Comparison of various quality measures between economic evaluations of complementary and integrative medicine (CIM) and conventional medicine
| Economic evaluations of CIM | Cost-utility analyses (CUAs) across all medicine† | ||||||
|---|---|---|---|---|---|---|---|
| All full | 2001–2005 | 2006–2010 | Higher quality | CUAs | 1998–2001 | 2002–2005 | |
| n=114 | n=59 | n=55 | n=31 | n=27 | n=300 | n=637 | |
| Average percentage met of applicable items on BMJ checklist | 72 | 71 | 73 | 87 | 89 | ||
| Presented the study perspective clearly (%) | 61 | 58 | 64 | 87 | 93** | 74 | 83** |
| Presented the study time horizon (%) | 96 | 98* | 93* | 100 | 100* | 75 | 87* |
| Conducted and reported sensitivity analysis (%) | 32 | 22** | 44** | 100 | 93** | 93 | 84** |
| Discounted costs and health effects, where appropriate (%)§ | 60 | 25* | 76* | 94 | 100* | 85 | 84* |
| Stated year of currency for resource costs (%) | 59 | 54 | 60 | 77 | 78** | 83 | 85** |
| Separate reporting of resource use (trials), parameters (models) and unit costs (for transferability) | 52 | 51 | 53 | 71 | 70 | ||
| Disclosed funding sources (%) | 72 | 58* | 76* | 84 | 93* | 65‡* | |
| Industry sponsored (%) | 10 | 12 | 11 | 10 | 7 | 18‡ | |
| Average Tufts quality score (CUAs only) | 4.75*** | 4.25‡*** | |||||
*χ2 Test p value<0.001.
**χ2 Test p value<0.01.
***t Test p value=0.002; comparisons were made between CIM economic evaluations published 2001–2005 and those published 2006–2010, and between CUAs of CIM 2001–2010 and CUAs of all medicine 2002–2005.
†Data from table 3 in Neumann.77
‡Data from table 3 in Neumann et al.76 Industry sponsored was calculated as a percent all studies 1976–2001.
§Denominators for the percentages reported in this row are the number of studies which evaluated impacts past 1 year in either the base case or in sensitivity analyses. For the first five columns the denominators are 25, 8, 17, 16 and 11, respectively. This information was not available for the last two columns.