Panaich et al conducted 1:1 propensity score matching between angioplasty and stenting in patients with peripheral arterial disease to predict mortality/complication in hospital with the use of multivariate logistic regression model. The investigators concluded that endovascular stenting was associated with a lower rate of mortality/complication in hospital compared with angioplasty alone. I have concerns on their statistics.

First, Austin examined the performance of 1:1 propensity score matching, stratification on the propensity score, inverse probability of treatment weighting (IPTW) using the propensity score, and covariate adjustment using the propensity score to estimate marginal hazard ratios by Monte Carlo simulations. Of 4 methods, the investigator recognized that both propensity score matching and IPTW using the propensity score allow for the estimation of marginal hazard ratios with minimal bias, and IPTW using the propensity score resulted in estimates with lower mean squared error. From these estimations, statistical method by Panaich et al is relatively acceptable when estimating the relative effect of treatment on time-to-event outcomes.

Relating to the first query, Deb et al reported that the matching technique would be ideal in an observational study in which there are more patients in the control group than in the treatment group, and the resulting match is well balanced. They also mentioned that if 1:N matching was difficult or the investigator wants to avoid losing data, the IPTW technique might be more appropriate than propensity score matching. Although recent review presented that balance was more often checked in studies using propensity score matching and IPTW using the propensity score, 1:N matching should also be considered for more valid evaluation of endovascular stenting.

Finally, the investigators did not include smoking habit as a lifestyle factor for their analysis. In addition, there is a discrepancy in statistical result in the effect of age on mortality/complication in hospital.