OBJECTIVES: The study aimed to investigate two proposed modifications promoting the consistency and overall quality of the cTTO data.  METHODS: The effect of the modifications was tested in the Netherlands (NL) and Hong Kong (HK) using a 2-arm study with a crossover design.  In each jurisdiction, half of the interviewers started with the existing EQ-VT design (control arm), followed by the modified version of separating the BTD and WTD task in cTTO exercise (experimental arm) after completing 25 (NL) and 30 (HK) interviews.  Another half of the interviewers started with the modified version of cTTO exercise and switched back to the current version in the same manner as the other group.  The two study arms would be switched again after completing further 25 and 30 interviews respectively until they have jointly reached the target sample.  The FB module was offered to all respondents collecting their feedback on the initial responses and giving the opportunity to make changes.  The impacts were evaluated through the proportion of respondents with WTD responses and consistency of observed values for the health status.  RESULTS: A total of  404 and 403 subjects were recruited in NL and HK respectively.  About one-third of the respondents used the FB module to remove one or more responses (around 5% in total TTO responses) in both jurisdictions.  This lowered the proportion of respondents with one or more inconsistent responses from 17.8% to 10.6% (p=0.003) in NL and from 31.2% to 22.3% (p=0.003) in HK.  However, these two jurisdictions reported opposite results of inconsistent responses regarding the separation of the BTD and WTD task and more inconsistent results were found in the experimental arm in Hong Kong.  CONCLUSIONS: The findings support the decision to implement a FB module in TTO tasks, but not the separation of the BTD and WTD task.