Table 5. Multivariable GEE logistic regression of surgical complications with surgeon volume analyses categorized by piecewise linear splines.

Comparison	Nationwide Inpatient Sample			State Inpatient Database
	Adjusted odds ratio (95% CI)	P value		Adjusted odds ratio (95% CI)	P value
1-unit SV increase when volumes ≤3	1.04 (0.84–1.29)	0.72		–	–
1-unit SV increase when volumes [4, 33]	1.03 (1.01–1.05)	0.004		–	–
1-unit SV increase when volumes ≥34	0.99 (0.96–1.01)	0.20		–	–
1-unit SV increase for all SV volumes	–	–		1.00 (1.00–1.01)	0.32
Age (1 unit)	1.01 (1.00–1.02)	0.23		1.00 (0.99–1.02)	0.37
Male vs. female	0.66 (0.49–0.89)	0.007		1.01 (0.77–1.32)	0.95
Race
Black vs. White	1.56 (0.81–3.02)	0.18		1.23 (0.62–2.45)	0.55
Hispanic vs. White	0.96 (0.64–1.42)	0.82		–	–
Other races vs. White	1.00 (0.61–1.63)	>0.99		0.86 (0.63–1.17)	0.33
Payor
Medicaid vs. private insurance	0.68 (0.36–1.28)	0.23		0.59 (0.39–0.89)	0.013
Medicare vs. private insurance	0.78 (0.48–1.26)	0.30		0.91 (0.60–1.37)	0.65
Other insurance vs. private insurance	0.92 (0.50–1.68)	0.78		0.46 (0.26–0.81)	0.007
Hospital size
Large vs. small bed size hospital	0.88 (0.55–1.40)	0.59		–	–
Medium vs. small bed size hospital	0.78 (0.44–1.37)	0.38		–	–
Teaching vs. non-teaching hospital	0.81 (0.57–1.15)	0.24		–	–
Hospital geographic region
Midwest vs. south	0.84 (0.43–1.64)	0.61		–	–
Northeast vs. south	0.77 (0.52–1.13)	0.18		–	–
West vs. south	0.79 (0.37–1.7)	0.54		–	–
Urgent vs. not urgent admission	1.61 (1.02–2.56)	0.042		1.15 (0.81–1.63)	0.45
Elixhauser comorbidity (1 unit)	1.03 (1.00–1.05)	0.02		0.94 (0.92–0.96)	<0.0001

GEE, generalized estimating equation; CI, confidence interval; SV, surgeon volume.