Characteristics associated with response to subcutaneously administered anti-CGRP monoclonal antibody medications in a real-world community cohort of persons living with migraine: A retrospective clinical and genetic study.

OBJECTIVE: To evaluate response to anti-calcitonin gene-related peptide (CGRP) migraine preventives in a real-world community cohort of persons living with migraine and to identify clinical and genetic characteristics associated with efficacious response. BACKGROUND: Erenumab-aooeb, fremanezumab-vrfm, and galcanezumab-gnlm target CGRP or its receptor; however, many patients are non-responsive. METHODS: In this retrospective clinical and genetic study, we identified 1077 adult patients who satisfied the International Classification of Headache Disorders, 3rd edition, criteria for migraine without aura, migraine with aura, or chronic migraine and who were prescribed an anti-CGRP migraine preventive between May 2018 and May 2021. Screening of 558 patients identified 289 with data at baseline and first follow-up visits; data were available for 161 patients at a second follow-up visit. The primary outcome was migraine days per month (MDM). In 198 genotyped patients, we evaluated associations between responders (i.e., patients with ≥50% reduction in MDM at follow-up) and genes involved in CGRP signaling or pharmacological response, and genetic and polygenic risk scores. RESULTS: The median time to first follow-up was 4.4 (0.9-22) months after preventive start. At the second follow-up, 5.7 (0.9-13) months later, 145 patients had continued on the same preventive. Preventives had strong, persistent effects in reducing MDM in responders (follow-up 1: η2 = 0.26, follow-up 2: η2 = 0.22). At the first but not second follow-up: galcanezumab had a larger effect than erenumab, while no difference was seen at either follow-up between galcanezumab and fremanezumab or fremanezumab and erenumab. The decrease in MDM at follow-up was generally proportional to baseline MDM, larger in females, and increased with months on medication. At the first follow-up only, patients with prior hospitalization for migraine or who had not responded to more preventive regimens had a smaller decrease in MDM. Reasons for stopping or switching a preventive varied between medications and were often related to cost and insurance coverage. At both follow-ups, patient tolerance (1: 92.2% [262/284]; 2: 95.2% [141/145]) and continued use (1: 77.5% [224/289]; 2: 80.6% [116/145]) were high and similar across preventives. Response consistency (always non-responders: 31.7% [46/145]; always responders: 56.5% [82/145], and one-time only responders: 11.7% [17/145]) was also similar across preventives. Non-responder status had nominally significant associations with rs12615320-G in RAMP1 (odds ratio [95% confidence interval]: 4.7 [1.5, 14.7]), and rs4680-A in COMT (0.6[0.4, 0.9]). Non-responders had a lower mean genetic risk score than responders (1.0 vs. 1.1; t(df) = -1.75(174.84), p = 0.041), and the fraction of responders increased with genetic and polygenic risk score percentile. CONCLUSIONS: In this real-world setting, anti-CGRP preventives reduced MDM persistently and had similar and large effect sizes on MDM reduction; however, clinical and genetic factors influenced response.

Early prediction of Alzheimer's disease and related dementias using real-world electronic health records.

INTRODUCTION: This study aims to explore machine learning (ML) methods for early prediction of Alzheimer's disease (AD) and related dementias (ADRD) using the real-world electronic health records (EHRs). METHODS: A total of 23,835 ADRD and 1,038,643 control patients were identified from the OneFlorida+ Research Consortium. Two ML methods were used to develop the prediction models. Both knowledge-driven and data-driven approaches were explored. Four computable phenotyping algorithms were tested. RESULTS: The gradient boosting tree (GBT) models trained with the data-driven approach achieved the best area under the curve (AUC) scores of 0.939, 0.906, 0.884, and 0.854 for early prediction of ADRD 0, 1, 3, or 5 years before diagnosis, respectively. A number of important clinical and sociodemographic factors were identified. DISCUSSION: We tested various settings and showed the predictive ability of using ML approaches for early prediction of ADRD with EHRs. The models can help identify high-risk individuals for early informed preventive or prognostic clinical decisions.

Optimizing the electronic medical record to improve patient care and conduct quality improvement initiatives in a concussion specialty clinic.

Objective: To use the electronic medical record (EMR) to optimize patient care, facilitate documentation, and support quality improvement and practice-based research in a concussion (mild traumatic brain injury; mTBI) clinic.Methods: We built a customized structured clinical documentation support (SCDS) toolkit for patients in a concussion specialty clinic. The toolkit collected hundreds of fields of discrete, standardized data. Autoscored and interpreted score tests include the Generalized Anxiety Disorder 7-item scale, Center for Epidemiology Studies Depression scale, Insomnia Severity Index, and Glasgow Coma Scale. Additionally, quantitative score measures are related to immediate memory, concentration, and delayed recall. All of this data collection occurred in a standard appointment length.Results: To date, we evaluated 619 patients at an initial office visit after an mTBI. We provided a description of our toolkit development process, and a summary of the data electronically captured using the toolkit.Conclusions: The electronic medical record can be used to effectively structure and standardize care in a concussion clinic. The toolkit supports the delivery of care consistent with Best Practices, provides opportunities for point of care decision support, and writes comprehensive progress notes that can be communicated to other providers.

Sensitivity and Specificity of Diagnostic Criteria for Progressive Supranuclear Palsy.

BACKGROUND: In 2017, the International Parkinson and Movement Disorder Society put forward new clinical criteria for the diagnosis of PSP, recognizing diverse PSP phenotypes. In this study, we compared the sensitivity and specificity of the new criteria with the National Institutes of Neurological Disease and Society for Progressive Supranuclear Palsy criteria at different times. METHODS: Patients with clinical parkinsonism, clinical and/or neuropathological diagnosis of PSP, were identified from the Society for Progressive Supranuclear Palsy brain bank. All patients had neuropathologic diagnoses and detailed clinical examination performed by a neurologist at 1 of the 3 Mayo Clinic sites, in Florida, Arizona, and Minnesota. Clinical symptoms and signs were abstracted retrospectively in a blinded fashion and used to determine whether patients met either diagnostic criterion. Patients were divided into early and late disease stage groups using a 3-year cutoff. RESULTS: A total of 129 patients were included, of whom 66 had PSP pathology (51%). The remainder had other neurodegenerative diseases. The overall sensitivity of the International Parkinson and Movement Disorder Society criteria was 87.9%, compared with 45.5% for the National Institutes of Neurological Disease and Society for Progressive Supranuclear Palsy criteria, whereas the specificity of the International Parkinson and Movement Disorder Society probable PSP criteria was 85.7%, compared with 90.5% for the National Institutes of Neurological Disease and Society for Progressive Supranuclear Palsy. Individual patients were noted to have features of multiple PSP phenotypes. CONCLUSION: The International Parkinson and Movement Disorder Society criteria recognize several phenotypes of progressive supranuclear palsy and hence have higher sensitivity than the previous criteria. © 2019 International Parkinson and Movement Disorder Society.

Design and Implementation of Structured Clinical Documentation Support Tools for Treating Stroke Patients.

BACKGROUND AND PURPOSE: Standardized electronic medical record tools provide an opportunity to efficiently provide care that conforms to Best Practices and supports quality improvement and practice-based research initiatives. METHODS: We describe the development of a customized structured clinical documentation "toolkit" that standardizes patient data collection to conform to Best Practices for treating patients with stroke. The toolkit collects patients' demographic information, relevant score test measures, and captures information on disability, treatment, and outcomes. RESULTS: We describe here our creation and implementation of the toolkits and provide example screenshots. As of August 1, 2018, we have evaluated 2332 patients at an initial visit for a possible stroke. We provide basic descriptive data gathered from the use of the toolkits, demonstrating their utility in collecting patient data in a manner that supports both quality clinical care and research initiatives. CONCLUSIONS: We have developed an EMR toolkit to support Best Practices in the care of patients with stroke. We discuss quality improvement projects and current research initiatives using the toolkit. This toolkit is being shared with other Departments of Neurology as part of the Neurology Practice-Based Research Network.

Structured Clinical Documentation to Improve Quality and Support Practice-Based Research in Headache.

OBJECTIVE: To use the electronic medical record (EMR) to optimize patient care, facilitate documentation, and support quality improvement and practice-based research, in a headache specialty clinic. BACKGROUND: Many physicians enter data into the EMR as unstructured free text and not as discrete data. This makes it challenging to use data for quality improvement or research initiatives. METHODS: We describe the process of building a customized structured clinical documentation support toolkit, specific for patients seen in a headache specialty clinic. The content was developed through frequent physician meetings to reach consensus on elements that define clinical Best Practices. Tasks were assigned to the care team and data mapped to the progress note. RESULTS: The toolkit collects hundreds of fields of discrete, standardized data. Auto scored and interpreted score tests include the Generalized Anxiety Disorder 7-item, Center for Epidemiology Studies Depression Scale, Migraine Disability Assessment questionnaire, Insomnia Sleep Index, and Migraine-Specific Quality of Life. We have developed Best Practice Advisories (BPA) and other clinical documentation support tools that alert physicians, when appropriate. As of April 1, 2018, we have used the toolkits at 4346 initial patient visits. We provide screenshots of our toolkits, details of data fields collected, and diagnoses of patients at the initial visit. CONCLUSIONS: The EMR can be used to effectively structure and standardize headache clinic visits for quality improvement and practice-based research. We are sharing our proprietary toolkit with other clinics as part of the Neurology Practice-Based Research Network. These tools are also facilitating clinical research enrollment and a pragmatic trial of comparative effectiveness at the point-of-care among migraine patients.

Investigating ioflupane I123 injection and single photon emission tomography as an imaging biomarker for long-term sequelae following mild traumatic brain injury.

OBJECTIVE: To determine whether there were differences in clinical outcomes for ioflupane I123 injection (DaTscan) and single photon emission tomography consistent with early Parkinson's disease (PD) among individuals with a history of mild traumatic brain injury (mTBI). METHODS: We performed a case-control study among patients presenting to the Emergency Room (ER) during 2006-2013 with mTBI (cases, n = 34) or without mTBI (controls, n = 33). We performed clinical and imaging measurements in cases and controls at least 1-year post-presentation to the ER (average three years four months). RESULTS: All DaTscans obtained were qualitatively normal. There were no qualitative DaTscan differences between cases and controls. There was, however, a significant increase in caudate asymmetry in controls versus cases (p = 0.02), but this finding was no longer significant after correction for multiple comparisons. There was a suggestion of a trend of poorer clinical score test measures among those with mTBI, although the overall mean score difference between cases and controls was not clinically significant. CONCLUSION: Our small study does not provide support for DaTscan changes suggestive of PD in the one to seven years following mTBI. A trend towards poorer clinical measures was seen but was not clinically relevant in our small sample. Further work in a large population is necessary to support these findings.

Structured clinical documentation in the electronic medical record to improve quality and to support practice-based research in epilepsy.

OBJECTIVE: Using the electronic medical record (EMR) to capture structured clinical data at the point of care would be a practical way to support quality improvement and practice-based research in epilepsy. METHODS: We describe our stepwise process for building structured clinical documentation support tools in the EMR that define best practices in epilepsy, and we describe how we incorporated these toolkits into our clinical workflow. RESULTS: These tools write notes and capture hundreds of fields of data including several score tests: Generalized Anxiety Disorder-7 items, Neurological Disorders Depression Inventory for Epilepsy, Epworth Sleepiness Scale, Quality of Life in Epilepsy-10 items, Montreal Cognitive Assessment/Short Test of Mental Status, and Medical Research Council Prognostic Index. The tools summarize brain imaging, blood laboratory, and electroencephalography results, and document neuromodulation treatments. The tools provide Best Practices Advisories and other clinical decision support when appropriate. The tools prompt enrollment in a DNA biobanking study. We have thus far enrolled 231 patients for initial visits and are starting our first annual follow-up visits and provide a brief description of our cohort. SIGNIFICANCE: We are sharing these EMR tools and captured data with other epilepsy clinics as part of a Neurology Practice Based Research Network, and are using the tools to conduct pragmatic trials using subgroup-based adaptive designs.

MAPT haplotype H1G is associated with increased risk of dementia with Lewy bodies.

INTRODUCTION: The MAPT H1 haplotype has been associated with several neurodegenerative diseases. We were interested in exploring the role of MAPT haplotypic variation in risk of dementia with Lewy bodies (DLB). METHOD: We genotyped six MAPT haplotype tagging SNPs and screened 431 clinical DLB cases, 347 pathologically defined high-likelihood DLB cases, and 1049 controls. RESULT: We performed haplotypic association tests and detected an association with the protective H2 haplotype in our combined series (odds ratio [OR] = 0.75). We fine-mapped the locus and identified a relatively rare haplotype, H1G, that is associated with an increased risk of DLB (OR = 3.30, P = .0017). This association was replicated in our pathologically defined series (OR = 2.26, P = .035). DISCUSSION: These results support a role for H1 and specifically H1G in susceptibility to DLB. However, the exact functional variant at the locus is still unknown, and additional studies are warranted to fully explain genetic risk of DLB at the MAPT locus.

Comprehensive research synopsis and systematic meta-analyses in Parkinson's disease genetics: The PDGene database.

More than 800 published genetic association studies have implicated dozens of potential risk loci in Parkinson's disease (PD). To facilitate the interpretation of these findings, we have created a dedicated online resource, PDGene, that comprehensively collects and meta-analyzes all published studies in the field. A systematic literature screen of -27,000 articles yielded 828 eligible articles from which relevant data were extracted. In addition, individual-level data from three publicly available genome-wide association studies (GWAS) were obtained and subjected to genotype imputation and analysis. Overall, we performed meta-analyses on more than seven million polymorphisms originating either from GWAS datasets and/or from smaller scale PD association studies. Meta-analyses on 147 SNPs were supplemented by unpublished GWAS data from up to 16,452 PD cases and 48,810 controls. Eleven loci showed genome-wide significant (P < 5 × 10(-8)) association with disease risk: BST1, CCDC62/HIP1R, DGKQ/GAK, GBA, LRRK2, MAPT, MCCC1/LAMP3, PARK16, SNCA, STK39, and SYT11/RAB25. In addition, we identified novel evidence for genome-wide significant association with a polymorphism in ITGA8 (rs7077361, OR 0.88, P  =  1.3 × 10(-8)). All meta-analysis results are freely available on a dedicated online database (www.pdgene.org), which is cross-linked with a customized track on the UCSC Genome Browser. Our study provides an exhaustive and up-to-date summary of the status of PD genetics research that can be readily scaled to include the results of future large-scale genetics projects, including next-generation sequencing studies.

A road map for efficient and reliable human genome epidemiology.

Networks of investigators have begun sharing best practices, tools and methods for analysis of associations between genetic variation and common diseases. A Network of Investigator Networks has been set up to drive the process, sponsored by the Human Genome Epidemiology Network. A workshop is planned to develop consensus guidelines for reporting results of genetic association studies. Published literature databases will be integrated, and unpublished data, including 'negative' studies, will be captured by online journals and through investigator networks. Systematic reviews will be expanded to include more meta-analyses of individual-level data and prospective meta-analyses. Field synopses will offer regularly updated overviews.

Translation initiator EIF4G1 mutations in familial Parkinson disease.

Genome-wide analysis of a multi-incident family with autosomal-dominant parkinsonism has implicated a locus on chromosomal region 3q26-q28. Linkage and disease segregation is explained by a missense mutation c.3614G>A (p.Arg1205His) in eukaryotic translation initiation factor 4-gamma (EIF4G1). Subsequent sequence and genotype analysis identified EIF4G1 c.1505C>T (p.Ala502Val), c.2056G>T (p.Gly686Cys), c.3490A>C (p.Ser1164Arg), c.3589C>T (p.Arg1197Trp) and c.3614G>A (p.Arg1205His) substitutions in affected subjects with familial parkinsonism and idiopathic Lewy body disease but not in control subjects. Despite different countries of origin, persons with EIF4G1 c.1505C>T (p.Ala502Val) or c.3614G>A (p.Arg1205His) mutations appear to share haplotypes consistent with ancestral founders. eIF4G1 p.Ala502Val and p.Arg1205His disrupt eIF4E or eIF3e binding, although the wild-type protein does not, and render mutant cells more vulnerable to reactive oxidative species. EIF4G1 mutations implicate mRNA translation initiation in familial parkinsonism and highlight a convergent pathway for monogenic, toxin and perhaps virally-induced Parkinson disease.

Alpha-synuclein repeat variants and survival in Parkinson's disease.

OBJECTIVES: To determine whether α-synuclein dinucleotide repeat (REP1) genotypes are associated with survival in Parkinson's disease (PD). METHODS: Investigators from the Genetic Epidemiology of Parkinson's Disease Consortium provided REP1 genotypes and baseline and follow-up clinical data for cases. The primary outcome was time to death. Cox proportional hazards regression models were used to assess the association of REP1 genotypes with survival. RESULTS: Twenty-one sites contributed data for 6,154 cases. There was no significant association between α-synuclein REP1 genotypes and survival in PD. However, there was a significant association between REP1 genotypes and age at onset of PD (hazard ratio: 1.06; 95% confidence interval: 1.01-1.10; P value = 0.01). CONCLUSIONS: In our large consortium study, α-synuclein REP1 genotypes were not associated with survival in PD. Further studies of α-synuclein's role in disease progression and long-term outcomes are needed.

TICS-M scores in an oldest-old normative cohort identified by computable phenotype.

Objective: To (1) examine the distribution of Telephone Interview for Cognitive Status modified (TICS-m) scores in oldest-old individuals (age 85 and above) identified as cognitively healthy by a previously validated electronic health records-based computable phenotype (CP) and (2) to compare different cutoff scores for cognitive impairment in this population. Method: CP identified 24,024 persons, 470 were contacted and 252 consented and completed the assessment. Associations of TICS-m score with age, sex, and educational categories (<10 years, 11-15 years, and >16 years) were examined. The number of participants perceived as impaired was studied with commonly used cutoff scores (27-31). Results: TICS-m score ranged from 18 to 44 with a mean of 32.6 (SD = 4.7) in older adults aged 85-99 years old. A linear regression model including (range-restricted) age, education, and sex, showed beta estimates comparable to previous findings. Different cutoff scores (27 to 31) generated slightly lower MCI and dementia prevalence rates of participants meeting the criteria for the impairments than studies of younger elderly using traditional recruitment methods. Conclusions: The use of validated computable phenotype to identify a normative cohort generated a normative distribution for the TICS-m consistent with prior findings from more effortful approaches to cohort identification and established expected TICS-m performance in the oldest-old population.

Lipid Trajectories Improve Risk Models for Alzheimer's Disease and Mild Cognitive Impairment.

To assess the relationship between lipids and cognitive dysfunction, we retrospectively analyzed blood-lipid levels in clinically well-characterized individuals with stable mild cognitive impairment (MCI) or Alzheimer's disease (AD) over the decade prior to first cognitive symptoms. In this case/control cohort study, AD and MCI cases were diagnosed using DSM-IV criteria; MCI cases had not progressed to dementia for ≥5 years; and controls were propensity matched to cases at age of symptom onset (MCI: 116 cases, 435 controls; AD: 215 cases, 483 controls). Participants were grouped based on longitudinal trajectories and quintile of variability independent of the mean (VIM) for total cholesterol, HDL-C, LDL-C, non-HDL-C and ln(triglycerides). Models for the risk of cognitive dysfunction evaluated trajectory and VIM groups, APOE genotype, polygenic risk scores (PRS) for AD and lipid levels, age, comorbidities, and longitudinal correlates of blood-lipid concentrations. Lower HDL-C trajectories (OR = 3.8, 95% CI = 1.3-11.3) and the lowest VIM quintile of non-HDL-C (OR = 2.2, 95% CI = 1.3-3.0) were associated with higher MCI risk. Lower HDL-C trajectories (OR = 3.0, 95% CI = 1.6-5.7) and the lowest VIM quintile of total cholesterol (OR = 2.4, 95% CI = 1.5-3.9) were associated with higher AD risk. The inclusion of lipid-trajectory and VIM groups improved risk-model predictive performance independent of APOE genotype or PRS for AD and lipid levels. These results provide an important real-world perspective on the influence of lipid metabolism and blood-lipid levels on the development of stable MCI and AD.

Mutations in CIZ1 cause adult onset primary cervical dystonia.

OBJECTIVE: Primary dystonia is usually of adult onset, can be familial, and frequently involves the cervical musculature. Our goal was to identify the causal mutation in a family with adult onset, primary cervical dystonia. METHODS: Linkage and haplotype analyses were combined with solution-based whole-exome capture and massively parallel sequencing in a large Caucasian pedigree with adult onset, primary cervical dystonia to identify a cosegregating mutation. High-throughput screening and Sanger sequencing were completed in 308 Caucasians with familial or sporadic adult onset cervical dystonia and matching controls for sequence variants in this mutant gene. RESULTS: Exome sequencing led to the identification of an exonic splicing enhancer mutation in exon 7 of CIZ1 (c.790A>G, p.S264G), which encodes CIZ1, Cip1-interacting zinc finger protein 1. CIZ1 is a p21(Cip1/Waf1) -interacting zinc finger protein expressed in brain and involved in DNA synthesis and cell-cycle control. Using a minigene assay, we showed that c.790A>G altered CIZ1 splicing patterns. The p.S264G mutation also altered the nuclear localization of CIZ1. Screening in subjects with adult-onset cervical dystonia identified 2 additional CIZ1 missense mutations (p.P47S and p.R672M). INTERPRETATION: Mutations in CIZ1 may cause adult onset, primary cervical dystonia, possibly by precipitating neurodevelopmental abnormalities that manifest in adults and/or G1/S cell-cycle dysregulation in the mature central nervous system.

Population-specific frequencies for LRRK2 susceptibility variants in the Genetic Epidemiology of Parkinson's Disease (GEO-PD) Consortium.

BACKGROUND: Variants within the leucine-rich repeat kinase 2 gene are recognized as the most frequent genetic cause of Parkinson's disease. Leucine-rich repeat kinase 2 variation related to disease susceptibility displays many features that reflect the nature of complex, late-onset sporadic disorders like Parkinson's disease. METHODS: The Genetic Epidemiology of Parkinson's Disease Consortium recently performed the largest genetic association study for variants in the leucine-rich repeat kinase 2 gene across 23 different sites in 15 countries. RESULTS: Herein, we detail the allele frequencies for the novel risk factors (p.A419V and p.M1646T) and the protective haplotype (p.N551K-R1398H-K1423K) nominated in the original publication. Simple population allele frequencies not only can provide insight into the clinical relevance of specific variants but also can help genetically define patient groups. CONCLUSIONS: Establishing individual patient-based genomic susceptibility profiles that incorporate both risk factors and protective factors will determine future diagnostic and treatment strategies.

Evaluation of a Computable Phenotype for Successful Cognitive Aging.

Objective: To establish, apply, and evaluate a computable phenotype for the recruitment of individuals with successful cognitive aging. Participants and Methods: Interviews with 10 aging experts identified electronic health record (EHR)-available variables representing successful aging among individuals aged 85 years and older. On the basis of the identified variables, we developed a rule-based computable phenotype algorithm composed of 17 eligibility criteria. Starting September 1, 2019, we applied the computable phenotype algorithm to all living persons aged 85 years and older at the University of Florida Health, which identified 24,024 individuals. This sample was comprised of 13,841 (58%) women, 13,906 (58%) Whites, and 16,557 (69%) non-Hispanics. A priori permission to be contacted for research had been obtained for 11,898 individuals, of whom 470 responded to study announcements and 333 consented to evaluation. Then, we contacted those who consented to evaluate whether their cognitive and functional status clinically met out successful cognitive aging criteria of a modified Telephone Interview for Cognitive Status score of more than 27 and Geriatric Depression Scale of less than 6. The study was completed on December 31, 2022. Results: Of the 45% of living persons aged 85 years and older included in the University of Florida Health EHR database identified by the computable phenotype as successfully aged, approximately 4% of these responded to study announcements and 333 consented, of which 218 (65%) met successful cognitive aging criteria through direct evaluation. Conclusion: The study evaluated a computable phenotype algorithm for the recruitment of individuals for a successful aging study using large-scale EHRs. Our study provides proof of concept of using big data and informatics as aids for the recruitment of individuals for prospective cohort studies.

Assessment of ChatGPT's performance on neurology written board examination questions.

Background and objectives: ChatGPT has shown promise in healthcare. To assess the utility of this novel tool in healthcare education, we evaluated ChatGPT's performance in answering neurology board exam questions. Methods: Neurology board-style examination questions were accessed from BoardVitals, a commercial neurology question bank. ChatGPT was provided a full question prompt and multiple answer choices. First attempts and additional attempts up to three tries were given to ChatGPT to select the correct answer. A total of 560 questions (14 blocks of 40 questions) were used, although any image-based questions were disregarded due to ChatGPT's inability to process visual input. The artificial intelligence (AI) answers were then compared with human user data provided by the question bank to gauge its performance. Results: Out of 509 eligible questions over 14 question blocks, ChatGPT correctly answered 335 questions (65.8%) on the first attempt/iteration and 383 (75.3%) over three attempts/iterations, scoring at approximately the 26th and 50th percentiles, respectively. The highest performing subjects were pain (100%), epilepsy & seizures (85%) and genetic (82%) while the lowest performing subjects were imaging/diagnostic studies (27%), critical care (41%) and cranial nerves (48%). Discussion: This study found that ChatGPT performed similarly to its human counterparts. The accuracy of the AI increased with multiple attempts and performance fell within the expected range of neurology resident learners. This study demonstrates ChatGPT's potential in processing specialised medical information. Future studies would better define the scope to which AI would be able to integrate into medical decision making.