| Advantages | |
| Easy to develop and administer | | Explicit (evidence based) criteria |
| High face validity, since experts define ‘good’ and ‘bad’ care | | Reproducible |
| Self-updating through use of experts | | Easy to explain low score in terms of criteria—which may narrow score of improvement efforts |
| Reflects the full scope of clinical decisions that apply to a particular patient | Can be conducted by researchers rather than expert clinicians, once the criteria have been agreed, reducing costs |
| Involves physicians and other expert clinicians in the quality of care process | | |
| |
| Disadvantages | |
| Requires (expensive) clinical experts | | Require training of reviewers |
| More arbitrary than evidence based | | Need to be updated constantly |
| Developed principally for inpatient care | | Limited scope in terms of content and context (relevant populations) |
| Poor reproducibility of judgements | | Does not capture the subtleties of health care (eg, contraindications) |
| | Expensive |
| | Potential for gaming |
| | Need to decide how to analyse multiple criteria |
| | Possible bias if different numbers of criteria apply to patients between comparative sites, particularly if some criteria are harder to meet than others |