Verbal fluency tests are one of the most commonly used measures of executive functioning in neuropsychological testing and play an important role in the assessment, diagnosis and care planning of patients with a variety of conditions, including brain tumour. There is little conclusive evidence about which factors may influence verbal fluency outcomes. No studies to date have investigated the interactions between a comprehensive range of demographic variables, mood scores, tumour factors and key cognitive skills with the focus of verbal fluency outcomes in brain tumour patients. Similarly, clarification is required across studies assessing the localisation effects of verbal fluency skills. To address these gaps in the evidence base this study used a retrospective cohort design of cross-sectional data from patients with brain tumours, to investigate their performance of both phonemic and semantic verbal fluency. More specifically this study used simple linear and multiple regression calculations to analyse the interactions between these variables and other potentially important factors such as localisation, depression and anxiety (using the HADS), age, gender, education, premorbid functioning (using the TOPF), semantic memory (using the GNT), and tumour type.
The results showed that an increase in phonemic fluency performance was significantly correlated with being educated, an increase in semantic memory, and an increase in premorbid functioning. Phonemic fluency was also significantly correlated with localisation. In general, a decrease in phonemic fluency was significantly associated with tumours in the left frontal lobe. An increase in semantic fluency was correlated with an increase in semantic memory. No other factors showed significant associations with phonemic or semantic fluency. The outcomes from the hierarchical multiple regressions indicated that localisation, gender, education, tumour type, depression, semantic memory, and premorbid functioning when combined can predict phonemic fluency variance. Combining localisation effects, semantic memory, depression and education together do not result in a model that predicts semantic fluency, as within this model the only significant relationship was between semantic memory and semantic fluency. These findings show that, for brain tumour patients, it is important to take into consideration tumour localisation, education, semantic memory, and premorbid functioning when assessing and care planning for deteriorations in phonemic fluency. Similar patients with deteriorations in semantic fluency need to have their results considered in light of performance in semantic memory tests.
