Case Study 2 — The Successful Weight Loss That Wasn't: Ivor, 79

A composite clinical case. ⚠️ The physiology, the screening tools and the failure mode are real; the person is constructed.

Setup

Ivor Petrakis is 79. He lives alone. His wife died fourteen months ago.

He came to the surgery about a shoulder problem and was weighed as routine.

⚠️ Every number on that table moved in the direction a health system is set up to celebrate.

And the note in his record from eight months earlier read:

"Weight down 9 kg. Well done. Encouraged to continue."

What was actually happening

Nobody had asked why.

Fourteen kilos in two years, unintentional, in a 79-year-old. ⚠️ That is not a success. It is a symptom, and it should have triggered an investigation at the first 5%.

The history took eleven minutes once someone asked:

The measurements nobody had taken

💡 Aha moment. ⚠️ He had lost 14 kg and a large fraction of it was muscle.

The BMI improved. The blood pressure improved. The A1c improved. ⚠️ And he could no longer stand up from a chair without his arms, had fallen once, and was at high risk of the fracture that would end his independence.

⚠️ Every metric the system watches got better while the thing that determines whether he stays in his house got substantially worse.

This is §25.12's inversion, and it is not a rare presentation. It is a common one.

What "well done" cost

⚠️ The note eight months earlier is the failure, and it's worth being precise about why.

At that point he had lost 9 kg — about 10% — unintentionally. The appropriate response was: ⚠️ why? Screen. Investigate. Weigh again in a month.

What happened instead was encouragement.

Which ⚠️ confirmed to him that the trajectory was good, delayed investigation by eight months, **and — this is the part he said out loud — **

"I thought I was finally doing something right."

⚠️ A health system that celebrates weight loss without asking whether it was intentional will systematically miss malnutrition, cancer, depression, dementia and dysphagia in older adults, because all of them present this way.

What changed

⚠️ Note how little of it is dietary advice.

1. ⚠️ Dentist. Denture relined within three weeks. He started eating meat again. ⚠️ §25.14 called this one of the two most cost-effective interventions in the chapter and here it is.

2. ⚠️ Medication review. Two changes made; the dry-mouth agent swapped. Taste improved.

3. Investigation completed. ⚠️ Nothing sinister found — which is the outcome you hope for and cannot assume.

4. B12 and vitamin D corrected (Chapter 13).

5. ⚠️ Home food fortification, not supplements first. Milk powder into porridge and soup. Cheese into everything. Oil and butter used generously. Nut butter on toast. Full-fat milk. ⚠️ Everything Part IV recommended, run in reverse — and correct here (§25.14).

6. Protein at every eating occasion. ⚠️ Target ~1.2 g/kg = ~86 g/day, from eggs, tinned fish, cheese, milk, yoghurt, chicken. Not a powder.

7. ⚠️ Resistance training. A referral to a community strength-and-balance programme. ⚠️ He resisted this hardest and it did the most.

8. ⚠️ And the intervention that isn't nutrition: a lunch club, twice a week.

⚠️ That last one was the hardest to arrange and it changed his intake more than anything else on the list. People eat more with other people, and the effect is not small.

Eight months later

⚠️ He gained six kilos, his A1c went up, his blood pressure went up — and he can stand up from a chair.

⚠️ If you find that trade uncomfortable, that discomfort is the chapter.

Every framework that scores this as a deterioration is using metrics designed for a fifty-year-old.

What he said:

"I didn't know I was ill. I thought I was just old."

Analysis

1. ⚠️ Fourteen kilos of unintentional loss in a 79-year-old is a symptom, not an achievement, and should have triggered investigation at the first 5%.

2. Every conventional metric improved while function collapsed. ⚠️ BMI, blood pressure and A1c all moved "correctly" while he lost the ability to rise from a chair.

3. ⚠️ "Well done. Encouraged to continue" delayed investigation by eight months and told him the trajectory was right.

4. The causes were: bereavement, a loose denture, two medications, isolation, transport, and low mood. ⚠️ Not one of them is a dietary problem, and all of them produced a dietary outcome.

5. ⚠️ The dental review and the medication review were the two highest-yield interventions — §25.14, exactly as claimed.

6. Home fortification beat supplements — ⚠️ cheaper, more acceptable, and it was food he already recognized.

7. ⚠️ The lunch club changed intake more than any nutritional instruction.

8. And the resistance training was resisted hardest and mattered most (§25.13). ⚠️ Protein without loading does much less.

Discussion Questions

1. ⚠️ Design the alert. What should a primary care system do automatically when an older adult's weight falls 5%? What does it cost, and what's the false-positive burden?

2. "Well done. Encouraged to continue." ⚠️ Was that clinician negligent, or working from guidelines written for a different population? What would have to change?

3. ⚠️ His A1c and blood pressure went UP and the outcome was better. How should that be recorded in a system built on those metrics? Does the record have anywhere to put "can stand up unaided"?

4. The lunch club changed intake most. ⚠️ Is social prescribing a nutrition intervention? Who should fund it, and out of whose budget?

5. ⚠️ He said: "I didn't know I was ill. I thought I was just old." How much illness in later life is absorbed into "getting old"? What would help someone tell the difference?

6. ⚠️ Nothing sinister was found on investigation. Would this case read differently if something had been? Should it?

Your Turn

⚠️ This section is for anyone over about 70, and for anyone who cares for someone who is.

Step 1 — Get a weight, and get last year's.

Now: _ kg · A year ago: _ kg · Change: ____%**

⚠️ Unintentional loss of 5% or more in 6–12 months warrants investigation. Not encouragement.

Step 2 — The two questions that outperform dietary advice (§25.14):

⚠️ "When did you last see a dentist, and do your teeth or dentures work?" _ ⚠️ "How many medications, and which were added in the last year?" _

Step 3 — The function tests. ⚠️ No equipment needed:

• [ ] Can they rise from a chair without using their arms?
• [ ] Can they carry shopping up stairs?
• [ ] Has there been a fall — including one they didn't report?
• [ ] Do they open jars?

⚠️ These tell you more about nutritional status than a BMI does.

Step 4 — Count the protein at one day's eating. ⚠️ Target roughly 1.0–1.2 g/kg with ~0.4 g/kg per meal. Most older adults eating alone are far below it, and breakfast is usually the emptiest meal.

Step 5 — ⚠️ The question nobody asks:

"How many meals a week do you eat with another person?" ____

⚠️ If the answer is zero, that is a modifiable finding with a larger effect than most of this book.

Step 6 — And fortify, don't restrict.

⚠️ Milk powder into soup and porridge · cheese into everything · full-fat milk · oil and butter used generously · nut butter · custard, cream, ice cream.

⚠️ If that list makes you uncomfortable after twenty-four chapters, reread §25.12. The advice inverts, and this is what the inversion looks like on a plate.