New Confusion During an Inpatient Stay

Learning objectives#

• Recognize hyperactive, hypoactive, and mixed delirium through acute change, impaired attention, and fluctuation from baseline.
• Use a validated assessment tool without confusing a positive screen with a complete cause assessment.
• Build a prioritized search for infection, hypoxia, medicines, metabolic disturbance, pain, retention, constipation, neurologic disease, and environmental contributors.
• Choose multicomponent non-drug care first while reserving medication for narrowly defined distress or immediate safety risk.
• Assess capacity and discharge readiness as separate, time-sensitive questions rather than consequences of an age or diagnosis label.

Initial presentation#

An 84-year-old woman is on the third day of a medical admission for fever, cough, hypoxemia, and a right lower-lobe pneumonia. At admission she had low blood pressure and organ dysfunction consistent with sepsis; she received prompt hospital treatment, and by the second day her blood pressure, oxygen need, fever, and inflammatory markers were improving.

Before this illness she lived beside her son's home, prepared meals, managed a weekly medicine organizer, used a cane outside, and called her sister daily. Her son reports occasional forgotten names but no hallucinations, wandering, nighttime reversal, or loss of routine function. She normally wears hearing aids. Her history includes hypertension, overactive bladder, and osteoarthritis.

Overnight, a nurse documented that she tried to climb out of bed, pulled at her intravenous line, and insisted she was in a train station. She slept briefly after receiving a sedating antihistamine from an as-needed order. During morning rounds she is drowsy, answers slowly, and says the month is November. She follows one simple command but loses track of a two-part request. When her son arrives after lunch, she is more alert and recognizes him, then becomes suspicious of staff twenty minutes later.

Temperature is 37.2 C, blood pressure 136/72 mm Hg, pulse 88 per minute, respiratory rate 20 per minute, and oxygen saturation 93 percent with the same low-flow oxygen used overnight. She looks tired but not toxic. Lungs have right basal crackles. Heart rhythm is regular. The abdomen is mildly distended without focal tenderness. There is no facial asymmetry, arm drift, new speech deficit, meningismus, or witnessed seizure. A urinary catheter placed during initial resuscitation was removed that morning, and she has not voided since.

A urine culture collected from the catheter the previous day is now reported as growing bacteria, and a team member proposes broadening antibiotics for a urinary infection and discharging her the next morning if she remains afebrile.

Problem representation#

This is an older inpatient recovering from sepsis and pneumonia who has developed an acute, fluctuating disturbance of attention, awareness, thinking, perception, sleep, and motor activity over hours; the course alternates between agitation and reduced responsiveness. Baseline function was substantially better, and there is no current focal neurologic syndrome.

The organizing diagnosis is mixed delirium during an acute hospital stay, but that label is only the beginning. Infection recovery, oxygenation, and a newly administered anticholinergic sedative may all contribute. So may urinary retention, constipation, and electrolyte disturbance. So may pain, sleep fragmentation, sensory deprivation, and unfamiliar surroundings. Recurrent sepsis, stroke, and intracranial bleeding remain cannot-miss alternatives until proportionately assessed. So do seizure, hypoglycemia, medication toxicity, and withdrawal. Neither of the morning's two proposals survives that framing. The positive catheter urine culture is not proof of a symptomatic urinary infection, and the planned discharge is premature, because absence of fever does not establish cognitive recovery, decision-making ability, mobility, medicine safety, or adequate support at home.

Prioritized differential#

1. Multifactorial delirium related to acute illness and hospital stressors#

Reasoning for: Onset over hours, fluctuation within the day, and impaired attention are characteristic. So are disorientation, altered perception, and changing motor activity. Age, severe infection, and hypoxemia all raise risk. So do sleep disruption, immobility, and sensory impairment. So do unfamiliar surroundings and multiple medicines.

Reasoning against or still uncertain: Delirium describes the syndrome but does not identify the causes. The pneumonia is improving, so attributing every symptom to the original infection could miss a new complication or an iatrogenic contributor.

2. Medicine-associated delirium or withdrawal#

The sedating first-generation antihistamine has anticholinergic effects and may worsen confusion, urinary retention, dry mouth, and sedation. Her home overactive-bladder medicine may add anticholinergic burden. Opioids, benzodiazepines, and corticosteroids require review. So do sleep medicines and other centrally active drugs. So do abrupt omissions of alcohol, benzodiazepines, antidepressants, or other medicines that can cause withdrawal.

The medicine is a plausible contributor, not a complete explanation. Stopping every medicine without considering indication, withdrawal, pain, and physiologic stability could create additional harm.

3. Persistent or new infection, hypoxia, or sepsis#

Pneumonia and systemic inflammation can precipitate delirium, and older adults may deteriorate without high fever. Worsening oxygen need, respiratory effort, or blood pressure would demand rapid reassessment. So would worsening lactate, cultures, examination, or organ function. Aspiration, line infection, skin infection, abdominal infection, and Clostridioides difficile disease remain conditional alternatives. Her improving trajectory lowers concern for uncontrolled pneumonia without eliminating it, and repeated assessment across the day tells the team more than one normal temperature does.

4. Urinary retention, constipation, dehydration, pain, and sleep disruption#

Failure to void after catheter removal, abdominal distention, immobility, anticholinergic burden, and a disrupted day-night cycle support several reversible contributors. Pain can present as withdrawal or agitation in someone who cannot describe it clearly. Dehydration and poor intake can coexist with a normal blood pressure.

5. Metabolic or organ dysfunction#

Glucose, sodium, and calcium disturbances can alter cognition. So can kidney, liver, acid-base, and medication disturbances. Testing targets plausible dangers; thyroid, vitamin, toxicology, or endocrine studies follow specific clues.

6. Acute neurologic disease#

Why it cannot be missed: Stroke, intracranial hemorrhage, subdural hematoma, meningitis, encephalitis, and nonconvulsive seizure can resemble delirium.

Case evidence: There is no focal deficit, severe headache, or meningismus. There is no head injury, anticoagulant concern, or witnessed seizure. Systemic triggers make delirium more likely. New focal findings, trauma, seizure features, unexplained low consciousness, or failure to improve would prompt targeted neurologic testing.

7. Dementia, depression, or a primary psychiatric disorder#

A slowly progressive neurocognitive disorder may increase vulnerability and may coexist, but it does not explain an abrupt fluctuating change by itself. Dementia assessment during active delirium can mislabel temporary impairment as a permanent baseline. New late-life psychosis is also less likely than delirium when attention and arousal fluctuate with acute illness.

8. Catheter-associated bacteriuria without symptomatic infection#

The culture may reflect colonization after catheter use. IDSA recommends looking for other causes and observing older patients with bacteriuria and delirium when local urinary symptoms and systemic signs are absent. If fever, hemodynamic instability, or flank pain appeared, infection management would change. The same would follow from dysuria or another compatible syndrome. Antimicrobial stewardship and sepsis vigilance are complementary, not competing, goals.

Focused history and examination#

Establish baseline and time course#

• Ask the patient what she has noticed, while recognizing that recall may be incomplete.
• With permission, ask family about baseline cognition, function, communication, sensory aids, medicines, and timing.
• Review nursing observations across shifts for fluctuation, motor activity, perception, intake, elimination, mobility, pain, and sleep.
• Check whether language discordance, hearing loss, aphasia, low vision, or an unfamiliar communication style is being mistaken for confusion.

Search for dangerous and reversible contributors#

• Repeat vital signs and oxygen needs; ask about cough, breathlessness, chest symptoms, aspiration, diarrhea, urinary symptoms, pain, headache, trauma, and seizure clues.
• Reconcile every inpatient, home, nonprescription, and recently stopped medicine, including actual administration times and renal function.
• Quantify fluid and food intake, vomiting, bowel movements, urine output, and mobility.
• Ask sensitively about alcohol and sedative use so withdrawal is neither missed nor presumed.
• Examine the heart, lungs, abdomen, bladder, skin, mouth, lines, joints, and potential pressure injuries. Perform a focused neurologic examination and assess gait only when safe.

Hypoactive delirium deserves special attention. A person who is withdrawn, slow, sleepy, eating less, and moving less may be at equal or greater risk than someone calling out. Sedation can make the ward appear calmer while the underlying syndrome worsens.

Diagnostic strategy#

Confirm the syndrome with the right tool#

The team uses the 4AT because NICE recommends it when delirium indicators are identified outside critical care or postoperative recovery, and the assessment documents alertness, orientation, attention, and acute change or fluctuation. A positive result supports prompt expert diagnostic assessment; it is not a laboratory test that identifies the cause. In critical care, CAM-ICU or the Intensive Care Delirium Screening Checklist is used instead.

Serial bedside observation matters because delirium fluctuates. A normal conversation at noon does not invalidate documented inattention overnight. The record should state the syndrome, suspected contributors, baseline source, assessment result, and current safety needs. NICE also recommends documenting delirium in the primary care record so it is not lost at transition.

Target tests to the clinical picture#

Immediate checks include bedside glucose, oxygenation, and current vital signs. They include medicine administration history, bladder volume, and a focused neurologic examination. In this case, useful laboratory testing includes a complete blood count, electrolytes, and kidney and liver measures. It includes glucose, calcium, and acid-base testing if respiratory or metabolic change is suspected. Infection studies and chest imaging are repeated only if the clinical trajectory raises concern for progression, a new source, or a complication.

A urine test is not a default delirium test. It is chosen when symptoms, systemic infection features, or another indication makes the result actionable, and catheter cultures often detect incidental bacteriuria. Routine head computed tomography is likewise unnecessary when systemic contributors are evident and no trauma or focal feature exists. Imaging follows focal deficits, head injury, or anticoagulant context. It follows severe headache, unexpectedly low consciousness, or an unexplained course. Electroencephalography and lumbar puncture are reserved for compatible seizure or central nervous system syndromes. The finding drives the scan, not the confusion.

Progressive results and interpretation#

The 4AT is positive because of altered alertness, errors on orientation and attention tasks, and a documented acute fluctuating change. This supports delirium. It does not establish that pneumonia, urine bacteria, or any one medicine is the cause.

Bedside glucose is normal. Oxygen saturation improves to her recent baseline after the nasal cannula is repositioned. A bladder scan shows substantial retention, and catheterization under the team's urinary-retention plan relieves discomfort. Her son brings the hearing aids, which had been left at home. The medication record confirms the overnight sedating antihistamine and reveals that her anticholinergic bladder medicine continued despite retention.

Laboratory testing shows sodium 129 mmol/L, down from 137 mmol/L on admission, with mildly worse kidney function and no new leukocytosis. Review of intake reveals that she drank little while receiving hypotonic maintenance fluid. Pneumonia markers and respiratory findings continue to improve. Abdominal examination and history support constipation, with no bowel movement since admission.

The urine culture grew a common organism, but she has no urinary symptoms, new fever, or hemodynamic instability. Because bacteriuria is common after catheter use, antibiotics are not broadened solely for that result. Surveillance for a genuine urinary or systemic syndrome continues.

The best explanation is mixed delirium with interacting contributors: recent severe infection, mild hypoxemia, and hyponatremia and reduced intake. The contributors include urinary retention, constipation, and anticholinergic medicine burden. They include sleep interruption, immobility, and loss of hearing aids. None needs to be declared the single cause. Improvement after correction will strengthen but not mathematically prove the model.

Management plan#

Correct causes and reduce new insults#

The team continues appropriate pneumonia treatment while reassessing duration and spectrum according to the confirmed source and clinical response, and fluids and sodium correction are individualized to volume status, heart and kidney function, cause of hyponatremia, and safe correction limits. Exact fluid and electrolyte protocols are omitted here on purpose. Overly rapid sodium correction can cause severe neurologic harm.

Urinary retention and constipation are treated, catheter need is reviewed daily, and pain is assessed with verbal and nonverbal cues. The sedating antihistamine is removed from the as-needed list. The bladder medicine and every other anticholinergic or centrally active agent are reviewed with pharmacy and the prescriber. Necessary medicines are not withheld without considering withdrawal, pain, seizure risk, or loss of disease control.

Deliver a multicomponent delirium plan#

Non-drug care is active treatment, not decoration. The ward plan includes:

• frequent calm reorientation with a visible clock and calendar;
• hearing aids, glasses, dentures, and communication in the preferred language;
• daylight, daytime activity, and reduced overnight interruption when safe;
• supervised mobility and range-of-motion work rather than bed rest;
• food, fluid, oral care, and assistance matched to swallowing and functional needs;
• family presence with the patient's agreement, using familiar conversation rather than repeated testing;
• consistent staff and room location where feasible;
• removal of unnecessary lines, alarms, and catheters;
• fall, aspiration, and pressure-injury precautions that preserve mobility and dignity.

Physical restraint is not a convenience response to wandering or line pulling. It can increase fear, injury, immobility, and agitation. The team first changes the environment, assigns observation appropriate to risk, treats discomfort, removes unnecessary devices, and uses verbal and nonverbal de-escalation.

Reserve medication for a narrow indication#

An antipsychotic is not routine delirium treatment and does not replace cause correction or supportive care. If the patient remains severely distressed or presents an immediate risk of substantial harm after de-escalation fails, a short, cautious, lowest-appropriate intervention may be considered under local policy and specialist review. Exact dosing is intentionally omitted.

Before use, the clinician reviews corrected QT interval, other QT-prolonging medicines, and electrolyte disturbance. The review covers Parkinson disease, dementia with Lewy bodies, and stroke risk. It covers sedation, swallowing, aspiration, falls, and prior reactions. Need is reassessed frequently, and continuation at discharge requires an explicit indication and stop plan. Benzodiazepines can worsen delirium in many older adults but may be necessary for alcohol or sedative withdrawal, seizures, or another specific indication. The diagnosis determines the exception.

Escalation, referral, and safety net#

Emergency reassessment is required for new focal weakness, facial droop, or speech change. It is required for seizure, severe headache, or head trauma. It is required for meningismus, profound reduction in consciousness, or severe hypoxemia. It is required for chest pain, shock, or rapidly worsening infection. It is required for a dangerous glucose, sodium, or calcium disturbance. It is required for a dangerous kidney, liver, or acid-base disturbance. Delirium can be the first visible sign of acute deterioration, so new confusion must not be managed only with a sedative.

Geriatrics or hospital medicine review is useful when causes are multiple, function has declined, medicines are complex, or discharge is uncertain. Neurology becomes important for focal findings, suspected seizure, unexplained persistent impairment, or an atypical course. Psychiatry can help with severe behavioral symptoms or diagnostic uncertainty but should not be used to reclassify a medical delirium as a primary psychiatric condition.

The immediate safety net is shared across shifts: who monitors fluctuation, oxygen, intake, urine, stool, mobility, and pending results; what triggers the rapid-response system; and which non-drug approaches work for this person. Family members are invited to report abrupt changes but are not made responsible for ward safety.

Communication, shared decisions, and equity#

The clinician tells the patient during a clearer period and repeats the explanation for her son with consent: "Your thinking and attention have changed quickly and are fluctuating. This is called delirium. It often has several causes. We are treating the infection and also checking oxygen, medicines, sodium, and bladder. We are checking bowels, sleep, hearing, and mobility. It may improve, but we will not assume you are ready to go home until we reassess."

Delirium does not erase the person. Staff address her directly, explain each touch and procedure, avoid arguing about a hallucination, and respond to the fear behind it. A professional interpreter is used when needed. Hearing loss, low vision, literacy, cultural communication, and disability can mimic or intensify apparent cognitive impairment if no accommodations are made.

Capacity is assessed for the specific decision and time under local law. Delirium raises concern without automatically removing capacity. The team treats barriers, supports communication, and chooses a clearer time. It checks whether she can understand, retain, weigh, and communicate the relevant choice. An unwise choice is not itself incapacity.

If she cannot make a time-sensitive decision, the responsible team follows the applicable legal process, advance directives, authorized decision-maker rules, and the person's known values. Capacity is revisited as the delirium changes. The NICE capacity guidance is jurisdiction-specific to England and Wales, but its decision-specific, support-first principles illustrate why a global incapacity label is unsafe.

Follow-up and contingencies#

Over the next two days, sodium and kidney function improve gradually, urinary retention does not recur after the medication plan is changed, bowel function returns, oxygen is discontinued, and sleep becomes less fragmented. Her attention is better in the morning but still declines late in the day. She can discuss why she is in hospital and participate in choices, yet she needs cueing for medicines and transfers.

The team does not equate partial improvement with baseline recovery. Discharge readiness includes:

• stable infection, oxygenation, electrolytes, hydration, and organ function;
• a documented cognitive and functional comparison with baseline;
• safe mobility, toileting, nutrition, swallowing, and fall plan;
• complete medication reconciliation, especially removal of temporary sedatives;
• assessment of the specific decisions required for discharge;
• caregiver availability and consent, without assuming family can provide continuous care;
• written instructions, equipment, transport, home services, and timely follow-up;
• a named clinician responsible for pending results and recovery review.

She is discharged only after a short additional period of recovery, therapy assessment, and an agreed temporary supervision plan. The discharge summary names delirium, suspected contributors, medicine changes, current function, and the positive urine culture that did not represent a treated symptomatic infection. This prevents a later clinician from assuming that confusion proved a urinary infection.

Follow-up asks whether attention and sleep returned to baseline and whether new functional support is still needed. It asks whether infection, retention, or constipation recurred. It asks whether dehydration, falls, or medicine problems recurred. Persistent or newly recognized cognitive impairment is evaluated after the acute syndrome has had time to resolve. The team does not promise complete recovery, and it does not diagnose dementia solely from inpatient performance.

Reasoning traps and alternative pathways#

• Calling all confusion dementia: Abrupt onset, inattention, and fluctuation point toward delirium even when dementia may coexist.
• Recognizing only agitation: Withdrawal, drowsiness, slow responses, reduced eating, and immobility can mark hypoactive delirium.
• Treating the screen as the diagnosis and cause: A tool supports recognition; it does not replace clinical diagnosis or the search for contributors.
• Anchoring on the original infection: Improving pneumonia does not exclude retention, medicine effects, sodium disturbance, stroke, or a new infection.
• Equating bacteriuria with urinary infection: Culture results require symptoms and systemic context, especially after catheter use.
• Sedating the behavior: A calmer patient may be more obtunded, aspirate, fall, or lose the ability to participate while the cause remains untreated.
• Using restraints for system convenience: Unfamiliar alarms, devices, room changes, sensory deprivation, and staffing patterns may be modifiable causes of distress.
• Assuming incapacity from diagnosis: Capacity is decision-specific, time-specific, and assessed after support.
• Discharging when fever resolves: Cognition, function, medicines, support, pending results, and contingency planning determine transition safety.

Alternative branches remain open. A new unilateral deficit would activate a stroke pathway. Recurrent fever, hypotension, or rising oxygen need would reopen infection and sepsis evaluation. Rhythmic movements or unexplained persistent unresponsiveness could require electroencephalography. Tremor, autonomic activation, and a compatible substance history would raise withdrawal. Persistent decline after medical recovery would prompt a structured neurocognitive assessment rather than indefinite use of the delirium label.

Evidence limits and what could change#

Delirium is heterogeneous, and no single laboratory test confirms it or identifies its cause. NICE updated its assessment recommendation in 2023 to use the 4AT outside critical care, while much of the prevention and treatment evidence remains older and supports multicomponent, tailored care. Benefits of any one component are difficult to isolate because the intervention works as a system.

Evidence does not support routine antipsychotic use to shorten delirium, and medication harms are especially relevant in older adults. NICE allows narrowly selected short-term haloperidol after failed de-escalation for distress or danger, with major cardiac and neurologic cautions. American Geriatrics Society criteria similarly warn against anticholinergics, benzodiazepines, and routine antipsychotics in older adults at risk. Individual indications and withdrawal syndromes can create exceptions.

The IDSA recommendation against treating bacteriuria alone in an older person with delirium is strong but based on low or very low certainty for some outcomes. It does not apply when local urinary symptoms, fever, hemodynamic instability, or a sepsis syndrome makes infection plausible. CDC sepsis guidance also emphasizes frequent reassessment and post-sepsis cognitive and functional needs.

Legal rules for capacity, surrogate decisions, privacy, and discharge vary by jurisdiction. This case provides clinical reasoning principles, not a legal determination. Baseline dementia, severe sensory impairment, or palliative goals would materially change the pathway. So would alcohol withdrawal, intensive-care treatment, a neurologic emergency, or unavailable home support.

Key points#

• Acute change, impaired attention, and fluctuation identify delirium more reliably than agitation alone; hypoactive delirium is easy to miss.
• Delirium is usually multifactorial, so infection, oxygenation, medicines, metabolism, pain, bladder, bowels, sleep, mobility, and sensory needs are assessed together.
• A positive urine culture after catheter use does not establish symptomatic infection or justify antibiotics without compatible clinical evidence.
• Multicomponent non-drug care and cause correction come first; sedating medication is reserved for a narrow, reviewed safety or distress indication.
• Capacity and discharge readiness are separate, time-sensitive assessments, and the transition plan must document baseline, residual impairment, medicine changes, support, and contingencies.