The "Miserable Minority" following mild traumatic brain injury: who are they and do meta-analyses hide them?

Ruff et al. (1994; Ruff, Camenzuli, & Mueller, 1996) hypothesized that some mild traumatic brain injury (MTBI) patients will suffer chronic symptomatic complaints and impairments, identifying this subgroup as the "miserable minority." However, several meta-analyses of the effects of MTBI have been published (e.g., Rohling et al., 2011) showing no significant cognitive impairments following recovery. Recently Pertab, James, and Bigler (2009) suggested that meta-analysis might be obscuring impairments in some MTBI patients, presenting a hypothetical score distribution to illustrate their claim. Our statistical analyses of their hypothetical figure and of several other potential distributions containing an impaired subgroup that varied as a function of effect size and base rate of occurrence did not support the existence of a miserable minority that is obscured in meta-analyses by the larger group of MTBI patients experiencing full recovery. Indeed, given our recent published MTBI effect size of -0.07 (Rohling et al., 2011), for an impaired subgroup to exist, the level of impairment would have to be just under a tenth of a standard deviation, equivalent to a WMS-IV Index score value of 1 point. At effect sizes this small, any cut score chosen on a test to diagnose patients would result in more false positives than true positives. This greatly increases the risk of misdiagnosis in persons who are susceptible to misattribution, expectancy effects, and "diagnosis threat," thereby increasing the risk of iatrogenic illness.