Published in the January 2011 issue of Today’s Hospitalist
You’d think that a medical condition afflicting more than one-third of inpatients that has the same long-term morbidity and mortality risk as heart attacks would have the full attention of hospitals and hospitalists.
But relatively speaking, it doesn’t. Hospitalists frequently don’t screen for delirium, and it often goes undetected until patients start pulling out their tubes. Even then, delirium can be addressed haphazardly, without a systematic approach that can reduce its duration and morbidity.
"The biggest problem is that it’s not recognized, and many hospitalists don’t systematically assess for it," says David Frenz, MD, a behavioral health hospitalist at HealthEast Care System in St. Paul, Minn. "A lot of doctors, including hospitalists, don’t take it seriously enough, and sometimes don’t even chart it. And once it develops, doctors do what they always do: Throw medications at it."
But delirium’s disturbing statistics “one-year mortality rates of 35% to 40% after discharge and an annual price tag in the U.S. of nearly $7 billion ” warrant a full-on campaign.
"Once you identify it, delirium is a psychiatric emergency," Dr. Frenz adds. "It immediately becomes the No. 1 issue on the problem list."
Delirium is that important, he explains, because morbidity goes way up and patients can become persistently confused. "It can take months, and sometimes may never clear “and then you’re in the difficult position of having to tell the family."
Suspect it in everyone
Hospitalists recognize that many cases of delirium can be traced to medications. (Other top offenders include infection, stroke, dehydration and myocardial infarction.) Given the complexity of patients that hospitalists treat, Dr. Frenz thinks physicians should suspect delirium in each and every one of their patients until proven otherwise.
He cites this common scenario: The 81-year-old post-op hip fracture patient on opioid pain meds and a laundry list of ortho-ordered PRNs for everything from nausea and spasm to anxiety, insomnia and narcotic-associated itch. By day 2, she’s taken the PRNs and is mildly confused. By day 3, when she’s agitated and headed for full-blown delirium, the nurse loads her up on benzodiazepines.
"Post-op order sets are a witches’ brew," Dr. Frenz says. "If you keep patients on the medications that caused the delirium “opioids, antihistamines, anticholinergics ” and then add medications to control behaviors, you’re exacerbating the problem."
While benzodiazepines are often used in the ICU to treat confusion, he adds, they tend to fuel and prolong delirium. A better move, Dr. Frenz advises, is to go straight to haloperidol for behavioral control.
Viviane Alfandary, MD, a hospitalist in Walnut Creek, Calif., knows the "drug-deluged" patient well. But these days, Dr. Alfandary is on the lookout for drugs even beyond the post-op PRNs that can be, she notes, "hugely problematic. I’ve even had patients become delirious on Lidocaine patches or Benadryl."
Dr. Alfandary points to other offenders, such as antibiotics like levofloxacin and ceftriaxone. And hospitalists, she thinks, may not be aware of the anticholinergic effects of certain commonly prescribed drugs. Those include new drugs for bladder control and urinary retention, such as oxybutynin and tolterodine (Detrol), and memory drugs like donepezil.
Not the usual suspects
It’s also important to look beyond the elderly for delirium, a step that some hospitalists may fail to take.
Dr. Frenz recalls a recent post-op orthopedic patient who was transferred to his hospital’s psychiatric unit because the staff at another hospital thought she was suicidal. No one even considered that she might have delirium, which she did. "No one is looking for delirium in a 25-year-old day-surgery patient," he says.
Dr. Alfandary, who practices at John Muir Medical Center, offers two more common non-traditional delirium patients: young diabetics and patients with adrenal insufficiency. She has treated two diabetic patients in their 20s with undiagnosed groin abscesses that caused delirium, not the diabetic ketoacidosis that was suspected. For the young patient with adrenal insufficiency, the culprit was the poor absorption of steroids through her J-tube.
And according to Seattle hospitalist Sandeep Sachdeva, MD, delirium in middle-aged patients with medical issues is similarly overlooked.
"It’s usually the elderly and infirm," says Dr. Sachdeva, who practices at Swedish Medical Center, "but there are plenty of others “like the 55-year-old with bad vasculopathy who has had TIAs, isn’t physically active and is on multiple medications. He’s as much at risk as the 85-yearold with medical issues, so you have to keep an open mind. We are doing a disservice if we talk about delirium as just an old person’s disease."
Checking status at every visit
To screen for delirium, Dr. Frenz takes an informal but effective approach. He obtains baseline mental status on all patients “and rechecks status at every visit “using what he calls his "30-second assessment." With patients who can talk, he asks them to tell him who they are, where they are, and the date and day of the week. He also asks them to spell a five-letter word backwards, and he may test memory by "spotting" the patient three words and asking them to recall them five minutes later.
"You can pick up a lot of subtle changes “evolving delirium “that way," he notes. Dr. Frenz also says that hospitalists could consider having nurses conduct that assessment.
Dr. Alfandary starts by zeroing in on patients at high risk: all elderly patients and those of any age with chronic conditions or multiple comorbidities. She also takes both a sleep and constipation history; while delirium is "almost always infection or medications, or both, there’s usually a component of undiagnosed constipation or urinary retention," she says. She also takes a "sun-downing" history for patients with diagnosed or suspected dementia.
And to assess patients’ baseline mental status, Dr. Alfandary relies on family members whenever they’re available. That’s particularly crucial in identifying the so-called "quiet" or complacent patient with delirium who goes undetected much more frequently than one who’s raving.
"With my elderly patients, I talk to the family members almost every day, because they’re the ones who can give me the most accurate information about the patient’s mental status," Dr. Alfandary says. "A lot of times, I won’t be able to tell if the patient is delirious, but a family member will."
It’s also family members who reveal the undisclosed alcohol abuse or long-term benzodiazepine use that can set off delirium two days after a hip replacement. "We’ve got a lot of 80-year-olds who are still on the two-martinis-before-dinner regimen, and they don’t tell us or the surgeon," Dr. Alfandary points out. "But the family member knows."
Edward Ma, MD, a hospitalist with Medical Inpatient Care Associates in West Chester, Pa., concurs.
"The way to lower the threshold on delirium detection is discussions with family members," Dr. Ma says, adding that he always tries to get buy-in from the family that a patient is ready “physically, socially and mentally “for discharge.
"They’re the ones who’ll tell you, ‘Dad could put on his shirt yesterday, but today he’s trying to use it as toilet paper.’ " Or, Dr. Ma adds, he may be treating a pneumonia patient where the family tells him the patient’s breathing looks great but she seems confused. " ‘Mom said she saw a dog in the room,’ " he has heard, "or, ‘She’s asking for dad, who died 15 years ago.’ " That simple exchange, says Dr. Ma, has stopped him from discharging a patient who he thought was otherwise coming along well.
Moving toward protocols
When it comes to screening and treatment, hospitalist Martin Johns, MD, says his hospital, Gifford Medical Center, doesn’t have a delirium protocol. But because Gifford is a 25-bed critical access hospital, hospitalists and nurses have the relative luxury of staying in close contact with fewer patients during the hospitalization. Even then, Dr. Johns says, it can be hard to differentiate between delirium and mental illness.
"We use the mini-mental status exam that’s been modified for patients who have dementia already," Dr. Johns points out. "If there’s still ambiguity, we have our speech therapists do full cognitive testing as an objective way to assess patients who keep bouncing back."
But larger hospitals don’t have the luxury of that close contact. Seattle’s Swedish Medical Center has launched an aggressive educational campaign to increase the recognition of delirium. That grew out of the medical center’s initiative for its Brain Attack Response Team (BART), a rapid response team for in-house strokes.
"Many patients we were called for “45% to 50% “actually had delirium, not focal neurological disruption," recalls Dr. Sachdeva, a BART member. "But those BART tools helped us more quickly correct some obvious causes of delirium “opioids and anticholinergics “and to look for CO2 retention in COPD and blood sugars in the possible closet alcoholic patients."
In addition to learning what to look for, nurses and physicians are now charged with documenting delirium in the electronic medical record. The next step will be to improve system-wide use of nonpharmacologic and other interventions. (See, "How to prevent and treat delirium without meds." below)
But while there’s no hospital-wide delirium protocol yet, Dr. Sachdeva says he’s now much more proactive in his approach.
He includes chart notes indicating "patient at risk for delirium" and shares those concerns with the family. "I tell them that so they won’t freak out if the patient seems confused," Dr. Sachdeva says. "At least then they know it’s a possibility, and that leads to a discussion on prevention strategies." He also tells nurses to call right away when they suspect delirium.
The good news
On the upside, when delirium is caught early, clearing it might be a matter of a few days. The first step is to remove offending or exacerbating medications as quickly as possible. If the drugs can’t be stopped, doses should be reduced “although Dr. Sachdeva points out that patients may be sensitive to even lower doses of the offending medication. And for patients at high risk, start with lower doses.
"Medical doctors are famous for medicating away every problem," Dr. Johns says, "but it’s important to keep in mind that you don’t have to start with the highest dose of Dilaudid for everyone’s pain."
When someone else has done the prescribing, especially for elderly patients who are now delirious, Dr. Alfandary may cut those doses by as much as 75%. Frail elderly patients with vertebral fractures who develop delirium may get 2.5 milligrams of Flexeril instead of 10, and 1 milligram of Valium instead of 2 or 5 mg.
"It’s the same thing with Percocet," she says. "I might use a quarter of a pill as a starting point."
If, however, alcohol withdrawal is the issue and family members indicate Mom or Dad probably won’t stop drinking once discharged, Dr. Alfandary might skip the "benzos" altogether and just give the patient alcohol ” which is stocked in the hospital pharmacy “on the same schedule that he or she is used to at home.
"That’s why it’s important to get the alcohol history and ask what time the patient has his first drink," she points out. "For some, it’s 10 o’clock in the morning."
Bonnie Darves is a freelance health care writer based in Seattle.
There’s a great deal that hospitalists can do to reduce the duration of delirium: Get rid of as many PRNs as possible, minimize the use of catheters, and get naloxone on board right away if patients are over-sedated or over-medicated for pain.
However, experts say that at least one-third of delirium cases may be preventable ” and even more may have their negative effects minimized “primarily through the use of non-drug interventions. Nonpharmacologic interventions are often underutilized, in part because they’re resource intensive, admits Christopher Mertz, PsyD, LP, a clinical neuropsychologist with the HealthEast Care System in St. Paul, Minn.
That may be changing, due to the growing awareness of delirium’s profound impact in terms of cost and morbidity. At Seattle’s Swedish Medical Center, for instance, nurses were quick to embrace a delirium-detection initiative and to form a multidisciplinary team.
"Usually when we ask nursing to help with a new initiative, there’s resistance because they’re already so busy," says Swedish psychiatrist Arpan Waghray, MD, who helped launched the campaign. "But there was none with this project because they know, of course, that it takes so much more in resources to deal with even one confused patient."
Here are nonpharmacologic interventions to consider:
But it’s also important to avoid under-stimulation. "Don’t have the shade drawn so the room is like a cave," he says. "If you leave patients alone in their delirium, they’ll have no basis for reality." Get post-op patients up and in a chair as soon as possible, preferably near a window. And according to Edward Ma, MD, a hospitalist with Medical Inpatient Care Associates in West Chester, Pa., measures like playing music in the patient’s room make "all the difference."
Finally, be aware that delirium can have serious psychological effects. While that strengthens the case for prevention and early detection, those effects themselves may need to be dealt with.
"With delirium, there’s always the potential for post-traumatic stress disorder," Dr. Mertz points out. "While some folks are completely snowed, others may remember things that happened when they were delirious, and that can be very upsetting."