Table 2.

Weighted combination (%) of the 7 candidate learners that define the 11 Super Learners for estimating the denominators of the IPW weights. Each candidate learner is defined by a type of algorithm (’glm’ or ’polyclass’) and a subset of explanatory variables identified by univariate regressions based on a particular significance level (p-value).

Learner	glm	glm	glm	glm	glm	polyclass	polyclass
Subset of explanatory variables	p≤1e-30	p≤1e-10	p≤1e-5	p≤0.1	p≤1	p≤1e-30	p≤1
met initiation at t=0	0	0	0	8.3	6	9.8	76
sul initiation at t=0	0	0	0	0	14	35	50
met+sul initiation at t=0	13	0	4.9	0	28	0.37	53
other therapy initiation at t=0	0	0	0	2.3	34	0	64
met initiation at t>0	0	0	0	0	0	16	84
sul initiation at t>0	0	0	0	0	5.2	59	36
met+sul initiation at t>0	0	0	0	0	9.8	19	71
other therapy initiation at t>0	0	0	0	14	24	30	32
censoring by health plan disenrollment	0	0	0	0	32	40	28
censoring by death	0	0	54	3.1	7	0	36
insufficient GFR monitoring	0	0	0	0	0	0	100