Non-pharmacological Management of Delirium: A Proactive Approach

One of the most challenging patients in health care is the “agitated delirious patient.” Just as challenging, for different reasons, is the lethargic confused patient who “just isn’t getting better.” The patient depicted by the drawing above was both.

The geriatric medicine team was consulted on day seven of this hospitalization because of his persistent confusion and inability of the primary team to “find placement.” According to his wife at the bedside, her 80 year old husband had fallen and fractured his hip four weeks ago, had hip surgery, “which went well” but had some “problems with mild confusion” after the surgery. He went to a skilled facility “for rehab” and then developed pneumonia within two weeks, requiring a second hospitalization. He returned to the skilled facility and was making some progress but then developed “shakes” and more confusion, which caused him to be admitted for this, his third hospitalization. The primary team had done a complete workup for reversible causes (including CT head, MRI head, lumbar puncture, EEG, and several urinalyses and cultures).

The patient was on a typical medical-surgical floor, not the intensive care unit (ICU). On our exam, we observed a rather “big” man (i.e. not thin or frail) in bed, occasionally opening his eyes and moving various parts of his body (his head at times, then his extremities as much as he could). Every few seconds, he seemed to get a surge of energy, would mumble something, and then try to sit up, at which time the nurse would say, “now, lie back down.” He would try again, then would wear out, and then close his eyes again. While observing in order to get non-verbal clues from the patient, we counted the number of restraints and tethers: NINE. What can one do in this situation?

The medical workup had been exhausted, and the patient was not getting any better. Although the primary team and nurses had been trying to maintain “non-pharmacological” measures, the “reactive” mode of management took over: The patient seemed to be a “danger to himself” and had been combative towards the staff during typical care so physical restraints were used. Then, various antipsychotics were tried, which only seemed to make things worse. If non-pharmacological approaches are going to be successful, they have to be proactive. The following are three examples of proactive principles in the non-pharmacological management of delirium:

  1. The brain works better when it is upright
  2. Delirium goes down as ambulation goes up
  3. The “T-A-DA method” of managing delirium is not magic: it is hard work and based on 18 years of experience in a hospital unit that cares for delirious older patients without the use of any physical restraints.

1) The brain works better when it is upright.

The anecdotal evidence: our patient above. One by one, we removed the tethers and the restraints. We got to a point that he was free enough to allow us to sit him up. Although he was a big man, we were able to get his legs over the side of the bed. He did not have good sitting balance, but the longer he sat up, the more alert he became, and as he became more alert, he was able to sit without falling over. We almost stopped there, but I knew we couldn’t because the patient started to move, as if he wanted to get out of the bed. He couldn’t step very well, but, eventually, he made it to the chair. As he got his breath, he looked over at his wife and said “eh…that’s…better” -- his first intelligible words in a week.

During the next two days, we implemented the mainstay of our “treatment:” getting him out of bed. Our other treatments (interventions) included getting the urinary catheter out, getting the feeding tube out, sitting him in a chair to eat and drink and stopping the antipsychotics. The delirium started to clear, and he was discharged to the skilled facility on our fourth day of consultation in a better place than when he came in.

The indirect evidence: Studies of healthy young volunteers subjected to prolonged bed rest have shown changes in gray matter and white matter of the brain, negative effects on executive function and changes in certain cytokines. Although this research may seem space ages away from older frail patients in the hospital bed, it does point to the detrimental effect of bed rest on the brain.*

2) Delirium goes down as ambulation goes up.

The indirect evidence: In a landmark trial of delirium prevention in a medical-surgical hospital population, delirium incidence was approximately one-third less in the intervention group compared to usual care. The intervention consisted of multiple components (protocol for sleep, fluid repletion, attention to hearing/vision, reorientation, decreasing unnecessary medications and early ambulation). However, probably the most important part of the intervention was the early ambulation.

The direct evidence: Early mobility among patients in the ICU is associated with a decreased number of days with delirium. In a study of ICU patients on the ventilator, early therapy (day 1-2) compared to usual care (initial therapy day 7 on average) resulted in 2 days (SD 0-6) of delirium in the ICU compared to 4 days (SD 2-7), respectively (p Value=.03).

3) The T-A-DA method of management: Tolerate, Anticipate and Don’t Agitate.

A note of caution: although on the surface this may appear to be a low-key and passive approach, it is very proactive and takes a significant amount of training, culture change and effort.

Tolerate: Although tolerating certain behaviors may seem dangerous and even contrary to our training to keep patients safe, (e.g. when patients try to get out of bed by themselves), allowing patients to respond naturally to their situation while under close observation (which often means standing or sitting very close by), gives the patient some semblance of control in their confused state. More importantly, it also allows the healthcare professional to get clues about what might be bothering the patient. Imagine a patient so confused that he or she cannot communicate the need to empty his/her bladder. Climbing out of bed might be the first symptom of a full bladder!

Anticipate: It should be anticipated that patients with delirium will pull on anything that is not normally present. This is not a pejorative principle, but one of preparedness. If this happens, the “ready” caregiver has some options:

  • “Hide” these unnatural attachments.
  • Use a decoy. For example, tape a false IV on (not in) the patients non-dominant arm.
  • When an “attachment” is needed, try to use it briefly, then get rid of it or hide it. For example, give IV fluids as boluses, instead of a continuous rate. Cover up the precious IV in between the boluses.
  • Try to get rid of attachments that are not completely necessary. The seemingly standard telemetry monitor and oxygen tubing that most patients get are two overused attachments.
  • When attachments are necessary, stay flexible in their use. For example, it might seem imperative that we have minute-by-minute recordings of heart rate and rhythm for a patient with uncontrolled atrial fibrillation, but getting the patient to wear the monitor an average of 30 minutes per hour might be better than agitating the patient.
  • Getting out of bed is as natural as eating and toileting. This action is so anticipated and encouraged that “standby observation” should be the standard rather than standby assistance.

Don’t agitate: This is one of the most obvious yet subtle principles of delirium care. There are numerous potential “agitators” in the hospital environment, some of which will agitate certain delirious patients while calming others. Lights, visitors, television and music are just a few of these. Some agitators are predictable, many are not. Reorientation is one of the unpredictable techniques. If it helps, good; but if it’s not helping, don’t keep doing it.

The T-A-DA method is based on nearly two decades of experience in a specialized unit called the Delirium Room (DR). The DR is a 4-bed unit, within an Acute Care of the Elderly Unit. It is free of physical restraints and emphasizes the non-pharmacological approach to care of older hospitalized patients with delirium. Based on two different retrospective sets of data, negative outcomes associated with delirium, such as loss of function, longer hospital stay and increased mortality, can be decreased to levels seen in patients without delirium. Falls are also less frequent in the DR compared to typical rooms on the ACE Unit. The concept has been used in other countries (Singapore, Hong Kong and Australia) with some studies showing positive effects on delirium outcomes.

In summary, whether delirium is hyperactive, hypoactive or both, there is a role for a proactive non-pharmacological approach consisting of at least three principles:

  1. The brain works better when it is upright
  2. Delirium goes down as ambulation goes up
  3. The T-A-DA method: Tolerate, Anticipate, Don’t Agitate

*These studies were done related to research on how space flight affects the brain and cognition


Joseph H. Flaherty, MD
Professor and Associate Chair of Medicine
Geriatrics Division
Saint Louis University School of Medicine



Dr. Joe Flaherty's post describing three proactive principles for the care of delirious patients is sage advice for all clinicians. It is always better to begin treatment with non-pharmacological interventions if for no other reason than they are generally safe. What Dr. Flaherty presents is practice-based evidence accumulated over his 20+ years of clinical experience- evidence not to be ignored! I will add some of my own observations to his list of pro-active, non-pharmacological approaches that support brain/cognitive function. Then I’d like to comment on the empirical evidence (or lack of) for non-pharmacological interventions in the treatment of delirium.
Other non-pharmacological approaches that may support brain/cognitive function in the delirious patient (some are part of the widely-used HELP program) include:
- Ensure that sensory aides (glasses; hearing aids, amplifiers) are in good working condition and worn at all times. This helps the patient interpret their environment more accurately and reduces confusion.
- Use good communication skills when speaking with a delirious patient. Because attention can be impaired, speak in clear, short sentences. Avoid the use of “elder speak” (i.e. words such as “honey,” sweetie pie”) with older patients; it is associated with increased agitation, particularly in those with dementia.1
- Provide cues that orient the patient to their environment: use signage to indicate the bathroom; reduce excess noise; adjust lighting levels to reflect the hour of the day; use sitters appropriately.
- Offer activities that provide cognitive stimulation: a favorite game; simple puzzles; engaging conversation with a family member.
- Avoid unnecessary medications.
As Dr. Flaherty points out, much of this seems simple or low-key, but implementation takes a culture change. This is not easy to do. I encourage readers to access Dr. Jim Rudolf’s August, 2015 post on “Building a Quality Program Around Delirium.” It outlines ways to implement culture change.
Now my second point. Recent systematic reviews indicate that there is empirical evidence for the efficacy of multi-component non-pharmacological interventions when used to prevent delirium2, but little evidence to support their use in the treatment of delirium. 3 One reason for these findings may be that, with few exceptions, the individual components that made up these multi-component interventions were rarely implemented in a systematic, consistent fashion or in a “dose” that was likely to produce an effect. An exception may be early ambulation and/or use of occupational/physical therapy in critically ill patients.4 Rigorous prospective clinical trials are needed to test the independent contribution of individual components of interventions used for the treatment of delirium. A recent example of such a clinical trial is that reported by Jackson and colleagues.5 Their intervention of combined cognitive and physical therapy improved executive function and instrumental activities of daily living in cognitively impaired survivors of critical illness when compared to physical therapy only.5 All interventions require staff time regardless of complexity, and using only those components with known benefits, rather than a “kitchen sink” approach, will improve the quality and cost of delirium care.

1. Herman RE, Williams KN. (2009). Elderspeak's influence on resistiveness to care: focus on behavioral events. Am J Alzheimers Dis Other Demen. 2009 Oct-Nov;24(5):417-23. doi: 10.1177/1533317509341949.
2. Hshieh, T., Yue, J., Oh, E., Puelle, M et. al. (2015). Effectiveness of multicomponent nonpharmacological delirium interventions: A meta-analysis. JAMA Internal Medicine 175 (4), 512-520. doi: 10.1001/jamainternmed.2014.7779.
3. Abraha, I., Trotta, F, Rimland, J. et. al. (2015). Efficacy of non-pharmacological interventions to prevent and treat delirium in older patients: A systematic overview. The SENATOR project ONTOP Series. Plos One 10(6): e0123090. doi: 10.1371/journal.pone.0123090.
4. Schweickert, W., Pohlman, M., Pohlman, A., Nigos, C., Pawlik, C., Esbrook, E. et. Al. (2009). Early physical and occupational therapy in mechanically ventilated, critically ill patients: A randomized controlled trial. The Lancet, 373, 1874-1882.
5. Jackson, J., Ely, W., Morey, M. (2012). Cognitive and physical rehabilitation in intensive care unit survivors: Results from the RETURN randomized controlled pilot investigation. Critical Care Medicine, 40 (4), 1088-1097.


Excellent article


Great comments about hypoactive delirium, which I not infrequently get called to evaluate in my role as a psychiatric consultant, although the primary team usually misidentifies it as depression, which it mimics. Hypoactive delirium can also evolve into a catatonic variant of delirium,

We had a great Clinical Problems in Consultation in Psychiatry (CPCP) today by one of our clinical pharmacy residents, Luke Watson, Pharm.D. on pharmacologic treatment of hypoactive delirium, which you can view at