Table 4
COSMIN adapted the GRADE approach for development/content validity
| QUALITY OF EVIDENCE: the starting point is always HIGH | ||
 | High | |
| Moderate | ||
| Low | ||
| Very Low | ||
| Risk of bias | 1: Serious | The content validity study is of doubtful quality. The content validity rating of the content validity study is insufficient (−) OR indeterminate (?) OR inconsistent (±) |
| 2: Very serious | No content validity study OR content validity study of insufficient quality (−) AND The development study is of doubtful quality. The content validity rating of the development study is indeterminate (?) OR inconsistent (±) | |
| 3: Very serious | No content validity study OR content validity study of insufficient quality (−) AND No development study or development study is of inadequate quality. The content validity rating of the development study is insufficient (−) | |
| Inconsistency | 1: Serious | The combination of the scores for development study, content validity study and reviewer’s rating is rated inconsistent (±) |
| Indirectness | 1: Serious | A content validity study was performed in a cancer population but not representative of the population of interest (eg, patients with head and neck cancer vs patients with cancer, palliative questionnaire assessed in patients with non-palliative cancer) |
| 2: Very serious | A content validity study was performed in a non-cancer population. | |
COSMIN, Consensus-based Standards for the selection of health Measurement Instruments; GRADE, Grading of Recommendations Assessment, Development and Evaluation.