Supplementary Online Content

Transcription

1 Supplementary Online Content Tilburt JC, Wynia MK, Sheeler RD, et al. Views of US physicians about controlling health care costs. JAMA. doi: /jama Appendix A. Survey Items from Physicians, Health Care Costs, and Society Appendix B. Development of cost-consciousness scale Appendix C. Table e4. Associations between key physician characteristics and enthusiasm ratings ( very enthusiastic versus somewhat or no enthusiasm ) for key ACA cost-containment strategies; odds ratios shown are from a single multivariate logistic regression model including all listed variables. eerences This supplementary material has been provided by the authors to give readers additional information about their work.

2 Appendix A 2

3 3

4 4

5 5

6 6

7 Appendix B The distributions of physician responses to survey items of interest for scale development were obtained using basic descriptive statistics. Before using items or scales measuring cost-consciousness and professional role in cost containment, we reviewed histograms, calculated skew and kurtosis, and then proceeded with principle components factor analysis with varimax orthogonal rotation. 1 We retained factors based on criteria λ>1(the Eigenvalue rule) 2 or using the scree test, 3 reviewing 1, 2, 3 and 4-factor solutions. Items that failed to load >0.3 on any factor were dropped from subsequent solutions; items loading >0.4 were considered in the analysis of what domains were represented by factor(s), and analyses were repeated in an iterative fashion in consultation and with feedback from scale development experts. In all analyses, Factor 1 had λ>2, often λ >2.5, with other factors λ <1. We also examined Cronbach alpha scores for items comprising each factor as each 1, 2, 3, and 4-factor solution was assessed. We settled on a 1-factor solution with 11 items that included items conceptually coherent with our a priori objectives (to what extent physicians pay attention to and believe it is their job to address health care costs including cost to society in their work). Internal consistency of the resulting scale was moderate (Cronbach s raw alpha=0.77). The Eigenvalues of each item in our final scale are shown in the table below: Eigenvalues of the Reduced Correlation Matrix: Total = Average = Eigenvalue Difference Proportion Cumulative

8 Appendix C etable 4. Associations between key physician characteristics and enthusiasm ratings ( very enthusiastic versus somewhat or no enthusiasm ) for key ACA cost-containment strategies; odds ratios shown are from a single multivariate logistic regression model including all listed variables. Improving Quality & Efficiency of Care Improving Conditions for Evidence-Based Decisions Promoting continuity of care Promoting chronic disease care coordination Expanding access to free preventive Expanding electronic health records Rooting out fraud and abuse Expanding access to quality and safety data Promoting headto-head trials of competing treatments Limiting corporate influence on physician behavior CHARACTERISTIC Age (years) Less than (76) 50 or greater 1084 (75) Sex Female 635 (85) Male 1237 (71) Region South 597 (74) Midwest 419 (73) ortheast 420 (79) West 422 (76) Practice Setting Type Small/solo 360 (74) Group/HMO 1200 (76) 738 (71) 985 ( ) (68) 587 (78) 0.4* 1136 ( ) (65) 551 (68) 396 ( ) (72) ( ) (70) 387 ( ) (68) 306 (63) ( ) (70) ( ) 514 (49) 660 (46) 413 (56) 761 (44) 0.5* ( ) 358 (44) 294 ( ) (54) 257 ( ) (48) ( ) (45) 198 (41) 752 (48) 1.4 (-1.9) ( ) 401 (39) 456 (32) 269 (36) 588 (34) 0.7* ( ) 261 (32) 235 ( ) (42) ( ) (33) 1.5* 178 (-2.1) (31) 106 (22) 574 (36) 1.2 ( ) 0.8 (0.6-) 748 (72) 988 (68) 546 (73) 1190 (69) ( ) 0.9 ( ) 511 (49) 747 (52) 408 (55) 850 (49) 0.8 (0.6-) 564 (70) 410 (51) ( ) (68) ( ) (48) ( ) (73) ( ) (51) 1.6* ( ) (68) ( ) (51) 326 (66) 1108 (70) 1.7* ( ) 221 (46) 810 (51) ( ) 510 (49) (-1.6) (51) 350 (47) (0.6-) (52) 412 (51) ( ) (51) 264 ( ) (50) 280 ( ) (49) 205 (42) (-1.7) (52) 656 (64) 897 ( ) (63) 490 (67) (-1.6) (62) 508 (63) 350 ( ) (64) 347 ( ) (65) 339 ( ) (60) 315 (65) 1.6* 985 ( ) (63) ( ) 0.7 (0.6-) 0.9 ( ) ( ) ( ) ( ) 8

9 Improving Quality & Efficiency of Care Improving Conditions for Evidence-Based Decisions Promoting continuity of care Promoting chronic disease care coordination Expanding access to free preventive Expanding electronic health records Rooting out fraud and abuse Expanding access to quality and safety data Promoting headto-head trials of competing treatments Limiting corporate influence on physician behavior CHARACTERISTIC City/state/fed government 246 (75) Medical school 48 (83) Compensation Type Billing only 743 (74) Salary plus bonus 639 (75) Salary only 345 (78) *p-value < ( ) 1.9 ( ) 241 (73) 46 (78) 656 (65) 609 ( ) (72) ( ) (72) 1.3 ( ) 1.8 ( ) 174 (53) 38 (64) 410 (41) ( ) (52) ( ) (51) 1.3 ( ) 2.1 (-4.6) 147 (45) 22 (37) 264 (26) 1.4* 339 (-1.9) (40) (-1.9) (42) 2.3* ( ) 1.5 ( ) 258 (79) 35 (59) 1.6 (-2.6) 0.7 ( ) 178 (54) 33 (57) 678 (67) 475 (47) 1.6* ( ) (72) ( ) (54) 1.8* ( ) (72) ( ) (52) 1.3 ( ) 1.6 ( ) 165 (50) 36 (61) 515 (51) (-1.7) (50) 206 ( ) (46) 1.6 (-2.4) 2.3 (-4.9) 202 (62) 39 (67) 645 (65) (0.7-) (62) (0.5-) (64) 0.9 ( ) ( ) 0.9 ( ) ( ) 9

10 etable 4 (continued) Changing How Care Gets Paid For Cutting Payment to Physicians Directly Limiting access to expensive treatments with little net benefit Using costeffectiveness data to determine available treatments Higher patient copays High deductible health plans Penalizing providers for avoidable readmissions Paying a network of practices a fixed, bundled price Eliminating fee-forservice payment models Allowing Medicare payment cuts to doctors to take effect Reducing compensation for the highest-paid specialties Age (years) Less than (47) 50 or greater 781 (54) Sex Female 367 (49) Male 898 (52) Region South 381 (47) 1.4* (-1.8) 483 (46) 687 (48) 356 (48) 814 (47) ( ) 357 (44) Midwest (55) ( ) (51) ortheast (48) ( ) (43) West (57) (-1.9) (51) Practice Setting Type ( ) 168 (16) (13) 320 (18) ( ) ( ) ( ) (13) (-1.8) (20) ( ) 160 (15) (11) 326 (19) 1.4 (-2.0) ( ) (16) ( ) (16) 96 ( ) ( ) 53 (5) (5) * ( ) ( ) (5) ( ) (4) ( ) ( ) (7) 49 (7) ( ) ( ) (9) ( ) ( ) 70 (7) ( ) (7) 55 (8) 120 ( ) (7) 36 (5) ( ) (11) ( ) (9) (-3.1) 13 (1) 22 ( ) (2) 7 (1) 28 ( ) (2) 8 (1) ( ) (1) ( ) (2) 3.0* 9 ( ) (2) 235 (23) ( ) (25) 223 (30) ( ) (21) 146 (18) ( ) (26) ( ) (25) ( ) (29) 1.3 (-1.7) 0.7* ( ) 1.5 (-2.2) 1.4 (-2.0) 1.8* ( ) Small/solo 213 (44) Group/HMO 830 (53) 203 (42) 771 (49) 1.4 (-1.9) (18) 1.3 (-1.8) ( ) (5) ( ) 24 (5) 96 ( ) 15 (3) ( ) (7) 10 (2) ( ) (1) 105 (22) ( ) (24) ( ) 10

11 City/state/fed government (55) ( ) Medical school (57) ( ) 173 (48) 29 (51) 1.2 ( ) 1.4 ( ) 50 (15) 4 (7) ( ) 0.4 ( ) 53 (16) 5 (8) 1.2 ( ) 0.5 ( ) 21 4 (7) ( ) 1.3 ( ) (10) ( ) 2 (3) 0.6 ( ) 37 (12) 8 (14) 2.3 ( ) 2.6 ( ) 5 (2) 1 (2) 0.8 ( ) 0.9 ( ) 85 (26) 17 (29) 0.9 ( ) ( ) Changing How Care Gets Paid For Cutting Payment to Physicians Directly Limiting access to expensive treatments with little net benefit Using costeffectiveness data to determine available treatments Higher patient copays High deductible health plans Penalizing providers for avoidable readmissions Paying a network of practices a fixed, bundled price Eliminating fee-forservice payment models Allowing Medicare payment cuts to doctors to take effect Reducing compensation for the highest-paid specialties Practice Compensation Type Billing only 485 (48) Salary plus bonus 437 (52) Salary only 246 (56) *p-value < (44) 419 ( ) (49) ( ) (50) 197 (20) ( ) (15) ( ) (15) 198 (20) (0.5-) (15) ( ) (14) 55 (5) (0.5-) (5) (0.4-) (7) 39 (4) 56 ( ) (7) ( ) (9) 29 (3) (-3.3) (9) 2.4* 57 ( ) (13) 17 (2) 3.3* 11 ( ) (1) 4.3* 7 ( ) (2) 188 (19) ( ) (24) 294 ( ) (34) 1.4 (-1.9) 2.2* ( ) 11

12 eerences 1. Kim J, Mueller C. Factor Analysis: Statistical Methods and Practical Issues. ewbury Park, CA: Sage Publications, Inc; Kaiser H. The application of electronic computers to factor analysis. Educational and Psychological Measurement. 1960;20: Cattell R. The scree test for the number of factors. Multivariate Behavioral Research. 1966;1: