David Darby’s research while affiliated with Monash University (Australia) and other places

Background Cognitive impairment is one of the most common and debilitating symptoms of relapsing–remitting multiple sclerosis (RRMS). Digital cognitive biomarkers require less time and resources and are rapidly gaining popularity in clinical settings. We examined the longitudinal trajectory of the iPad‐based Processing Speed Test (PST) and predictors of PST scores. Methods We prospectively enrolled RRMS patients between 2017 and 2021 across six Australian MS centres. Longitudinal data was analysed with mixed effect modelling and latent class mixed models. We then examined whether latent class group membership predicted confirmed decrease in correct PST responses. Results We recruited a total of 1093 participants, of which 724 had complete baseline data with a median follow up duration of 2 years. At a population level, PST trajectory was stable. A small practice effect was present up to the 4th visit. Age, baseline disability, T2 lesion volume, male sex and depression were associated with lower correct PST responses, whilst years of education and full/part‐time employment were associated with more correct PST responses. We identified four latent class trajectories of PST. The worst latent class was typified by low baseline PST and lack of a practice effect. Being in the worst latent class was associated with a greater hazard of time to sustained 5% decrease in PST (HR 2.84, 95% CI 1.16–6.94, p = 0.02). Conclusion Worse baseline cognitive performance and lack of a practice effect predicted future cognitive decline in RRMS.

Background: Many dementia and Alzheimer's disease (AD) registries operate at local or national levels without standardization or comprehensive real-world data (RWD) collection. This initiative sought to achieve consensus among experts on priority outcomes and measures for clinical practice in caring for patients with symptomatic AD, particularly in the mild cognitive impairment and mild to moderate dementia stages. Objective: The primary aim was to define a minimum dataset (MDS) and extended dataset (EDS) to collect RWD in the new International Registry for AD and Other Dementias (InRAD) and other AD registries. The MDS and EDS focus on informing routine clinical practice, covering relevant comorbidities and safety, and are designed to be easily integrated into existing data capture systems. Methods and results: An international steering committee (ISC) of AD clinician experts lead the initiative. The first drafts of the MDS and EDS were developed based on a previous global inter-societal Delphi consensus on outcome measures for AD. Based on the ISC discussions, a survey was devised and sent to a wider stakeholder group. The ISC discussed the survey results, resulting in a consensus MDS and EDS covering: patient profile and demographics; lifestyle and anthropometrics; co-morbidities and diagnostics; imaging; treatment; clinical characterization; safety; discontinuation; laboratory tests; patient and care partner outcomes; and interface functionality. Conclusion: By learning from successful examples in other clinical areas, addressing current limitations, and proactively enhancing data quality and analytical rigor, the InRAD registry will be a foundation to contribute to improving patient care and outcomes in neurodegenerative diseases.

Survival analysis holds a crucial role across diverse disciplines, such as economics, engineering and healthcare. It empowers researchers to analyze both time-invariant and time-varying data, encompassing phenomena like customer churn, material degradation and various medical outcomes. Given the complexity and heterogeneity of such data, recent endeavors have demonstrated successful integration of deep learning methodologies to address limitations in conventional statistical approaches. However, current methods typically involve cluttered probability distribution function (PDF), have lower sensitivity in censoring prediction, only model static datasets, or only rely on recurrent neural networks for dynamic modelling. In this paper, we propose a novel survival regression method capable of producing high-quality unimodal PDFs without any prior distribution assumption, by optimizing novel Margin-Mean-Variance loss and leveraging the flexibility of Transformer to handle both temporal and non-temporal data, coined UniSurv. Extensive experiments on several datasets demonstrate that UniSurv places a significantly higher emphasis on censoring compared to other methods.

BACKGROUND: Many dementia and Alzheimers disease (AD) registries operate at local or national levels without standardization or comprehensive real-world data (RWD) collection. This initiative sought to achieve consensus among experts on priority outcomes and measures for clinical practice in caring for patients with symptomatic AD, particularly in the mild cognitive impairment and mild to moderate dementia stages. OBJECTIVE: The primary aim was to define a minimum dataset (MDS) and extended dataset (EDS) to collect RWD in the new International Registry for AD and Other Dementias (InRAD) and other AD registries. The MDS and EDS focus on informing routine clinical practice, covering relevant comorbidities and safety, and are designed to be easily integrated into existing data capture systems. METHODS AND RESULTS: An international steering committee (ISC) of AD clinician experts lead the initiative. The first drafts of the MDS and EDS were developed based on a previous global inter-societal Delphi consensus on outcome measures for AD. Based on the ISC discussions, a survey was devised and sent to a wider stakeholder group. The ISC discussed the survey results, resulting in a consensus MDS and EDS covering: patient profile and demographics; lifestyle and anthropometrics; co-morbidities and diagnostics; imaging; treatment; clinical characterization; safety; discontinuation; laboratory tests; patient and care partner outcomes; and interface functionality. CONCLUSION: By learning from successful examples in other clinical areas, addressing current limitations, and proactively enhancing data quality and analytical rigor, the InRAD registry will be a foundation to contribute to improving patient care and outcomes in neurodegenerative diseases.

Background Upper limb dysfunction is a common debilitating feature of relapsing‐remitting multiple sclerosis (RRMS). We aimed to examine the longitudinal trajectory of the iPad®‐based Manual Dexterity Test (MDT) and predictors of change over time. Methods We prospectively enrolled RRMS patients (limited to Expanded Disability Status Scale (EDSS) < 4). Longitudinal data was analysed using mixed‐effect modelling and latent class mixed models. We then examined whether group membership in latent classes predicted confirmed slowing in MDT. Results Seven hundred and twenty‐one participants had complete data for analysis. At a population level, MDT remained stable over time. No practice effect was seen. Baseline disability and T2 lesion volume were the strongest predictors of longitudinal MDT performance. We identified two latent class trajectories of MDT. The slower latent class was typified by greater variability and a weak association with confirmed worsening of MDT and EDSS. When compared to trajectory analysis stratified by baseline MDT, latent class analysis (LCA) was able to identify those at greater risk of confirmed slowing, signifying the importance of latent processes in upper limb function in pwMS. Conclusion In this cohort of mild to moderate RRMS, MDT scores remained stable over time with no evidence of a practice effect at a population level. Trajectory analysis based on LCA identified a cohort with greater variability and risk of disability progression and domain specific worsening. Our findings demonstrate the importance of latent processes in determining upper limb function in pwMS.

Background and Objectives Many neurodegenerative syndromes present with impairment of frontal networks, especially frontoinsular networks affecting social and emotional cognition. People presenting with frontal network impairments may be considered for a frontotemporal dementia (FTD) diagnosis. We sought to examine the diagnostic mix of patients referred with frontal network impairments to a single cognitive neurology service. Methods A retrospective review was conducted of all patients seen between January 2010 and December 2019 at the Eastern Cognitive Disorders Clinic, a quaternary cognitive neurology clinic in Melbourne, Australia. Patients were included if they met the following criteria: (1) were referred for suspected FTD or with a preexisting diagnosis of a FTD syndrome, (2) were referred for ‘frontal behaviors’ (i.e., disinhibition, disorganization, poor judgment, loss of empathy, apathy) and/or had an informant report of behavior change, and (3) had available referral documents and clinical consensus diagnosis. Referral diagnosis was compared against final diagnosis adjudicated by a consensus multidisciplinary team. Case details including age of symptom onset, Cambridge Behavioural Inventory-Revised scores, psychiatric history, and Charlson Comorbidity Index were compared against the final diagnosis. Results In total, 161 patients aged 42–82 years (mean = 64.5, SD = 9.0; 74.5% men) met inclusion criteria. The commonest final diagnosis was a FTD syndrome (44.6%: 26.7% behavioral variant FTD (bvFTD), 9.3% progressive supranuclear palsy, 6.2% semantic dementia, 1.2% corticobasal syndrome, and 1.2% FTD/motor neuron disease). A primary psychiatric disorder (PPD) was the next commonest diagnosis (15.5%), followed by vascular cognitive impairment (VCI, 10.6%), Alzheimer disease (AD, 9.9%), and other neurologic diagnoses (6.2%). A final diagnosis of bvFTD was associated with higher rates of medical comorbidities and more eating behavior abnormalities compared with a diagnosis of PPD. Screening cognitive tests and preexisting psychiatric history did not distinguish these 2 groups. Discussion A broad spectrum of neurologic and psychiatric disorders may present with impairments to frontal networks. Almost half of patients referred had a final FTD syndrome diagnosis, with bvFTD the commonest final diagnosis. People with PPD, VCI, and AD present with similar clinical profiles but are distinguishable using MRI and FDG-PET imaging. Medical and psychiatric comorbidities are common in people with bvFTD.

Survival analysis holds a crucial role across diverse disciplines, such as economics, engineering and healthcare. It empowers researchers to analyze both time-invariant and time-varying data, encompassing phenomena like customer churn, material degradation and various medical outcomes. Given the complexity and heterogeneity of such data, recent endeavors have demonstrated successful integration of deep learning methodologies to address limitations in conventional statistical approaches. However, current methods typically involve cluttered probability distribution function (PDF), have lower sensitivity in censoring prediction, only model static datasets, or only rely on recurrent neural networks for dynamic modelling. In this paper, we propose a novel survival regression method capable of producing high-quality unimodal PDFs without any prior distribution assumption, by optimizing novel Margin-Mean-Variance loss and leveraging the flexibility of Transformer to handle both temporal and non-temporal data, coined UniSurv. Extensive experiments on several datasets demonstrate that UniSurv places a significantly higher emphasis on censoring compared to other methods.

Background/Objectives Frontal variant Alzheimer’s Disease (fvAD) is relatively recently characterised condition which has been traditionally regarded as a young onset dementia. Classically the diagnosis relied primarily on clinical criteria. However, with the advent of new biomarkers and imaging technology, we expect to see a better reflection of the true prevalence of this condition in the community. Methods We conducted a 10-year retrospective audit of all patients presenting to an outpatient specialist cognitive neurology clinic from 2009 to 2019. Results 626 patients were seen over this period with 334 receiving a diagnosis of dementia. In the 65 and over age group, Alzheimer’s Dementia (AD) was the predominant diagnosis accounting for 59% of all dementia. Under 65 years, AD and behavioural variant Frontotemporal Dementia (bvFTD) patients represented 40% and 34% of the diagnoses, respectively. A total of ten patients were diagnosed with fvAD, seven of whom were aged over 65 years. Conclusion Frontal variant AD was an uncommon diagnosis in our study making up 3% of the total dementia diagnosis over 10 years. The majority of those were in the late onset dementia group, meaning that dysexecutive profiles in older people with dementia are still most likely due to AD pathology. We note that the use of imaging techniques including FDG-PET, and new plasma biomarkers, may have assisted in better identifying a relatively recently characterised condition. Our findings would suggest that fvAD is less common in those under 65, with most patients in that group being diagnosed with bvFTD.

Background Cognitive impairment is one of the most common and debilitating symptoms of relapsing remitting multiple sclerosis (RRMS). Digital cognitive biomarkers require less time and resources and are rapidly gaining popularity in clinical settings. We examined the longitudinal trajectory of the iPad-based Processing Speed Test (PST) and predictors of change over time. Methods We prospectively enrolled relapsing-remitting multiple sclerosis (RRMS) patients with an EDSS score of less than four. Longitudinal data was analysed with mixed effect modelling and latent class mixed models. Results At a population level, PST trajectory was stable. A small practice effect was present up to the 4th visit. Age, baseline disability, T2 lesion volume, male gender and depression were associated with less correct PST responses, whilst years of education and full/part-time employment were associated with more correct PST responses.We identified four trajectories of processing speed with latent class analysis. The lowest latent class was typified by the lack of a practice effect and was associated with a greater hazard of time to sustained 5% decrease in PST (HR 2.84, 95%CI 1.16–6.94, p=0.02). Conclusion In this cohort of mild to moderate RRMS, PST scores remained largely stable over time. Membership in the worst latent class trajectory was associated with a sustained 5% PST decrease. Poor cognitive performance at baseline and the lack of a practice effect is a predictor of future cognitive decline and should prompt early intervention for maximising cognitive function such as treatment escalation.