Available via license: CC BY-NC 4.0
Content may be subject to copyright.
Vol.:(0123456789)
PharmacoEconomics - Open
https://doi.org/10.1007/s41669-023-00419-w
ORIGINAL RESEARCH ARTICLE
Content Validity andPsychometric Evaluation oftheFunctional
Assessment ofChronic Illness Therapy–Fatigue (FACIT–Fatigue)
inPatients withCrohn’s Disease andUlcerative Colitis
EdwardV.LoftusJr1 · AshwinN.Ananthakrishnan2· Wan‑JuLee3· YuriSanchezGonzalez3· KristinaAFitzgerald3·
KoriWallace3· WenZhou3· LeighannLitcher‑Kelly4· SarahB.Ollis4· SylviaSu4· SilvioDanese5
Accepted: 30 April 2023
© The Author(s) 2023
Abstract
Background Patients with Crohn’s disease (CD) or ulcerative colitis (UC) frequently experience fatigue, although it is often
overlooked in medical research and practice.
Aims To explore patients’ experience of fatigue and evaluate content validity, psychometric properties, and score interpret-
ability of the Functional Assessment of Chronic Illness Therapy–Fatigue (FACIT–Fatigue) in patients with CD or UC.
Methods Concept elicitation and cognitive interviews were conducted with participants aged ≥15years with moderately-to-
severely active CD (N=30) or UC (N=33). To evaluate psychometric properties (reliability and construct validity) and inter-
pretation of FACIT–Fatigue scores, data from two clinical trials were analyzed [ADVANCE (CD): N=850; U-ACHIEVE
(UC): 248]. Meaningful within-person change was estimated using anchor-based methods.
Results Almost all interview participants reported experiencing fatigue. Over 30 unique fatigue-related impacts were reported
per condition. The FACIT–Fatigue was interpretable for most patients. FACIT–Fatigue items had good internal consistency
(Cronbach’s α0.86–0.88 for CD and 0.94–0.96 for UC); the total score displayed acceptable test–retest reliability (intraclass
correlation coefficients >0.60 for CD and >0.90 for UC). FACIT–Fatigue scores had acceptable convergent validity with
similar measures. A 7–10 point improvement for CD and 4–9 point improvement for UC on the FACIT–Fatigue total score
may represent meaningful improvements.
Conclusions These results highlight the importance of fatigue among adolescents and adults with CD or UC and provide
evidence that the FACIT–Fatigue is content valid and produces reliable, valid, and interpretable scores in these populations.
Care should be taken if using the questionnaire with adolescents who may be less familiar with the word “fatigue.”
Clinical trial registration numbers NCT03105128 (date of registration: 4 April 2017) and NCT02819635 (date of registra-
tion: 28 June 2016).
1 Introduction
Crohn’s disease (CD) and ulcerative colitis (UC) are charac-
terized by chronic relapsing and remitting inflammation of
the gastrointestinal (GI) tract. While patients with CD and
UC experience similar symptoms[1–3], the two conditions
differ with respect to the location and extent of inflamma-
tion in the GI tract. CD can affect any portion of the GI tract
and presents as transmural inflammation involving all tissue
layers of the bowel wall, while UC is manifested as diffuse
mucosal inflammation of the colon and/or rectum[4].
One impactful, yet often overlooked, symptom of CD and
UC is fatigue[4–7]. The prevalence of fatigue among CD
and UC patients, based on previous literature, ranges from
approximately 24% to 87%[7–11]. The pooled prevalence
of fatigue in CD and UC was 47% based on a random-effects
meta-analysis, compared with only 5% in healthy individu-
als. For individuals with active disease, the pooled preva-
lence of fatigue was 72%, compared with 47% for those in
remission[7]. Common risk factors for fatigue in CD and
UC include sleep disturbance, anxiety, depression, vitamin
and mineral deficiencies, and anemia[7]. Fatigue is associ-
ated with impaired health-related quality of life, including
high disability, decreased physical function, and negative
effects on work productivity[7]. The prevalence and bur-
den of fatigue demonstrates its importance for consideration
when developing interventions for CD and UC[7].
Fatigue, however, is not conceptually straightforward—
it is a multifaceted symptom that may be experienced in
various ways, both within and across individuals. As
Extended author information available on the last page of the article
E.V.Loftus Jr et al.
Key Points for Decision Makers
Fatigue is an important symptom experienced by indi-
viduals with moderately-to-severely active CD and UC,
with substantial impact to their lives.
The FACIT–Fatigue is comprehensible, relevant, and
comprehensive for assessing fatigue and fatigue-related
impacts for use with individuals with moderately-to-
severely active CD and UC.
FACIT–Fatigue is a reliable and valid tool that is
sensitive to change, and an improvement in the FACIT–
Fatigue total score of 7–10 points or 4–9 points may be
clinically meaningful for patients with moderately-to-
severely active CD and UC, respectively, in a clinical
trial setting.
such, it is important to consider an assessment approach
accounting for diverse aspects of fatigue.
One common instrument to measure the patient experi-
ence of fatigue is the Functional Assessment of Chronic
Illness Therapy–Fatigue (FACIT–Fatigue), which assesses
concepts related to the severity and impacts of fatigue over
the past 7days [12]. While there are many questionnaires
that could be considered for the evaluation of fatigue, and
even some that assess fatigue specifically among those
with inflammatory bowel disease (IBD), such as the
Inflammatory Bowel Disease Questionnaire (IBDQ) [13],
the FACIT–Fatigue captures varied fatigue-related con-
cepts across its 13 items, allowing for a nuanced assess-
ment (as opposed to a single item broadly evaluating
fatigue).
While the FACIT–Fatigue was originally developed to
assess anemia-related fatigue in cancer patients[14–16],
and has been widely used in a variety of disease areas, the
appropriateness of the FACIT–Fatigue for use in CD and
UC has also been supported by prior publications[11, 17].
Previous studies on IBD have used the FACIT–Fatigue to
evaluate the severity of fatigue in patients with IBD and
have reported mean scores that are associated with moderate
or severe fatigue[17–19], while lower fatigue was associ-
ated with improvements in other symptoms related to treat-
ment[5, 20]. In addition, previous research has aimed to
understand the reliability, validity, and sensitivity to change
of the FACIT–Fatigue in CD and UC[17, 20–22].
To expand on this previous research and to confirm that
the FACIT–Fatigue (13-item version) is appropriate for use
in registrational clinical trials for CD and UC, the current
study aimed to evaluate the content validity (readability,
relevance, comprehensibility, and comprehensiveness),
psychometric performance, and score interpretation of the
FACIT–Fatigue for adolescents and adults with clinician-
confirmed, moderately-to-severely active CD or UC.
2 Methods
2.1 Evaluation ofContent Validity
Trained interviewers conducted open-ended, semi-structured
qualitative interviews with adolescents and adults in the
USA with moderately-to-severely active CD or UC to (1)
identify, describe, and substantiate the important and rel-
evant CD- and UC-related fatigue experiences and impacts
(concept elicitation) and (2) evaluate the readability, com-
prehensibility, relevance, and comprehensiveness of the
FACIT–Fatigue (cognitive debriefing). The study protocol
and all study documents were approved by a centralized
independent review board.
2.1.1 Participants
Adolescents (i.e., 15–17years of age) and adults (i.e.,
≥18 years of age) with a clinician-confirmed diagno-
sis of moderately-to-severely active CD or UC were
recruited from clinical sites within the USA. The study
inclusion/exclusion criteria specified, among other crite-
ria, that Health Insurance Portability and Accountability
Act (HIPAA) authorization, informed consent to partici-
pate for adult participants, and parental permission and
assent for adolescent participants be obtained. Participants
≥15years of age and ≤80years old, fluent in US English,
and with a clinician-confirmed diagnosis of moderate-to-
severe CD or UC were considered for inclusion in the
study. Full inclusion and exclusion criteria can be found
in Supplementary Table1 in the Online Resource 1: Sup-
plementary tables and figures.
2.1.2 Conduct ofQualitative Interviews
The qualitative interviews were conducted either in per-
son or over the phone using a semi-structured interview
guide. During the interviews, participants were first
asked to briefly discuss whether they have experienced
fatigue related to their CD or UC and to describe how
their disease-related fatigue felt and how it impacted their
lives. Participants were subsequently asked to complete
the FACIT–Fatigue either on an electronic device or
by using screenshots of the questionnaire while “think-
ing aloud” about the process they used to arrive at each
answer to identify any words, terms, or concepts within
Evaluation of FACIT–Fatigue for Patients with CD and UC
the questionnaire that they did not understand or did not
interpret as intended. Following the “think-aloud” process,
participants were asked additional questions designed to
evaluate the content of the FACIT–Fatigue, including its
comprehensibility, readability, relevance, and compre-
hensiveness. Interviews were audio recorded, following
participants’ verbal consent, and subsequently transcribed
and anonymized.
2.1.3 Qualitative Coding andData Analysis
Data were analyzed separately for CD and UC. Each tran-
script underwent qualitative coding to organize and catalog
participants’ CD- or UC-related fatigue experience, feed-
back, and responses on the instructions, items, and response
options of the FACIT–Fatigue. All transcripts were coded
in ATLAS.ti (ATLAS.ti Scientific Software Development
GmbH, Berlin, Germany); coded quotations were then fur-
ther reviewed and aggregated by themes in study findings
tables.
2.2 Evaluation ofPsychometric Performance
andInterpretability ofScores oftheFACIT–
Fatigue
Data from two clinical programs [risankizumab for CD Phase
3 (NCT03105128; hereafter identified as ADVANCE)[23]
and upadacitinib for UC Phase 2b (NCT02819635; hereaf-
ter identified as U-ACHIEVE)[24] were used for the psy-
chometric and score interpretation analyses. The following
properties of the FACIT–Fatigue scores were evaluated for
CD and UC separately: quality of completion of the ques-
tionnaire, total score distribution, reliability [internal con-
sistency using Cronbach’s alpha and test–retest reliability
using intraclass correlation coefficient (ICC)], and valid-
ity (convergent and discriminant validity, known-groups
methods, and sensitivity to change). Further, analyses were
conducted to establish meaningful within-person change
(MWPC) estimates for the FACIT–Fatigue total score.
During the clinical trials, participants completed the
FACIT–Fatigue questionnaire (along with other question-
naires/assessments) during clinic visits using an electronic
tablet device. The methodology of these trials is described
fully elsewhere[23–25].
2.2.1 Participants
Scores on the FACIT–Fatigue and other clinical outcome
assessments (both patient reported and clinician reported)
were included in the psychometric and score interpreta-
tion analyses for CD and UC separately. Analysis popula-
tions were drawn from those participants in each clinical
trial who were randomized to an active treatment or pla-
cebo group, but required that participants had scores on
the FACIT–Fatigue from one or more clinic visits. For CD,
participants with scores on the FACIT–Fatigue at baseline,
week 4, and/or week 12 of the clinical trial were included in
the psychometric analyses. For UC, participants with scores
on the FACIT–Fatigue at baseline, week 2, week 4, and/or
week 8 were included. While all participants in the psycho-
metric analyses were randomized, scores from all treatment
groups were collapsed for all analyses (i.e., analyses did
not examine the differences between treatment and placebo
groups).
2.2.2 Assessments
The version of the FACIT–Fatigue (version 4) under evalu-
ation includes 13 items measuring fatigue and its impacts.
All items include the same 5-point verbal rating scale rang-
ing from 0 (“not at all”) to 4 (“very much”) with a recall
period of the “past 7days.” Items 7 (i.e., “I have energy”)
and 8 (i.e., “I am able to do my usual activities”) express a
positive connotation, while other items on the 5-point scale
reflect a negative connotation (e.g., “I am too tired to eat”)
related to fatigue. To calculate a total score, the scores of all
items except for items 7 and 8 are reversed by subtracting
the response from 4. After reversing the items (specifically,
items 1–6 and 9–13), a total score is calculated by summing
the individual item scores. The FACIT–Fatigue (version
4) has a maximum score of 52 and a minimum score of 0,
where a higher score equates to less fatigue. Item numbers
will be referenced as 1–13; however, the associated item
bank numbers are as follows: item 1 (HI7), item 2 (HI12),
item 3 (An1), item 4 (An2), item 5 (An3), item 6 (An4), item
7 (An5), item 8 (An7), item 9 (An8), item 10 (An12), item
11 (An14), item 12 (An15), and item 13 (An16).
In addition to the FACIT–Fatigue, scores of other assess-
ments were used to support the analysis. The Crohn’s Dis-
ease Activity Index (CDAI) and the Crohn’s Symptoms
Severity Questionnaire (CSS) were used for CD, and the
Adapted Mayo Score and the Ulcerative Colitis Symptoms
Questionnaire (UC-SQ) were used for UC as disease-spe-
cific assessments. Impact and quality-of-life assessments,
namely the Work Productivity and Activity Impairment
Questionnaire (WPAI: CD and WPAI: UC), the IBDQ,
and the 36-Item Short Form Survey Version 2 (SF-36v2®),
were used for both CD and UC. The five-level EQ-5D (EQ-
5D-5L) was used for CD/UC as a general health and utilities
assessment. In addition, a disease-specific Global Impres-
sion of Change (PGIC) was used in both CD and UC, while
a disease-specific Patient Global Impression of Severity
(PGIS) was used for CD only. The PGIC and PGIS items
asked patients to rate the change in, or the severity of, their
E.V.Loftus Jr et al.
Table 1 Reference measures for psychometric and score interpretation analyses (CD and UC)
Measure Concepts measured Included in
CD and/or UC
analyses
Recall period Scoring
Disease activity assessments
Crohn’s Disease Activity Index (CDAI)a[31] Abdominal pain, stool frequency, and general
well-being, in addition to presence of other
conditions, treatment for diarrhea, abdomi-
nal mass, hematocrit, and body weight/
standard weight
CD The past 7days Summing the weighted individual scores of 8
itemsb
Adapted Mayo Score Scoring Systemc [32] Stool frequency, rectal bleeding, and endos-
copy results
UC Current state Subscores from each of the three domains each
ranging from 0 to 3; total score ranges from
0 to 9b
Symptom assessments
Crohn’s Symptoms Severity Questionnaire
(CSS)d
CD-related gastrointestinal and nongastroin-
testinal symptoms (frequency and intensity)
CD The past 7days 14 items with overall symptom score calcu-
lated by combining ratings from individual
items (possible scores range from 14 to 70)b
UC Symptom Questionnaire (UC-SQ)dUC-related gastrointestinal symptoms and
nongastrointestinal symptoms
UC The last week 17 items with the total score calculated by
summing the rating for each itemb
Impacts and quality of life assessments
Work Productivity and Activity Index
(WPAI) for CD or UC (WPAI: CD and
WPAI: UC)d [33, 34]
Impact of disease on absenteeism, pres-
enteeism, productivity loss, and activity
impairment
CD and UC The past 7days Six items; scores are expressed as impairment
percentages, adjusting for hours worked
according to the WPAI score algorithmb
Inflammatory Bowel Disease Questionnaire
(IBDQ)d[13]
Dimensions of symptoms and health-related
quality of life: bowel symptoms, systemic
symptoms, social function, emotional
function
CD and UC The past 2weeks 32 items split into four domains; each domain
score can be computed with scores ranging
from 10 to 70, 5 to 35, 12 to 84, and 5 to 35;
total scores range from 32 to 224e
36-Item Short Form Fatigue (SF-36v2®)d
[35]
Physical functioning, bodily pain, role limita-
tions due to physical health problems, role
limitations due to personal or emotional
problems, emotional well-being, social
functioning, energy/fatigue, and general
health problems
CD and UC The past 4weeks 36 items that, when scored, can be aggregated
into two summary responses: the physical
and mental component summary scores;
each item is scored on a 0–100 range; scores
represent the percentage of total possible
scores; items in the same scale are then aver-
aged together
Global assessments
The Patient Global Impression of Severity
(PGIS)d
Severity of the overall symptoms due to CD CD The past week One item rated on a 7-point verbal rating scale
from “Absent” to “Very severe”a
The Patient Global Impression of Change
(PGIC)d
Overall change in their UC or CD symptoms CD and UC Since before treatment began One item rated on a 7-point verbal rating scale
ranging from “Very much improved” to
“Very much worse”a
Evaluation of FACIT–Fatigue for Patients with CD and UC
CD or UC symptoms overall. All assessments are described
in Table1.
2.2.3 Statistical Analyses
Descriptive characteristics, reliability, validity, and score
interpretation analyses were conducted on the clinical trial
datasets for CD (ADVANCE) and UC (U-ACHIEVE) sepa-
rately. Table2 summarizes each set of analyses and bench-
marks for determining acceptable results, where appropriate.
3 Results
3.1 Qualitative Research forContent Validity inCD
andUC
3.1.1 Participant Demographics
Between January 2020 and September 2020, 30 individuals
with CD and 33 individuals with UC who metall the study
inclusion and none of the exclusion criteria (as described
in Supplementary Table1 in the Online Resource 1: Sup-
plementary tables and figures) were recruited from four US
clinical sites. Participants with clinician-confirmed moder-
ate-to-severe CD (n=20 adults and n=10 adolescents) and
UC (n=22 adults and n=11 adolescents) participated in
qualitative hybrid concept elicitation and cognitive debrief-
ing interviews. Participant demographic and health informa-
tion are provided in Table3.
3.1.2 Fatigue Experience
The open-ended concept elicitation portion of the interviews
confirmed that almost all participants experienced fatigue
due to CD or UC (CD: n=30/30, 100.0%; UC: n=32/33,
97.1%). Participants described fatigue as: “a feeling of
physical or mental tiredness that comes in waves, exhaus-
tion, weakness, lethargy, lack of motivation, lack of energy,
wanting to go to bed or relax, sleepiness, being worn out,
sluggish or slow, being unalert, lifeless, achy, and/or feel-
ing drained.” When asked to rate fatigue on a 0–10 numeric
rating scale with 0=not bothersome at all to 10=most
bothersome, CD participants reported a mean bother rat-
ing of 6.8 [median=7, interquartile range (IQR)=5–8]
and UC participants reported a mean bother rating of 6.2
(median=6.5, IQR=5–8).
Fatigue was additionally reported to impact quality of
life. Participants reported fatigue-related impacts (n=32
unique impacts for CD and n=33 unique impacts for UC)
across nine domains (activities of daily living, cognitive
function, emotional function, leisure activities, physical
activities, relationships, sleep, social function, and work/
Table 1 (continued)
Measure Concepts measured Included in
CD and/or UC
analyses
Recall period Scoring
General health and utilities assessment
Five-level EQ-5D (EQ-5D-5L)d [36] Mobility, self-care, usual activities, pain/dis-
comfort, and anxiety/depression
Self-evaluated health status
CD and UC Current health (i.e., “today”) EQ-5D descriptive system: Each of the five
dimensions includes fives levelsa
EQ-5D visual analog scale (VAS): A verti-
cal VAS ranging from 0 (worst health you
can imagine) to 100 (best health you can
imagine)a
CD Crohn’s disease, UC ulcerative colitis
a Composite measure (patient, clinician, laboratory)
b Higher scores indicate greater disease severity or greater negative impact
c Clinician-rated measure
d Patient-rated measure
e Higher scores indicate better outcomes
E.V.Loftus Jr et al.
Table 2 Summary of psychometric analyses for FACIT–Fatigue scores (CD and UC)
Analysis Description Timepoint(s) for CD Timepoint(s) for UC
Quality of completion Evaluate missing scores for the FACIT–
Fatigue items and total score
Baseline, week4, and week12 Baseline, week2, and week8
Item and total score distributions Descriptive statistics to summarize FACIT–
Fatigue item and total scores, as well as
to describe the ceiling/ floor effects in the
target measure. While there is no generally
agreed-upon threshold that defines a floor or
ceiling effect, items were considered to have
an extreme floor/ceiling effect if more than
40% of participants endorsed the lowest or
highest option, respectively[37]
Baseline and week12 Baseline and week8
Inter-item correlation Evaluate how highly items within the
FACIT–Fatigue correlate to each other to
understand conceptual overlap. A strong
correlation was defined as ≥0.70 to ≤0.90,
a moderate correlation as ≥0.30 to <0.70,
and a weak correlation as <0.30[38]
Week12 Week8
Item-total correlation Evaluate the extent to which each item
contributes to the total score. The threshold
for acceptable item-total correlations was
≥0.3[26]
Week12 Week8
Internal consistency reliability Evaluate the degree to which individual items
are measuring the same general concept.
Cronbach’s alpha coefficient typically
ranges from 0 to 1.0, with higher estimates
indicative of stronger internal consist-
ency among items. Internal consistency
was considered to be met if α≥0.70 [39].
Cronbach’s alpha was calculated with each
item removed from its respective score to
assess the impact
Baseline and week12 Baseline and week8
Test–retest reliability Evaluate the degree to which items produce
stable, reliable scores under similar condi-
tions
For CD, stability was defined in two ways:
participants who selected (1) the same
response on the PGIS at baseline and week
4 or baseline and week 12 and (2) “no
change” for CD symptoms on the PGIC at
week 4 and week 12.
For UC, stability was defined in two ways:
(1) participants who selected “no change”
for UC symptoms on the PGIC at week 2 or
(2)no change between baseline and week 2
on scores for UC-SQ item 6 (tired, lacking
energy during the past week).
Test–retest reliability ICCs were calculated
with a two-way mixed-effects model
without interaction [ICC(3A,1)]. An ICC
of 0.70 or greater was used as evidence of
acceptable test–retest reliability for a scale
[40, 41]
Baseline, week 4, and week12 Baseline and week2
Evaluation of FACIT–Fatigue for Patients with CD and UC
school). The most frequently reported CD-related fatigue
impacts were lack of motivation (n=14, 46.7%), limited
social interactions (n=11, 36.7%), difficulty starting things
(n=11, 36.7%), and limitation to household chores (n=11,
36.7%). The most frequently reported UC-related impacts
were limited physical activity (n=18, 56.3%), limitation to
household chores (n=12, 37.5%), limited social interactions
(n=10, 30.3%), and limited productivity at work or school
(n=9, 28.1%).
Table 2 (continued)
Analysis Description Timepoint(s) for CD Timepoint(s) for UC
Convergent/discriminant validity Evaluate the degree to which scores produced
by FACIT–Fatigue correlate with scores
produced by other measures that should the-
oretically be associated with them. A strong
correlation defined as ≥0.70 to ≤0.90,
moderate correlation as ≥0.30 to <0.70,
and a weak correlation as <0.30[38]
For CD, concurrent assessments were the
CDAI, PGIS, IBDQ, SF-36v2®, EQ-5D-5L,
and WPAI: CD
For UC, concurrent assessments were the
Adapted Mayo Score, EQ-5D-5L, IBDQ,
SF-36v2®, UC-SQ, and WPAI: UC
Week12 Week8
Known-groups methods Evaluate the degree to which scores produced
by the FACIT–Fatigue are capable of
distinguishing among groups hypothesized
a priori as being clinically distinct
For CD, known groups were defined using
the CDAI, IBDQ, and PGIS. The classifica-
tion values for the CDAI were <150 for
remission versus ≥150 for nonremission,
and for the IBDQ total score, ≥170 for
remission versus <170 for nonremission
For UC, known groups were defined using
the Adapted Mayo Score, UC-SQ item 6
(felt tired or lacking energy during the past
week), and IBDQ item2 (feeling of fatigue
or being tired and worn out)
Week12 Week8
Sensitivity to change Evaluate the degree to which the change of
scores produced by the FACIT–Fatigue
change in concert with the change of scores
produced by other concurrent measures
(same as those listed in the convergent
validity above)
Baseline and week12 Baseline and week8
Anchor-based methods Estimate the MWPC for the FACIT–Fatigue
total score. For CD, anchors were the PGIS
and PGIC, and for UC, the anchor was the
PGIC. Estimates were supplemented by
eCDF curves, PDF curves, and ROC curves
Baseline and week12 Baseline and week8
Distribution-based methods Estimate the clinically important difference
between groups for the FACIT–Fatigue
total score
MCID1: 0.5 of an SD[42] at Baseline;
MCID2: SEM:
SEM = SDatBaseline ∗ √
1
− reliabilit y
a
Baseline Baseline
CD Crohn’s disease, CDAI Crohn’s Disease Activity Index, eCDF empirical cumulative distribution function, EQ-5D-5L Five-level EQ-5D,
FACIT–Fatigue Functional Assessment of Chronic Illness Therapy–Fatigue, IBDQ Inflammatory Bowel Disease Questionnaire, ICC intraclass
correlation coefficient, MCID minimal clinically important difference, MWPC meaningful within-person change, PDF probability distribution
function, PGIC Patient Global Impression of Change, PGIS Patient Global Impression of Severity, ROC receiver operating characteristic, SD
standard deviation, SEM standard error of measurement, SF-36v2® 36-Item Short Form Survey Version 2, UC ulcerative colitis, UC-SQ Ulcera-
tive Colitis Symptoms Questionnaire, WPAI: CD Work Productivity and Activity Impairment Questionnaire for Crohn’s Disease
a Cronbach’s alpha at baseline
E.V.Loftus Jr et al.
Table 3 Patient interviews:
Participant-and clinician-
reported demographic and
health information (CD and
UC)
Characteristic CD total sample
(N=30)
n (%)a
UC total sample
(N=33)
n (%)a
Age, in yearsb
Range (minimum–maximum) 15.1–75.4 15.1–80.1
Mean (SD) 36.6 (19.2) 38.3 (19.8)
Sexb
Female 16 (53.3%) 17 (51.5%)
Male 14 (46.7%) 16 (48.5%)
Spanish/Hispanic/Latinoc
Not Spanish/Hispanic/Latino 24 (80.0%) 32 (97.0%)
Mexican/Mexican American, Chicano 5 (16.7%) –
Hispanic, unspecified 1 (3.3%) –
Puerto Rican – 1 (3.0%)
Racec
White 22 (73.3%) 24 (72.7%)
Hispanic 6 (20.0%) –
Black or African American 2 (6.7%) 7 (21.2%)
Asian – 1 (3.0%)
Unspecified – 1 (3.0%)
Education (adults)c;d
High school diploma or GED 2 (6.7%) 3 (9.1%)
Some college or associate degree 8 (26.7%) 5 (15.2%)
College or university degree 5 (16.7%) 10 (30.3%)
Graduate or professional degree 4 (13.3%) 4 (12.1%)
Other: Technical school 1 (3.3%) –
Education (adolescents)c;d
9th grade 2 (6.7%) –
10th grade 1 (3.3%) 2 (6.1%)
11th grade 2 (6.7%) 2 (6.1%)
12th grade 5 (16.7%) 7 (21.2%)
Clinician-reported severity of CD or UC (moderate or severe)b
Moderate 19 (63.3%) 19 (57.6%)
Severe 11 (36.7%) 14 (42.4%)
Current medication useb
5-ASA (e.g., mesalamine, sulfasalazine) 10 (33.3%) 20 (60.6%)
Advanced therapy (e.g., biologics) 11 (36.7%) 13 (39.4%)
Immunomodulators (e.g., azathioprine or 6-mercaptopurine) 7 (23.3%) 5 (15.2%)
Corticosteroids 13 (43.3%) 21 (63.6%)
Time since diagnosis (in years)b
Range (minimum–maximum) 1–45 1-35
Mean (SD) 9.7 (10.2) 6.0 (5.8)
Other health conditionsc;e
None 20 (66.7%) 15 (45.5%)
Depression/anxiety 3 (10.0%) 2 (6.1%)
Diabetes: Type II 2 (6.7%) 2 (6.1%)
High blood pressure 2 (6.7%) 7 (21.2%)
Rheumatoid arthritis 2 (6.7%) –
High cholesterol 1 (3.3%) 5 (15.2%)
Unspecified – 6 (18.2%)
Disease extent for UC onlyf
E1 (proctitis) – 24 (72.7%)
Evaluation of FACIT–Fatigue for Patients with CD and UC
3.1.3 Cognitive Debriefing ofFACIT–Fatigue
During the interviews, participants completed the
FACIT–Fatigue and, across items, 40.0–90.0% of partici-
pants with CD and 33.3–84.8% of participants with UC
reported experiencing the assessed concepts within the recall
period (Supplementary Table2 in the Online Resource 1:
Supplementary tables and figures). During the cognitive
debriefing of the FACIT–Fatigue, ≥70.0% of CD par-
ticipants and ≥93.0% of UC participants interpreted the
instructions, recall period, items, and response options as
intended, demonstrating the readability and comprehensibil-
ity of the questionnaire. The primary interpretation issue for
CD was that six participants (n=6, 20%), mostly adoles-
cents (n=5/6, 83.3%), reported being unfamiliar with the
term “fatigue” or did not interpret the concept as intended
(e.g., interpreted as nausea, dizziness, pain, sickness, or
the act of taking a break/resting). Participants who did not
understand the word “fatigue” were ultimately provided a
definition for the purpose of understanding whether it was
relevant experience. Overall, all CD and UC participants
(n=30 and n=33, 100.0%, respectively) reported that the
FACIT–Fatigue items measured concepts that are relevant
to their experience, which was further supported by the
spontaneous descriptions provided in the concept elicita-
tion portion of the interviews. Participants were asked to
report whether anything relevant to their fatigue experience
was missing from the questionnaire, while there were mul-
tiple suggestions, the only concept mentioned by more than
one person per condition was mental health/mental fatigue
(reported by n=6 with CD and n=1 with UC)—this was
described as an impact to one’s mental “state” or “mental
health effects” due to fatigue which can lead to changes in
one’s thinking process or mood.
3.2 CD Psychometric Evaluation andScore
Interpretation
A total of 850 patients from the ADVANCE study were
included in the psychometric and score interpretation
analysis. Participants’ ages ranged from 16 to 79years
[mean=37.5years; standard deviation (SD)=13.3years];
slightly less than half (45.9%) of the sample was female
(Table4).
3.2.1 CD: Quality ofCompletion andScore Distribution
Quality of completion for the FACIT–Fatigue for the
CD psychometric analysis population was high. At least
95.0% of participants had complete data across all analy-
sis timepoints. The mean FACIT–Fatigue total score was
25.09 at baseline (n=836, SD=11.29), 31.07 (n=825,
SD=11.81) at week 4, and 34.98 (n=778, SD=11.87) at
week 12. While item 10 (too tired to eat) and item 11 (need
help doing usual activities) demonstrated floor effects at
baseline (>40% of participants endorsing the lowest option,
“not at all,” indicating no experience of the symptom within
the recall period), improvement for these impacts was still
demonstrated over time for the sample.
3.2.2 CD: Inter‑item andItem‑Total Correlations
Inter-item correlations using data from week 12 were moder-
ate to strong (r=0.32–0.90). The strongest correlation was
between item 5 (trouble starting things) and item 6 (trouble
finishing things) (r=0.90 at week 12). Inter-item correla-
tions for the ADVANCE study at week 12 are summarized
in Supplementary Table3 in the Online Resource 1: Supple-
mentary tables and figures. For item-total correlations, the
Table 3 (continued) Characteristic CD total sample
(N=30)
n (%)a
UC total sample
(N=33)
n (%)a
E2 (distal to splenic flexure) – 12 (36.4%)
E3 (proximal to splenic flexure) – 16 (48.5%)
5-ASA 5-aminosalicylic acid, CD Crohn’s disease, SD standard deviation, UC ulcerative colitis
a Unless otherwise indicated
b Clinician-reported information
c Participant-reported information
d Not mutually exclusive
e Other health conditions were reported by a one participant with CD or UC each: arthritis, cancer (lung
cancer), fibromyalgia, ovarian cysts, thyroid disease, uterine fibroids, Wolff–Parkinson–White syndrome,
migraine headaches
f Extent of disease based upon endoscopy findings: (E1: disease limited to rectum with or without sigmoid
involvement; E2: UC present in descending colon up to, but not proximal to, splenic flexure; E3: evidence
of UC proximal to the splenic flexure)[43]
E.V.Loftus Jr et al.
magnitude between each item in the FACIT–Fatigue scale
and the FACIT–Fatigue total score across analysis time-
points ranged between r=0.57 and r=0.92.
3.2.3 CD: Reliability
Score reliability for the FACIT–Fatigue questionnaire
was assessed in two ways: internal consistency reliability
and test-retest reliability. For the first type of reliability,
Cronbach’s α for the FACIT–Fatigue total score ranged from
0.86 to 0.88 from baseline to week 12, exceeding stand-
ard thresholds for acceptable internal consistency. Removal
of any item did not markedly improve internal consistency
reliability.
Among participants who were considered “stable” (i.e.,
those that either chose the same score on the PGIS for dis-
ease activity at baseline and week 4 or week 4 and week
12, or selected “no change” on the PGIC at week 4 and
week 12, depending on the timepoint being evaluated), the
ICCs for the FACIT–Fatigue total score ranged from 0.63
[95%confidence interval (CI) 0.50–0.72] to 0.73 (95%CI,
0.64–0.79). Interpretively, the ICCs ranged from slightly
below to minimally above the a priori-defined acceptable
threshold of 0.70.
3.2.4 CD: Convergent andDiscriminant Validity
The correlations between the FACIT–Fatigue total score
and scores on almost all concurrent measures were either
as strong as or stronger than expected in the correct
directions, demonstrating evidence of convergent valid-
ity. Weaker correlations were hypothesized between the
FACIT–Fatigue and the measures used for discriminant
validity analyses (EQ-5D-5L, SF-36v2®, IBDQ, and
WPAI: CD), but moderate relationships were observed,
indicating fatigue is more strongly related to the constructs
assessed by the discriminant measures than expected.
Detailed results are presented in Table5, summarizing the
hypothesized relationships between the FACIT–Fatigue
and other measures and the observed correlations between
them.
3.2.5 CD: Known‑Groups Analysis
FACIT–Fatigue scores were strongly differentiated
between clinically distinct groups on all measures, as
expected. More specifically, FACIT–Fatigue total scores
demonstrated a 10.6–11.2 point difference between groups
classified as remission versus nonremission on the CDAI
and a 14.9–16.6 point difference in remission versus
nonremission groups using the IBDQ, and there was a
monotonic decrease in the total score by PGIS group as
severity improved; all comparisons were statistically sig-
nificant (p<0.001) (Table6).
3.2.6 CD: Sensitivity toChange
Moderate-to-strong correlations were observed
(0.430–0.701) between the FACIT–Fatigue change score
and change scores on concurrent measures from base-
line to week 12, except for the EQ-5D-5L mobility and
self-care domains and the WPAI: CD work-time-missed
domain. These three domains were weakly correlated
with the FACIT–Fatigue change score (0.287, 0.236, and
0.269, respectively). Stronger correlations were observed
between FACIT–Fatigue change scores and conceptually
similar measures such as the SF-36v2 Physical Compo-
nent Summary and IBDQ item 2 (0.624, 0.658, and 0.701,
respectively).
3.2.7 CD: Interpretation ofScores—Anchor‑Based Methods
andSupportive Analyses
For CD, FACIT–Fatigue total score improved in parallel
with each PGIS and PGIC anchor level. One and two point
improvements on the PGIS generated a total score change
of 6.17–10.03 on the FACIT–Fatigue (Table7). Anchor
groups on the PGIC (“minimally improved” or “much
improved”) generated a 6.83–13.63 point improvement in
FACIT–Fatigue total score (Table8).
Considering additional evidence from empirical cumu-
lative distribution function (eCDF) curves and probability
distribution function (PDF) curves, it is reasonable to con-
clude that a change on the FACIT–Fatigue of 7–10 points
may indicate meaningful improvement (i.e., an estimate of
MWPC). Based on receiver operating characteristic (ROC)
curves, a 9 point change in the total score is recommended
as a threshold for determining meaningful improvement
Table 4 Clinical trial sample demographics (ADVANCE and
U-ACHIEVE)
max maximum, min minimum, SD standard deviation
Characteristic CD UC
Age (years)
N 850 248
Mean (SD) 37.5 (13.3) 42.3 (14.1)
Median 34.0 41
Min–max 16.0–79.0 18.0–75.0
Gender (n, %)
Male 390 (45.9%) 99 (39.9%)
Female 460 (54.1%) 149 (60.1%)
Evaluation of FACIT–Fatigue for Patients with CD and UC
(Supplementary Table4 and Supplementary Figs.1–3 in
the Online Resource 1: Supplementary tables and figures).
3.3 UC Psychometric Evaluation andScore
Interpretation
A total of 248 patients from the U-ACHIEVE study were
included in the psychometric analysis. Participants’ ages
ranged from 18 to 75years (mean=42.3years, SD=14.1
years); 60% of the sample was female (Table4).
3.3.1 UC: Quality ofCompletion andScore Distribution
Quality of completion for the UC psychometric analy-
sis population was high. At least 92.6% of the population
had complete data across analysis timepoints. The mean
FACIT–Fatigue total score was 28.7 at baseline (n=238,
SD=11.7) and 38.4 (n=214, SD=11.4) at week 8.
In general, respondents used the entire range of the
response scale for the FACIT–Fatigue items across assess-
ment timepoints, and scores trended toward improvement
over time.
3.3.2 UC: Inter‑item andItem‑Total Correlations
Week 8 correlations between FACIT–Fatigue items were
moderate to strong in magnitude (from 0.49 to 0.94). Similar
to the CD results, the highest correlations were between items
5 (trouble starting things) and 6 (trouble finishing things) for
all timepoints. The lowest correlations were between items
8 (usual activities) and 9 (sleep during the day), which were
negatively correlated due to the reverse score direction of
the items. Item-total correlations across analysis timepoints
ranged between 0.52 and 0.86, exceeding the threshold for
acceptable item-total correlations of ≥0.3[26].
3.3.3 UC: Reliability
Overall, Cronbach’s α ranged from 0.94 to 0.96 from base-
line to week 8, exceeding the standard thresholds for accept-
able internal consistency. Removing items for the UC dataset
did not improve the α coefficient at any of the timepoints.
Among participants who (1)selected “no change” on the
PGIC at week 2 or (2)selected the same response at both
baseline and week 2 on UC-SQ item 6 (tired, lacking energy
Table 5 Spearman correlation coefficients between FACIT–Fatigue total score and other assessments at baseline, week 4, and week 12 for the
ADVANCE study for CD
CD Crohn’s disease, CDAI Crohn’s Disease Activity Index, EQ-5D-5L Five-level EQ-5D, EQ-5D-VAS EQ-5D Visual Analog Scale, FACIT–
Fatigue Functional Assessment of Chronic Illness Therapy–Fatigue, IBDQ Inflammatory Bowel Disease Questionnaire, PGIS Patient Global
Impression of Severity, SF-23v2® 36-Item Short-Form Survey, version 2, WPAI: CD Work Productivity and Activity Impairment Questionnaire:
Crohn’s Disease
+++ strong relationship (>0.70 but ≤0.90), ++ moderate relationship (>0.30 but ≤0.70), + weak relationship (≤0.30)
Instrument/score Hypothesized rela-
tionship to FACIT–
Fatigue
FACIT–Fatigue total score
Baseline (N=850) Week 4 (N=841) Week 12 (N=819)
CDAI ++ −0.31 (++) −0.48 (++) −0.56 (++)
PGIS ++ −0.58 (++) −0.62 (++) −0.64 (++)
IBDQ total score ++ 0.72 (+++) 0.77 (+++) 0.82 (+++)
SF-36v2® Physical component summary score ++ 0.59 (++) 0.63 (++) 0.70 (++)
EQ-5D-5L Mobility ++ − 0.44 (++) − 0.42 (++) − 0.43 (++)
EQ-5D-5L Self-care ++ −0.33 (++) −0.31 (++) −0.32 (++)
EQ-5D-5L Usual activities ++ −0.58 (++) −0.61 (++) −0.61 (++)
WPAI: CD presenteeism (impairment at work/reduced
on-the-job effectiveness)
++ −0.47 (++) −0.59 (++) −0.57 (++)
WPAI: CD work productivity loss (overall work impair-
ment/absenteeism plus presenteeism)
++ −0.47 (++) −0.58 (++) −0.52 (++)
WPAI: CD activity impairment ++ −0.60 (++) −0.66 (++) −0.66 (++)
EQ-5D-5L anxiety/depression + −0.41 (++) −0.54 (++) −0.53 (++)
EQ-5D-5L Pain/discomfort + −0.51 (++) −0.55 (++) −0.60 (++)
EQ-5D-VAS + 0.53 (++) 0.61 (++) 0.64 (++)
WPAI: CD absenteeism + −0.33 (++) −0.41 (++) −0.38 (++)
SF-36v2® mental component summary score + 0.57 (++) 0.63 (++) 0.73 (++)
E.V.Loftus Jr et al.
during the past week), the FACIT–Fatigue total showed ade-
quate test–retest reliability (ICC=0.908 and ICC=0.945
for PGIC and UC-SQ item 6 definitions of stability, respec-
tively). All items demonstrated acceptable test–retest reli-
ability for one or both stability definitions.
3.3.4 UC: Convergent andDiscriminant Validity
Among the UC population, the correlations between the
FACIT–Fatigue total score and scores on almost all con-
current measures were either as strong as or stronger than
Table 6 Known-groups comparisons for FACIT–Fatigue total score at week 12 for the ADVANCE study for CD (N=841)
CD Crohn’s disease, CDAI Crohn’s Disease Activity Index, FACIT–Fatigue Functional Assessment of Chronic Illness Therapy–Fatigue, IBDQ
Inflammatory Bowel Disease Questionnaire, SD standard deviation, WPAI: CD Work Productivity and Activity Impairment Questionnaire:
Crohn’s Disease
* p-value <0.05
Comparison group nMean (SD) Median p-Value
CDAI
Remission: CDAI<150 326 41.41 (8.50) 43.50 <0.001*
Non-remission: CDAI≥150 429 30.22 (11.72) 31.00
IBDQ total score
IBDQ total score ≥170 remission 380 43.45 (6.17) 45.00 <0.001*
IBDQ total score <170 nonremission 398 26.88 (10.24) 27.00
PGIS
0 Absent: No symptoms 67 46.45 (4.45) 47.00 <0.001*
1 Minimal: Can be easily ignored without symptoms 203 42.04 (8.17) 44.00
2 Mild: Can be ignored with effort 163 36.02 (9.23) 37.00
3 Moderate: Cannot be ignored but does not influence my daily activities 123 34.59 (9.39) 36.00
4 Moderately severe: Cannot be ignored and occasionally limits my daily activities 129 27.50 (9.67) 27.00
5 Severe: Cannot be ignored and often limits my concentration on daily activities 66 21.02 (9.89) 19.50
6 Very severe: Cannot be ignored and markedly limits my daily activities 22 15.86 (12.65) 12.00
Table 7 Anchor-based estimates FACIT–Fatigue score by PGIS-stratified anchor categories from baseline to week 12 for the ADVANCE study
for CD
CD Crohn’s disease, FACIT–Fatigue Functional Assessment of Chronic Illness Therapy–Fatigue, PGIS Patient Global Impression of Severity,
SD standard deviation
a The within-group p-value is from a Wilcoxon Signed Rank test on change scores at each level of PGIS (seven level) response
b The between-groups p-value is from a Kruskal–Wallis testing distributional shift in change scores between PGIS (seven level) response groups
Bolding represents the primary PGIS anchor definitions for evaluation of the change in the FACIT-Fatiguescore: “1 point improvement” and “2
point improvement”
* p-value <0.05
Score Change in PGIS (base-
line to week 12)
NBaseline
Mean (SD)
Week 12
Mean (SD)
Mean change (SD) Within-
group
p-valuea
Between-groups
p-valueb
FACIT–Fatigue total
score
Most improved (4 point
or greater improve-
ment)
120 20.09 (10.44) 41.90 (8.95) 21.81 (10.54) <0.001* <0.001*
Much improved (3
point improvement)
158 25.62 (10.55) 39.82 (8.74) 14.20 (9.61) <0.001*
More improved (2
point improvement) 148 25.87 (10.61) 35.90 (10.45) 10.03 (8.88) <0.001*
Improved (1 point
improvement) 161 26.14 (11.23) 32.30 (11.26) 6.17 (7.74) <0.001*
No change 118 26.64 (12.18) 27.54 (12.85) 0.90 (7.64) 0.042*
Worsened (1 point or
greater deterioration)
39 25.08 (11.99) 25.38 (10.89) 0.31 (11.69) 0.630
Evaluation of FACIT–Fatigue for Patients with CD and UC
expected in the correct directions, demonstrating evidence
of convergent validity. Exceptions were the self-care domain
of the EQ-5D-5L and the Adapted Mayo Score, which had
weaker than expected correlation with the FACIT–Fatigue
total score in the UC sample. Weak correlations were
hypothesized between the FACIT–Fatigue and the meas-
ures used for discriminant validity analyses (EQ-5D-5L, SF-
36v2®, IBDQ, and WPAI: UC), but moderate relationships
were observed, indicating fatigue is more strongly related
to the constructs assessed by the discriminant measures
than expected. Detailed results are presented in Table9,
summarizing the hypothesized relationships between the
FACIT–Fatigue and other measures, and the observed cor-
relations between them.
3.3.5 UC: Known‑Groups Analysis
FACIT–Fatigue total scores demonstrated a 0.9–3.6 point
difference between groups classified as remission versus
nonremission on the Adapted Mayo Score, which was not
statistically significant at week 2 but was statistically signifi-
cant at week 8 (Table10).
3.3.6 UC: Sensitivity toChange
Moderate-to-strong correlations were observed (0.45–0.76)
between the FACIT–Fatigue change score and change scores
on concurrent measures, except for the EQ-5D-5L mobil-
ity domain, EQ-5D-5L self-care domain, and WPAI:UC
work-time-missed domains. These were weakly correlated
with the FACIT–Fatigue change score (0.20, 0.20, and 0.35,
respectively). Stronger correlations were observed between
FACIT–Fatigue change scores and conceptually similar
measures such as the SF-36v2 Physical Component Sum-
mary and IBDQ Total Score (0.61 and 0.76, respectively).
3.3.7 UC: Interpretation ofScores—Anchor‑Based Methods
andSupportive Analyses
The FACIT–Fatigue total score improved in parallel with
each PGIC anchor level. Anchor groups on the PGIC (“min-
imally improved” or “much improved”) generated a total
score change of 4.24–9.33 points for phase 2b U-ACHIEVE
substudy 1 (Table11).
Considering all additional evidence (results for eCDF,
PDF, and ROC curves are in Online Resource 1: Supple-
mentary tables and figures), a change on the FACIT–Fatigue
between 4 and 9 points can be recommended as meaning-
ful improvement thresholds (i.e., an estimate of MWPC).
Results are presented in Supplementary Table5 and Supple-
mentary Figs.4–6 in the Online Resource 1: Supplementary
tables and figures.
4 Discussion/limitations
This study demonstrated the importance and relevance
of fatigue among IBD patients as well as the validity of
the FACIT–Fatigue in measuring fatigue in moderate-to-
severe IBD. To the knowledge of the authors, these were
the first studies to thoroughly evaluate the content validity
of the FACIT–Fatigue among adolescent and adult patients
with moderately-to-severely active CD and UC separately.
Table 8 Anchor-based estimates FACIT–Fatigue score by PGIC-stratified anchor categories from baseline to week 12 for the ADVANCE study
for CD
CD Crohn’s disease, PGIC Patient Global Impression of Change, SD standard deviation
a The within-group p-value is from a Wilcoxon signed rank test on change scores at each level of PGIC response
b The between-groups p-value is from a Kruskal–Wallis testing distributional shift in change scores between PGIC (seven level) response groups
Bolding represents the primary PGIC anchor definitions for evaluation of the change in the FACIT-Fatiguescore: “Much Improved” or “Mini-
mally improved”
* p-value <0.05
Score PGIC at week 12 NBaseline
Mean (SD)
Week 12
Mean (SD)
Mean change (SD) Within-
group
p-valuea
Between-groups p
-valueb
FACIT–Fatigue total
score
Very much improved
(PGIC=1)
118 26.41 (11.90) 44.44 (7.50) 18.03 (11.54) <0.001* <0.001*
Much improved
(PGIC=2) 288 24.77 (10.76) 38.40 (9.58) 13.63 (10.20) <0.001*
Minimally improved
(PGIC=3) 192 25.33 (11.28) 32.17 (10.34) 6.83 (9.34) <0.001*
No change (PGIC=4) 110 24.75 (11.96) 27.60 (12.31) 2.85 (8.23) <0.001*
Worsened (PGIC>4) 52 23.44 (11.80) 22.21 (11.23) −1.23 (8.53) 0.451
E.V.Loftus Jr et al.
Results from the qualitative study provide details on the
experience of fatigue and evidence supporting the use of
the FACIT–Fatigue in these populations.
Specifically, during concept elicitation and cognitive
debriefing interviews, all but one participant [i.e., over 98%
of participants (n=62/63)] reported experiencing fatigue.
Participants reported cognitive, physical, social, work/
school, sleep, leisure, emotional, and relationship impacts,
as well as impacts to activities of daily living due to fatigue.
Patients rated fatigue above a 6.0 on a scale from 0 (not
at all bothersome) to 10 (most bothersome). In general,
patients interpreted the FACIT–Fatigue content as intended
and reported the questionnaire was relevant and compre-
hensive. However, a subset of CD patients (n=6/63, 9.5%,
mostly adolescents) reported being unfamiliar with the term
“fatigue” (although upon receiving clarification, all of them
reported experiencing it). This result poses a limitation to
the content validity of the questionnaire for use with ado-
lescents. Researchers may consider providing a definition of
fatigue when administering the FACIT–Fatigue, especially
within adolescent populations. If individuals are not pro-
vided with a definition of the term, there is risk that they
may not be responding about the intended concept. This
study was limited in sample size and greater research may
be needed to better understand the extent of adolescent
understanding of fatigue and whether it may vary between
CD and UC.
Further, analyses using two clinical trial datasets demon-
strated that the scores on the FACIT–Fatigue had (1)good
internal consistency and acceptable test–retest reliability
and (2)acceptable convergent and discriminant validity,
known-groups results, and sensitivity to change. These
results aligned with findings from Tinsley etal. [17] where
the reliability and validity of the FACIT–Fatigue was evalu-
ated among adults with CD or UC. While floor effects were
observed for items 10 and 11 among CD patients at base-
line, improvements in item-level scores still occurred over
time. It may be that the floor effects observed at baseline
do not indicate restriction in the FACIT–Fatigue response
scale but likely reflect participants’ actual levels of fatigue-
related impact. Overall, although the FACIT–Fatigue was
not developed for use in IBD, results from this research pro-
vide evidence that fatigue is relevant and highly important to
CD and UC patients and that the FACIT–Fatigue has sound
psychometric properties in this patient population.
This study was also the first to summarize thresholds
for MWPC for the FACIT–Fatigue total score for patients
with moderately-to-severely active CD or UC and provides
Table 9 Spearman correlation coefficients between FACIT–Fatigue total score and other assessments at baseline, week 2, and week 8 for UC for
U-ACHIEVE substudy 1 for UC
EQ-5D-5L Five-level EQ-5D, EQ-5D-VAS EQ-5D Visual Analog Scale, FACIT–Fatigue Functional Assessment of Chronic Illness Therapy–
Fatigue, IBDQ Inflammatory Bowel Disease Questionnaire, SF-23v2® 36-Item Short-Form Survey, version 2, UC ulcerative colitis, UC-SQ
Ulcerative Colitis Symptoms Questionnaire, WPAI: UC Work Productivity and Activity Impairment Questionnaire: Ulcerative Colitis
+++ strong relationship (≥0.70 but ≤0.90), ++ moderate relationship (>0.30 but ≤0.70), + weak relationship (≤0.30)
Assessment/score Hypothesized relation-
ship to FACIT–Fatigue
FACIT–Fatigue total score
Baseline (N=248) Week 2 (N=248) Week 8 (N=231)
Adapted Mayo Score + −0.19 (+) – −0.38 (++)
IBDQ Bowel symptoms + 0.61 (++) 0.61 (++) 0.59 (++)
IBDQ Systemic symptoms + 0.78 (+++) 0.80 (+++) 0.73 (+++)
IBDQ Emotional function + 0.68 (++) 0.71 (+++) 0.73 (+++)
IBDQ Social function + 0.64 (++) 0.65 (++) 0.66 (++)
EQ-5D-5L Mobility ++ −0.34 (++) −0.36 (++) −0.40 (++)
EQ-5D-5L Self-care ++ −0.26 (+) −0.27 (+) − 0.30 (+)
EQ-5D-5L Usual activities ++ −0.66 (++) −0.68 (++) −0.70 (+++)
EQ-5D-5L Pain/discomfort + −0.58 (++) −0.57 (++) −0.59 (++)
EQ-5D-5L Anxiety/depression + −0.55 (++) −0.58 (++) −0.54 (++)
EQ-5D-VAS + 0.60 (++) 0.63 (++) 0.71 (+++)
SF-36v2® Physical component summary + 0.62 (++) 0.64 (++) 0.65 (++)
SF-36v2® Mental component summary + 0.70 (+++) 0.67 (++) 0.73 (+++)
UC-SQ + −0.68 (++) −0.66 (++) −0.66 (++)
WPAI: UC absenteeism (work time missed) + −0.45 (++) − 0.43 (++) −0.32 (++)
WPAI: UC presenteeism (impairment at work) + −0.58 (++) −0.71 (+++) −0.66 (++)
WPAI: UC overall work impairment + −0.63 (++) −0.70 (+++) −0.62 (++)
WPAI: UC activity productivity + −0.58 (++) −0.65 (++) −0.72 (+++)
Evaluation of FACIT–Fatigue for Patients with CD and UC
guides for interpreting within-patient score changes in
this population. A variety of quantitative methods were
employed to generate reliable estimates, and during the
qualitative interviews, patients described meaningful
changes in fatigue, which provided context to aid in the
interpretation of quantitative MWPC estimates. MWPC
estimates differed by condition, and the threshold was
higher for the analyses of the CD clinical trial data com-
pared with the UC data (7–10 point improvement for CD
and 4–9 point improvement for UC). This aligns with the
qualitative results; the degree of bother of fatigue was gen-
erally higher among CD patients than UC patients, which
may explain such differences. The ranges of MWPC esti-
mates for the FACIT–Fatigue can be used as a guide for
future clinical research for each condition, but it should
be noted that if the target patient population or other trial
characteristics (e.g., number of study weeks, administra-
tion schedule) differs, MWPC estimates may be impacted
as well. Therefore, these ranges are presented to a start-
ing place or aid in developing endpoints based on the
FACIT–Fatigue scores for future research in CD and
UC. After triangulating all evidence including ROC and
eCDF graphs, an improvement of 9 and 5 points on the
FACIT–Fatigue for CD and UC, respectively, could be con-
sidered for future clinical research in these target patient
populations. These MWPC estimates were applied in the
ADVANCE and phase 3 U-ACHIEVE trials, which demon-
strated significantly greater portions of patients in treatment
groups versus placebo were responders[27–30].
Data for psychometric evaluation of the FACIT–Fatigue
came from two randomized clinical trials. The sample sizes
for psychometric analyses were large, and data were of high
quality. Patients had little missing data on the FACIT–Fatigue
across timepoints, and data were captured in a standardized
way during the trials.
While relying on data from clinical trials provided large
samples of quality data, the test–retest analysis populations
were limited because, due to the nature of clinical research,
there were challenges to identifying a subgroup of “stable”
patients (i.e., patients with unchanging disease status) across
timepoints to include in analyses primarily due to the long
duration between assessment timepoints and the administra-
tion of treatment. ICCs for CD ranged from slightly below to
minimally above the threshold for acceptability. Given con-
straints in the available data, the ICCs for CD and UC should
be interpreted with caution, as they likely underestimate the
reliability of the scores. Using a fatigue-specific anchor (UC-
SQ item 6) to identify stable patients in UC yielded higher
ICCs. Additional research should evaluate the test–retest reli-
ability of the FACIT–Fatigue in CD and UC patients using a
larger sample of stable patients and define stability specific
to fatigue.
Table 10 Known-groups comparisons for FACIT–Fatigue total score
at week 2 and week 8 for U-ACHIEVE substudy 1 for UC
Clinical remission on the Adapted Mayo is defined as having a stool
frequency subscore of 0 or 1 and not greater than the baseline score
and rectal bleeding subscore of 0 and endoscopic subscore of 0 or 1;
nonremission is defined as individuals not in the remission state
FACIT–Fatigue Functional Assessment of Chronic Illness Therapy–
Fatigue, IBDQ Inflammatory Bowel Disease Questionnaire, SD
standard deviation, UC ulcerative colitis, UC-SQ Ulcerative Colitis
Symptoms Questionnaire
a p-Values are from a Mann–Whitney U test for comparison of means
between two groups and a Kruskal–Wallis test for more than two groups
* p-value <0.05
Comparison group NMean (SD) Median p-Valuea
Week 2 (N=248)
Adapted Mayo Score
Clinical remission 28 34.5 (11.7) 34.5 0.378
Nonremission 173 33.6 (12.1) 36.0
UC-SQ Item 6. During the past week, have you felt tired or lacking
energy?
Not at all 12 49.3 (1.6) 49.5 <0.001*
A little bit 37 41.7 (7.0) 44.0
Somewhat 28 34.4 (7.7) 35.0
Quite a bit 23 23.2 (7.6) 26.0
Very much 16 12.9 (7.7) 13.5
IBDQ Item 2. How often has the feeling of fatigue or of being tired
and worn out been a problem for you during the last 2 weeks?
All of the time 19 13.1 (7.0) 13.0 <0.001*
Most of the time 35 22.4 (8.8) 22.0
A good bit of the time 45 27.8 (8.1) 27.0
Some of the time 52 36.4 (7.7) 38.0
A little of the time 49 41.4 (7.0) 43.0
Hardly any of the time 29 45.6 (4.5) 47.0
None of the time 9 45.0 (7.8) 49.0
Week 8 (N=248)
Adapted Mayo Score
Clinical remission 27 41.6 (10.0) 44.0 0.041
Nonremission 167 38.0 (11.5) 42.0
UC-SQ Item 6. During the past week, have you felt tired or lacking
energy?
Not at all 25 49.0 (4.1) 50.0 <0.001*
A little bit 55 41.9 (5.3) 42.0
Somewhat 22 37.7 (8.7) 39.5
Quite a bit 13 26.5 (7.9) 24.0
Very much 10 16.6 (10.8) 11.5
IBDQ Item 2. How often has the feeling of fatigue or of being tired
and worn out been a problem for you during the last 2 weeks?
All of the time 11 17.0 (13.6) 12.0 <0.001*
Most of the time 15 21.2 (9.3) 19.0
A good bit of the time 28 30.7 (9.4) 30.0
Some of the time 41 39.1 (7.1) 39.0
A little of the time 54 40.9 (6.6) 42.0
Hardly any of the time 43 45.6 (5.1) 47.0
None of the time 22 49.4 (3.2) 50.0
E.V.Loftus Jr et al.
Additional paths forward for future research may include
comparing the FACIT–Fatigue with other multi-item fatigue
scales, evaluating the FACIT–Fatigue among individuals with
mild disease, and evaluating the differences in how fatigue
impacts quality of life between CD and UC.
5 Conclusions
This study demonstrated the relevance and importance of
fatigue to patients’ disease experience in moderately-to-
severely active CD and UC. Cognitive debriefing results
provided evidence that the FACIT–Fatigue can be inter-
preted as intended and has comprehensive coverage of
fatigue and fatigue-related impacts among adults with
moderately-to-severely active CD or UC; however, some
caution may be needed when using this questionnaire
with adolescents who may have difficulty understanding
the word “fatigue.” Psychometric analyses indicate that
the scores generated by the FACIT–Fatigue demonstrate
acceptable reliability, construct validity, and sensitivity
to change. Further, a 7–10 point improvement for CD and
a 4–9 point improvement for UC on the FACIT–Fatigue
total score may represent meaningful improvements and
may serve as context for future research in individuals with
moderate-to-severe CD and UC.
Supplementary Information The online version contains supplemen-
tary material available at https:// doi. org/ 10. 1007/ s41669- 023- 00419-w .
Acknowledgements The authors would like to thank Alan Shields,
Roger Lamoureux, Emma Pranschke, Ryan Hoffman, Naomi Suminski,
Doug Huntington, Jeff Mcdonald, and Alejandro Moreno-Koehler for
their important contributions to the completion of this research.
Declarations
Authorship statement EVL is acting as the submission's guarantor.
AbbVie and Adelphi Values authors collaborated on the conception
of the study design, conduct of research, data collection, analysis, and
interpretation of data. All authors had access to the data results and par-
ticipated in the development, review, and approval of this manuscript.
No honoraria or payments were made for authorship. All authors have
approved the final version of the article, including the authorship list.
Declaration of personal interests Edward Loftus has received
financial support for research from AbbVie, Bristol-Myers Squibb,
Celgene, Genentech, Gilead, Gossamer Bio, Janssen, Pfizer, Receptos,
Robarts Clinical Trials, Takeda, Theravance, and UCB; and
received consultancy fees from AbbVie, Amgen, Arena, Boehringer
Ingelheim, Bristol-Myers Squibb, Calibr, Celgene, Eli Lilly, Fresenius
Kabi, Genentech, Gilead, Gossamer Bio, Iterative Scopes, Janssen,
Morphic, Ono Pharma, Pfizer, Protagonist, Scipher Medicine, Sun
Pharma, Surrozen, Takeda, and UCB. Ashwin Ananthakrishnan has
received funding from the US National Institutes of Health, served on
scientific advisory boards for Gilead and received research funding
from Pfizer. Wan-Ju Lee, Yuri Sanchez Gonzalez, Kristina Fitzgerald,
Kori Wallace, and Wen Zhou are full-time employees of AbbVie and
may hold AbbVie stock and/or stock options. Leighann Litcher-Kelly,
Sarah Ollis, and Sylvia Su are employed by Adelphi Values LLC, which
received payment from AbbVie Inc. to support the research activities
presented in this publication. Silvio Danese has received consultancy
fees from AbbVie, Allergan, Amgen, AstraZeneca, Biogen, Boehringer
Ingelheim, Celgene, Celltrion, Ferring, Gilead, Hospira, Janssen,
Johnson & Johnson, MSD, Mundipharma, Pfizer, Roche, Sandoz,
Table 11 Anchor-based estimates FACIT–Fatigue score by PGIC stratified anchor categories for U-ACHIEVE substudy 1 for UC (baseline to
week 8)
FACIT–Fatigue Functional Assessment of Chronic Illness Therapy–Fatigue, PGIC Patient Global Impression of Change, SD standard deviation
a The within-group p-value is from a Wilcoxon signed rank test on change scores at each level of PGIC response
b The between-groups p-value is from a Kruskal–Wallis testing distributional shift in change scores between PGIC (seven level) response groups
c There were no individuals who endorsed “Very much worsened” (PGIC=7)
Bolding represents the primary PGIC anchor definitions for evaluation of the change in the FACIT-Fatiguescore: “Much Improved” or “Mini-
mally improved”
* p-value <0.05
Score PGIC at week 12cNBaseline
Mean (SD)
Week 12
Mean (SD)
Mean change (SD) Within-
group
p-valuea
Between-groups
p-valueb
U-ACHIEVE substudy 1
FACIT–Fatigue total
score
Very much improved
(PGIC=1)
59 27.69 (12.91) 44.64 (8.96) 16.95 (11.40) <0.001* <0.001*
Much improved
(PGIC=2) 64 30.94 (10.66) 40.27 (7.73) 9.33 (8.44) <0.001*
Minimally improved
(PGIC=3) 38 28.05 (11.55) 32.29 (12.74) 4.24 (7.66) <0.001*
No change (PGIC=4) 29 31.34 (11.57) 35.17 (10.19) 3.83 (6.09) 0.002*
Worsened (PGIC=5) 7 29.14 (12.38) 30.29 (11.93) 1.14 (6.36) 0.500
Much worsened
(PGIC=6)
8 24.88 (12.92) 22.38 (15.35) −2.50 (4.90) 0.188
Evaluation of FACIT–Fatigue for Patients with CD and UC
Takeda, TiGenix, UCB, and Vifor Pharma. No honoraria or payments
were made for authorship.
Declaration of funding interests This work was supported by AbbVie,
Inc. AbbVie participated in the study design, research, data collection,
analysis and interpretation of data, writing, reviewing, and approving
the publication. Writing support was provided by Naomi Suminski
and Emma Pranschke of Adelphi Values and funded by AbbVie, Inc.
Data availability statement The data that support the findings of this
research may be available from AbbVie, but restrictions apply to the
availability of these data, which were used under license for the current
research, and so are not publicly available. Data are however avail-
able from the authors upon reasonable request and with permission
of AbbVie.
Ethics approval Ethical approval for the interview study was obtained
via a centralized independent review board. Ethical approval for the
clinical trials was obtained separately.
Consent to participate All interview and clinical trial participants
provided consent (or assent with parental/guardian consent, as appro-
priate).
Consent for publication (from patients/participants) Not applicable.
Code availability May be available upon reasonable request.
Open Access This article is licensed under a Creative Commons Attri-
bution-NonCommercial 4.0 International License, which permits any
non-commercial use, sharing, adaptation, distribution and reproduction
in any medium or format, as long as you give appropriate credit to the
original author(s) and the source, provide a link to the Creative Com-
mons licence, and indicate if changes were made. The images or other
third party material in this article are included in the article's Creative
Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article's Creative Commons
licence and your intended use is not permitted by statutory regula-
tion or exceeds the permitted use, you will need to obtain permission
directly from the copyright holder. To view a copy of this licence, visit
http:// creat iveco mmons. org/ licen ses/ by- nc/4. 0/.
References
1. Danese S, Fiocchi C. Ulcerative colitis. N Engl J Med.
2011;365(18):1713–25.
2. National Institute of Diabetes and Digestive and Kidney Dis-
eases. Ulcerative Colitis. 09/2014; https:// www. niddk. nih.
gov/ health- infor mation/ diges tive- disea ses/ ulcer ative- colit is.
Accessed 29 Mar 2022.
3. Panaccione R. Mechanisms of inflammatory bowel disease.
Gastroenterol Hepatol. 2013;9(8):529–32.
4. Gohil K, Carramusa B. Ulcerative colitis and Crohn’s disease.
P T. 2014;39(8):576–7.
5. Villoria A, García V, Dosal A, etal. Fatigue in out-patients with
inflammatory bowel disease: prevalence and predictive factors.
PLoS ONE. 2017;12(7): e0181435.
6. Klusmann B, Fleer J, Trovote KA, etal. Trajectories of
fatigue in inflammatory bowel disease. Inflamm Bowel Dis.
2021;27(12):1919–30.
7. D’Silva A, Fox DE, Nasser Y, etal. Prevalence and risk fac-
tors for fatigue in adults with inflammatory bowel disease: a
systematic review with meta-analysis. Clin Gastroenterol Hepa-
tol. 2022;20(5):994–1009.
8. Bager P, Befrits R, Wikman O, etal. Fatigue in out-patients
with inflammatory bowel disease is common and multifactorial.
Aliment Pharmacol Ther. 2012;35(1):133–41.
9. Borren NZ, van der Woude CJ, Ananthakrishnan AN. Fatigue
in IBD: epidemiology, pathophysiology and management. Nat
Rev Gastroenterol Hepatol. 2019;16(4):247–59.
10. Jelsness-Jørgensen LP, Bernklev T, Henriksen M, Torp R,
Moum BA. Chronic fatigue is more prevalent in patients with
inflammatory bowel disease than in healthy controls. Inflamm
Bowel Dis. 2011;17(7):1564–72.
11. Nocerino A, Nguyen A, Agrawal M, Mone A, Lakhani K, Swa-
minath A. Fatigue in inflammatory bowel diseases: etiologies
and management. Adv Ther. 2020;37(1):97–112.
12. FACIT Fatigue Scale (Version 4). 1997; http:// www. ser. es/ wp-
conte nt/ uploa ds/ 2015/ 03/ FACIT-F_ INDICE. pdf. Accessed 29
Mar 2022.
13. Guyatt G, Mitchell A, Irvine EJ, etal. A new measure of health
status for clinical trials in inflammatory bowel disease. Gastro-
enterology. 1989;96(3):804–10.
14. Yellen SB, Cella DF, Webster K, Blendowski C, Kaplan E.
Measuring fatigue and other anemia-related symptoms with the
Functional Assessment of Cancer Therapy (FACT) measure-
ment system. J Pain Symptom Manage. 1997;13(2):63–74.
15. Meysami A, Yari H, Haghighat S, Montazeri A. The Functional
Assessment of Chronic Illness Therapy-Fatigue (FACIT-Fatigue
Subscale): validity and reliability of the Iranian version. Oncol
Res Treat. 2017;40(12):789–93.
16. FACIT.org. FACIT-F: Functional Assessment of Chronic Ill-
ness Therapy-Fatigue. https:// www. facit. org/ measu res/ FA CIT-F.
Accessed 29 Mar 2022.
17. Tinsley A, Macklin EA, Korzenik JR, Sands BE. Validation
of the functional assessment of chronic illness therapy-fatigue
(FACIT-F) in patients with inflammatory bowel disease. Ali-
mentPharmacol Ther. 2011;34(11–12):1328–36.
18. Conley S, Proctor DD, Jeon S, Sandler RS, Redeker NS. Symp-
tom clusters in adults with inflammatory bowel disease. Res
Nurs Health. 2017;40(5):424–34.
19. Williet N, Sarter H, Gower-Rousseau C, etal. Patient-reported
outcomes in a French Nationwide Survey of inflammatory bowel
disease patients. J Crohns Colitis. 2017;11(2):165–74.
20. Loftus EV, Feagan BG, Colombel JF, etal. Effects of adali-
mumab maintenance therapy on health-related quality of life of
patients with Crohn’s disease: patient-reported outcomes of the
CHARM trial. Am J Gastroenterol. 2008;103(12):3132–41.
21. Farrell D, Artom M, Czuber-Dochan W, Jelsness-Jør-
gensen LP, Norton C, Savage E. Interventions for fatigue in
inflammatory bowel disease. Cochrane Database Syst Rev.
2020;4(4):Cd012005.
22. Saraiva S, Cortez-Pinto J, Barosa R, etal. Evaluation of fatigue
in inflammatory bowel disease—a useful tool in daily practice.
Scand J Gastroenterol. 2019;54(4):465–70.
23. D’Haens G, Panaccione R, Baert F, et al. Risankizumab as
induction therapy for Crohn’s disease: results from the phase
3 ADVANCE and MOTIVATE induction trials. Lancet.
2022;399(10340):2015–30.
24. Sandborn WJ, Ghosh S, Panes J, etal. Efficacy of upadacitinib
in a randomized trial of patients with active ulcerative colitis.
Gastroenterology. 2020;158(8):2139-2149.e2114.
25. Ghosh S, Sanchez Gonzalez Y, Zhou W, etal. Upadacitinib
treatment improves symptoms of bowel urgency and abdominal
pain, and correlates with quality of life improvements in patients
with moderate to severe ulcerative colitis. J Crohn’s Colitis.
2021;15:2022.
E.V.Loftus Jr et al.
26. Thompson B, Vacha-Haase T. Psychometrics is datametrics: the
test is not reliable. Educ Psychol Measur. 2000;60(2):174–95.
27. Danese S, Tran J, D’Haens G, etal. OP08 The effects of main-
tenance therapy with upadacitinib on abdominal pain, bowel
urgency, and fatigue in patients with moderately to severely active
Ulcerative Colitis: Phase 3 U-ACHIEVE maintenance results. J
Crohn’s Colitis. 2022;16(Supplement_1):i008-i010.
28. Danese S, D’Haens G, Rubin D, etal. Patients with ulcerative coli-
tis report improvements in abdominal pain, bowel urgency, and
fatigue with 8-week upadacitinib treatment in two phase 3 trials:
U-ACHIEVE and U-ACCOMPLISH. United Eur Gastroenterol J.
2021;9(S8):25.
29. Peyrin-Biroulet L, Louis E, Ghosh S, etal. DOP88 Effect of
risankizumab on patient-reported outcomes in patients with
Crohn’s Disease who had an inadequate response or intoler-
ance to conventional and/or biologic treatments: Results from
phase 3 MOTIVATE and ADVANCE trials. J Crohn’s Colitis.
2021;15(1):S119–21.
30. Peyrin-Biroulet L, Ghosh S, Lee SD, etal. Effect of risankizumab
on health-related quality of life in patients with Crohn’s disease:
results from phase 3 MOTIVATE, ADVANCE and FORTIFY
clinical trials. Aliment Pharmacol Ther. 2023;57(5):496–508.
31. Best WR, Becktel JM, Singleton JW, Kern F Jr. Development of
a Crohn’s disease activity index. National Cooperative Crohn’s
Disease Study. Gastroenterology. 1976;70(3):439–44.
32. Schroeder KW, Tremaine WJ, Ilstrup DM. Coated oral 5-amino-
salicylic acid therapy for mildly to moderately active ulcerative
colitis. A randomized study. N Engl J Med. 1987;317(26):1625–9.
33. Reilly MC, Gerlier L, Brabant Y, Brown M. Validity, reli-
ability, and responsiveness of the work productivity and activ-
ity impairment questionnaire in Crohn’s disease. Clin Ther.
2008;30(2):393–404.
34. Reilly MC, Zbrozek AS, Dukes EM. The validity and reproduc-
ibility of a work productivity and activity impairment instrument.
Pharmacoeconomics. 1993;4(5):353–65.
35. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health
survey (SF-36). I. Conceptual framework and item selection. Med
Care. 1992;30(6):473–83.
36. Herdman M, Gudex C, Lloyd A, Janssen M, Kind P, Parkin D.
Development and preliminary testing of the new five-level version
of EQ-5D (EQ-5D-5L). Qual Life Res. 2011;20(10):1727–36.
37. Wang L, Zhang Z, McArdle JJ, Salthouse TA. Investigating ceil-
ing effects in longitudinal data analysis. Multivar Behav Res.
2009;43(3):476–96.
38. Hinkle DE, Jurs SG, Wiersma W. Applied statistics for the behav-
ioral sciences. 2nd ed. Boston: Houghton Mifflin; 1988.
39. Cronbach LJ. Coefficient alpha and the internal structure of tests.
Psychometrika. 1951;16(3):297–334.
40. Bland JM, Altman DG. Cronbach’s alpha. BMJ.
1997;314(7080):572.
41. Weir J. Quantifying test-retest reliability using the intraclass
correlation coefficient and the SEM. J Strength Cond Res.
2005;19(1):231–40.
42. Norman GR, Sloan JA, Wyrwich KW. Interpretation of changes in
health-related quality of life: the remarkable universality of half a
standard deviation. Med Care. 2003;41(5):582–92.
43. Basson M. How is ulcerative colitis (UC) defined based on the
extent of disease? https:// www. medsc ape. com/ answe rs/ 183084-
13754/ how- is- ulcer ative- colit is- uc- defin ed- based- on- the- extent-
of- disea se. Accessed 29 Mar 2022.
Authors and Aliations
EdwardV.LoftusJr1 · AshwinN.Ananthakrishnan2· Wan‑JuLee3· YuriSanchezGonzalez3· KristinaAFitzgerald3·
KoriWallace3· WenZhou3· LeighannLitcher‑Kelly4· SarahB.Ollis4· SylviaSu4· SilvioDanese5
* Edward V. Loftus Jr
loftus.edward@mayo.edu
1 Mayo Clinic College ofMedicine andScience, Rochester,
MN, USA
2 Massachusetts General Hospital, Boston, MA, USA
3 AbbVie Inc, Chicago, IL, USA
4 Adelphi Values, Boston, MA, USA
5 Humanitas University, Milan, Italy