124 The study protocol of this prospective cohort study was approved by the Medical Ethics Committee of the Netherlands (2013.301(A2018.307)-NL45051.029.13), and is part of the NET-QUBIC cohort study.10 Patient data about age, sex, tumor stage,11 tumor site, and treatment were used. Patients were assessed before primary treatment (baseline, M0), and 3 (M3), 6 (M6), 12 (M12), and 24 months after treatment (M24). At every assessment, the primary outcome measure in the present study (100 mL Water Swallow Test (WST)) was performed. 100 mL Water Swallow Test During the WST, a subject is asked to drink 100 mL of water as quickly as is comfortably possible. The time to swallow 100 mL (in seconds) and the number of swallows are counted, both by the subject and the researcher. Timing starts when the water touches the bottom lip, and stops when the larynx comes to rest after the last swallow.12 Persons fail the test when they cough or choke post swallow, have a wet voice quality post swallow, or are unable to drink the whole 100 mL.13 When a person is unable to drink the 100 mL, the residual water is measured and noted. As shown in previous research, the number of swallows had an excellent reliability (Intraclass correlation coefficient (ICC)=0.923) when comparing test and retest, while the swallowing duration had a slightly lower reliability (ICC=0.893). Swallowing duration needed a larger Smallest Detectable Change (SDC%) and Standard Error of Measurement (SEM%) (16.5% versus 52.8%, and 5.9% versus 19.1%, respectively) in comparison to the number of swallows.14 Therefore, in the current study, the number of swallows was chosen as primary outcome measure. A higher number of swallows indicates more swallowing problems. Data from previous research was used to calculate a cut-off value.14 A value larger than two standard deviations from the mean value of healthy subjects was used to indicate swallowing problems in patients with HNC (≥8 swallows needed to drink 100 mL of water).15 Swallowing dysfunction was defined as a failure on the WST and/or a value above the cut-off value of eight number of swallows needed to swallow 100 mL of water.15 Apart from the cut-off value, the SDC found in previous research (0.79 swallows) indicates whether the difference between measurements is a real difference and not a measurement error.14 Statistical Analyses Descriptive statistics were used to describe the study population. A Kruskal-Wallis H test was performed to examine differences in age between primary treatment groups, and a chi-square test was run to test for differences in sex, tumor site and tumor stage between primary treatment groups. A linear mixed-effects model (LMM) analysis was conducted to investigate changes over time in number of swallows, and the association with patient and clinical factors.16 Akaike’s Information Criterion (AIC) was used to select the most