Abstract Aims Patients with brain tumours and headache commonly have poorer cognitive skills, either overtly or covertly, when cognitively tested. Cognitive changes reflect, tumour mass, fronto-temporal location or hydrocephalus, Previous work has demonstrated that the “semantic Verbal Fluency Test (SVFT) -“How many animals can you think of in a minute?” is a useful fast screening test for cognitive issues. Median SVFT in patients with brain tumour on admission is 10 animals. Most GPs can now order “direct access cerebral imaging (DACI)” in patients with headache suspicious of cancer. The waiting times for scanning can be many weeks. The aim of this study was to determine whether low SVFT scores: might be useful to help stratify or expedite DACI. We present data from referrals through and electronic Protocol Based Referral (PBR) pathway for CT scanning over 3 years, to determine whether SVFT might be a useful adjunct to history and examination. Method From 2017, in Edinburgh/Lothians, Scotland, an electronic PBR was developed with involvement of Primary Care Cancer Lead, PBR lead, Neurology, Neurosurgery and Neuro-Imaging for outpatient imaging of patients in the community with Headache Suspicious of Cancer, to expedite their scans. The PBR sat alongside the routine outpatient DACI system. If the forms were correctly filled in Neuro-Radiology prioritised their appointments. The referrer (GP) was asked to complete the ePBR form and SVFT at the time of referral. Other data were also gathered, including: Past Medical History of cancer; other symptoms/signs; and co-morbid conditions and medications filled automatically from the GP system. This formed the dataset. We also retrospectively assessed a) whether English was first language b) past history of Pain Clinic Attendance or Functional Illness and subsequent final diagnosis of headache/condition, through evaluation of electronic GP referral letters through SCI Gateway system of those cases where SVFT was recorded. Results Between March 2017 - November 2019, 669 scans through PBR pathway. (62% females; Mean age 53 years: 60% cases <60 years). SVFT was completed in only 381/669 (57%). Median SVFT was 17. Eleven of 381 cases had cancer (2.9%). 10 cases with cancer had SVFT <17 animals (median 10) (5.32%). One case had SVFT >=17 (35 animals) (0.5%) - CT scan showed small multiple intra-cerebral calcified and non-calcified lesions, consistent with metastases. 12% with PMH cancer had a tumour. Other possible reasons for low SVFT were: co-existing presumed dementia/mild cognitive impairment (19); non native English speakers (12); headache after traumatic brain injury (5); significant small vessel disease/vascular(5); intracranial cysts (4)(pineal / arachnoid, Giant Cell Arteritis (4) (all new - symptomatic); Chiari 1 malformations (2), PMH – encephalitis (1). Interestingly, there were 53 cases with known psychiatric/pain conditions on drugs (e.g. codeine/antidepressants/antipsychotics) with SVFT < 17 words/min. Conclusion People with Headache "Suspicious of Cancer" + SVFT <17 words in a minute are more likely to have a tumour (5.32% vs 0.5%) or other secondary cause for poor cognition. Other probable causes /associations, with SVFT <17 are age, poor English skills, co-existing dementia. SVFT score may be a useful adjunct or “red flag,” to consider, to expedite DACI scan in patients with “Headache Suspicious of Cancer”. A SVFT >=17 in those with Headache Suspicious of Cancer, does not exclude the possibility of an intracranial tumour. Excluding cases with recognised causes for low SVFT e.g. dementia and those with existing chronic pain/psychiatric disease further increases the likelihood of a secondary cause for headache. SVFT should be tested in the persons native language. A larger prospective study is required to establish whether these pilot study data and to examine whether chronic pain, functional neurology are negative predictive factors for secondary headache.