WE AGREE with Narrow et al1 that the National Institute of Mental Health Epidemiologic Catchment Area Program (ECA)2 and National Comorbidity Survey (NCS)3 have 3 problems: implausibly high disorder prevalence rates, discrepancies between rates, and limited usefulness for estimating treatment need. The authors propose adding a clinical significance criterion (CS) to each disorder's DSM4,5 criteria to help solve these problems. Their CS essentially requires that symptoms lead to outpatient service contact or interfere with life or activities "a lot." Unsurprisingly, the more restrictive criteria yield lower prevalence rates. The authors claim that adding the CS reduces discrepancies between ECA and NCS rates and that diagnostic validity increases, as evidenced by higher rates of suicidal ideation or attempts and disability.
Wakefield JC, Spitzer RL. Lowered Estimates—but of What?. Arch Gen Psychiatry. 2002;59(2):129-130. doi:10.1001/archpsyc.59.2.129