Table 2 |
|||
| Baseline cohort characteristics for the electronic diabetes case-finding model* | |||
| Characteristics | Derivation | Validation | P-value** |
| n | 1417 | 341 | |
| Age, mean years (SD) | 49.3 (14.4) | 45.0 (14.6) | <.001 |
| Race/ethnicity,% | .83 | ||
| Hispanic | 44 | 40 | |
| White | 23 | 24 | |
| Black | 25 | 27 | |
| Other | 9 | 9 | |
| Male,% | 47 | 52 | .13 |
| Primary payer,% | .15 | ||
| Commercial | 18 | 16 | |
| Medicare | 9 | 6 | |
| Medicaid | 13 | 13 | |
| Self-pay | 30 | 29 | |
| Charity | 30 | 36 | |
| Encounters, mean no. (SD) | |||
| All | 7.86 (9.03) | 6.65 (9.06) | .01 |
| Primary care | 2.25 (3.90) | 1.99 (3.60) | .45 |
| Specialty care | 4.31 (6.86) | 3.62 (7.08) | .02 |
| Urgent care and ED | 0.81 (2.58) | 0.61 (0.89) | .48 |
| Inpatient | 0.49 (0.97) | 0.43 (0.88) | .25 |
*Derivation and validation cohorts defined as per Figure 1.
**Student t-test for age; χ2 tests for race/ethnicity, male sex, and primary payer; Wilcoxon rank-sum test for encounters.
Makam et al.
Makam et al. BMC Medical Informatics and Decision Making 2013 13:81 doi:10.1186/1472-6947-13-81
