Table 4.
Weighted meta-regression of the effects of study design, methodological quality and assay parameters on diagnostic accuracy of carbohydrate antigen 19-9
| Covariate | Number of studies | Coefficient | RDOR (95%CI) | P value |
| Study design and quality | ||||
| STARD ≥ 13 | 30 | 0.564 | 1.76 (0.14-22.68) | 0.652 |
| QUADAS ≥ 10 | 29 | -0.666 | 0.51 (0.06-4.11) | 0.514 |
| Consecutive or random design | 32 | 0.924 | 2.52 (0.26-24.68) | 0.411 |
| Cross-sectional design | 18 | -0.512 | 0.60 (0.28-1.28) | 0.178 |
| Blinded design | 0 | ND | ND | ND |
| Prospective design | 34 | ND | ND | ND |
| Assay method or cut-off value | ||||
| RIA | 17 | -0.619 | 0.54 (0.12-2.51) | 0.413 |
| ELISA | 11 | -0.737 | 0.48 (0.10-2.26) | 0.336 |
| EIA | 3 | 0.425 | 1.53 (0.29-8.14) | 0.604 |
| Cut-off of 37 U/mL | 30 | 0.553 | 1.74 (0.36-8.36) | 0.474 |
| Cut-off of 100 U/mL | 3 | 0.890 | 2.43 (0.72-8.26) | 0.146 |
EIA: Enzyme immunoassay; ELISA: Enzyme-linked immunosorbent assay; ND: Not done; RIA: Radioimmunoassay; STARD: Standards for Reporting Diagnostic Accuracy; QUADAS: Quality Assessment for Studies of Diagnostic Accuracy.