Klaske van Sluis

28 2.3. Results Ten studies [27, 34, 35, 37, 38, 42, 44, 49, 52, 54] did not mention any inclusion criteria. Eleven studies [27, 33–36, 38, 42, 44, 46, 46, 50, 54] failed to describe their method of selection and recruitment of participants. Two studies [40, 45] provided a detailed description of the selection process. The remaining articles [30, 32, 37, 39, 43, 47–49] mentioned the selection process briefly. Most patients were recruited via a clinical setting or via support groups. Treatment details of the included groups were only provided in about half of the studies [27, 30–33, 38, 39, 41, 45, 47, 49, 52]. This variable was indicated as present when treatment details were provided. Nevertheless, surgical details, e.g. use of flaps during the surgery were not provided in any of the included studies. 2.3.1 Acoustic outcomes In Table 2.3 acoustic outcomes for the included studies that reported on the pri- mary (e.g. F0, HNR, MPT) and secondary outcomes (e.g. jitter, shimmer, inten- sity, spectral tilt) are presented. Comparative results for the different speaker groups are shown. None of the studies performed acoustical analysis on ELS, therefore ELS is not discussed in this section. Fundamental frequency Thirteen papers [27, 32–35, 37, 38, 40–42, 45, 49, 52] ( n =443) reported fun- damental frequency (F0) outcomes, including the level A categorized study of Shim et al. [32] (Figure 2.). Measurements are presented for evaluations in sus- tained vowels and in running speech. No distinction between male and female speakers is made. Most studies did not make this distinction since the sound source, the PE-segement, is similar in both groups. Not all F0 outcomes could be taken in account because in some studies the reporting was only range of F0 or in boxplots [37, 40]. Higher F0 values are designated as better [1]. The total range of F0 values for all groups of speakers in vowels and running speech is 64 - 179 Hz, which is reported in twelve studies [27, 32–35, 38, 40–42, 45, 49, 52]. The mean F0 value of 227 Hz for TES [52] and the mean F0 value of 246 Hz for ES [41] are considered outliers and were therefore excluded. The level A rated study of Shim et al. [32] showed non-significant higher mean F0 values for ES compared to healthy speakers, resp. 131 Hz and 124 Hz. Higher mean F0 values are found for healthy speakers compared to ES and TES ( n =7) [27, 33, 34, 38, 40, 42, 49]. In two studies this difference was significant [27, 34]. For the speech rehabilitation methods higher F0 values are seen in the group of TES as compared to ES ( n =7) [27, 33–35, 42, 49, 49]. In four studies this difference was significant [27, 33, 34, 52]