Associations between characteristics and death among 2950 males with hemophilia in six US states, 1993-1995
| Characteristic . | Relative risk . | 95% CI . | P . |
|---|---|---|---|
| Age (10-year increase) | 1.6 | 1.4-1.7 | <.001 |
| Race (white vs all other) | 1.1 | 0.8-1.6 | .56 |
| State of residence (vs New York) | |||
| Colorado | 0.4 | 0.2-0.7 | <.001 |
| Georgia | 0.8 | 0.6-1.2 | .33 |
| Louisiana | 1.4 | 0.9-2.3 | .13 |
| Massachusetts | 0.7 | 0.5-1.1 | .11 |
| Oklahoma | 0.8 | 0.5-1.4 | .52 |
| Hemophilia type (B vs A) | 0.8 | 0.5-1.2 | .35 |
| Disease severity (vs mild) | |||
| Moderate | 0.7 | 0.5-1.1 | .17 |
| Severe | 1.0 | 0.7-1.5 | .89 |
| Insurance type (vs private/HMO) | |||
| Medicare/Medicaid | 1.4 | 1.1-1.9 | .01 |
| Other/None | 0.9 | 0.3-2.1 | .75 |
| Inhibitor present | 1.6 | 0.8-3.0 | .18 |
| Liver disease | 2.4 | 1.5-3.9 | <.001 |
| HIV-infected without AIDS | 4.7 | 3.0-7.2 | <.001 |
| AIDS | 33.5 | 22.7-49.5 | <.001 |
| Hemophilia care source (HTC vs non-HTC) | 0.6 | 0.5-0.8 | .002 |
| Characteristic . | Relative risk . | 95% CI . | P . |
|---|---|---|---|
| Age (10-year increase) | 1.6 | 1.4-1.7 | <.001 |
| Race (white vs all other) | 1.1 | 0.8-1.6 | .56 |
| State of residence (vs New York) | |||
| Colorado | 0.4 | 0.2-0.7 | <.001 |
| Georgia | 0.8 | 0.6-1.2 | .33 |
| Louisiana | 1.4 | 0.9-2.3 | .13 |
| Massachusetts | 0.7 | 0.5-1.1 | .11 |
| Oklahoma | 0.8 | 0.5-1.4 | .52 |
| Hemophilia type (B vs A) | 0.8 | 0.5-1.2 | .35 |
| Disease severity (vs mild) | |||
| Moderate | 0.7 | 0.5-1.1 | .17 |
| Severe | 1.0 | 0.7-1.5 | .89 |
| Insurance type (vs private/HMO) | |||
| Medicare/Medicaid | 1.4 | 1.1-1.9 | .01 |
| Other/None | 0.9 | 0.3-2.1 | .75 |
| Inhibitor present | 1.6 | 0.8-3.0 | .18 |
| Liver disease | 2.4 | 1.5-3.9 | <.001 |
| HIV-infected without AIDS | 4.7 | 3.0-7.2 | <.001 |
| AIDS | 33.5 | 22.7-49.5 | <.001 |
| Hemophilia care source (HTC vs non-HTC) | 0.6 | 0.5-0.8 | .002 |
All characteristics were included in a multivariate analysis, using proportional hazards regression. Each relative risk has been adjusted for the effects of all other characteristics in the model. CI indicates confidence interval; HMO, health maintenance organization; HTC, hemophilia treatment center.