We conducted the present study to determine which clinical parameters for diagnosing halitosis are most reliable in the development of an accurate prognosis for oral malodor treatment. Clinical evaluation included a questionnaire, oral examination and volatile sulfur compounds (VSC) measurement, and organoleptic test (OLT) at baseline. Ninety-two patients classified with oral pathologic halitosis were treated in a similar manner. Improved treatment outcome was defined as < 0.25 ppm of total VSC, < or = 2 OLT score, and < 30 on the "sad feeling" scale as subjective stress level by oral malodor 6 months after baseline. Thirty-four patients demonstrated improvement due to the treatment. In a multiple logistic model, pocket depth, OLT score, intra-oral discomfort, and self-perception of oral malodor at baseline displayed significant association with "not improved outcome". These results suggest that malodor- and periodontal-disease-related parameters and self-estimation of malodor at baseline are effective for prediction of outcome.