BACKGROUND/AIMS: Mild cognitive impairment (MCI) is present in up to 34% of patients with early-stage Parkinson disease (PD); however, it is difficult to detect subtle impairment without objective cognitive testing. METHODS: Data were obtained from the Parkinson Progression Marker Initiative. All 341 participants were administered the Montreal Cognitive Assessment (MoCA) and a brief neuropsychological battery. Participants were classified as PD-MCI if MoCA was <26 or if they scored >/=1 standard deviation below the normative mean in 2 or more domains, based upon established criteria. The sensitivity/specificity for the clinical detection of PD-MCI was determined. RESULTS: Overall accuracy for clinical detection of PD-MCI was 67.4%. Although clinical determination was highly specific (96.3%; 95% confidence interval [CI]: 0.92-0.98), sensitivity was poor (32.0%; 95% CI: 0.25-0.40). CONCLUSION: Identifying MCI in early-stage PD based on clinical interview alone appears to be insufficient. The inclusion of objective cognitive tests allowing for normative sample comparisons is needed to increase the detection of cognitive impairment in this population.