Objectives: To investigate the occurrence and clinical parameters associated with oral malodour. Methods: Retrospective study of all new adult patients (
³16yrs) visiting oral hygiene department of the University of Pretoria during January and October 2004 (N=896; 41.4% male), and treated by senior oral hygiene students under the supervision of clinical instructors. Medical history, including smoking status was collected by means of self-reporting using standardised medical questionnaires. During the first consultation, the following clinical parameters were also recorded: DMFT, plaque index (PI), pocket depths, bleeding on probing and tongue coating status (TCS), as well as regular use of additional cleaning aids other than tooth-brushing (floss, mouth rinses and tongue cleaning). Malodour was diagnosed if the average level of volatile sulfur compound (VSC) concentration assessed by Halimeter was
³120ppb and the organoleptic measurement using a 0-5 point scale was
³3. Data analysis included chi-square, t-tests and logistic regression. Results: Prevalence of oral malodour was 15.1% and 20.9% using organoleptic and Halimeter respectively. Organoleptic ratings and VSC values were significantly correlated (Pearson r=0.39). Subjects with oral malodour had higher mean PI (55.7% vs 44.5%; p<0.01) and VSC value (254.4ppb vs 78.2ppb; p<0.01). Those who reported tongue cleaning were less likely to present with tongue coating (17% vs 23%; p<0.05). Significant predictors of oral malodour in a multivariate analysis were periodontal status (Odds Ratio [OR] =3.17; 95% Confidence Interval [CI] 1.58-6.34), TCS (OR = 2.59; 1.66-4.04), flossing (OR = 0.49; 0.29-0.82), PI (OR = 1.02; 1.0-1.02), filled teeth (OR = 0.93; 0.89-0.98) and being female (OR = 0.50; 0.33-0.75). Conclusions: Periodontal status followed by TCS play the most important role in determining oral malodour in the studied population. The modest correlation between organoleptic and Halimeter readings, suggest factors other than VSCs may be associated with oral malodour in the studied population.