Sherrie L. Kelly

The rapid integration of artificial intelligence (AI) into clinical practice necessitates urgent restructuring of medical education and physician assessment to ensure that future physicians are proficient and responsible users of AI tools. Despite the existence of core AI competencies, the current state of AI education in Canadian undergraduate medical education is highly inconsistent and disjointed, and available data indicate that most medical students receive minimal to no formal AI training even as they anticipate that AI will profoundly shape their future careers. National policy, specifically the Pan-Canadian AI for Health Guiding Principles, has advanced the agenda by calling for AI literacy among health professionals and emphasizing core values such as equity, robust data practices, and Indigenous-led data governance. However, these principles offer limited practical guidance on the educational and regulatory mechanisms required for effective implementation. We contend that this critical implementation deficit arises from a traditional, sequential reform model in which faculty development, curriculum change, and regulatory updates occur in isolation. This slow, siloed approach is fundamentally inadequate for addressing AI's inherent speed, opacity, and significant equity risks. To overcome this challenge, we propose a 3-lever concurrent implementation framework that provides a conceptual lens to address the interdependencies among faculty development, curriculum, and regulation. This model posits that AI competencies transition from abstract requirements to practical application only when 3 levers-clinician-educator capacity, digitally enabled learning environments, and regulatory and assessment reform-are activated simultaneously and in alignment. This Viewpoint extends existing AI competency frameworks by theorizing AI curriculum implementation as a problem of concurrent lever activation and by outlining minimum concurrent actions for deans and regulators that can be adapted to competency-based medical education systems. Although illustrated with Canadian examples, the framework is designed to be transferable beyond Canada and offers a testable, licensure-level blueprint for embedding AI competence in medical education.

UNSTRUCTURED The rapid integration of artificial intelligence (AI) into clinical practice necessitates urgent restructuring of medical education and physician assessment to ensure future doctors are proficient and responsible users of AI tools. Despite the existence of core AI competencies, the current state of AI education within Canadian Undergraduate Medical Education (UGME) is highly inconsistent and disjointed, and available data indicate that most medical students receive minimal to no formal AI training even as they anticipate AI profoundly shaping their future careers. National policy, specifically the Pan-Canadian AI for Health (AI4H) Guiding Principles, has advanced the agenda by calling for AI literacy among health professionals and emphasizing core values such as equity, robust data practices, and Indigenous-led data governance. However, these principles offer limited practical guidance on the educational and regulatory mechanisms required for effective implementation. We contend that this critical implementation deficit arises from a traditional, sequential reform model in which faculty development, curriculum change, and regulatory updates occur in isolation. This slow, siloed approach is fundamentally inadequate for addressing AI’s inherent speed, opacity, and significant equity risks. To overcome this, we propose a three-lever concurrent implementation framework. This model posits that AI competencies transition from abstract requirements to practical application only when three levers—clinician-educator capacity, digitally enabled learning environments, and regulatory and assessment reform—are activated simultaneously and in alignment. This Viewpoint extends existing AI competency frameworks by theorizing AI curriculum implementation as a problem of concurrent lever activation and by outlining minimum concurrent actions for deans and regulators that can be adapted to competency-based medical education systems. Although illustrated with Canadian examples, the framework is designed to be transferable beyond Canada and offers a testable, licensure-level blueprint for embedding AI competence in medical education.

Background: The recommendation of the RTS,S/AS01 malaria vaccine for use in children in moderate to high transmission areas by the WHO in 2021 provided another crucial tool in reducing malaria morbidity and mortality. As countries introduce the RTS,S vaccine, there is an interest in exploring vaccination schedules that align with existing routine health touch points and to optimize vaccine coverage. Methods: This study used OpenMalaria, an individual-based, stochastic model of malaria transmission and disease progression, to assess the impact of different vaccination timing schedules and coverage on uncomplicated malaria cases, severe malaria cases, and malaria-related deaths across different archetypal transmission settings. We ran simulations comparing fourth dose protection against infection and timings at six-, nine-, 12-, 15- and 18-month intervals following the primary series, including exploring ranges of plausible protection assumptions for many of the dose timings. Results: The primary series substantially reduces the malaria burden across transmission settings, regardless of timing of the fourth dose. While the number of cases averted varies by baseline prevalence, our modelling suggests that the primary series is responsible for around 54-93% of the total clinical and severe cases averted across transmission archetypes in perennial settings. Additionally, a fourth dose of the vaccine could avert additional 9-46% malaria cases and 7-41% severe cases in addition to the primary series alone. We find potential flexibility in timing this dose, especially when focused on 6 to 12 months following the primary series. Conclusions: In addition to the primary series, use of a fourth dose of the RTS,S vaccine can be tailored to country specific context and linked to existing health touch points to ensure adequate coverage of this dose. In particular, a fourth dose delivered between 15- and 21-months of age with high population coverage will likely avert the largest proportion of cases of clinical malaria, severe malaria, and deaths in perennial settings, across transmission intensities. Trial registration: N/A

New malaria vaccine development builds on groundbreaking recommendations and roll-out of two approved pre-erythrocytic vaccines (PEVs); RTS,S/AS01 and R21/Matrix-M. Whilst these vaccines are effective in reducing childhood malaria within yearly routine immunization programs or seasonal vaccination, there is little evidence on how different PEV efficacies, durations of protection, and spacing between doses influence the potential to avert uncomplicated and severe childhood malaria. Mainly, lacking understanding of the required vaccine properties and delivery strategies that lead to an effective childhood vaccine with multi-year protection. We used an individual-based model of malaria transmission informed by trial data to quantify trade-offs between PEV performance properties and impact across different endemicities, deployment schedules, and coverage levels. We found that deploying a vaccine with 90% initial efficacy against infection, with a six to 12-month half-life duration of protection, co-administered with a blood-stage drug, followed by yearly boosters, results in 60–80% yearly incidence reduction, consistent with seasonal RTS,S and R21 trials. Halting vaccination after five years, leads to sustained protection of at least a 35% incidence reduction in children less than six years in the 12 months following cessation in settings where PfPR_2–10< 30%. Increasing the half-life duration to 12–18 months or reaching more children provides the same health impact with lower vaccine efficacy. Without a booster (fourth dose), high efficacy against infection (>90%) and longer half-life duration (>12 months) are required to sustain impact beyond primary vaccination, averting up to half the preceding year’s burden. The contribution of each property to the overall impact varies by setting and clinical endpoint, indicating that public health goals should dictate key vaccine performance criteria. Overall, our findings support the need for well-defined target product profiles for long duration vaccines linking priority use cases of where, how, and to whom to deploy new malaria vaccines, to maximize public health impact.

Background: Characterising disparities in HIV infection across populations by gender, age, and HIV risk is key information to guide intervention priorities. We compared 9 models representing 15 different settings across Africa to assess how indicators measuring HIV acquisitions, transmissions, or potential long-term infections influence estimates of the contribution of different populations to new infections, including key populations (KPs, including female sex workers (FSW), their clients, men who have sex with men). Methods: We evaluated four indicators: I ₁ ) acquisition indicator measuring the annual fraction of all new infections acquired by a specific population, I ₂ ) direct transmission indicator measuring the annual fraction of all new infections directly transmitted by a specific population, I ₃ ) 1-year and I ₄ ) 10-year transmission population-attributable fractions ( tPAFs ). tPAFs measure the fraction of new infections averted if transmission involving a specific population was blocked over a specific time period. We compared estimates of the four indicators across 7 populations and 15 settings and assessed if the contribution of specific populations is ranked differently across indicators for 10 settings. Findings: Indicators identified distinct priority populations as the largest contributors: The acquisition indicator (I ₁ ) identified women aged 25+ years outside KPs as contributing the most to acquired infections in 8/10 settings in 2020, but to direct transmissions (I ₂ ) in only two settings. In 6/10 settings, the 10-year tPAFs (I ₄ ) identified non-KP men aged 25+ years and clients of FSW as the largest contributors to HIV transmission. Notably, non-KP women aged 15-24 years acquired (I ₁ ) more infections in 2020 (median of 1·7-fold across models) than they directly transmitted (I ₂ ), while non-KP men aged 25+ years and clients of FSWs transmitted more infections than they acquired in all but one model (median: 1·4 and 1·6-fold, respectively). Estimates of the 10-year tPAFs accounting for transmission in the long-term were substantially larger than the direct transmission indicator for all populations, especially for FSW (median: 2·0-fold). Interpretation: Indicators that reflect HIV acquisitions and transmissions over the short and long term can be utilised to capture the complexity of HIV epidemics across different populations and timeframes. The added nuance would improve the effectiveness of the HIV prevention response across all populations at risk. Funding: NIH, MRC. Research in context: Evidence before this study: Measures of the distribution of HIV acquisition across population groups are commonly used for assessing the contribution of populations to new HIV infections and prevention priorities. However, alternative indicators documented in the literature reflect transmissions or potential long-term effects. It is unclear how the choice of indicator affects the identification of populations that require additional prevention and treatment efforts to accelerate progress towards ending AIDS. We searched PubMed on March 08, 2025, with the terms (HIV) AND (Africa*) AND (acqui*) AND (transm*) AND (model*), with no language or publication date restriction, and identified no meta-assessment or mathematical model comparison studying differences in estimates of the fraction of all infections acquired and transmitted by a population when using different epidemiological indicators. Added value of this study: Using estimates from 9 models representing 15 different epidemic settings across Africa, we studied indicators of HIV epidemic contribution for 7 populations, including female sex workers, their clients, men who have sex with men, and non-key populations stratified by gender and age. We measured four commonly reported indicators of HIV contribution. One focused on acquired infections and the other three focused on transmissions. We found that estimates from these different indicators can differ greatly for the same model and population, to the extent that they identify different populations for prioritising interventions to accelerate HIV incidence declines. The acquisition-focused indicator (i.e. fraction of all infections acquired by a given population), the most used and communicated by UNAIDS, substantially underestimates the large contribution of men, and particularly male clients of female sex workers, to ongoing HIV transmission. Implications of all the available evidence: The choice of indicators measuring a population’s contribution to the HIV epidemic should be carefully considered and precisely defined. Modelling teams working in partnership with government, implementers, funders, and community members should systematically report both acquisition- and (long-term) transmission-focused indicators, instead of only measuring acquisitions in the short term as currently done. Multiple indicators will more comprehensively capture the potential impact of prevention efforts addressing acquisition and transmission risks of different vulnerable populations.

Vaccines, monoclonal antibodies, and long-acting injectables are being developed to prevent Plasmodium falciparum malaria. These therapeutics may target multiple stages of the parasite life cycle; evidence is needed to articulate their benefits with chemoprevention and prioritise candidates for clinical development. We used an individual-based malaria transmission model to estimate the health impact of combining new therapeutics with seasonal malaria chemoprevention (SMC). Our modelling framework used emulator-based methods with models of pre-liver and blood stage therapeutic dynamics. We evaluated the benefit of combining therapeutics with SMC in children under five by estimating reductions in the cumulative incidence of uncomplicated and severe malaria, relative to SMC or the new therapeutic alone, during and five years after deployment. New therapeutics may require extended pre-liver stage duration or multi-stage activity to combine with SMC. For three SMC cycles in a high transmission setting, a pre-liver stage therapeutic with partial initial efficacy (>50%) required a protection half-life >230 days to reduce cumulative severe cases by >5% five years after deployment stopped (>23% during interventions). Longer protection was needed when combined with four or five SMC cycles. Combining SMC with a multi-stage therapeutic increased public health impact both during and after deployment. Combining SMC with malaria therapeutics active against multiple stages of the parasite life cycle can improve the effectiveness of SMC, highlighting the need to prioritise the clinical development of these therapeutics for combination with malaria chemoprevention.

New malaria vaccine development builds on groundbreaking recommendations and roll-out of two approved pre-erythrocytic vaccines (PEVs); RTS,S/AS01 and R21/MM. Whilst these vaccines are effective in reducing childhood malaria within yearly routine immunization programs or seasonal vaccination, there is little evidence on how different PEV efficacies, durations of protection, and spacing between doses influence the potential to avert uncomplicated and severe childhood malaria. Mainly, lacking understanding of the required vaccine properties and delivery strategies that lead to an effective childhood vaccine with multi-year protection. We used an individual-based model of malaria transmission informed by trial data to quantify trade-offs between PEV performance properties and impact across different endemicities, deployment schedules, and coverage levels. We found that deploying a vaccine with 90% initial efficacy, with a six to 12-month half-life duration of protection, co-administered with a blood-stage drug, followed by yearly boosters, results in 60-80% yearly incidence reduction, consistent with seasonal RTS,S and R21 trials. Halting vaccination after five years, leads to sustained protection of at least a 35% incidence reduction in children Pf PR _2-10 90%) and longer half-life duration (>12 months) are required to sustain impact beyond primary vaccination, averting up to half the preceding year’s burden. The contribution of each property to the overall impact varies by setting and clinical endpoint, indicating that public health goals should dictate key vaccine performance criteria. Overall, our findings support the need for well-defined target product profiles for long duration vaccines linking priority use cases of where, how, and to whom to deploy new malaria vaccines, to maximize public health impact.

Background: Seasonal malaria chemoprevention (SMC) is recommended for disease control in settings with moderate to high Plasmodium falciparum transmission and currently depends on administration of sulfadoxine-pyrimethamine with amodiaquine. However, poor regimenadherence and the increasedfrequencyof sulfadoxine-pyrimethamine resistant parasite mutations may threaten SMC’s effectiveness. We need guidance to de-risk the development of drug compounds for malaria prevention. Methods: We combined an individual-based malaria transmission model that has explicit parasite growth with drug pharmacokinetic/pharmacodynamic models. We modelled SMC drug attributes for several possible modes-of-action, linked to their potential public health impact. Global sensitivity analyses identified trade-offs between drug elimination half-life, maximum killing effect, and SMC coverage, and optimisation identified minimum requirements to maximise malaria burden reductions. Findings: Model predictions show that preventing infection for the entire period between SMC cycles is more important than drug curative efficacy for clinical disease effectiveness outcomes, but similarly important for impact on prevalence. When four SMC cycles are deployed to children under five years with high levels of coverage (69% of children receiving all cycles), drug candidates require a duration of protection half-life of >23 days (elimination half-life >10 days) to achieve >75% clinical incidence and severe disease reductions (measured over the intervention period in the target population, compared with no intervention across a range of modelled scenarios). High coverage is critical to achieve these targets, requiring >60% of children received all SMC cycles and >90% of children at least one cycle regardless of the drug’s duration. Interpretation: While efficacy is crucial for malaria prevalence reductions, chemoprevention development should select drug candidates for their duration of protection to maximise burden reductions, with the duration half-life determiningcycle timing. Explicitlydesigning or selectingdrug properties to increase communityuptake is paramount. Funding: Bill & Melinda Gates Foundation and the Swiss National Science Foundation.

Sciris aims to streamline the development of scientific software by making it easier to perform common tasks. Sciris provides classes and functions that simplify access to frequently used low-level functionality in the core libraries of the scientific Python ecosystem (such as numpy for math and matplotlib for plotting), as well as in libraries of broader scope (such as multiprocess for parallelization and pickle for saving and loading objects). While low-level functionality is valuable for developing robust software applications, it can divert focus from the scientific problems being solved. Some of Sciris' key features include: ensuring consistent dictionary, list, and array types (e.g., enabling users to provide inputs as either lists or arrays); enabling ordered dictionary elements to be referenced by index; simplifying datetime arithmetic by allowing date input in multiple formats, including strings; simplifying the saving and loading of files and complex objects; and simplifying the parallel execution of code. With Sciris, users can often achieve the same functionality with fewer lines of code, avoid reinventing the wheel, and spend less time looking up recipes on Stack Overflow. This can make writing scientific code in Python faster, more pleasant, and more accessible, especially for people without extensive training in software development.

Background Seasonal malaria chemoprevention (SMC) with sulfadoxine-pyrimethamine (SP) plus amodiaquine (AQ) prevents millions of clinical malaria cases in children under five in Africa’s Sahel region. However, parasites partially resistant to SP (with “quintuple” mutations) potentially threaten SMC protective effectiveness. We evaluated its spread and clinical consequences. Methods An individual-based malaria transmission model with explicit parasite dynamics and drug pharmacological models, was used to identify and quantify the influence of factors driving quintuple mutant spread and predict the time needed for the mutant to spread from 1% to 50% of inoculations for several SMC deployment strategies. We estimated the impact of this spread on SMC effectiveness against clinical malaria. Findings Higher transmission intensity, SMC coverage, and expanded age range of chemoprevention promoted mutant spread. SMC implementation in a high transmission setting (40% parasite prevalence in children aged 2-10 years) with four monthly cycles to children aged three months to five years (with 95% initial coverage declining each cycle), the mutant requires 53·1 years (95% CI 50·5–56·0) to spread from 1% to 50% of inoculations. This time increased in lower transmission settings and reduced by half when SMC was extended to children under ten, or reduced by 10-13 years when an additional monthly cycle of SMC was deployed. For the same setting, the effective reduction in clinical cases in children receiving SMC was 79·0% (95% CI 77·8–80·8) and 60·4% (95% CI 58·6–62·3) during the months of SMC implementation when the mutant was absent or fixed in the population, respectively. Interpretation SMC with SP+AQ leads to a relatively slow spread of SP-resistant quintuple mutants and remains effective at preventing clinical malaria despite the mutant spread. SMC with SP+AQ should be considered in seasonal settings where this mutant is already prevalent. Funding Swiss National Science Foundation and Marie Curie Individual Fellowship.