294
differences between physician and patient global assessment [ 41 ]. Although the
actual questions asked for patient-rated global assessments are not standardized,
they are valid and sensitive to change [ 33 , 42 , 43 ].
Short-Form Health Survey
Developed in 1992, the Short-Form Health Survey , or SF-36, examines the impact
of disease on a patient’s mental and physical well-being. It assesses eight domains
that may be affected by disease: physical and social functioning, role limitations
due to both physical health and emotional problems, physical pain, general mental
health, energy/fatigue, and general health perceptions [ 44 ]. The domains are each
scored on a multi-item scale, using the Likert method of summated ratings, and are
further subcategorized into the Physical Component Summary (PCS) (0–100) and
the Mental Component Summary (MCS) (0–100) [ 44 , 45 ]. This tool was developed
to be comprehensive, but shorter than previous questionnaires used to investigate
the quality of life impact and burden of disease. The SF-36 assesses the infl uence of
any disease on overall well-being, rather than specifi c manifestations of a rheumatic
disease. One benefi t of using such a general score is that it may detect an unex-
pected clinical event during a trial, which would otherwise be missed by measures
that are more specifi c [ 43 ].
The SF-36 has been shown to be a reliable assessment of health-related quality of
life in several rheumatic diseases [ 46 ]. Danieli et al. examined the use of the SF-36
for determining HRQoL in SSc and found that, when compared to fi ndings in RA,
the SF-36 appears to correlate well to HRQoL in SSc [ 46 ]. The SF-36 is strongly
correlated with HAQ-DI scores in SSc [ 43 , 47 ] Adequate validity of the SF-36,
including satisfactory internal consistency, has been determined [ 43 , 48 ]. Reliability
of the SF-36 in SSc, however, has not been fully investigated [ 49 ]. When compared
to the HAQ-DI for diffuse cutaneous SSc (dcSSc), the SF-36 appears to be more
responsive to patient and physician global assessment, but less responsive to clinical
measures, such as variation in skin score and percent forced vital capacity (FVC)
predicted. The increased clinical responsiveness of the HAQ-DI may be because it
is geared more toward musculoskeletal disease than the generic SF-36 [ 43 ].
Data from the SF-36 can been used to generate an indirect preference-based
measure: the SF-6D [ 50 ]. Preference-based measures are often used in determining
HRQoL for economic analysis. In SSc, the SF-6D has excellent test–retest reliabil-
ity, and low fl oor and ceiling effects [ 51 ], and it is associated with the HAQ-DI and
pain scores. As with the SF-36, the SF-6D lacks strong correlations with some SSc
clinical measures such as change in skin and lung involvement [ 51 ].
Symptom Burden Index
Though the HAQ and the SF-36 have their merits, they are still relatively general. In
an attempt to develop a scale more specifi c to scleroderma, Kallen et al. created the
Symptom Burden Index (SBI), published in 2010 [ 52 ]. The SBI examines eight
R.E. Pellar et al.