Understanding and Preventing Loss to Follow-up: Experiences From the Spinal Cord Injury Model Systems
Top Spinal Cord Inj Rehabil, 2018 · DOI: 10.1310/sci2402-97 · Published: January 1, 2018
Simple Explanation
Longitudinal research faces challenges due to participants dropping out, which can bias the study results. To ensure the findings are widely applicable, it's important to understand what causes this attrition and focus on retaining those at high risk of not participating. This study examined factors linked to loss to follow-up (FU) among 25,871 individuals with spinal cord injury (SCI) in the National Spinal Cord Injury Database, defining loss to FU as the absence of research information from eligible participants. The study found that people who are marginalized in society are more likely to be lost to FU. These findings can help identify individuals less likely to participate in follow-up, allowing for targeted interventions to improve their response rate.
Key Findings
- 1Loss to FU rates were 23.1% at post-injury year 1 and 32.9% at year 5, exceeding 40% between years 20 and 35.
- 2People at risk of being marginalized, including non-whites, those with less education, the unemployed, victims of violence, and those without health insurance, had the highest odds of being lost to FU across all post-injury years.
- 3The study found that the FU rate varied by study sites and was improved in recent injury cohorts.
Research Summary
Practical Implications
Targeted Interventions
The study's findings can be used to identify individuals at high risk of loss to follow-up, allowing for targeted interventions to improve their response rate.
Improved Study Design
The study highlights the importance of incorporating strategies to maximize follow-up participation from initial enrollment in longitudinal studies.
Policy Implications
The study underscores the need to address socioeconomic disparities that contribute to loss to follow-up, ensuring equitable representation in research.
Study Limitations
- 1The study findings may not be generalizable to all people with SCI in the United States, as participants were recruited from SCIMS centers, not a population-based sample.
- 2Deaths and other technical reasons for non-participation were excluded, limiting the assessment of study attrition's complete picture and bias.
- 3The study was not able to examine factors such as data collection practices, retention strategies, and interviewers’ skills due to NSCID design constraints.