Skip to main content

Table 3 Extension: heterogeneous effects by practice size (subsample of “non-switchers”)

From: Gender composition in the work environment and physicians’ income from Medicare Part B fee-for-service payments: evidence from longitudinal data

 

log(payments)

log(patients)

log(payments per patient)

 

(1)

(2)

(3)

(4)

(5)

(6)

(7)

(8)

(9)

 

Solo

Small

Large

Solo

Small

Large

Solo

Small

Large

Share of women within specialty

−0.434

0.283

1.282

−0.155

0.091

−0.065

−1.189*

−0.798

−0.767

(1.073)

(1.600)

(3.052)

(1.020)

(0.710)

(0.760)

(0.679)

(0.576)

(0.649)

 

[−2.598,1.730]

[−2.943,3.509]

[−4.872,7.436]

[−2.211,1.902]

[−1.342,1.524]

[−1.598,1.468]

[−2.558,0.180]

[−1.959,0.364]

[−2.076,0.542]

1(women) x (Share of women within specialty)

3.061***

3.789***

2.263***

2.358***

2.822***

1.942***

0.961***

1.013***

0.783***

(0.539)

(0.663)

(0.548)

(0.321)

(0.378)

(0.183)

(0.217)

(0.284)

(0.188)

 

[1.973,4.148]

[2.452,5.127]

[1.158,3.368]

[1.711,3.004]

[2.060,3.583]

[1.573,2.311]

[0.522,1.399]

[0.440,1.586]

[0.404,1.162]

Share of women within practice

0.056*

0.007

−0.013

0.032***

0.016**

0.019

0.013

−0.003

−0.019

(0.029)

(0.012)

(0.041)

(0.012)

(0.006)

(0.014)

(0.018)

(0.005)

(0.015)

 

[−0.002,0.115]

[−0.018,0.032]

[−0.097,0.070]

[0.008,0.056]

[0.004,0.027]

[−0.009,0.047]

[−0.023,0.049]

[−0.014,0.008]

[−0.049,0.011]

1(women) x (Share of women within practice)

−0.012

0.018

0.058

−0.005

−0.000

−0.013

−0.019

0.014**

0.029

(0.058)

(0.012)

(0.088)

(0.012)

(0.010)

(0.021)

(0.029)

(0.006)

(0.020)

 

[−0.129,0.106]

[−0.007,0.042]

[−0.120,0.237]

[−0.029,0.019]

[−0.020,0.019]

[−0.055,0.029]

[−0.077,0.039]

[0.001,0.027]

[−0.012,0.070]

Average patient risk score

0.053

0.005

−0.012

0.032

−0.045

−0.036*

0.063**

0.070***

0.060*

(0.075)

(0.042)

(0.066)

(0.020)

(0.028)

(0.018)

(0.030)

(0.026)

(0.030)

 

[−0.097,0.204]

[−0.080,0.089]

[−0.145,0.121]

[−0.008,0.072]

[−0.102,0.012]

[−0.073,0.000]

[0.003,0.123]

[0.018,0.123]

[−0.000,0.119]

Average patient age

0.034***

0.043***

0.041***

0.022***

0.023***

0.031***

0.011***

0.016***

0.009**

 

(0.003)

(0.007)

(0.005)

(0.004)

(0.002)

(0.003)

(0.003)

(0.005)

(0.004)

 

[0.029,0.040]

[0.030,0.056]

[0.030,0.051]

[0.014,0.031]

[0.018,0.028]

[0.026,0.037]

[0.004,0.017]

[0.005,0.026]

[0.001,0.017]

Constant

7.632***

7.271***

7.182***

3.551***

3.895***

3.271***

4.384***

3.950***

4.114***

 

(0.471)

(0.661)

(0.826)

(0.412)

(0.313)

(0.320)

(0.299)

(0.429)

(0.373)

 

[6.682,8.581]

[5.938,8.604]

[5.515,8.849]

[2.721,4.381]

[3.264,4.526]

[2.626,3.917]

[3.780,4.988]

[3.084,4.816]

[3.361,4.867]

Year FE

YES

YES

YES

YES

YES

YES

YES

YES

YES

Physician FE

YES

YES

YES

YES

YES

YES

YES

YES

YES

Observations

265,869

726,237

1,361,016

265,869

726,237

1,361,016

265,869

726,237

1,361,016

R2

0.883

0.883

0.887

0.900

0.911

0.914

0.919

0.912

0.944

  1. Authors’ analysis of 2012–2020 CMS Physician Utilization and Payment Data. All regressions also include the share of physicians among all providers at the practice level, the share of physicians among all providers at the specialty level, and the total number of physicians at the practice level. “Year FE” and “physician FE” refer to year and physician fixed effects, respectively. The analysis focuses on within-physician variations and thus defines practice sizes based on that of the practice a physician worked at in 2012. Solo practices are defined as those with only one physician in the CMS data; small practices are those with 2–10 physicians; large practices are those with 11 or more physicians. Medicare payments include only those for Part B FFS services rendered by a physician, excluding those for laboratory work, tests and imaging, vaccination, and similar services that are not necessarily provided by a physician. All payments have been converted to real 2020 dollars. Standard errors clustered at the specialty level are reported in parentheses. 95% confidence intervals are reported in the following row in brackets. */**/*** represent p-values below 0.1/0.05/0.01, respectively