|
FAQ's
|
|
Q |
|
A |
|
What guidelines should be
followed when including other measures with the SF-36® Health
Survey in the same questionnaire?
|
|
|
| |
When combining the SF-36®
with other measures, it is best to put the SF-36® first, to be
consistent with the standard followed when normative SF-36® data
were gathered. The ordering of questions may affect responses. A clear break in
the questionnaire between each measure or topic helps in standardizing the
context and informing the respondent. Use of a similar print font and style of
presentation for questions and response choices throughout the questionnaire,
across measures and topics, reduces errors.
|
|
|
Q |
|
A |
|
When should the physical
and mental summary scales from the SF-36® or SF-12®
be used?
|
|
|
| |
The summary scales for
the SF-36® and SF-12® Health Surveys are
recommended when: (a) there is a need to limit the number of outcomes being
analyzed; (b) a general effect (across sub-scales) in the physical or mental
health domain is expected. Because the physical component summary (PCS) and
mental component summary (MCS) each measure hundreds of levels of health and
extend the range of measurement to higher and/or lower levels than the eight
sub-scales, individuals rarely score at the very top or very bottom of the PCS
or MCS scales; this is a major advantage.
In cases where summary measures are appropriate, the choice between the 36-item
or 12-item versions is largely practical and depends on study objectives. The
SF-12® reproduces the SF-36® summary scales
(PCS and MCS) very well and it is much shorter. The SF-36® profile
has the advantage of providing more information about the nature of differences
in physical and mental health outcomes. Some medical conditions and most
treatments have specific health effects that tend to be concentrated in some
scales. For example, arthritis typically has its greatest impact on the Bodily
Pain Scale, whereas hepatitis impacts most on the Vitality Scale. When there is
a need to examine all eight scales, the SF-36® is recommended
over the SF-12®, which achieves less precision for all eight
scales.
|
|
|
Q |
|
A |
|
Distribution of copies of
the SF-36® Health Survey under different labels is confusing;
for example, what is the difference between the SF-36® and the
RAND-36 versions?
|
|
|
| |
The RAND-36 is an exact
replica of the content of the SF-36®. However, because RAND
uses different scoring algorithms for two of the 8 scales (Bodily Pain, General
Health), their results for those scales are not comparable with the standard
SF-36®. The differences are largest for the Bodily Pain scale;
RAND scores are five points or more higher for more than one-third of
respondents. Studies cited in the SF-36® bibliography have
shown that the RAND-36 scoring does not meet scaling and scoring assumptions as
well as the standard SF-36® scoring in the U.S. or in other
countries. For these reasons and because the SF-36® is readily
available, the RAND scoring is rarely used. Fewer than 10 of the first 300
publications about either version used the RAND-36 scoring.
|
|
|
Q |
|
A |
|
What is the relationship
between the SF-36® and the Functional Independence Measure
(FIM) in studies of outcomes of rehabilitation services?
|
|
|
| |
The 1988-1995 SF-36®
annotated bibliography includes three publications about the FIM and the SF-36®.
(They are identified under FIM in the keyword index.) A fourth paper entitled
"Extending the Range of the Functional Independence Measure (FIM) with SF-36®
Items," by Heinemann, Segal, Schall and Wright (from the Rehabilitation
Foundation in Chicago), presented at the First International Outcome
Measurement Conference, May 31, 1996, Chicago, IL, is also informative.
Heinemann showed that items from the SF-36® Physical
Functioning (PF) scale raised the ceiling (most favorable score) of
favorably-scored FIMs for two-thirds of stroke survivors. The FIM lowered the
floor (least-favorable score) on the SF-36® PF for one-sixth of
those patients. Thus, SF-36® and FIM can substantially improve
each other. They also found that: (a) not all FIM items define the same generic
PF scale (e.g., bowell/bladder control are disease specific), (b) the FIM
worked better with a 3-choice response scale than with the usual 7-choice
scale. Most interesting, with respect to the next generation of multi-purpose
PF outcome measures, six items from the SF-36® and four items
from the FIM defined a PF scale that was better than either the SF-36®
or the FIM among the 38 stroke survivors studied. The Heinemann et al results
agree with studies of the SF-36® and the FIM, SIP, ADL, IADL,
FSQ, NHP, DUKE, QWB, and other PF measures ongoing at the Health Institute in
Boston. We need more studies like the Heinemann et al study to guide the
development of the next generation of PF measures.
|
|
|
Q |
|
A |
|
What is the difference
between the acute and standard forms?
|
|
|
| |
There are two versions
each of the SF-36® and SF-12® instruments. The
Standard versions use the 4 week recall period and the Acute versions use the 1
week recall period. The acute version is particularly useful in instances when
the effects of treatment are quick in time. Both surveys are included in the
SF-36® and SF-12® manuals. Keller et.al., 1997
published an article that provides empirical results from a study comparing
both recall periods. Both versions conform to the assumptions used in scoring
the SF-36® and SF-12® as well as reliability
and validity standards.
|
|
|
Q |
|
A |
|
What is the Medical
Outcomes Study and how do I obtain more information on the instruments that
were developed and used?
|
|
|
| |
The Medical Outcomes
Study (MOS) was a quasi-experimental study of the variations in physician
practice styles and patient outcomes in different health care delivery systems.
It spanned the decade of the 1980's. Stewart and Ware, 1992 edited a text that
provides a comprehensive account of the measures developed for the Medical
Outcomes Study.
|
|
|
Q |
|
A |
|
How do I gain access to
the databases from the Medical Outcomes Study?
|
|
|
| |
Sociometrics has recently
made available for purchase data sets that are available to the public from the
Medical Outcomes Study. For more information go to:
Sociometrics
|
|
|
Q |
|
A |
|
How do I find information
on other health questionnaires?
|
|
|
| |
For more information on
other health questionnaires there is a text edited by McDowell and Newell, 1996
. This book is a great reference for learning more about health questionnaires
and choosing the appropriate instrument for your research purpose.
|
|
|
Q |
|
A |
|
Is it feasible to use the
instrument as a pre and post measure of change after an intervention?
|
|
|
| |
It is recommended that
you use the instrument to assess outcomes of interventions. If you look in the
published medical literature you will see that many researchers have used the
SF-36® and SF-12® to do just that. The
instrument is very sensitive to change.
|
|
|
Q |
|
A |
|
Has the SF-36®
been translated in languages other than English?
|
|
|
| |
The SF-36®
has already been translated in numerous languages. The translation work has
been under the auspices of the International Quality of Life Project Assessment
(IQOLA), housed at the Health Assessment Lab in Boston. If you are interested
in obtaining copies of translations or finding more information please contact
info@iqola.org
|
|
