On the eight scales of the Medical Outcomes Study Short‐Form 36‐Item Health Survey (SF‐36), Version 2, we compared the clinically important difference (CID) thresholds for change over time developed by three separate expert panels of physicians with experience in quality of life assessment among patients with chronic obstructive pulmonary disease (COPD), asthma, and heart disease.
We used a modified Delphi technique combined with a face‐to‐face panel meeting within each disease to organize and conduct the consensus process among the expert panelists, who were familiar with the assessment and evaluations of health‐related quality of life (HRQL) measures among patients with the panel‐specific disease.
Each of the expert panels first determined the magnitude of the smallest numerically possible change on each SF‐36 scale, referred to as a state change, and then built their CIDs from this metric. All three panels attained consensus on the scale changes that constituted small, moderate, and large clinically important SF‐36 change scores. The CIDs established by the heart disease panel were generally greater than the CIDs agreed on by the asthma and COPD panels.
These panel‐derived thresholds reflect possible differences in disease management among the represented panel‐specific diseases, and are all greater than the minimal CID thresholds previously developed for the SF‐36 scales among patients with arthritis. If confirmed among patients with the relevant diseases and those patients' physicians, these disease‐specific CIDs could assist both researchers and practicing clinicians in the use and interpretation of HRQL changes over time.