In all medical studies, there are patients who are “lost to follow-up.” They stop taking part in the ongoing appointments, surveys, or tests. In other words, they drop out. In long-term studies, as many as 25 percent of patients end up dropping out. That’s a significant number, which could skew the results. Researchers often assume that these dropouts had worse outcomes than the rest of the patients in the study.
These authors put this theory to the test. They tracked down patients who had a total knee replacement (TKR) as part of a study and who later dropped out. It took two years and a private detective, but all 123 dropouts were found. Each person completed surveys about their knee function, which were compared to the other 440 patients who stayed in the study.
The authors ran separate comparisons for the two different types of artificial joints that were used in the study. Results showed no significant differences in rates of implant failure, overall outcomes, or survival between the dropouts and those still in the study. The main differences were in gender (more dropouts were women) and knee score (dropouts had slightly worse knee function and pain).
This study suggests that dropouts don’t necessarily skew research results, at least not in this case. The authors stress that it is still important to keep patients from leaving the study, but that it may not be worth the cost or effort to track them down.