HW-005
She Was Fine at Church on Sunday
This is the opening act of a member case, with the reasoning unlocked so you can see exactly what you get. Members work the whole case in the library: six decision points that each open only once you’ve committed to an answer, then the debrief, self-assessment and reference card.
- From
- Dr Lena Marsh, emergency physician
- To
- You — GP hospitalist on call
- When
- Tuesday 14 October 2025, 14:20
- Via
- Telephone, from the ED
"Afternoon, it’s Lena in Emerg. Seventy-eight-year-old woman, Margaret, brought in by her daughter after forty-eight hours of progressive confusion. She was fine at church on Sunday. By Tuesday she doesn't know where she is. The ED has started ceftriaxone for a UTI — her urinalysis showed nitrites and leukocytes. Sodium is 126. Can you take over?"
| Item | Value | Reference |
|---|---|---|
| Sodium | 126 | 135–145 mmol/L |
| Urinalysis — nitrites | Positive | Negative |
| Urinalysis — leukocytes | Positive | Negative |
| Duration of confusion | 48 | hours |
| Started in ED | Ceftriaxone | — |
Her daughter adds two things on the way past: Margaret takes hydrochlorothiazide, and she was started on zopiclone about two weeks ago for insomnia.
- The ED started antibiotics for a UTI on the basis of a positive urinalysis. She has no dysuria, frequency, fever, or suprapubic pain. Is this a UTI?
- Her sodium is 126 and she is on hydrochlorothiazide. Is that the cause of her confusion?
- She was started on zopiclone two weeks ago. Is that relevant?
- What do you do about the ceftriaxone?
Show the expert reasoning
This is one of the most important teaching points in hospitalist medicine. A positive urinalysis in a confused elderly patient is not a diagnosis. It is a distraction.
Per IDSA 2019 guidelines, bacteriuria accompanied by delirium without localising genitourinary symptoms should not be treated with antibiotics. Asymptomatic bacteriuria is present in twenty-five to fifty per cent of women over seventy. Treating it does not improve delirium, and it raises the risk of C. difficile.
The two things actually driving this presentation are on the medication list: thiazide-associated hyponatraemia (sodium 126 on hydrochlorothiazide) and a new sedative-hypnotic (zopiclone — a Beers Criteria medication). Two iatrogenic causes. No infection. Stop the ceftriaxone.
When a patient is confused, read the medication list before you read the urine. New medications — sedative-hypnotics and thiazides especially — are a more likely cause of delirium in an elderly woman than a positive urinalysis.
The case runs on through the sodium correction and its ceiling, the urine culture that comes back positive two days later and still should not be treated, the zopiclone discontinuation and what you tell the daughter about sleep, and a discharge plan built around medication reconciliation rather than an antibiotic course.
Every citation in the full debrief carries its publication year.
This is one of forty.
Join free to work every open case in full — decision points, expert reasoning, pearls, self-assessment and reference card.
Join free Open the library