Comparing results of concurrent and retrospective designs in a hospital utilization review

Hospital utilization reviews are performed on the basis of lists of explicit criteria, such as the Appropriateness Evaluation Protocol, both concurrently and retrospectively, in an increasing number of settings as part of efforts to improve the performance of hospitals and to reduce health care costs. Retrospective data collection has advantages in terms of expenses and ease of sampling, but relies on the quality of medical records. We report on a comparison between concurrent and retrospective data collection performed simultaneously and independently by two reviewers on the same hospital stays in the regional St-Loup Hospital. Results suggest that retrospective data collection produces higher rates of inappropriate hospital utilization, due to a limited number of criteria that are recorded concurrently, but are not found in the retrospective reading of medical records. These results should encourage a further investigation of the comparability between concurrent and retrospective designs in other settings.