Table 2.
Year | Ambulatory Visits, millions (SE) | Percentage of Ambulatory Visits Resulting in Referral, % (SE) | Ambulatory Visits Resulting in Referral, millions (SE) | p-value for time trend | p-value for difference in trends | |||
---|---|---|---|---|---|---|---|---|
1999 | 2009 | 1999 | 2009 | 1999 | 2009 | |||
Solo Practice | ||||||||
No | 494 (42.4) | 701 (66.2) | 4.87 (0.37) | 9.37 (0.59) | 24.1 (2.8) | 65.7 (7.4) | <0.001 | 0.392 |
Yes | 263 (25.7) | 336 (33.3) | 3.41 (0.44) | 7.03 (0.50) | 8.95 (1.6) | 23.7 (2.9) | <0.001 | |
MD Full or Part Owner? | ||||||||
No | 241 (24.5) | 315 (34.5) | 4.65 (0.49) | 11.1 (0.94) | 11.2 (1.7) | 34.9 (4.8) | <0.001 | 0.016 |
Yes | 516 (44.2) | 723 (61.0) | 4.23 (0.36) | 7.54 (0.55) | 21.8 (2.8) | 54.5 (5.9) | <0.001 | |
E-mail Patient Consults? | 2001* | 2001* | 2001* | |||||
No | 816 (66) | 962 (83.1) | 5.43 (0.34) | 8.55 (0.49) | 44.3 (4.2) | 82.2 (8.4) | <0.001 | 0.759 |
Yes | 64.7 (14) | 76.0 (13.7) | 8.22 (1.72) | 9.43 (1.64) | 5.32 (1.8) | 7.16 (1.7) | 0.427 | |
Telephone Patient Consults? | ||||||||
No | 307 (27.9) | 501 (48.1) | 5.18 (0.51) | 8.40 (0.73) | 15.9 (2.2) | 42.1 (5.3) | 0.008 | 0.915 |
Yes | 574 (53.1) | 537 (54.7) | 5.87 (0.47) | 8.81 (0.61) | 33.7 (3.8) | 47.3 (5.6) | <0.001 | |
Electronic Medical Records | 2003* | 2003* | 2003* | |||||
No | 758 (65.9) | 690 (58.8) | 5.97 (0.39) | 8.45 (0.57) | 45.3 (5.0) | 58.3 (6.5) | 0.001 | 0.697 |
Yes | 148 (22.2) | 348 (41.7) | 6.03 (0.78) | 8.94 (0.83) | 8.90 (1.8) | 31.1 (4.6) | 0.030 | |
>50% Income Managed Care? | ||||||||
No | 608 (56.9) | 650 (58.7) | 5.50 (0.42) | 8.70 (0.58) | 33.4 (4.0) | 56.6 (6.6) | <0.001 | 0.007 |
Yes | 298 (32.7) | 388 (41.1) | 6.94 (0.59) | 8.46 (0.73) | 20.7 (3.2) | 32.8 (4.4) | 0.540 | |
>50% Income Medicaid? | ||||||||
No | 872 (72.5) | 966 (82.1) | 5.97 (0.37) | 8.82 (0.50) | 52.1 (5.4) | 85.2 (8.5) | <0.001 | 0.365 |
Yes | 33.7 (7.6) | 72.3 (14.1) | 6.20 (2.30) | 5.77 (0.95) | 2.09 (1.0) | 4.17 (1.1) | 0.678 |
P-values for time trend calculated using logistic regression for trend over interval from the earliest year available to 2009 in each subgroup. P-values for difference in trends within each group calculated using analysis of covariance (ANCOVA).
For the e-mail and telephone consult characteristics, the earliest year available was 2001, for the electronic medical record, HMO, and Medicaid variables, 2003 was the earliest year.