Issues of loss to follow up in a longitudinal study of traumatic brain injury

Longitudinal studies in neurological trauma and disease have faced many challenges not experienced by other areas of medical research. The inclusion of people who experience cognitive dysfunction due to their neurological disorder makes them unique. Studies of other areas of medicine usually exclude people displaying cognitive disorders as they are difficult to recruit, retain and gain useful information from. Retaining participants within longitudinal studies is important so that the study's validity is maintained, leading to the production of generalizable results. This is the key objective of cohort studies, to ensure the benefit of the research can be made to the broader population (Booker, Harding, & Benzeval, 2011; McGonagle, Couper, & Schoeni, 2011; Newberry, Sherwood, Hricic, Bradley, Kuo and Crago, 2011; Ribislk, Walton, Mowbray, Luke, Davidson, Bootsmiller 1991). Traumatic brain injury (TBI) is among the most common causes of serious neurological disorder in adults and research is the key to understanding the needs of those who sustain TBI and the needs of those who are the carers of people with TBI (Helps, Henley, & Harrison, 2004-05; Zasler, Katz, & Zafonte, 2007, p.3). The most effective way to carry out this research is through longitudinal studies, although these are difficult to establish and maintain. This is because recruitment often involves consent by proxy due to levels of disorientation, confusion or coma experienced by the potential participant. In addition retention within the study relies on the memory of the participant or caregiver and regular reminders by research staff. Furthermore lengthy interviews are often needed for the information to meet the needs of the research. These three factors make TBI studies time consuming and expensive. Increasing competition and accountability for financial and human resources, in an environment where people are less inclined to participate in research, means it is important to understand the cohort thoroughly to maximise the information gained and ensure part of the study population is not excluded. (Booker et al., 2011; Newberry et al., 2011; Ribislk et al., 1991). Once time is allowed for recruitment and initial interview/assessment the priority becomes the maintenance of the cohort between review points. Without the adequate retention of participants a study loses information, loses power and also faces the possibility that it is no longer relevant to the total TBI population which the study was established to help. Loss to follow up in TBI studies is an issue that requires consideration. TBI research has been identified as a difficult area of research which experiences a loss to follow up rate of 30-50% over a one to two year follow up (Corrigan, Bogner, Mysiw, Clinchot, & Fugate, 1997; Corrigan, Harrison-Felix, Bogner, Dijkers, Terrill & Whiteneck 2003). Such high losses leave a gap in our knowledge as there is no way of knowing how these individuals recover and what effect their injury has on their lives and those around them. It may further indicate an area of unmet need in the community that is yet to be identified, leaving the people affected exposed to medical and psychological impediments. In addition to exposing an area of unmet need it may be possible to predict who is at risk of dropping out of studies. If people at risk of non-participation in studies and in therapy can be recognised at the time of trauma strategies can be established to assist people to return for follow up interview or treatment. This information is needed by researchers and clinicians to underpin the methodology of future research and the direction of clinical practice ( Zasler et al., 2007, p.38) The influential work of Corrigan et al., (2003) stated participants were lost to follow up if they could not be located, refused to be interviewed, died, made not response to contact or were unable to be identified.