Response

We read with great interest the recent paper by Liu et al, who found that long-term exposure to air pollutants is positively associated with an increased risk of schizophrenia.^{Reference Liu, Li, Ma, Tang, Chen and Tian1} However, we note several biases in their use of the Cox proportional hazards (CoxPH) model that the authors did not address.

The established criteria may result in mixed censoring outcomes, i.e. right-censoring and interval-censoring events.^{Reference Thomas, Turakhia and Pencina2,Reference Leung, Elashoff and Afifi3} Events of schizophrenia diagnosed from medical records could result in interval censoring if they occurred between follow-up visits, and right-censoring if diagnosed between the end of follow-up and the time of data analysis. The CoxPH model primarily handles right-censored data. In contrast, the accelerated failure time (AFT) model is often preferred for scenarios involving various types of censored data.^{Reference Lee, Rondeau and Haneuse4} The AFT model can effectively handle left-censored, right-censored and interval-censored data by appropriately adjusting the likelihood function.^{Reference Ning, Qin and Shen5} By using the ‘survival’ and ‘icenReg’ packages, mixed censored data can be fitted and analysed and event times can be estimated.^{Reference Crowther, Royston and Clements6}

In conclusion, we believe that a re-evaluation considering the potential impact of censoring events is necessary. Further research is anticipated to provide more empirical data and clearer insights into this field.