Online Program

Return to main conference page
Tuesday, January 7
Tue, Jan 7, 7:45 AM - 8:45 AM
Pacific D
Continental Breakfast & Poster Session II

Minimally Important Difference in Cost savings: Is Effect Size a Good Benchmark (307898)

*Mary Dooley, Medical University of South Carolina 
Kit N Simpson, Medical University of South Carolina 

Keywords: minimally important difference, MID, costs, cost data

Healthcare costs in the United States are increasing at alarming rates, creating a heavy burden for an already pressured healthcare system. In response, studies to assess costs are abounding but their design and planning is hampered by the lack of indicators of magnitude of an important effect size for costs incurred in different settings. Other areas, such as quality of life measurement, use guiding rules for minimally important difference (MID) to judge effect magnitude. A MID is defined as the smallest difference of a score perceived as beneficial that would result in a change of the patient’s management. MIDs are identified by three methods: 1) anchor-based, 2) distribution-based, and 3) consensus-based. MIDs are useful for planning studies using quality of life and clinical assessment tools. However, we have found no publications establishing a MID for costs. We report the MIDs derived using the distribution-based method for cost data from hospital admissions studies and out-patient intervention studies. Our results are useful for informing cost study planning and power estimation.