WEDNESDAY, June 3, 2026 (HealthDay News) -- Structured interviews may not be as reliable for diagnosing behavioral health conditions as previously thought, according to a review published online May 28 in JAMA Network Open.Weiyi Xie, from McMaster University in Hamilton, Ontario, Canada, and colleagues conducted a systematic literature review and meta-analysis to estimate the test-retest reliability of standardized diagnostic interviews (SDIs) used to classify common adult psychiatric disorders.Based on 46 studies included in the meta-analysis (8,146 participants), the researchers found that the pooled estimate of SDI test-retest reliability was κ = 0.69. There was substantial between-study heterogeneity (Q534 = 23,578.7; I2 = 93 percent). For substance use disorders (SUDs), reliability was higher than for mental disorders (κ = 0.72 [292 estimates] versus 0.65 [243 estimates]; z = 3.74). Reliability varied by mental disorders, ranging from κ = 0.55 for nonaffective psychoses to κ = 0.74 for bipolar disorders. For SUDs, reliability ranged from κ = 0.59 for hallucinogens to κ = 0.81 for opioids. Diagnostic criteria partially explained between-study variation in SUDs, while methodological quality indicators (e.g., small sample size and retest interval) did not."Our findings show that these interviews are not as reliable or consistent as many people believe," senior author Laura Duncan, also from McMaster University, said in a statement. "These differences suggest that structured interviews work better for conditions with clearer behaviors or timelines than for disorders that rely heavily on personal experiences and interpretation."Two authors disclosed ties to the pharmaceutical industry.Abstract/Full Text.Sign up for our weekly HealthDay newsletter