It is clear that the lack of standardization in threshold criteria and measuring techniques is the cause for this large variation. For example, a study that was conducted in Sweden (Soder et al. 2000), which used “strong, noticeably unpleasant odor” (i.e., “Foetor ex ore”) as the criterion for breath odors, reported 2.4% prevalence in a population of 1681 subjects. Using such a high threshold that is equivalent to an odor judge score “4” (i.e., “strong malodor”) on a 0–5 scale yielded a similarly low prevalence (2.14%) in a study done in Swaziland (Bornstein et al. 2009). However, in the same study, when an odor judge score of “2” and above was applied as a threshold for malodor (i.e., “mild, clearly noticeable malodor”), prevalence reached 31.5%.

Some of these studies regarded odor judge scores of “2” and above as the threshold value for breath odors (Bornstein et al. 2009; Iwanicka-Grzegorek et al. 2005; Liu et al. 2006; Rosenberg et al. 2007). Using this criterion, they reported that 27.5–31.5% of the subjects were positive for breath odors, concluding that the prevalence of breath odors in the general population is a little over 25%.

A few studies based their breath odor assessment solely on oral VSC levels (Miyazaki et al. 1995; Miyazaki et al. 1997), as measured using a sulfide monitor (i.e., Halimeter), whereas others used VSC measurements adjacent to organoleptic scorings (Bornstein et al. 2009; Iwanicka-Grzegorek et al. 2005; Liu et al. 2006). These studies regarded different VSC concentration values as cutoff points for breath odor conformation (e.g., 75, 110, 125 ppb, respectively). It is important to stress that although VSC readings using Halimeter are in high correlation with organoleptic scores and their value is an important parameter for clinical follow up, the absolute value of a given reading would vary between different instruments depending on the sensor’s freshness. Therefore, setting a universal cutoff value would be impractical.