Social desirability bias in face to face interviews

摘要

We present our reply to the comments by Tin SS et al.w Serum high-density lipoprotein (HDL) cholesterol levels were determined with an automated system using the colorimetric method (Architect C8000, Abbott Park, IL, USA); a routine and common measurement method. The measurement was not manual. While the higher percentage of smoking observed in men may have contributed to lower HDL-cholesterol levels in this group, there are no significant differences with respect to gender and smoking status between group-I (moderate to high-risk Duke Treadmill Score (DTS)) and group-II (low-risk DTS). Hence, it is unlikely for smoking status to have played a role in the sex-specific difference we have observed in this study. We agree that age has an impact on DTS. The adjustment of DTS according to age with the relevant P value and the adjusted odds ratio with 95% confidence interval can be found in Table 2.