Table 2:

Tabular form of mathematical prediction model’s (MPMs) base equations. Risk variables are categorized into demographical (subject reported) and radiological (clinician reported) factors. Units are coded in clinical terms; for use in the equation(s), sex (F=1,M=0) and presence (Y=1,N=0) are numerically coded. To obtain a prediction value for a given MPM, multiply each coeffiecnt by the subject’s risk variable value and take the summation with the base intercept/offset. The resulting number is the x in the logistic equation: êx/((1 + êx)) = risk prediction. For example, performing the VA MPM prediction for a 62-year-old, never-smoker, with a 10mm nodule would yield x = (62*0.0779 + 0*2.061 + 0*0.0567 + 10*0.112 – 8.404) = −2.454; plugging into the logistic equation would yield a risk prediction = 0.079.

	Risk variable	Units	MPM Coefficient
			MC	VA	PU	BU
Demographical	Age	Years	0.0391	0.0779	0.07	0.0287
	Sex	F/M				0.6011
	Ever Smoker	Y/N	0.7917	2.061
	Time of smoking cessation	Years		0.0567
	Cancer history	Y/N	1.3388
	Family history of cancer	Y/N			1.267
	Family history of lung cancer	Y/N				0.2961
Radiological	Emphysema	Y/N				0.2953
	Upper lobe	Y/N	0.7838			0.6581
	Diametera	MM	0.1274	0.1 12	0.0676	−5.3854a
	Spiculation	Y/N	1.0407		0.736	0.7729
	Smooth Border	Y/N			−1.408
	Calcification	Y/N			−1.615
	Nodule type	Solid : Y/N				0
		Part Solid: Y/N				0.377
		Non-Solid: Y/N				−0.1276
	Nodule count	Count				−0.0824
	Base Intercept/Offset		−6.872	−8.404	−4.496	0.2761

Definition of abbreviations: MPM – mathematical prediction model; MC – Mayo Clinic; VA – Veteran’s Affairs; BU – Brock University; PU – Peking University; F – Female; M – Male; Y – Presence; N – Absence.

^aIn the BU model, nodule size is defined by (diameter in millimeters/10)^−0.5