Emerging Subspecialties: Pediatric Movement Disorders Neurology.

Pediatric movement disorders (PMD) neurologists care for infants, children, and adolescents with conditions that disrupt typical movement; serving as important subspecialist child neurologists in both academic and private practice settings. In contrast to adult movement disorders neurologists whose "bread and butter" is hypokinetic Parkinson disease, PMD subspecialty practice is often dominated by hyperkinetic movement disorders including tics, dystonia, chorea, tremor, and myoclonus. PMD neurology practice intersects with a variety of subspecialties, including neonatology, developmental pediatrics, rehabilitation medicine, epilepsy, child & adolescent psychiatry, psychology, orthopedics, genetics & metabolism, and neurosurgery. Over the past several decades, significant advancements in the PMD field have included operationalizing definitions for distinct movement disorders, recognizing the spectrum of clinical phenotypes, expanding research on genetic and neuroimmunologic causes of movement disorders, and advancing available treatments. Subspecialty training in PMD provides trainees with advanced clinical, diagnostic, procedural, and management skills that reflect the complexities of contemporary practice. The child neurologist who is fascinated by the intricacies of child motor development, appreciates the power of observation skills coupled with a thoughtful physical examination, and is excited by the challenge of the unknown may be well-suited to a career as a PMD specialist.

Should we be concerned? A qualitative study of educators' perceptions of medical student wellbeing in domestic violence training.

INTRODUCTION: Domestic violence (DV) is common in the Australian community so it is likely that there will be medical students who are affected personally by DV. Some of these students may find DV training confronting or even re-traumatising. A trauma-informed medical education (TIME) framework utilising trauma-informed care principles may minimise this risk to students. We aimed to explore educators' perceptions of student well-being in Australian medical school DV training. METHOD: This descriptive qualitative study interviewed 13 educators with experience teaching DV in Australian medical schools using an interpretivist methodology and a TIME framework. Interview data was thematically analysed to identify themes. RESULTS: Four key themes included (1) educators thrown in at the deep end; (2) keeping students emotionally safe; (3) a trauma-informed learning environment and; (4) challenges of student DV disclosures. Few of the participants had received training in DV. Educators used methods such as trigger warnings and ground rules to improve student's emotional safety. Experienced educators dealt with disclosures of DV by students which led to role confusion. DISCUSSION: There is a need for increased training of medical educators that includes awareness and implementation of TIME principles when training medical students in DV as well as increased supports and resources for educators.

Four decades of regional wet deposition, local bulk deposition, and stream-water chemistry show the influence of nearby land use on forested streams in Central Appalachia.

Hydrologic monitoring began on two headwater streams (<1 km2) on the University of Kentucky's Robinson Forest in 1971. We evaluated stream-water (1974-2013) and bulk-deposition (wet + dust) (1984-2013) chemistry in the context of regional wet-deposition patterns that showed decreases in both sulfate and nitrate concentrations as well as proximal surface-mine expansion. Decadal time steps (1974-83, 1984-93, 1994-2003, 2004-2013) were used to quantify change. Comparison of the first two decades showed similarly decreased sulfate (minimum flow-adjusted annual-mean concentration of ≈13.5 mg/L in 1982 to 8.8 mg/L in 1992) and increased pH (6.6-6.8) in both streams, reflecting contemporaneous changes in both bulk and wet deposition. In contrast, concentrations of nitrate (0.14 to >0.25 mg/L) and base cations increased between these two decades, coinciding with expansion of surface mining between 1985 and 1995. In 2004, stream-water pH (6.7 in 2004), sulfate (9.2 mg/L), and nitrate (>0.11 mg/L) were similar to 1982, despite wet-deposition concentrations being lower. Base-cation concentrations were higher in the stream adjacent to ongoing surface mining relative to the stream situated near the middle of the experimental forest. However, pH decreased to approximately 5.7 by 2013 for both streams, which, combined with a shift in dominant cations from calcium to magnesium and potassium, indicates that the soil-buffering capacity of this landscape has been exceeded. Ratios of bulk deposition and stream-water concentrations indicate enrichment of sulfate (1.7-25.2) and cations (0.5-64.8), but not nitrogen (0.1-5.6), indicating that the Forest is not nitrogen saturated and that ongoing changes in water-quality are sulfate driven. When concentrations were adjusted to account for changes in streamflow (climate) over the 4 decades, external influences (land management/regulation) explained most change. The amount and direction of change differed among constituents, both between consecutive decades and between the first and last decades, reflecting the influence of localized surface mining even as regional wet deposition continued to improve due to the Clean Air Act. The implication is that localized stressors have the potential to out-pace the benefits of national environmental policies for communities that depend on local water-resources in similar environments.

Deep Brain Stimulation for Pediatric Dystonia: A Review of the Literature and Suggested Programming Algorithm.

Deep brain stimulation (DBS) is an established intervention for use in pediatric movement disorders, especially dystonia. Although multiple publications have provided guidelines for deep brain stimulation patient selection and programming in adults, there are no evidence-based or consensus statements published for pediatrics. The result is lack of standardized care and underutilization of this effective treatment. To this end, we assembled a focus group of 13 pediatric movement disorder specialists and 1 neurosurgeon experienced in pediatric deep brain stimulation to review recent literature and current practices and propose a standardized approach to candidate selection, implantation target site selection, and programming algorithms. For pediatric dystonia, we provide algorithms for (1) programming for initial session and follow-up sessions, and (2) troubleshooting side effects encountered during programming. We discuss common side effects, how they present, and recommendations for management. This topical review serves as a resource for movement disorders specialists interested in using deep brain stimulation for pediatric dystonia.

Predictive value of video alone in diagnosis of epileptic vs paroxysmal nonepileptic events in children.

OBJECTIVE: Previous studies examined the use of video-based diagnosis and the predictive value of videos for differentiation of epileptic seizures (ES) from paroxysmal nonepileptic events (PNEE) in the adult population. However, there are no such published studies strictly on the pediatric population. Using video-EEG diagnosis as a gold standard, we aimed to determine the diagnostic predictive value of videos of habitual events with or without additional clinical data in differentiating the PNEE from ES in children. METHODS: Consecutive admissions to our epilepsy monitoring unit between June 2020 and December 2020 were analyzed for events of interest. Four child neurologists blinded to the patient's diagnosis formulated a diagnostic impression based upon the review of the video alone and again after having access to basic clinical information, in addition to the video. Features of the video which helped to make a diagnosis were identified by the reviewers as a part of a survey. RESULTS: A total of 54 patients were included (ES n = 24, PNEE n = 30). Diagnostic accuracy was calculated for each reviewer and combined across all the ratings. Diagnostic accuracy by video alone was 74.5% (sensitivity 80.8%, specificity 66.7%). Providing reviewers with basic clinical information in addition to the videos significantly improved diagnostic accuracy compared to viewing the videos alone. Inter-rater reliability between four reviewers based on the video alone showed moderate agreement (κ = 0.51) and unchanged when additional clinical data were presented (κ = 0.51). The ES group was significantly more likely to demonstrate changes in facial expression, generalized stiffening, repetitive eye blinks, and eye deviation when compared with the PNEE group, which was more likely to display bilateral myoclonic jerking. CONCLUSIONS: Video review of habitual events by Child Neurologists may be helpful in reliably distinguishing ES from PNEE in children, even without included clinical information.

Building a practice-based research network for healthcare integration: a protocol paper for a mixed-method project.

INTRODUCTION: Practice-based research networks (PBRNs) are sustained collaborations between healthcare professionals, researchers and members of the community that develop, conduct and report on research relevant to local needs. While PBRNs have traditionally been focused towards primary care practices and their patients, there has been increasing interest in how they may help facilitate healthcare integration. Yet, little is known on the ways in which PBRNs can best integrate with the broader healthcare system, in particular Advanced Health Research and Translation Centres. The overall project aim is to build a sustainable collaboration between a PBRN and an Advanced Health Research and Translation Centre to generate a research platform suitable for planning, undertaking and translating research to improve care across the healthcare continuum. METHODS AND ANALYSIS: We will use a developmental evaluation design. Our iterative approach will be informed by a programme logic model and consists of: preparation work (pre-implementation assessment, literature review, community and stakeholder engagement), adaptation and building for a sustainable collaboration (strategy for recruitment and sustainment of members) and planning for network action (designing and implementing priority initiatives, monitoring and follow-up). ETHICS AND DISSEMINATION: This project was approved by the Monash Health ethics committee (ERM Reference Number: 76281; Monash Health Ref: RES-21-0000-392L) and the Monash University Human Research ethics committee (Reference Number: 29786). Dissemination will take place via various channels, including relevant national and international committees and conferences, peer-reviewed journals and social media. Continuous dissemination to and communication with all participants in this project as well as other relevant stakeholders will help strengthen and sustain the network.

Advance care planning in dementia: a qualitative study of Australian general practitioners.

As the population ages, so does the prevalence of dementia, a condition in which timely advance care planning (ACP) is recommended. This study explored the barriers and enablers of ACP for Australian GPs caring for people with dementia. Semi-structured qualitative telephone interviews informed by the theoretical domains framework (TDF) were conducted with 16 GPs from the south-east region of metropolitan Melbourne. The most prevalent domains of the TDF were: (1) environmental context and resources; (2) beliefs about consequences; and (3) social/professional role and identity. Further thematic analysis found that: (1) ACP was felt to be within the scope of general practice, but more so before the onset of dementia because lack of confidence in capacity assessment acted as a barrier once a diagnosis was made; (2) beliefs about the perceived benefits of ACP motivated GPs to engage patients in ACP; and (3) doctors felt that patients were often reluctant to discuss ACP, especially in the time-pressured context of the standard consultation. This study's use of the TDF provides direction for potential interventions to alleviate challenges faced by Australian GPs with ACP in dementia. Improving ACP in general practice may require a focus on public health campaigns to educate patients about the benefits of ACP before the onset of dementia. GPs' difficulty assessing a patient's decisional capacity may be addressed through skills development workshops.

Genome-wide analyses of platinum-induced ototoxicity in childhood cancer patients: Results of GO-CAT and United Kingdom MAGIC consortia.

Hearing loss (ototoxicity) is a major adverse effect of cisplatin and carboplatin chemotherapy. The aim of this study is to identify novel genetic variants that play a role in platinum-induced ototoxicity. Therefore, a genome-wide association study was performed in the Genetics of Childhood Cancer Treatment (GO-CAT) cohort (n = 261) and the United Kingdom Molecular Genetics of Adverse Drug Reactions in Children Study (United Kingdom MAGIC) cohort (n = 248). Results of both cohorts were combined in a meta-analysis. In primary analysis, patients with SIOP Boston Ototoxicity Scale grade ≥1 were considered cases, and patients with grade 0 were controls. Variants with a p-value <10-5 were replicated in previously published data by the PanCareLIFE cohort (n = 390). No genome-wide significant associations were found, but variants in TSPAN5, RBBP4P5, AC010090.1 and RNU6-38P were suggestively associated with platinum-induced ototoxicity. The lowest p-value was found for rs7671702 in TSPAN5 (odds ratio 2.0 (95% confidence interval 1.5-2.7), p-value 5.0 × 10-7). None of the associations were significant in the replication cohort, although the effect directions were consistent among all cohorts. Validation and functional understanding of these genetic variants could lead to more insights in the development of platinum-induced ototoxicity.

Use and performance of machine learning models for type 2 diabetes prediction in clinical and community care settings: Protocol for a systematic review and meta-analysis of predictive modeling studies.

OBJECTIVE: Machine learning involves the use of algorithms without explicit instructions. Of late, machine learning models have been widely applied for the prediction of type 2 diabetes. However, no evidence synthesis of the performance of these prediction models of type 2 diabetes is available. We aim to identify machine learning prediction models for type 2 diabetes in clinical and community care settings and determine their predictive performance. METHODS: The systematic review of English language machine learning predictive modeling studies in 12 databases will be conducted. Studies predicting type 2 diabetes in predefined clinical or community settings are eligible. Standard CHARMS and TRIPOD guidelines will guide data extraction. Methodological quality will be assessed using a predefined risk of bias assessment tool. The extent of validation will be categorized by Reilly-Evans levels. Primary outcomes include model performance metrics of discrimination ability, calibration, and classification accuracy. Secondary outcomes include candidate predictors, algorithms used, level of validation, and intended use of models. The random-effects meta-analysis of c-indices will be performed to evaluate discrimination abilities. The c-indices will be pooled per prediction model, per model type, and per algorithm. Publication bias will be assessed through funnel plots and regression tests. Sensitivity analysis will be conducted to estimate the effects of study quality and missing data on primary outcome. The sources of heterogeneity will be assessed through meta-regression. Subgroup analyses will be performed for primary outcomes. ETHICS AND DISSEMINATION: No ethics approval is required, as no primary or personal data are collected. Findings will be disseminated through scientific sessions and peer-reviewed journals. PROSPERO REGISTRATION NUMBER: CRD42019130886.

Cu4MnGe2S7 and Cu2MnGeS4: two polar thiogermanates exhibiting second harmonic generation in the infrared and structures derived from hexagonal diamond.

The new, quaternary diamond-like semiconductor (DLS) Cu4MnGe2S7 was prepared at high-temperature from a stoichiometric reaction of the elements under vacuum. Single crystal X-ray diffraction data were used to solve and refine the structure in the polar space group Cc. Cu4MnGe2S7 features [Ge2S7]6- units and adopts the Cu5Si2S7 structure type that can be considered a derivative of the hexagonal diamond structure. The DLS Cu2MnGeS4 with the wurtz-stannite structure was similarly prepared at a lower temperature. The achievement of relatively phase-pure samples, confirmed by X-ray powder diffraction data, was nontrival as differential thermal analysis shows an incongruent melting behaviour for both compounds at relatively high temperature. The dark red Cu2MnGeS4 and Cu4MnGe2S7 compounds exhibit direct optical bandgaps of 2.21 and 1.98 eV, respectively. The infrared (IR) spectra indicate potentially wide windows of optical transparency up to 25 µm for both materials. Using the Kurtz-Perry powder method, the second-order nonlinear optical susceptibility, χ(2), values for Cu2MnGeS4 and Cu4MnGe2S7 were estimated to be 16.9 ± 2.0 pm V-1 and 2.33 ± 0.86 pm V-1, respectively, by comparing with an optical-quality standard reference material, AgGaSe2 (AGSe). Cu2MnGeS4 was found to be phase matchable at λ = 3100 nm, whereas Cu4MnGe2S7 was determined to be non-phase matchable at λ = 1600 nm. The weak SHG response of Cu4MnGe2S7 precluded phase-matching studies at longer wavelengths. The laser-induced damage threshold (LIDT) for Cu2MnGeS4 was estimated to be ∼0.1 GW cm-2 at λ = 1064 nm (pulse width: τ = 30 ps), while the LIDT for Cu4MnGe2S7 could not be ascertained due to its weak response. The significant variance in NLO properties can be reasoned using the results from electronic structure calculations.

Sourcing Antibiotic-Resistant Escherichia coli in Aquatic Ecosystems: A Combined Laboratory and Field Module.

The emergence of antibiotic-resistant bacteria represents a growing threat in aquatic ecosystems. In this combined field and laboratory activity, students will determine whether Escherichia coli, an indicator bacteria species commonly found in aquatic ecosystems, shows signs of resistance to common antibiotics. In addition, students will use molecular biology techniques to identify whether Escherichia coli cells sourced from different hosts (i.e., phylogroups) show different patterns of antibiotic resistance. This activity will help students to gain experience in environmental microbiology, environmental science, molecular biology, and public health. This module is also designed to provide instructors with flexibility to pick and choose activities that best meet the needs of their class or research program.

Pragmatic cluster randomised control trial using Vaxcards as an age-appropriate tool to incentivise and educate school students about vaccination.

OBJECTIVE: This trial aimed to determine if return rates of consent forms for vaccination could be improved when Vaxcards were offered as an incentive to school children. SETTING: Nineteen schools in South East Melbourne participated. INTERVENTIONS: Students in the experimental arm received a pack of Vaxcards when they returned their government consent form. OUTCOME MEASURES: Return of 'yes' consent forms for vaccination as part of a local government council vaccine programme was the primary outcome of this trial. Return rates were compared between the intervention and control schools and with historical return rates. RESULTS: Secondary school students (N=3087) from 19 schools participated. Compared with historical returns, a small global reduction in 'yes' responses to consent forms of -4.21% in human papilloma virus consent 'yes' responses and -4.69% for diphtheria, tetanus and pertussis was observed across all schools. No difference between the experimental and control groups was observed. CONCLUSIONS: Low 'yes' consent rates and reduction in consent rates between 2018 and 2019 for all groups are concerning. This finding highlights the need for behaviour change interventions across all groups to increase vaccine confidence. Lack of effect of incentivisation with Vaxcards in this study may have been due to the timing of receiving the cards (after the decision to vaccinate had been made, not before) and the limited intensity of the intervention. Optimising the timing and the intensity of exposure to Vaxcards could improve the outcome. TRIAL REGISTRATION NUMBER: ACTRN12618001753246.

An Online, Person-Centered, Risk Factor Management Program to Prevent Cognitive Decline: Protocol for A Prospective Behavior-Modification Blinded Endpoint Randomized Controlled Trial.

BACKGROUND: Several modifiable risk factors for dementia have been identified, although the extent to which their modification leads to improved cognitive outcomes remains unclear. OBJECTIVE: The primary aim is to test the hypothesis that a behavior modification intervention program targeting personalized risk factors prevents cognitive decline in community-dwelling, middle-aged adults with a family history of dementia. METHODS: This is a prospective, risk factor management, blinded endpoint, randomized, controlled trial, where 1510 cognitively normal, community-dwelling adults aged 40-70 years old will be recruited. Participants will be screened for risk factors related to vascular health (including physical inactivity), mental health, sleep, and cognitive/social engagement. The intervention is an online person-centered risk factor management program: BetterBrains. Participants randomized to intervention will receive telehealth-based person-centered goal setting, motivational interviewing, and follow-up support, health care provider communication and community linkage for management of known modifiable risk factors of dementia. Psychoeducational health information will be provided to both control and intervention groups. RESULTS: The primary outcome is favorable cognitive performance at 24-months post-baseline, defined as the absence of decline on one or more of the following cognitive tests: (a) Cogstate Detection, (b) Cogstate One Card Learning, (c) Cogstate One Back, and (d) Cognitive Function Instrument total score. CONCLUSION: We will test the hypothesis that the BetterBrains intervention program can prevent cognitive decline. By leveraging existing community services and using a risk factor management pathway that tailors the intervention to each participant, we maximize likelihood for engagement, long-term adherence, and for preserving cognitive function in at-risk individuals.

Tabletop Board Game Elements and Gamification Interventions for Health Behavior Change: Realist Review and Proposal of a Game Design Framework.

BACKGROUND: Games, when used as interventional tools, can influence behavior change by incentivizing, reinforcing, educating, providing feedback loops, prompting, persuading, or providing meaning, fun, and community. However, not all game elements will appeal to all consumers equally, and different elements might work for different people and in different contexts. OBJECTIVE: The aim of this study was to conduct a realist review of tabletop games targeting behavior change and to propose a framework for designing effective behavior change games. METHODS: A realist review was conducted to inform program theory in the development of tabletop games for health behavior change. The context, mechanisms used to change behavior, and outcomes of included studies were reviewed through a realist lens. RESULTS: Thirty-one papers met the eligibility criteria and were included in the review. Several design methods were identified that enhanced the efficacy of the games to change behavior. These included design by local teams, pilot testing, clearly defined targets of behavior change, conscious attention to all aspects of game design, including game mechanics, dynamics, aesthetics, and the elicitation of emotions. Delivery with other mediums, leveraging behavioral insights, prior training for delivery, and repeated play were also important. Some design elements that were found to reduce efficacy included limited replayability or lack of fun for immersive engagement. CONCLUSIONS: Game designers need to consider all aspects of the context and the mechanisms to achieve the desired behavior change outcomes. Careful design thinking should include consideration of the game mechanics, dynamics, aesthetics, emotions, and contexts of the game and the players. People who know the players and the contexts well should design the games or have significant input. Testing in real-world settings is likely to lead to better outcomes. Careful selection and purposeful design of the behavior change mechanisms at play is essential. Fun and enjoyment of the player should be considered, as without engagement, there will be no desired intervention effect.

Patient-Chosen Gap Payment: an exploratory qualitative review of patients and general practitioner attitudes toward an alternative funding model for general practice.

We explored patients' and GPs' perceptions of an alternative payment system, a Patient-Chosen Gap Payment, where a gap fee is determined by the patient based on their perceived value of the service, including the choice to pay nothing. Semi-structured, in-depth interviews held with GPs (n=10) and patients (n=10) were audio-recorded, transcribed and analysed for emerging themes. We found three emergent themes: (1) the cost of quality: health care was difficult to value for both GPs and patients; there was belief in universal coverage and the importance of quality, but trade-offs in quality of care were a common perception; (2) the doctor-patient relationship: patient-centred care was a common goal and perceived as a good measure of quality care and a way for patients to place a value on the service/care; and (3) the business of general practice: participants wanted to see sustainable business models for primary care that incentivised quality of care. A Patient-Chosen Gap Payment (PCGP) funding model could incentivise doctors to provide better care without limiting access to health care. Further research is needed to model real-world application.

Synthesis and evaluation of etoposide and podophyllotoxin analogs against topoisomerase IIα and HCT-116 cells.

Etoposide is a widely-used anticancer agent that targets human type II topoisomerases. Evidence suggests that metabolism of etoposide in myeloid progenitor cells is associated with translocations involved in leukemia development. Previous studies suggest halogenation at the C-2' position of etoposide reduces metabolism. Halogens were introduced into the C-2' position by electrophilic aromatic halogenation onto etoposide (ETOP, 1), podophyllotoxin (PPT, 2), and 4-dimethylepipodophyllotoxin (DMEP, 3), and to bridge the gap of knowledge regarding the activity of these metabolically stable analogs. Five halogenated analogs (6-10) were synthesized. Analogs 8-10 displayed variable ability to inhibit DNA relaxation. Analog 9 was the only analog to show concentration-dependent enhancement of Top2-mediated DNA cleavage. Dose response assay results indicated that 8 and 10 were most effective at decreasing the viability of HCT-116 and A549 cancer cell lines in culture. Flow cytometry with 8 and 10 in HCT-116 cells provide evidence of sub-G1 cell populations indicative of apoptosis. Taken together, these results indicate C-2' halogenation of etoposide and its precursors, although metabolically stable, decreases overall activity relative to etoposide.

Mortality prediction model for the triage of COVID-19, pneumonia, and mechanically ventilated ICU patients: A retrospective study.

RATIONALE: Prediction of patients at risk for mortality can help triage patients and assist in resource allocation. OBJECTIVES: Develop and evaluate a machine learning-based algorithm which accurately predicts mortality in COVID-19, pneumonia, and mechanically ventilated patients. METHODS: Retrospective study of 53,001 total ICU patients, including 9166 patients with pneumonia and 25,895 mechanically ventilated patients, performed on the MIMIC dataset. An additional retrospective analysis was performed on a community hospital dataset containing 114 patients positive for SARS-COV-2 by PCR test. The outcome of interest was in-hospital patient mortality. RESULTS: When trained and tested on the MIMIC dataset, the XGBoost predictor obtained area under the receiver operating characteristic (AUROC) values of 0.82, 0.81, 0.77, and 0.75 for mortality prediction on mechanically ventilated patients at 12-, 24-, 48-, and 72- hour windows, respectively, and AUROCs of 0.87, 0.78, 0.77, and 0.734 for mortality prediction on pneumonia patients at 12-, 24-, 48-, and 72- hour windows, respectively. The predictor outperformed the qSOFA, MEWS and CURB-65 risk scores at all prediction windows. When tested on the community hospital dataset, the predictor obtained AUROCs of 0.91, 0.90, 0.86, and 0.87 for mortality prediction on COVID-19 patients at 12-, 24-, 48-, and 72- hour windows, respectively, outperforming the qSOFA, MEWS and CURB-65 risk scores at all prediction windows. CONCLUSIONS: This machine learning-based algorithm is a useful predictive tool for anticipating patient mortality at clinically useful timepoints, and is capable of accurate mortality prediction for mechanically ventilated patients as well as those diagnosed with pneumonia and COVID-19.

Use and performance of machine learning models for type 2 diabetes prediction in community settings: A systematic review and meta-analysis.

OBJECTIVE: We aimed to identify machine learning (ML) models for type 2 diabetes (T2DM) prediction in community settings and determine their predictive performance. METHOD: Systematic review of ML predictive modelling studies in 13 databases since 2009 was conducted. Primary outcomes included metrics of discrimination, calibration, and classification. Secondary outcomes included important variables, level of validation, and intended use of models. Meta-analysis of c-indices, subgroup analyses, meta-regression, publication bias assessments and sensitivity analyses were conducted. RESULTS: Twenty-three studies (40 prediction models) were included. Studies with high-, moderate-, and low- risk of bias were 3, 14, and 6 respectively. All studies conducted internal validation whereas none conducted external validation of their models. Twenty studies provided classification metrics to varying extents whereas only 7 studies performed model calibration. Eighteen studies reported information on both the variables used for model development and the feature importance. Twelve studies highlighted potential applicability of their models for T2DM screening. Meta-analysis produced a good pooled c-index (0.812). Sources of heterogeneity were identified through subgroup analyses and meta-regression. Issues pertaining to methodological quality and reporting were observed. CONCLUSIONS: We found evidence of good performance of ML models for T2DM prediction in the community. Improvements to methodology, reporting and validation are needed before they can be used at scale.