Case Study 1 — Walt Goes for Remission: Eleven Years, and the Question Nobody Asked

Walt Prosser is an illustrative composite carried through this book — Chapter 11 (the fibre ramp), Chapter 13 (B12), Chapter 16 (the supplement audit and the biotin–troponin ER visit), and here. ⚠️ The remission protocol, the dose-response and the medication management are real; the person is constructed. ⚠️ This describes a supervised clinical intervention, not a self-directed one.**


Setup

Walt is 68. Type 2 diabetes for eleven years.

A1c 6.9% (down from 7.4% after Chapters 11 and 16)
Weight 103 kg, BMI 33.6
Waist ⚠️ 116 cm — never previously measured
Medication Metformin 1 g twice daily
BP 148/88
Triglycerides 2.6 mmol/L
ALT ⚠️ 61 U/L — flagged on three previous panels, never explained to him
Lp(a) ⚠️ 92 nmol/L — never measured until now

He came in about his knees.

⚠️ And in eleven years of appointments, nobody had ever said the word "remission" to him.


The conversation that changed it

He asked what he thought was a resigned question:

"It just gets worse, doesn't it. That's what diabetes does."

⚠️ That is what he had understood for eleven years, and it is what the structure of his care had implied — annual reviews, medication escalation, and the word "progressive."

The answer:

"Sometimes. Not always. There's a reasonable chance you could come off medication entirely — and we can tell you roughly what it would take."

⚠️ Then the numbers, from §26.3:

⚠️ Weight lost ⚠️ Remission at 12 months, in DiRECT
≥15 kg ~86%
10–15 kg ~57%
5–10 kg ~34%
<5 kg ⚠️ ~7%

Walt looked at the table for a while and said:

"So it's not 'lose some weight and see.' There's a number."

⚠️ That is the entire value of the dose-response gradient, and it is why §26.3 argues it matters more than the headline remission rate.


The two honest cautions he was given first

1. ⚠️ Duration. Eleven years is longer than DiRECT's recruitment window of up to six. Remission is less likely with longer duration, because beta-cell function is less recoverable.

⚠️ He was told this explicitly, before starting, with a number attached: "Your chances are lower than those figures. I don't know by how much. It's still worth trying if you want to."

2. ⚠️ Medication. He was on metformin only, which is low hypoglycaemia risk — ⚠️ but the protocol still required prescriber involvement, and his blood pressure medication needed monitoring because it commonly requires reduction as weight falls.

⚠️ Had he been on gliclazide or insulin, this would have been a different and considerably more carefully managed conversation (§26.4). A person on a sulfonylurea who starts an 850 kcal formula diet without medication adjustment is at real risk of hypoglycaemia, and it can be severe.


What he actually did

A structured programme, supervised:

Phase
1 ⚠️ Total diet replacement — formula products, ~850 kcal/day, ~12 weeks Metformin stopped at start; BP medication reviewed
2 Structured food reintroduction — ~6 weeks, stepwise Weight monitored weekly
3 ⚠️ Long-term maintenance support — monthly contact, then quarterly The phase that determines whether it holds (Chapter 24 §24.2)

⚠️ What he found hardest, in his words:

Weeks 1–2: "Genuinely horrible. I thought about food constantly." Weeks 3–12: "It got easier around day ten. Then it was just boring." ⚠️ Reintroduction: "This was the frightening bit. I'd stopped trusting myself with food."

⚠️ That last line is why phase 2 is structured rather than "go back to normal" — Chapter 24 Case Study 1's age-39 attempt failed precisely there.


The numbers, in the order they moved

⚠️ §26.13b, observed.

Week What changed
1 ⚠️ Blood glucose fell immediately — before meaningful weight loss. This is the liver responding, not the fat coming off
2 Blood pressure down; ⚠️ BP medication reduced
3 ⚠️ Triglycerides 2.6 → 1.4 mmol/L
8 ⚠️ ALT 61 → 34 U/L — liver fat, the fastest structural change available
12 ⚠️ A1c 6.9% → 5.7% — first meaningful reading, as expected
12 Weight 103 → 88 kg (−15 kg)
Waist ⚠️ 116 → 101 cm

💡 Aha moment. ⚠️ Glucose improved in week one, before the weight did.

This surprises almost everyone, and it is §26.1b's mechanism showing itself: the earliest effect of severe energy restriction is a rapid fall in LIVER fat, which reduces hepatic glucose output.

⚠️ The pancreas takes longer. The scale takes longer still. Which means "nothing's happening yet" in week two is usually wrong — the wrong thing is being measured.


At 12 months

Baseline 12 months
Weight 103 kg ⚠️ 91 kg (regained 3 from the nadir)
A1c 6.9% ⚠️ 5.9%, off all diabetes medication
⚠️ Remission status ⚠️ Achieved and maintained
BP 148/88 128/78, on a reduced dose
Triglycerides 2.6 1.3
ALT 61 29
Waist 116 cm 103 cm
⚠️ Lp(a) 92 nmol/L ⚠️ 91 nmol/L

⚠️ Read the last row.

Everything on that panel moved except the one thing that was never going to.

⚠️ And knowing that in advance mattered — because when the Lp(a) came back unchanged, Walt did not conclude the programme had failed. He had been told at the start which number was tier 4 (§26.1).

What his Lp(a) did change: his cardiovascular risk is higher than his other numbers suggest, so everything else is now managed more aggressively — including a statin, which he takes without regarding it as a defeat.


Analysis

1. ⚠️ Eleven years, and nobody had said "remission." The single most consequential fact in the case.

2. The dose-response table converted a vague goal into a target. ⚠️ "There's a number" is what made him agree.

3. ⚠️ He was told his chances were lower than the trial figures, before starting. Duration matters, and setting the expectation honestly is what allows a partial result to feel like a result.

4. ⚠️ Glucose moved in week one, before the weight. Liver fat first, pancreas later, scale later still — §26.1b's mechanism, visible.

5. The reintroduction phase was the frightening part, and it is where unsupported programmes fail (Chapter 24 Case Study 1).

6. ⚠️ He regained 3 kg and remained in remission. Remission is not all-or-nothing at a single weight, and framing it that way would have made a good outcome feel like a failure.

7. ⚠️ Lp(a) didn't move, and he had been warned. The forewarning is what stopped an unchanged number from discrediting a successful intervention.

8. And his ALT — flagged on three panels over several years — had never been explained. ⚠️ Fatty liver was the most reversible thing on his chart and nobody had named it.


Discussion Questions

  1. ⚠️ Eleven years without the word "remission." Why? Give a structural explanation, not a blaming one — and say what would have to change in how diabetes reviews are conducted.

  2. DiRECT recruited people diagnosed within six years; Walt was at eleven. ⚠️ Was offering him the programme justified? How should evidence be extrapolated beyond its recruitment criteria?

  3. ⚠️ He was told his chances were lower and given no number. Is "I don't know by how much" good practice or an evasion? What would you have said?

  4. Glucose moved before weight. ⚠️ How much of medicine's messaging about diabetes would change if this were widely understood?

  5. ⚠️ He regained 3 kg and stayed in remission. How should remission be defined and monitored so that normal weight fluctuation doesn't read as failure?

  6. His ALT was flagged three times and never explained. ⚠️ Design the system change. What stops an abnormal result from being noted and ignored?


Your Turn

⚠️ This is for anyone with type 2 diabetes or prediabetes, or anyone advising someone who has.

Step 1 — Establish the basics:

Years since diagnosis: ____ (⚠️ shorter is better for remission) Current A1c: _ · Weight: _ kg · Waist: ____ cm ⚠️ Medications, specifically: any insulin or sulfonylurea? ____

Step 2 — Find your number. ⚠️ Using §26.3's gradient:

Weight loss From my current weight, that means Remission likelihood
5% ____ kg Low
10% ____ kg Moderate
⚠️ 15 kg or more ____ ⚠️ Highest

Step 3 — ⚠️ The question to take to your appointment:

"Am I a candidate for remission, and how much weight loss would it take in my case?"

⚠️ If the answer is no, ask why — duration, beta-cell reserve, other conditions. A reason is useful. Silence isn't.

Step 4 — Sort your panel by §26.1's tiers, and ⚠️ specifically ask whether your Lp(a) has ever been measured.

Step 5 — ⚠️ The safety step, and it is not optional:

If you take insulin or a sulfonylurea, you must speak to your prescriber BEFORE reducing carbohydrate or energy intake. Not after. Not "I'll mention it at my review."

Step 6 — And write the maintenance plan before the loss plan (Chapter 24 §24.14).

⚠️ "What will I still be doing in three years?"

Walt's answer was: weigh weekly, walk the dog, cook five nights, and a quarterly appointment he doesn't cancel. ⚠️ Four things, none of them a diet.