NCBI Bookshelf. A service of the National Library of Medicine, National Institutes of Health.
Greer N, Rossom R, Anderson P, et al. Delirium: Screening, Prevention, and Diagnosis – A Systematic Review of the Evidence [Internet]. Washington (DC): Department of Veterans Affairs (US); 2011 Sep.
REVIEWER COMMENT | RESPONSE |
---|---|
1. Are the objectives, scope, and methods for this review clearly described? | |
Yes. The incidence of delirium is a significant complication of hospitalization that warrants further review. The ability for identification and prevention of delirium in medically ill patients is a current need. The objectives of this study were clearly stated and it appears that a large data base of research was examined to address the key questions posed by this review. | |
Yes | |
Yes | |
Yes | |
Yes. I think Key Question #1 includes multiple disparate elements “effectiveness” is really answered by question #3 diagnostic accuracy, as is vary in results. In the summary, only does screening improve clinical outcomes is answered. | Effectiveness is not adequately addressed by KQ3 “diagnostic accuracy”. While there was no direct evidence of the effectiveness and harms of screening for delirium we have described in the KQ1 results section the pieces of chain of evidence that would need to be addressed for indirect evidence of effectiveness. |
Yes | |
2. Is there any indication of bias in our synthesis of the evidence? | |
No. | |
No | |
No | |
No | |
No. Honestly, I have a sense of bias, but it is hard to identify the source. I am a little worried that your questions are so narrow that a naïve reader will say… well, there is nothing new here since 1970. When in fact, it is pretty clear that delirium is associated with mortality, that some drugs are used more commonly in patients who develop delirium, that haldol can attenuate the consequences of delirium, that benzodiazepines in patients at risk should be avoided……… | The scope of this report was not to assess all pharmacologic interventions that increase a person's risk of delirium. However, we have added categories of medications widely recognized to be associated with delirium. We also have described that delirium is associated with mortality. |
No | |
3. Are there any published or unpublished studies that we may have overlooked? | |
No. Not to my knowledge. | |
No. This is an amazing compendium of information, and I have little to add, especially given comment about authors' awareness of, and plans to include information from, June 2011 article in Annals of Internal Medicine. | Thank you |
No | |
There are pending publications from 2 studies (Boustani and Marcantonio) on cholinesterase inhibitors and their role in delirium prevention. | Our inclusion criteria required articles be published in peer review manuscripts. |
I don't know the literature sufficiently to know. | |
I am not aware of any studies that were overlooked. | |
4. Are there any clinical performance measures, programs, quality improvement measures, patient care services, or conferences that will be directly affected by this report? If so, please provide detail. | Thank you – we will share these suggestions with the people responsible for dissemination of the report. |
Presently OQP reviews inpatient records for evidence of elevated risk for delirium. Because these efforts are still in a relatively early stage, not much attention has been drawn to them—but as the data and the outcome correlations become more robust, educational efforts can be undertaken to support use of the QI and thereby, to enhance quality of inpt. care. Some of these data were presented at a recent VA conference (EES) in Indianapolis that focused on safety enhancement in different health delivery settings. In the two preceding years (2009, 2010), national conferences concerning delirium prevention, recognition, and management were also held in Boston and Baltimore. Plans are just beginning for a “Emergency Rooms and the Elderly Veteran” conference for Spring 2012. OQP is also examining a proposal from GEC to adapt a number of the “Assessing Care of Vulnerable Elderly” QIs to VA—several of these have to do with documenting mental status upon hospital admission in order to have a baseline against which subsequent mental status may be compared. The Office of Geriatrics and Extended Care currently supports several demonstration pilots (about to embark on their 3rd years of funding) specifically directed to delirium prevention in different settings: in San Francisco, an Acute Care for the Elderly unit; in Connecticut, a home-based presurgical assessment followed by postdischarge transition management; in Boston, a “Delirium Toolbox” for reducing risk factors in recent admissions with demonstrated elevated risk for delirium; in Durham, a caregiver education program to assist with behaviors associated with cognitive decline; in Indianapolis, a transition management approach that begins during an inpatient stay; and in New Orleans, Portland, Boise, and Honolulu, a “Hospital at Home” that provides an inpatient level of care in the home for targeted diagnoses, with complete avoidance of delirium. There is a national Dementia Steering Committee that developed a strategic plan and has educational, clinical, and research activities underway. Because dementia is one of the most concerning risk factors for delirium onset, this group's awareness of this information will unquestionably be of interest. The final report of the USH-chartered “Healthcare Workforce for Aging Veterans” Executive Taskforce has been the subject of three briefings with Dr. Petzel and, with his approval, is about to be presented to the National Leadership Board—it recommends focusing resources over the next 5 years on ensuring universal access within VHA to a single program in each of the inpatient, outpatient, and extended care areas—and for inpatient, that program is Geriatric Consultation, specifically targeting prevention, recognition, and management of inpatient delirium. Finally, the Deputy Under Secretary for Health for Operations and Management last month approved the formation of a Delirium Field Advisory Committee, charged with advising the GEC office on projects, programs, and activities that hold promise for enhancing awareness of and familiarity about delirium on the part of providers across the continuum of care. | |
There is a potential for diagnosis of delirium with some of the tools reviewed. There is some low level evidence of preventive medications and possibly staff education that are useful in preventing delirium. The prevention of delirium could affect performance measures such as length of stay, length of ICU stay, decrease in morbidity, and decrease in NHPPD. This could have a positive impact on patient flow and improved discharge to home settings. | |
The annual American Delirium Society conference and EES conferences during the last 3 years will be significantly impacted by these findings. In general, studies in the VA are nearly non-existent, yet VA eligible, VA using patients are sicker than any others in the country (Kazis data). In particular, younger veterans (Vietnam Era) have significant loads of comorbidity (often associated with PTSD as a contributing factor) and really need to be included in the “high risk” category although they don't meet usual age criteria. We may also see a need for OEF/OIF vets to be included for the same reason. | |
Inpatient nurses provide direct care to patients with delirium. The evidence in this report about nonpharmacological interventions to prevent delirium will be especially relevant to nurses in the acute care setting. Once the report is released, the Office of Nursing Services Evidence Based Practice Group will work with the Geriatric Nursing Field Advisory Committee to discuss how the information from this report on evidence-based nonpharmacological interventions can be disseminated to staff nurses and how we might enlist facilities to trial these evidence-based interventions. | |
This report has the potential to impact the standard of care relative to screening of older Veterans for delirium at point of care. | |
There is an ongoing quality improvement project in 5 – 7 ICUs measuring CAM ICU and RASS scores | |
5. Please provide any recommendations on how this report can be revised to more directly address or assist implementation needs. | |
Screening
| Screening
|
Diagnosis
| Diagnosis a., b‥ These are important points and we have included this information in the findings for KQ3 and the conclusions. Conclusions a., b. Thank you for the suggestions. We have attempted to address them in the Conclusions and Future Research Needs sections. |
The report points out the great amount of evidence in the field that is nonetheless non-definitive in its clinical application. The VA population (see above) really does require separate investigation. See Comments below. | We agree that the findings are generally of low-quality and/or insufficient. |
Given the evidence presented, it is clear that much more research is needed to identify valid and reliable means of improving detection of delirium. A screening measure that can be universally implemented is needed. The CAM alone does not seem sufficient for this purpose – it requires supplemental mental status testing prior to completion. (Key Question #3) Recommendations for who to screen based on currently available evidence (older, sensory or cognitively impaired) should be highlighted. (Key Question #1) | We again emphasize that there are no data about the effectiveness and harms of screening for delirium in hospitalized medical patients. Therefore, we disagree that a screening measure that can be universally implemented is needed (or at least that such an instrument “should be implemented”). The current evidence does not permit making recommendations on who to screen. |
More emphasis may also be placed on the non-pharmacological interventions for delirium prevention based in the evidence. These are low-cost, low-risk interventions.(Key Question #2) | We have added a table of risk ratios for the non-pharmacological interventions and more detail about the components of the multicomponent interventions. |
Limited evidence was reviewed regarding the need for education among providers that fail to recognize delirium across settings where Veterans receive care. (Key Question #1) | We have attempted to address this in the Key Question 1 conclusions. |
Flip questions 1 and 3. In the summary, when a reader starts with “no convincing improvement in clinical outcomes, no convincing difference with different drugs, …. Many people won't get to 3. They want validation that their standard of care is fine. There is a way to measure brain dysfunction (which we call delirium like in the 18th century). | The questions are listed in the order originally agreed upon. No further change. |
Additional Comments: | |
I think this was a very thorough review of the literature and it was disheartening to see that there is little substantiated evidence on screening, identification and prevention of delirium. I did find 2 typos – page 18, first paragraph, states “following up” should state “follow up” Page 37 – typo of control group “if”29-60% and should be control group “of” 29-60% | Thank you for your comments. We have corrected the typos. |
Investigations regarding deliriums that may be provoked ONLY by certain medication use (in the absence of other causes) would be very helpful; they may well have different prognoses than the multifactorial ones. This could help greatly because it would offer some “clean” recommendations that could easily be implemented very quickly through the VA. | Thank you for the suggestion. We have included this in the Future Research section. |
This is an important and complex topic for all staff who care for Veterans with delirium, in particular nursing staff who are with these Veterans 24/7 and understand the profound distress this condition causes for both Veterans and their family members. My comments are as follows: | |
Introduction-page 4 | |
Para 1: The 3 reasons that this review was undertaken are not listed in the order that the 3 key questions are discussed throughout this report (same inconsistency appears in the first paragraph of the Executive Summary) | We have corrected to ensure consistency. |
Para 2: The authors state that they were “careful to make important distinctions between screening for delirium and diagnosis of delirium.” This distinction is somewhat confusing in that the discussion of screening (para 1, page 13) suggests that the purpose of screening is to detect a condition before symptoms occur and the CAM is mentioned as a screen for delirium. Later, however, in the discussion of KQ3, CAM is discussed as a diagnostic tool (Key Question #3). Since the CAM items all address identifiable symptoms, is the CAM a screening test or a diagnostic tool or both? Are there any screening instruments for delirium that detect delirium in the preclinical state? Or are the delirium “screening instruments” really diagnostic instruments (tools)? | CAM could be used as both as a screening instrument in hospitalized patients (individuals without identifiable signs or symptoms of delirium) or as a diagnostic tool (patients with some signs or symptoms that are consistent with but not definitely determined to be delirium (e.g., a patient with confusion). KQ1 and the overarching goal of this report was to assess the effectiveness and harms as a screening tool including in individuals who may be at increased risk due to patient factors (e.g., age, personal history of delirium), index disease type or severity (e.g., stroke, ICU) or co-existing medical conditions/medications that are not directly the reason for admission (e.g. use of narcotics in a patient admitted for COPD). KQ3 assessed the use of CAM as both a diagnostic and screening tool as many of the studies evaluated patients with signs/ symptoms potentially compatible with delirium. |
Background (page 4) | |
In the 3rd sentence, paragraph 4, underlying causes of delirium are listed. The next sentence mentions “risk factors.” Are underlying causes of delirium different from risk factors? Is so, what are the risk factors for delirium? Are orthopedic and cardiac surgeries risk factors for delirium or underlying causes of delirium? Most of the pharmacological studies discussed in KQ2 targeted patients who underwent either cardiac or orthopedic surgery yet surgery is not mentioned in para 4 on page 13 either as an underlying cause or risk factor for delirium. | We clarified our use of the term “risk factors”. Causality is a strong term that definitely ascribes the outcome to the risk factor. We have clarified regarding surgery. |
Key Question 1 (page 13) | |
There is no discussion of who (MD, nurse, other staff?) would likely perform screening. In the screening studies/guidelines reviewed, was there mention of who completes the screening? This is an important question given that often the first contact a patient has in the inpatient setting is with a nurse. | This is a policy issue beyond the scope of the review. Screening if found to be effective could be implemented by several lines of health care staff including nurses and physicians and could be done at the admitting floor or in the clinic/emergency room where the admission decision was made. If screening for delirium is effective then future research should be conducted to assess the most effective/efficient methods for implementation. |
Key Question 2 | |
Pharmacolgocial Studies | |
Several different pharmacologic studies are discussed (pages 21-24 and 30-31). Most of the pharmacological studies with the exception of Dautezenberg et al (cholinesterase inhibitor) target patients who either underwent orthopedic or cardiac surgery. The report Conclusions on page 45 state, “Low level evidence suggests that pharmacologic strategies using analgesia via fascia iliaca compartmental block, antipsychotic, and lighter anesthesia may be useful in delirium prevention.”
| We have clarified regarding surgery. |
Non-pharmacological Studies | |
On page 25 the report mentions that 9 multi-component studies consisted of interventions that significantly decrease the incidence of delirium. In the report Conclusions (page 45), multi-component interventions are again mentioned. It might be helpful for those staff interested in implementing multi-component interventions if examples were given of the intervention bundles trialed in some of these studies. | We have added information about the interventions in the multicomponent studies. |
Overall Organization | |
While the discussion of each key question requires a somewhat different approach, there seems to be some inconsistencies in the overall organization of this report.
| We have corrected these inconsistencies. |
Page 6/88 Key Question #1. Consider adding the positives… Lacking direct evidence, ¾ criteria establishing an indirect link between screening and outcomes for delirium were satisfied: 1) patents with delirium have worse outcomes, 2) systematic screening likely improves detection, and 3)harms associated with screening are likely minimal. However, we viewed evidence that treatments for delirium are effective is mixed. | We have modified this section. Without a systematic review of the evidence for each criterion, we are hesitant to say that the criteria were satisfied. |
Page 17/88 Paragraph 1. Consider adding after Screening for disease or condition is warranted if the disease is serious …‥ if treatment or therapeutic decisions would be altered in the presence of the condition. | Thank you – we have modified this statement. |
This is an excellent, thorough review that emphasizes the need for research in delirium detection and prevention. I learned a lot by reading it. | Thank you. |