Table 4.
Characteristics of the Group of Physicians that Met the Performance Standard Compared to the Outlier Group of Physicians that Did Not
| Met the Standard (N = 919) | Did Not Meet the Standard (N = 38) | ||||
|---|---|---|---|---|---|
| Characteristics * | Mean (SD) | Mean (SD) | Test statistic ‖ | P-value | Effect size ¶ |
| Overall clinical competence ratings † | 6.5 (1.2) | 5.9 (1.2) | 2.59 | 0.01 | 0.45 |
| Professional behavior/attitude ratings † | 7.0 (1.3) | 6.4 (1.3) | 2.75 | 0.006 | 0.40 |
| IM certification exam scores (1st attempt) ‡ | 459.6 (102.9) | 410.9 (91.2) | 2.79 | 0.005 | 0.49 |
| MOC exam scores (1st attempt) ‡,§ | 514.8 (91.8) | 443.2 (92.2) | 4.38 | <.001 | 0.78 |
| Solo practice | 29% | 47% | 5.49 | 0.02 | 0.08 |
IM, Internal Medicine; MOC, Maintenance of Certification.
* Physician subspecialty, age, gender, birth/training country, or other practice characteristics were not significantly associated with the group of physicians that did not meet the standard.
† Program director ratings were based on a nine-point rating scale and were done at the end of the third year of residency training.
‡ Scores were statistically equated to be comparable over time and scaled to have a mean = 500 and SD = 100.
§ MOC exam scores were available for 749 physicians in our sample.
‖t-tests were used to assess the significance of each difference, except for solo practice, in which case the chi-squared test was used.
¶ The Cohen’s d standardized mean difference was used to measure the magnitude of the difference, except for solo practice, in which case the phi coefficient was used.