Evidence-Based Medicine: How One Research Paper Saved a Life at UPMED Hospital

The Night Everything Changed in ICU Ward 3

The call came at 2:17 AM on a Wednesday.

Dr. Sumaira Talib was on her surgical ICU rotation at UPMED Hospitals, Lahore, running on three hours of sleep and her second cup of hospital chai, when the nurse rang from Ward 3. A 58-year-old male patient, admitted two days earlier with severe intra-abdominal sepsis following a perforated duodenal ulcer, was deteriorating rapidly. His blood pressure was dropping despite aggressive fluid resuscitation and maximum-dose vasopressors. His lactate was climbing. His urine output had slowed to a trickle. The standard sepsis protocol, the one every doctor in the hospital had followed for years, was not working.

Sumaira stood beside the bed, staring at the monitor as the numbers moved in the wrong direction. The senior registrar on call had already seen the patient and ordered the textbook adjustments: increase noradrenaline, add dobutamine, repeat cultures; everything by the book, but the book was not saving this man.

And then Sumaira did something that would change the way every member of The Research Clinic thought about medical research forever.
She opened PubMed on her phone.

Not to pass the time. Not to study for an exam. She searched because she remembered something Dr. Muhammad Yaqoob had said during their very first Research Clinic session: “Every treatment you give your patient was once somebody’s research question. If you don’t read research, you are practising yesterday’s medicine on today’s patients.”

With trembling fingers, she typed: “glucocorticoids therapy septic shock.” Near the top of the results was a landmark randomized controlled trial, the ADRENAL trial, published in the New England Journal of Medicine in 2018, including 3,658 mechanically ventilated patients with septic shock. Hydrocortisone did not reduce deaths at 90 days (27.9% versus 28.8% with placebo), but patients who received hydrocortisone came out of shock faster, which was a secondary outcome.

Sumaira called the on-call consultant. She did not just say, “I think we should try steroids.” She said: “Sir, the ADRENAL trial, a large multi-centre RCT in the NEJM, found that hydrocortisone did not reduce deaths at 90 days, but it did help patients come out of shock faster. Our patient is still in shock on high-dose vasopressors, and current sepsis guidelines suggest considering hydrocortisone in this situation. Can we consider adding it?”

The consultant paused. Then he said yes.

By morning, the patient’s blood pressure had stabilized. By the following day, vasopressors were being weaned. By day five, the patient was transferred out of the ICU. He went home two weeks later to his family in Shahdara.

Did the hydrocortisone save him? We cannot say. One patient getting better after a drug does not prove that the drug worked. He was also receiving antibiotics, fluids, surgical source control and good nursing care, and some patients recover anyway. That is exactly why we need trials like ADRENAL, which compare large groups of patients. What Sumaira did right was not “finding a cure”. She found the best evidence, understood what it did and did not show, and brought it to her consultant.

When the story reached The Research Clinic’s next session, the room was electric. Dr. Hammad Ali, who usually sat at the back making jokes, was leaning forward. Dr. Zunaira Malik had tears in her eyes. Dr. Hassan Raza had stopped asking for shortcuts.

And Dr. Junaid Rashid, the fifteen-year veteran who had once said that his patients being alive was his research, sat very still and said quietly: “Sumaira, how did you know where to look?”

That question opened the door to the most important lesson of our journey.

What Is Evidence-Based Medicine?

I stood at the whiteboard that morning and wrote three words: Evidence-Based Medicine. Then I asked a simple question: “What does this actually mean?”

Hammad raised his hand first. “It means treating patients based on evidence, sir. Based on research.”

“Close,” I said, “but incomplete.”

Evidence-based medicine (EBM) is the conscientious, explicit, and judicious use of the best available evidence, combined with clinical expertise and patient values, to make decisions about individual patient care. That definition comes from Dr. David Sackett, the father of EBM, and every word in it matters. (Ref: Click Here)

Notice: it is not just about research papers. EBM sits on three pillars. The first is the best available research evidence, based on what studies and trials have shown. The second is clinical expertise, the judgment a doctor develops through years of practice. The third is patient values and preferences because the right treatment for one patient may not be right for another.

Sumaira did not just quote a paper. She matched high-quality evidence to a specific clinical scenario, presented it to an experienced consultant who applied his clinical judgment, and together they chose a treatment that aligned with the goal of saving the patient’s life. That is evidence-based medicine in action.

Dr. Junaid raised an objection that many senior doctors share: “I have been treating sepsis for fifteen years. My experience is my evidence.”
“Your experience is invaluable, Junaid sahib,” I replied. “But experience alone can mislead us. A doctor who has treated 500 patients a certain way may believe that method works, without realizing that 50 of those patients would have done better with a different approach. That is where research comes in: it shows us what we cannot see from our own limited sample of cases.”

Something shifted in the room. Research was no longer abstract. It was not about promotions, PMDC requirements, or HEC categories. It was about the patient in bed 7 of ICU Ward 3 and a PGR Year 2 trainee who knew how to find and use evidence.

The Evidence Hierarchy: Not All Evidence Is Created Equal

As we progressed, I introduced them to the research design pyramid, which ranks studies according to the strength of their evidence. I drew the pyramid on the board, and together we filled in each level. At the base, I placed expert opinions and background information that were useful but largely based on personal experience. “These are the weakest forms of evidence,” I explained, “but still the foundation of scientific curiosity.” Moving up, I added case reports and case series, which describe one or a few patients in detail. “That’s where most discoveries start,” Junaid remarked.

Higher up the pyramid were observational studies, including cross-sectional, case-control, and cohort designs. They learned to observe what happens naturally in people without altering their treatments. Junaid remembered a cross-sectional study on hypertension by his colleague; it helped him see patterns but couldn’t prove what caused them. “These studies are like looking at life as it is,” Sumaira said. “We watch, but we don’t interfere.”

I then pointed to the pyramid’s upper layers. “Now we move toward the stronger designs,” I said. Just above the observational studies, I placed the quasi-experimental and non-randomised controlled trials (non-RCTs). These studies test interventions without full randomisation, usually because it isn’t feasible or ethical to assign participants randomly. “They still compare treatment effects,” I explained, “but with a slightly higher chance of bias.”

Then I turned to the very top of the pyramid. “Here are the strongest designs,” I said. They learned that randomised controlled trials (RCTs) are considered the gold standard because participants are randomly assigned to groups, minimising bias and ensuring fairness. “That’s why drug trials use randomisation,” Junaid remarked. At the peak stood systematic reviews and meta-analyses, which combine findings from many high-quality studies. “These give us the most reliable conclusions,” Sumaira added, completing the pyramid.

Over time, they also learned about the quality of evidence, a measure of how much confidence one can have in a study’s findings. High-quality evidence, they found, comes from well-planned research with large samples and minimal bias. Low-quality evidence often arises from small, poorly structured, or subjective studies. “In clinical practice,” Junaid said, “doctors rely more on high-quality studies because they reflect the real effect of a treatment.”

I also warned them about bias, one of the biggest challenges in research. I explained that bias doesn’t mean dishonesty; it means unintentional errors that make results lean in one direction. Sumaira realised that if her study on surgical site infection after laparoscopic appendicectomy included mostly younger patients, the results would be less useful for older people. She also noted how some journals prefer to publish only positive findings, leaving out neutral ones, a form of publication bias.

Zunaira raised her hand quietly from the back. “Sir, where do clinical guidelines fit?”

I smiled. “Excellent question, Zunaira. Clinical practice guidelines, such as those from WHO, NICE, or the AHA, are usually based on systematic reviews of the best available evidence. They translate research into practical recommendations. So they sit alongside the top of the pyramid as applied evidence. When you follow a clinical guideline, you are indirectly using the best available research evidence.”

Image Source: https://www.sciencedirect.com/science/article/pii/S0002916522047554?via%3Dihub

Why Evidence-Based Medicine Matters Especially in Pakistan

Evidence-based medicine is not a luxury for Western hospitals. It is perhaps more important in Pakistan’s healthcare system than anywhere else, precisely because resources are limited.

Consider this: when a government hospital in Lahore has a limited drug budget, every prescribing decision carries weight. Giving an expensive antibiotic that a well-designed RCT has shown to be no better than a cheaper alternative is not just bad medicine; it drains resources that could be used to treat other patients. Evidence-based medicine in Pakistan is not just about quality. It is about justice. Dr. Bushra Fatima, who had been quiet, spoke up: “So when we publish research from Pakistani hospitals, we are not just building our CVs. We are creating evidence that other Pakistani doctors can use.”

“Exactly,” I said. “And that is the deeper purpose of The Research Clinic. Yes, you will learn to publish. Yes, it will help your career. But the ultimate goal is this: every piece of research you produce adds to the body of evidence that helps patients. Your research question could become someone else’s treatment protocol. Your data could save a life in an ICU in Peshawar, or a rural BHU in Tharparkar, or a teaching hospital in Karachi.”

I looked around the room. “Research is not rocket science; it is a skill, and skills can be taught. But it is also a responsibility. When you know how to find and use evidence, and you choose not to, you are choosing to give your patients less than they deserve.”