A closer look at the baseline-observation-carried-forward (BOCF)

Area of Science:

• Clinical trials methodology
• Statistical analysis in clinical research
• Pain management

Background:

• Missing data is a common challenge in clinical trials, particularly in studies with early treatment discontinuation.
• The baseline-observation-carried-forward (BOCF) method is one approach to address missing data, but its modifications are crucial for accurate interpretation.
• Understanding the reasons for treatment discontinuation (e.g., adverse events, lack of efficacy) is vital for appropriate data imputation.

Purpose of the Study:

• To examine modifications of the baseline-observation-carried-forward (BOCF) approach for handling missing data in clinical trials.
• To evaluate the impact of different statistical methods on study outcomes, considering treatment-related and non-treatment-related reasons for discontinuation.
• To assess the efficacy of duloxetine compared to placebo in chronic pain trials using various data imputation techniques.

Main Methods:

• Two placebo-controlled duloxetine chronic pain trials were analyzed.
• Missing data in the primary efficacy outcome were handled using five statistical methods: mixed-model repeated measures (MMRM), last-observation-carried-forward (LOCF), BOCF, modified BOCF (mBOCF), and aeBOCF.
• Reasons for discontinuation were categorized as treatment-related (adverse events [AEs], lack of efficacy) and non-treatment-related.

Main Results:

• Duloxetine demonstrated superiority over placebo in reducing average pain ratings across most methods, including MMRM, LOCF, BOCF, and mBOCF.
• The aeBOCF method showed duloxetine superior to placebo in one study and numerically better in the other.
• Statistical significance (P-values) varied slightly across the different analytical methods employed.

Conclusions:

• The choice of statistical method for handling missing data significantly influences the interpretation of clinical trial results.
• It is essential to consider the actual treatment outcomes of patients and the reasons for discontinuation.
• Employing sensitivity analyses with different statistical approaches can enhance the robustness and accuracy of study findings.