Examination of Prospective Diagnostic Thresholds in Premenstrual Dysphoric Disorder

Premenstrual Dysphoric Disorder (PMDD) is characterized by extreme mood symptoms that emerge cyclically in the late luteal phase of the menstrual cycle and subside during menses. About 3-8% of women meet the full DSM-5 criteria for PMDD, which requires the presentation of five symptoms (at least one being an emotional symptom) following the cyclical pattern per cycle, while another 10-11% of women can be characterized as having a less severe menstrually-related mood disorder (MRMD), which requires just one emotional symptom to follow the cyclical pattern per cycle. Women with signs of MRMD have less severe emotional symptoms than those with PMDD. However, both PMDD and MRMD can cause impairment and distress severe enough to warrant treatment. Diagnosis of PMDD or MRMD must follow strict criteria established by the DSM-5 over a period of at least two months of prospective daily symptom ratings. Since retrospective reports of these symptoms is highly subjective and inaccurate, valid diagnosis requires prospectively evaluating daily symptoms against the diagnostic criteria. Typically, symptoms are evaluated through visual inspection of daily ratings across multiple cycles; however, differences in interpretation of the diagnostic criteria during these visual inspections may lead to inconsistencies. Furthermore, the complex diagnostic criteria outlined in the DSM-5 means there is a high risk of diagnostic error due to differing practices and errors in clinical judgment. These concerns suggest the need for a more reliable method for making valid prospective diagnoses. In response to this need, Eisenlohr-Moul et al. (under review) have developed the Carolina Premenstrual Assessment Scoring System (or C-PASS), a computerized standardized diagnostic system aimed at creating shared meaning for PMDD and MRMD across laboratories. Although the C-PASS, described in more detail below, represents an essential step forward in our standardization of these diagnoses, additional work is needed to verify that the cut points selected for making the diagnosis are adequate for identifying the subgroup of women that need treatment. The purpose of the present paper is to use descriptive statistics and ROC analyses to examine how changes to these diagnostic cutoffs influence the frequency and appropriateness of diagnoses.Bachelor of Science in Public Healt