Validation of a novel prediction model for early mortality in adult trauma patients in three public university hospitals in urban India

Table 2 Outline of models

Model	Parameter	β^a (SE)
Basic^b	SBP, SBF 1	−0.026 (0.006)
	SBP, SBF 2	−0.044 (0.036)
	SBP, SBF 3	0.419 (0.244)
	GCS	−0.228 (0.031)
	Intercept	2.302 (0.577)
Basic + HR^b	SBP, SBF 1	−0.024 (0.008)
	SBP, SBF 2	−0.042 (0.038)
	SBP, SBF 3	0.378 (0.256)
	HR, SBF 1	−0.003 (0.013)
	HR, SBF 2	−0.018 (0.121)
	HR, SBF 3	0.225 (0.559)
	GCS	−0.229 (0.032)
	Intercept	2.214 (0.662)
Kondo et al.	GCS	−0.254 (0.030)
	Age < 60 years	−0.225 (0.354)
	SBP < 60	0^c
	60 ≤ SBP ≤ 120	−2.067 (0.525)
	SBP > 120	−2.511 (0.585)
	Intercept	1.919 (0.616)
Perel et al.^d	Age	−0.067 (0.142)
	Age²	0.002 (0.003)
	Age³	−1.3·10^-5 (<0.001)
	SBP	0.019 (0.024)
	SBP²	−0.001 (<0.001)
	SBP³	2.3·10^-6 (<0.001)
	GCS	−0.679 (0.188)
	GCS²	0.024 (0.010)
	Intercept	3.678 (2.016)
Systolic blood pressure model	SBP, SBF 1	−0.285 (0.030)
	SBP, SBF 2	−0.079 (0.247)
	SBP, SBF 3	−0.031 (0.006)
	Intercept	−0.052 (0.033)
Glasgow coma scale model	GCS	0.533 (0.222)
	Intercept	0.595 (0.460)

Coefficients (β) of all models were estimated in the updating sample
^aAfter shrinkage
^bAs reported in the derivation study [22]. In the basic and basic + HR models systolic blood pressure and heart rate were modelled using restricted cubic splines. ^cReference. ^dAge and systolic blood pressure were modelled using cubic terms whereas Glasgow coma scale was modelled using a quadratic term. Abbreviations: GCS Glasgow coma scale, HR heart rate, SBP systolic blood pressure, SBF spline basis function

ISSN: 1471-227X