Session summary

Managing type 2 diabetes in primary care: what changed in 2026

A packed afternoon session took a hard look at how first-line prescribing has shifted, why the cardiovascular evidence keeps rewriting the running order, and what all of it means for a ten-minute appointment.

RC Dr Rachel ConleyConsultant in Diabetes & Endocrinology · session lead Recorded 14 May 2026 Runtime 58 min 1.0 CPD credit

Full session 58:04

The state of play

Why the running order moved

For the best part of a decade, the answer to "what do I start first?" was reassuringly boring. That is no longer the case, and the session opened by being honest about it.

Metformin still holds its place at the top of the page for most people, Conley was quick to say — but the interesting decisions now happen straight afterwards, and they happen faster than they used to. Where the old sequence waited for a patient to fail one therapy before reaching for the next, the current thinking is to look at the whole person up front: their weight, their kidneys, and above all their cardiovascular risk. The drug that best protects the heart may now earn its place before the drug that best lowers a number on a lab report.

That reframing sounds abstract until you put it in front of a real list of patients. The session used a running case — a newly diagnosed 58-year-old with a history of heart failure — to show how quickly the modern algorithm diverges from the one many of us learned. The point was not that the old approach was wrong, but that it was built for a question we are no longer only asking.

We spent years treating the glucose and hoping the heart would follow. The evidence has quietly reversed that instinct.

— Dr Rachel Conley, opening remarks

What follows is the shape of the argument as she made it, section by section, with the slides and clips that did the heavy lifting.

Why the guidelines shifted

Three forces moved at once, Conley argued. The first was a run of large cardiovascular outcome trials that were designed to prove safety and ended up proving benefit. The second was the arrival of agents that treat weight and glucose as a single problem rather than two. The third — less glamorous but arguably more important — was the slow accumulation of real-world data showing that the trial results held up in ordinary practice, not just in carefully selected study populations.

The clip below is the two minutes where the room went quiet: a side-by-side of the 2019 and 2026 decision trees, annotated live.

Clip · 2:112:11

The evidence base

To keep the trials honest, the session laid the headline studies out in a single table rather than talking around them. The columns Conley cared about were not the p-values but the practical ones: who was studied, what actually changed, and whether the finding was strong enough to move a guideline.

The table did something a slide full of forest plots could not: it made the pattern obvious. Every guideline-moving trial in the top three rows shares a shape — a cardiovascular or renal endpoint, not a glucose one. That is the tell.

A note on cardiovascular outcomes

Conley was careful here, and it is worth repeating her caveat. A trial that reduces heart-failure hospitalisations in a population selected for heart failure does not automatically justify the same drug in a low-risk 40-year-old with a mildly raised HbA1c. The benefit travels with the risk. Reading the top of the table as "start everyone on an SGLT2 inhibitor" is exactly the over-reach the evidence does not support.

---

Practical prescribing

All of which is interesting, but the room was full of people who have to make this decision on a Tuesday morning with eight minutes left. The second half got practical.

First-line choices

The working rule Conley offered was deliberately simple, because simple rules survive contact with a busy clinic. Start with metformin unless there is a reason not to. Then, before reaching for a second glucose-lowering agent by reflex, ask whether the patient has established cardiovascular disease, heart failure, or chronic kidney disease. If they do, the second agent is chosen for that condition, and the glucose benefit comes along for free.

The slide that summarised it is below — she called it "the only slide you need to photograph", and judging by the sea of phones that went up, the room agreed.

Slide 22Diabetes in Primary Care · Conley 2026

She was candid about the friction, too. These agents cost more than the older ones, some require a conversation about injection technique, and a few carry side effects that are better discussed before the first prescription than after the first phone call. None of that changes the direction of travel; it just means the conversation is longer than it used to be.

When to escalate

Escalation, she argued, is where most of the avoidable harm hides — not in starting treatment, but in leaving it unchanged for years while the numbers drift. Her checklist for a review appointment was refreshingly short:

• Is the HbA1c at the agreed target for this specific patient — not a universal number, but the one you set together?
• Has anything changed in their cardiovascular or renal picture since you last chose a drug?
• Are they actually taking it, and if not, is the barrier cost, side effects, or complexity — because each has a different fix?
• Is there a de-prescribing case? Sometimes the right escalation is subtraction.

---

From the room

The recording captures the argument; it doesn't quite capture the atmosphere. A few images from the afternoon, and the exchange that stayed with people afterwards.

The exchange everyone quoted afterwards came near the end, when a GP raised the question the whole room had been circling.

I came in thinking this was a talk about drugs. I'm leaving thinking it was a talk about which patient is in front of me.

— Delegate feedback, collected after the session

That, more than any single trial, was the session's thesis: the science has not just given us better tools, it has changed the first question we ask.

---

Resources & further reading

Everything referenced in the session, plus the slide deck and the recording, is available to registered delegates. CPD is claimable until the end of the year.

Next in this track: Cardiovascular risk in the under-40s and De-prescribing in complex multimorbidity — both recorded at MIMS Learning Live 2026.