Thank you to all the contributors to this issue of the newsletter! We have a new, regular section, “From the Field,” with articles from members. In this issue Jenny Spencer reflects on the wide variety of opinions laypeople have about cancer screenings. Ryan Suk provides insightful commentary on the crucial role of decision-making science in an increasingly AI-enabled world. Kine Pedersen describes the foundational work on cervical cancer prevention that came out of a collaboration started at SMDM more than a decade ago, and Carlo Federici provides an overview of the SUSTAIN-HTA Initiative which will present beta versions of the project’s new tools for the SMDM community to engage with at the annual conference in June.

The newsletter welcomes content for the From the Field section from all members. If you have an idea for an article you would like to include in a future issue, let me know at hcolby@iu.edu. We also welcome submissions of other content including upcoming opportunities and workshops, jobs, opinion pieces, or even brief tutorials.

Thank you for reading the newsletter, and if you haven’t done so already, don’t forget to register for the annual conference in Oslo!

April update: Even if the weather’s unpredictable and our next meeting theme is “Medical Decision Making in Uncertain Times,” there’s a 100% chance of a successful meeting in Oslo - so pack a lightweight rain jacket and your sensitivity analyses.

For our first international conference, we saw record interest: nearly 500 abstracts from six continents and 30+ countries. Submissions span North America (267: United States, Canada), Europe (167: Netherlands, United Kingdom, Norway, Austria, Italy, Germany, Switzerland, Ireland, Belgium, France, Spain, Portugal, Denmark, Sweden, Finland, Hungary, Greece), Asia (35: Singapore, Malaysia, Taiwan, India, China, Korea, Thailand, Philippines, Israel, Saudi Arabia, Lebanon, Iran, Indonesia, Armenia), Oceania (23: Australia), Africa (3: Egypt, Ghana, Tanzania), and South America (1: Colombia). We’re also proud to support participation with 13 scholarships: five Clinical Scholarships, seven LMIC/Trainee Scholarships, and one Dependent Care award. Together, these numbers reflect our commitment to access, inclusion, and a truly global community. In politically challenging times, our scientific community has a special responsibility to collaborate across borders, uphold scientific rigor, evidence, and integrity, and translate decision science into better health for all.

Beyond the conference, momentum for decision science continues to grow in policy and practice. For the first time, Germany’s HTA agency IQWiG has built and implemented its own decision model to evaluate colorectal cancer screening strategies (e.g., start age, test intervals) for G-BA comparisons. After first relying on an external model for breast cancer in 2022, this in-house effort marks a notable step in embedding decision analysis within German HTA and exemplifies the field’s expanding impact -reflecting the kind of methods development and real-world application advocated by the SMDM community.

We’re excited to unveil SMDM’s redesigned website, built to make it easier to find resources, connect with colleagues, and share opportunities. Visit to explore key pages, find conferences, webinars, guidelines, teaching materials, and SIG pages, and post openings to the SMDM Job Board (faculty, postdocs, internships, industry roles). We invite your feedback - send comments, feature requests, or bug reports; we’re listening. 

We also completed a seamless transition to our new project management partner, MWS Partners - so smooth many may not have noticed. Thanks to careful preparation and the innovative team that has supported us in recent years, the move preserved continuity while benefiting from MWS’s framework for high-quality operations, responsive engagement, and enhanced member support. We’re excited to build on this strong foundation together.

Want to get involved? Visit the SMDM committee and SIG information desk at the Annual Meeting in Oslo to learn how you can contribute throughout the year.

Please join us in congratulating the 2026 SMDM Annual Award recipients and thanking our Awards Committee for its work: Uwe Siebert (Career Achievement Award); Brian J. Zikmund-Fisher (Distinguished Service Award); Jagpreet Chhatwal and Rachael Fleurence (John M. Eisenberg Award for Practical Application of Medical Decision Making Research, jointly recognized for their collaborative work); Saniya Singh (International Early Scholars Award in Honor of Murray Krahn); Alyssa Bilinski (Early Scholars Award in Health Policy in Honor of Sandy Schwartz); and Ashley Leech (Outstanding Paper by an Early Investigator). These honorees will be celebrated at the Leadership Awards Ceremony on Tuesday, 30 June during the SMDM 48th Annual Meeting in Oslo (28 June–1 July 2026). 

We look forward to seeing you at our 48th Annual Meeting in Oslo - our first international conference - or at our online networking and education events. 

Beate Jahn

President, Society for Medical Decision Making

Everyone has an opinion about cancer screening

Jenny Spencer, PhD

I’m chatty and I talk to a lot of strangers. Inevitably, I am asked what I do. When I answer that I research cancer screening, I am usually treated to some very strongly held opinions about the topic. These opinions are often well-earned, perhaps the product of supporting close family members through diagnosis and treatment or their own experiences with scary abnormal findings, but these opinions also reflect some common misunderstandings around cancer screening guidelines. 

“It’s just the evil insurance companies not wanting to cover it”, says the woman cutting my hair. Cancer screening guidelines in the United States come from multiple sources – the American Cancer Society, professional organizations, but perhaps most notably the US Preventive Services Task Force, as USPSTF recommendations are now connected to insurance coverage. These recommendations are informed by systematic reviews of evidence from trials and large observational studies as well as through simulation modeling studies. These models recreate the “natural history” of cancer – estimating what would happen in the absence of any screening and then layering screening scenarios on top of this baseline to evaluate the benefits and potential harms of alternative screening strategies. For example, in the 2024 breast cancer screening recommendations, six different simulation models each evaluated 36 different combinations of when to start or stop screening, how often to screen, and which screening technologies to use.

“If you had cancer, wouldn’t you want to know as soon as possible?” says the friendly barista as they pour my Americano. An important part of any screening test is identifying the *just right* amount of testing. Of course, with too little testing there will be potentially preventable deaths from cancer, but over-testing can result in a high number of false positives or an excess of resources spent to gain very little in overall health. The USPSTF guidelines are based on considerations of tradeoffs between harms and benefits of screening and attention to where screening begins to reach diminishing returns. In the modeling conducted for the 2021 colorectal cancer screening guidelines, the model outcomes are compared as a ratio of the additional life years gained through screening vs. number of lifetime screens as well as the number of complications per 1,000 – identifying the points where the longevity benefits from more screening start to flatten. 

“I don’t know. I just don’t trust it.” says a friend of a friend of a friend at a backyard barbeque. People can be resistant to changes in screening guidelines – especially when they involve screening less. But these changes aren’t arbitrary – they reflect a lot of scientific progress. Pap smears for cervical cancer screening were recommended annually until 1996 when improvements in testing methods and better understanding of the disease motivated a move to every 3 years. In 2012, the ability to test for HPV (the Human Papillomavirus that causes cervical cancer) and a better understanding of risk stratification across HPV types offered further opportunities to increase the interval between tests to 5 years. Current draft recommendations for cervical cancer incorporate new technologies for self-collected testing methods. Cervical cancer risk will decline further as the impacts of HPV vaccination become realized – so there are certainly more changes to come. 

Sometimes the people I’m talking to are reassured. Sometimes not. The challenge is that guidelines are made for populations and screening decisions are made by individuals. Individuals are complex and have to weigh their own individual risk, experiences, and preferences. But that, as I tell my new friends, is a conversation for the other half of SMDM.  

In an AI-Focused Era, the Case for Decision Science

Ryan Suk, PhD, MS

When I recently gave a guest lecture on decision modeling and economic evaluation, a senior colleague asked me, “Do you do this with AI?” I remember pausing for quite a while, partly because I was not entirely sure what the question meant. But I was also struck by something deeper… how quickly, in this moment, many forms of rigorous analytic work are being absorbed into one broad and often blurry conversation about AI.

To be clear, I am not resistant to AI. I do actively incorporate it into my analytics work where it is useful, such as pattern recognition and accelerating certain forms of evidence synthesis. But the encounter stayed with me because it revealed a conceptual flattening that should give our field pause. Decision modeling and economic evaluation are not simply older tools waiting to be rebranded in the language of AI. They are structured around a different set of concerns.

AI encompasses a wide range of methods. In many of its most visible current applications, it is oriented toward extracting patterns from data and generating predictions, classifications, or other outputs. Some approaches also aim to learn policies that maximize a specified objective. But decision science is distinguished by its explicit focus on defining the decision problem itself, including the objectives that those methods often take as given. It asks what choices are available, what outcomes matter, what trade-offs are at stake, whose perspective counts, how uncertainty should be handled, and what counts as value. Economic evaluation sharpens that logic further by making opportunity cost explicit, asking not only whether something can be done, but whether it should be done relative to competing uses of finite resources.

That distinction is not semantic. It is practical. In health care, a highly accurate model is not the same thing as a valuable intervention. As a growing body of health care AI scholarship has emphasized, enthusiasm often outpaces evidence of improved outcomes in real-world settings, where value depends heavily on workflow, implementation, training, and organizational context. 

This is precisely where decision scientists should be especially confident about our role. We should articulate how decision science can shape an AI-focused era.

Decision science helps define the relevant decision before optimizing the wrong one.

Decision science evaluates interventions in terms of consequences, costs, uncertainty, equity, and their trade-offs.

Decision science recognizes that implementation shapes whether an innovation has value at all.

In other words, this moment does not diminish decision science. It makes it more necessary. If AI becomes one more input into health care decision-making, then decision science can help determine how that input is evaluated, where it fits, when it adds value, and when it does not. 

So perhaps the question of whether we “do our work with AI” was more valid than I initially realized, because it leads to something deeper: whether AI-focused work is paying enough attention to the decisions it is meant to inform. If decision scientists are to help shape that conversation, then we need to ask: how do we define and assert our role in it? 

How SMDM collaborations and decision modeling informed cervical cancer prevention policies in Norway

Kine Pedersen, PhD, Associate Professor at the University of Oslo

Medical decision making often requires acting in the face of incomplete, evolving, and sometimes conflicting evidence. Over the past decade, my work has focused on using disease simulation models to inform cervical cancer prevention policies in the context of technological and epidemiological change, within the context of Norway. 

The discovery that human papillomavirus (HPV) causes nearly all cervical cancers led to the development of both highly effective vaccines and HPV-based screening tests, which fundamentally reshaped prevention strategies. Since 2009, many countries, including Norway, have introduced HPV vaccination programs and revised screening guidelines. These shifts have required continuous evaluation, and simulation modeling has played a central role in supporting these policy decisions. 

Much of this work has been conducted in close collaboration with national stakeholders, including the Cancer Registry and public health authorities responsible for screening and vaccination programs. Within these collaborations, model-based analyses have served as a tool to quantify tradeoffs, project long-term outcomes, and inform real-world decisions. In our work, we use a modeling framework developed by researchers at Harvard T.H. Chan School of Public Health, that links a transmission model of HPV infection with a disease model reflecting cervical carcinogenesis. This framework allows us to simulate the pathway from infection to disease, while layering on interventions such as vaccination and screening. The cervical cancer modeling collaboration between Harvard University and Oslo University in Norway started more than 15 years ago, at an SMDM meeting, and has since resulted in more than 30 analyses and policy recommendations. 

One application of our work, published in Medical Decision Making, explored how cervical cancer prevention policies perform when evaluated jointly rather than in isolation. Using our modeling framework, we compared a restricted analysis that evaluated optimal HPV vaccine choice under current screening guidelines, to a comprehensive analysis including alternative screening and vaccination strategy combinations. A central insight was that the value of vaccination depended on the intensity of screening. Evaluating these interventions independently can therefore lead to suboptimal, or even misleading, policy conclusions. By explicitly modeling their interaction, the analysis highlights the importance of coordinated, program-level decision making. This perspective is becoming increasingly important as vaccinated cohorts age into screening programs.

In our latest study, published in the Annals of Internal Medicine in February 2026, we extend on this work by evaluating how cervical cancer screening can be tailored based on a woman’s age at HPV vaccination. Using our modeling framework, we project lifetime outcomes under a range of alternative screening strategies for women who received the HPV vaccine between the ages of 12 and 30 years. We varied ages to start screening, screening intervals and total lifetime test, while incorporating both benefits (cancer prevention) and harms (follow-up procedures). Our results suggest that across all vaccination ages and vaccine types, less frequent screening, involving fewer lifetime tests and longer intervals, was consistently preferred, particularly for women vaccinated at younger ages.

As population risk of HPV and consequently cervical cancer declines, the marginal benefit of intensive interventions diminishes, while the relative importance of harms increases. Simulation models are uniquely suited to capturing these dynamics, allowing us to move beyond one-size-fits-all recommendations toward more risk-based, individualized strategies. At their core, decision models provide a structured way to synthesize evidence, test assumptions, and explore the consequences of alternative choices. The experience from cervical cancer prevention illustrates how decision models can help shape adaptive medical decisions – and how SMDM can facilitate international collaborations to make it possible.

Bridging Innovation and Practice in HTA: The SUSTAIN-HTA Initiative

Carlo Federici, PhD

Health Technology Assessment (HTA) is facing increasing pressure to adapt to a rapidly evolving landscape of health technologies, policy requirements, and methodological expectations. Advances such as precision medicine, smaller target populations, and more complex evidence generation pathways are challenging traditional HTA approaches. In this context, the SUSTAIN-HTA project has been launched as a Horizon Europe Coordination and Support Action (2024–2027) to help bridge the gap between methodological innovation and real-world HTA practice.

SUSTAIN-HTA brings together a consortium of HTA bodies, academic institutions, and other stakeholders with the shared goal of supporting the uptake of innovative and fit-for-purpose HTA methods. While significant progress has been made over the past decade in developing new methodologies—often through EU-funded research—these advances have not always translated into routine use by HTA agencies. SUSTAIN-HTA directly addresses this challenge by building a structured, sustainable framework to identify the methodological needs of HTA bodies and align them with ongoing research and innovation efforts.

At the heart of the project is the development of a set of tools and activities designed to support implementation. These include a horizon scanning tool to systematically identify priority methodological needs, a living methods observatory to track and synthesize methodological developments from research projects and the broader literature, and a sandbox environment where new approaches can be tested and validated before being adopted in practice. In parallel, the SUSTAIN-HTA project is promoting initiatives focused on stakeholder engagement and capacity building. A dialogue platform is being developed as a structured matchmaking service between academia and HTA bodies to support collaboration on activities including methodological research, technical support, the development of operational guidelines, and training. A fellowship program, the definition of a validated HTA competency framework, and an online learning platform with HTA-relevant content also aim to strengthen HTA capacity in innovative methods across EU Member States.

SUSTAIN-HTA will be featured at the upcoming 48th Annual Meeting of the Society for Medical Decision Making (SMDM) in Oslo, offering an opportunity for the medical decision science community to engage with the project’s work. The session will present the beta versions of the project’s core tools and provide an overview of how they can be used to support the integration of innovative methods into HTA processes. The session is designed to be interactive, inviting feedback from SMDM members and fostering discussion on how SUSTAIN-HTA can align with ongoing developments in medical decision making. Researchers, practitioners, and policymakers within the SMDM community are invited to join the session to discuss practical approaches to implementing methodological innovation in HTA and to explore opportunities for collaboration. Participants will have the opportunity to engage directly with the project team, explore the tools under development, and consider how these resources can support their own work.

By aligning methodological research with the needs of decision-makers, SUSTAIN-HTA aims to strengthen the role of HTA in supporting high-quality, evidence-based healthcare decisions. The project team considers this session in Oslo a valuable opportunity to further engage with the HTA community, ensuring that innovation in HTA methods translates into meaningful impact in practice.

All are welcome!

Cost-Effectiveness and Decision Modeling using R Workshop

25-27 June, 2026

University of Oslo

Kick-off your SMDM 2026 experience with a hands-on workshop: Decision Modeling in R

Join the Decision Analysis in R for Technologies in Health (DARTH) workgroup, in partnership with the University of Oslo, in a hands-on workshop happening June 25-27 2026. This workshop will cover building decision tree, cohort, and microsimulation models and conducting sensitivity analyses during three days of live coding sessions and hands-on exercises. You are expected to have a basic understanding of decision modeling and R. A pre-course module will be provided to brush up on your R basics. This workshop is a unique opportunity to improve your skills in developing health economic models in R and to expand your network within SMDM.