According to trial data handling, what is the impact of data loss?

• A. Loss of data may bias results and lead to incorrect safety conclusions.
• B. Data loss has no impact on trial conclusions.
• C. Data loss only affects formatting.
• D. Data loss improves generalizability.

Answer: (A)

Data loss changes what we can learn from a trial. When data are missing, especially if the missingness is related to the outcome or to the treatment itself, the results can become biased. In other words, the observed data may no longer reflect what would have happened if everyone had complete data, which can distort estimates of how well a treatment works and, importantly, how safe it appears. The bigger the amount of missing data and the more that missingness is linked to outcomes (for example, people with worse results or more adverse events are more likely to drop out), the greater the risk of biased conclusions.

This is why trial data handling emphasizes preventing data loss when possible, documenting why data are missing, and applying appropriate analytical methods or sensitivity analyses to assess how missing data might affect conclusions. For instance, if participants experiencing adverse events drop out, safety findings could underestimate risks; if those with poor response discontinue, efficacy could be overstated. The other choices imply no impact, limited impact, or only formatting issues, which isn’t accurate in the context of GCP data handling.