These definitions were primarily designed for clinical trials, but their use in clinical practice was predicted and rapidly became very popular, especially in Europe. All of these definitions include the PGA, and this is the only patient-reported outcome measure included in the definition of target.
The PGA is a practical, easy-to-administer measure, with good validity on its face and test-retest reliability. Its incorporation in these criteria was justified mainly by its relationship with disease activity and ability to distinguish active treatments from placebo.10 Its inclusion was also informed by the need to have a representation of the patient’s view in an era of patient-centered care philosophy.11
The use of the PGA in the definition of target has raised numerous concerns, which can be coalesced in three main domains: 1) The PGA has a poor relationship with disease activity, blurring the target for immunosuppressive therapy; 2) including the PGA increases the risk of overtreatment; and 3) the PGA provides a poor representation of the patient’s needs, but creates the illusion the patient is properly cared for.
PGA & Disease Activity
We, and others, have robustly demonstrated the PGA is more strongly associated with pain, function, fatigue, physical well-being and psychological domains, such as depression, anxiety and personality traits, as well as comorbidities (e.g., fibromyalgia, osteoarthritis) or years of formal education, than with markers of inflammation. This is especially true in the lower levels of disease activity, in which the definition of the target is more decisive.12-14
The discrepancy between the PGA and disease activity tends to increase with disease duration and age, when cumulative damage and comorbid causes of pain and function loss become more prominent.
We have also demonstrated that there is no difference between subclinical inflammation, as evaluated by extensive ultrasound examination and the ACR/EULAR, Boolean-based remission (i.e., 4V remission) and PGA near-remission (i.e., the PGA is the only Boolean criteria >1) descriptions.15
The PGA is essentially a measure of disease impact and not of disease activity.
That is one of the bases for our position that the PGA blurs the target and our proposal to remove it from the current definition of remission.12
In addition, a three-variable definition of remission (without the PGA) better predicts radiographic damage over two years than the full 4V-remission definition. This was demonstrated through individual patient data meta-analysis from 11 randomized controlled trials that included 5,792 patients.16
Risk of Overtreatment
Our group, among many others, has provided evidence that of the four variables included in the Boolean definition of remission (i.e., swollen and tender joints, CRP and PGA) the PGA is the most important single reason patients don’t reach remission. It is responsible for 60–80% of patients who don’t reach remission, depending on the studies considered.