close

Вход

Забыли?

вход по аккаунту

?

The face scaleA brief nonverbal method for assessing patient mood.

код для вставкиСкачать
906
BRIEF REPORT
THE FACE SCALE: A BRIEF, NONVERBAL METHOD FOR ASSESSING
PATIENT MOOD
CHRISTOPHER D. LORISH and RICHARD MAISIAK
Validity and reliability studies were conducted on
the Face Scale, a very brief, pictorial scale of mood
which uses a sequence of 20 faces and does not require
reading literacy. Correlational and experimental evidence of the Face Scale’s construct validity is presented,
as well as its test-retest reliability. Recommendations
are made for its use as a screening tool and for additional validity studies.
Depressed mood often occurs in arthritis patients (1,2) and may affect their response to the disease
and treatment (3,4). Although there are now several
instruinents designed to assess the status of arthritis
patients, few assess patient mood, and none make that
assessment in under 2 minutes. Several instruments,
such as the Beck Depression Inventory ( 5 ) or the Zung
Self-RatingDepression Scale (6),have been developed
solely to assess patient mood. Other instruments
which measure aspects of health status (7) may also
indirectly assess mood. Scales for assessment of arthritis patients tend to be based on verbal materials
and normally contain several stem sentences and
alternate responses. These require from 10-30 minutes
to complete, a serious limitation when used in a busy
clinic. Another limitation of the scales is that they
must be read to patients who have low literacy skills or
severe disabilities, and this further increases the time and
labor commitment needed to obtain the data.
-__
From the Office of Educational Development, School of
Medicine, University of Alabama at Birmingham.
Supported by NIH Multipurpose Arthritis Center grant P60
AM-20614.
Christopher D. Lorish, PhD; Richard Maisiak, PhD.
Address reprint requests to Christopher Lorish, PhD,
Office of Educational Development, Volker Hall, L210-C, Birmingham, AX, 35294. Master copies of the Face Scale may be obtained
from the authors.
Submitted for publication August 1, 1985; accepted in
revised form February 24, 1986.
Arthritis and Rheumatism, Vol. 29, No. 7 (July 1986)
Recent studies have shown that the number of
Americans who are functionally illiterate has been
underestimated and has been increasing (8). Data from
a recent study conducted in an arthritis clinic showed
that 35% of the patients had a grade-school education
only (9). Most patients had to be read questionnaires,
including those questionnaires designed for selfadministration. Therefore, the development of a
nonverbal measure of mood could be valuable in
minimizing the time needed to obtain data on mood
from readers and nonreaders alike, especially when a
battery of psychosocial data is needed.
Aitken and Zeally (10) developed the nonverbal
Visual Analogue Scale, a simple 100-mm line anchored
at its ends by “normal mood” and “extreme depression.” The scale correlated well with verbal measurement of mood and was easily completed by subjects.
The use of a simple line to assess mood, however,
lacks content validity, since it has little apparent
relationship to a patient’s visual cues of moods.
Sad facial expressions and weeping are common indicators of depression (5). Ekman et a1 (11,121
found that facial feature variations are universal, valid
indicators of mood, i.e., happiness and sadness, and
that mouth and eye variations were important for
distinguishing degrees of those emotions. Our purpose
was to assess the ability of the Face Scale, a brief,
nonverbal pictorial scale that uses images of the human face, to reliably and validly assess patient mood.
METHODS
Description of the Face Scale. The Face Scale
(Figure 1) contains 20 drawings of a single face,
arranged in serial order by rows, with each face
depicting a slightly different mood state. A graphic
BRIEF REPORTS
907
IKSTRCICTIONS Thr lace\ hclnu prr horn \cr! happt a1 Ihc lop 10 vcr\ rad 81 Ihe boltom. Chcrl
ihr face u h r h hrii <hoar ihc wa, you hatr fell i n d r lodm
1
2
3
4
5
6
7
8
9
10
I1
12
13
14
15
16
17
18
19
20
Figure 1. The Face Scale.
artist was consulted so that the faces would be portrayed as genderless and multiethnic. Subtle changes
in the eyes, eyebrows, and mouth were used to represent slightly different levels of mood. They are arranged in decreasing order of mood and numbered
from 1-20, with 1 representing the most positive mood
and 20 representing the most negative mood. As the
examiner pointed at the faces, the following instructions were given to each patient: “The faces below go
from very happy at the top to very sad at the bottom.
Point to the face which best shows the way you have
felt inside today.”
Testing procedures. The Face Scale was tested
in 2 studies: 1 to assess validity, and the other to
assess short-term reliability. In the first study, 174
consecutive rheumatoid arthritis patients at an inpatient rehabilitation facility were asked to complete 5
instruments, including the Face Scale, during a 9month period. The patients were admitted to the unit
for treatment during periods of flare, and were discharged after 10-14 days. Subjects had a mean age of
54 years, and an average of 10 years of education.
Sixty-four percent were female, 80% were white, 75%
had a functional status level (13) of 2, and 25% had a
functional status Ievel of 3.
Patients also completed various other moodrelated scales. The Sickness Impact Profile (SIP) (14)
uses 136 dichotomous items to assess function in 12
categories of daily living. A scale score is a ratio
computed by summing the value associated with each
item marked “yes” and dividing by the total values of
all items in the scale. The Emotional Behavior
subscale contains 9 items, all relative to mood. Higher
scores indicate greater dysfunction.
The Beck Depression Scale consists of 21 categories of symptoms or attitudes derived from a study
of depressed patients (5). Each category contains a
graded series of 4 or 5 self-evaluative statements
which reflect the range of severity of the symptom.
Seven items were eliminated because their content
reflected symptoms that were potentially diseaserelated. Values from 0-3 were assigned to each statement. The total score was the sum of the category
scores. Higher scores indicate depressed mood.
The Bradburn Affect Balance Scales consist of
a negative-affect scale and positive-affect scale (15).
Each scale consists of 5 true-false statements concerning positive or negative feelings. The score for each
scale is the sum of true responses to the statement
made by the respondent; therefore, scores may range
from 0-5. A total score was computed by subtracting
the positive scale score from 5 and adding it to the
negative scale score. Total scores range from 0-10;
higher scores indicate depressed mood.
The Pain Scale, a locally developed 10-point
visual analog scale, consists of a graduated line labeled
at one end “least pain experienced with arthritis” and at
the other end “worst pain experienced with arthritis.”
Patients voluntarily completed the Face Scale,
the Bradburn Affect Balance Scales, and the Pain
Scale on admission, at discharge, and at the first
followup clinic visit 1-3 months after discharge. Patients also voluntarily completed the Beck Depression
Scale only when admitted and when discharged, and
the Sickness Impact Profile only at admission and at
the first followup clinic visit (approximately 6 weeks
post-discharge). The specific order of test adrninistrations within the same interviewing session varied
randomly from patient to patient. All patients were
read the instructions and all the questions for each
instrument, except for the Face Scale, by the same
interviewer. The interviewer was not part of the
clinical staff, and had practiced administering the
instruments in pilot studies before the actual data
BRIEF REPORTS
Table 1. Coefficients of the correlation of Face Scale scores with
other imeasures of mood and status of arthritis patients (n = 174)
Measures
r
Mood-related measures
Becjk Depression Scale
Sickness Impact Profile
Eimotional Behavior
Bradburn Positive Affect
Bradburn Negative Affect
SeKreported pain rating
Patient status measures
Mobility
Social interaction
A s
Years of schooling
0.49*
0.49*
-0.37*
0.37*
0.44*
0.24*
0.35*
0.04
-0.11
* P < 0.01.
collection. The administration of the SIP, Beck, and
Bradburn scales by an interviewer should yield results
similar to those seen with self-administration (5,15,16).
T o assess test-retest reliability, a similar group
of 37 rheumatoid arthritis inpatients was included in a
second study. This group was asked to voluntarily
complete the Face Scale when admitted to the unit and
then again 45 minutes later, using the same instruction
set as, the first study group. A 45-minute period was
chosen because a longer time period would have
increased chances for mood changes due to an interspersed activity, such as an assuring doctor’s visit
with the patient (which might result in an enhanced
mood) o r an unpleasant blood-drawing (resulting in a
decreased mood).
RESULTS
Correlational evidence. One indication of a
test’s validity is its pattern of correlation with other
measures. A valid test tends to show convergent
validity by correlating well with other known meaTabk 2.
sures of the same concept. Discriminant validity is
shown by a lack of correlation with known measures
of unrelated concepts. The Pearson correlations between the Face Scale and other instruments which
measure mood and mental status are presented in
Table 1 . The results show statistically significant ( P <
0.01) correlations with other standardized measures of
mood, indicating convergent validity, and low nonsignificant correlations with age and years of schooling,
indicating divergent validity. Correlations with the SIP
Emotional Behavior Scale, which assesses a construct
conceptually related to mood, were also higher than
correlations for the SIP Mobility or Social Interaction
Scales. The magnitudes of the correlation coefficients
were similar for patient ranges with different levels of
education (<5 versus 2 5 years), indicating that the
responses of low literacy patients were not biased by
the question format. The convergent correlation coefficients may have been higher if the Face Scale had
used the same response time frames required by the
other scales. For example, directions for the Bradburn
Affect Balance Scale use the “past few weeks” as a
time frame, while directions for the SIP use “the past
month.
Experimental evidence. Another assessment of
scale construct validity is its ability to detect change
when change is expected. Treatment during hospitalization should produce significant improvements in
patient mood and health status. The mean scale scores
of the patients at admission, discharge, and at the
followup clinic visit for different instruments are presented in Table 2. Sample sizes differ because of study
dropouts or unusable responses. Paired-comparison
t-tests indicated significant (P < 0.05) changes in
patient mood from admission to discharge, but not
from discharge to followup clinic visit for each scale.
Thus, all scales detected a significant improvement in
”
Mean scale values for rheumatoid arthritis patients
Hospital
admission
Face Scale
Bradburn Positive Affect
Bradburn Negative Affect
Pain Scale
Beck Depression Scale
Sickness Impact Profile
Emotiional Behavior
Total
8.70
2.00
2.30
5.90
5.10
34.4
27.6
No.
Hospital
discharge
No.
t-statistic
Clinic
followup
I74
171
171
180
166
5.5*
3.1*
1.2*
3.2*
2.2*
176
176
176
177
156
6.9
4.2
5.6
12.8
8.7
6.20
2.40
I .60
4.10
NGt
135
135
135
133
183
181
NG
NG
NG
NG
5.1
5.0
24.5*
22.2*
131
129
* Significant change from previous measurement (P< 0.05, paired !-test).
t NG = not given.
No.
BRIEF REPORTS
patient status. The magnitude of the t-statistics presented in Table 2 provided an indication of each
scale’s sensitivity to change. The Face Scale was
second only to the Beck Scale in the size of the
t-statistics for all mood scales.
Since patients were given less help in completing the Face Scale than in completing the other scales,
less-educated patients might have difficulty completing
the Face Scale properly. To investigate this possibility, Mann-Whitney U tests were used to compare the
mean scores for each scale between patient groups of
different educational levels. No significant ( P < 0.05)
differences were found between more-educated groups
(>5th grade) and less-educated groups (55th grade) on
any scale.
Test-retest reliability. The Pearson correlation
coefficient between patient scores on the 2 administrations of the Face Scale was 0.81.
DISCUSSION
The results provide initial support for the use of
the Face Scale as a brief, valid method for assessing
the mood of arthritis patients. The Face Scale was
easily completed by patients with a minimal amount of
guidance, regardless of their literacy levels, required a
minute or less of their time, and showed a high
test-retest reliability over a 45-minute interval. The
Face Scale correlated well with other, more complex
measures of patient mood and was sensitive to
changes in patients’ status produced by treatment.
There are few reports of correlations between depression scores and other arthritis health status measures.
The Face Scale appears to be a quick, simple, and
useful research tool for assessing mood in groups of
arthritis patients. The scale is not intended to diagnose
clinical depression, but can be useful as a screening
tool prior to more extensive patient evaluation. The
scale might also be especially useful with patients who
do not speak English and in the comparison of the
effects of arthritis with other diseases. The study was
limited by the use of disparate time intervals in the
instruction sets among tests. Further research on the
use of the Face Scale with longer time intervals would
be useful.
Acknowledgments. We wish to thank the Treatment
Team and patients at Spain Rehabilitation Center for their
909
patience and cooperation during the conduct of this study.
Special thanks to Cindy Ellis and Ellen Johnson for their
helpful comments and suggestions.
REFERENCES
1. Robinson ET: Depression in rheumatoid arthritis. J R
Coll Gen Pract 27:423-427, 1977
2. Rirnon R: Depression in rheumatoid arthritis. Ann Clin
Res 6:171-174, 1979
3. Vignos PJ: Comprehensive care and psychosocial factors in rehabilitation in chronic rheumatoid arthritis: a
controlled study. J Chronic Dis 25:457467, 1972
4. Ehrlich GE: Social, economic, psychologic, and sexual
outcomes in rheumatoid arthritis. Am J Med 75:27-34,
I983
5 . Beck AT, Ward CH, Mendelsohn M, Mock J, Erbaugh
J: An inventory for measuring depression. Arch Gen
Psychiatry 4561-571, 1961
6. Zung W: A self-rating depression scale. Arch Gen Psychiatry 12:63-70, 1965
7. Liang MH, Larson MG, Cullen KE, Schwartz JA:
Comparative measurement efficiency and sensitivity of
five health status instruments for arthritis research.
Arthritis Rheum 28542-547, 1985
8. Kozol J: Illiterate America. New York, Doubleday, 1985
9. Lorish CL, Parker J, Brown S: Effective patient education: a quasi-experiment comparing an individualized
strategy with a routinized strategy. Arthritis Rheum
28:1289-1297, 1985
10. Aitken RCB, Zeally AR: Measurement of moods. Br J
Hosp Med 4:214224, 1970
11. Ekman P, Oster H: Facial expressions of emotions.
Annu Rev Psychol 30:527-554, 1979
12. Ekman P, Fiesen W, O’Sullivan M, Scherer K: Relative
importance of face, body, and speech in judgement of
personality or affect. J Pers SOC Psychol 38:27&277,
1980
13. Steinbrocker 0, Traeger CH, Batterman RC: Therapeutic criteria in rheumatoid arthritis. JAMA 140:659-662,
1949
14. Bergner M, Bobbitt RA, Pollard WE, Martin DP, Gilson
BS: The sickness impact profile: validation of a health
status measure. Med Care 1457-67, 1976
15. Bradburn NM: The Structure of Psychological WellBeing. Chicago, Aldine, 1969
16. Conn J , Bobbitt RA, Berger M: Administration procedures and interviewing training for the Sickness Impact
Profile. Seattle, Washington, University of Washington,
Department of Health Services, 1978
Документ
Категория
Без категории
Просмотров
2
Размер файла
373 Кб
Теги
patients, scales, facer, brief, method, nonverbal, assessing, moody
1/--страниц
Пожаловаться на содержимое документа