Table 4. Physician-Level and Group-Level Characteristics Associated With Generalized Frequent Use of 2 or More Screening Tests (ie, Cross-Test Score ≥2) Among 2340 Physiciansa.
| Variable | Odds Ratio (95% CI)b | 80% Interval Odds Ratiob,c |
|---|---|---|
| Physician-Level Fixed Effects | ||
| Male vs female | 1.29 (1.01-1.64)d | NA |
| IMG vs domestic | 0.64 (0.49-0.84)d | NA |
| Years since medical school graduation | 1.03 (1.02-1.04)d | NA |
| Group-Level Fixed Effects | ||
| Primary care modele | ||
| Family health organization vs family health group | 0.90 (0.66-1.23) | (0.44-1.85) |
| Family health team vs family health group | 0.49 (0.34-0.71)d | (0.23-1.07) |
| No. of patients per physician | ||
| <30-60 vs <0 | 0.67 (0.43-1.03) | (0.28-1.57) |
| >60 to 110 vs <30 | 0.91 (0.60-1.38) | (0.39-2.10) |
| >110 to 210 vs <30 | 0.69 (0.44-1.07) | (0.29-1.62) |
| >210 vs <30 | 0.81 (0.53-1.26) | (0.35-1.91) |
| Group-Level Random Effectsb | ||
| Variance (SE) | 0.26 (0.12) | NA |
| Median odds ratioc | 1.62 (1.39-1.86) | NA |
| Intracluster correlation coefficient, %c,f | 7.27 | NA |
Abbreviations: IMG, international medical graduate; NA, not applicable.
All reported values are based on SAS PROC GLIMMIX output; model estimation method is Laplace; denominator df estimation method is between and within; covariance structure is standard variance.
Adjusted for all other factors present in the model and table, as well as the quarter and season in which the visit occurred.
Estimated using group-level variance estimate.
Significant at P < .05.
Represents the primary care patient enrollment model that informs group organization and remuneration. Suppressed comparison of other vs family health group because of low observed frequency of the other category.
Calculated using the latent response formulation (ie, set residual variance equal to 3.29).