Characteristics of Participants who Fail to Complete Research Studies

Past Studies

Past Studies demonstrate that up to one half of participants in long-term outcome studies for traumatic brain injury are unable to be located once they leave inpatient treatment. The loss of participants from a research project can effect the final results of the study. Because the participants who do not complete a study may have characteristics that are different from the participants who do complete a study, the final results may be overly representative of those that did complete the project. When this happens, the research results are termed “biased” or slanted. Biased research results are concerning because they can prevent researchers from drawing generalizable conclusions. There has been little research about the effects of participant dropout in traumatic brain injury outcome studies.

This Study

This Study analyzed data about traumatic brain injury research participants from three settings. The participants were adolescents and adults hospitalized with a traumatic brain injury. The characteristics were compared for research participants that were both lost and found for one and two-year follow-ups. Subjects were considered “lost” from the study when no information or only limited information was collected from them for any reason, including that they could not be located, death, and refusal.

About 42% of the participants were lost in the first year. Two of the three facilities in this study attempted to contact individuals in the second year, whether or not they were lost or found in the first year. During the second year, 44.9% to 48.6% of the participants were lost. The most common reasons for participant loss were that they were not locatable or, if located they refused directly or did not respond. Individuals who were not available during the first and second year follow-up tended to be individuals identified as socioeconomically disadvantaged, individuals identified as belonging to a racial and or an ethnic minority, those unemployed, those with little education, and those dependent on public funding for hospital payment. They also were more likely to have a pre-injury history of alcohol or other drug abuse and had received violence-related traumatic brain injuries, either from self-infliction or from the violent act of another person. Individuals with more severe injuries, who were institutionalized or dependent on caregivers, were located more frequently.

Who May Be Affected By These Findings

Researchers, healthcare providers

Caveats

The researchers suggest that participant dropout may be a problem directly associated with traumatic brain injury. For instance, some participants are not able to complete a research study because of death. Others may have a transient lifestyle associated with a history of violence, disadvantage and/or substance abuse, which makes them more difficult to locate.

Bottom Line

Individuals who dropout of traumatic brain injury research studies appeared to have a history of alcohol or substance abuse, a violence-related traumatic brain injury, and weresocioeconomically disadvantaged. The researchers revealed that between 42% and 48.6% of participants dropped out during the first and second follow-up years of the project. Because of this, a study’s results may not accurately reflect the original group as a whole. In the future, researchers are challenged to seek out methods to enable projects to have generalizable conclusions.

Please take a moment to comment on the value of this abstract:
Click here to take a brief survey