close

Вход

Забыли?

вход по аккаунту

?

Evidence of the validity of the arthritis impact measurement scales.

код для вставкиСкачать
93
BRIEF REPORT
EVIDENCE OF THE VALIDITY OF THE ARTHRITIS IMPACT
MEASUREMENT SCALES
MARILYN K. POTTS and KENNETH D. BRANDT
The Arthritis Impact Measurement Scales
(AIMS) were administered to 120 patients with rheumatoid arthritis. These patients also rated the importance
of various aspects of their treatment. Our findings
support the validity of the pain, anxiety, depression,
dexterity, and physical activity subscales of the AIMS.
The household activity subscale, however, may not be
appropriate for use with men, and the activities of daily
living subscale may be insensitive to mild impairment.
In recent years, the number of instruments
available for measurement of health status in general
populations has grown markedly. Questionnaires designed to measure the clinical status of patients with
arthritis, e.g., the Arthritis Impact Measurement
Scales (AIMS) (I), the Stanford Health Assessment
Questionnaire (2), and the Functional Status Index (3)
now are widely used.
Convergent validity is the extent to which
scores on a health status indicator are correlated with
scores on another instrument that addresses a related
concept. This is one aspect of construct validity, i.e.,
the degree to which scores on the indicator are assoFrom the Department of Medicine, Indiana University
School of Medicine, Indianapolis.
Supported in part by Multipurpose Arthritis Center grant
AM-20582 from the National Institutes of Health.
Marilyn K. Potts, MS: Director, Community/Health Services Research Component, Indiana University Multipurpose Arthritis Center; Kenneth D. Brandt, MD: Professor of Medicine and
Head, Rheumatology Division, Indiana University School of Medicine and Director, Indiana University Multipurpose Arthritis
Center.
Address reprint requests to Marilyn K. Potts, MS, Indiana
University Arthritis Center, 541 Clinical Drive, Room 492, Indianapolis, IN 46223.
Submitted for publication January 21, 1986; accepted in
revised form July 9, 1986.
Arthritis and Rheumatism, Vol. 30, No. 1 (January 1987)
ciated with measures of other relevant variables, behaviors, or responses (4).
One way to assess the validity of a clinical
status indicator is to examine the associations between
patients’ ratings of the importance of various aspects
of their treatment and their scores on pertinent sections of the indicator. If the instrument is valid, those
patients who view pain relief as an important aspect of
their treatment should indicate on the instrument that
they are, indeed, experiencing pain. Similarly, patients
who consider it important to improve their ambulation
status should indicate that their mobility is impaired.
Thus, we hypothesized that, on a questionnaire
designed to measure the views of patients with rheumatoid arthritis about the importance of various aspects of their treatment, specific responses would be
associated with the patients’ scores on relevant subscales of the AIMS. A study was conducted to determine the extent to which patient-physician agreement
on the importance of specific facets of a treatment
program were associated with clinical status outcomes
(5). The data on which our analysis was based were
obtained in the course of that study.
METHODS
Subject selection. The study was conducted at
the Rheumatology Clinic of Indiana University Hospital. Each patient who attended this clinic received
care from 1 of 6 staff rheumatologists. Patients with a
diagnosis of definite or classic rheumatoid arthritis (6)
and who attended the clinic between August 1, 1984
and April 30, 1985 were asked, at a routine visit, to
participate in the study. Patients were enrolled
sequentially.
94
Table 1. Twelve aspects of an overall treatment program for
patients with rheumatoid arthritis*
1. Be told about tests and diagnostic procedures (why done,
results)
2. Learn how to protect joints from stressful motion
3. Obtain pain relief
4. Have billing procedures explained
5 . Learn how arthritis can cause joint pain, swelling, and
stiffness
6. Increase ability to get around
7. Learn how to use heat or cold on affected joints
8. Obtain devices to help perform daily activities (writing,
dressing, cooking, etc.)
9. Discuss emotional concerns
10. Learn how to do proper exercises for affected joints
1 1. Receive sexual counseling
12. Be informed of possible side-effects of medicines
* Items 3, 6, 8, and 9 were used in the present analysis.
Importance questionnaire. Immediately following a clinic visit, each patient completed a questionnaire
on his or her views of the importance of 12 aspects of the
overall treatment program (Table 1). Our previous research had shown each of these items to be important to
patients with rheumatoid arthritis (7). The present study
deals only with items 3, 6, 8, and 9. These were chosen
because they correspond to the content of 1 or more
subscales of the AIMS.
Patients were asked to rate the extent to which
they considered each item to be an important aspect Of
their current treatment, on a 5-point scale where 1 =
not at all important and 5 = very important.
status assessment. The
questionnaire was completed by participating patients and
returned to the investigators by mail 1 week after the
clinic visit. This self-administered questionnaire required 15-20 minutes to Complete. It consisted Of 45
items which created 9 subscales. The number Of items
in each subscale ranged from 4-7. Scores were standardized on a scale of 0-10: 0 represented good health
status and 10 represented poor health status.
Psychometric properties of the AIMS (e.g.,
internal consistency, 2-week test-retest reliability)
are documented elsewhere (8). AIMS scores are
known to be sensitive to changes in the patient’s
health status (9,lO).
Eight of the 9 AIMS subscales were used in the
current analysis: mobility, physical activity, dexterity,
household activity, activities of daily living (ADL),
pain, depression, and anxiety. The social activity
subscale was not used, because no item in our “importance” questionnaire corresponded to its content.
As stated above, this questionnaire was developed for
use in a larger study. In pilot tests for that study,
results showed that patients did not view increased
BRIEF REPORTS
Table 2. Rheumatoid arthritis patients’ assessment of the importance of various aspects of their treatment, and their responses to
selected subscales of the Arthritis Impact Measurement Scales
(AIMS)
Mean 2 SD
Aspect of treatment*
Obtain pain relief
Discuss emotional concerns
Obtain devices to help perform
daily activities
Increase ability to “get around”
AIMS subscalet
Mobility
Physical activity
Dexterity
Household activity
Activities of daily living
Pain
Anxiety
Depression
4.0 2 1.5
2.6 2 1.6
2.8 2 1.7
3.6
2 1.7
2.4 2 3.5
6.8 2 2.5
4.9 2 3.2
1.3 % 1.6
2.4 f 1.9
5.6 2 2.4
3.4 t 1.9
2.5 2 1.8
* Scored on a scale of 1-5, where 1 = not at all important and 5 =
very important.
t Scores standardized on a scale of 0-10, where 0 = good health
status and 10 = poor health status.
social activity as an important aspect of their medical
treatment.
RESULTS
Characteristics of study participants. One hundred twenty patients participated in the study. A]though no patient refused to participate, 10 (8%) failed
to return the AIMS. The enrollment period was extended to allow recruitment of 10 additional, substitute
participants.
The mean age of the 120 participating patients
was 53 years (range 2443). Most were female (8 I%),
white (95%), and married (74%). Only 11% had not
completed high
- school, while 31% had completed at
least 4 years of college or postgraduate training. Thirty-three percent were employed outside the home,
24% were homemakers, and 18% were retired or
unemployed. The remaining 25% were disabled. The
mean duration of rheumatoid arthritis at the time of
enrollment was 12 years (range 1-39). The patients had
visited the rheumatology clinic an average of 20 times
prior to enrollment (range 3-78 visits).
Scores on “importance” and clinical status questionnaires. Based on their responses to the “importance” questionnaire, our patients appeared to view
items 3, 6, 8, and 9 (Table 1) as moderately important
aspects of their care. Thus, mean ratings for these
items ranged from 2.6 for “discuss emotional concerns” to 4.0 for “obtain pain relief” (Table 2).
Mean responses to the AIMS subscales indicated that the average disease severity of our patients
9:,
BRIEF REPORTS
was mild to moderate (range 1.3 for household activity-6.8 for physical activity) (Table 2).
Correlations between importance ratings and
AIMS scores. Pain, anxiety, depression, dexterity, and
physical activity subscales. Patients who viewed pain
relief as an important aspect of their treatment program tended to exhibit high levels of pain on the pain
subscale of the AIMS (P < 0.01) (Table 3). Those who
considered it important to discuss their emotional
concerns were more likely than other patients to
exhibit high levels of psychological distress on both
the anxiety and the depression subscales (P < 0.001
and P < 0.01, respectively). Patients who expressed a
desire to obtain ADL devices tended to indicate on the
dexterity subscale that they were experiencing problems with this aspect of their clinical status (P < 0.01).
Finally, patients who considered it important to increase their ability to “get around” were more likely
than other patients to indicate on the physical activity
subscale that their ambulation status was impaired
(P < 0.01).
Our data thus provide evidence in support of
the validity of the pain, anxiety, depression, dexterity,
and physical activity subscales of the AIMS. Furthermore, the significance of each correlation with these
subscales was not altered when age, sex, duration of
arthritis, or the number of visits to the rheumatology
clinic prior to enrollment in the study was controlled.
The significant correlations described above
were of modest magnitude. This may be explained by
the possibility that patients’ assessment of the importance of various aspects of their treatment, and their
scores on the relevant AIMS subscales, were not
related as straightforwardly as hypothesized originally. For example, some patients with low pain
scores may have valued pain relief highly because they
anticipated the return of their pain. Others with high
pain scores may have discounted the value of treatment for pain because they had developed an attitude
of defeatism regarding the physician’s ability to relieve
their discomfort.
The fact that several additional correlations
noted in Table 3 neared a significance level of 0.05
(i.e., pain subscale with “discuss emotional concerns”
and pain and anxiety subscales with “obtain assistive
devices”) does not detract from the validity of the
AIMS. Thus, it is reasonable to expect that patients
experiencing pain might wish to discuss their emotional concerns, and that pain and anxiety might have
reciprocal effects on patients’ views regarding the
importance of assistive devices.
Household activity subscafe. In contrast to the
above results, our patients’ views regarding the importance of obtaining ADL devices were not associated
with their scores on the household activity subscale
(Table 3). In an attempt to account for this lack of
association, we considered the possibility that items
within the household activity subscale might be more
relevant to women than to men, insofar as this
subscale deals primarily with activities traditionally
perceived as “feminine” (e.g., shopping, cooking,
doing housework and laundry). We therefore analyzed
male and female subgroups separately. The results
indicated that, in women, but not in men, responses on
the importance of obtaining ADL devices were indeed
associated with scores on the household activity
subscale (r = 0.22, P < 0.05 for women; r = -0.13,
P not significant for men). This suggests that the
household activity subscale may not be appropriate for
use with most men, and that investigators should use
Table 3. Correlation matrix between rheumatoid arthritis patients’ ratings of the importance of
various aspects of their treatment, and their scores on selected subscales of the Arthritis Impact
Measurement Scales (AIMS)*
Components of treatment
AIMS
subscale
Pain
Anxiety
Depression
Dexterity
Household activity
Activities of daily living
Physical activity
Mobility
~~
~~
~
Obtain
pain
relief
Discuss
emotional
concem s
Obtain
assistive
devices
Increase
ability to
“get around”
0.32t0
0.10
0.08
-0.03
0.13
0.10
0.08
0.15
0.45$0
0.29t0
-0.08
0.10
0.12
0.18
0.20
-0.03
0.26t0
0.088
0.15
0.01
0.14
-0.03
-0.03
0.06
0.20
-0.08
0.25t0
0.130
0.06
-0.041
0.10
-0.08
~~
* Correlation matrix determined by Pearson’s r.
t P < 0.01.
f P < 0.001.
0 Correlation coefficients that were predicted to be positive and significant.
96
this section of the AIMS with caution when studying
samples in which most subjects are male.
ADL subscale. No association was noted between our patients’ assessment of the importance of
obtaining assistive devices and their scores on the
ADL subscale (Table 3). These unexpected results
may have occurred because our patients, as a group,
were only minimally impaired with respect to their
ADL status. Thus, even though they indicated a desire
to obtain assistive devices, this subscale may not have
been sensitive enough to identify mild impairment.
To test this hypothesis, we divided the patients
into 2 groups, based on scores above or below the
median on the ADL subscale. We then analyzed the
resulting subgroups separately. The subgroup that had
greater impairment and scored above the median on
the ADL subscale tended to indicate that obtaining
assistive devices was an important aspect of their care
(r = 0.24, P < 0.05), while the remaining patients did
not. These findings support our hypothesis that the
ADL subscale is insensitive to mild impairment, and
suggest that this subscale may not be valid when used
with patients whose ADL status is not significantly
affected by their disease.
Mobility subscafe. Patients’ ratings of the importance of increasing their ability to “get around”
were not associated with their scores on the mobility
subscale (Table 3). To determine whether the strength
of this association might vary according to the degree
of mobility impairment, we again divided our patients
into above-median and below-median subgroups.
When these subgroups were analyzed separately, no
association was apparent between patients’ views, in
either subgroup, of the importance of increasing their
mobility and their scores on the mobility subscale.
Thus, our data do not provide evidence in
support of the validity of the mobility subscale. Two
explanations may account for this finding. First, our
patients had attended the rheumatology clinic for a
relatively long time (mean 20 visits), and each patient
had visited the clinic at least 3 times prior to enrollment. Although the mobility of some patients may
have been impaired, it is likely that they had already
received ambulation aids, and had been instructed in
appropriate exercises for their lower extremities. They
may have assumed that further attention to their
mobility-related problems was not an important aspect
of their current treatment program. Second, 2 of the 4
items included in the mobility subscale of the AIMS
refer to the patient’s ability to travel within the community (i.e., to use public transportation, to travel
without the assistance of another person). Since social
BRIEF REPORTS
and environmental factors, in addition to physical
factors, may affect the response to these items, the
mobility subscale may not specifically reflect the patient’s ambulation status.
DISCUSSION
Our findings support the validity of several
AIMS subscales. However, the results suggest that the
household activity subscale is more appropriate for
use with women than with men, and that the ADL
subscale is more valid when used with patients whose
ADL status is severely impaired.
Acknowledgments. We thank Drs. Robert Meenan
and Barry Katz for their helpful comments. Roberta Fehrman provided excellent secretarial support.
REFERENCES
1. Meenan RF, Gertman PM, Mason JH: Measuring health
status in arthritis: the Arthritis Impact Measurement
Scales. Arthritis Rheum 23:14&152, 1980
2. Fries JF, Spitz P, Kraines RG, Holman HR: Measurement of patient outcome in arthritis. Arthritis Rheum
23~137-145, 1980
3. Jette AM: Functional status instrument: reliability of a
chronic disease evaluation instntment. Arch Phys Med
Rehabil61:395-401, 1980
4. Aiken LR: Standardization, reliability, and validity,
Psychological Testing and Assessment. Third edition.
Boston, Allyn & Bacon, 1979, pp 51-74
5. Potts MK, Mazzuca SA, Brandt KD: Views of patients
and physicians regarding the importance of various
aspects of arthritis treatment: correlations with health
status and patient satisfaction. Patient Educ Couns (in
press)
6. Criteria for the classification of rheumatoid arthritis,
Primer on the Rheumatic Diseases. Eighth edition. Edited by GP Rodnan, HR Schumacher, NJ Zvaifler.
Atlanta, Arthritis Foundation, 1983, pp 207-208
7. Potts M, Weinberger M, Brandt KD: Views of patients
and providers regarding the importance of various aspects of an arthritis treatment program. J Rheumatol
11:71-75, 1984
8. Meenan RF, Gertman PM, Mason JH, Dunaif R: The
Arthritis Impact Measurement Scales: further investigations of a health status measure. Arthritis Rheum
25:1048-1053, 1982
9. Meenan RF, Anderson JJ, Kazis LE, Egger MJ, AltzSmith M, Samuelson CO Jr, Willkens RF, Solsky MA,
Hayes SP, Blocka KL, Weinstein A, Guttadauria M,
Kaplan SB, Klippel J: Outcome assessment in clinical
trials: evidence for the sensitivity of a health status
measure. Arthritis Rheum 27: 1344-1352, 1984
10. 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
Документ
Категория
Без категории
Просмотров
11
Размер файла
408 Кб
Теги
measurements, scales, evidence, validity, arthritis, impact
1/--страниц
Пожаловаться на содержимое документа