Olfactory tests and depression outcome
The whole group of patients
For the whole group of patients, the classical t-test indicated that the patients improved in terms of almost all olfactory components: odor threshold (T: t = 2.35, p = .021), identification (I: t = 3.9, p < .001) and general olfactory functions (TDI: t = 2.7, p = .008), while in case of odor discrimination no change was noticed (D: t = 1.42, p = .16) (Figure 2).
Bayesian t-test demonstrated slightly different results (Figure 3). While Bayes factor supported very strongly the presumption about change in odor identification performance (I: B10 = 112.77), neither odor threshold (T: B10 = 1.36) nor discrimination (D: B10 = 0.26) was shown to change, and in these cases Bayes factor supported H0. Instead, general olfactory functions were demonstrated to change very slightly (TDI: B10 = 3.1), which was probably driven by the big change in odor identification.
In terms of depression score, for the whole group classical t-test indicated no evidence supporting the difference between the first and the second measure in depression (t = 0.07, p = .94). Bayesian t-test results provided further strong evidence for the null hypothesis stating that both measures did not, indeed, differ from each other (B01 = 11.69).
Dysosmic patients
For the group of dysosmic patients, the classical t-test indicated that the dysosmic patients improved in terms of three out of four olfactory components: odor threshold (T: t = 3.42, p < .001), identification (I: t = 4.35, p < .001) and general olfactory functions (TDI: t = 3.24, p = .002), while odor discrimination value did not change (D: t = 1.55, p = .12) (see: Figure 2).
Bayesian t-test demonstrated further that the hypothesis about change in odor identification, threshold and general olfactory functions were respectively very strongly and strongly supported (for I: B10 =535.09, for T: B10 = 23.505, for TDI: B10 = 14.02). At the same time, odor discrimination data indicate the null hypothesis (D: B10 = 0.329).
Like in the case of the whole group, also here depression severity did not change significantly, which was indicated by the classic t-test (t = .47, p = .64). Bayesian test, additionally, provided strong evidence for the lack of difference between the first and the second measure (BF01 = 10.11)
Furthermore, no difference was found between the dysosmic and normosmic group in terms of depression severity neither for the first measure (t = 1.49, p = .14) nor the second one (t = 0.4, p = .69). Bayesian factor indicated that the severity of depression in the second measure was the same in both groups (B01 for the first measure = 1.44; for the second measure = 3.35).
Pearson Correlations
The whole group of patients
For the whole group of patients, change in depression severity correlated positively with the change in general olfactory functions (TDI: r = .25, p = .004) and in olfactory identification performance (I: r = .22, p = .011) (Table 4). This finding was respectively strongly and moderately confirmed by the Bayesian factor (for “TDI score” B10 =12.8, for I score: 5.51 (Supplemental Table 1).
Dysosmic patients
For the group of dysosmic patients change in depression severity correlated positively with the change in general olfactory functions (r = .28, p < .001) and in olfactory identification performance (r = .25, p = .003) (Table 5 and Figure 4), what was confirmed by Bayesian factor, strongly for general olfactory functions (B10 = 28.5) and moderately for odor identification (B10 = 8.8) (Supplemental Table 2).