Tips to help you recognize delirium in elderly patients
By Edward Doyle
When it comes to diagnosing delirium in elderly inpatients, Bruce A. Leff, FACP, finds that housestaff typically make one of two mistakes. They either assume that it's natural for a 90-year-old patient to be a little confused because of advancing age, or they hesitate to label patients as delirious out of a fear of stigmatizing them.
While you can chalk up either assumption to residents' inexperience, both reflect a deeper misunderstanding about delirium—one that's often shared by physicians long out of training. As presentations at two recent meetings for internists attest, doctors need to get beyond these basic misconceptions to be able to diagnose and care for delirious patients.
Dr. Leff, a geriatrician at Baltimore's Johns Hopkins University, told a group of internists at ACP's Annual Session in New Orleans last April that while delirium is relatively common in elderly inpatients, it doesn't get the recognition it deserves. While research has found that between 10% and 50% of elderly patients are delirious at admission, he said that physicians miss upwards of 60% of inpatient delirium cases.
Compounding that problem is the fact that many physicians simply take the wrong approach to diagnosing delirium. As Anne Fabiny, MD, a geriatrician at Harvard Medical School and director of geriatrics education at Beth Israel Deaconess Medical Center in Boston, told a group of hospitalists at a regional Society of Hospital Medicine (SHM) meeting this summer, questions about orientation do little to pin down problems affecting a patient's cognitive function.
While asking patients if they know the date may be a favorite diagnostic technique, she said, that line of questioning usually doesn't give you the information you need to make the right diagnosis.
Here are some tips for residents—and all physicians who care for elderly patients—on how to recognize and treat delirium.
A big part of the problem, Dr. Leff explained, is that many physicians aren 't exactly sure what to look for in a patient with delirium.
"Housestaff haven't been trained to see inattention," he said. "They may say a patient is 'out of it' or 'uncooperative.' That's when they need to step back from that assessment and consider the possibility of delirium." Many physicians, he told internists at the meeting, "never received a working vocabulary to describe someone's consciousness or the sum of their mental functions. We don't always have the tools in our toolbox."
And if you're trying to spot delirium by asking your patient to answer simple questions like where she is, today's date and the name of the current president, it may be time to take another approach. Such general questions aren't specific enough, said Dr. Fabiny, and patients can actually learn the answers to such questions if they're asked enough times.
"I've had patients who say, 'They keep asking me where I am and what day it is,' " Dr. Fabiny told the SHM group. " 'Now I know I'm at Beth Israel Deaconess and I know it's Tuesday.' "
If you're trying to distinguish delirium from dementia, you're better off assessing factors like the patient's attention level. "A demented person doesn't become inattentive until he's in the advanced stages of illness," Dr. Fabiny said. "A demented person living in a nursing home wouldn't be inattentive unless she is also delirious or very demented."
Dr. Leff said that techniques like the digit span test—asking patients to repeat a three-digit number, then a four-digit one, then a five-digit number and so on—can help you assess patients' attention level. Patients who aren't delirious can easily remember four- and five-digit numbers, while simple repetitive sequence can also yield clues.
The confusion assessment method
While these lines of questions can give you insight into a patient's state of consciousness, Dr. Leff was quick to point out that these techniques aren't enough to formally diagnose a patient as delirious. To do that, both he and Dr. Fabiny suggested a relatively recent tool.
Because inattention is such an important symptom of delirium, it is one of the primary diagnostic criteria of a screening tool known as the confusion assessment method (CAM). Dr. Fabiny said that the method is not only easy to use, but more effective than other tools like the Folstein Mini Mental State Exam.
"The Folstein tool," Dr. Fabiny said, "has been misused for years as a screening tool for delirium."
The confusion assessment method, which was published in the Dec. 15, 1990, Annals of Internal Medicine, uses four measures. To make a diagnosis of delirium, you need to confirm the presence of both of the first two factors, as well as either the third or the fourth.
Inattention. When using the confusion assessment method, you should first measure a patient's attention. If patients can't pay attention to you, that's a sign that may be suffering from delirium. "Do you have to constantly repeat questions because their attention wanders?" Dr. Leff asked. "You start a question, then they start looking at the wall a bit? They're awake, but they're just not focusing, and they come back to you when you ask a question. That's inattention."
The Folstein exam uses "serial sevens" to test the patient's attention, but Dr. Fabiny said older women often complain that they have poor math skills. An alternative test is to spell the word "world" backwards, but that's not particularly easy if English isn't your first language.
What's the best way to test for attention? Ask the patient to name the days of the week or the months of the year backwards. And if you have a patient who doesn't speak English well, Dr. Fabiny said, use a translator.
Acute onset/fluctuating course. Collect information about both of these aspects of the patient's mental state from ancillary sources like family members. "That's always an important source of information when you're trying to assess a pa tient's cognitive function," Dr. Fabiny said.
Disorganized thinking. "In my experience," she explained, "disorganized thinking is almost always present [in delirious patients]. The way you test for disorganized thinking is to determine whether the person is tangential, rambling, incoherent or irrelevant in her responses."
For instance, does the patient's conversation switch illogically from subject to subject? Do patients answer a different question than the one you asked? And are they noticing things in the room and talking about them while you're asking them a question?
Altered level of consciousness. Levels of consciousness are categorized as follows: alert (normal); vigilant (hyperalert); drowsy (arousable by voice alone); stupor (arousal requires physical stimulus); or coma (unarousable). If a patient falls asleep while you're speaking to him, said Dr. Fabiny, he is probably delirious, sleep-deprived or both.
Finding the cause
Once you've diagnosed delirium in a hospitalized patient, you need to identify its cause. Before you send your patient for a head CT or a lumbar puncture, however, Dr. Fabiny said it's important to look for more obvious explanations.
For instance, consider whether any drugs, either alone or in combination with other agents, are causing the patient's delirium. "I once had a teacher say that whether older patients are confused or not," Dr. Fabiny said, "a new symptom should always be considered a side effect of a medication until proven otherwise."
Dr. Leff agreed. "You have to start with medicines," he said. "In my experience, that will provide your answer in most cases."
While drugs like diphenhydramine (Benadryl) can obviously inter fere with cognitive function, Dr. Leff said he has seen nonsteroidals, COX-2 inhibitors and even antibiotics like ciprofloxacin cause delirium.
And while you're thinking about medications, take a moment to consider whether alcohol could be affecting your patient's state of mind.
Just because a patient is homebound, said Dr. Fabiny, doesn't mean she doesn't have access to alcohol. "I had one patient who was confined to a wheelchair but had alcohol delivered to her," she explained. "The patient could still use the phone."
Other common causes of delirium in the elderly include pain and infection. Because pain in older people can cause acute confusion, you need to explore whether the patient is suffering from untreated pain.
"In many cases of delirium with nursing home patients," Dr. Leff said, "it's caused by pain that's not easily identifiable, such as a dental problem the patient can't tell you about."
You also need to check the patient's bowels and bladder. According to Dr. Leff, fecal impaction or urinary retention can cause delirium, so a rectal exam may be necessary for a diagnosis. "One possibility is that the patient hasn't had a bowel movement in four days." he said. "Someone who is confused can't tell you that." Acute urinary retention in men can likewise cause mental confusion.
Also consider possible cardiovascular problems. Potential issues can range from atrial fibrillation to acute myocardial infarction.
"One thing that's noticeable about this list is that the central nervous system is not on it," Dr. Fabiny pointed out. "There is no indication for a head CT or a lumbar puncture based on any of these possible causes of delirium. Put those procedures on the bottom of your list and order them only if there is a specific indication for them."
At the same time, you need to "make sure that everyone gets an ECG, and that it is read," she added. "Make sure you know how much urine is in a man's bladder and ask about pain. Then, if you can't find anything, consider a head CT."
A more subtle cause of delirium, particularly in older patients, is reduced sensory input.
"An older patient can come into the hospital without confusion," Dr. Fabiny said, "but if he spends two days in there without his glasses, hearing aid and teeth, and the television is on all the time and the blinds are drawn, that can cause him to become confused, regardless of whatever underlying illness is present."
The good news is that to some degree, you can control environmental factors that lead to delirium. Dr. Fabiny said she asks older patients or their family members if patients need their eyeglasses, false teeth or any other essentials. She also checks to see that during the day, window blinds are open and the patient is up and out of bed when possible.
"Make sure they have everything they need to orient themselves to their environment," she explained.
Dr. Leff presented information from a clinical trial that examined a group of more than 800 elderly inpatients who were free of delirium when they were admitted. (Study findings were published in the March 4, 1999, New England Journal of Medicine and are online.) Researchers found that several strategies helped reduce the incidence of delirium in the group.
For example, staff talked to patients several times a day and tried to keep them up to date on current events. And to make sure patients were sleeping well, researchers used a nonpharmacologic "sleep drill." Instead of sedating patients with diphenhydramine or a benzodiazepine, they gave patients a massage and a glass of warm milk, and tried to reduce unnecessary noise in the hospital.
To further help patients get on a regular sleep cycle, staff got patients out of bed regularly, even if physical therapists weren't seeing the patient that day. They also reduced their use of restraints—even devices like IVs—that tend to keep patients tethered to their beds. Researchers found they could cut the incidence of delirium from 15% in a study control group to 9% in the intervention group.
According to Dr. Leff, pharmacologic therapy may be indicated in cases of severe agitation where the patient may be a threat to herself or others. But when he does use drugs for sedation, Dr. Leff said, he stays away from benzodiazepines. Research has shown that benzodiazepines produce side effects, he said, and don't effectively sedate these patients.
Instead, he uses a drug like haloperidol (Haldol) and starts with doses of 0.5 mg to 2 mg. He repeats the dose every 30 minutes until he's achieved sedation, and then identifies a maintenance dose.
Finally, when caring for patients with delirium, remember that the condition fluctuates. Everyone, especially the patient's family, needs to understand that patients will experience periods of lucidity as well as periods of confusion.
"I have house officers tell me that their patient was completely clear in the morning," Dr. Fabiny explained. "Then later in the day, they get a call from nursing saying she is confused. They start wondering if something is going on."
Over time, the patient's lucid periods should become longer while periods of confusion should shorten. Be concerned if the patient appears to be getting better but suddenly worsens.
"With these patients," Dr. Fabiny said, "you need to consider whether a new medical problem is manifesting itself. Older hospitalized patients are at increased risk for iatrogenic complications."
Finally, explain to patients' families that delirium may last for up to two months before it clears.
"When you're talking to the patient's family at discharge," said Dr. Fabiny, "educate them about the typical course of delirium. Explain that the medical problem has been treated and the patient is getting what she needs, but that the confusion may last a while."
She's seen patients bounce back to the hospital after they've been sent home with a spouse. "Twelve hours later, the patient is confused and the wife is panicked," she said. "She thinks he's sick again because nobody told her that this could go on for days, weeks or months."
Edward Doyle is a freelance health care writer in Chalfont, Pa.
The information included herein should never be used as a substitute for clinical judgment and does not represent an official position of ACP.
Internist Archives Quick Links
Internal Medicine 2014 Advance Program Now Online
Details about Internal Medicine 2014's robust offerings are now available on our Website! Use the online Schedule Planner to build and save your own schedule.
New from ACP: Free High Value Care Cases
Free MOC and CME Credit with easy online submission
HVC Cases provide:
Get started now with your ACP user name and password!