CorrectCare

Meta-Analysis Hones Estimate of Traumatic Brain Injury in Offenders

Health care providers who work with offender populations—both behind bars and in other settings—see plenty of cases of traumatic brain injury. But what is the actual prevalence rate? This is not an idle question, say researchers at the Medical University of South Carolina, Charleston. Given the many difficulties associated with TBI, a better grasp of prevalence would help in developing correctional policies aimed at improving screening, treatment and management of inmates with TBI, as well as public policies that could improve continuity of care and the social functioning of these individuals upon release to the community.

In the April 2010 issue of the Journal of Correctional Health Care, Eric Shiroma and colleagues report on findings from their meta-analysis of 20 epidemiologic studies on TBI in adult offender populations.

In the general population, prevalence of TBI is estimated. 8.5%. In contrast, the commonly reported rates in the medical literature concerning adult offenders are from 25% to 87%. Among juveniles, the reported range is even broader, from 4% to 74%. Such wide ranges reflect the fact that studies vary in factors such as how terms are defined, how screening is done and the specific population under study.

Shiroma and colleagues sought to systematically analyze the reported prevalence rates in terms of gender proportion, offender type, case definition of TBI and method of determining TBI. From this analysis of 20 studies that met specific criteria and a combined sample of 4,865 individuals, they calculated an overall estimate of TBI prevalence in adult offender populations of 60%. Using a 95% confidence interval, the estimated range is 48% to 72%. (The article describes statistical analysis methods in detail.)

Subgroup Findings
Clearly, this 60% figure comes with many conditions. The rate is overall, without adjusting for differing definitions, methods of determination, proportion of males in the sample (rates of TBI among males are about twice that of females) or other sample characteristics.

The researchers also conducted subgroup analysis based on many of these variables. One important distinction was the individual studies’ definition and severity of brain injury—for example, whether the injury was described as “moderate/severe,” or whether it produced a loss of consciousness. The estimated prevalence of TBI when the subject lost consciousness was 50% (confidence interval 40% to 61%).

TBI is inherently difficult to diagnose, especially in less severe cases, the researchers note, and thus use of extensive in-depth interviews by trained mental health professionals is the “gold standard” because it produces more complete and accurate histories compared to other methods. Studies that used such interviews found an estimated prevalence of 67%, higher than the overall prevalence.

By gender, analysis of the males-only subgroup reveals prevalence of 64%, lower than that of the females-only group, 70%. However, the females had a much larger confidence interval, 50% to 90%, compared to the men, 53% to 76%. In addition, looking only at the cases that involved loss of consciousness, prevalence was greater among men (59% +/- 12%) than women (55% +/- 14%).

The article breaks out other data, and notes study limitations, as well. It also summarizes the poor outcomes and costs associated with TBI. Ultimately, the authors argue that this study—which they believe to be the first of its kind—points to a need for more appropriate allocation of resources and formation of community partnerships to better manage a significant public health concern that afflicts incarcerated individuals at high rates.

[This article first appeared in the Winter 2010 issue of CorrectCare.]