Technical ReportPDF Available

A Technical Note on the PainChek™ System: A Web Portal and Mobile Medical Device for Assessing Pain in People With Dementia


Abstract and Figures

Background: Pain in dementia is predominant particularly in the advanced stages or in those who are unable to verbalize. Uncontrolled pain alters the course of behaviors in patients with dementia making them perturbed, unsettled, and devitalized. Current measures of assessing pain in this population group are inadequate and underutilized in clinical practice because they lack systematic evaluation and innovative design. Objective: To describe a novel method and system of pain assessment using a combination of technologies: automated facial recognition and analysis (AFRA), smart computing, affective computing, and cloud computing (Internet of Things) for people with advanced dementia. Methods and Results: Cognification and affective computing were used to conceptualize the system. A computerized clinical system was developed to address the challenging problem of identifying pain in non-verbal patients with dementia. The system is composed of a smart device enabled app (App) linked to a web admin portal (WAP). The App "PainChek TM " uses AFRA to identify facial action units indicative of pain presence, and user-fed clinical information to calculate a pain intensity score. The App has various functionalities including: pain assessment, pain monitoring, patient profiling, and data synchronization (into the WAP). The WAP serves as a database that collects the data obtained through the App in the clinical setting. These technologies can assist in addressing the various characteristics of pain (e.g., subjectivity, multidimensionality, and dynamicity). With over 750 paired assessments conducted, the App has been validated in two clinical studies (n = 74, age: 60-98 y), which showed sound psychometric properties: excellent concurrent validity (r = 0.882-0.911), interrater reliability (Kw = 0.74-0.86), internal consistency (α = 0.925-0.950), and excellent test-retest reliability (ICC = 0.904), while it possesses good predictive validity and discriminant validity. Clinimetric data revealed high accuracy (95.0%), sensitivity (96.1%), and specificity (91.4%) as well as excellent clinical utility (0.95). Conclusions: PainChek TM is a comprehensive and evidence-based pain management system. This novel approach has the potential to transform pain assessment in people who are unable to verbalize because it can be used by clinicians and carers in everyday clinical practice.
Content may be subject to copyright.
published: 12 June 2018
doi: 10.3389/fnagi.2018.00117
Frontiers in Aging Neuroscience | 1June 2018 | Volume 10 | Article 117
Edited by:
Donna M. Wilcock,
University of Kentucky, United States
Reviewed by:
Paola Sandroni,
Mayo Clinic, United States
Filippo Brighina,
Università degli Studi di Palermo, Italy
Mustafa Atee
Received: 17 January 2018
Accepted: 04 April 2018
Published: 12 June 2018
Atee M, Hoti K and Hughes JD (2018)
A Technical Note on the PainChekTM
System: A Web Portal and Mobile
Medical Device for Assessing Pain in
People With Dementia.
Front. Aging Neurosci. 10:117.
doi: 10.3389/fnagi.2018.00117
A Technical Note on the PainChekTM
System: A Web Portal and Mobile
Medical Device for Assessing Pain in
People With Dementia
Mustafa Atee 1
*, Kreshnik Hoti 1,2 and Jeffery D. Hughes 1
1School of Pharmacy and Biomedical Sciences, Faculty of Health Sciences, Curtin University, Bentley, WA, Australia,
2Division of Pharmacy, Faculty of Medicine, University of Pristina, Prishtina, Kosovo, Albania
Background: Pain in dementia is predominant particularly in the advanced stages or
in those who are unable to verbalize. Uncontrolled pain alters the course of behaviors
in patients with dementia making them perturbed, unsettled, and devitalized. Current
measures of assessing pain in this population group are inadequate and underutilized in
clinical practice because they lack systematic evaluation and innovative design.
Objective: To describe a novel method and system of pain assessment using a
combination of technologies: automated facial recognition and analysis (AFRA), smart
computing, affective computing, and cloud computing (Internet of Things) for people
with advanced dementia.
Methods and Results: Cognification and affective computing were used to
conceptualize the system. A computerized clinical system was developed to address
the challenging problem of identifying pain in non-verbal patients with dementia.
The system is composed of a smart device enabled app (App) linked to a web
admin portal (WAP). The App “PainChekTM” uses AFRA to identify facial action units
indicative of pain presence, and user-fed clinical information to calculate a pain intensity
score. The App has various functionalities including: pain assessment, pain monitoring,
patient profiling, and data synchronization (into the WAP). The WAP serves as a
database that collects the data obtained through the App in the clinical setting. These
technologies can assist in addressing the various characteristics of pain (e.g., subjectivity,
multidimensionality, and dynamicity). With over 750 paired assessments conducted,
the App has been validated in two clinical studies (n=74, age: 60–98 y), which
showed sound psychometric properties: excellent concurrent validity (r=0.882–0.911),
interrater reliability (Kw =0.74–0.86), internal consistency (α=0.925–0.950), and
excellent test-retest reliability (ICC =0.904), while it possesses good predictive
validity and discriminant validity. Clinimetric data revealed high accuracy (95.0%),
sensitivity (96.1%), and specificity (91.4%) as well as excellent clinical utility (0.95).
Atee et al. A Technical Note on the PainChekTM System
Conclusions: PainChekTM is a comprehensive and evidence-based pain management
system. This novel approach has the potential to transform pain assessment in people
who are unable to verbalize because it can be used by clinicians and carers in everyday
clinical practice.
Keywords: PainChekTM, pain assessment system, artificial intelligence, automated facial recognition, dementia,
smart device application, technology
In 2017, there are an estimated 962 million people aged 60 or
over in the world, comprising 13% of the global population
(United Nations Department of Economic Social Affairs
Population Division, 2017). This number is projected to
increase to 1.4 billion in 2030 and 2.1 billion in 2050, and
could rise to 3.1 billion in 2100 (United Nations Department
of Economic Social Affairs Population Division, 2017). The
incidence of dementia doubles beyond the age of 65 and becomes
prevalent by up to 50% over the age of 85 (Duthey, 2013).
Currently, there are 50 million people living with the condition
worldwide (ADI, 2017; WHO, 2017). Dementia is a clinical
neurodegenerative syndrome characterized by progressively
impaired cognition, communication (including pain self-
reporting), and comprehension as well as the lack of ability to
execute simple daily activities (Mitchell et al., 2009; WHO, 2017).
Pain is common (up to 80%) in people with dementia but it often
goes undetected and untreated, particularly in those who cannot
verbalize or express their needs (Hadjistavropoulos et al., 2014).
Uncontrolled pain alters the course of behaviors in patients with
dementia making them perturbed, unsettled, and devitalized
(Hadjistavropoulos et al., 2014).
Despite the availability of more than 35 observational-
behavioral pain assessment tools for adults with communication
difficulties (Hadjistavropoulos et al., 2014), including those
with advanced dementia, none are currently approved by
any regulatory agency, such as the United States’ Food and
Drug Administration (FDA), Australias Therapeutic Goods
Administration (TGA), and European Conformity (CE) mark.
As of July 2017, the electronic Pain Assessment Tool (ePAT)
[now known as PainChekTM] received regulatory clearance as a
Class 1 medical device for pain assessment and monitoring in
adults who cannot verbalize (e.g., those with dementia) from the
TGA and CE marking (ARTG, 2017). The mobile application
(App) as a medical device has also been approved for use in
other non-verbal adult populations such as those with other
neurodegenerative disorders, intellectual disability, traumatic
brain injury, aphasia, those receiving palliative care, and post
stroke patients (ARTG, 2017). Further, none of the existing tools
has their own electronic database, which collects data in real
time. This is an important feature of clinical tools because it
identifies the need for therapeutic intervention(s) in a timely
fashion, which if successful lead(s) to improvement in patient
outcomes. To achieve this goal we have developed an online
secure portal linked to the tool App that can be accessed through
various computing devices (e.g., PC, smart tablet, smartphone).
Current pain assessment tools in dementia also lack the
innovative design and advanced technological characteristics.
In a large meta-review of pain assessment tools in dementia
Lichtner et al. (2014), argue the need for new tools that
contain innovative characteristics to be able to transform the
process of pain assessment in non-verbal older adults with
This note aims to describe a novel system called PainChekTM
focusing on its conceptual foundation, clinical and technical
contents, clinical use, and practical tips for use in clinical
Conceptual Foundation of PainChekTM
System (Figure 1)
In designing the PainChekTM system, the following
conceptualizations were considered:
1. The subjective nature of pain i.e. individualized experience of
pain as per its definition by the International Association for
the Study of Pain (IASP) (Merskey and Bogduk, 1994).
2. The multi-dimensionality, complexity, and dynamicity of pain
as a construct (Merskey and Bogduk, 1994).
3. The American Geriatric Society (AGS) Indicators of
Persistent Pain were selected as a basic framework to enrich
comprehensiveness and to meet the objectives of the tool
(AGS Panel on Persistent Pain in Older Persons, 2002).
4. The temporality of pain and related behaviors, so that trends
and patterns of pain scores provide a comprehensive clinical
picture of the patient under assessment.
5. Objective description of key pain behaviors, such as Facial
Action Coding System (FACS)—pain relevant expressions
(Ekman et al., 1978).
6. Items sensitive to the presence and intensity of pain
were selected on the basis of current evidence (clinical
guidelines, previous studies, and other pain assessment tools
in dementia).
7. Simple scoring mechanism. For clinicians and carers, it is
difficult and highly subjective to make a distinction between
whether a patient has mild, moderate or severe pain-related
behaviors (Flaherty, 1996). We adopted binary scoring in the
PainChekTM pain scale, because such mathematical basis is
more predictive of event outcomes and less prone to error than
ordinal rating (Ridley, 2002). These criteria are also linked to
improved accuracy (Ridley, 2002).
Frontiers in Aging Neuroscience | 2June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
FIGURE 1 | Conceptual model of the PainChekTM system.
8. Innovative technologies were considered in developing
the system. Cognification and affective computing were
conceptualized as a model in designing the App to provide
a synergistic effect on the use of the tool (Kelly, 2016).
Cognification integrates artificial (emotional) intelligence (AI)
or affective computing e.g., automated facial recognition and
analysis (AFRA), smart computing, and “Internet of Things”
(IoT). Automation was integrated because the FACS requires
lengthy training, and a certified skilled observer (coder), which
render its use in clinical settings impractical (Craig et al.,
2011). Smart device technology was selected because they are
mobile, miniaturized, cost-efficient, easy to use, have high
processing power, and they allow interoperability. IoT and
cloud computing allow data management in real time, and
transfer of data among different networks. Further, the App
does not need to be connected to the internet while in use. A
glossary of terminology used in the technical characteristics of
the PainChekTM is presented in Table 1.
Rationale for PainChekTM Development
There are various perspectives as to why we have an urgent need
to develop a valid, reliable, and more objective pain assessment
system for people with dementia. From the patient’s perspective,
dementia limits their verbal and cognitive abilities to report
the presence, nature, location, and/or intensity of pain. These
impairments combined with subjective assessment of pain by
health professionals and carers are primarily responsible for
the failure to identify pain in this group. This then leads
to adverse health outcomes, such as behavioral disturbances,
use of inappropriate medications, poor quality of life, and
premature death (Schneider et al., 2005; Holmes et al., 2008).
From the carer’s perspective, care is less burdensome when pain
related-behavioral problems are well managed. There are also
many potential benefits from proper pain management to the
organization, which include staff efficiency and productivity, staff
retention, and better quality of care for their patients.
The PainChekTM System
The PainChekTM system is a software system which is comprised
of the following components:
a) Mobile Application (App)
b) Web Admin Portal (WAP)
PainChekTM is intended to be used to assess and monitor pain
in people who cannot verbalize such as people with dementia or
communication difficulties (ARTG, 2017).
A glossary of terminology used to describe the psychometric
and clinimetric properties of pain assessment tools is displayed
in Table 2. A comprehensive account of the clinical and technical
characteristics of the PainChekTM system is described in Table 3.
The PainChekTM App
PainChekTM is a point of care software application (App) that is
compatible with Android and iOS smart devices. The tool uses
automated facial recognition technology in real time to identify
nine facial micro-expressions called action units (AUs), which are
derived from the FACS (Atee et al., 2017a). These facial AUs are
validated indicators of the presence of pain (Prkachin, 1992, 2009;
Prkachin and Solomon, 2008; Craig et al., 2011; Kunz, 2014; Kunz
and Lautenbacher, 2014). These data are then combined with
other non-facial pain cues (also known as communicative and
protective pain behaviors) such as vocalizations, movements, and
behaviors inputted by the user to automatically calculate a pain
severity score (Atee et al., 2017a). The App includes a number of
components, which are outlined in Table 3.
The PainChekTM pain scale is composed of 42 items
distributed across six domains (Table 3;Atee et al., 2017a). Using
Frontiers in Aging Neuroscience | 3June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
TABLE 1 | Glossary of technical terms used.
Concept Definition
Cognification The process of making objects smarter by combining, connecting and/or integrating 2 or more technologies; one of which is AI (Kelly, 2016).
Also known as “artificial smartness”.
Artificial Intelligence (AI) “The scientific understanding of the mechanisms underlying thought and intelligent behavior and their embodiment in machines” (AAAI,
Smart computing A generation of integrated hardware, software, and network technologies that provide IT systems with real-time awareness of the real world
and advanced analytics to help people make more intelligent decisions about alternatives and actions that will optimize processes (Bartels
et al., 2009).
Cloud computing A model for enabling ubiquitous, convenient, on-demand network access to a shared pool of configurable computing resources (e.g.,
networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or
service provider interaction (Mell and Grance, 2011).
Affective computing Computing that relates to, arises from, or influences emotions (Picard, 1997).
Internet of Things (IoT) The networked interconnection of everyday objects, which are often equipped with ubiquitous intelligence. Also known as “Internet of
Objects” (Xia et al., 2012).
Deep learning A pattern recognition technique that allows computational models that are composed of multiple processing layers to learn representations of
data with multiple levels of abstraction (LeCun et al., 2015).
Smart device An electronic or digital mobile device that has advanced computational processing power, possess multiple capabilities (e.g., voice and video
communication, data storage), can operate independently and interactively by being linked to other devices or networks via various wireless
connections e.g., Wi-Fi, Bluetooth (Bamidis et al., 2015).
Android A mobile operating system designed primarily for touchscreen devices such as smartphone and tablet computers, and for other electronics
such as smart televisions (Android TV), and smart watches (Bamidis et al., 2015).
iOS A mobile operating system developed by Apple, which works in a similar way to the Android.
TABLE 2 | Glossary of terms used to describe psychometric and clinimetric properties of pain assessment tools.
Concept Definition
Validity The degree to which an instrument measures what it is intended to measure (Polit and Hungler, 1991).
Concurrent validity The degree to which scores on an instrument are correlated with some external criterion, measured at the same time (Polit and Hungler,
Discriminant validity An approach to construct validation that involves assessing the degree to which a single method of measuring two constructs yields
different results (i.e., discriminates the two; Polit and Hungler, 1991).
Predictive validity The degree to which an instrument can predict some criterion observed at a future time (Polit and Hungler, 1991).
Reliability The degree of consistency or dependability (i.e., repeatability) with which an instrument measures the attribute it is designed to measure.
Interrater reliability The degree to which two raters or observers, operating independently, assign the same ratings or values for an attribute being measured
(Polit and Hungler, 1991).
Test-retest reliability A procedure used to determine the stability of measurements over time (Waltz et al., 1991).
Internal consistency The degree to which two or more measures are essentially measuring the same construct (Portney and Mary, 2009).
Sensitivity (SE) Probability that a test result will be positive when the disease is present (true positive rate; Altman et al., 2000).
Specificity (SP) Probability that a test result will be negative when the disease is not present (true negative rate; Altman et al., 2000).
Accuracy Overall probability that a patient will be correctly classified (Altman et al., 2000).
Clinical utility The usefulness of the measure for decision making (van Herk et al., 2007).
Clinical Utility Index (CUI) The overall value of a test for combined screening and case finding (Mitchell, 2010).
the smart-device camera to capture a short video of a person’s
face, the App automatically identifies the face in real time, then
maps the face to analyze facial expressions (using a built-in AI
algorithm) indicative of the presence of pain. This step provides
a score for Domain 1. The user of the App then completes the
checklists in Domains 2–6, to give rise a numerical pain score
which fits one of the following categories: no pain (0–6), mild
pain, (7–11), moderate pain (12–15), or severe pain (16) (Atee
et al., 2017a).
The PainChekTM App (Figures 2A–M) is commercially
available through PainChek Ltd, and is demonstrated in
this animated video:
The Web Administration Portal (WAP)
The PainChekTM Web Administration Portal is a secure website
that allows administrators to manage patient data, and activate
new users. The WAP is a cloud hosted web application that can be
accessed via a dedicated URL using any of the following internet
browsers: Chrome (version 59.0 or later), Mozilla (version 54.0
or later), Internet Explorer (version 11 or later). The WAP
of PainChekTM is supported through the operating systems of
Windows (7 or later), or Macintosh (OS X Mavericks 10.9
or later). The portal is currently hosted on Amazon Webs
Services using the Amazon Elastic Compute Cloud (Amazon
EC2) (AWS, 2017). Figure 3 illustrates a screenshot of the WAP.
Characteristics of the WAP are also summarized in Table 3.
Frontiers in Aging Neuroscience | 4June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
TABLE 3 | Clinical and technical characteristics of PainChekTM system.
Smart device enabled application A point of care mobile application, which consists of:
Pain scale (Figures 2A–F)
Pain assessment log (Figure 2J)
Pain chart (Figure 2K)
Local patient database
Medications and therapies log
Comments section (Figure 2M).
Web administration portal (Figure 3) A secure website that allows the management of patients and users data
Smart device A smart phone or tablet to deliver point of care pain assessments, and to capture temporal patterns of
pain scores
PC or smart device A computing device for WAP access
Operation system of the App Android or iOS
Operation procedure 1) Download and Install the App
2) Log in and set up user profile
3) Enter details for a new patient or select an existing patient.
Time to set up the App The iOS PainChek App is a 55 MB download. Assuming a download speed of 40 Mbs the average
speed of a mobile connection in Australia as of mid-2017 (as reported by,
the PainChek App should take around 10 to 15s to download.
Administration skills 1) Familiarity with the use of smart device
2) Familiarity with the patient undergoing assessment
3) Basic knowledge of pain behaviors in dementia.
Target users Clinicians and carers
Training needs 1) User competence on the use of smart device technology and operation of the App
2) Clinical competence on tool’s contents, domains, and descriptors.
Training resources 1) Video tutorials accessible through the PainChek website
2) FAQs (text and illustrating pictures) accessible through the website
3) Face-to-face workshop for enterprise users.
All materials are currently available in English but other languages are planned.
Pain scale (Figures 2A–F) 42 items distributed across 6 domains:
The Face (9 items), The Voice (9 items), The Movement (7 items), The Behavior (7 items), The Activity
(4 items), The Body (6 items)
Pain chart (Figure 2K) A graphical representation of pain scores over a period of time
Pain assessment log (Figure 2J) A list of pain assessments completed with their corresponding time and dates
Patient database A local repository of patients’ data including demographics
Medications and therapies log A local repository of medications and therapies of each patient
Front camera mode Automated facial analysis using the front camera of a smart device
Back camera mode Automated facial analysis using the back camera of a smart device
Manual mode Manually completed facial assessment (optional)
Scoring format Binary (yes/no) checklist
Scoring instructions 1) Observe the patient
2) Use the AFRA in the Face domain to detect facial action unit descriptors
3) Complete the corresponding checklists for the remaining non-facial domains
4) The App automatically calculates a pain intensity score, which conforms to one of the pain category
bands below.
Scoring interpretation (total pain scores) 0-6 (No Pain), 7-11 (Mild Pain), 12-15 (Moderate Pain), 16 (Severe Pain)
Ideal conditions of pain assessments 1) Assess pain at rest (e.g. sitting) and immediately after movement (e.g. repositioning)
2) Assess and re-assess (e.g., 1 h post-intervention).
Time to complete scoring of total scale 1 min
Time to complete scoring of the Face domain (automated) 3 s
Study 1 (Atee et al., 2017a) Design: prospective observational study; Setting: RACFs; Sampling: purposive convenience; Time line:
13 weeks, N: 40; Age: 60–98 years
Frontiers in Aging Neuroscience | 5June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
TABLE 3 | Continued
Study 2 (Atee et al., 2017b) Design: prospective observational study; Setting: RACFs; Sampling: purposive convenience; Time line:
10 weeks; N: 34; Age: 68–93 years
Study 3 (Hoti et al., 2018) Design: post-hoc statistical analyses based on Study 2 findings
Concurrent validity Excellent
Study 1: r =0.882 (95% CI: 0.857-0.903)
Study 2: r =0.911 (95% CI: 0.893-0.927)
Discriminant validity Good (regression model not significantly influenced by the timing of the assessment i.e. at rest vs. with
Study 1: p=0.795
Study 2: p=0.243
Internal consistency reliability Excellent homogeneity
Study 1: α=0.925
Study 2: α=0.950
Inter-rater reliability Good-to-excellent
Study 1: κw=0.74 (95% CI: 0.69-0.80)
Study 2: κw=0.86 (95% CI: 0.82-0.90)
Test-retest reliability Excellent
Study 2: ICC =0.904 (95% CI: 0.885-0.921)
Predictive validity Good based on the following data:
Study 2
Preintervention pain scores (i.e. at rest)
Mean =8.33 ±3.34; Median =9; Mode =10
Postintervention pain scores (i.e. post movement)
Mean =11.44 ±3.54; Median =11; Mode =13
t-test: p<0.0001
Clinical utility Excellent based on the following data:
Study 1
Contingency Table
Pain categories were derived using a contingency table with the APS. Pain intensity scores include 4
categories: no pain (0-6), mild pain (7-11), moderate pain (12-15), severe pain (16)
Study 3
Mitchell’s Clinical Utility Index
CUI (+)=0.936 (95% CI: 0.911-0.960)
CUI ()=0.801 (95% CI: 0.748-0.854)
CUI =0.95
ROC Curve
AUC =0.98 (95% CI: 0.96–0.99)
Optimal cut-off for pain =7
Accuracy Excellent based on the following data:
Study 3
Accuracy =95.0% (95% CI: 92.9%-97.1%)
SE =96.1% (95% CI: 93.9%-98.3%)
SP =91.4% (95% CI: 85.7%-97.1%)
Browser compatibility Chrome (version 59.0 or later), Mozilla (version 54.0 or later), Internet Explorer (version 11 or later)
Operating system Windows (7 or later), or Macintosh (OS X Mavericks 10.9 or later)
Data hosting product Amazon Elastic Compute Cloud (Amazon EC2) (AWS, 2017)
N, number of subjects with moderate to severe dementia; RACFs, residential aged care facilities; r, Pearson’s correlation coefficient; α, Cronbach alpha; κw, Weighted Kappa; ICC,
Intraclass correlation coefficient; CUI, Clinical Utility Index; AUC, Area Under Curve (of receiver-operator characteristic curve); ROC, receiver-operator characteristic curve; SE, Sensitivity;
SP, Specificity.
Clinical Studies
To date, three studies about the PainChekTM App have been
published (Table 3;Atee et al., 2017a,b; Hoti et al., 2018). In
blind comparisons with the Abbey Pain Scale, PainChekTM has
been clinically evaluated in aged care residents with moderate to
severe dementia in two prospective observational studies (Atee
et al., 2017a,b). The third study provided a comprehensive
clinimetric analysis on the performance of the App
(Hoti et al., 2018).
Study 1: Pain Assessment in Dementia: Evaluation of a Point-
of-Care Technological Solution (Atee et al., 2017a).
In this study, Atee el al. provided an account of the
description, content, and conceptual synthesis as well as
the psychometric properties of the PainChekTM. The App
Frontiers in Aging Neuroscience | 6June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
FIGURE 2 | (A) PainChekTM pain assessment tool-The Face (Domain 1). (B) PainChekTM pain assessment tool-The Voice (Domain 2). (C) PainChekTM pain
assessment tool-The Movement (Domain 3). (D) PainChekTM pain assessment tool-The Behavior (Domain 4). (E) PainChekTM pain assessment tool-The Activity
(Domain 5). (F) PainChekTM pain assessment tool-The Body (Domain 6). (G) PainChekTM pain assessment tool-Summary screen. (H) PainChekTM pain assessment
tool–Saving assessment. (I) PainChekTM App—“Dashboard” screen. (J) PainChekTM App—“Assessments” log. (K) PainChekTM App—“Pain Chart.” (L) PainChekTM
App—“Pain Relief” list. (M) PainChekTM App—“Comments” section.
was tested in 40 residents, who underwent 353 paired
assessments during rest (n=209) and movement (n=144).
PainChekTM was demonstrated to have excellent concurrent
validity and internal consistency, together with good interrater
reliability and discriminant validity (Table 3;Atee et al.,
Study 2: Psychometric evaluation of the Electronic Pain
Assessment Tool (ePAT): An Innovative Instrument for Individuals
With Moderate-to-Severe Dementia (Atee et al., 2017b).
Based on 400 paired assessments, the psychometric properties
of the tool were further examined in 34 geriatric residents. Again,
the App demonstrated strong psychometric properties: excellent
Frontiers in Aging Neuroscience | 7June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
FIGURE 3 | PainChekTM Web Admin Portal (WAP).
concurrent validity, interrater reliability, internal consistency,
and excellent test-retest reliability. Discriminant validity and
predictive validity were both good (Table 3) (Atee et al., 2017b).
Study 3: Clinimetric Properties of the Electronic Pain
Assessment Tool (ePAT) for Aged-Care Residents With Moderate
to Severe Dementia (Hoti et al., 2018).
Hoti et al. further analyzed Study 2 data to confirm the cut-off
scores, and predictive validity of the tool, whilst also reporting on
its clinical utility. Using the ROC curve methodology, the cut-
off points for presence of pain was confirmed to be 7. The
study demonstrated the high accuracy, sensitivity and specificity
of the App in detecting pain in individuals with dementia. It
also demonstrated the excellent clinical utility of the App for
pain screening and case finding, as indicated by Mitchell’s Index
(Table 3) (Hoti et al., 2018).
Clinical Guide and Training on the Use of
A wide variety of training resources have been developed to
assist users with the operation of PainChekTM. Resources include
face-to-face training, and web-based materials.
The face-to-face training is available for institutional users.
The training is a 2.5 h program which comprises of four sessions:
(1) pain associated behaviors in people with dementia, (2)
pain assessment in people with dementia, (3) PainChekTM (e.g.,
contents, scoring, and administration), and (4) practical training
and clinical shadowing. The program consists of video and
written materials, which have been prepared or collated by
clinical researchers (who developed the PainChekTM system, and
experienced in the area of pain and dementia). The video clips
cover a wide range of pain relevant materials including AUs
typically displayed during pain (simulated). Similarly, a number
of facial images of people in pain and clinical cases are also
used during the training. A two-page clinical guide handout
(Figure 4), explaining the content, domains, corresponding
descriptors and recommended time frame of observation is also
available to the trained users. A clinical facilitator experienced
on the use of the system is responsible for running the training
For each patient, it is essential that pain assessments are
carried out during rest first. This is in order to determine baseline
scores for benchmarking purposes, which then followed by post
movement assessment to capture the nociceptive experience.
Kinesthetic activities such as bending and walking are part
of daily living. These activities involve joint activity, which
generates nociceptive signals, encoding the sensory aspect of pain
phenomenon (Breivik et al., 2008).
The web-based resources (
include user guides, video materials, and frequently asked
questions. All training resources are currently available in English
language, although translation to other languages is planned.
This article reports on an innovative pain assessment system that
includes a point of care App and WAP. The PainChekTM App
(Figure 2) is a newly TGA-approved and CE marked medical
device for pain assessment in non-verbal adult populations
(ARTG, 2017). The App is linked to WAP (Figure 3) to allow
capturing of data collected in the clinical setting.
The conceptual model around which the PainChekTM
was developed is multifaceted (Figure 1). This model
Frontiers in Aging Neuroscience | 8June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
FIGURE 4 | “Clinical Guide” handout.
entails subjectivity, multi-dimensionality, and dynamicity
of pain as a construct, objectivity, and comprehensiveness
of pain behaviors included in the tool, simple scoring,
and administration procedure of the pain scale, as well as
technological advancements and innovative characteristics of the
platform used. Of particular importance, AFRA were used for
the first time in a smart device enabled tool for pain assessment
targeted at people with dementia. This novel design (including
the AI-assisted scoring) facilitates the process of pain assessment
and allows pain scores to be obtained in a less subjective way.
Frontiers in Aging Neuroscience | 9June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
The system has password-enabled authentication to protect and
secure the data. Further, the binary scoring of pain scale reduces
the likelihood of user judgement bias when recording non-facial
items (Ridley, 2002). It also allows easier quantification of
patient’s characteristics in terms of pain which in turn facilitates
the prognostication of pain severity. This allows for conversion
of probability (present or absent) into a binary outcome which
informs the tool’s ordinal scale of pain severity (mild, moderate,
severe; Ridley, 2002). The latter is arrived through automated
summation of the total pain score (Figure 2G). Because
PainChekTM is a digital tool linked to the WAP, the technology
overcomes the limitations of guessing practices and poor
identification of the presence of pain by nurses and geriatricians
in people with severe cognitive impairment (Kovach et al.,
2000; Cohen-Mansfield and Lipson, 2002). Through systemizing
and structuring the process of assessment, PainChekTM will
overcome these challenges and improve multidisciplinary
communication among health care professionals. For its use in
clinical practice, the App does not require Internet connection.
Synchronizing the clinical data (collected through the App) with
the WAP even at a later time still preserves the actual time of
pain assessment as the App saves each assessment in real time
(Figures 2H,J). Electronic documentation of the system has the
capacity to allow patient profiling using their corresponding
pain assessment (Figure 2J) and management logs (Figure 2L).
The ability of the system to collect a large amount of pain data
over a period of time helps in identifying temporal patterns
(Figure 2K), which can offer useful clinical insights in terms
of patients’ responsiveness to interventions provided during a
specified period. PainChekTM also facilitates the interdisciplinary
communication among health care practitioners through
providing a more comprehensive and up-to-date picture of
patient’s pain. This approach was also suggested by Lichtner et al.
for decision support systems for pain management in patients
with dementia (Lichtner et al., 2015, 2016).
A primary deficiency in the development of majority of the
health related apps is the fact that they are designed by non-
health care practitioners (Lalloo et al., 2015). In contrast, our
system was developed by clinical researchers in collaboration
with AI and web interface engineers (Atee et al., 2017a). In
addition, most existing health apps including those related to
pain, lack evidence in their development and testing (Rosser
and Eccleston, 2011), whereas the PainChekTM App has been
conceptualized, and clinically tested around the best available
evidence in gerontology, pain, and dementia (Atee et al.,
2017a,b). Further, the PainChekTM App is to our knowledge
the only pain assessment tool in dementia that has regulatory
clearance in Australia (TGA) and Europe (CE mark), as a
medical device (ARTG, 2017). Through clinical studies, we have
also demonstrated that our approach in developing a hybrid
model (automated FACS pain relevant items and other clinical
indicators related to older adults with dementia) is a valid and
reliable method in evaluating pain (Atee et al., 2017a,b).
Our conceptualized model design was also supported by the
literature (Herr and Garand, 2001; AGS Panel on Persistent
Pain in Older Persons, 2002; Herr, 2002; Herr et al., 2006,
2011; Hadjistavropoulos et al., 2007, 2014; Beach et al., 2016).
Lints-Martindale et al. (2012) found that the AGS-recommended
pain behavioral domains are comprehensive and useful indicators
in recognizing painful episodes. Beach et al. (2016) reported that
integrating objective behavioral descriptors into observational
tools improve pain assessment in people with Alzheimer’s
dementia (AD). Defined as “Actions or postural displays that
are enacted during the experience of pain,” pain behaviors are
important manifestations that convey rich information about
patient’s pain severity (Hadjistavropoulos and Craig, 2002; Herr
et al., 2017). Of these behaviors, non-verbal expressions of pain
such as facial expressions, vocalizations, and body movements are
generally difficult to suppress (Martel et al., 2012). Of note, facial
expressions of pain have been widely researched because they
are “readily accessible, highly plastic, and are believed to be the
most specific, encodable form of pain behavior” (Williams, 2002).
Facial expressions are also one of the strongest indicators of pain
particularly in people with cognitive impairment or dementia
(Kunz et al., 2007, 2009). There is a significant increase of pain
behaviors in AD compared to healthy control (Lautenbacher
et al., 2013; Beach et al., 2017). Horgas et al. (2009) indicated that
the resultant numerical scores from summating pain behaviors
are closely linked to the self-report of pain. In their meta-analysis
Labus et al. (2003), also found that the use of multiple behavioral
domains has a synergistic effect on pain assessment because
the obtained scores are more representative of subjective pain
experience. McCahon et al. (2005) noted that observation of
pain behaviors is a valid and reliable assessment method for
use with patients with chronic pain. Although these findings
were drawn from samples of cognitively intact individuals with
chronic pain, similar trends were also observed in people with
cognitive impairments. A recent review of pain behaviors in
people with dementia by Herr et al. (2011, 2017) revealed
that these behaviors are strong indicators of the presence and
intensity of pain. Further, these pain behaviors configure the
item descriptors list of observational pain assessment tools in
verbal and non-verbal geriatric populations (Herr et al., 2011,
2017; Lichtner et al., 2014). In older adults with dementia, pain
behavior tools improved binary pain recognition (i.e., presence
or absence of pain) by up to 25.4%, and ordinal level of pain
intensity by up to 42.5% above chance (Lukas et al., 2013). It
is thus evident from the above that there is a consensus in the
literature about the predictability of pain related behaviors in
informing the assessment of clinical pain.
The App has been designed for administration by a wide
range of users including clinicians and carers (Atee et al., 2017a).
The training resources are comprehensive and diversified. The
tool is available as a smart device App compatible with various
mobile operating systems such as Android and iOS (Atee et al.,
2017a,b). Thus, these useful and unique characteristics cover
multiple aspects of clinical utility, such as scoring, administration
time and skills, and supporting materials which most of the other
tools are currently missing (Lichtner et al., 2014; Herr et al.,
In conclusion, evidence to date suggests that the PainChekTM
system offers a novel method that should make the process of
pain assessment and monitoring simpler and more objective for
clinicians and carers of patients who cannot verbalize their pain.
Frontiers in Aging Neuroscience | 10 June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
The PainChekTM system and materials reported in this
manuscript are commercially available from PainChek Ltd,
Sydney, NSW, Australia.
All relevant data is contained within the manuscript.
No additional datasets were generated for this
manuscript because the submitted work is a technology
All clinical studies of PainChekTM that involved human subjects
reported in this manuscript were conducted in accordance
with the World Medical Association’s Declaration of Helsinki,
the recommendations and policy statements of the Alzheimer’s
Association and the World Health Organization on assistive
technologies for people with dementia. These studies were
approved by the Human Research Ethics Committee of
Curtin University, Western Australia (HREC: HR10/2014),
and the participating aged care groups. Study 2 was also
registered with Therapeutic Goods Administration (TGA)
under the Clinical Trial Notification (CTN) Scheme (CT-
2016-CTN-04886-1 v1). Due to cognitive impairment of the
involved subjects, proxies gave authorized written informed
consent on their behalf. All data were de-identified to protect
MA, KH, and JH: conceived the idea; MA: drafted and organized
the manuscript, and conducted the literature search. All authors
reviewed, contributed to and approved the final version of the
The authors would like to acknowledge the contribution of an
Australian Government Research Training Program Scholarship
in supporting this research. The original research that led to
the development of the PainChekTM tool is part of a Ph.D.
project which was also supported by the Dementia Australia
Research Foundation (DARF) through grant funding and a
stipend scholarship. The content of the article is solely the
responsibility of the authors and does not necessarily represent
the official views of DARF. The project has been commercialized
into a spin-off start-up company (ePAT Pty Ltd), which has
been publicly listed as PainChek Ltd in the Australian Share
Securities (ASX) since October 2016. The current manuscript is
also sponsored by PainChek Ltd. The funding bodies had no role
in writing the manuscript.
Sincere thanks goes to Scott Robertson from PainChek Ltd for
providing some technical details about the PainChekTM system,
and gratitude to Mustafa Al Abbasi for assisting with placing the
images on the manuscript.
AAAI (1995-2013). Association for the Advancement of Artificial Intelligence
(AAAI). AI Overview. AAAI. Available online at:
ADI (2017). Alzheimer’s Disease International (ADI) (2017). About Dementia.
London: ADI [10/01/2018]. Available online at:
American Geriatrics Society (AGS) Panel on Persistent Pain in Older Persons.
(2002). The management of persistent pain in older persons. J. Am. Geriatr.
Soc. 50, S205–S224. doi: 10.1046/j.1532-5415.50.6s.1.x
ARTG (2017). Australian Government, Department of Health, Therapeutic
Goods Administration, eBS Australian Register of Therapeutic Goods (ARTG)
Medicines. Public Summary: ePAT Technologies Ltd- Information system
software, application program. Available online at: https://www.ebs.tga.
AWS (2017). Amazon Web Services (AWS). Amazon EC2: Secure and Resizable
Compute Capacity in the Cloud. AWS. Available online at:
Altman, D. G., Machin, D., Bryant, T. N., and Gardner,M. J. (eds.). (2000). Statistics
with Confidence, 2nd Edn. London: BMJ Books.
Lints-Martindale, A. C., Hadjistavropoulos, T., Lix, L. M., and Thorpe, L.
(2012). a comparative investigation of observational pain assessment
tools for older adults with dementia. Clin. J. Pain 28, 226–237.
doi: 10.1097/AJP.0b013e3182290d90
Atee, M., Hoti, K., and Hughes, J. D. (2017b). Psychometric evaluation
of the electronic Pain Assessment Tool (ePAT): an innovative
instrument for individuals with moderate to severe dementia.
Dement. Geriatr. Cogn. Disord. 44, 256–267. doi: 10.1159/0004
Atee, M., Hoti, K., Parsons, R., and Hughes, J. D. (2017a). Pain assessment in
dementia: evaluation of a point-of-care technological solution. J. Alzheimers
Dis. 60, 137–150. doi: 10.3233/JAD-170375
Bamidis, P., Tarnanas, I., Hadjileontiadis, L., and Tsolaki, M. (2015). Handbook
of Research on Innovations in the Diagnosis and Treatment of Dementia.
Hershy, PA: Medical Information Science Reference (an imprint of IGI
Bartels, A. H., Daley, E., Parker, A., Evelson, B., and Muteba, C. (2009). Smart
Computing Drives the New Era of IT Growth. Cambridge: UK Forrester
Beach, P. A., Huck, J. T., Miranda, M. M., Foley, K. T., and Bozoki, A. C. (2016).
Effects of Alzheimer Disease on the facial expression of pain. Clin. J. Pain 32,
478–487. doi: 10.1097/AJP.0000000000000302
Beach, P. A., Huck, J. T., Zhu, D. C., and Bozoki, A. C. (2017). Altered
behavioral and autonomic pain responses in Alzheimer’s Disease are associated
with dysfunctional affective, self-reflective and salience network resting-state
connectivity. Front. Aging Neurosci. 9:297. doi: 10.3389/fnagi.2017.00297
Breivik, H., Borchgrevink, P. C., Allen, S. M., Rosseland, L. A., Romundstad,
L., Breivik Hals, E. K., et al. (2008). Assessment of pain. BJA 101, 17–24.
doi: 10.1093/bja/aen103
Cohen-Mansfield, J., and Lipson, S. (2002). Pain in cognitively impaired nursing
home residents: how well are physicians diagnosing it? J. Am. Geriatr. Soc. 50,
1039–1044. doi: 10.1046/j.1532-5415.2002.50258.x
Craig, K. D., Prkachin, K. M., and Grunau, R. V. E. (2011). “The facial expression of
pain,” in Handbook of Pain Assessment, 3rd Edn, eds D. C. Turk and R. Melzack
(New York, NY: The Guilford Press), 117–133.
Frontiers in Aging Neuroscience | 11 June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
Duthey, B. (2013). Background Paper 6.11: Alzheimer Disease and Other Dementias.
A Public Health Approach to Innovation, Update on 2004 Background Paper.
Available online at:
Ekman, P., Friesen, W. V., and Hager, J. (1978). The Facial Action Coding System
(FACS): A Technique for the Measurement of Facial Action. Palo Alto, CA:
Consulting Psychologists Press.
Flaherty, S. A. (1996). Pain measurement tools for clinical practice and research. J.
Am. Assoc. Nurse Anesth. 64, 133–140.
Hadjistavropoulos, T., and Craig, K. D. (2002). A theoretical framework
for understanding self-report and observational measures of
pain: a communications model. Behav. Res. Ther. 40, 551–570.
doi: 10.1016/S0005-7967(01)00072-9
Hadjistavropoulos, T., Herr, K., Prkachin, K. M., Craig, K. D., Gibson, S. J., Lukas,
A., et al. (2014). Pain assessment in elderly adults with dementia. Lancet. Neurol.
13, 1216–1227. doi: 10.1016/S1474-4422(14)70103-6
Hadjistavropoulos, T., Herr, K., Turk, D. C., Fine, P. G., Dworkin, R. H.,
Helme, R., et al. (2007). An interdisciplinary expert consensus statement
on assessment of pain in older persons. Clin. J. Pain 23, S1–S43.
doi: 10.1097/AJP.0b013e31802be869
Herr, K. (2002). Pain assessment in cognitively impaired older adults. Am. J. Nurs.
102, 65–67. doi: 10.1097/00000446-200212000-00020
Herr, K. A., and Garand, L. (2001). Assessment and measurement of pain in older
adults. Clin. Geriatr. Med. 17, 457–478. doi: 10.1016/S0749-0690(05)70080-X
Herr, K., Bjoro, K., and Decker, S. (2006). Tools for assessment of pain in nonverbal
older adults with dementia: a state-of-the-science review. J. Pain Symptom.
Manage. 31, 170–192. doi: 10.1016/j.jpainsymman.2005.07.001
Herr, K., Coyne, P. J., McCaffery, M., Manworren, R., and Merkel, S. (2011).
Pain assessment in the patient unable to self-report: position statement
with clinical practice recommendations. Pain Manage. Nurs. 12, 230–250.
doi: 10.1016/j.pmn.2011.10.002
Herr, K., Zwakhalen, S., and Swafford, K. (2017). Observation of pain in dementia.
Curr. Alzheimer Res. 14, 486–500. doi: 10.2174/15672050136661606022
Holmes, H. M., Sachs, G. A., Shega, J. W., Hougham, G. W., Hayley, D., and
Dale, W. (2008). Integrating palliative medicine into the care of persons with
advanced dementia: identifying appropriate medication use. J. Am. Geriatr. Soc.
56, 1306–1311. doi: 10.1111/j.1532-5415.2008.01741.x
Horgas, A. L., Elliott, A. F., and Marsiske, M. (2009). Pain assessment in persons
with dementia: relationship between self-report and behavioral observation. J.
Am. Geriatr. Soc. 57, 126–132. doi: 10.1111/j.1532-5415.2008.02071.x
Hoti, K., Atee, M., and Hughes, J. D. (2018). Clinimetric properties of the electronic
Pain Assessment Tool (ePAT) for aged-care residents with moderate to severe
dementia. J. Pain Res. 11, 1037–1044. doi: 10.2147/JPR.S158793
Kelly, K. (2016). The Inevitable: Understanding the 12 Technological Forces thatWill
Shape Our Future. New York, NY: Viking.
Kovach, C. R., Griffie, J., Muchka, S., Noonan, P. E., and Weissman, D. E.
(2000). Nurses’ perceptions of pain assessment and treatment in the cognitively
impaired elderly. It’s not a guessing game. Clin. Nurse Spec. 14, 215–220.
doi: 10.1097/00002800-200009000-00011
Kunz, M. (2014). “Behavioural/facial markers of pain, emotion, cognition,” in Pain,
Emotion and Cognition: A Complex Nexus, eds G. Pickering, and S. Gibson
(Clermont-Ferrand: Springer), 123–33.
Kunz, M., and Lautenbacher, S. (2014). The faces of pain: a cluster analysis of
individual differences in facial activity patterns of pain. Eur. J. Pain 18, 813–823.
doi: 10.1002/j.1532-2149.2013.00421.x
Kunz, M., Mylius, V., Schepelmann, K., and Lautenbacher, S. (2009). Effects of age
and mild cognitive impairment on the pain response system. Gerontology 55,
674–682. doi: 10.1159/000235719
Kunz, M., Scharmann, S., Hemmeter, U., Schepelmann, K., and Lautenbacher,
S. (2007). The facial expression of pain in patients with dementia. Pain 133,
221–228. doi: 10.1016/j.pain.2007.09.007
Labus, J. S., Keefe, F. J., and Jensen, M. P. (2003). Self-reports of pain intensity
and direct observations of pain behavior: when are they correlated? Pain 102,
109–124. doi: 10.1016/s0304-3959(02)00354-8
Lalloo, C., Jibb, L. A., Rivera, J., Agarwal, A., and Stinson, J. N. (2015). “There’s
a Pain App for That” review of patient-targeted smartphone applications for
pain management. Clin. J. Pain 31, 557–563. doi: 10.1097/AJP.00000000000
Lautenbacher, S., Niewelt, B. G., and Kunz, M. (2013). Decoding pain from
the facial display of patients with dementia: a comparison of professional
and nonprofessional observers. Pain Med. 14, 469–477. doi: 10.1111/pme.
LeCun, Y., Bengio, Y., and Hinton, G. (2015). Deep learning. Nature 521, 436–444.
doi: 10.1038/nature14539
Lichtner, V., Dowding, D., Allcock, N., Keady, J., Sampson, E. L., Briggs, M.,
et al. (2016). The assessment and management of pain in patients with
dementia in hospital settings: a multi-case exploratory study from a decision
making perspective. BMC Health Serv. Res. 16:427. doi: 10.1186/s12913-016-
Lichtner, V., Dowding, D., and Closs, S. J. (2015). The relative meaning of absolute
numbers: the case of pain intensity scores as decision support systems for pain
management of patients with dementia. BMC Med. Inform. Decis. Mak. 15:111.
doi: 10.1186/s12911-015-0233-8
Lichtner, V., Dowding, D., Esterhuizen, P., Closs, S. J., Long, A. F., Corbett,
A., et al. (2014). Pain assessment for people with dementia: a systematic
review of systematic reviews of pain assessment tools. BMC Geriatr. 14:138.
doi: 10.1186/1471-2318-14-138
Lukas, A., Barber, J. B., Johnson, P., and Gibson, S. J. (2013). Observer-rated
pain assessment instruments improve both the detection of pain and the
evaluation of pain intensity in people with dementia. Eur. J. Pain 17, 1558–1568.
doi: 10.1002/j.1532-2149.2013.00336.x
Martel, M. O., Wideman, T. H., and Sullivan, M. J. (2012). Patients who display
protective pain behaviors are viewed as less likable, less dependable, and
less likely to return to work. Pain 153, 843–849. doi: 10.1016/j.pain.2012.
McCahon, S., Strong, J., Sharrj, R., and Cramond, T. (2005). Self-report and
pain behavior among patients with chronic pain. Clin. J. Pain 21, 223–231.
doi: 10.1097/00002508-200505000-00005
Mell, P., and Grance, T. (2011). The NIST Definition of Cloud Computing:
Recommendations of the National Institute of Standards and Technology.
National Institute of Standards and Technology (NIST), Gaithersburg, MD:
US Department of Commerce, Contract No. NIST Special Publication,
Merskey, H., and Bogduk, N. (eds.) (1994). IASP Task Force on Taxonomy:
Classification of Chronic Pain: Description of Chronic Pain Syndromes and
Definition of Pain Terms. Seattle, WA: IASP Press.
Mitchell, A. J. (2010). “How do we know when a screening test is clinically useful?,
in Coyne Screening for Depression in Clinical Practice : an Evidence-based Guide
eds A. J. Mitchell and J. C. Coyne (Oxford; New York, NY: Oxford University
Press), 395.
Mitchell, S. L., Teno, J. M., Kiely, D. K., Shaffer, M. L., Jones, R. N., Prigerson, H.
G., et al. (2009). The clinical course of advanced dementia. New Engl. J Med.
361, 1529–1538. doi: 10.1056/NEJMoa0902234
Picard, R. W. (1997). Affective Computing. Cambridge: MIT press.
Polit, D. F., and Hungler, B. P. (1991). Nursing Research: Principals and Methods.
Philadelphia, PA: JB Lippincott.
Portney, L., and Mary, P. (2009). Foundations of Clinical Research: Applications to
Practice. Upper Saddle River, NJ: Pearson/Prentice Hall.
Prkachin, K. M. (1992). The consistency of facial expressions of pain: a
comparison across modalities. Pain 51, 297–306. doi: 10.1016/0304-3959(92)90
Prkachin, K. M. (2009). Assessing pain by facial expression: facial expression as
nexus. Pain Res. Manage. 14, 53–58. doi: 10.1155/2009/542964
Prkachin, K. M., and Solomon, P. E. (2008). The structure, reliability
and validity of pain expression: evidence from patients with
shoulder pain. Pain 139, 267–274. doi: 10.1016/j.pain.2008.
Ridley, S. A. (2002). Uncertainty and scoring systems. Anaesthesia 57, 761–767.
doi: 10.1046/j.1365-2044.2002.02619.x
Rosser, B. A., and Eccleston, C. (2011). Smartphone applications for pain
management. J. Telemed. Telecare 17, 308–312. doi: 10.1258/jtt.2011.101102
Schneider, L. S., Dagerman, K. S., and Insel, P. (2005). Risk of death with
atypical antipsychotic drug treatment for dementia - meta-analysis
Frontiers in Aging Neuroscience | 12 June 2018 | Volume 10 | Article 117
Atee et al. A Technical Note on the PainChekTM System
of randomized placebo-controlled trials. JAMA 294, 1934–1943.
doi: 10.1001/jama.294.15.1934
United Nations Department of Economic and Social Affairs Population Division
(2017). World Population Prospects: The 2017 Revision, Key Findings and
Advance Tables. New York, NY: United Nations.
van Herk, R., van Dijk, M., Baar, F. P. M., Tibboel, D., and de Wit,
R. (2007). Observation scales for pain assessment in older adults with
cognitive impairments or communication difficulties. Nurs. Res. 56, 34–43.
doi: 10.1097/00006199-200701000-00005
Waltz, C. F., Strickland, O. L., and Lenz, E. R. (1991). Measurement in Nursing
Research. Philadelphia, PA: F.A Davis.
Williams, A. C. (2002). Facial expression of pain: an evolutionary account. Behav.
Brain Sci. 25, 439–455. doi: 10.1017/S0140525X02000080
World Health Organisation (WHO) (2017). Dementia: Fact sheet. Geneva: WHO
Media Centre. Available online at:
fs362/en/ (Accessed January 10, 2018).
Xia, F., Yang, L. T., Wang, L., and Vinel, A. (2012). Editorial: internet of things. Int.
J. Commun. Syst. 25, 1101–1102. doi: 10.1002/dac.2417
Conflict of Interest Statement: All authors are shareholders in PainChek Ltd,
which is marketing the PainChekTM instrument. They also have a patent
application titled “A pain assessment method and system” (PCT/AU2015/000501),
which is currently under national phase examination since February 2, 2017. MA
is a Research Scientist for PainChek Ltd while serving as a Research Fellow and
Ph.D. Candidate with the School of Pharmacy and Biomedical Sciences, Curtin
University. KH is employed as a consultant by PainChek Ltd while serving as an
Assistant Professor at University of Pristina, and an Adjunct Senior Lecturer at
the School of Pharmacy and Biomedical Sciences, Curtin University. JH holds the
position of Chief Scientific Officer of PainChek Ltd while serving as a Professor at
the School of Pharmacy and Biomedical Sciences, Curtin University.
Copyright © 2018 Atee, Hoti and Hughes. This is an open-access article distributed
under the terms of the Creative Commons Attribution License (CC BY). The use,
distribution or reproduction in other forums is permitted, provided the original
author(s) and the copyright owner are credited and that the original publication
in this journal is cited, in accordance with accepted academic practice. No use,
distribution or reproduction is permitted which does not comply with these terms.
Frontiers in Aging Neuroscience | 13 June 2018 | Volume 10 | Article 117
... Currently, there is no published real-world data attached to any of the existing pain assessment tools, which makes it difficult to evaluate exactly what items are commonly encountered in those living with cognitive impairment, who are in pain. This lack of realworld assessment data can be addressed where data are collected digitally in clinical practice to be pooled and analyzed on a large scale through the use of a central electronic portal (18). Scoring mechanism systems of observational tools can also affect the profiling of pain behavior items. ...
... The majority of the tools have ordinal scoring, which makes the process of item profiling difficult or even insurmountable in some cases. In contrast, checklists and binary ratings of items (e.g., in PACSLAC-II and PainChek R ) allow an easier identification of pain behavior patterns including those related to facial expressions (11,18). ...
... PainChek R is the first pain observational assessment tool to utilize the combination of artificial intelligence (AI) and smart automation in a mobile application, allowing pain assessment at the point-of-care (18). ...
Full-text available
Pain is common in people living with dementia (PLWD), including those with limited verbal skills. Facial expressions are key behavioral indicators of the pain experience in this group. However, there is a lack of real-world studies to report the prevalence and associations of pain-relevant facial micro-expressions in PLWD. In this observational retrospective study, pain related facial features were studied in a sample of 3,144 PLWD [mean age 83.3 years (SD = 9.0); 59.0% female] using the Face domain of PainChek®, a point-of-care medical device application. Pain assessments were completed by 389 users from two national dementia-specific care programs and 34 Australian aged care homes. Our analysis focused on the frequency, distribution, and associations of facial action units [AU(s)] with respect to various pain intensity groups. A total of 22,194 pain assessments were completed. Of the AUs present, AU7 (eyelid tightening) was the most frequent facial expression (48.6%) detected, followed by AU43 (closing eyes; 42.9%) and AU6 (cheek raising; 42.1%) during severe pain. AU20 (horizontal mouth stretch) was the most predictive facial action of higher pain scores. Eye-related AUs (AU6, AU7, AU43) and brow-related AUs (AU4) were more common than mouth-related AUs (e.g., AU20, AU25) during higher pain intensities. No significant effect was found for age or gender. These findings offer further understanding of facial expressions during clinical pain in PLWD and confirm the usefulness of artificial intelligence (AI)-enabled real-time analysis of the face as part of the assessment of pain in aged care clinical practice.
... This instrument utilizes automated facial recognition technology to detect the presence of facial micro-expressions indicative of the presence of pain. Atee, Hoti, and Hughes (2018) assessed the psychometric properties of ePAT by applying it in noncommunicative patients with dementia. The scale obtained an excellent concurrent validity (r = .882, ...
... This review also identified three articles showing new ways to measure pain in non-communicative patients. The electronic Pain Assessment Tool (ePAT) and the Facial Action Coding System (FACS) both utilize automated facial recognition technology to detect the presence of facial micro-expressions indicative of the presence of pain (Atee et al., 2018;Browne et al., 2019). Another new way to detect pain is measuring pupillary diameter that varies under the influence of the sympathetic as well as parasympathetic systems, increasing proportionally with pain intensity (Charier et al., 2019). ...
Full-text available
The aim of this review was to investigate the tools used to identify pain in patients who are non-communicative and therefore unable to communicate pain to their caregivers. A systematic review of the studies published with predetermined eligibility criteria was undertaken. A total of 32 studies were identified and reviewed. Among the articles included, we observed two different types of outcome measures: physiological and behavioral measurements. Eleven articles referred to physiological measurements and 20 referred to behavioral measurements, with one article referring to both types of measurements. All the papers described at least one instrument or measure, which was implemented to measure pain in non-communicative patients. The results confirmed the agreement among behavioral measurements of pain, while physiological measurements should be studied more in depth. This review suggests that the use of frequent items in each scale and in each physiological instrument may help healthcare professionals and caregivers to overcome the barrier of prognosis uncertainty and to identify patients that could benefit from their use Life Span and Disability Canegallo V. et al. (2022)
... Pain was assessed using PainChek ® , a valid and reliable regulatory cleared pain assessment medical device in the form of a point-of-care app for people living with cognitive impairment including those living with dementia (Atee et al., 2017a(Atee et al., , 2018. The app has a 42-item pain scale that are distributed over six domains: Face (9 items), Voice, (9 items), Movement (7 items), Behaviour (7 items), Activity (4 items) and Body (6 items). ...
... PainChek ® scores above 6 (i.e. ⩾7) indicate the presence of pain (Atee et al., 2017a(Atee et al., , 2017b(Atee et al., , 2018. ...
Objective Younger-onset dementia accounts for about 5–10% of all dementias in Australia. Little data is available on neuropsychiatric symptoms in people with younger-onset dementia compared to those with older-onset dementia. This study aims to compare the types of neuropsychiatric symptoms and their clinico-demographic characteristics of people with younger-onset dementia and older-onset dementia who are referred to a specific dementia support service. Methods A 2-year retrospective observational cross-sectional analysis was undertaken on referrals with neuropsychiatric symptoms from Dementia Support Australia programmes. Neuropsychiatric symptoms were measured using the Neuropsychiatric Inventory total severity scores and distress scores. Contributing factors to neuropsychiatric symptoms for dementia groups were examined. Logistic regression was used to examine the relationship between individual neuropsychiatric symptoms and having older-onset dementia vs younger-onset dementia. Results Of the 15,952 referrals, about 5% ( n = 729, mean age: 60.7 years, standard deviation = 5.4) were individuals with younger-onset dementia. Referrals with older-onset dementia were more likely to be female (56%), whereas referrals with younger-onset dementia were more likely to be male (54%). There was a four times greater rate of frontotemporal dementia for those with younger-onset dementia (16.0%, n = 117) compared to those with older-onset dementia (2.8%, n = 427), χ ² (1) = 366.2, p < 0.001. Referrals with younger-onset dementia were more likely to be referred from community settings and those with older-onset dementia were more likely to be from residential aged care. Overall, there was no difference in the severity and distress of neuropsychiatric symptoms between the two groups. Contributing factors to neuropsychiatric symptoms were different between the groups, with pain being more frequently endorsed for individuals with older-onset dementia whereas communication difficulties were more commonly identified for those with younger-onset dementia. Conclusion Clinico-demographics of referrals with younger-onset dementia differ from those with older-onset dementia. There were some differences in the characteristics of neuropsychiatric symptoms between younger-onset dementia and older-onset dementia. Our findings have implications for service provision and support for people with dementia at different ages.
... PainChek Universal data will be stored in a repository within the PainChek secure cloud database. 31 Data will only be accessible by the research team members via a password-protected web administration portal. InterRAI data will be stored on a secure server at the University of Queensland accessible only by the research team via password. ...
Full-text available
Introduction Hospitalised older adults are prone to functional deterioration, which is more evident in frail older patients and can be further exacerbated by pain. Two interventions that have the potential to prevent progression of frailty and improve patient outcomes in hospitalised older adults but have yet to be subject to clinical trials are nurse-led volunteer support and technology-driven assessment of pain. Methods and analysis This single-centre, prospective, non-blinded, cluster randomised controlled trial will compare the efficacy of nurse-led volunteer support, technology-driven pain assessment and the combination of the two interventions to usual care for hospitalised older adults. Prior to commencing recruitment, the intervention and control conditions will be randomised across four wards. Recruitment will continue for 12 months. Data will be collected on admission, at discharge and at 30 days post discharge, with additional data collected during hospitalisation comprising records of pain assessment and volunteer support activity. The primary outcome of this study will be the change in frailty between both admission and discharge, and admission and 30 days, and secondary outcomes include length of stay, adverse events, discharge destination, quality of life, depression, cognitive function, functional independence, pain scores, pain management intervention (type and frequency) and unplanned 30-day readmissions. Stakeholder evaluation and an economic analysis of the interventions will also be conducted. Ethics and dissemination Ethical approval has been granted by Human Research Ethics Committees at Ramsay Health Care WA|SA (number: 2057) and Edith Cowan University (number: 2021-02210-SAUNDERS). The findings will be disseminated through conference presentations, peer-reviewed publications and social media. Trial registration number ACTRN12620001173987.
... Commercial tools for pain assessment informed by the existing literature on automated assessment have already been developed and marketed and there is every reason to believe that this trend will continue. For example, Painchek ( is a smartphone app-based device that combines a facial expression assessment component with input from five other domains (voice, movement, behavior, activity, body) to yield a pain score for application in geriatric and pediatric settings (92). It goes without saying that the development and marketing of tools for clinical assessment should be based on knowledge about automated assessment that is grounded in the empirical literature, consistent with the best-established technological solutions, has been subjected to rigorous validation procedures, and informed by understanding of issues of bias raised above. ...
Full-text available
Pain is often characterized as a fundamentally subjective phenomenon; however, all pain assessment reduces the experience to observables, with strengths and limitations. Most evidence about pain derives from observations of pain-related behavior. There has been considerable progress in articulating the properties of behavioral indices of pain; especially, but not exclusively those based on facial expression. An abundant literature shows that a limited subset of facial actions, with homologs in several non-human species, encode pain intensity across the lifespan. Unfortunately, acquiring such measures remains prohibitively impractical in many settings because it requires trained human observers and is laborious. The advent of the field of affective computing, which applies computer vision and machine learning (CVML) techniques to the recognition of behavior, raised the prospect that advanced technology might overcome some of the constraints limiting behavioral pain assessment in clinical and research settings. Studies have shown that it is indeed possible, through CVML, to develop systems that track facial expressions of pain. There has since been an explosion of research testing models for automated pain assessment. More recently, researchers have explored the feasibility of multimodal measurement of pain-related behaviors. Commercial products that purport to enable automatic, real-time measurement of pain expression have also appeared. Though progress has been made, this field remains in its infancy and there is risk of overpromising on what can be delivered. Insufficient adherence to conventional principles for developing valid measures and drawing appropriate generalizations to identifiable populations could lead to scientifically dubious and clinically risky claims. There is a particular need for the development of databases containing samples from various settings in which pain may or may not occur, meticulously annotated according to standards that would permit sharing, subject to international privacy standards. Researchers and users need to be sensitive to the limitations of the technology (for e.g., the potential reification of biases that are irrelevant to the assessment of pain) and its potentially problematic social implications.
... Also, there's potential to explore digital or gamelike apps to make pediatric self-measurements more engaging for children. Apps like electronic Pain Assessment Tool (ePAT) or PainChek® used in older adult patients, use AI technology for facial analysis to evaluate the presence and intensity of pain [41,42]. Such digital apps and game-like formats for PRO assessments will help children be more interested in providing measurements and improve the evaluation of symptoms that children struggle to verbalize or articulate. ...
Full-text available
Objective To provide an assessment of the quality of the most frequently used self-reported, generic patient-reported outcome measures (PROMs) that measure health-related quality of life (HRQoL) in children against the good research practices recommended by ISPOR task force for the pediatric population. Method Literature search was conducted on OvidSP database to identify the generic pediatric PROMs used in published clinical studies. The quality of PROMs used in more than ten clinical studies were descriptively evaluated against the ISPOR task force’s good research practices. Results Six PROMs were evaluated, namely Pediatric Quality-of-Life inventory 4.0 (PedsQL), Child Health Questionnaire (CHQ), KIDSCREEN, KINDL, DISABKIDS and Child Health and Illness Profile (CHIP). All PROMs, except KIDSCREEN, had versions for different age ranges. Domains of physical, social, emotional health and school activities were common across all the instruments, while domains of family activities, parent relations, independence, and self-esteem were not present in all. Children’s input was sought during the development process of PROMs. Likert scales were used in all the instruments, supplemented with faces (smileys) in instruments for children under 8 years. KIDSCREEN and DISABKIDS were developed in a European collaboration project considering the cross-cultural impact during development. Conclusion The comparison of the instruments highlights differences in the versions for different pediatric age groups. None of the PROMs fulfill all the good research practices recommended by the ISPOR task force. Further research is needed to define which age-appropriate domains are important for older children and adolescents.
Introduction Dementia is characterized by global cognitive dysfunction, which can cause difficulties in performing Activities of Daily Living (ADLs) and Instrumental Activities of Daily Living (IADLs), leaving people with dementia (PwD) who do not have the proper support extremely vulnerable. Dementia management should involve preventative methods, including during the stage of mild cognitive impairment (MCI). Lay-caregivers are found to have poorer health in all three domains of the biopsychosocial model, as a result of the burden of care. New assistive technologies (ATs) have been designed to help care for PwD. ATs aim to be more affordable and widely available than human workers, achieving greater health equity and quality of life for all. Methods To identify relevant articles, a literature search using PubMed was undertaken by one independent reviewer: S.L.C. The keywords of “dementia”, “technology”, and “management” were used, with no date of publication limitations, which revealed 571 results. Results 44 articles were included in this review. Articles regarding new technologies to diagnose dementia or MCI were not included. ATs aim to help facilitate aging-in-place, reduce medical costs, and rates of caregiver burnout, by helping maintain patient functioning. Discussion Legal issues in the form of workplace safety laws, data privacy laws and regulations, and health care ethics are major barriers to implementation that need to be resolved. The hope is that artificial intelligence (AI) systems may be able to advance what they are able to perceive and help uncover new knowledge and management options for dementia and MCI.
The implementation of artificial intelligence (AI) and Internet of Things (IoT) technology in neurosurgery offers the opportunity to advance clinical practice significantly, pushing the limits of diagnostics, decision making, and prognostication, and allowing more accurate interventions to be performed with fewer errors. AI and machine learning offer the potential for automation. In addition, owing to the broad distribution of powerful, portable wearables and wireless devices, the development of learning-based IoT applications for medical usage has been accelerated, providing real-time monitoring and potentially streamlining complicated procedures, thus saving the surgeon’s time. Together, the exploitation of AI along with IoT big data technologies, and cloud computing, have introduced the highly advanced networked surgical operating room, the Smart Cyber Operating Theater (SCOT).The work presented in this chapter reviews the basic concepts of AI and IoT, their application in neurosurgery and the neurosurgical operating theater, and recent trends. Their current and future perspectives are also discussed to assess their effect on clinical practice and surgical precision objectively.KeywordsArtificial intelligenceMachine learningAutomationIoTSCOT
Full-text available
Aim This study aims to test the feasibility of the PainChek app to assess pain for people with dementia living in residential aged care facilities (RACFs). It will also identify the optimal dosage and efficacy of a social robot (personal assistant robot [PARO]) intervention on chronic pain for people with dementia. Design This is a feasibility randomized controlled trial with three groups. Methods Forty-five residents living with dementia and chronic pain will be recruited from one RACF. The intervention consists of an individual 15-min non-facilitated session with a PARO robot twice a day (Group 1), a PARO robot once a day (Group 2), or a Plush-Toy (non-robotic PARO) once a day (Group 3) from Monday to Friday for 4 weeks. Participants will be followed at 4 and 8 weeks after baseline assessments. The primary outcome will be the feasibility of using the PainChek app to measure changes in pain levels before and after each session. Secondary outcomes include staff-rated pain levels, neuropsychiatric symptoms, quality of life and changes in psychotropic and analgesic medication use. Participants, staff and family perceptions of using PARO and the PainChek app will be collected after the 4-week intervention. Discussion This study will test the use of the PainChek app and PARO to improve pain management for people with dementia. Results from this study will help determine its usefulness, feasibility and acceptability for pain management in people with dementia living in RACFs. Impact As pain is a significant problem for people with dementia, this project will generate evidence on the use of the PainChek to measure the efficacy of a social robot intervention that has the potential to improve the quality of pain care in people with dementia. Trial Registration Australian and New Zealand Clinical Trials Registry number (ACTRN12621000837820) date registered 30/06/2021.
Full-text available
Purpose Accurate pain assessment is critical to detect pain and facilitate effective pain management in dementia patients. The electronic Pain Assessment Tool (ePAT) is a point-of-care solution that uses automated facial analysis in conjunction with other clinical indicators to evaluate the presence and intensity of pain in patients with dementia. This study aimed to examine clini-metric properties (clinical utility and predictive validity) of the ePAT in this population group. Methods Data were extracted from a prospective validation (observational) study of the ePAT in dementia patients who were ≥65 years of age, living in a facility for ≥3 months, and had Psychogeriatric Assessment Scales – cognitive scores ≥10. The study was conducted in two residential aged-care facilities in Perth, Western Australia, where residents were sampled using purposive convenience strategy. Predictive validity was measured using accuracy statistics (sensitivity, specificity, positive predictive value, and negative predictive value). Positive and negative clinical utility index (CUI) scores were calculated using Mitchell’s formula. Calculations were based on comparison with the Abbey Pain Scale, which was used as a criterion reference. Results A total of 400 paired pain assessments for 34 residents (mean age 85.5±6.3 years, range 68.0–93.2 years) with moderate–severe dementia (Psychogeriatric Assessment Scales – cognitive score 11–21) were included in the analysis. Of those, 303 episodes were classified as pain by the ePAT based on a cutoff score of 7. Unadjusted prevalence findings were sensitivity 96.1% (95% CI 93.9%–98.3%), specificity 91.4% (95% CI 85.7%–97.1%), accuracy 95.0% (95% CI 92.9%–97.1%), positive predictive value 97.4% (95% CI 95.6%–99.2%), negative predictive value 87.6% (95% CI 81.1%–94.2%), CUI⁺ 0.936 (95% CI 0.911–0.960), CUI⁻ 0.801 (95% CI 0.748–0.854). Conclusion The clinimetric properties demonstrated were excellent, thus supporting the clinical usefulness of the ePAT when identifying pain in patients with moderate–severe dementia.
Full-text available
Background/aims: Pain is common in aged care residents with dementia; yet it often goes undetected. A novel tool, the electronic Pain Assessment Tool (ePAT), was developed to address this challenging problem. We investigated the psychometric properties of the ePAT. Methods: In a 10-week prospective observational study, the ePAT was evaluated by comparison against the Abbey Pain Scale (APS). Pain assessments were blindly co-performed by the ePAT rater against the nursing staff of two residential aged care facilities. The residents were assessed twice by each rater: at rest and following movement. Results: The study involved 34 residents aged 85.5 ± 6.3 years, predominantly with severe dementia (Psychogeriatric Assessment Scale - Cognitive Impairment score = 19.7 ± 2.5). Four hundred paired assessments (n = 204 during rest; n = 196 following movement) were performed. Concurrent validity (r = 0.911) and all reliability measures (κw = 0.857; intraclass correlation coefficient = 0.904; α = 0.950) were excellent, while discriminant validity and predictive validity were good. Conclusion: The ePAT is a suitable tool for the assessment of pain in this vulnerable population.
Full-text available
Controls: Autonomic measures (HR change intercept and mean HR change) were reduced in severe vs. mildly affected AD patients. Group functional connectivity differences associated with greater pain behavior reactivity in patients included: connectivity within a temporal limbic network (TLN) and between the TLN and ventromedial prefrontal cortex (vmPFC); between default mode network (DMN) subcomponents; between the DMN and ventral salience network (vSN). Reduced HR responses within the AD group were associated with connectivity changes within the DMN and vSN-specifically the precuneus and vmPFC. Discriminant classification indicated HR-related connectivity within the vSN to the vmPFC best distinguished AD severity. Thus, altered behavioral and autonomic pain responses in AD reflects dysfunction of networks and structures subserving affective, self-reflective, salience and autonomic regulation.
Full-text available
Pain is common among people with moderate to severe dementia, but inability of patients to self-report means it often goes undetected and untreated. We developed the electronic Pain Assessment Tool (ePAT) to address this issue. A point-of-care App, it utilizes facial recognition technology to detect facial micro-expressions indicative of pain. ePAT also records the presence of pain-related behaviors under five additional domains (Voice, Movement, Behavior, Activity, and Body). In this observational study, we assessed the psychometric properties of ePAT compared to the Abbey Pain Scale (APS). Forty aged care residents (70% females) over the age of 60 years, with moderate to severe dementia and a history of pain-related condition(s) were recruited into the study. Three hundred and fifty-three paired pain assessments (either at rest or post-movement) were recorded and analyzed. The ePAT demonstrated excellent concurrent validity (r = 0.882, 95% CI: 0.857-0.903) and good discriminant validity. Inter-rater reliability score was good overall (weighted κ= 0.74, 95% CI: 0.68-0.80) while internal consistency was excellent. ePAT has psychometric properties which make it suitable for use in non-communicative patients with dementia. ePAT also has the advantage of automated facial expression assessment which provides objective and reproducible evidence of the presence of pain.
Full-text available
Background Pain is often poorly managed in people who have a dementia. Little is known about how this patient population is managed in hospital, with research to date focused mainly on care homes. This study aimed to investigate how pain is recognised, assessed and managed in patients with dementia in a range of acute hospital wards, to inform the development of a decision support tool to improve pain management for this group. MethodsA qualitative, multi-site exploratory case study. Data were collected in four hospitals in England and Scotland. Methods included non-participant observations, audits of patient records, semi-structured interviews with staff and carers, and analysis of hospital ward documents. Thematic analysis was performed through the lens of decision making theory. ResultsStaff generally relied on patients’ self-report of pain. For patients with dementia, however, communication difficulties experienced because of their condition, the organisational context, and time frames of staff interactions, hindered patients’ ability to provide staff with information about their pain experience. This potentially undermined the trials of medications used to provide pain relief to each patient and assessments of their responses to these treatments. Furthermore, given the multidisciplinary environment, a patient’s communication about their pain involved several members of staff, each having to make sense of the patient’s pain as in an ‘overall picture’. Information about patients’ pain, elicited in different ways, at different times and by different health care staff, was fragmented in paper-based documentation. Re-assembling the pieces to form a ‘patient specific picture of the pain’ required collective staff memory, ‘mental computation’ and time. Conclusions There is a need for an efficient method of eliciting and centralizing all pain-related information for patients with dementia, which is distributed in time and between personnel. Such a method should give an overall picture of a patient’s pain which is rapidly accessible to all involved in their care. This would provide a much-needed basis for making decisions to support the effective management of the pain of older people with dementia in hospital.
There is no shortage of suggested methods to screen for depression, including clinical interviews. Assuming these are applied to a group containing patients with depression and patients without depression, how do we decide which are the optimal methods? In addition, how can tests be compared and how can tests be combined? This chapter discusses the methods used to compared scales and tools. The terms diagnosis and screening both refer to the application of an agreed method to confirm those with a condition and to exclude those without the condition (for discussion see Chapter 2). When attempting to separate depressed versus non-depressed individuals there is always an overlap of symptoms (or biological markers) (see Chapter 1, Fig. 1); therefore, a perfect test based on current tests is unobtainable. Testing may be focused on those at high risk of the condition (such as screening for depression after myocardial infarction) or applied to a wider population (screening for depression in all primary care patients). The former is a high-prevalence setting, which favors the ability to confirm a condition, whereas the latter is a low-prevalence setting, which favors the ability to refute a condition. It is often forgotten that the clinical process of making a diagnosis is a form of screening itself. Here the tool is the clinician’s clinical skill and the sample is all patients seen by the clinician. If a clinician is attuned to the concept of depression, has a high index of suspicion, and asks the right questions, then it is likely he or she will have high personal diagnostic accuracy. If the clinician is unconfident, inexperienced, and untrained, it is less likely that he or she will be able to make a correct diagnosis (see Table 5.1 and Chapter 3). Some literature suggests that the added value of screening tools for depression is apparent only in the latter situation. A diagnostic test for depression is designed to help the clinician elicit and weigh symptoms and signs to make a diagnosis. How, then, is this achieved, and how does a screening test work in scientific terms?
Recognition of pain in older persons with dementia is a considerable challenge to quality pain care for this vulnerable population. Without recognition, pain cannot be thoroughly evaluated and effectively treated. Observing for pain-related behaviors is the most researched means of identifying the presence or likelihood of pain in persons with moderate to severe dementia, or those who are unable to self-report their pain. The purpose of this paper is to discuss the state of observation of pain, primarily focusing on pain behavior tool development, providing an overview of current pain tools and discussing the challenges at this stage of the science, including the issue of assessing pain intensity . We also recommend a number of areas to prioritize future research with the goal to advance effective pain assessment in older persons with dementia. Central to these recommendations is the refinement of existing tools to incorporate those behaviors most predictive of pain in persons with dementia as the science progresses in this area. The future of pain observation in dementia is poised for considerable advancement through these refinements of tools and techniques. Improving our ability to detect and evaluate pain in the vulnerable population unable to self-report their pain through the results of these suggested research prioirities will likely assist in addressing the related suffering that results from unrecognized and untreated pain.
Technology is playing an increasing role in the lives of the elderly. One of the most prevalent developments for the aging population is the use of technological innovations for intervention and treatment of individuals with mental impairments. The Handbook of Research on Innovations in the Diagnosis and Treatment of Dementia offers empirical research and theoretical analyses on the cognitive impairment of the aging. Featuring studies in gerotechnology, this book is an essential resource for researchers, students, and practitioners in the field of geriatrics who are interested in the emerging research, clinical practices, therapy, and technological innovations concerning the development and treatment of dementia.