The Shirom-Melamed Burnout Questionnaire/Measure (SMBQ/SMBM) is a self-report instrument frequently used for assessing degree of burnout and screening for stress-related exhaustion disorder. The aim of the present study was three-fold. First, to examine reliability and construct validity of different versions of SMBM with 6–22 items in a clinical context. Second, to examine the criterion validity by assessing sensitivity and specificity and determining clinical cut-offs for these versions of the SMBM, and third to examine the prevalence of burnout in a general population and primary care sample using the proposed cut-offs. Two Swedish samples were used for the first two purposes: a clinical sample of patients diagnosed with exhaustion disorder (n = 149), and a matched sample of healthy controls (n = 60). For the third purpose a sample from the general population (n = 3406), and a primary care clinical sample (n = 326) was used. The modified versions of the SMBM showed good internal consistency, construct validity, dimensionality and model fit on the clinical exhaustion disorder sample, as well as configural measurement invariance across clinical and non-clinical samples. The sensitivity (94.6%–95.3%) and specificity (93.3%–95.0%) in identifying cases with exhaustion disorder based on the cut-off of 4.0 for the 19-, 16- and 11-items versions, and on the cut-off of 3.75 for the 6-item version was high. The prevalence of burnout was 81.2% in the primary care sample and 16.6% in the general population sample. The findings indicate that the SMBM is a useful instrument for screening for exhaustion disorder and burnout.