Search

Trip Database Blog

Liberating the literature

Open Search
Author

jrbtrip

Thinking Out Loud – Emergency Departments as Systems

We noticed a clump of Qs around emergency departments (ED) systems, workforce, and patient experience. And following on from the approach we used yesterday, we analysed and expanded these into wider topics

1. Reducing Demand / Preventing Admissions

Existing Qs:

  • Effectiveness of community-based models of emergency care.
  • Best strategies to prevent unnecessary admissions.
  • Effectiveness of advance care planning in preventing admissions.

Potential additional Qs:

  • What is the impact of urgent care centres, walk-in clinics, and out-of-hours GP services on ED demand?
  • How effective are paramedic-led interventions (e.g., treat-and-refer pathways, community paramedicine) in reducing ED conveyance?
  • Do public education campaigns (on appropriate ED use) reduce unnecessary visits?
  • What role do integrated care systems (linking primary, community, and social care) play in reducing ED demand?
  • What is the cost-effectiveness of these demand-reduction strategies?

2. Workforce & Staffing in EDs

Existing Qs:

  • Impact of ED layout on staffing levels.
  • Frameworks for ensuring safe nursing staffing.
  • International evidence base for safe staffing.

Potential additional Qs:

  • How do staff-to-patient ratios correlate with patient safety outcomes in EDs?
  • What is the impact of skill mix (nurses, nurse practitioners, physician associates, consultants) on ED performance and safety?
  • How does burnout and turnover among ED nurses and physicians affect patient outcomes?
  • What is the evidence for flexible staffing models (e.g., surge staffing during peaks) in maintaining safety?
  • How does the physical environment (e.g., single rooms vs. open bays, digital monitoring systems) influence staff workload and efficiency?

3. Stakeholder Perspectives & Communication

Existing Qs:

  • Perceived role of EDs among the public, professionals, and policymakers.
  • Interventions to improve nurse–family communication in paediatric EDs.

Potential additional Qs:

  • How do patients with frequent ED use perceive the role of emergency departments?
  • What is the impact of shared decision-making tools on communication and satisfaction in the ED?
  • How do cultural and language barriers affect communication and outcomes in ED settings?
  • What interventions improve staff–patient communication in high-stress environments (e.g., triage, resus)?
  • How do media portrayals of EDs shape public expectations and demand?

Bringing these strands together, what stands out is just how multi-dimensional the evidence needs to be. Emergency departments are not only clinical environments but also systems under pressure, workplaces with unique staffing challenges, and touchpoints where public expectations, professional realities, and policy goals all collide.

By clustering the questions in this way, we can start to see where the gaps lie: for example, plenty is known about demand reduction through community models, but far less about the cultural narratives that shape how people view and use EDs. Likewise, staffing frameworks exist, but how they interact with design, technology, and wellbeing is less clear.

This sort of mapping doesn’t provide the answers, but it does highlight the terrain — showing where a stronger evidence base could make the biggest difference to practice and policy.

We’ll continue to explore these clusters in future posts. In the meantime, we’d love to hear from readers: which of these areas feels most pressing in your context? And are there other questions you’d add to the mix?

0

Thinking out loud – stroke Q&A clusters

In recent AskTrip activity we’re seeing clusters of related Q&As around stroke. These clusters may reflect how evidence is used in practice. We tried mapping the questions along a stroke care continuum – from Acute & Emergency through Secondary Prevention to Rehabilitation & Recovery – and then added logical “next questions” we haven’t been asked yet.

So, below are a list of Qs, those with a hyperlink have been asked already and suggested Qs are listed as ‘supplementary’.

As mentioned in the title this is a ‘thinking out loud’ post – seeing what things look like. It’s helpful to air these ideas…. One can see issues immediately, for instance ‘What are the most effective secondary prevention strategies for reducing stroke recurrence?‘ and ‘What are the current best practices for managing patients with a history of stroke to prevent recurrence?‘ are very similar in scope. But that’s the nature of posting this sort of thing – helps you highlight the issues.

While there are still some rough edges, the bottom-up nature of this approach feels refreshing. I can’t help but wonder: might this become part of Trip/AskTrip’s future?

A structured map of evidence questions from acute care to recovery

1. Acute & Emergency Management

2. Secondary Prevention (Reducing Recurrence)

3. Rehabilitation & Recovery

Cross-cutting Priorities

  • Supplementary questions:
    • How should patients and caregivers be educated about stroke warning signs and secondary prevention?
    • What are the cost-effective models of long-term follow-up in primary vs. specialist care?
    • How can access to stroke rehab services be improved in underserved populations?

1

113 Questions in a Day: What Clinicians Are Asking on AskTrip

Yesterday was a momentous day for AskTrip. We recorded the highest number of clinical questions ever asked in a single day – 113 in total.

That’s 113 moments where a health professional turned to AskTrip for support: to check a management decision, clarify a diagnosis, weigh risks and benefits, or simply explore the evidence behind a difficult case. To mark the occasion, we took a closer look at what those questions were about – and what they reveal about the daily reality of medicine.

The Constant Search for Better Treatment

It’s no surprise that most questions revolved around treatment and therapeutics. Clinicians want to know: What’s the best, safest option for my patient?

  • Should methotrexate be taken at a particular time of day?
  • Is aspirin a valid long-term option after anticoagulation for pulmonary embolism?
  • What are the benefits and risks of SGLT2 inhibitors in elderly patients with diabetes and heart failure?

These queries show not just an appetite for the latest trials and guidelines, but also a desire to tailor care to unique patient circumstances — like whether stenting is safe in someone with a nickel allergy or why bile acids might be elevated after a cholecystectomy when bilirubin is normal.

Surgery: When to Cut, and How to Do It Better

Surgery questions revealed two strands of curiosity: when to intervene and how to do it better.

  • Should a neonatal hernia be repaired early, and if so, when?
  • Should proximal humerus fractures be managed surgically or non-surgically?
  • How does robotic prostate surgery compare to conventional approaches in cost and outcomes?
  • Is transoral thyroidectomy a safer, less invasive option than open surgery?

These questions highlight a thoughtful balancing of risks and benefits, as well as a hunger for innovations that promise quicker recovery and fewer complications.

The Rise of “Prehab” and Non-Drug Strategies

One of the strongest clusters was around prehabilitation — preparing patients physically and mentally before major interventions like surgery or CAR-T therapy.

  • What is the evidence for prehabilitation in thoracic surgery?
  • How does it affect recovery after esophagectomy or cancer treatment?

This shows a shift from reactive medicine to proactive strengthening, where the goal is not just survival but resilience and long-term outcomes.

Other questions highlighted rehabilitation and lifestyle: the role of exercise in chronic fatigue syndrome, the best exercises for thumb extension, and safe activity for patients with PICC lines. These aren’t about treating disease alone, but about restoring function and quality of life.

Complications and Safety First

Again and again, clinicians asked not just “Does it work?” but “What could go wrong?”

  • Can proton pump inhibitors cause myalgia?
  • Is intracameral cefuroxime safe in penicillin-allergic patients?
  • What complications occur after augmentation mastopexy or breast reduction?

This emphasis on adverse effects shows how safety considerations shape clinical decisions as much as effectiveness.

Beyond the Bedside

Not all questions were about patient management. Some reached into the systems that underpin healthcare:

  • How does plagiarism in nursing programs impact education quality?
  • What are the benefits of grounded theory research in healthcare?
  • Does using a template reduce variation in nursing records?

These reflect a broader concern with the integrity of training, the quality of evidence, and the consistency of documentation.

Children and Adolescents in Focus

Children and young people also featured prominently:

  • Do sleep disorders contribute to anxiety and depression?
  • What’s the evidence for scoliosis screening in Europe?
  • Are team sports or meditation beneficial for children?
  • How should screen time be limited?

These queries show clinicians thinking beyond immediate symptoms, grappling with prevention, wellbeing, and the challenges of modern childhood.

Non-Pharmacological Interventions

About one in six questions were not about drugs at all, but about lifestyle, rehabilitation, or supportive care.

  • What lifestyle changes can slow cognitive decline?
  • How should screen time be managed in children?
  • What is the role of exercise in chronic fatigue syndrome?
  • What are the most effective prehabilitation interventions before surgery?

This cluster shows a strong appetite for evidence beyond prescribing — emphasising prevention, recovery, and wellbeing.

What Stands Out from 113 Questions

Looking across the day’s record activity, three things stand out:

  1. Breadth of curiosity – From thumb exercises to the global burden of dementia, clinicians are asking at every scale.
  2. Safety vs efficacy – Many questions probed not “does it work?” but “is it safe?”
  3. System-level thinking – Alongside bedside care, clinicians are worried about education, documentation, and societal health.

Why This Matters

Guidelines and textbooks provide frameworks, but frontline clinicians constantly face edge cases, overlaps, and grey zones. The 113 questions asked yesterday show where evidence support is most needed — in diabetes, dementia, oncology, paediatrics, and in the systems that support safe care.

Closing Thought

Clinical questions aren’t abstract. They emerge from real patients, puzzling scans, unexpected complications, and the human urge to do better. Yesterday’s record-breaking 113 questions are more than just a number — they’re a window into the everyday challenges of healthcare, and a reminder that curiosity is alive and well in medicine.

At AskTrip, we’re proud to help clinicians find answers to those questions — big and small — that matter most to their patients.

0

Hallucinations in AskTrip – Let’s Be Honest About Them

At AskTrip, we’ve always believed that transparency builds trust. That’s why I want to talk about something that’s getting a lot of attention in the world of AI: hallucinations.

What are hallucinations?

In simple terms, hallucinations are when a large language model (LLM) generates something that sounds convincing but isn’t entirely accurate. These models are incredibly powerful, but they don’t “understand” in the way humans do. Most of the time this works brilliantly, but sometimes it can slip.

How we keep an eye on quality

We don’t just leave this to chance. AskTrip has an active quality control system in place that monitors for hallucinations and other errors. We log, track, and learn from every issue that we find. On top of that, we’re finalising a test bed – a safe environment where we can trial new methods specifically aimed at reducing hallucinations – and we’re doing this in collaboration with AI experts.

The kinds of hallucinations we’ve seen

Being upfront means sharing real examples. Here are three patterns we’ve spotted:

  1. Condition mismatch – A paper was returned as though it was relevant to one condition, but in fact, it wasn’t.
  2. Inserted numbers – The LLM provided a recovery figure. The number itself was correct (from the paper), but the way it was presented made it look like it came from somewhere else.
  3. Inference over quotation – Not quite a hallucination, but worth noting. Sometimes the LLM infers from a study instead of sticking strictly to the words on the page.

How often does this happen?

Thankfully, not very often. Importantly, none so far have drastically changed the clinical answer — but even minor inaccuracies can matter in a clinical setting. That’s why we take this so seriously, and why it’s equally important that users uphold their responsibility too. This is why we require all users to agree to a responsibility statement, which includes checking the facts and applying their own critical judgement.

What we’re doing about it

We’re working hard to make AskTrip even more reliable. That means:

  • Partnering with AI experts.
  • Stress-testing new approaches in our test bed.
  • Constantly monitoring, learning, and refining.

Why this matters for you

As a user, it’s important you know that hallucinations can happen. We’ll always be open about this. The frequency is low, we’re actively addressing it, and improvements are underway. But awareness is part of safe use – just as it is with any evidence-based tool.

Pulling it all together

So here’s the bottom line: hallucinations exist. We’re aware of them. We’re working hard to reduce them. And we want you, our users, to be aware too.

AskTrip is built on trust – and that means being transparent, even when it’s uncomfortable. By working together, we can keep improving and make evidence access safer and more reliable for everyone.

1

AskTrip: Beyond Trip

AskTrip currently generates answers from content in the Trip Database. When little is available, we “back fill” using ChatGPT. While answers are clearly labelled, relying on ChatGPT alone (or mostly) doesn’t feel entirely comfortable.

Twenty years ago, when we answered clinical questions manually, we often had to search beyond Trip or Medline to find reliable evidence. That spirit of search expansion has inspired Beyond Trip (working name). If AskTrip finds little or no evidence in Trip, it will now automatically search other sources to strengthen the answer.

Our approach now searches both OpenAlex and Google Scholar – two vast, general academic databases. Even when limited to peer-reviewed medical journals, this still represents a huge increase in coverage compared to Trip alone.

Take one example question What psychiatric adverse effects are associated with the use of antileukotrienes in asthma treatment? that was asked today. AskTrip’s standard answer cited 2 references. With Beyond Trip, the system retrieved 11 references, including:

  • Adverse drug reactions of leukotriene receptor antagonists in children with asthma: a systematic review
  • Neuropsychiatric reactions with the use of montelukast
  • Psychiatric adverse effects of montelukast—a nationwide cohort study
  • Suspected Adverse Drug Reactions Associated with Leukotriene Receptor Antagonists Versus First Line Asthma Medications: A National Registry-Pharmacology Approach
  • Risk of psychiatric adverse events among montelukast users

Beyond Trip won’t activate for the majority of questions. But when AskTrip turns up little or no supporting evidence, it will automatically engage, taking around 60 seconds longer while drawing on a much broader pool of references.

We expect to release it within the next two weeks. Believe me, this is a major step forward for AskTrip 🙂

1

Responding to Concerns About AI at Trip

As part of our commitment to quality with AskTrip (and Trip more broadly), we actively encourage feedback. We recently received the following comment, which I’d like to respond to in case it reflects a view held by others:

“I believe that generative AI tools should not be promoted and positioned as equivalent to expert searching, and feel that it is completely inappropriate that TRIP has devoted so many resources to this. There is extensive evidence that AI tools lack the precision and recall of equivalent systematic searches performed by human beings, and treating them as search engines — especially in medicine and healthcare — causes serious risks when it comes to the reliability of evidence used to support clinical practice. Generative AI is fancy predictive text, not a search engine, and the fact that TRIP has devoted significant resources towards this pivot to AI is extremely disappointing. This irresponsibility has meant that I am less likely to use TRIP as a database, and less likely to recommend it to the healthcare professionals I support.“

My response is:

Thank you for taking the time to share your concerns. We take all feedback seriously, and it’s important to us to listen and reflect when people raise issues around the use of AI in healthcare.

We’d like to reassure you on a few points. AskTrip is not designed to replace systematic searches or the expertise of information professionals. Instead, it builds on our nearly 30 years’ experience in making high-quality evidence accessible to healthcare professionals. The system is supported by extensive quality-control processes, which we’ve written about in more detail on our blog. These safeguards mean that AskTrip is very different from generic generative AI tools, even if it may look similar on the surface.

We also want to emphasise that using AskTrip is entirely optional – it sits alongside the existing Trip Database, which continues to work as it always has. For some clinicians, especially those without ready access to specialist librarian support, AskTrip provides an additional way to quickly access evidence in a clinically relevant timeframe. For others, it won’t be the right fit, and that’s absolutely fine.

More broadly, we recognise that AI is here to stay. The real challenge – for Trip and for information specialists – is to understand where it adds value, where it falls short, and how to use it responsibly in service of healthcare professionals. Ultimately, both Trip and expert searchers need to offer solutions that meet user needs. If we don’t, clinicians will inevitably look elsewhere.

Finally, on resources – while we have invested in this area, it’s relative and has not been at the expense of our core database. We remain committed, as ever, to delivering trusted evidence to healthcare professionals worldwide.

We share your belief that evidence in healthcare needs to be robust, reliable, and used responsibly. That’s why we’re keen to be transparent and to have these conversations.

0

AskTrip in Trip

We’ve just rolled out an exciting new feature that brings the Q&A power of AskTrip directly into the Trip Database. Using AI, the system predicts questions based on a user’s search terms and the articles they view. By analysing session activity and identifying user intentions, it suggests the most relevant questions to support clinical decision-making.

If a user runs a simple search, there’s little indication of intent, so no questions are shown. Once an article is clicked, the AI gains enough context to understand the intention and generate relevant questions. In this example, a user searches for obesity children (indicating intent) and these are the suggested questions:

The user then scrolls down and clicks on the article Surgery for the treatment of obesity in children and adolescents. This indicates an interest in surgery, so the questions update (appearing above the clicked article):

As a user clicks on additional articles, the suggested questions are updated further.

In short, AskTrip transforms a user’s browsing into a dynamic, question-driven experience—helping clinicians move from search to evidence faster, with AI guiding them to the answers that matter most.

0

Quality Control in Action: Guidelines Reflect Yesterday’s Evidence

Because quality control is central to the growth of AskTrip, we invest a lot of time in it. I’d like to share a couple of examples where we take a systematic review and ask AskTrip the very clinical question the review set out to answer.

Example 1: A Systematic Review And Meta-Analysis Of Randomized Trials Of Therapeutic Intraarticular Facet Joint Injections In Chronic Axial Spinal Pain

AskTrip question: What is the evidence for intra-articular facet joint injections in treating chronic axial spinal pain?

Using the neutral(ish) ChatGPT 5 we asked it to compare the results:

Both sources agree that intra-articular facet joint injections offer at best short-term pain relief in chronic axial spinal pain, with limited or low-certainty evidence and weak/negative support from guidelines. The systematic review/meta-analysis (ONE) takes a narrow, RCT-only lens and downgrades the evidence to Level IV with low certainty, stressing the absence of robust long-term benefit. The broader narrative and guideline-based synthesis (TWO) reaches a similar conclusion but adds clinical context: short-term improvements are sometimes seen, yet effects are transient, major guidelines (e.g., NICE, BMJ) recommend against routine use, and alternatives such as radiofrequency ablation generally provide more durable relief. Thus, while both sources converge on limited efficacy, ONE emphasizes strict evidence grading, whereas TWO highlights comparative effectiveness, guideline positions, and practical considerations such as imaging and safety.

Example 2: Prophylactic Antibiotics for Upper Gastrointestinal Bleeding in Patients With Cirrhosis: A Systematic Review and Bayesian Meta-Analysis

AskTrip question: In patients with cirrhosis and upper gastrointestinal bleeding, should prophylactic antibiotics be administered to reduce mortality or complications like infection or re-bleeding?

ChatGPT5 comparison:

The 2024 systematic review and Bayesian meta-analysis casts doubt on the mortality benefit of prophylactic antibiotics in cirrhotic patients with upper GI bleeding, showing that shorter or no prophylaxis was likely non-inferior for mortality and rebleeding, though antibiotics did reduce reported infections; overall, it highlights low–moderate quality evidence and questions the current 5–7 day guideline standard. In contrast, the AskTrip answer aligns with NICE and earlier meta-analyses, presenting prophylactic antibiotics as evidence-based standard care that reduces mortality, infections, and rebleeding, particularly in decompensated cirrhosis, and recommending 5–7 days of treatment with ceftriaxone or quinolones depending on resistance. Thus, while the SR emphasises uncertainty and possible overtreatment, the AskTrip answer reflects guideline consensus and stronger claims of clinical benefit.

A really interesting finding, our answer reflects current guideline recommendations, which support prophylactic antibiotics in cirrhotic patients with upper GI bleeding. But a new 2025 systematic review questions the mortality benefit and suggests shorter or no courses may be just as effective. It’s a clear example of how new evidence can challenge established guidelines—and why keeping answers under review is so important.

0

The Key to Efficient Evidence Searching: Structure the Question First

Clinicians waste a lot of time searching for clinical evidence! In AskTrip, we’ve seen that when our automated answers are limited, it’s often not because the evidence doesn’t exist, but because the question itself was too vague.

Evidence searching is like diagnosis: a fuzzy question leads to fuzzy answers. The fastest way to get to the right evidence is to sharpen the question before you even touch the search bar.

Structure the Question: The Foundation of Evidence Retrieval

A clear, well-framed question is possibly the single biggest factor in cutting search time.

A vague query like “asthma treatment” returns thousands of scattered results. Reframed using PICO, the question becomes much more precise: “In children with asthma (Patient), how effective are inhaled corticosteroids (Intervention) compared with leukotriene antagonists (Comparator) in reducing exacerbations (Outcome)?”

This is the PICO framework:

  • Patient (or Problem)
  • Intervention
  • Comparator
  • Outcome

You don’t need every element every time, but just adding a comparator or outcome can transform your results from broad noise to focused evidence.

And to make this even easier, Trip includes a dedicated PICO interface with four search boxes—one for each PICO element. This helps you break down your question into its core components and avoid the common pitfall of vague searching.

Using PICO in Trip

Once you’ve identified the PICO elements, you can:

  • Enter them directly into Trip’s standard search, combining terms to sharpen your results. For the example question the search might be children asthma AND inhaled corticosteroids AND leukotriene antagonists AND exacerbations which generates just 265 results of which 39 are from the higher quality, secondary evidence.
  • Or use Trip’s dedicated PICO interface, which has four search boxes, one for each PICO element. Unlike the standard search, this isn’t designed to be exhaustive. Instead, it aims to return a handful of the most relevant documents, the ones most likely to answer your question quickly.

Looking Ahead: Smarter AI Support

We’re enhancing Trip’s PICO interface with AI and large language model (LLM) tools, so clinicians can automatically uncover more relevant evidence without extra effort.

This is just the beginning. In future blogs, we’ll explore how to speed up other stages of evidence searching—using filters effectively, navigating the evidence pyramid, and more.

0

Blog at WordPress.com.

Up ↑