Can inverse probability treatment weighting (IPTW) be used to assess differences of CRBSI rates between non-tunneled femoral and jugular CVCs in PICU patients?

Table 2 Final outcome analysis

Parameter	Estimate	Std error	95% CI	P-value
Causal relative risk	4.67	0.897	[0.872 to 25.05]	0.09
Causal risk difference	0.074	0.039	[− 0.019, 0.167]	0.06
Factual probability	0.0414	0.0166	[0.0088 to 0.0739]	X
Counterfactual probability	0.0279	0.0174	[− 0.0062, 0.0619]	X

Causal risk difference estimated from propensity score weighted data set using marginal structural model. Causal relative risk estimated from propensity score weights created from ps object generated by generalized boosted model and svyglm function extracted the output. Regression standardization stdReg R packages used to measure the factual, counterfactual probability

ISSN: 1471-2334