I have worked in healthcare for six years and I have always been interested in building my knowledge and keeping up to date with the latest information in my field. The problem is finding the time to do it properly.

Anyone who has done any kind of research knows how time consuming it is. Searching for articles, clicking through different sources, reading through information that overlaps, trying to make sense of it all. And after a long day managing a full caseload, back to back meetings, MDTs, and everything else that comes with the role — your brain has very little left to give.

CPD is encouraged by every Trust and every lead I have worked with. But finding the time has not got easier as I have progressed. If anything it has got harder. Going from Band 5 to Band 7 has meant more responsibility, not more time.

Perplexity fixes that problem. It does the searching, the reading, and the summarising for you and gives you the sources so you can go deeper when you have the time and headspace for it.

What Is Perplexity

Perplexity is a tool many people will not have heard of or used. But in my opinion it is one of the best AI search tools available when you want verified information.

Most AI tools generate answers from what they already know. Perplexity works differently. It searches the web first, finds the most current and up to date information, and then summarises everything in a clear and digestible way.

What makes it stand out is that it does not just give you the answer — it gives you the sources and citations alongside it so you can review and verify the information yourself. That is why it is my preferred tool for research. I know the information is coming directly from the web and not being generated from outdated training data.

Setting It Up

Perplexity is available as a free app on iOS and Android or at perplexity.ai on any browser. You will need to create a free account. That is it. No technical setup. No IT approval needed for personal use.

The interface is simple. A search bar at the bottom and your chat history on the left. Tap the plus sign at the bottom and you will see options including Deep Research which is a more powerful search available on the premium plan. For everyday clinical queries the standard free version is more than good enough.

[Perplexity home screen showing search bar]

[Options menu showing Deep Research]

Why Not Just Use Google

I know what you are thinking. I can just Google this.

You can. But the answer is speed and quality. You can find the same information on Google but it takes longer. In a clinical setting where you have a caseload, meetings, CPD, and everything else on your plate — every minute you can save counts.

Google gives you a list of links and you have to click through each one, read the page and make sense of it yourself. Perplexity does that work for you and shows you the sources so you can verify them.

I put the same question into Google. The AI Overview at the top referenced NG56. NG249 — the most recently updated guideline — only appeared after tapping Show More and scrolling down. Some people never get that far.

[First Google answer]

[After you click show more]

Google is not going anywhere. This is not about replacing it. It is about having a faster option when looking for current and verified information.

What Happens When You Search

I typed a simple question. What is the latest NICE guidelines on managing frailty in the community.

Once you hit search you will see the word Thinking appear. That is Perplexity actively searching the web in real time — not generating an answer from what it already knows. It is going out and finding the information before it responds.

[Thinking screen]

Once it has finished searching it gives you a brief explanation of where it sourced its answers from before summarising all the findings in plain English.

[Overview of answer with inline source references]

At the bottom it provides a key takeaways section. If you are short on time you can scroll straight to that and get the essential points without reading the full breakdown.

[Key takeaways section]

Below that you will find all the sources it drew from. Click any one of them and you go directly to the original page to verify the information yourself.

[Sources section]

The Finding That Stopped Me

II wanted to test Perplexity on something real. I had been meaning to check whether there were any recent updates to how frailty is managed in the community. Simple enough question.

What is the latest NICE guidelines on managing frailty in the community.

No instructions. No special formatting. Just a plain question the same way you would ask a colleague.

Within seconds Perplexity returned NG249. A guideline NICE published in April 2025. The summary positioned it within older adult care including frailty practice. I had not seen it before. And I am not sure many other clinicians had either.

[PERPLEXITY RESULT SHOWING NG249]

But here is where it got interesting.

I clicked through to the source. NG249 on the NICE website is titled Falls: assessment and prevention in older people and in people 50 and over at higher risk. It is a falls guideline. Not a frailty guideline.

Now the two overlap a lot. Frail older adults fall. Falls assessment sits within comprehensive geriatric assessment. The clinical populations are largely the same. Which is why Perplexity surfaced it for a frailty query. But the labelling matters. If I had taken the summary at face value and passed it on as the latest frailty update I would have been wrong. And anyone following that framing would have been looking in the wrong place.

This is exactly why source verification is the whole game.

Perplexity found me a useful guideline I did not know existed. It also framed it in a way that did not quite match what NICE actually published. Both of those things are true at the same time. That is not a reason to stop using the tool. It is a reason to understand what the tool is actually doing.

NG249 is worth your time. The 2025 update replaced guidance from 2013 and falls in older adults is a significant area of community OT and frailty adjacent practice. But I would have introduced it incorrectly without clicking through to NICE first.

The tool finds the information. The verification is where your clinical judgement starts.

The Limitations

Perplexity pulls its information from the open web which means it can draw from low quality or unreliable sources alongside reputable ones. You cannot rely on it without checking where the information is coming from. That is why the citations matter — always look at the sources before acting on anything it tells you.

The depth of the answer is also dependent on the quality of your prompt. A vague question gets a surface level answer. The more specific and clear your prompt the more detailed and useful the response.

And because Perplexity is built as a search tool it struggles with reasoning and strategy. It will find the information you need but it will not help you decide what to do with it. It needs to be used alongside other tools rather than as a standalone solution. More on that next week.

The best way to think about Perplexity is as a research assistant. Use it to find and verify information. Do not rely on it to make decisions for you.

Where To Start This Week

Do not try to overhaul how you research overnight. Start small.

Pick one thing you have been meaning to look into but have not had the time. A condition you see regularly but want to understand better. A guideline you know exists but have never read properly. A topic that came up in an MDT and you meant to follow up on.

Type it into Perplexity as a plain question. See what comes back. Check the sources. That one action takes less than five minutes and will show you more about what this tool can do than anything I have written here.

If you want to go deeper use one of the prompts below.

Two Prompts Worth Saving

Perplexity is best used as a research assistant. And like any tool the quality of what you get out depends on the quality of what you put in. A vague question gets a surface level answer. A well structured prompt gets something genuinely useful.

The two prompts below are designed to help you get the most out of Perplexity for clinical research — and to make sure the information comes back in a format that is clear, sourced, and easy to act on.

Prompt One — Literature Review

Best used in Perplexity

Act as a PhD researcher in healthcare. I need a detailed literature review on [topic]. Include key theories and frameworks, major studies from the last 5 years, alternative viewpoints, research gaps and citations.

Prompt Two — Clinical Guidance Summary

Best used in Perplexity

Summarise current UK and NICE guidance for [condition]. Present it as a concise structured explanation covering management, key red flags, recommended investigations and when escalation is required. Keep it brief, clinically practical and include direct source links.

Opinion

Most clinicians are not bad at finding information. They are bad at finding the right information quickly enough to act on it.

The NHS runs on guidelines, evidence, and protocols. The expectation is that clinicians stay current, apply the latest evidence, and make decisions that reflect best practice. What nobody accounts for is the time that takes. Reading a full NICE guideline update takes an hour you do not have. Searching through Google results and clicking through pages takes time you do not have. Making sense of evidence from multiple sources takes time you do not have.

NG249 is one of the most operationally important NICE updates of 2025 for community practice with older adults. A clinician who missed it is not negligent. They are just busy.

Tools like Perplexity remove one specific friction — the gap between you and the current evidence you need to do your job properly. They do not remove the need for clinical judgement. The example earlier in this issue makes that clear. The tool surfaced a real guideline and described it in a way that did not quite match the source. The clinician’s job is still to click through, read the actual guidance, and decide what it means for their service.

That is the right relationship between the clinician and the tool. The tool finds. The clinician verifies. Together they go faster than either could alone.
The tools exist. They are free. They are available on your phone right now.

The question is not whether you have time to use them. It is whether you have time not to.

IG Guidance

Perplexity is a search tool. You are asking it questions about guidelines, research, and services — not uploading patient information. That keeps it well within the standing rule for this newsletter.

Nothing about a real patient goes in. Not a name. Not an NHS number. Not the combination of details that could identify someone even after the obvious markers are removed.

Perplexity is a consumer product. It has not been through NHS procurement or assessed against DCB0129 or DCB0160. Until your IG lead tells you otherwise treat it accordingly. For research, CPD, looking up guidelines, finding services in your area — these are the use cases the tool is built for and they sit well within reasonable use.

If you ever find yourself including patient details in a search query — stop. There is no version of that query that is safe to run.

If you are ever unsure ask your IG lead before you search.

In Case You Missed It

NHS App AI Triage — Over one million patients can now book GP appointments directly through the NHS App using an AI triage tool called Smart Triage. Patients describe their symptoms and receive an instant triage to the right appointment without waiting for a receptionist or doctor to review the request. Source: Digital Health, April 2026.

NHS AI Scribing Expanding — Buckinghamshire Healthcare NHS Trust has signed a contract with System C to embed ambient voice technology directly into their electronic patient record. The tool listens to consultations, automatically generates clinic letters, and completes outcome forms. It goes live in outpatient services in summer 2026. Source: Health Tech World, April 2026.

MHRA AI Regulation — Public consultation findings from over 770 responses show strong support for reform of AI regulation in healthcare but not a complete overhaul. Safety, oversight, and liability are the main concerns raised by clinicians, patients, and industry. Recommendations are expected later this year. Source: Health Tech Newspaper, April 2026.

That is all for Issue 03. Every week I will bring you something practical you can use and a view on where this space is heading. Next week I am going to show you how to use these tools together — and which one to reach for depending on what you actually need.

Clinically Intelligent drops every Wednesday. If you are not yet subscribed you can join here. If someone you know would find this useful, pass it on.

Updated 28/04/26. The information governance framework in this issue has been refined since first publication. The current standing framework is published in every issue from this point onwards.

Keep reading