Skip to main content
. Author manuscript; available in PMC: 2023 Nov 29.
Published in final edited form as: Am Econ J Econ Policy. 2023 Nov;15(4):37–74. doi: 10.1257/pol.20210625
Catchment Area Volume Hospital Volume
(1)
Mean
(2)
β
(3)
p-value
(4)
Mean
(5)
β
(6)
p-value
Bed Days 2.134 0199***
(0.053)
0.368 2.152 0.103**
(0.044)
0.540
Bassinet Days 2.310 0.068
(0.117)
0.786 2.353 0.029
(0.098)
0.820
RWP 0.503 0.012***
(0.003)
0.883 0.503 0.009***
(0.003)
0.615
Procedures 2.091 0.063
(0.096)
0.822 2.129 0.129
(0.103)
0.775
Diagnostics 0.095 0.074***
(0.021)
0.339 0.097 0.052***
(0.018)
0.621
C-Section 0.280 0.030***
(0.005)
0.712 0.280 0.027***
(0.005)
0.647
Severe Complication 0.033 −0.018***
(0.005)
0.298 0.036 −0.021***
(0.004)
0.081
Preventable Complication 0.183 −0.088***
(0.018)
0.507 0.191 −0.087***
(0.021)
0.532
Severe Acute Maternal Morbidity 0.011 −0.003
(0.002)
0.050 0.012 −0.001
(0.002)
0.157
28-Day Infant Mortality (per 1,000) 0.828 0.126
(0.771)
0.993 0.791 −0.531
(0.592)
0.909
1-Year Infant Mortality (per 1,000) 1.610 −0.292
(1.090)
0.498 1.593 −0.734
(0.800)
0.579
Unplanned Readmissions (per 1,000) 11.669 −3.511*
(2.064)
0.682 12.029 −6.879***
(2.086)
0.946
Unplanned Child Readmit (per 1,000) 35.444 −4.095
(4.675)
0.986 36.111 −7.327
(5.151)
0.536
30-Day Child Hosp. (per 1,000) 54.999 −17.458***
(5.558)
0.508 54.052 −15.618***
(5.965)
0.936
1-Year Unplanned Child Hosp. (per 1,000) 84.219 −3.460
(6.297)
0.122 84.296 −10.745
(6.506)
0.112
Any 1-Year Child Hosp. (per 1,000) 103.296 −16.180**
(6.340)
0.297 102.188 −20.088***
(7.225)
0.182
N 56,647 90,821

Notes: All estimates considered in this table use the first difference between the current and prior birth admission. Means are shown for the current birth. All regressions include controls for calendar year, age group, sponsor’s pay grade and occupation, and sponsor’s race. Columns 1–3 control for the total volume of surgical inpatient cases in 2004 in counties that fall within 40 miles of the nearest base. Columns 4–6 control for the number of surgical inpatient cases performed at the delivery hospital in 2004, the most recent year of data availability for off-base hospitals. Hospitals are identified using zip codes. In zip codes with more than one surgical hospital we take the average. Not all zip codes are present in the 2004 AHA data, resulting in a somewhat smaller sample. Standard errors are clustered at the base level.