possible to scroll so that the point of
truncation is off screen, it is more likely that a
truncated term is misinterpreted as being whole

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 45

Dynamic versus static truncation

Areas / questions for further study

• Emerging recommendations can be used to inform further work on dynamic versus static truncation

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 46

ntial

Theme 5

TRUNCATION MARKERS

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 47

Truncation markers

Text markers can be placed at the point of truncation and may be easier to implement than icons or images. Some text markers could be misread as

ers or numbersCol2
Creation of valved conduit between right atrium and pulmo /
Creation of valved conduit between atrium and ventricle of /
Creation of valved conduit between left ventricle of heart a /
Correction of tetralogy of Fallot using valved right ventricul /
Text markers such as “/” or
“>” may be easier to notice
than an ellipsis but may
also be misinterpreted as
letters or numbers
Text markers such as “/” or
“>” may be easier to notice
than an ellipsis but may
also be misinterpreted as
letters or numbers
Creation of valved conduit between right atrium and pulmo /Creation of valved conduit between right atrium and pulmo /
Creation of valved conduit between atrium and ventricle of /Creation of valved conduit between atrium and ventricle of /
Correction of tetralogy of Fallot using valved right ventricul /Correction of tetralogy of Fallot using valved right ventricul /
Creation of valved conduit between left ventricle of heart a /Creation of valved conduit between left ventricle of heart a /

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 48

Truncation markers

A sequence of text characters, such as ‘… >>’ or ‘*… /’ may be easier to notice than a text marker such as an ellipsis ‘…’

Symbols, icons or controls may be more noticeable than an ellipsis They may also take up more space (so more text has to be truncated)

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 49

Truncation markers

Icons or controls may distract too much from the text itself, especially if there are many of them in one view

KEY RISK - that
information (w
kind of truncat
marker) is assu
be unimportan
13-Oct-2009
hidden
1 day diarrhoea.
Liquid…
History of current
condition…
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
13-Oct-2009
ith any
ion
Shortness of
breath…
History of current
condition…
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
13-Oct-2009
med to
t
No haemoptysisHistory of current
condition…
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
PastNo history of MI/
Stroke/ …
Past medical history
Known absent: Epilepsy
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
Mild aortic
stenosis
Past medical history
Aortic stenosis - Mild
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
No cancerFamily history
Known absent: …
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
2008
(calculated)
Father died of MI
aged 55
Family history
Known present: …
13-Oct-2009
Dr Sophie Allen
**KEY RISK -**that
information (w
kind of truncat
marker) is assu
be unimportan
1997L index finger,
PIP joint…
Past surgical history
Joint injury, Open…
02-Feb-2007
Dr Tim Smith

18 December 2009 Copyright ©2013 Health and Social Care Information Centre

ial

Slide 50

Truncation markers

Markers can be placed at the point where text is truncated or in association with the text item

A marker may also be placed such that it is associated with the text but not part of it

The ellipsis can be placed at the point where text is truncated

13-Jun-2008international neuroblastoma pathology
xxclassification: Favourable histology…
13-Oct-2009
Dr Sophie Allen
13-Nov-1997emergency replacement of aneurysmal
xxbifurcation of aorta by anastomosis…
16-Jan-1999
Dr Sophie Allen
13-Jan-1990open fracture of forearm, lower end,
xxunspecified
16-Jan-2009
Dr Sophie Allen
13-Oct-1987pancreatic stone13-Oct-1987
Dr Sophie Allen

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 51

Truncation markers

Summary

  1. Text markers:

a) Ellipses are used in typography to mark omission from speech or writing of a word or words

that are superfluous or able to be understood from the context. It is also the term for a set of dots (…) used to indicate such an omission b) A text marker may be easier to implement and to place at the point of truncation than an

icon or control c) Text markers, such as ‘>>’ or ‘/’, may be easier to notice than ‘…’ but may also be

misinterpreted or mistaken for text d) A sequence of text characters, such as ‘’ or ‘/ … /’, may be easier to notice than a single

character such as the ellipsis ‘…’ e) Bold or colourful truncation markers may draw attention away from the text f) Placing a space between the text and truncation marker may cause the marker to wrap to

the beginning of a new line

  1. Symbols, icons and controls:

a) May be useful if a control for displaying the full text is to be placed near the truncated text b) May distract too much from the text itself c) A symbol or icon may be more noticeable than an ellipsis d) The use of larger markers or controls may take up more space and result in more text being

truncated

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 52

Truncation markers

Emerging RecommendationsEvidence
When text is truncated at the end,
display (at least) an ellipsis at the point
of truncation
• An ellipsis is more noticeable when displayed at the
end of a line of truncated text (than in the ‘middle’)
• Truncation without a truncation marker is easily
missed
• Without a truncation marker truncated text may
not be noticed in a dynamic width list in which text
is cut off by the right hand edge of the area. If the
user does not realise that the area can be resized,
the truncation may never be discovered.
If truncation is used in a selection list,
supplement the ellipsis with a symbol
for each truncated list item
• Positive feedback in user testing and risk
assessments
The completion of further work in this area should
include a review of this recommendation since it
depends heavily on the contents of the list, the type
of task it is designed to support, the type of
truncation used and whether it can be resized. A
guidance point would need to be more specific than
this

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 53

Truncation markers

Emerging RecommendationsEvidence
Do not support the use of the ellipsis
as a control (such as a button) for
accessing the full text of a truncated
term
• Derived from the Windows Style Guide
• The ellipsis is not commonly used as a control so
this function is not likely to be sufficiently
discoverable

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 54

Truncation markers

Areas / questions for further study

• A more detailed study of revealing full text in selection lists containing other types of data is needed to determine whether there is a generic approach that can be used in other contexts • Emerging recommendations can be used to inform further work on truncation markers

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 55

Theme 6

IDENTICAL TRUNCATED TERMS

Where ‘identical terms’ are ‘apparent duplicates’ created by the truncation

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 56

Identical truncated terms

When truncated, some terms may be identical. These apparently duplicate terms can be either marked, consolidated into one item or avoided

18 December 2009 Copyright ©2013 Health and Social Care Information Centre

onfidential

Slide 57

Identical truncated terms

If the presence of identical truncated terms in the list is easily avoided by changing the search criteria, a system could suggest words that would reduce the number of duplicates in the search results. A similar approach could be used to avoid the need for truncation altogether

appear next to one another in the list

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 58

Identical truncated terms

Summary

  1. When truncation causes identical terms to be displayed in a selection list:

a) it may be necessary to draw attention to each b) it may be possible to consolidate them so that they are only displayed once

  1. If the presence of identical truncated terms is easily avoided by changing the search criteria, a system could provide prompts to support narrowing the search results

  2. An improved search process can support access to shorter lists in which identical terms are much less likely

  3. Progressive disclosure can help to mitigate apparent duplicates since they can be expanded in context

  4. A rule that requires identical truncated terms to be expanded automatically would create a potentially confusing mixed list with some expanded and others truncated

  5. When many terms are truncated, it is more likely that there are similar (truncated) terms in a list, thus increasing the risk of mis-selection even when they are not identical

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 59

Identical truncated terms

Emerging RecommendationsEvidence
For fixed-width lists, when a list
contains identical truncated terms
(and when this can be determined)
display the list without truncation by
default
• An expert user familiar with the list can choose to
re-apply truncation (so this still allows truncation to
be used but only in certain circumstances)
This recommendation is as close as possible to a
rule stating that lists should never contain truncated
text if they can contain duplicate items (without
being unnecessarily restrictive). It should be tested
further
For fixed-width lists, when it is not
possible for the system to determine
whether a list contains identical
truncated terms, provide a notification
warning of the presence of identical
truncated terms
This recommendation is a reminder that when it is
possible for identical truncated terms to appear in a
list, a way of mitigating the associated risks is
needed

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 60

Identical truncated terms

Areas / questions for further study • Many of the recommendations refer specifically to fixed or dynamic-width lists. This repetition implies that further work is needed to clarify the priorities and the scenarios in which truncation must be excluded • Emerging recommendations can be used to inform further work on identical truncated terms

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 61

Theme 7

ACCESS to FULL TEXT for truncated terms

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 62

to for truncated terms

Access full text

When a selection list contains truncated terms, the full term must be easily available to support the task of comparing and making a selection

A fly-out for a selection list takes more space than a tooltip but does not obscure other information in the list

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 63

to for truncated terms

Access full text

Progressive disclosure can allow more than one truncated term to be displayed in full without obscuring other information

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 64

to for truncated terms

Access full text

A control could be provided that allows all truncated terms in the list to be revealed

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 65

to for truncated terms

Access full text

Summary

  1. Fly-outs and pop-ups

a) Show the full text in place of or in close proximity to the truncated text b) Hover is used to highlight an item in a selection list (immediately) and

display a fly-out (with a delay) c) A fly-out for a selection list takes more space but does not obscure other

information in the list d) A tooltip obscures other information in the same list or table

  1. Expand inline

a) Allows more than one item to be displayed in full b) Allows an item to be expanded in context without obscuring other

information

  1. Expand all truncated terms

a) A control could be provided that allows all truncated terms in the list to be

expanded, either just this once or as a setting for the search results list

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 66

to for truncated terms

Access full text

Emerging RecommendationsEvidence
If it is possible that a list may
contain truncated terms, allow
the user to display the list
without truncation
• If a list contains truncated terms, but the user is
sufficiently familiar with those terms and they are
truncated in places that allow them to be differentiated
without revealing the full text, then it may be acceptable
to work with the truncated list. If the user is not familiar
with the terms, they may choose (and it is safer for them)
to view the list without truncation
If it is possible that a list may
contain truncated text, give the
user control over the width of
the list
• In many cases, this allows a simple and direct method of
removing the truncation by resizing the list
• Larger lists may be appropriate when reviewing long terms
• When a list is resized, more text is displayed within them
This recommendation aims to encourage the provision of
dynamic-width lists as far as possible until more work is
completed that can generate more specific
recommendations or guidance

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 67

to for truncated terms

Access full text

Emerging RecommendationsEvidence
Ensure that the full text of a
truncated term is displayed and
acknowledged before an action
can be carried out on the
selected term
• This guidance aims to ensure that the full text of a term is
reviewed actively before any action is carried out on the
selection

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 68

to for truncated terms

Access full text

Areas / questions for further study

• Assuming that the need to access full text for a truncated term is expected to be infrequent, are the methods of accessing full text explored here too complex? • When assessing design alternatives for accessing full text, consider the full process including the steps immediately before and after the display of truncated terms • The primary purpose of accessing full text is to support the process of comparing and selecting so there should be a focus on considering mis-selection risks • Emerging recommendations can be used to inform further work on access to full text for truncated terms

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 69

Supporting Materials

The following additional supporting materials were used in the research of this area and creation of this document

  • Working Documents
- _Truncation Kick-off_ (PowerPoint)
- Risk Assessment Presentation - _Truncation ORA 4_ _[th]_ _-Nov-2009_ (PowerPoint)
- Risk Assessment Scenarios - _ORA Scenario_ (Visio)
- Risk Assessment Design Alternatives - _ORA Design Alternatives_ (Visio)
- Risk Assessment Findings - _ORA Findings_ (Visio)
- Risk Assessment Findings (illustrations) - _Truncation ORA – Findings_ (PowerPoint)
- _Truncation Design Log_ (Excel)
- _Truncation Design Challenges_ (PowerPoint) (Visio)
- _Truncation User Research Materials_ (PowerPoint) (Visio)
- _User Research Notes_ (Word)
- _Truncation Notes_ (Word)
- User Research Findings - _NHS CUI CAPS Truncation user feedback Nov 2009 0.0.0.2_ (PowerPoint)
  • NHS CFH Documents
- _Truncation – Master Hazard Log – v0.5 – 13 Nov 2009_ (Excel)
- _CFH_Position-paper_Rules-for-truncation-and-wrapping-of-coded-clinical-data_v10_27-Jun-2008_

(Word)

- Analysis of SNOMED CT terms (Excel)

To obtain any of these supporting materials please contact cuistakeholder.mailbox@nhs.net

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 70

Distribution

Reviewers and Distribution

NamePositionVersion ApprovedDate
Mike CareyNHS CFH Project Manager0.2.0.017-Dec-2009
Tim ChearmanNHS CFH Project Lead0.2.0.017-Dec-2009
Peter JohnsonClinical Architect0.2.0.017-Dec-2009
Frank CrossClinical Advisor0.2.0.017-Dec-2009
Lindsey ButlerClinical Safety Advisor0.2.0.017-Dec-2009
Greg ScottClinical Advisor0.2.0.017-Dec-2009
Priya ShahClinical Advisor0.2.0.017-Dec-2009

18 December 2009 Copyright ©2013 Health and Social Care Information Centre Slide 71