.webp)
Gen AI for medical writers is here to stay
(and it’s not as scary as we thought)
The buzz around generative (Gen) AI in medical communications, specifically medical writing, often comes with mixed feelings. “Will there still be a place for medical writers in the future?”, “Gen AI will take away the grunt work”, (followed inevitably by, “How will medical writers ever learn anything now?”), and “How should I integrate AI into my day-to-day as a medical writer?” are all common concerns.
It's easy to worry about the impact that Gen AI will have on medical writing. But when you remember that the role of a medical writer extends beyond medical writing, you quickly realise that Gen AI can offer so much more than role automation.
Amidst endless speculation about what impact Gen AI could have, I’d like to share my own experiences with it so far. I’ll describe how Gen AI has already changed my approach to medical writing, offer tips to get the most out of Gen AI, and explore the potential of this fledgling writer-chatbot partnership.
How does Gen AI work?
The world of AI is complex, so for simplicity I’ll focus on GPT-4 here. It’s like a highly intuitive guessing game player. It predicts the next most likely word or sequence of words based on the context it's given, drawing from a huge dataset that includes a wide array of language use across books, websites, and articles. Gen AI doesn't have understanding or awareness, but simply looks at patterns in data to generate text. This sometimes mimics reasoning or knowledge retrieval, but it is essentially based on statistical probabilities.
With this in mind, I see Gen AI as a sophisticated tool that enhances creativity and productivity but doesn't replace the nuanced understanding and judgment that medical writers bring to their work.
Gen AI empowers writers. It doesn’t replace human insight.
How did I learn how to use AI as a medical writer?
Joining the team at Camino has been a great catalyst for me. I’m lucky to be surrounded by people with diverse skillsets who celebrate agility and an openness to new technologies. I’m sure most of you have felt the impact of cumbersome AI policies, steering committees, or even outright bans that hinder innovation. At Camino, we are encouraged to experiment with Gen AI, share insights, and collectively push the boundaries of what's possible in medical communications while keeping within the limits of compliance. Working with Gen AI at Camino has contributed to my personal and professional development, keeping me ahead of the curve as a medical writer, something that I really value right now considering the uncertainty that our industry is facing. That’s not to say that this couldn’t be achieved in any environment, I just don’t think I would have become as familiar with Gen AI as quickly as I have.
When I think back, there have been two key learning points for me over the past few months.
Firstly, I designed an AI-augmented literature review, and it was a perfect first project to experience the way Gen AI can support medical writing. A mix of process, content analysis and content creation allowed me to work out what was possible, or not, with GPT-4. Reviewing AI outputs for countless hours, understanding that the model boasts a 90% accuracy rate (as per the MMLU benchmark), allows me to appreciate what such a level of precision means in the context of a 1,500-article literature search. Not only did this project help me to understand where Gen AI’s work ends and a writer’s begins, but it also set a new benchmark for efficiency and thoroughness in our literature review process. The impact on our clients is significant, as we can offer them a more comprehensive and nuanced analysis in a fraction of the time compared to traditional approaches. We can offer both breadth and depth, and because the human isn’t bogged down in the detail, it’s much easier to translate a trend into a strategic insight. Plus I now have a healthy scepticism over claims that Gen AI can be left alone to conduct a review of 200,000 articles over a lunch break, as was claimed in a promotional video of Google’s competing AI bot.
The second pivotal moment was when I was tasked with demonstrating Gen AI's capabilities during a live demonstration.
This required not just a theoretical understanding of Gen AI's potential but also the confidence to showcase it in front of clients (and we all know how erratic ChatGPT can be). After extensive testing, I decided to showcase how Gen AI can support with brand analysis and omnichannel planning. The demonstration was strangely exhilarating, and it was then that I really understood the how Gen AI could help empower writers. For our clients, aside from feeling “mind blown”, it illustrated how AI can transform their marketing strategies and content creation processes.
These experiences were eye-opening for me, as I started to see the potential of AI extending way beyond task automation for writers. It’s a way to make my writing and myself more efficient, more innovative, and more valuable to my team and to my clients. And if you’re worried that it could be difficult for medical writers to learn the job in an AI world, I’ve used ChatGPT to teach myself more in the last few months than in the last few years. I really wish I had it when I was starting out as a medical writer!
I have switched from a mindset of “AI could”, to “AI does”. It was a catalyst for me to bring Gen AI into my day-to-day and to share what I’ve learnt with my team.
What lessons have I learnt about Gen AI in medical writing?
Content creation: Rapidly generate first drafts, freeing up time for nuanced editing and refinement
Content personalisation: Discover creative ways to present information, ensuring it appeals to different audiences or even personalities
Analysis and interpretation: Condense complex information into accessible formats, which is especially beneficial for quickly mastering new therapy areas. However, always check the accuracy of AI-generated summaries!
Quality control: Use AI as an initial tool for error checking and receiving suggestions for improvements, though it's not a substitute for a human writer or editor
Brainstorming partner: Leverage AI to brainstorm ideas, which you can take to your team discussions
Process optimisation: Gain insights on enhancing project and writing processes
Coaching: Consult AI on how to tackle new challenges (there’s a lot of useful guidance from different industries in its vast database)
Motivation: AI can be a great motivator for your productivity and the creative process when you're feeling stuck – just prompt to get you going
Final words
My experience with Gen AI in medical writing over the past few months really has been transformative, enabling me to push my own boundaries and redefine what's possible not only for myself, but for my team and my clients by efficiently delivering innovative and valuable content. There came a point where thinking “it could” just wasn’t enough anymore, and I just had to get started.
Looking ahead, the road for Camino is one of continued innovation to remain agile and responsive to the changing environment. We are in the process of integrating AI across our work and internal processes, constantly testing and adapting as the technology evolves. As we move forwards, our commitment to leveraging AI in creative and impactful ways will ensure we are ready to meet the challenges of tomorrow.
You may also be interested in
Omnichannel-native: A mindset shift to deliver effective healthcare communication
A brief history of omnichannel within pharma communications, from disconnected digital tactics to the omnichannel-native mindset.
Agility: How to embrace change
Everything changes, everything stays the same. Does this old adage still ring true?
