On 2016 Apr 12, Oliver Kuss commented:

This is a clearly written paper on proportional hazards model for interval-censored data. I especially liked the authors sharing their SAS code for the piecewise constant model with which I was able to recalculate the results for the breast cancer data set as given by Lindsey/Ryan.

However, there are also two things that concern me a bit:

(1) The SAS code seems to be a only slightly shortened version of the %recurr macro of Lu/Liu (2008). It would have been fair by Gong/Fang to point to this source.

(2) The estimated hazard ratios for the breast cancer data set (table 6) do not coincide with the previous literature. For example, Lindsey/Ryan report 0.930 (0.287) as the hazard ratio from the piecewise constant model and I found the same results. However, the authors report 0.392 ...

Lindsey JC, Ryan LM. Tutorial in biostatistics methods for interval-censored data. Stat Med. 1998 Jan 30;17(2):219-38.

Lu L, Liu C. Analysis of Correlated Recurrent and Terminal Events Data in SAS®. www.lexjansen.com/nesug/nesug08/sa/sa16.pdf

On 2016 Apr 12, Oliver Kuss commented:

This is a clearly written paper on proportional hazards model for interval-censored data. I especially liked the authors sharing their SAS code for the piecewise constant model with which I was able to recalculate the results for the breast cancer data set as given by Lindsey/Ryan.

However, there are also two things that concern me a bit:

(1) The SAS code seems to be a only slightly shortened version of the %recurr macro of Lu/Liu (2008). It would have been fair by Gong/Fang to point to this source.

(2) The estimated hazard ratios for the breast cancer data set (table 6) do not coincide with the previous literature. For example, Lindsey/Ryan report 0.930 (0.287) as the hazard ratio from the piecewise constant model and I found the same results. However, the authors report 0.392 ...

Lindsey JC, Ryan LM. Tutorial in biostatistics methods for interval-censored data. Stat Med. 1998 Jan 30;17(2):219-38.

Lu L, Liu C. Analysis of Correlated Recurrent and Terminal Events Data in SAS®. www.lexjansen.com/nesug/nesug08/sa/sa16.pdf