Quality of care of many diseases, such as diabetes, hypertension, hyperlipidemia, and osteoporosis, can be assessed effectively from information in usual medical records concerning blood tests, blood pressure, bone density, etc. However, quality of care of rheumatoid arthritis (RA), as well as most rheumatic diseases, cannot be assessed from usual medical records. The primary basis for this problem involves limitations of laboratory tests and the absence of a single "gold standard" measure in RA Therefore, indices which include laboratory tests, joint counts, and patient questionnaires have been developed. These indices are collected in all RA clinical trials and other clinical research, but not in usual clinical care, a phenomenon which may limit severely possible assessment and improvement of quality and patient outcomes. Patient questionnaires and joint counts, rather than laboratory tests or radiographs in a medical record, are the best measures to assess and monitor RA patient status. Patient questionnaires are the most significant clinical prognostic markers for severe long-term RA outcomes, such as work disability, costs and premature mortality, and are more cost-effective and easily-collected than formal quantitative joint counts in busy clinical settings. The value of patient questionnaires and joint counts in RA is reviewed in three examples from the authors' research concerning premature mortality in RA, changes in patient clinical status between 1985 and 2000, and a QUEST-RA global perspective, to better evaluate the structure, processes, and outcomes of RA care.