On 29 March 2025, the programme committee informed me that my abstract (“Python, Politics, and Public Health”) had been accepted for a talk at EuroPython 2025 in Prague. As luck would have it, I ended up delivering it on the first day after my retirement from WHO. The talk thus became a manifesto for where I hope to go next, uniting my work in public health with my long-standing love of Python.

I’ve divided my notes into short sections, aiming to capture the essence of the conference from many perspectives. I want to jog happy memories among those who attended, and to entice others to join next year. Skip to the parts that interest you.

As delighted as I was to hear the news that the abstract proposal had been accepted, many questions remained: how would the Python community respond to a retired public health official from a different world? Would there be kindred spirits interested in the health and social impact of code?

I am embarrassed by the arrogance in my abstract where I said:

“Python is everywhere—except, it seems, in public health. ... Despite its versatility, Python remains underutilised in public health. Why does this gap exist? How can we bridge the divide between Python as a tool for developers and its potential as an engine for hacking health?”

I could not have strayed farther from the truth. Indeed, in hallway conversations, I had several encounters with participants who have worked in either clinical or public health applications of Python and who would appreciate opportunities to do so again in the future. Indeed, the many health-related talks summarised below serve as a good illustration of some of the essential functions of public health.

Public Health Talks at EP2025

Python, Politics, and Public Health

In my own presentation, I gave a selection of historical examples of how I have used Python to either analyse public health data with political relevance or to help advocate for public health policies. The war stories covered:

• a 2015 network analysis of a Twitter conversation on vaping and how it turned out to have a surprising connection to Brexit,
• a data pipeline for the graphics in a report on women’s health in Europe and how its long tail has culminated in a programme on violence against women, and
• the visualisation of a thought experiment that I used to illustrate the inequity in chronic diseases (NCDs) between East and West in Europe.

My examples illustrated the application of Python's data visualisation tools to public health advocacy, both within WHO as well as in discussion with policy makers in the Member States.

Two phrases that I used seemed to resonate with the audience; some of the participants approached me afterwards to react to them:

1. Code as social organisation—by this I meant that many of the inequities in society have structural causes (poverty, education, access to health services), and code is part of the causal chain in all of these determinants.
2. Code as activism—one would love to live in a world where code is a neutral artefact, but this is far from the truth. Code generates a host of benefits, yet it also empowers weapons, enables surveillance capitalism, and contributes to inequity in society. Can we envisage ways, I asked, where at least code does no harm? Or where it is used more actively to advocate for public good?

The AI-generated report of the poll are summarised in this Annex.

At the end of my presentation, I asked the audience to respond to a very open (and intentionally vague) question: How might Python and its community bring about a vision of an "open source public health operating system"? The answers they gave were summarised by in the annexed AI-generated report.

Let me now pay tribute to other speakers who gave impressive examples of the intersection between Python and population health. The following section summarises key presentations that reflected different ways in which Python can be used in both clinical and public health applications.

Fairlearn: assessing and mitigating harm in AI systems (Tamara Atanasoska)

Weerts H, Dudík M, Edgar R, Jalali A, Lutz R, Madaio M. Fairlearn: Assessing and Improving Fairness of AI Systems. Journal of Machine Learning Research. 2023; 24. (accessed Aug 2, 2025)

Tamara started off by presenting examples of unfair AI systems. They included a sentencing model in the US justice system that was racially biased, and child care applications in the Netherlands that were processed by an AI model biased against migrants. She pointed out that fairness concerns can arise across the whole machine learning lifecycle and thus needs disaggregated evaluation; assessing fairness is a process not a solution. She pointed out that algorithms and data frames sit within decisions made by humans and institutions at all stages in the process.

She explained core concepts in handling fairness, such as: looking out for abstraction traps, attending to construct validity, defining harms (e.g. to quality of service, allocation of resources, and representation), and then bringing all these together with fairness metrics. She introduced Fairlearn, a community-led OSS library, that is scikit-learn compatible and helps integrate fairness into one’s pipeline.

She gave a health example: readmission of people with diabetes released from hospitals in the US. The question: are the aftercare resources allocated fairly by race? The ten-minute tutorial on the Fairlearn site covers this example.

Afterwards Tamara and I chatted about the fact that fairness and privacy are at loggerheads in public health. The privacy laws that protect patients make it impossible to collect the datasets that can emit fairness metrics. Possibly the European Health Data Space could create opportunities to resolve this paradox. Possibly, Fairlearn could be used at facility level to analyse equity of access, provision, and outcomes within each facility's own population and this could be mandated by national or European legislation.

Stopping Epidemics using the Atomica Python Tool (Eloisa Pérez Bennetts)

Sadly, Eloisa’s talk was scheduled in the same time slot as mine, so I was not able to attend it. Her abstract, briefly summarised, said:

atomicateam/atomica. 2025. https://github.com/atomicateam/atomica (accessed Aug 2, 2025).

Infectious diseases kill over 13 million people each year, and finding the best response to outbreaks depends on many shifting factors. Atomica, a free and open-source Python tool, makes it easier to build and analyse disease models that incorporate real-world data, interventions, and budget constraints. [The] talk demonstrated how to use Atomica to simulate a typhoid epidemic and guide funding decisions for maximum public health impact.

Though I could not attend her talk, I was lucky to get a chance to catch up with her in person later and, together with Jannis Lübbe (see below), to discuss her work on Atomica—a tool that has the advantage not just of modelling an outbreak from known characteristics of the pathogen, but also to bring in social factors and resource allocation decisions to guide the response.

Bowring AL, Martin-Hughes R, Brink D ten, et al. Optimising TB investments in Belarus, Moldova, Kyrgyz Republic, Tajikistan and Uzbekistan: An allocative efficiency analysis. PLOS Global Public Health 2025; 5: e0004548.

Later I read her most recent paper: “Optimising TB investments in Belarus, Moldova, Kyrgyz Republic, Tajikistan and Uzbekistan: An allocative efficiency analysis”—discovering that not only were we working in the same region of Europe, but also that her team was bringing these new tools to bear on matters deeply connected with the TB-Free Central Asia Initiative recently launched by WHO/Europe.

WHO/Europe. TB-Free Central Asia Initiative. 2025. (accessed July 27, 2025)

Eloisa, Jannis, and I had a delightful discussion about their projects and noted that we were not exceptional in our shared interest, but that there are many participants at this conference who are applying Python to health and social concerns. I lose track of who of us brought up the idea that this might be a theme to be proposed for an “Open Space” or an explicit thread of presentations in next year’s EuroPython.

Offline Disaster Relief Coordination (Jannis Lübbe)

Technisches Hilfswerk. Wikipedia. 2025. (accessed July 27, 2025).

Somewhere I heard that good design starts with defining the constraints. There could not be a better example of this than Jannis’ project for mapping disaster relief efforts. Germany’s Federal Agency for Technical Relief is a civil protection support agency where 97% of the eighty thousand members are volunteers, like Jannis.

JL’s project assumes a number of realistic constraints facing the staff and helpers during a disaster. They cannot assume that the command and control vehicle has a power supply beyond fuel for its own engine. They must assume they have no internet access in any form. They cannot assume that the vehicle has its own computer (this was not clear whether this was a budget issue or an additional constraint imposed by the fact that the generator is in another vehicle, likely being deployed elsewhere).

The problem: how to accurately map the affected locality using only local resources and how to use that map to deploy assistance wherever it is needed?

JL presented the (real and not infrequent) example of a WWII bomb discovered at a construction site. THW is called in to evacuate the surrounding buildings before controlled detonation. He demonstrated how his Raspberry Pi-powered Python project provides the following features:

1. Local map downloaded from OSM and available fully offline, with resolution down to the house number.
2. Ability to delineate the affected area (by annotating the map using basic shapes e.g. to draw a 1km evacuation circle around the bomb thus defining buildings within and outside the affected area).
3. Ability to locate specific points of intervention (e.g. a marker on a house where residents are refusing to evacuate and where police need to be called in).
4. In the absence of GPS services, the location of teams can be added to the system using a LORAWAN gateway to receive positional information from relief workers.

Jannis and I had many conversations about the utility of such a system in disaster relief and civil protection, including during conflict.

Schafer M, Strohmeier M, Lenders V, Martinovic I, Wilhelm M. Bringing up OpenSky: A large-scale ADS-B sensor network for research. In: IPSN-14 Proceedings of the 13th International Symposium on Information Processing in Sensor Networks. Berlin, Germany: IEEE, 2014: 83–94.

I also learned about Jannis’ other volunteer work with the OpenSky Network, that provides flight data for social impact and academic research. The paper that launched the OpenSky Network has over 250 citations with applications that include studies on environmental impact of aviation, on the modelling of COVID-19, and on the effects of the Ukraine war on global air transportation.

Privacy-Enhancing Cancer Data Processing (Florian Stefan)

Florian introduced the data pipeline for Flatiron Health, a company that provides privacy-enhanced cancer data for health research (usually providing Real World Evidence — RWE — for pharma companies testing novel chemotherapeutic agents).

He described a model of data processing with several strengths:

1. All raw data is owned by the health care facility and resides on their servers.
2. Anonymised data is then transferred to AWS for pre-processing on cloud instances managed by Flatiron.
3. From then on various tools are used to clean, normalise, and harmonise the data before providing access to the client.
4. The data flow involves engineers, data scientists, and clinicians — a division of labour that allows quality control, iteration, experimentation, and innovation.
5. The pipeline allows Flatiron to switch between Snowflake, the standard data provider for the external clients, and a simple notebook interface that permits internal users to experiment on the pipeline, seeking efficiencies, testing quality, and adding features easily.

In chatting with Florian afterwards, I asked how they handled image data (very important for cancer patients, but they could have embedded identifiers). In certain markets, with strict privacy rules, image data is discarded. In the US it is retained. The conflict between privacy rules and population benefit kept emerging throughout these talks.

I asked him about applications of their data in public health research and he pointed out to an intriguing example of breast cancer in men, a cancer that is real enough, but so rare that no company has the incentive to run expensive clinical trials for such a small group of patients. Using Real World Evidence it becomes a possibility to at least investigate the potential impact of existing chemotherapeutic agents for this new application.

Prenatal Diagnosis of Genetic Diseases (Helena Gómez Pozo, Marina Moro López)

HGP and MML, two computational biologists, gave a gentle introduction to genetics and DNA sequencing (explained very elegantly to a lay audience) as the basis for their talk. For context, they stated that 10% of the population suffer from one of 6,000 genetic diseases and pointed out that prenatal diagnosis applies to inherited genetic disorders.

Their talk focused on a project looking for Epidermolysis bullosa, a condition occurring in 1:6000 of newborns. They illustrated their workflow. They began by downloading reference gene sequences from the NIH database. They wrote a script using tkinter to provide a simple GUI making the tool accessible to non-programmers. Their script read off the reference and patient sequences, aligned the two, identified the positions of mutations, and displaying the difference. Effectively a git diff on a strand of DNA.

Apart from introducing the audience to their methods, they made several, highly valid ethical points:

• The tool, even though a “simple” script, is regarded as a medical device and has to be regulated by EU regulations for in vitro medical devices. This has implications for anyone thinking of building a library of open source public health tools...
• The system is very simple for programmers to understand, and even to replicate, but their admonition was “don’t try this at home” as this sort of diagnostic work should only be done in the context of genetic counselling.
• They also counselled the audience to be cautious in the face of overly-optimistic media reports of advances in genetics. They discussed the example of mis-reporting success in the “treatment” of Down’s Syndrome, with reporters getting the science wrong quite dramatically.
• They reminded the audience that the human genome is under the protection of the universal declaration on human rights — one of the safeguards against eugenics.

If I felt any tensions or self-consciousness, they dissipated at the Speakers’ Dinner. Three hours that flowed serenely and imperceptibly in the Manes restaurant overlooking the Vltava River, spanning sunset and early evening. The dinner gives you an easy entry point to conversations, since all the other guests are speakers: you just have to ask what their speech is about and you have a conversation starter. (The image quilt below is sewn from the speaker profile photos on the EuroPython 2025 website).

It was different with Georgi Ker (Board of Directors, Python Software Foundation). Our chat went straight to the theme of community, an aspect of Python that I have observed with interest since my first EuroPython a decade ago in Florence. We discussed three attributes of the community:

• The ubiquity of PyLadies at Python meetings and in the life of the language: I am used to attending public health meetings where gender representation has to be very carefully planned, where vocal criticism of manels (panels composed entirely of men) is frequent, and where the invitation of female speakers that is also geographically diverse is a constant challenge. In Python, the creation of PyLadies has been hugely effective and their leadership in the community is organic and evident everywhere. The PyLadies presence goes well beyond mere gender representation, it brings an inclusiveness to the language itself and I would suggest is one of the reasons for Python’s success, as much as any internal feature.

⊕

In the main conference, later, I learned about the PacMan rule. It goes like this: if there is a circle of people talking in a hallway or over tea and a new person approaches, the circle should open up to include the newcomer. Clever naming, and such an enabler!

• The power and detail of the EuroPython Society code of conduct stands as role model even for conferences outside the Python community. It is detailed, transparent, and strictly enforced. I thoroughly enjoyed reading it, noting the transparency reports that accompany it, the frequent mentions of it during the conference, and the clear reporting lines for incidents. I think it is a model of good practice; I tip my virtual hat to all involved in creating, maintaining, and implementing it.

• A third element: the young median age of the conference. In public health policy conferences, the age of the participants gradually advances with the seniority of the representation. In one sense it is natural that the conferences I am used to attending include older people from the executive levels, and often there has to be a search for young talent to “represent” that age cohort. But this creates the problem of how to select the representatives, and how to go beyond tokenism. In EuroPython, youth are just there, they are the community.

Keynotes

Two keynote speakers had more to say on the theme of community or served as effective role models.

• Python core developers are gods to mere mortals like me. In a modest, humorous, intensely personal talk, Savannah Bailey demystified her journey into becoming a core developer. From triaging issues, to tinkering with documentation, to contributions to the standard library (she focused her energy on argparse), to eventually leveraging her DevOps skills to contributing to the Python JIT compiler, she made it all seem so accessible, so inviting. A place for everyone.

⊕

I couldn’t unsee the fact that the two accented vowels in Sebastián Ramírez’s name are AI. Was this foreshadowing? And is this a stale joke by now?

• Sebastián Ramírez (@tiangolo), developer of FastAPI, a library that has 3 Million downloads... a day, gave forty-five minutes of rapid-fire tips. The sort of talk that’s wasted if you take notes, and lost if you don’t. I just decided to set my notebook aside and enjoy his burst of Colombian energy. He broke stereotypes on innovation. “Solve a problem,” he said, “It’s not about innovation or disruption. Just solve a problem, and innovation follows”. And he gave surprisingly practical advice (“Brush your teeth,” he said, “No, seriously. Brush your teeth. Especially at night. Or they will fall off.”) He did have a lot to say about community including this gem: “Avoid the dark places,” he said, “Hacker News, and Reddit.” (Amazing, I thought. Even @tiangolo gets haters.) He joked that he is more a Markdown programmer than a Python programmer since he emphasises the writing of extensive documentation. He did not so much give a recipe for community, he just stood out as an embodiment of its principles.

Python, the documentary

We thronged to the Forum Hall to watch a sneak preview of The Python Documentary that will be released on cult.repo soon. We watched highlights of two major milestones in the life of Python: the transition from a male-dominated to a more diverse community, and the painful switch from version 2 to version 3 of the language. Language, community. The inescapable double helix of Python’s DNA.

The panel pictured below, Brett Cannon, Armin Ronacher, Paul Everitt (yes, a manel, with one male and one female moderator) exchanged war stories and discussed the future of both Python and its people.

Armin Ronacher’s revolt during the 2-to-3 transition had featured in the video. “Time healed and I came back, of course” he said. Brett Cannon noted how they had underestimated the extent of the existing codebase and the resistance they would encounter in the switch. Paul Everitt remembered how the transition contributed to Zope’s not surviving. All of them expressed the worry they had had at the time, that Python would not survive or risk fracturing the community, as has happened in other languages. They recalled dark times and lessons learned.

They spoke about growing the community, about mentorship, role models, about older members needing to “get out of the way”, about elders giving support to newcomers but consciously doing so in a subordinate role, and about celebrating the achievement that is PyLadies.

Their final note was one of optimism: AI is recruiting a new cohort into the Python community; we should be welcoming them, they concluded. As of course we should.

Savannah Bailey's keynote invited the audience deeper into the heart of the language. Not everyone will be a core developer, but she laid out a host of other ways of getting involved that are accessible to anyone in the audience.

One of them, I believe, is giving a talk on some aspect of Python that you are passionate about or expert in. If anyone reads this with an interest in submitting an abstract for next year, here are some lessons I took away from speakers who packed the room and kept the audience engaged.

If you are not a core developer, or you are not an expert on the frontiers of AI, then you need to work a bit harder 😉 to draw people in:

• Propose a deep dive into a highly focused skill or technique. It doesn't matter if the subject feels niche, only that it is useful or fun.

• Make a promise in the proposal that the audience will go away with something they can replicate, and then deliver on your promise.

• A highly successful technique, that I saw repeated over and over: start with an ordinary script, and layer on features stepwise, revealing new power in the language, the library, or the skill that you are demonstrating.

• Note that you do not need for the theme to be at the cutting edge just that the audience will go away with a deeper practical knowledge, one thing they can try out (even if AI in the title was a huge draw, so were tutorials on concurrency and caching).

• This also applies for more esoteric themes — a talk “slithering through audio data” also filled the room because it held the double promise that the audience would be entertained and that participants would be able “to identify key python libraries for exploring audio datasets, state what outputs we can get from the python libraries, and explain what the outputs tell us about the data”.

• Remember for next time that the majority of participants are first timers at the conference; there is always a fund of beginners (to the community, not necessarily to the language) and there is no shame in catering to their needs.

• Showing code is very useful, but far from essential. It can even be a distraction if there is so much of it that it can’t be read at the back of the room. The best uses of code that I saw were by people who used reveal.js to progressively highlight single lines in a wall of code as they spoke.

• Humour was as prevalent in the successful talks as it was in the community. Perhaps you have heard the adage that you should never crack a joke when public speaking as you are sure to offend someone. But I have also learned over time that clowns are universally funny across cultures, and you can use that to your advantage. You don't need to perform slapstick, but gentle self-deprecation can be both funny and humble. This applies to both the proposal (e.g., a gentle joke to explain how you learned from a mistake) and to the presentation itself. It makes you come across as modest and endearing, especially if the rest of the talk shows how expert or successful you really are at the solution.

In short, it's about making the tool, the language, the community, or the skill that you are presenting more accessible, in bite-sized chunks, with vision, humility, and humour. I just hope that doesn't sound more daunting than inviting.

The lightning talks stunned me. I kept thinking it was time for me to leave, since they came at the end of each intense day, but like a forbidden snack, I kept staying for just one more. Ultra-short presentations with sudden shifts from the emotional, to the weird, to the hilarious. The Lightning Talks serve as a sampler of the community at large, incredibly diverse, incredibly joyful:

• Armin Ronacher, star developer whose tools have been used by every single person in the hall, moved us all: “Python gave me my life and this community made me who I am”, he said, paying tribute to those who had inducted him into the language.
• Then a medley:
  • knitting as a binary pattern, whether knitting kittens or fractals;
  • how to pronounce PyPI (it is Py, P, I, not PyPy, which is something else entirely);
  • weird and wonderful Olympic mascots;
  • building underwater, Python-driven robots;
  • the thriving Python Community in Africa;
  • riding Python-powered e-bikes from Switzerland to Prague (for the conference) and back;
  • And an amazing teen (was he the youngest at the conference?) who set the audience abuzz with applause and hilarity as he showed off his school project: playing the buzzer on an Arduino.

A flash mob of musicians from the audience playing the language anthem: always look on the bright side of life. Photo credit: Gauden Galea

TL;DR for Selected Talks

Here are my summaries of the extensive notes I took from talks I attended, with personal lessons and reactions. Any inaccuracies or errors are, of course, my own.

📦 Caching in Python (Akshita Gupta)

Wow, this was a masterclass in content and style. Akshita Gupta gets my award for "highest bandwidth" presentation: a deep dive into caching in Python with a fully visual set of slides, the most explained with the least words on screen.

AG took us on a journey through caching, from DNS caches in browsers, to the use of functools decorators, to distributed systems built on Redis and Memcached, to JobLib for parallel execution of data-intensive tasks. She layered on the patterns: starting with storing the results of function given a set of inputs; later, looking for the same inputs and retrieving the results; on a miss, computing and storing the new result. She covered setting policies on expiry, eviction, and refreshing of cached entries.

I came away keen to start experimenting with optimising some of my own code. And... am I tempted to try to improve my times in last December's Advent of Code? Baby steps: start with the more routine use of the functools.lru_cache and functools.cache decorators.

🌐 Democratizing Serverless AI Apps (Yuichiro Tachibana)

Yuichiro Tachibana converted a prosaic data processing script step by step into a serverless Streamlit AI app using Pyodide and Stlite, finally layering on transformers.js.py for sentiment analysis on the data set. The result: a fully client-side AI-enabled web app (HTML-only), that preserves privacy, enables easy sharing, and reduces cost.

I had a double-take towards the end of the talk, realising belatedly that YT (whitphx) is himself the developer of both stlite and transformers.js.py, the libraries he was so expertly yet playfully demonstrating. I shall definitely explore these.

📚 PyScript + EduBlocks (Josh Lowe)

Josh Lowe runs EduBlocks, a block-based Python learning site for schoolchildren. He demonstrated how he needs to achieve massive scale (300K monthly users and over a million projects) without incurring any server costs.

His secret: Josh uses PyScript. No backend infrastructure needed; everything happens on the client side.

I have to admit that I tried out PyScript too early after its release and I wasn't impressed. But Josh unveiled a totally different world of possibilities. Amazing! Maybe I won’t be writing another EduBlocks web app, but it would certainly be interesting to try adding PyScript to mini-apps hosted on “static” pages, such as on this blog. Watch this space.

🧠 Generative AI's Promises & Pitfalls (Keynote by Nerea Luis)

Nerea Luis' keynote covered GenAI’s rapid rise, shifting tech roles, and blurring lines of accountability: who can claim to have written the code when AI pair programmers are used? Who takes the blame when AI agents are in the loop?

Her key message to developers: cultivate adaptable AI habits, not tool dependency. Over the next five years, she said, expect significant, unforeseeable change, particularly in areas such as search, information analysis, service consumption, communication, and the development of digital products and projects.

She introduced us (or was that just me?) to two invaluable resources:

1. the OWASP lists for new vulnerabilities (e.g., the OWASP Top 10 for LLM Applications 2025), and
2. the AI CEO letters site for AI dreams at the head of large companies.

⚙️ Async Python Basics (Krzysztof Wilczyński & Mateusz Zaremba)

KW and MZ used a clever metaphor for concurrency: they passed the Python-baton to each other as they took turns speaking. It stood for control flow, as Python shifts from executing to waiting. Loved it. Photo credit: Krzysztof Wilczyński & Mateusz Zaremba.

KW and MZ gave us fast, accessible introduction to Python concurrency: async/await, coroutines, the event loop, and asyncio.gather. They started with the standard library and delved deep into the logic and pitfalls of concurrency. Tools also mentioned: anyio, asyncer.

Finally, I felt I understood the concepts -- I had had threads, processes, and concurrency in a jumble in my mind, and I just avoided all of that. KW and MZ invited me in again, another neglected space to explore.

🐍 Python Performance “Snake Oil” (Sindre Nistad)

With tongue firmly in cheek and sheer virtuosity, Sindre delivered real speed gains in Python leveraging C, Rust, Numba, and others. He first introduced the original Mandelbrot set, plotted in asterisks on a dot matrix printer, then he went on to animate it, step by step, live, at high FPS, zooming rapidly down as far as Python's floating point numbers would allow him.

Sindre has a deadpan comedy style, funny and brilliant, with punchlines in Python. I even loved the snake oil pun for Python performance (geddit?). Not sure there was much Python code left on screen by the end, after all the C, Rust, and Numba took over. Nor that I will be able to emulate any of this any time soon. It was just astounding to watch. It felt like staring into a wind machine, as close to flying as possible.

🤖 Automating the Unloved Project (Sebastian Witowski)

Sebastian told us how he automated a side project (stock-trading signals) with Python, n8n, and Telegram bots. He had started the project to "scratch his own itch" while day-trading during the lockdowns. He discovered that he could package it for others to use, and some people subscribed.

However, he fell out of love with it. He stopped trading and no longer had any use for it. Customer support started taking too much of his time, with little income to show for it. So he automated himself out of it, handing over on-boarding, trial subscriptions, renewal information, and other repetitive tasks to a combination of scripts and n8n flow, with Telegram bots for when his intervention (approval, rejection, exceptions) was needed.

His key lessons to the audience: thinking of a side project? Don't buy a new house yet. And automate early, set boundaries, don’t confuse passive income with no effort, and plan your exit. Candid, and practical.

🛠️ Building & Deploying AI Agents in VS Code (Sharon Xu)

Sharon Xu used Microsoft’s AI Toolkit in VS Code to build and deploy an AI agent locally, live, in real time. She used GitHub Copilot to create both front and back ends, and deployed to the cloud. It took her less than 30 minutes, including a few minutes of projector glitch.

This talk made agents and MCP much more accessible and provided an elegant example of a basic but effective app at the end. SX made it look so effortless, as experts always do, I guess. Added to my list for exploring.

Am I concerned?

Why should I be? After all, did I not witness many signs of energy, growth and support? Is not Python sitting right at the heart of the AI revolution?

The Community, out in full force...

I have written enthusiastically about the diversity, the joy, and the power of the Python Community, and I hope that I have provided ample evidence of the power and versatility of the language itself. Truly the idiom still holds, batteries included, not only in the standard library, but in the vibrant ecosystem and the energy of the people involved. It was amazing to witness the sheer creativity and virtuosity on prominent display at EP2025.

⊕A big vote of thanks goes to the organisers and volunteers seen here in a group photo at the closing ceremony. These people not only organised the formal conference, but also the tutorials, the sprints, the language summits, the "open spaces", and the social events!

With Numerous Supporters...

Thanks are also due to the sponsors. They supported the meeting financially. And their representatives inserted themselves organically into the life of the conference. They deserve our appreciation.

⊕The reps manned the stands, gave their own technical talks, asked attendees to solve Rubik's cubes or to scrape websites in friendly contests, distributed swag, and held recruitment chats with anyone interested.

import this

I remember (not sure how accurately) that David Crystal, in "The Stories of English", partly ascribed the global dominance of the English language to its remarkable capacity for adaptation and incorporation, borrowing vocabulary and structure, surviving by accretion of words and grammar.

I was reminded of that idea during the conference, where Pythonistas rubbed shoulders with Rustaceans, where summits were held on C and Rust and WebAssembly with equal prominence to any of the other, more Pythonic concerns, such as packaging or AI, and where Python felt more solid in the browser, with Pyodide and PyScript. And this is nothing new to Python: C extensions have always been a staple and even NumPy has some legacy Fortran code lurking inside it.

⊕ PEP 20: The Zen of Python.

Maybe, when the Zen of Python was first written, the strength of the language lay not in any of the nineteen precepts but in the very invocation: import this.

So why the niggling concern?

⊕

Source: The PSF has paused our Grants Program. Accessed 17 August 2025.

Soon after we closed the conference in celebration, there was an announcement on the conference Discord channel that the Python Software Foundation has paused its grants program for 2025.

In the announcement, the Board pointed out the massive rise in Python popularity, but they also show this graph and commented:

By contrast, the PSF’s available resources have decreased over the last three years and are projected to decrease again in 2025 due to increasing PyCon US and foundation costs. Much of the growth shown before 2022 in the chart below comes from grants we received that were restricted funds dedicated to specific staffing and other programs. The result is increasing global community activity against a decreasing pool of available funds.

I could not fail to hear familiar echoes of what is happening in WHO. The fall in income. The need to shift whatever donations are received from earmarked grants to flexible funding. The possibility that a global good may be choked from lack of resources. The dissonance between the centrality of a major global good (whether Python or global health), and the drop in funding.

Was 2022 a peak? Or was 2024 an outlier?

Perhaps I am overthinking it. Surely all will be well. Surely companies and countries profiting from the AI boom will step in. Surely the community will not takes its gains for granted and rise to the challenge.

And am I really going to end this celebratory essay with an anti-climax? Am I? If I do, it will only be because I care, because I would be interested to explore what I can do, and because I would like to spur others to act: to give of their time, money, or ideas.

(And anyway, who on Earth reads down to the last paragraph in a long, five-part blog?)

This banner was my first experience of the power of Large Language Models—I believe it was GPT 3 at the time—to generate working code from a hand-drawn sketch. Here is the input I gave it, together with a prompt asking it to produce a `p5.js` sketch following the design.

In late 2023, I vibe-coded (the term did not exist at the time) a Processing sketch to serve as a public banner during webinars. I used it to remind participants that the time to the Fourth UN High Level Meeting on NCDs and Mental Health was growing short and that we had a lot to do before then.

I set up this banner when the count was 100, hence the name, but I publish this post when the number has dwindled to five.

The countdown became almost a marker of my own journey to retirement and life after WHO. I will keep this code repository running and switch the label to mark weeks beyond the High Level Meeting—a tiny voice, a reminder, and a conscience-prick to policy-makers and politicians—I assume they have consciences—mutely asking them to accelerate work on their commitments to 2030.

Here's the Github repo and the banner itself, more a reminder now than a call to action.

I started my coding life in the 80s with a ZX Spectrum and a book on Basic. I was hooked. I've been coding ever since. I've worked in a variety of languages and environments, including Hypertalk, Perl, PHP, Processing, Python, R, and ZX Basic (in alphabetical order, because I loved each of them at different times). These days, my mainstay is Python, a happy coincidence given its current dominance in the field of AI. It also explains why I have chosen to use Lektor for this site.

But I will try any new language or framework, just for the thrill. My hard drive is littered with half-finished projects and experiments. Here are a few of them, in no particular order. In recent months I have been able to take them to a fuller stage of completion with the assistance of AI copilots.

Context

At the end of my talk at EuroPython 2025, I ran a quick brainstorm among the audience in the room. I generated this report on the spot, but there was no time to discuss it on the day. Several promising ideas did emerge, and I share the report here as Annex to my "Health and Society" post.

Acknowledgment

This poll was conducted using the Pera tool (an internal WHO/Europe app) managed and created by Nebi Bekiri and Miguel Guadarrama (I had a hand in its ideation and the MVP creation as did Nuwan Weerasinghe). Sincere gratitude to Nebi for sanctioning this use.

Poll Question

How might Python and its community bring about a vision of an "open source public health operating system"?

LLM Prompt

You are an expert meeting facilitator. You are collecting inputs from a gathering of Python programmers asked to submit their ideas on possible intersections between Python, its community, and the idea of an open source public health operating system. You are requested to summarise basic concepts, opportunities, reservations or cautions. Be professional in tone and style. Limit the report to three pages maximum. Be ready to use Python in-jokes or allusions where useful but do this sparingly (e.g. the "essential functions of public health" as the "standard library" of the discipline).

AI-Generated Report

Introduction

This document summarises the feedback collected from a group of Python programmers on the potential intersections between Python, its community, and the vision of an open source public health operating system. This summary aims to provide a clear and concise overview of the community's insights and suggestions.

Experience with Open Source and Community Management

• Relevance: Several participants emphasised the importance of leveraging the community's experience with open source and community management. This foundational knowledge is crucial for the development and maintenance of an open source public health operating system.

Education and Training

• University Modules: Incorporating public health and statistics modules into university curricula, specifically using Python and tools like Matplotlib, was a recurring theme. This approach aims to build a strong foundation for future contributors.
• Medical Practitioner Training: Training medical practitioners in Python could enhance their ability to analyse and visualise public health data effectively.

Hackathons and Sprints

• Public Health Data Visualisations: Organising hackathons focused on public health data visualisations can foster innovation and collaboration within the community.
• Conference Sprints: Adding sprints to conferences that utilise open health data for various regions or countries can help explore new insights and applications.

Conferences and Summits

• PyCon Summits: Holding specific summits at major PyCon conferences dedicated to public health topics can increase awareness and engagement.
• Involvement of Organizations: Involving organizations like the Python Software Foundation (PSF) in public health discussions at international forums such as the WHO can bridge gaps between technology and health sectors.

Standardization and Tools

• Common Standards: Developing common standards for accessing public health data is crucial. Providing tools that enable the development of useful applications can significantly impact public health initiatives.

Collaboration Platforms

• Marketplace for Groups: Creating a marketplace for groups focusing on specific tasks, where participants can sign up for sprints, can enhance collaboration and project management.
• Discussion Groups: Establishing discussion groups can facilitate continuous dialogue and idea exchange among community members.

Data Privacy and Security

• Sensitive Data: Handling public health data involves significant privacy and security concerns. Ensuring robust data protection measures is essential to maintain trust and compliance with regulations.

Resource Allocation

• Sustainability: Sustaining an open source public health operating system requires substantial resources, including funding, time, and expertise. Ensuring long-term commitment and support is critical.

Inclusivity and Accessibility

• Broad Participation: Ensuring that the initiative is inclusive and accessible to a diverse range of participants, including those from different regions and backgrounds, is vital for its success.

Conclusion

Key areas of focus include leveraging existing open source and community management experience, enhancing education and training, organizing hackathons and sprints, and establishing common standards and collaboration platforms. Addressing reservations around data privacy, resource allocation, and inclusivity will be essential to the initiative's success.

← Back to "Health and Society" post.

In January, 2025, the US Surgeon General released an Advisory on alcohol consumption and cancer risk. He briefly summarised the evidence on alcohol and cancer, including underlying biological mechanisms, and concluded that alcohol use is a leading preventable cause of cancer in the USA and globally, causing around 100 000 and 750 000 cancer cases annually, respectively. The carcinogenicity of alcohol is not a new concept. More than three decades ago, the International Agency for Research on Cancer (IARC) and the Continuous Update Project of the World Cancer Research Fund/American Institute for Cancer Research concluded that there was sufficient evidence that alcohol causes certain cancers; the current list of alcohol-attributable cancer sites published by IARC includes cancers of the oral cavity, oropharynx, hypopharynx, oesophagus (squamous cell carcinoma), colon, rectum, liver, and intra-hepatic bile duct, larynx, and female breast.

Citation

Lancet Public Health. 2025. 10 (5). e358-e359.

Download citation in Medline format

Despite their established effectiveness, uptake of the WHO best buys for tackling non-communicable diseases (NCDs) has been uneven and disappointing. Here we introduce the "quick buys", an evidence-based set of cost-effective interventions with measurable public health impacts within five years. We reviewed 49 interventions previously established as cost-effective (<$I20,000 per disability-adjusted life-year averted) to identify the earliest possible detectable effect on high-level population health targets. Using a strict evidence hierarchy, including Cochrane and systematic reviews, we estimated the effects of each intervention against global targets agreed upon by countries. Quick buys were defined as those interventions that could exhibit measurable effects within 5 years, aligning with average electoral cycles in across the WHO European Region. Of the 49 interventions, 25 qualified as quick buys, including those relating to tobacco (n = 5), alcohol (n = 4), unhealthy diet (n = 3), physical inactivity (n = 1), cardiovascular disease (n = 3), diabetes (n = 4), chronic respiratory disease (n = 1), and cancer (n = 4). These findings not only offer guidance to policymakers deciding on interventions that align with short-term political cycles but also have the potential to accelerate progress to global health targets, particularly the 2030 Sustainable Development Goal of reducing premature NCD mortality by one-third.

Citation

Lancet Reg Health Eur. 2025. 52. 101281.

Download citation in Medline format

This study aimed to evaluate the effectiveness of educational health message and menu default options on a Meal Delivery App (MDA) in nudging consumers towards reduced-salt options in restaurants of China. We conducted a nudge-based intervention trial on an MDA named ELEME in China. Along with a control group, five intervention groups were formed utilizing different combinations of the three nudging treatments, including setting up a salt submenu for "reduced salt" default, a salt submenu for "regular salt" default, and a conventional educational health message on the ordering page. We recruited 903 restaurants from cities across different geographic regions and assigned them into either the control group or one of the five intervention groups. After analyzing 870,942 meal orders, the results showed that the inclusion of a health message on the restaurants' ordering page was not effective to influence consumers to choose reduced-salt dishes (P > 0.1). A salt submenu that explicitly asked for consumers' preference for "reduced salt" or "regular salt" significantly increased consumers' likelihoods of choosing reduced-salt dishes in the unadjusted model (P < 0.1). Applying choice architecture changes as nudge-based interventions on MDAs were proven to be effective to promote consumers to order reduced salt dishes, and we consider these findings to have real-world implications for policymakers, researchers, and the private sectors.

Citation

Public Health. 2025. 242. 250-255.

Download citation in Medline format

Innovation leadership in public health has become increasingly crucial in an era of complex global health challenges and rapid technological advancements. This commentary explores the concept of innovation leadership, its importance in public health, and the key factors that influence its effectiveness. As the world grapples with unprecedented health crises and the need for sustainable solutions, the demand for innovation in public health has intensified. However, the sector often struggles to fully embrace innovation due to various factors, including lack of agility, complex regulatory pathways, incompatible organizational culture, and deficient change management skills. At the heart of these challenges lies the critical need for innovation leadership to create an enabling environment that spurs innovation to accelerate health improvements and sustainable public health impact.

Citation

Eur J Public Health. 2025. 35 (Supplement_2). ii19-ii20.

Download citation in Medline format

Current BMI-based measures of obesity can both underestimate and overestimate adiposity and provide inadequate information about health at the individual level, which undermines medically-sound approaches to health care and policy. This Commission sought to define clinical obesity as a condition of illness that, akin to the notion of chronic disease in other medical specialties, directly results from the effect of excess adiposity on the function of organs and tissues. The specific aim of the Commission was to establish objective criteria for disease diagnosis, aiding clinical decision making and prioritisation of therapeutic interventions and public health strategies. To this end, a group of 58 experts—representing multiple medical specialties and countries—discussed available evidence and participated in a consensus development process. Among these commissioners were people with lived experience of obesity to ensure consideration of patients’ perspectives. The Commission defines obesity as a condition characterised by excess adiposity, with or without abnormal distribution or function of adipose tissue, and with causes that are multifactorial and still incompletely understood. We define clinical obesity as a chronic, systemic illness characterised by alterations in the function of tissues, organs, the entire individual, or a combination thereof, due to excess adiposity. Clinical obesity can lead to severe end-organ damage, causing life-altering and potentially life-threatening complications (eg, heart attack, stroke, and renal failure). We define preclinical obesity as a state of excess adiposity with preserved function of other tissues and organs and a varying, but generally increased, risk of developing clinical obesity and several other non-communicable diseases (eg, type 2 diabetes, cardiovascular disease, certain types of cancer, and mental disorders). Although the risk of mortality and obesity-associated diseases can rise as a continuum across increasing levels of fat mass, we differentiate between preclinical and clinical obesity (ie, health illness) for clinical and policy-related purposes. We recommend that BMI should be used only as a surrogate measure of health risk at a population level, for epidemiological studies, or for screening purposes, rather than as an individual measure of health. Excess adiposity should be confirmed by either direct measurement of body fat, where available, or at least one anthropometric criterion (eg, waist circumference, waist-to-hip ratio, or waist-to-height ratio) in addition to BMI, using validated methods and cutoff points appropriate to age, gender, and ethnicity. In people with very high BMI (ie, >40 kg/m), however, excess adiposity can pragmatically be assumed, and no further confirmation is required. We also recommend that people with confirmed obesity status (ie, excess adiposity with or without abnormal organ or tissue function) should be assessed for clinical obesity. The diagnosis of clinical obesity requires one or both of the following main criteria: evidence of reduced organ or tissue function due to obesity (ie, signs, symptoms, or diagnostic tests showing abnormalities in the function of one or more tissue or organ system); or substantial, age-adjusted limitations of daily activities reflecting the specific effect of obesity on mobility, other basic activities of daily living (eg, bathing, dressing, toileting, continence, and eating), or both. People with clinical obesity should receive timely, evidence-based treatment, with the aim to induce improvement (or remission, when possible) of clinical manifestations of obesity and prevent progression to end-organ damage. People with preclinical obesity should undergo evidence-based health counselling, monitoring of their health status over time, and, when applicable, appropriate intervention to reduce risk of developing clinical obesity and other obesity-related diseases, as appropriate for the level of individual health risk. Policy makers and health authorities should ensure adequate and equitable access to available evidence-based treatments for individuals with clinical obesity, as appropriate for people with a chronic and potentially life-threatening illness. Public health strategies to reduce the incidence and prevalence of obesity at population levels must be based on current scientific evidence, rather than unproven assumptions that blame individual responsibility for the development of obesity. Weight-based bias and stigma are major obstacles in efforts to effectively prevent and treat obesity; health-care professionals and policy makers should receive proper training to address this important issue of obesity. All recommendations presented in this Commission have been agreed with the highest level of consensus among the commissioners (grade of agreement 90–100%) and have been endorsed by 76 organisations worldwide, including scientific societies and patient advocacy groups.

Citation

Lancet Diabetes Endocrinol. 2025. 13 (3). 221-262.

Download citation in Medline format

Despite decades of research on alcohol’s harms, the reality remains striking: the health risks of alcohol are largely misunderstood or downplayed, both in the policy arena and by the public. Alcohol is a leading cause of preventable death in the WHO European Region, responsible for nearly 800,000 deaths annually. The consequences extend far beyond the individual drinker, with societal costs that are staggering. In high-income countries, alcohol use is estimated to cost up to 2.6% of GDP annually due to health care expenses, lost productivity, and increased burdens on justice systems. This makes alcohol’s harms not just a public health problem; it is an economic issue and a matter of social justice.

Citation

Lancet Reg Health Eur. 2025. 48. 101179.

Download citation in Medline format

and colleagues argue that the digital divide may exacerbate global disparities in prevention and control of non-communicable diseases and enlarge health inequity when effective digital health solutions are unevenly distributed

Citation

BMJ. 2024. 387. e076768.

Download citation in Medline format

BMJ. 2024. 387. e076764.

Download citation in Medline format

Alcohol health-warning labels are a policy option that can contribute to the reduction of alcohol-related harms, but their effects and public perception depend on their content and format. Our study aimed to investigate the effect of health warnings on knowledge that alcohol causes cancer, the perceptions of three different message topics (responsible drinking, general health harm of alcohol, and alcohol causing cancer), and the role of images included with the cancer message. In this online survey experiment, distributed in 14 European countries and targeting adults of the legal alcohol-purchase age who consumed alcohol, participants were randomly allocated to one of six label conditions using a pseudorandom number generator stratified by survey language before completing a questionnaire with items measuring knowledge and label perceptions. Effect on knowledge was assessed as a primary outcome by comparing participants who had increased knowledge after exposure to labels with the rest of the sample, for the six label conditions. Label perceptions were compared between label conditions as secondary outcomes. 19 110 participants completed the survey and were eligible for analysis. Our results showed that a third of the participants exposed to the cancer message increased their knowledge of alcohol causing cancer (increase for 1131 [32·5%, 95% CI 29·8 to 35·2] of 3409 participants [weighted percentage] for text-only message; increase for 1096 [33·3%, 30·4 to 36·2] of 3198 [weighted percentage] for message inlcuding pictogram; and increase for 1030 [32·5%, 29·6 to 35·4] of 3242 [weighted percentage] for message including graphic image), compared with an increase for 76 (2·4%, -1·2 to 6·0) of 3018 participants who viewed the control message. Logistic regression showed that cancer messages increased knowledge compared with the control label (odds ratio [OR] 20·20, 95% CI 15·88 to 26·12; OR 21·16, 16·62 to 27·38; OR 20·61, 16·19 to 26·68). Cancer messages had the highest perceived impact and relevance, followed by general health harm and responsibility messages. Text-only and pictogram cancer messages were seen as clear, comprehensive, and acceptable, whereas those including an image of a patient with cancer had lower acceptability and the highest avoidance rating of all the labels. The only identified interaction between perceptions and experimental conditions (with gender) indicated higher comprehensibility and acceptability ratings of cancer labels than responsibility messages and control labels by women, with the results reversed in men. Health warnings are an effective policy option to increase knowledge of alcohol causing cancer, with a generalisable effect across several countries. Europeans consider alcohol health-warning labels to be comprehensible and acceptable, with cancer-specific health warnings having the highest perceived impact and relevance. EU4Health.

Citation

Lancet Public Health. 2024. 9 (7). e470-e480.

Download citation in Medline format

WHO has identified climate change as the greatest health threat of the 21st century. The 2015 Rockefeller Foundation–Lancet Commission found that six of nine planetary boundaries had been breached, including land system change, freshwater depletion, and biodiversity loss, with ocean acidification approaching the boundary. These planetary boundaries interact with climate change and will also pose considerable challenges to health. Also, non-communicable diseases (NCDs), are the leading cause of death globally, responsible for approximately 90% of deaths. Climate change has a range of effects on health, including through exposure to extreme heat, floods, and droughts; air pollution from wildfires; increased risks of a range of infectious diseases and vector-borne diseases; undernutrition; and population displacement. However, the complex relationships between climate change and NCDs are not widely understood. In this Comment, we explore the synergism between climate change and NCDs and also show how climate change mitigation and adaptation actions can also reduce NCD risks.

Citation

Lancet Planet Health. 2024. 8 (7). e430-e431.

Download citation in Medline format

The WHO Western Pacific region bears disproportionate deaths from non-communicable diseases (NCDs), with increased overall NCD proportional mortality over the past two decades. The disease burden of mental health increased, resulting from rapid ageing, enhanced stress, and the COVID-19 pandemic, but it was largely neglected. The highly diverse cultures, religions, political systems, socioeconomic contexts, lifestyles, and environmental factors probably have led to massive disparities across countries in NCD mortality, risk factors, and NCD management. Geographically, East Asia had the lowest NCD mortality whilst Pacific islands had the highest. Economic booms, ageing, nutrition transition, social stress, prevalent tobacco use, and fast-increasing obesity and hyperglycaemia are important drivers of NCDs. Men tended to have more adverse behavioural and metabolic risk factors. Rural residents are catching up with their urban counterparts in metabolic risk factors and conditions. Sustainable strategies tailored to NCD patterns are needed to fight the NCD epidemic and related disparities.

Citation

Lancet Reg Health West Pac. 2024. 43. 100938.

Download citation in Medline format

BMC Public Health. 2024. 24 (1). 74.

Download citation in Medline format

Alcohol increases cancer risk, but less is known about public awareness of this link. This scoping review summarizes recent findings on the public awareness of alcohol as a cancer risk factor in European Union and UK. Four databases (Web of Science, MEDLINE, PsycInfo, CINAHL) were searched for papers containing data on awareness of alcohol as cancer risk factor in EU or UK published between January 2017 and December 2022, and complemented with grey literature searches. In total, 45 studies were included covering 18 EU countries (Austria, Belgium, Cyprus, Czechia, Denmark, Finland, France, Germany, Hungary, Ireland, Italy, Netherlands, Poland, Romania, Slovakia, Slovenia, Spain, Sweden) and UK, presenting data collected between 2009 and 2022. Studies covered general population (17 studied a nationally representative sample), women, health professionals, patients and young people. Awareness of alcohol causing cancer in general was higher and studied more often than awareness of alcohol's impact on specific cancers. Among the EU general population, awareness of the link between alcohol and breast cancer ranged between 10% and 20%, head and neck cancer 15-25%, colorectal and oesophagus cancer 15-45% and liver cancer 40%. Awareness was higher among young people and specialized health professions and lower among women (the latter specifically for the breast cancer). While awareness rates varied depending on the exact question wording, many studies showed low awareness of the alcohol-cancer link, especially for specific types such as breast and colon cancer. Public should be better informed about alcohol consumption-related cancer risk.

Citation

Eur J Public Health. 2023. 33 (6). 1128-1147.

Download citation in Medline format

Chinese urban residents consume more salt from meals prepared outside home than in the past. The purpose of this study is to understand Chinese consumer demand for salt reduction as expressed through their orders on meal delivery apps (MDAs), restaurants' willingness to promote salt reduction, and the extent to which restaurants comply with reduced salt requests. We analyzed consumer comments extracted from 718 restaurants on a Chinese MDA called ELEME for orders made in the July-December 2020 timeframe. A self-designed questionnaire was distributed to the restaurant managers to assess restaurants' attitude towards salt reduction upon signing up for the study, and laboratory validation was conducted to test whether dishes ordered with reduced salt requests by consumers actually contained less salt. A total of 25,982 (0.7%) orders out of 3,630,798 orders contained consumer comments. Of the consumer comments, 40.6% (10,549) were about requests for less salt in dishes. Totally 91.5% of 421 surveyed restaurants showed a willingness to respond to consumers' reduced salt requests. The median sodium content measured in the reduced-salt dishes by the laboratory was significantly lower than that in their regular salt counterparts (P < 0.05). We observed substantial consumer demand for salt reduction while ordering meals on the MDA and that restaurants did, in response, reduce the sodium content in the meals they provided. As meals delivered via MDAs comprise an increasing proportion of outside foods consumed, there is an opportunity for public health experts and policy makers to work with MDAs and restaurants to promote healthier food selections. ChiCTR2100047729.

Citation

BMC Public Health. 2023. 23 (1). 2000.

Download citation in Medline format

The rapidly growing field of digital meal delivery platforms has transformed the out of home (OOH) food environment, presenting both opportunities and challenges for public health. This paper introduces the development and potential of a novel digital platform designed for monitoring the OOH food environment. Drawing on publicly available data from meal delivery applications, this platform provides valuable insights into the landscape of digital food offerings, such as the most common restaurants per region, average caloric content per meal type, and energy value per monetary unit. This research addresses the current void in regulations for this digital environment, particularly around food labeling and provision of nutrition information. Even though the platform has significantly improved our understanding of the digital food ecosystem, it highlights gaps, primarily due to the lack of publicly available individual data and inconsistencies in provided information. Despite these challenges, the proposed digital platform holds considerable promise for better understanding the digital food environment, supporting healthier food choices, and informing future policy interventions aimed at regulating the online food environment. This research advocates for mandatory regulations in the digital food sector to ensure comprehensive, comparable, and transparent nutrition information and equality in access to nutritious foods.

Citation

Nutrients. 2023. 15 (18).

Download citation in Medline format

Digital tools have an important role to play in meeting the health demands of ageing societies. However, current technological design paradigms often marginalize older people. We adopted a lean, user-centred approach to prototype the Avatar for Global Access to Technology for Healthy Ageing (Agatha), an interactive one-stop shop for healthy ageing promotion. Building on this experience, we present a vision for an integrated approach to "digital healthy ageing". Older people consulted predominantly associated "healthy ageing" with disease avoidance. Digital healthy ageing should take a more holistic approach, covering self-care, prevention, and active ageing. It should also consider social determinants of health in old age, including access to information and digital health literacy, as they interact with poverty, education, access to health services and other structural factors. We use this framework to map out key areas of innovation and explore policy priorities and opportunities for innovation practitioners.

Citation

Lancet Reg Health West Pac. 2023. 35. 100649.

Download citation in Medline format

A systematic scoping review of digital contact tracing (DCT) interventions for COVID-19 was conducted to describe the implementation, adoption, use and effectiveness of DCT interventions implemented as part of the COVID-19 response in the Western Pacific Region (WPR). A systematic search identified 341 studies and 128 grey literature sources, of which 18 studies and 41 grey literature sources were included. 17 (46%) WPR countries and areas implemented DCT interventions. Adoption ranged from 14.6% to 92.7% in different adult populations and epidemiological contexts. Trust in authorities, and privacy concerns and beliefs, were the most frequent determinants of adoption and use. Only two studies analysed DCT effectiveness, which showed limited to no effectiveness of DCT interventions in low transmission settings. Overall, there is limited evidence available to evaluate the contribution of DCT to mitigating COVID-19 in the WPR. Preparedness for future health emergencies should include developing robust frameworks for DCT effectiveness evaluations.

Citation

Lancet Reg Health West Pac. 2023. 34. 100647.

Download citation in Medline format

Lancet Reg Health West Pac. 2022. 29. 100621.

Download citation in Medline format

The use of digital health technologies was an integral part to China's early response to coronavirus disease 2019 (COVID-19). Existing literatures have analyzed and discussed implemented digital health innovations from the perspective of technologies, whereas how policy mechanisms contributed to the formulation of the digital health landscape for COVID-19 was overlooked. This study aimed to examine the contexts and key mechanisms in China's rapid mobilization of digital health interventions in response to COVID-19, and to document and share lessons learned. Policy documents were identified and retrieved from government portals and recognized media outlets. Data on digital health interventions were collected through three consecutive surveys administered between 23 January 2020 and 31 March 2020 by China Academy of Information and Communication Technology (CAICT) affiliated to the Ministry of Industry and Information Technology (MIIT). Participants were member companies of the Internet Health alliance established by MIIT and the National Health Commission (NHC) in June 2016. Self-report digital interventions focusing on social and economic recovery were excluded. Two hundred and sixty-six unique digital health interventions meeting our criteria were extracted from 175 narratives on digital health interventions submitted by 116 participating companies. Thematic analysis was conducted to describe the scope and priority of policies advocating for the use of digital health technologies and the implementation pattern of digital health interventions. Data limitations precluded an evaluation of the impact of digital health interventions over a longer time frame. Between January and March 2020, national policy directives promoting the use of digital technologies for the containment of COVID-19 collectively advocated for use cases in emergency planning and preparedness, public health response, and clinical services. Interventions to strengthen clinical services were mentioned more than the other two themes ( = 15, 62.5% (15/24)). Using digital technologies for public health response was mentioned much less than clinical services ( = 5, 20.8% (5/24)). Emergency planning and preparedness was least mentioned ( = 4, 16.7% (4/24)). Interventions in support of clinical services disproportionately favored healthcare facilities in less resource-constraint settings. Digital health interventions shared the same pattern of distribution. More digital health technologies were implemented in clinical services ( = 103, 38.7% (103/266)) than that in public health response ( = 91, 34.2% (91/266)). Emergency planning and preparedness had the least self-reported digital health interventions ( = 72, 27.1% (72/266)). We further identified case studies under each theme in which the wide use of digital health technologies highlighted contextual factors and key enabling mechanisms. The contextual factors and key enabling mechanisms through the use of policy instruments to promote digital health interventions for COVID-19 in China include pathway of policy directives influencing the private sector using a decentralized system, the booming digital health landscape before COVID-19, agility of the public sector in introducing regulatory flexibilities and incentives to mobilize the private sector.

Citation

Intell Med. 2021. 1 (1). 29-36.

Download citation in Medline format

Lancet. 2019. 394 (10204). 1123-1125.

Download citation in Medline format

Distinguished Chancellor,
Distinguished Rector and Pro-Rectors,
Honourable Minister for Education,
Honourable Parliamentary Secretary for Youth and Sports,
Members of the Academic Body,
Fellow Graduands,
Distinguished Guests,
Ladies and Gentlemen,

Download the PDF version of the speech.

I feel privileged and deeply grateful for the honour that you are bestowing upon me today.

Twenty years ago, I joined the World Health Organization. I have worked in the Pacific, in Asia, in Europe, and now head the WHO office in China. I have been privileged in my work to touch the health of hundreds of millions; and the work is its own reward.

Yet the news that my Alma Mater had chosen to award me this honour surprised and moved me to the core. It reminded me that fully half my career has been in Malta and that there was legacy and an abiding link here, f'din l-art ħelwa. Thank you indeed: Rector, Pro-Rectors, Senate, Council, Dr Roberta Sammut for your sponsorship, and Prof Julian Mamo for your oration.

Ladies and gentlemen,

As I was thinking what to say on this podium, Jin Yong passed away last month. Jin Yong was a hugely popular writer in China, and a giant of the wuxia or kung fu chivalry genre. At the recommendation of a friend, I picked up "A Hero Born", one of his novels, recently out in a new translation. Immersed in a fantasy world built around the turbulent years of medieval China, I followed the coming of age of Guo Jing, slow learner, loyal son, reluctant hero, and ultimately valiant warrior, as he acquired his skills under the tutelage of his devoted, if sometimes bungling, shifu, or teachers.

In wuxia literature, one's identity is locked to that of one's teachers. As I read the book, I reflected on how true this is in real life and on how much of one's achievements, whether intellectual or moral, are really due to our shifu. I have been blessed by great teachers, the greatest of whom are my parents, sitting here in the audience, but also the members of the MUSEUM society, the brothers and teachers at De La Salle College, and the academics here, at the University of Malta, where I acquired my medical degree, where I was privileged to serve as the Executive Director of the Institute of Health Care, during its formative years, and where I lectured in public health at the Faculty of Medicine and Surgery.

The list of my teachers would be endless so I limit myself to naming three, because they can only be here in spirit, in the hope that none of the living will feel left out: Prof George Xuereb, the rector under whom I studied, Father Peter Serracino-Inglott, the rector under whom I was appointed to the IHC, and Prof Herbert Gilles, the department head who opened my career up to international health.

As I look back, one other shifu comes to mind. Despite my never having met him, he provided me with a philosophical framework for everything that followed. I have learned his moves and made them mine. Kindly allow me to re-interpret my shifu's lessons for the modern era, as I have come to think of them after two decades of international work.

At the age of sixteen, reading "A Portrait of the Artist as a Young Man" by James Joyce, I was amazed to discover my own biography written decades before I was even born: the hero was another young man, Stephen Daedalus, searching for identity and purpose, struggling with sin and sanctity, living in a politically-divided Catholic society on what was then a relatively poor island nation.

In navigating this world, Stephen learned a set of what I shall describe as magic gongfu moves, and I shall give them imaginary titles. Let me call them: The Three Rejections, The Three Defences, and The Three Aesthetic Criteria.

First, The Three Rejections. James Joyce says:

"You talk to me of nationality, language, religion. I shall try to fly by those nets."

He regards nationality, language, and religion as traps to be avoided. Turning oneself away from country, language, and religion is a difficult move to understand and to learn. It may seem ungrateful or heretic, unpatriotic, even offensive to contemplate in this forum. It takes many years of practice to truly master.

Yet I would argue that the move is easier to understand and even essential to master in these days. Every time a politician proclaims the inherent superiority of one race over another, The Three Rejections are needed. Every time the politics of identity trump the politics of unity, The Three Rejections are needed. Every time membership of a geographic, linguistic, or religious category is used to justify a course of action, then some of our humanity is lost, we chip away at peace and justice, we put our common future in danger, and The Three Rejections grow ever more urgent.

Second, The Three Defences. Joyce describes them with these words:

"I will try to express myself in some mode of life or art as freely as I can and as wholly as I can, using for my defence the only arms I allow myself to use -- silence, exile, and cunning."

Silence grows ever rarer these days. Amid the hubbub of social media, we need to reclaim the space for creativity, for reflection and learning amid the noise of the political and social sphere. We need promote our work, of course, but we also need to let the work speak for itself. We need to value personal worth over personal brands.

Exile, in turn, is an effort to compete and contribute on a larger scale, leaving comfort, safety, and familiarity behind. Exile provides a stage as large as the whole world but affords no back up in case of failure. In our case, exile has been thrilling but also the hardest skill to master. Our departure was voluntary. We left a country that we loved and that has always welcomed us back, but still it was difficult. I am grateful to my wife and children, also here in the audience today, who have unstintingly shared the adventure these past two decades.

Cunning was Joyce's third defence. Cunning is not meant here in a negative sense; it is not manipulation. In my interpretation, cunning, or street wisdom, an internal style of gongfu or neigong, is essential to size up one's surroundings, to anticipate the lateral moves of opponents, to avoid being outmanoeuvred even when outnumbered, to compete in fraught arenas for scarce resources, to fly lightly along informal routes in dense bureaucracies, yet never to lose sight of youth's ideals.

Finally, Joyce, in defining The Three Aesthetic Criteria, adapts a quotation from Thomas Aquinas thus:

"Three things are needed for beauty: wholeness, harmony, and radiance."

In academia and in science, ideals of beauty may be particularly difficult to translate or to master, yet this deeper understanding can also be attained with practice.

Beauty is inextricably bound to truth, in science as in art. The search for truth, the garnering and valuing of empirical facts and forging of the theories that bind them; these are skills that are particularly under threat these days, from political slant or spin, and from publics that need simplified explanations consistent with their pre-existing world view. Much of public health and human development can only be attained with the participation of the whole of society, with harmonious operation between actors and ideologies, and with an eye on the ultimate goal of lasting well-being. When we attain peace, equity, and well-being, that will be truth and beauty.

Among all these moves, the dearest for me is Harmony, because it is particularly difficult to attain in a turbulent world, and because it is essential to fulfilment. I can only wish for all of us gathered here that in our life and work we may ever seek to know and enjoy the magic of Harmony.

Thank you.

Tobacco use is a leading but preventable cause of non-communicable diseases and premature death. The legislature has a key role in setting tobacco control policies. Smoking trends are decreasing thanks to the introduction of effective tobacco control policies in Turkey and these policies may have been shaped by how politicians' interpreted social problems that were prominent during the development and implementation of tobacco regulations. This paper explores the long-term national relationship between tobacco consumption, tobacco control policies and the associated political discourse in Turkey, considering the varying influences through national leadership on this important public health agenda. This relationship is studied by comparing a time series analysis of tobacco consumption trends with a policy analysis of the minutes of deliberations at the Grand National Assembly of Turkey (GNAT). This study uses Bayesian time series analysis in order investigate whether the tobacco control policies and related activities influenced the annual per adult cigarette consumption in Turkey. We used a novel method to identify change points in tobacco trends and whether they correspond with key policy changes intended to alter usage after adjusting for the effect of other non-policy related covariates, such as the purchasing power. The policy analysis included an examination of the minutes of deliberations at the GNAT-which is the Turkish parliament and unicameral Turkish legislature-1 year before and 1 year after the break years associated with an increase or decrease in tobacco consumption. Tobacco consumption increased with the encouragement of tobacco production and the entrance of multinational companies in the country in 1976 and 1993, respectively. The National Tobacco Law of 1996 and comprehensive amendments in 2008, including smoke-free public places and tax increases, appear to have helped reduce tobacco consumption in Turkey. The focus of Parliamentary discussions throughout this period changed, becoming less supportive of tobacco over time. However, throughout the period there remained discussions focussing on concerns around the implications for the economy and the privatization agenda, national agriculture and the welfare of farmers. Effective control appears to require certain political ingredients to be implemented: politicians who are well informed on tobacco control measures and understand the range of issues surrounding the policies (not only those directly health-related); and supportive public health information in the community. Evidence-based public health policy should be introduced to the politicians.

Citation

Eur J Public Health. 2018. 28 (suppl_2). 17-21.

Download citation in Medline format

Eur J Public Health. 2018. 28 (suppl_2). 1-3.

Download citation in Medline format

Lancet Diabetes Endocrinol. 2018. 6 (8). 595-598.

Download citation in Medline format

This study aims to present information on the surveillance, policy developments, and implementation of physical activity policies in the 28 European Union (EU) countries. Data was collected on the implementation of the EU Recommendation on health-enhancing physical activity (HEPA) across sectors. In line with the monitoring framework proposed in the Recommendation, a questionnaire was designed to capture information on 23 physical activity indicators. Of the 27 EU countries that responded to the survey, 22 have implemented actions on more than 10 indicators, four countries have implemented more than 20 indicators, and one country has fully addressed and implemented all of the 23 indicators of the monitoring framework. The data collected under this HEPA monitoring framework provided, for the first time, an overview of the implementation of HEPA-related policies and actions at the national level throughout the EU. Areas that need more investment are the "Senior Citizens" sector followed by the "Work Environment", and the "Environment, Urban Planning, and Public Safety" sectors. This information also enabled comparison of the state of play of HEPA policy implementation between EU Member States and facilitated the exchange of good practices.

Citation

Health Policy. 2018. 122 (5). 519-527.

Download citation in Medline format

To project the prevalence of obesity across the WHO European region and examine whether the WHO target of halting obesity at 2010 levels by 2025 is achievable. BMI data were collected from online databases and the literature. Past and present BMI trends were extrapolated to 2025 using a non-linear categorical regression model fitted to nationally representative survey data. Where only 1 year of data was available, a flat trend was assumed. Where no data were available, proxy country data was used adjusted for demographics. By 2025, obesity is projected to increase in 44 countries. If present trends continue, 33 of the 53 countries are projected to have an obesity prevalence of 20% or more. The highest prevalence is projected for Ireland (43%, 95% confidence interval (CI): 28-58%). Lithuania, Finland, and the Netherlands were each estimated to have an absolute increase of 2 percentage points in the prevalence of obesity between 2015 and 2025. The quality of BMI data across Europe is highly variable, with fewer than 50% of the 53 countries having measured nationally representative data and often not enough data to interpret projections meaningfully. Nevertheless, the prevalence of obesity in the European Region appears to be increasing in most countries and, with it, the health and economic burden of its associated diseases. This paints a concerning picture of the future burden of obesity-related noncommunicable diseases across the region. Greater and continued effort for the implementation of effective preventive policies and interventions is required from governments.

Citation

Obes Facts. 2018. 11 (5). 360-371.

Download citation in Medline format

Neural tube (NT) formation in the spinal region of the mammalian embryo involves a wave of "zippering" that passes down the elongating spinal axis, uniting the neural fold tips in the dorsal midline. Failure of this closure process leads to open spina bifida, a common cause of severe neurologic disability in humans. Here, we combined a tissue-level strain-mapping workflow with laser ablation of live-imaged mouse embryos to investigate the biomechanics of mammalian spinal closure. Ablation of the zippering point at the embryonic dorsal midline causes far-reaching, rapid separation of the elevating neural folds. Strain analysis revealed tissue expansion around the zippering point after ablation, but predominant tissue constriction in the caudal and ventral neural plate zone. This zone is biomechanically coupled to the zippering point by a supracellular F-actin network, which includes an actin cable running along the neural fold tips. Pharmacologic inhibition of F-actin or laser ablation of the cable causes neural fold separation. At the most advanced somite stages, when completion of spinal closure is imminent, the cable forms a continuous ring around the neuropore, and simultaneously, a new caudal-to-rostral zippering point arises. Laser ablation of this new closure initiation point causes neural fold separation, demonstrating its biomechanical activity. Failure of spinal closure in pre-spina bifida mutant embryos is associated with altered tissue biomechanics, as indicated by greater neuropore widening after ablation. Thus, this study identifies biomechanical coupling of the entire region of active spinal neurulation in the mouse embryo as a prerequisite for successful NT closure.

Citation

Proc Natl Acad Sci U S A. 2017. 114 (26). E5177-E5186.

Download citation in Medline format

Int J Epidemiol. 2017. 46 (3). 869.

Download citation in Medline format

Tobacco smoking harms health, so why do people smoke and fail to quit? An explanation originating in behavioural economics suggests a role for time-discounting, which describes how the value of a reward, such as better health, decreases with delay to its receipt. A large number of studies test the relationship of time-discounting with tobacco outcomes but the temporal pattern of this relationship and its variation according to measurement methods remain unclear. We review the association between time-discounting and smoking across (i) the life course, from initiation to cessation, and (ii) diverse discount measures. We identified 69 relevant studies in Web of Science and PubMed. We synthesized findings across methodologies and evaluated discount measures, study quality and cross-disciplinary fertilization. In 44 out of 54 studies, smokers more greatly discounted the future than non-smokers and, in longitudinal studies, higher discounting predicted future smoking. Smokers with lower time-discount rates achieved higher quit rates. Findings were consistent across studies measuring discount rates using hypothetical monetary or cigarette reward scenarios. The methodological quality of the majority of studies was rated as 'moderate' and co-citation analysis revealed an isolation of economics journals and a dearth of studies in public health. There is moderate yet consistent evidence that high time-discounting is a risk factor for smoking and unsuccessful cessation. Policy scenarios assuming a flat rate of population discounting may inadequately capture smokers' perceptions of costs and benefits.

Citation

Int J Epidemiol. 2017. 46 (3). 860-869.

Download citation in Medline format

There is an increasing policy commitment to address the avoidable burdens of unhealthy diet, overweight and obesity. However, to design effective policies, it is important to understand why people make unhealthy dietary choices. Research from behavioural economics suggests a critical role for time discounting, which describes how people's value of a reward, such as better health, decreases with delay to its receipt. We systematically reviewed the literature on the relationship of time discounting with unhealthy diets, overweight and obesity in Web of Science and PubMed. We identified 41 studies that met our inclusion criteria as they examined the association between time discount rates and (i) unhealthy food consumption; (ii) overweight and (iii) response to dietary and weight loss interventions. Nineteen out of 25 cross-sectional studies found time discount rates positively associated with overweight, obesity and unhealthy diets. Experimental studies indicated that lower time discounting was associated with greater weight loss. Findings varied by how time discount rates were measured; stronger results were observed for food than monetary-based measurements. Network co-citation analysis revealed a concentration of research in nutrition journals. Overall, there is moderate evidence that high time discounting is a significant risk factor for unhealthy diets, overweight and obesity and may serve as an important target for intervention. © 2016 The Authors Obesity Reviews published by John Wiley & Sons Ltd on behalf of International Association for the Study of Obesity (IASO).

Citation

Obes Rev. 2016. 17 (9). 810-9.

Download citation in Medline format

This paper presents a typology of available evidence to inform physical activity policy. It aims to refine the distinction between three types of evidence relating to physical activity and to compare these types for the purpose of clarifying potential research gaps. A scoping review explored the extent, range and nature of three types of physical activity-related evidence available in reviews: (I) health outcomes/risk factors, (II) interventions and (III) policy-making. A six-step qualitative, iterative process with expert consultation guided data coding and analysis in EPPI Reviewer 4. 856 Type I reviews, 350 Type II reviews and 40 Type III reviews were identified. Type I reviews heavily focused on obesity issues (18 %). Reviews of a systematic nature were more prominent in the Type II (>50 %). Type III reviews tended to conflate research about policy intervention effectiveness and research about policymaking processes. The majority of reviews came from the United States, United Kingdom, Australia and Canada. Although evidence gaps exist regarding evidence Types I and II, the most prominent gap regards Type III, i.e. research pertaining to physical activity policymaking. The findings presented herein will be used to inform physical activity policy development and future research.

Citation

Int J Public Health. 2016. 61 (5). 553-63.

Download citation in Medline format