80
A Quantitative RheuMetric Physician Checklist to Assess
Patient Status
The importance of expressing clinical phenomena as quantitative data beyond nar-
rative descriptions has been a basis for extensive attention to patient self-report
questionnaires, as noted in this chapter and most of the entire book. It is curious that
relatively little attention has been directed to advance more physician information
from narrative descriptions to quantitative scores. To be sure, 3 of the 7 RA Core
data set measures used in all clinical trials are swollen joint count, tender joint
count, and physician/assessor global estimate of patient status ( DOCGL) [ 68 ]. (The
abbreviation “DOCGL” rather than “PGA” is used to avoid confusion, as “PGA”
appears in the rheumatology literature to represent either patient and physician esti-
mates of disease activity in different reports.) At the same time, important limita-
tions are seen of both formal joint counts and DOCGL as quantitative measures.
Limitations of the joint count include: (1) poor reproducibility [ 82 – 88 ], with a
requirement to be performed by the same observer at each visit; (2) likelihood to
improve with placebo treatment more than the other 5 RA Core Data Set measures
[ 91 ]; (3) similar or lower relative effi ciencies than global or patient measures to
document differences between active and control treatments in clinical trials [ 117 ]
(Fig. 3.8 ); (4) improvement over 5 years while joint damage and functional disabil-
ity may progress [ 118 , 119 ]; (5) lower sensitivity to detect infl ammatory activity
than ultrasound and magnetic resonance imaging (MRI) [ 120 ]; (6) most visits to a
rheumatologist in routine care do not include a formal quantitative joint count [ 54 ,
121 ], unless required to provide a biological therapy.
DOCGL has the highest relative effi ciency of all RA Core Data Set measures to
distinguish active from control treatment in more clinical trials than joint counts,
laboratory tests, and patient self-report measures [ 117 ]. In a clinical trial, the physi-
cian global estimate often meets its initial purpose to assess disease activity.
However, a DOCGL may be sensitive not only to disease activity or reversible signs
and symptoms, but also organ damage (e.g., to joints in RA, kidneys in SLE, and
muscles in polymyositis) or irreversible symptoms, and/or distress (e.g., fi bromyal-
gia and depression) in which patients may have high levels of pain, fatigue, and
other symptoms, which remain unexplained by reversible or irreversible physical
fi ndings or laboratory values.
This matter appears more prominent when applied in routine clinical practice, in
which patients have not been selected for high levels of infl ammatory activity, as
seen in clinical trials [ 122 , 123 ].
These considerations have led to development of a RheuMetric (formerly called
RHEUMDOC, but name changed to avoid possible confusion with an electronic
medical record of the same name) checklist (Appendix C) to record quantitative
physician scores [ 76 , 124 ]. RheuMetric includes DOCGL score, supplemented by 3
separate (0–10) physician global estimate subscales for infl ammation or reversible
fi ndings (DOCINF), damage or irreversible fi ndings (DOCDAM), and distress
(DOCSTR) (previously termed DOCNON) [ 76 ]. A rationale for these subscales
includes:
T. Pincus et al.