Related Experiment Videos
A Validation and Calibration Process for Self-reported Tobacco Use With Participants' Cotinine Levels: An Example
Chao Huang1, Zoe Roberts2, Rebecca Cannings-John3
1Hull York Medical School, University of Hull, Hull, UK.
Introduction:
Reducing smoking in pregnancy was a primary outcome in our Building Blocks trial of the Family Nurse Partnership. We calibrated maternal reports of smoking using cotinine values derived from urine samples to assess tobacco use. This involves identifying the extent to which an individual accurately reports smoking and requires complete and synchronized data collection over time. However, some urine samples may be missed or collected at a different time from self-report (non-synchronized.
Methods:
We used statistical validation processes to address both non-synchronized and incomplete data. First, we examined consistency in reporting behaviors at baseline and follow-up for participants grouped by extent of non-synchronized time of collection. Second, we used data from complete cases to infer values for mothers with missing urine samples at follow-up. We then used Markov chain transition rate matrix constructed to assess the robustness of such inferences.
Results:
Maternal underreporting and overreporting of smoking were consistent across the 870 participants grouped by different levels of noncontemporary data collection (Breslow-Day test: p = .24; chi-square test: p = .69). Using participants' baseline reporting behaviors to infer their follow-ups provided comparable smoking outcomes (4.5 cigarettes/day with SD of 5.5) to the simulated counterparts (4.5 cigarettes/day with SD of 6.0).
Conclusion:
We have demonstrated consistent reporting behavior over time and minimal impact due to nonaligned follow-up urine sample collection. For studies collecting smoking data, this proposed method provided a pragmatic solution to facilitate the calibration process of self-reported tobacco use and retain adequate power without introducing undue bias.
Implications:
Synchronized and completed data collection is essential but very often hard to achieve in smoking related studies. When violated, proper statistical validation process should be followed to minimize the potential bias and loss of power in trial analyses. For this purpose, we provided the Building Block trial as an example to demonstrate how to deal with the non-synchronization and incompleteness issues in data collection.