Regression methods for investigating risk factors of chronic kidney disease outcomes: the state of the art

BMC Nephrology

Table 4 Examples of available software that handle statistical challenges in progression of CKD

Statistical issue	Software
Statistical issue	SAS	R	STATA
Exactly known time-to-event outcome
Survival regression models	PROC PHREG	survival	stcox
Competing risks models
Cause-specific model	PROC PHREG	survival	stcox
Fine and Gray model	PSHREG macro	cmprsk	stcrreg
Multistate models	PROC PHREG	mstate	stcox
		msm
		tdc.msm
Interval-censored time-to-event outcome
Survival regression models	PROC LIFEREG	intcox	intcens
	EMICM macro	survival
	ICSTEST macro	SmoothHazard
	ICE macro	SmoothHazard
Competing risk with death		SmoothHazard
Competing risk with death		msm
Multistate models		msm
Quantitative outcomes
Generalized estimating equations	PROC GENMOD	gee	xtgee
		geepack
		yags
Mixed models	PROC GLIMMIX	lme	xtmixed
	PROC MIXED	glmer	GLLAMM
	PROC NLMIXED	glmer	GLLAMM
Identification of subpopulation of trajectories
Latent class growth analysis	PROC TRAJ
Latent class mixed model	SASRTM macro	lcmm	GLLAMM
Informative drop-out censoring
Shared random-effects models	PROC NLMIXED	jm	jmre1
Shared random-effects models	CGEE2 macro	jm	jmre1
Joint latent class models		lcmm

ISSN: 1471-2369