Statistically significant relationships between the number of PV patients in respondents' practices and diagnostic testing
| Diagnostic test . | None, % . | 1-5 patients, % . | More than 5 patients, % . | χ2P . |
|---|---|---|---|---|
| EPO level | 60.5 | 73.1 | 79.1 | .008 |
| CT scan | 13.2 | 28 | 32.3 | .03 |
| Chest x-ray | 18.4 | 20.6 | 28.1 | .02 |
| Progenitor cell assays | 15.8 | 8.2 | 4.4 | .003 |
| Diagnostic test . | None, % . | 1-5 patients, % . | More than 5 patients, % . | χ2P . |
|---|---|---|---|---|
| EPO level | 60.5 | 73.1 | 79.1 | .008 |
| CT scan | 13.2 | 28 | 32.3 | .03 |
| Chest x-ray | 18.4 | 20.6 | 28.1 | .02 |
| Progenitor cell assays | 15.8 | 8.2 | 4.4 | .003 |
Each row represents the results of a Pearson χ2test computed on a contingency table for the number of PV patients in a respondent's practice and whether or not a respondent uses a diagnostic test. To conserve space, we present only the affirmative responses to our question and report results only for tests where we found statistically significant variation (P < .05).