No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Discriminative Accuracy of Physician and Nurse Predictions for Survival and Functional Outcomes 6 Months After an ICU Admission
Abstract
Importance:
Predictions of long-term survival and functional outcomes influence decision making for critically ill patients, yet little is known regarding their accuracy.
Objective:
To determine the discriminative accuracy of intensive care unit (ICU) physicians and nurses in predicting 6-month patient mortality and morbidity, including ambulation, toileting, and cognition.
Design, setting, and participants:
Prospective cohort study conducted in 5 ICUs in 3 hospitals in Philadelphia, Pennsylvania, and enrolling patients who spent at least 3 days in the ICU from October 2013 until May 2014 and required mechanical ventilation, vasopressors, or both. These patients' attending physicians and bedside nurses were also enrolled. Follow-up was completed in December 2014.
Main outcomes and measures:
ICU physicians' and nurses' binary predictions of in-hospital mortality and 6-month outcomes, including mortality, return to original residence, ability to toilet independently, ability to ambulate up 10 stairs independently, and ability to remember most things, think clearly, and solve day-to-day problems (ie, normal cognition). For each outcome, physicians and nurses provided a dichotomous prediction and rated their confidence in that prediction on a 5-point Likert scale. Outcomes were assessed via interviews with surviving patients or their surrogates at 6 months. Discriminative accuracy was measured using positive and negative likelihood ratios (LRs), C statistics, and other operating characteristics.
Results:
Among 340 patients approached, 303 (89%) consented (median age, 62 years [interquartile range, 53-71]; 57% men; 32% African American); 6-month follow-up was completed for 299 (99%), of whom 169 (57%) were alive. Predictions were made by 47 physicians and 128 nurses. Physicians most accurately predicted 6-month mortality (positive LR, 5.91 [95% CI, 3.74-9.32]; negative LR, 0.41 [95% CI, 0.33-0.52]; C statistic, 0.76 [95% CI, 0.72-0.81]) and least accurately predicted cognition (positive LR, 2.36 [95% CI, 1.36-4.12]; negative LR, 0.75 [95% CI, 0.61-0.92]; C statistic, 0.61 [95% CI, 0.54-0.68]). Nurses most accurately predicted in-hospital mortality (positive LR, 4.71 [95% CI, 2.94-7.56]; negative LR, 0.61 [95% CI, 0.49-0.75]; C statistic, 0.68 [95% CI, 0.62-0.74]) and least accurately predicted cognition (positive LR, 1.50 [95% CI, 0.86-2.60]; negative LR, 0.88 [95% CI, 0.73-1.06]; C statistic, 0.55 [95% CI, 0.48-0.62]). Discriminative accuracy was higher when physicians and nurses were confident about their predictions (eg, for physicians' confident predictions of 6-month mortality: positive LR, 33.00 [95% CI, 8.34-130.63]; negative LR, 0.18 [95% CI, 0.09-0.35]; C statistic, 0.90 [95% CI, 0.84-0.96]). Compared with a predictive model including objective clinical variables, a model that also included physician and nurse predictions had significantly higher discriminative accuracy for in-hospital mortality, 6-month mortality, and return to original residence (P < .01 for all).
Conclusions and relevance:
ICU physicians' and nurses' discriminative accuracy in predicting 6-month outcomes of critically ill patients varied depending on the outcome being predicted and confidence of the predictors. Further research is needed to better understand how clinicians derive prognostic estimates of long-term outcomes.
... These findings support other studies that have shown that nursing notes may predict short-term patient outcomes more accurately than physician notes [29,30]. Nurses frequently summarize patients' situations by describing their symptoms, as well as their nursing actions and responses, without the restriction of structured forms [31][32][33]. ...
... Thus, nursing notes serve as a snapshot of patients' current statuses and exhibit a higher degree of freedom compared to physicians' notes, which provide a problem-focused summary. In a prospective cohort of patients who are critically ill, nurses predicted in-hospital mortality slightly more accurately than physicians, whereas the latter predicted long-term outcomes more accurately [29]. Huang et al [30] applied natural language processing to free-text nursing notes to predict multiple outcomes, including prolonged hospital stay or mortality, using the Multiparameter Intelligent Monitoring of Intensive Care III. ...
Background
Nursing narratives are an intriguing feature in the prediction of short-term clinical outcomes. However, it is unclear which nursing narratives significantly impact the prediction of postoperative length of stay (LOS) in deep learning models.
Objective
Therefore, we applied the Reverse Time Attention (RETAIN) model to predict LOS, entering nursing narratives as the main input.
Methods
A total of 354 patients who underwent ovarian cancer surgery at the Seoul National University Bundang Hospital from 2014 to 2020 were retrospectively enrolled. Nursing narratives collected within 3 postoperative days were used to predict prolonged LOS (≥10 days). The physician’s assessment was conducted based on a retrospective review of the physician’s note within the same period of the data model used.
Results
The model performed better than the physician’s assessment (area under the receiver operating curve of 0.81 vs 0.58; P =.02). Nursing narratives entered on the first day were the most influential predictors in prolonged LOS. The likelihood of prolonged LOS increased if the physician had to check the patient often and if the patient received intravenous fluids or intravenous patient-controlled analgesia late.
Conclusions
The use of the RETAIN model on nursing narratives predicted postoperative LOS effectively for patients who underwent ovarian cancer surgery. These findings suggest that accurate and interpretable deep learning information obtained shortly after surgery may accurately predict prolonged LOS.
... However, the accuracy of the estimations was influenced by different factors for the medical staff. Prior studies have analyzed factors related to discordance in prognosis between physicians and surrogate decision makers of critically ill patients and the discriminative accuracy of physician and nurse predictions 10,11 . However, these studies did not provide accurate prognostic accuracy. ...
... Most patients came from the emergency department, neurosurgery, general surgery, orthopaedic surgery, and cardiac surgery. The average APACHE II score was 12 (interquartile range, [9][10][11][12][13][14][15][16][17][18]. The mean SOFA score was 5 (interquartile range, [3][4][5][6][7][8]. ...
The proportion of correctly predicted prognoses and factors associated with prediction accuracy are unknown. The objective of this study was to explore the accuracy of physician and nurse predictions of 28-day mortality in the ICU. This was a prospective observational single-center study. All medical staff in the ICU have access to patient data, can communicate with patients or clients, and can independently predict the prognosis of patients within 24 h of patient admission. The only question of the questionnaire survey was: What is the patient’s outcome on day 28 (alive or death)? There were 2155 questionnaires completed by 18 physicians and 1916 submitted by 15 nurses. In the 312 patients included, the 28-day mortality rates were predicted by physicians and nurses. The overall proportion of correct prognosis prediction was 90.1% for physicians and 64.4% for nurses (P = 0.000). There was no significant difference in the overall correct proportion and average correct proportion among physicians with different seniority levels. The overall correct proportion and average correct proportion increased among nurses with seniority. Physicians in the ICU can moderately predict 28-day mortality in critically ill patients. Nurses with a seniority of less than 10 years in ICU cannot accurately predict 28-day mortality in critically ill patients. However, the accuracy of nurses’ prediction of patients’ 28-day prognosis increased with their seniority in the ICU.
... Others, including thought leaders and clinicians, propose that timelimited trials should be used when clinicians believe with near certainty that the patient will not survive, but our committee concludes this is an overly narrow conception. By focusing on the acknowledgment and management of uncertainty instead of on "poor prognosis," time-limited trials may safeguard against well-described prognostic inaccuracies and biases about which patients are likely to benefit from critical care (18)(19)(20)(21). Our conclusion is further supported by the dynamic and fraught nature of prognostication, because of ever-evolving technology and the inability of clinicians' prognostic estimates to account for individual patients' priorities and values about acceptable health states. ...
... For example, the largest investigation of timelimited trials to date included patients whom a clinician deemed to be at risk for "potentially nonbeneficial" ICU treatments (38). Although this practical approach resembles usual care, inaccuracies and biases in physicians' prognostication for mortality and other nonmortal outcomes (e.g., quality of life) are well described (18,21), and quantitative, prognostic algorithms have not fared much better (19,43,44). Furthermore, assessment of "benefit" from medical treatments is highly influenced by patients' and surrogates' cultural norms and expectations (45,46). ...
In critical care, the specific, structured approach to patient care known as a "time-limited trial" has been promoted in the literature to help patients, surrogate decision makers, and clinicians navigate consequential decisions about life-sustaining therapy in the face of uncertainty. Despite promotion of the time-limited trial approach, a lack of consensus about its definition and essential elements prevents optimal clinical use and rigorous evaluation of its impact. The objectives of this American Thoracic Society Workshop Committee were to establish a consensus definition of a time-limited trial in critical care, identify the essential elements for conducting a time-limited trial, and prioritize directions for future work. We achieved these objectives through a structured search of the literature, a modified Delphi process with 100 interdisciplinary and interprofessional stakeholders, and iterative committee discussions. We conclude that a time-limited trial for patients with critical illness is a collaborative plan among clinicians and a patient and/or their surrogate decision maker(s) to use life-sustaining therapy for a defined duration, after which the patient's response to therapy informs the decision to continue care directed towards recovery, transition to care focused exclusively on comfort, or extend the trial's duration. The plan's 16 essential elements follow four sequential phases: consider, plan, support, reassess. We acknowledge considerable gaps in evidence about the impact of time-limited trials and highlight a concern that, if inadequately implemented, time-limited trials may perpetuate unintended harm. Future work is needed to better implement this defined, specific approach to care in practice through a person-centered equity lens and to evaluate its impact on patients, surrogates, and clinicians.
... However, the ability of physicians to predict mortality is uncertain in critically ill patients [20,26]. The prediction of functional outcomes is even more uncertain and less studied [5,7], and few studies have investigated physicians' ability to predict functional outcome specifically for patients with msTBI [25]. There are prognostic models, such as the CRASH and IMPACT scores [6,22], validated in large cohorts [19], but rarely used in daily practice for many reasons. ...
... This notion of collegiality and improvement of the physician's prognostic capacities has been highlighted several times under the term wisdom of the crowds [11,12]. We also identified that physicians' level of confidence would be a factor in the accuracy of the predictions, as shown in another study conducted for patients hospitalized for more than 3 days in the ICU [7]. This feeling reflects the physician's ability to appreciate probability. ...
Background
The functional prognosis of severe traumatic brain injury (TBI) during the acute phase is often poor and uncertain. We aimed to quantify the elements that shade the degree of uncertainty in prognostic determination of TBI and to better understand the role of clinical experience in prognostic quality.
Methods
This was an observational, prospective, multicenter study. The medical records of 16 patients with moderate or severe TBI in 2020 were randomly drawn from a previous study and submitted to two groups of physicians: senior and junior. The senior physician group had graduated from a critical care fellowship, and the junior physician group had at least 3 years of anesthesia and critical care residency. They were asked for each patient, based on the reading of clinical data and CT images of the first 24 h, to determine the probability of an unfavorable outcome (Glasgow Outcome Scale < 4) at 6 months between 0 and 100, and their level of confidence. These estimations were compared with the actual evolution.
Results
Eighteen senior physicians and 18 junior physicians in 4 neuro-intensive care units were included in 2021. We observed that senior physicians performed better than junior physicians, with 73% (95% confidence interval (CI) 65–79) and 62% (95% CI 56–67) correct predictions, respectively, in the senior and junior groups (p = 0.006). The risk factors for incorrect prediction were junior group (OR 1.71, 95% CI 1.15–2.55), low confidence in the estimation (OR 1.76, 95% CI 1.18–2.63), and low level of agreement on prediction between senior physicians (OR 6.78, 95% CI 3.45–13.35).
Conclusions
Determining functional prognosis in the acute phase of severe TBI involves uncertainty. This uncertainty should be modulated by the experience and confidence of the physician, and especially on the degree of agreement between physicians.
... U nrecognized clinical deterioration is the most common cause of unplanned ICU transfer in hospital (1). Although physicians and nurses are able to predict death, critical illness, and recovery in hospital (e.g., area under the receiver operating characteristic curve [AUC] values ranging from 0.70 to 0.85) (2)(3)(4)(5), early warning systems are designed to systematically predict a patient's likelihood of clinical deterioration with the goal of protocolizing care escalation and expediting early intervention to prevent deterioration (6)(7)(8)(9) Simple early warning systems generate a pointsbased score from a small number of inputs (e.g., patient vital signs and mental status) (10,11) and have been widely implemented with varying effectiveness in reducing mortality (6). They are generally less accurate than physicians and nurses at predicting in-hospital mortality (4,(12)(13)(14)(15)(16)(17). ...
... Yet, we have poor understanding of how early warning systems interact with clinical judgment. Numerous studies have demonstrated that clinicians are modestly accurate in predicting death, critical illness, or recovery (2)(3)(4)(5)37). Simpler risk prediction tools are generally inferior to clinical judgment (4,(12)(13)(14)(15)(16)(17)37), but advances in data science have improved prediction of clinical deterioration (9,19,38). ...
Unlabelled:
Hospital early warning systems that use machine learning (ML) to predict clinical deterioration are increasingly being used to aid clinical decision-making. However, it is not known how ML predictions complement physician and nurse judgment. Our objective was to train and validate a ML model to predict patient deterioration and compare model predictions with real-world physician and nurse predictions.
Design:
Retrospective and prospective cohort study.
Setting:
Academic tertiary care hospital.
Patients:
Adult general internal medicine hospitalizations.
Measurements and main results:
We developed and validated a neural network model to predict in-hospital death and ICU admission in 23,528 hospitalizations between April 2011 and April 2019. We then compared model predictions with 3,374 prospectively collected predictions from nurses, residents, and attending physicians about their own patients in 960 hospitalizations between April 30, and August 28, 2019. ML model predictions achieved clinician-level accuracy for predicting ICU admission or death (ML median F1 score 0.32 [interquartile range (IQR) 0.30-0.34], AUC 0.77 [IQ 0.76-0.78]; clinicians median F1-score 0.33 [IQR 0.30-0.35], AUC 0.64 [IQR 0.63-0.66]). ML predictions were more accurate than clinicians for ICU admission. Of all ICU admissions and deaths, 36% occurred in hospitalizations where the model and clinicians disagreed. Combining human and model predictions detected 49% of clinical deterioration events, improving sensitivity by 16% compared with clinicians alone and 24% compared with the model alone while maintaining a positive predictive value of 33%, thus keeping false alarms at a clinically acceptable level.
Conclusions:
ML models can complement clinician judgment to predict clinical deterioration in hospital. These findings demonstrate important opportunities for human-computer collaboration to improve prognostication and personalized medicine in hospital.
... Previous research has demonstrated the ability of ICU nurses to accurately anticipate and predict survival and functional outcomes of patients. 32,33 Compared with predictive models limited to objective clinical measures, models that include subjective nurse and physician predictions had significantly higher discriminative accuracy for the patient outcomes. 32 Our findings demonstrate the importance of the subjective motor assessment performed by the bedside nurse, which should be taken into consideration during interdisciplinary rounds and decision-making in the ICU. ...
... 32,33 Compared with predictive models limited to objective clinical measures, models that include subjective nurse and physician predictions had significantly higher discriminative accuracy for the patient outcomes. 32 Our findings demonstrate the importance of the subjective motor assessment performed by the bedside nurse, which should be taken into consideration during interdisciplinary rounds and decision-making in the ICU. ...
Objective
In many institutions, intensive care unit (ICU) nurses assess their patients’ muscle function as part of their routine bedside examination. We tested the research hypothesis that this subjective examination of muscle function prior to extubation predicts tracheostomy requirement.
Methods
Adult, mechanically ventilated patients admitted to 7 ICUs at Beth Israel Deaconess Medical Center (BIDMC) between 2008 and 2019 were included in this observational study. Assessment of motor function was performed every four hours by ICU nurses. Multivariable logistic regression analysis controlled for acute disease severity, delirium risk assessment through the confusion assessment method for the ICU (CAM-ICU), and pre-defined predictors of extubation failure was applied to examine the association of motor function and tracheostomy within 30 days after extubation.
Results
Within 30 days after extubation, 891 of 9609 (9.3%) included patients required a tracheostomy. The inability to spontaneously move and hold extremities against gravity within 24 h prior to extubation was associated with significantly higher odds of 30-day tracheostomy (adjusted OR 1.56, 95% CI 1.27−1.91, p < 0.001, adjusted absolute risk difference (aARD) 2.8% (p < 0.001)). The effect was magnified among patients who were mechanically ventilated for >7 days (aARD 21.8%, 95% CI 12.4−31.2%, p-for-interaction = 0.015).
Conclusions
ICU nurses’ subjective assessment of motor function is associated with 30-day tracheostomy risk, independent of known risk factors. Muscle function measurements by nursing staff in the ICU should be discussed during interprofessional rounds.
... Survivors of critical illness and their families are often unaware of the physical, mental, and cognitive problems collectively referred to as post-intensive care syndrome (PICS), as well as the impaired quality of life (QoL), that can persist for months, or even years, after intensive care unit (ICU) discharge, consequently underestimating the long-term impact of critical illness [1,2]. Physicians too, are often overoptimistic and struggle to adequately estimate long-term prognosis [3][4][5][6]. This is understandable, as QoL is subjective and not solely determined by disability, making it challenging to estimate [7,8]. ...
To evaluate the effect of discussing personalized predictions of long-term quality of life (QoL) on patient and family experiences and outcomes, and on experiences of ICU clinicians.
We conducted a randomized clinical trial in two Dutch hospitals, assigning adult ICU patients to receive usual care or the intervention: discussing the expected long-term QoL based on a validated prediction model, during a family meeting in the ICU. Primary outcome was patient and family experience with shared decision-making (CollaboRATE, range 0–100), evaluated < 3 days after the family meeting. Secondary outcomes included ICU professionals’ experiences (Collaboration and Satisfaction about Care Decisions [CSACD] and Ethical Decision-Making Climate Questionnaire [EDMCQ]), symptoms of anxiety and depression among patients and family, and patients’ QoL 3 months and 1 year post-ICU.
160 patients were included, of whom 81 were randomized to receive the intervention and 79 to receive usual care. No significant differences were seen in patients’ and family members’ experiences (median CollaboRATE score 89 [IQR 85–100] in the intervention arm vs 93 [IQR 85–100] in the usual care arm, p = 0.6). The outcomes of patients did not differ, whereas at 1 year post-ICU family members in the usual care group reported a larger increase in depression symptoms (mean 2.3 [SD 4.2] vs 0.2 [SD 3.9], p = 0.04). Regarding ICU professionals’ experiences, an improvement in CSACD score was observed post-intervention (median 40 [IQR 34–45] vs 37 [IQR 32–43], p = 0.01), while no significant change in EDMCQ was found.
Incorporating personalized predictions of long-term QoL in family meetings had no measurable effect on patients’ and family members’ experiences. However, a positive effect on family members’ symptoms of depression and ICU professionals’ experienced collaboration was observed.
This study was registered at ClinicalTrials.gov: NCT05155150.
... P rognostication for critically ill patients with acute respiratory failure (ARF) is fraught with uncertainty (1)(2)(3), which contributes to the challenge of decision-making about the use of life-sustaining therapies like invasive mechanical ventilation. Time-limited trials (TLTs) are promoted by experts in palliative and critical care as a specific approach to navigate these uncertain but consequential courses of action (4,5). ...
IMPORTANCE
A time-limited trial (TLT) is a collaborative plan among clinicians, patients, and families to use life-sustaining therapy for a defined duration, after which the patient’s response informs whether to continue care directed toward recovery or shift the focus toward comfort. TLTs are a promising approach to help navigate uncertainty in critical illness, yet little is known about their current use.
OBJECTIVES
To characterize TLT use in patients with acute respiratory failure (ARF).
DESIGN, SETTING, AND PARTICIPANTS
Prospective 12-month observational cohort study at an U.S. academic medical center of adult ICU patients with ARF receiving invasive mechanical ventilation for greater than or equal to 48 hours.
MAIN OUTCOMES AND MEASURES
Primary exposure was TLT participation, identified by patients’ ICU physician. Patient characteristics, care delivery elements, and hospital outcomes were extracted from the electronic medical record.
RESULTS
Among 176 eligible patients, 36 (20.5%) participated in a TLT. Among 18 ICU attending physicians, nine (50%) participated in greater than or equal to 1 TLT (frequency 0–39% of patients cared for). Median TLT duration was 3.0 days (interquartile range [IQR], 3.0–4.5 d). TLT patients had a higher mean age (67.4 yr [ sd , 12.0 yr] vs. 60.0 yr [ sd , 16.0 yr]; p < 0.01), higher Charlson Comorbidity Index (5.1 [ sd , 2.2] vs. 3.8 [ sd , 2.6]; p < 0.01), and similar Sequential Organ Failure Assessment score (9.6 [ sd , 3.3] vs. 9.5 [ sd , 3.7]; p = 0.93), compared with non-TLT patients. TLT patients were more likely to die or be discharged to hospice (80.6% vs. 42.1%; p < 0.05) and had shorter ICU length of stay (median, 5.7 d [IQR, 4.0–9.0 d] vs. 10.3 d [IQR, 5.5–14.5 d]; p < 0.01).
CONCLUSIONS AND RELEVANCE
In this study, approximately one in five patients with ARF participated in a TLT. Our findings suggest TLTs are used primarily in patients near end of life but with substantial physician variation, highlighting a need for evidence to guide optimal use.
... Unfortunately, clinicians are frequently unable to accurately predict important outcomes for patients with critical illness (such predictions may be even worse for patients who are ambulatory and outpatient). 41 Although we frequently have information about the average level of risk associated with a particular procedure across many patients, we forget that each patient is unique and each clinical context is different. So even if the risk associated with a procedure is low, if a patient experiences an adverse outcome, their realized risk is 100%. ...
Many seriously ill patients undergo surgical interventions. Palliative care clinicians may not be familiar with the nuances involved in perioperative care, however they can play a valuable role in enabling the delivery of patient-centered and goal-concordant perioperative care. The interval of time surrounding a surgical intervention is fraught with medical, psychosocial, and relational risks, many of which palliative care clinicians may be well-positioned to navigate. A perioperative palliative care consult may involve exploring gaps between clinician and patient expectations, facilitating continuity of symptom management or helping patients to designate a surrogate decision-maker before undergoing anesthesia. Palliative care clinicians may also be called upon to direct discussions around perioperative management of modified code status orders and to engage around the goal-concordance of proposed interventions. This article, written by a team of surgeons and anesthesiologists, many with subspecialty training in palliative medicine and/or ethics, offers ten tips to support palliative care clinicians and facilitate comprehensive discussion as they engage with patients and clinicians considering surgical interventions.
... With diverse definitions of QoL, difficulties for clinicians to objectively assess patients' QoL are common and add a need for comprehensive questionnaires as measurement proxies for QoL. Improvements in survival are increasingly challenging to demonstrate in trials, and intensive care is no exception to these needs and problems; the generic HRQoL questionnaires SF-36 and EQ-5D are commonly used but are both developed for a general population [8]. Thus, their dismal results in measuring ICU survivors' views on important issues are not surprising [9,10]. ...
Background
Quality of life (QoL) is a key outcome measure in healthcare. However, the heterogeneity in its definitions presents challenges in the objective evaluation of improvement. Universal questionnaires, tailored for a broad demographic group, inadequately represent the unique experiences of intensive care unit (ICU) survivors, including a lack of ability to discriminate issues related to QoL from issues that do not.
Methods
Using a 218-item, 13-domain provisional questionnaire, we assessed 395 adult ICU survivors, with a minimum 72-h stay at one of three Swedish university hospital ICUs, at 6 months to three years post-discharge. Their responses were compared to those of 195 controls, matched for age and sex and randomly recruited from the Swedish Population Registry. By multi-group exploratory factor analysis, we compared dimensionality in QoL perceptions between the two groups, emphasising patterns of correlation to 13 domain-specific QoL questions. Model fit was assessed using information criteria. Internal consistency reliability for each scale was determined using McDonald’s omega or Cronbach’s alpha. All analyses were conducted using Mplus, applying full information maximum likelihood to handle missing data.
Results
All domains except Cognition had a subset of questions correlating to the domain-specific QoL question in at least the ICU survivor group. The similarity between the two groups varied, with Physical health, Sexual health and Gastrointestinal (GI) functions mainly correlating the same issues to QoL in the two groups. In contrast, Fatigue, Pain, Mental health, activities of daily living, Sleep, Sensory functions and Work life showed considerable differences. In all, about one-fourth of the issues correlated to QoL in the ICU survivor group and about one-tenth of the issues in the control group.
Conclusions
We found most issues experienced by ICU survivors to be unrelated to quality of life. Our findings indicate that the consequences of post-ICU issues may play a more significant role in affecting QoL than the issues themselves; issues restricting and affecting social life and work life were more related to QoL in ICU survivors than in non-ICU-treated controls. Caution is advised before associating all post-ICU problems with an effect on quality of life.
Trial registration : ClinicalTrials.gov Ref# NCT02767180; Registered 28 April 2016.
... Shared decision-making has been recommended to overcome these challenges (Davidson et al., 2017;Kon et al., 2016), which requires family engagement in decision-making and consultation regarding the patient's goals and preferences. Shared decision-making is recommended for treatment decisions involving uncertainty, high-risk, or potentially unacceptable patient outcomes (Detsky et al., 2017;Kruser et al., 2019). ...
Aim
This study aims to define and investigate characteristics, antecedents, and consequences of the concept of family engagement in caring for patients with infectious diseases hospitalised in intensive care units.
Design
This is a three‐phase hybrid model study (theoretical, fieldwork, and analytical phase).
Methods
The York University Guidelines were used in the theoretical phase, and ultimately, 16 pieces of literature related to the subject under study from 2011 to 2021 were reviewed. The content analysis was used for fieldwork phases; eight participants were interviewed. Then, the theoretical and fieldwork findings were compared, integrated, and analysed.
Results
This concept has characteristics such as; awareness, belief, perception, and willingness of the nurse to engage the family; a sense of responsibility, willingness, and sacrifice of the family; the physical or virtual presence of the family; triangular interaction between the nurse, patient, and family; perception and identifying the goals; education and information transfer; team collaboration; delegation of responsibility to the family; decision making; and protection of the family. Antecedents include the availability of infrastructure; patient, family, and nurse conditions; and the quality implementation of engagement. The consequences include positive consequences related to the patient, family, nursing, and society, as well as some negative consequences. This study provided a comprehensive perception of family engagement in the care of patients with infectious diseases in intensive care units and defined it more clearly, showing its characteristics, antecedents, and consequences.
Patient or Public Contribution
Eight participants were interviewed, including five nurses, two family caregivers, and one patient.
... 3 6 They are also not designed to predict life-limiting clinical conditions (LLCCs) and may not achieve precise clinical decisions in individualised patients. 7 Emergency physicians need a practical method to help identify patients who are seriously ill and at the highest risk of mortality, and to ensure that such patients and their families have the opportunity to align their expectations of care with the severity of their conditions after being admitted to the hospital. 8 ...
Objectives
This study aims to test the ability of the surprise question (SQ), when asked to emergency physicians (EPs), to predict in-hospital mortality among adults admitted to an emergency room (ER).
Methods
This prospective cohort study at an academic medical centre included consecutive patients 18 years or older who received care in the ER and were subsequently admitted to the hospital from 20 April 2018 to 20 October 2018. EPs were required to answer the SQ for all patients who were being admitted to hospital. The primary outcome was in-hospital mortality.
Results
The cohort included 725 adults (mean (SD) age, 60 (17) years, 51% men) from 58 128 emergency department (ED) visits. The mortality rates were 20.6% for 30-day all-cause in-hospital mortality and 23.6% for in-hospital mortality. The diagnostic test characteristics of the SQ have a sensitivity of 53.7% and specificity of 87.1%, and a relative risk of 4.02 (95% CI 3.15 to 5.13), p<0.01). The positive and negative predictive values were 57% and 86%, respectively; the positive likelihood ratio was 4.1 and negative likelihood ratio was 0.53; and the accuracy was 79.2%.
Conclusions
We found that asking the SQ to EPs may be a useful tool to identify patients in the ED with a high risk of in-hospital mortality.
... It might be ascribed that surrogate decision-making is a complex task in the ICU environment, and the prognosis judgment of ICU doctors may be wrong. Thus, the surrogate decision-making interventions may cause patients to lose life support prematurely; otherwise, this part of patients may survive for a longer time [57]. In addition, personal characteristics related to SDM, such as coping strategies and competitive responsibilities, may also affect the effect of this intervention. ...
Background
Numerous studies have explored care interventions to improve the psychological outcome of intensive care unit (ICU) patients, but inconclusive evidence makes it difficult for decision-makers, managers, and clinicians to get familiar with all available literature and find appropriate interventions. This umbrella review aimed to analyze the relationship between care intervention and psychological outcomes of ICU patients based on existing systematic reviews.
Methods
An umbrella review of evidence across systematic reviews and meta-analyses published between 1987 and 2023 was undertaken. We systematically searched reviews that examined the association between care intervention and the improvement of adverse psychological outcomes in ICU patients using PubMed, EMBASE, Web of Science, Cochrane Library, and manual reference screening. The measurement tool (AMSTAR 2) was applied to evaluate the methodological quality of included studies. The excess significance bias, between-study heterogeneity expressed by I², small-study effect, and evidence class were estimated.
Results
A total of 5110 articles were initially identified from the search databases and nine of them were included in the analysis. By applying standardized criteria, only weak evidence was observed in 13 associations, even though most included reviews were of moderate to high methodological quality. These associations pertained to eight interventions (music therapy, early rehabilitation, post-ICU follow-up, ICU diary, information intervention, preoperative education, communication and psychological support, surrogate decision-making) and five psychological outcomes (post-intensive care syndrome, transfer anxiety, post-traumatic stress disorder, anxiety, and depression). Weak or null association was shown among the rest of the associations (e.g., weak association between music therapy and maternal anxiety or stress level).
Conclusions
The evidence of these eight supporting interventions to improve the adverse psychological outcomes of ICU patients and caregivers was weak. Data from more and better-designed studies with larger sample sizes are needed to establish robust evidence.
... When compared to different scoring systems, physicians' predictions of mortality have been found to be more accurate 1,3-5 ; however, both have only a moderate ability to discriminate between survivors and nonsurvivors. ICU nurses have also been shown to be able to accurately predict patient outcome and mortality, 6 and to be more accurate than scoring systems. 5 Prognostic assessments of veterinary patients influence management decisions and may be relied upon by owners for decision-making on concepts such as euthanasia. ...
Objective
To determine whether emergency staff and students can predict patient outcome within 24 hours of admission, comparing the accuracy of clinician prognostication with outcome prediction by Acute Patient Physiologic and Laboratory Evaluation (APPLE)fast scoring and identifying whether experience or mood would be associated with accuracy.
Design
Prospective observational study between April 2020 and March 2021.
Setting
University teaching hospital.
Animals
One hundred and sixty‐one dogs admitted through an Emergency Service were assessed. Where data were available, an APPLEfast score was calculated per patient. An APPLEfast score of >25 was deemed a predictor for mortality.
Interventions
None.
Measurements and Main Results
Emergency staff and students were asked to complete surveys about dogs admitted to the emergency room. All clinicopathological data were available for review, and the animals were available for examination. Data collected included opinions on whether the patient would be discharged from hospital, a mood score, position, and experience in Emergency and Critical Care. One‐hundred and twenty‐five dogs (77.6%) were discharged; 36 dogs (22.4%) died or were euthanized. Two hundred and sixty‐six responses were obtained; 202 responses (75.9%) predicted the correct outcome. Students, interns, residents, faculty, and nurses predicted the correct outcome in 81.4%, 58.3%, 83.3%, 82.1%, and 65.5% of cases, respectively. Of 64 incorrect predictions, 43 (67.2%) predicted death in hospital. APPLEfast scores were obtained in 121 cases, predicting the correct outcome in 83 cases (68.6%). Of 38 cases in which APPLEfast was incorrect, 27 (71.1%) were dogs surviving to discharge. Mean APPLEfast score was 22.9 (± 6.2). There was no difference in outcome prediction accuracy between staff and APPLEfast scores (P = 0.13). Neither experience nor mood score was associated with outcome prediction ability (P = 0.55 and P = 0.74, respectively).
Conclusions
Outcome prediction accuracy by staff is not significantly different to APPLEfast scoring where a cutoff of >25 is used to predict mortality. When predictions were incorrect, they often predicted nonsurvival.
... 3 6 They are also not designed to predict life-limiting clinical conditions (LLCCs) and may not achieve precise clinical decisions in individualised patients. 7 Emergency physicians need a practical method to help identify patients who are seriously ill and at the highest risk of mortality, and to ensure that such patients and their families have the opportunity to align their expectations of care with the severity of their conditions after being admitted to the hospital. 8 ...
... This becomes especially problematic if the initial prediction was incorrect (a false-positive), which could result in the patient not receiving the potentially beneficial care. While these issues have existed even before AI and ML are developed (because predictions of clinicians are sometimes inaccurate), 91 there is growing concern that AI and ML might amplify the bias due to self-fulfilling feedback loops (Fig. 6). If a model trained on biased data is applied to guide clinical decision-making, and the new data influenced by the model's results are then used as input data again to "improve" the model, there is a risk that the initial biases will be reinforced and amplified. ...
Aim:
Artificial intelligence (AI) and machine learning (ML) are important areas of computer science that have recently attracted attention for their application to medicine. However, as techniques continue to advance and become more complex, it is increasingly challenging for clinicians to stay abreast of the latest research. This overview aims to translate research concepts and potential concerns to healthcare professionals interested in applying AI and ML to resuscitation research but who are not experts in the field.
Main text:
We present various research including prediction models using structured and unstructured data, exploring treatment heterogeneity, reinforcement learning, language processing, and large-scale language models. These studies potentially offer valuable insights for optimizing treatment strategies and clinical workflows. However, implementing AI and ML in clinical settings presents its own set of challenges. The availability of high-quality and reliable data is crucial for developing accurate ML models. A rigorous validation process and the integration of ML into clinical practice is essential for practical implementation. We furthermore highlight the potential risks associated with self-fulfilling prophecies and feedback loops, emphasizing the importance of transparency, interpretability, and trustworthiness in AI and ML models. These issues need to be addressed in order to establish reliable and trustworthy AI and ML models.
Conclusion:
In this article, we overview concepts and examples of AI and ML research in the resuscitation field. Moving forward, appropriate understanding of ML and collaboration with relevant experts will be essential for researchers and clinicians to overcome the challenges and harness the full potential of AI and ML in resuscitation.
... This entailed providers attempting to determine who was most likely to benefit from the resources of an ICU and from whom these resources should be withheld or withdrawn. This crisis-fueled approach proved problematic because, in addition to the newness of the disease with its many unknowns, prognostic accuracy of providers in predicting outcomes has been to shown to be only slightly better than chance [18]. This prognostic uncertainty is also a significant driver of the wide variability in admitting practices for patients with mTBI. ...
... It might be ascribed to that surrogate decision-making is a complex task in the ICU environment, and the prognosis judgment of ICU doctors may be wrong. Thus, the surrogate decision-making interventions may cause patients to lose life support prematurely, otherwise this part of patients may survive for a longer time [43]. In addition, personal characteristics related to SDM, such as coping strategies and competitive responsibilities, may also affect the effect of this intervention. ...
Background
Numerous studies have explored care interventions to improve the psychological outcome of intensive care unit (ICU) patients, but inconclusive evidence makes it difficult for decision-makers, managers and clinicians to get familiar with all available literatures and find appropriate interventions. This umbrella review aimed to analyze the relationship between care intervention and psychological outcomes of ICU patients bases on existing systematic reviews.
Methods
Umbrella review of evidence across systematic reviews and meta-analyses published between 1987 and 2020 was undertaken. We systematically searched primary studies that examined the association between care intervention and the improvement of adverse psychological outcomes in ICU patients using PubMed, EMBASE, web of science, Cochrane library, and manual reference screening. The measurement tool (AMSTAR 2) was applied to evaluate the methodological quality of included studies. The excess significance bias, between-study heterogeneity expressed by I2, small study effect and evidence class were estimated.
Results
A total of 5110 articles were initially identified from the search databases and nine of them were included in analysis. By applying standardized criteria, only week evidence was observed in 13 associations, even though most included reviews were of moderate to high methodological quality. These associations pertained to eight interventions (music therapy, early rehabilitation, post ICU follow up, ICU diary, information intervention, preoperative education, communication and psychological support, surrogate decision-making) and five psychological outcomes (post intensive care syndrome, transfer anxiety, post-traumatic stress disorder, anxiety and depression). Weak or null association was shown among the rest of the associations (e.g., weak association between music therapy and maternal anxiety or stress level).
Conclusions
The evidence of these eight supporting interventions to improve the adverse psychological outcomes of ICU patients and caregivers was weak. Data from more and better-designed studies with lager sample size are needed to establish robust evidence.
... For the former, there is abundant evidence that physicians can generally beat simple benchmarks (like uninformative forecasts) in offering nearer-term prognostication for patients, though they tend towards overestimating the probability of good outcomes (Brandt et al. 2006;Cheon et al. 2016;Glare et al. 2003;Hoesseini et al. 2020;Rojas et al. 2020). In many realms, physicians can discriminate between patients likely to have certain good outcomes and those likely to have bad outcomes (Detsky et al. 2017). ...
Expectations about future events underlie practically every decision we make, including those in medical research. This paper reviews five studies undertaken to assess how well medical experts could predict the outcomes of clinical trials. It explains why expert trial forecasting was the focus of study and argues that forecasting skill affords insights into the quality of expert judgment and might be harnessed to improve decision-making in care, policy, and research. The paper also addresses potential criticisms of the research agenda and summarizes key findings from the five studies of trial forecasting. Together, the studies suggest that trials frequently deliver surprising results to expert communities and that individual experts are often uninformative when it comes to forecasting trial outcome and recruitment. However, the findings also suggest that expert forecasts often contain a "signal" about whether a trial will be positive, especially when forecasts are aggregated. The paper concludes with needs for further research and tentative policy recommendations.
... Previous studies have shown that clinician prediction performance for outcomes in hospitalized patients may vary according to clinical experience of the physician who complete the survey [28,39]. In our multivariate model, the physician prediction was significantly associated with RRT requirement, independently of the clinical experience of the physician. ...
Purpose:
Identifying patients who will receive renal replacement therapy (RRT) during intensive care unit (ICU) stay is a major challenge for intensivists. The objective of this study was to evaluate the performance of physicians in predicting the need for RRT at ICU admission and at acute kidney injury (AKI) diagnosis.
Methods:
Prospective, multicenter study including all adult patients hospitalized in 16 ICUs in October 2020. Physician prediction was estimated at ICU admission and at AKI diagnosis, according to a visual Likert scale. Discrimination, risk stratification and benefit of physician estimation were assessed. Mixed logistic regression models of variables associated with risk of receiving RRT, with and without physician estimation, were compared.
Results:
Six hundred and forty-nine patients were included, 270 (41.6%) developed AKI and 77 (11.8%) received RRT. At ICU admission and at AKI diagnosis, a model including physician prediction, the experience of the physician, SOFA score, serum creatinine and diuresis to determine need for RRT performed better than a model without physician estimation with an area under the ROC curve of 0.90 [95% CI 0.86-0.94, p < 0.008 (at ICU admission)] and 0.89 [95% CI 0.83-0.93, p = 0.0014 (at AKI diagnosis)]. In multivariate analysis, physician prediction was strongly associated with the need for RRT, independently of creatinine levels, diuresis, SOFA score and the experience of the doctor who made the prediction.
Conclusion:
As physicians are able to stratify patients at high risk of RRT, physician judgement should be taken into account when designing new randomized studies focusing on RRT initiation during AKI.
... Allowing the anticipated clinical course to impact our sense of hope becomes even more fraught when considering that our ability to independently predict patient outcomes is limited. Physician prediction has failed to reliably predict extubation readiness (5, 6), ICU readmission (7), and long-term functional outcomes in critically ill patients (8,9). When predicting in-hospital mortality, experienced physicians do better than random chance but still get it wrong much of the time (9). ...
... Other known categories of prognostic indicators that can be rapidly assessed at the bedside include number of organ failures (or SOFA score), functional ability or co-morbidity by a simple score such as a modified ASA score, etc [58][59][60]. Intuitive survival prognostication by clinicians is also important and has been shown to be at least equivalent to individual prognostic scores [61,62]. ...
ICU resources are not unlimited, and in many circumstances frontline clinicians may be called upon to prioritize among patients who are referred for ICU care. When resources, manifested as ICU bed availability, are insufficient to offer ICU admission for all, some patients must be refused. There are concerns regarding this process of triage in the very elderly. On the one hand, some are concerned that the very elderly will be preferentially refused as a result of age discrimination, whereas others are concerned that the very elderly may receive inappropriately aggressive ICU care resulting in unnecessary suffering and pain, as well as consuming resources better redirected to the young. This chapter will categorize the underlying reasons why very elderly patients may be refused ICU admission but focus on the process of triage in resource-restricted settings. A modified utilitarian approach will be used for justification of triage across all age groups, including the very elderly. Suggested methods of prioritization that avoid the need to directly and substantially consider chronological age will be described. A decision-making framework, coupled with a prioritization tool, is presented to inform the process of frontline triage.
... 8,9 The ICU admission decisions making by physicians had significantly different ICU admission rates, which are affected by the type and seniority of physicians. [10][11][12] The resource availability is associated with better survival and increasing resource availability may improve patients' outcomes. 13 Therefore, the development of an accurate predictive model including objective clinical variables in the preoperative assessment is required to guide the allocation of resources such as ICU beds. ...
Background
To develop a highly discriminative machine learning model for the prediction of intensive care unit admission (>24h) using the easily available preoperative information from electronic health records. An accurate prediction model for ICU admission after surgery is of great importance for surgical risk assessment and appropriate utilization of ICU resources.
Method
Data were collected retrospectively from a large hospital, comprising 135,442 adult patients who underwent surgery except for cardiac surgery between 1 January 2014, and 31 July 2018 in China. Multiple existing predictive machine learning algorithms were explored to construct the prediction model, including logistic regression, random forest, adaptive boosting, and gradient boosting machine. Four secondary analyses were conducted to improve the interpretability of the results.
Results
A total of 2702 (2.0%) patients were admitted to the intensive care unit postoperatively. The gradient boosting machine model attained the highest area under the receiver operating characteristic curve of 0.90. The machine learning models predicted intensive care unit admission better than the American Society of Anesthesiologists Physical Status (area under the receiver operating characteristic curve: 0.68). The gradient boosting machine recognized several features as highly significant predictors for postoperatively intensive care unit admission. By applying subgroup analysis and secondary analysis, we found that patients with operations on the digestive, respiratory, and vascular systems had higher probabilities for intensive care unit admission.
Conclusion
Compared with conventional American Society of Anesthesiologists Physical Status and logistic regression model, the gradient boosting machine could improve the performance in the prediction of intensive care unit admission. Machine learning models could be used to improve the discrimination and identify the need for intensive care unit admission after surgery in elective noncardiac surgical patients, which could help manage the surgical risk.
... (4,5) They are also not designed to predict life support limitations (LSLs) and may not achieve precise clinical decisions in individualized patients. (6) There are several physical functional performance (PFP) scales that are useful to predict outcomes in clinical contexts other than the ICU. (7) The Karnofsky Performance Status (KPS) scale classifies patients according to the degree of their functional disabilities and loss of autonomy. ...
Objective
To assess whether scales of physical functional performance and the surprise question (“Would I be surprised if this patient died in 6 months?”) predict life support limitations and mortality in critically ill nonsurgical patients.
Methods
We included 114 patients admitted from the Emergency Department to an intensive care unit in this prospective cohort. Physical functional performance was assessed by the Palliative Prognostic Score, Karnofsky Performance Status, and the Katz Activities of Daily Living scale. Two intensivists responded to the surprise question.
Results
The proposed physical functional performance scores were significantly lower in patients with life support limitations and those who died during the hospital stay. A negative response to the surprise question was more frequent in the same subset of patients. Adjusted univariable analysis showed an increased odds ratio for life support limitations and death regarding the activities of daily living scale (1.35 [1.01 - 1.78] and 1.34 [1.0 - 1.79], respectively) and a negative response for the surprise question (42.35 [11.62 - 154.43] and 47.79 [11.41 - 200.25], respectively); with a p < 0.05 for all results.
Conclusion
All physical functional performance scales showed lower scores in nonsurvivors and patients with life support limitations. The activities of daily living score and the surprise question increased the odds of life support limitations and mortality in our cohort of nonsurgical intensive care unit patients admitted from the Emergency Department.
Keywords:
Physical functional performance; Karnofsky Performance Status; Activities of daily living; Palliative care; Intensive care units
... (4,5) They are also not designed to predict life support limitations (LSLs) and may not achieve precise clinical decisions in individualized patients. (6) There are several physical functional performance (PFP) scales that are useful to predict outcomes in clinical contexts other than the ICU. (7) The Karnofsky Performance Status (KPS) scale classifies patients according to the degree of their functional disabilities and loss of autonomy. ...
Objective:
To assess whether scales of physical functional performance and the surprise question ("Would I be surprised if this patient died in 6 months?") predict life support limitations and mortality in critically ill nonsurgical patients.
Methods:
We included 114 patients admitted from the Emergency Department to an intensive care unit in this prospective cohort. Physical functional performance was assessed by the Palliative Prognostic Score, Karnofsky Performance Status, and the Katz Activities of Daily Living scale. Two intensivists responded to the surprise question.
Results:
The proposed physical functional performance scores were significantly lower in patients with life support limitations and those who died during the hospital stay. A negative response to the surprise question was more frequent in the same subset of patients. Adjusted univariable analysis showed an increased odds ratio for life support limitations and death regarding the activities of daily living scale (1.35 [1.01 - 1.78] and 1.34 [1.0 - 1.79], respectively) and a negative response for the surprise question (42.35 [11.62 - 154.43] and 47.79 [11.41 - 200.25], respectively); with a p < 0.05 for all results.
Conclusion:
All physical functional performance scales showed lower scores in nonsurvivors and patients with life support limitations. The activities of daily living score and the surprise question increased the odds of life support limitations and mortality in our cohort of nonsurgical intensive care unit patients admitted from the Emergency Department.
... There are scant data on the ability of intensivists to predict long term quality of life. It has been found that, while intensivists can reasonably accurately predict mortality at 6 months for patients already in the ICU, they are less accurate at predicting functional outcome (which may correlate with quality of life) in survivors, 6 and the authors of the ANZICS guidance on pandemic tirage do note that estimating quality of life following intensive care is challenging. ...
The initial weeks and months of the coronavirus disease 2019 (COVID-19) pandemic in early 2020 were a time of incredible uncertainty for the intensive care community. Reports were emerging, particularly from northern Italy, of intensive care units (ICUs) being rapidly overwhelmed by large numbers of patients with respiratory failure from COVID-19, and of intensivists having to make seemingly impossible decisions about who to admit to the ICU and who to leave on theward. 1 These decisions were being made with the reasonable expectation that those left on the ward would die. There were widespread reports of age being used as a determinant of ICU admission, which generated some controversy both in the medical literature and in the lay press.
Background
Clinical uncertainty is associated with increased resource utilization, worsened health-related quality of life for patients, and provider burnout, particularly during critical illness. Existing data are limited, because determining uncertainty from notes typically requires manual, qualitative review. We sought to develop a consensus list of descriptors of clinical uncertainty and then, using a thematic analysis approach, describe how respondents consider their use in intensive care unit (ICU) notes, such that future work can extract uncertainty data at scale.
Design
We conducted a Delphi consensus study with physicians across multiple institutions nationally who care for critically ill patients or patients with advanced illnesses. Participants were given a definition for clinical uncertainty and collaborated through multiple rounds to determine which words represent uncertainty in clinician notes. We also administered surveys that included open-ended questions to participants about clinical uncertainty. Following derivation of a consensus list, we analyzed participant responses using thematic analysis to understand the role of uncertainty in clinical documentation.
Results
Nineteen physicians participated in at least 2 of the Delphi rounds. Consensus was achieved for 44 words or phrases over 5 rounds of the Delphi process. Clinicians described comfort with using uncertainty terms and used them in a variety of ways: documenting and processing the diagnostic thinking process, enlisting help, identifying incomplete information, and practicing transparency to reflect uncertainty that was present.
Conclusions
Using a consensus process, we created an uncertainty lexicon that can be used for uncertainty data extraction from the medical record. We demonstrate that physicians, particularly in the ICU, are comfortable with uncertainty and document uncertainty terms frequently to convey the complexity and ambiguity that is pervasive in critical illness.
Highlights
Question: What words do physicians caring for critically ill patients use to document clinical uncertainty, and why? Findings: A consensus list of 44 words or phrases was identified by a group of experts. Physicians expressed comfort with using these words in the electronic health record. Meaning: Physicians are comfortable with uncertainty words and document them frequently to convey the complexity and ambiguity that is pervasive in critical illness.
Rationale:
Despite functional impairments, ICU survivors can perceive their quality of life as acceptable.
Objectives:
To investigate discrepancies between calculated health, based on self-reported physical, mental and cognitive functioning, and perceived health one year after ICU admission.
Methods:
Data from an ongoing prospective multicenter cohort study, MONITOR-IC, were used. Patient-reported physical, mental and cognitive functioning, and perceived health (EQ-VAS, range 0-100) one year post-ICU of patients admitted to one of eleven participating ICUs between July 2016 and September 2021 were analyzed. The relationship between functional outcomes and perceived health was modeled using linear regression. Calculated health for each patient was estimated using this model and compared to patients' perceived health, the difference reflecting a discrepancy. Based on a minimal clinically important difference of eight points, three groups were defined: patients who rated their health better than calculated (positive discrepancy), patients who rated their health worse than calculated (negative discrepancy), and patients whose perceived health was concordant with their calculated health.
Results:
2,545 patients were analyzed, of whom 45.0% (n = 1,146) showed a discrepancy between calculated and perceived health. Patients with a negative discrepancy rated their health significantly lower (median 50, IQR 36 - 66) than patients with a positive discrepancy (median 84, IQR 75 - 90). Importantly, there were no significant differences in physical, mental and cognitive functioning between patients with a negative and positive discrepancy. Patients with a negative discrepancy had a higher education level and were more often unemployed.
Conclusions:
One year post-ICU, almost half of ICU survivors showed a discrepancy between calculated health and perceived health.
Background
More than 350,000 U.S. infants are admitted to the neonatal intensive care unit (NICU) annually and likely experience discomfort. Although nurse perceptions of infant symptoms, suffering, and quality of life (QOL) are valuable, the availability of standardized assessment tools to measure these concepts are limited.
Purpose
To provide preliminary evidence of the internal structure, reliability, and validity of the Nurse Perception of Infant Condition (NPIC) scale.
Methods
Infants were enrolled from a Level IV NICU in the U.S. Midwest. Nurses reported on their perceptions of the infant symptom experience and their expectations for infant survival. Weekly behavioral observations of infants were obtained before and after standard delivery of care to obtain a comfort score.
Results
237 nurses who cared for 73 infants completed 569 surveys over 28 months. All NPIC items were significantly correlated with each other ( P < .001). Factor analysis revealed strong evidence of a 2-factor structure (survival and suffering subscales). Both subscales demonstrated good to excellent internal consistency. Together the 2 factors explained 82% of the variability in the scale responses. Limited validity evidence was found.
Implications for Practice and Research
Evidence was found to support the internal structure and reliability of the NPIC scale. However, further item development and refinement is needed to increase the utility NPIC scale in clinical and research settings. The development of improved assessments of the infant NICU experience is warranted. Nurse perceptions of infant suffering or poor QOL may have implications for their expectations for infant survival and possibly care delivery.
Background
End-of-life (EOL) care is associated with high resource utilization. Recognizing and effectively communicating that EOL is near promotes more patient-centered care, while decreasing futile interventions. We hypothesize that provider assessment of futility during the surgical intensive care unit (SICU) admission would result in higher rates of Do Not Resuscitate (DNR).
Methods
We performed a retrospective review of a prospective SICU registry of all deceased patients across a health system, 2018–2022. The registry included a subjective provider assessment of patient’s expected survival. We employed multivariable logistic regression to adjust for clinical factors while assessing for association between code status at death and provider’s survival assessment with attention to race-based differences.
Results
746 patients—105 (14.1%) traumatically injured and 641 (85.9%) non-traumatically injured—died over 4.5 years in the SICU (mortality rate 5.9%). 26.3% of these deaths were expected by the ICU provider. 40.9% of trauma patients were full code at the time of death, compared with 15.6% of non-traumatically injured patients. Expected death was associated with increased odds of DNR code status for non-traumatically injured patients (OR 1.8, 95% CI 1.03 to 3.18), but not for traumatically injured patients (OR 0.82, 95% CI 0.22 to 3.08). After adjusting for demographic and clinical characteristics, black patients were less likely to be DNR at the time of death (OR 0.49, 95% CI 0.32 to 0.75).
Conclusion
20% of patients who died in our SICU had not declared a DNR status, with injured black patients more likely to remain full code at the time of death. Further evaluation of this cohort to optimize recognition and communication of EOL is needed to avoid unnecessary suffering.
Level of evidence
Level III/prognostic and epidemiological.
The prognostication of long-term functional outcomes remains challenging in patients with traumatic brain injury (TBI). Our aim was to demonstrate that intensive care unit (ICU) variables are not efficient to predict 6-month functional outcome in survivors with moderate to severe TBI (msTBI) but are mostly associated with mortality, which leads to a mortality bias for models predicting a composite outcome of mortality and severe disability.
We analyzed the data from the multicenter randomized controlled Continuous Hyperosmolar Therapy in Traumatic Brain-Injured Patients trial and developed predictive models using machine learning methods and baseline characteristics and predictors collected during ICU stay. We compared our models’ predictions of 6-month binary Glasgow Outcome Scale extended (GOS-E) score in all patients with msTBI (unfavorable GOS-E 1–4 vs. favorable GOS-E 5–8) with mortality (GOS-E 1 vs. GOS-E 2–8) and binary functional outcome in survivors with msTBI (severe disability GOS-E 2–4 vs. moderate to no disability GOS-E 5–8). We investigated the link between ICU variables and long-term functional outcomes in survivors with msTBI using predictive modeling and factor analysis of mixed data and validated our hypotheses on the International Mission for Prognosis and Analysis of Clinical Trials in TBI (IMPACT) model.
Based on data from 370 patients with msTBI and classically used ICU variables, the prediction of the 6-month outcome in survivors was inefficient (mean area under the receiver operating characteristic 0.52). Using factor analysis of mixed data graph, we demonstrated that high-variance ICU variables were not associated with outcome in survivors with msTBI (p = 0.15 for dimension 1, p = 0.53 for dimension 2) but mostly with mortality (p < 0.001 for dimension 1), leading to a mortality bias for models predicting a composite outcome of mortality and severe disability. We finally identified this mortality bias in the IMPACT model.
We demonstrated using machine learning–based predictive models that classically used ICU variables are strongly associated with mortality but not with 6-month outcome in survivors with msTBI, leading to a mortality bias when predicting a composite outcome of mortality and severe disability.
Prognosis determines major decisions regarding treatment for critically ill patients. Statistical models have been developed to predict the probability of survival and other outcomes of intensive care. Although they were trained on the characteristics of large patient cohorts, they often do not represent very old patients (age ≥ 80 years) appropriately. Moreover, the heterogeneity within this particular group impairs the utility of statistical predictions for informing decision-making in very old individuals. In addition to these methodological problems, the diversity of cultural attitudes, available resources as well as variations of legal and professional norms limit the generalisability of prediction models, especially in patients with complex multi-morbidity and pre-existing functional impairments. Thus, current approaches to prognosticating outcomes in very old patients are imperfect and can generate substantial uncertainty about optimal trajectories of critical care in the individual. This article presents the state of the art and new approaches to predicting outcomes of intensive care for these patients. Special emphasis has been given to the integration of predictions into the decision-making for individual patients. This requires quantification of prognostic uncertainty and a careful alignment of decisions with the preferences of patients, who might prioritise functional outcomes over survival. Since the performance of outcome predictions for the individual patient may improve over time, time-limited trials in intensive care may be an appropriate way to increase the confidence in decisions about life-sustaining treatment.
Background
In South Africa (SA), intensive care is faced with the challenge of resource scarcity as well as an increasing demand for intensive care unit (ICU) services. ICU services are expensive, and practitioners in low- to middle-income countries experience daily the consequences of limited resources. Critically limited resources necessitate that rationing and triage (prioritisation) decisions are frequently necessary in SA, particularly in the publicly funded health sector.
Purpose
The purpose of this consensus statement is to examine key questions that arise when considering the status of ICU resources in SA, and more specifically ICU admission, rationing and triage decisions. The accompanying guideline in this issue is intended to guide frontline triage policy and ensure the best utilisation of intensive care in SA, while maintaining a fair distribution of available resources. Fair and efficient triage is important to ensure the ongoing provision of high-quality care to adult patients referred for intensive care.
Recommendations
In response to 14 key questions developed using a modified Delphi technique, 29 recommendations were formulated and graded using an adapted GRADE score. The 14 key questions addressed the status of the provision of ICU services in SA, the degree of resource restriction, the efficiency of resource management, the need for triage, and how triage could be most justly implemented. Important recommendations included the need to formally recognise and accurately quantify the provision of ICU services in SA by national audit; actively seek additional resources from governmental bodies; consider methods to maximise the efficiency of ICU care; evaluate lower level of care alternatives; develop a triage guideline to assist policy-makers and frontline practitioners to implement triage decisions in an efficient and fair way; measure and audit the consequence of triage; and promote research to improve the accuracy and consistency of triage decisions. The consensus document and guideline should be reviewed and revised appropriately within 5 years.
Conclusion
In recognition of the absolute need to limit patient access to ICU because of the lack of sufficient intensive care resources in public hospitals, recommendations and a guideline have been developed to guide policy-making and assist frontline triage decision-making in SA. These documents are not a complete plan for quality practice but rather the beginning of a long-term initiative to engage clinicians, the public and administrators in appropriate triage decision-making, and promote systems that will ultimately maximise the efficient and fair use of available ICU resources.
Since treatment in an intensive care unit (ICU) can be burdensome and the outcome may be disappointing, the benefit and harm of an ICU treatment should be assessed repeatedly to adhere to the patient’s wishes. This should be done within a process of careful and shared decision-making, in which clinicians and patients or their surrogate decision-makers make rational, evidence-based, and individualized healthcare decisions. In this way, prognosis, proportionality, and patient preferences can be aligned in order to provide proportional and goal-concordant care. In this chapter, we well expound the process of making individualized decisions concerning life-sustaining treatments in the ICU. We will describe the prognosis of ICU patients concerning mortality, morbidity, and quality of life. We will explicit the concept of goal-concordant care, explain how advance care planning can or can’t help in decision-making and what the role of substituted judgment is in life-sustaining treatment decisions in incapacitated ICU patients.KeywordsIntensive care medicineEthicsCritical careLife-sustaining therapiesPrognosticationClinical ethicsAdvance care planningShared decision-makingGoal-concordant therapySurrogate decision-makersSubstituted judgment
Importance:
The ability to provide invasive mechanical ventilation (IMV) is a mainstay of modern intensive care; however, whether rates of IMV vary among countries is unclear.
Objective:
To estimate the per capita rates of IMV in adults across 3 high-income countries with large variation in per capita intensive care unit (ICU) bed availability.
Design, setting, and participants:
This cohort study examined 2018 data of patients aged 20 years or older who received IMV in England, Canada, and the US.
Exposure:
The country in which IMV was received.
Main outcomes and measures:
The main outcome was the age-standardized rate of IMV and ICU admissions in each country. Rates were stratified by age, specific diagnoses (acute myocardial infarction, pulmonary embolus, upper gastrointestinal bleed), and comorbidities (dementia, dialysis dependence). Data analyses were conducted between January 1, 2021, and December 1, 2022.
Results:
The study included 59 873 hospital admissions with IMV in England (median [IQR] patient age, 61 [47-72] years; 59% men, 41% women), 70 250 in Canada (median [IQR] patient age, 65 [54-74] years; 64% men, 36% women), and 1 614 768 in the US (median [IQR] patient age, 65 [54-74] years; 57% men, 43% women). The age-standardized rate per 100 000 population of IMV was the lowest in England (131; 95% CI, 130-132) compared with Canada (290; 95% CI, 288-292) and the US (614; 95% CI, 614-615). Stratified by age, per capita rates of IMV were more similar across countries among younger patients and diverged markedly in older patients. Among patients aged 80 years or older, the crude rate of IMV per 100 000 population was highest in the US (1788; 95% CI, 1781-1796) compared with Canada (694; 95% CI, 679-709) and England (209; 95% CI, 203-214). Concerning measured comorbidities, 6.3% of admitted patients who received IMV in the US had a diagnosis of dementia (vs 1.4% in England and 1.3% in Canada). Similarly, 5.6% of admitted patients in the US were dependent on dialysis prior to receiving IMV (vs 1.3% in England and 0.3% in Canada).
Conclusions and relevance:
This cohort study found that patients in the US received IMV at a rate 4 times higher than in England and twice that in Canada in 2018. The greatest divergence was in the use of IMV among older adults, and patient characteristics among those who received IMV varied markedly. The differences in overall use of IMV among these countries highlight the need to better understand patient-, clinician-, and systems-level choices associated with the varied use of a limited and expensive resource.
Background
Predicting functional outcome in critically ill patients with traumatic brain injury (TBI) strongly influences end-of-life decisions and information for surrogate decision makers. Despite well-validated prognostic models, clinicians most often rely on their subjective perception of prognosis. In this study, we aimed to compare physicians’ predictions with the International Mission on Prognosis and Analysis of Clinical Trials in TBI (IMPACT) prognostic model for predicting an unfavorable functional outcome at 6 months after moderate or severe TBI.MethodsPREDICT-TBI is a prospective study of patients with moderate to severe TBI. Patients were admitted to a neurocritical care unit and were excluded if they died or had withdrawal of life-sustaining treatments within the first 24 h. In a paired study design, we compared the accuracy of physician prediction on day 1 with the prediction of the IMPACT model as two diagnostic tests in predicting unfavorable outcome 6 months after TBI. Unfavorable outcome was assessed by the Glasgow Outcome Scale from 1 to 3 by using a structured telephone interview. The primary end point was the difference between the discrimination ability of the physician and the IMPACT model assessed by the area under the curve.ResultsOf the 93 patients with inclusion and exclusion criteria, 80 patients reached the primary end point. At 6 months, 29 patients (36%) had unfavorable outcome. A total of 31 clinicians participated in the study. Physicians’ predictions showed an area under the curve of 0.79 (95% confidence interval 0.68–0.89), against 0.80 (95% confidence interval 0.69–0.91) for the laboratory IMPACT model, with no statistical difference (p = 0.88). Both approaches were well calibrated. Agreement between physicians was moderate (κ = 0.56). Lack of experience was not associated with prediction accuracy (p = 0.58).Conclusions
Predictions made by physicians for functional outcome were overall moderately accurate, and no statistical difference was found with the IMPACT models, possibly due to a lack of power. The significant variability between physician assessments suggests prediction could be improved through peer reviewing, with the support of the IMPACT models, to provide a realistic expectation of outcome to families and guide discussions about end-of-life decisions.
Background:
The benefit of ICU in older patients is often debated. There is little knowledge on subjective impressions of excessive care in ICU clinicians combined with objective patient data in real-life cases.
Research question:
Is there a difference in treatment limitation decisions and one-year outcomes in patients above and below 75 years, with and without concordant perceptions of excessive care by two or more ICU clinicians?
Study design and methods:
Reanalysis of the prospective observational DISPROPRICUS study, performed in 56 ICUs. Clinicians (nurses, physicians) completed a daily questionnaire about the appropriateness of care for each of their patients during a 28 day period in 2014. We compared the cumulative incidence of patients with concordant perceptions of excessive care, treatment limitation decisions and the proportion of patients attaining the combined endpoint (death, poor quality of life or not being at home) at one year across age groups via Cox-regression with propensity score weighing and Fisher-exact tests.
Results:
Of 1641 patients, 405 (25%) were ≥75 years. The cumulative incidence of concordant perceptions of excessive care was higher in older patients (13.6% versus 8.5%, p<0.001).In patients with concordant perceptions of excessive care, we found no difference between age groups in risk of death (one-year mortality 83% in both groups, p=1, HR after weighing 1.11, 95%CI 0.74-1.65); treatment limitation decisions (33% versus 31%, HR after weighing 1.11, 95%CI 0.69-2.17) and reaching the combined endpoint at one year (90% versus 93%, p=0.546).In patients without concordant perceptions of excessive care, we found a difference in risk of death (one-year mortality 41% versus 30%, p<0.001, HR after weighting 1.38, 95%CI 1.11-1.73) and treatment limitation decisions (11% versus 5%, p<0.001; HR 2.11, 95%CI 1.37-3.27), though treatment limitation decisions were mostly documented prior to ICU admission. The risk of reaching the combined endpoint was higher in the older (61.6% versus 52.8%, p<0.001).
Interpretation:
Although the incidence of perceptions of excessive care is slightly higher in older patients, there is no difference in treatment limitation decisions and one-year outcomes between older and younger patients once patients are identified by concordant perceptions of excessive care. Additionally, in patients without concordant perceptions the outcomes are worse in the older, pleading against ageism in ICU clinicians.
Rationale: Small trials and professional recommendations support mobilization interventions to improve recovery among critically ill patients, but their real-world effectiveness is unknown. Objective: To evaluate a low-cost, multifaceted mobilization intervention. Methods: We conducted a stepped-wedge cluster-randomized trial across 12 ICUs with diverse case mixes. The primary and secondary samples included patients mechanically ventilated for ⩾48 hours who were ambulatory before admission, and all patients with ICU stays ⩾48 hours, respectively. The mobilization intervention included 1) designation and posting of daily mobilization goals; 2) interprofessional closed-loop communication coordinated by each ICU's facilitator; and 3) performance feedback. Measurements and Main Results: From March 4, 2019 through March 15, 2020, 848 and 1,069 patients were enrolled in the usual care and intervention phases in the primary sample, respectively. The intervention did not increase the primary outcome, patient's maximal Intensive Care Mobility Scale (range, 0-10) score within 48 hours before ICU discharge (estimated mean difference, 0.16; 95% confidence interval, -0.31 to 0.63; P = 0.51). More patients in the intervention (37.2%) than usual care (30.7%) groups achieved the prespecified secondary outcome of ability to stand before ICU discharge (odds ratio, 1.48; 95% confidence interval, 1.02 to 2.15; P = 0.04). Similar results were observed among the 7,115 patients in the secondary sample. The percentage of days on which patients received physical therapy mediated 90.1% of the intervention effect on standing. ICU mortality (31.5% vs. 29.0%), falls (0.7% vs. 0.4%), and unplanned extubations (2.0% vs. 1.8%) were similar between groups (all P > 0.3). Conclusions: A low-cost, multifaceted mobilization intervention did not improve overall mobility but improved patients' odds of standing and was safe. Clinical trial registered with www.clinicaltrials.gov (NCT03863470).
Intravenous (IV) fluids are commonly administered to critically ill children, but clinicians lack effective guidance for the correct dose and duration of therapy resulting in variation of prescribing habits which harm children. It is unknown if clinicians recognize the amount of IV fluid that patients receive. We aimed to determine clinician's accuracy in the identification of the volume of IV fluids patients will receive over the next 24 hours. Prospective cohort study enrolled all patients admitted to the pediatric intensive care unit (PICU) from May to August 2021 at the University of Michigan's C.S. Mott Children's Hospital PICU. For each patient, clinicians estimated the volume of IV fluid that patients will receive in the next 24 hours. The primary outcome was accuracy of the estimation defined as predicted volume of IV fluids versus actual volume administered within 10 mL/kg or 500 mL depending on patient's weight. We tested for differences in accuracy by clinician type using chi-square tests. There were 259 patients for whom 2,295 surveys were completed by 177 clinicians. Clinicians' estimates were accurate 48.8% of the time with a median difference of 10 (1–26) mL/kg. We found that accuracy varied between clinician type: bedside nurses were most accurate at 64.3%, and attendings were least accurate at 30.5%. PICU clinicians have poor recognition of the amount of IV fluids their patients will receive in the subsequent 24-hour period. Estimate accuracy varied by clinician's role and improved over time, which may suggest opportunities for improvement.
Aim:
To explore the cognition and practice on transitional care during the transfer from intensive care unit to a general ward among healthcare professionals in China.
Background:
Due to the significant differences in the medical and humanistic environment at home and abroad, the safety of patients during the transmission from intensive care unit to the general ward is often ignored when their conditions become stable. There are few qualitative studies on the cognition and practice on transitional care during the transfer from intensive care unit to the ward among healthcare professionals in China.
Methods:
With a qualitative research design, 20 medical and nursing staff in the neurosurgery intensive care unit and ward were interviewed from May 2021 to August 2021. NVivo 11.0 software was utilized for Colaizzi's (1978) method of data analysis.
Results:
Based on data analysis, perceptions of transitional care, the influencing factors for transitional care, and the recommendations for improving transitional care were obtained.
Conclusion:
To ensure the continuity of care and improve patient safety during the period from intensive care unit to a general ward in China, we should clarify the expectation for the content of intensive care unit transitional care services, establish the transitional nursing team, guide nursing work, standardize the handover mode and process from intensive care unit to the general ward, promote the communication and coordination of healthcare professionals, and improve the transitional nursing security system from the perspective of institutional level.
Implications for nursing management:
This study can be used as a guide to help health care professionals provide a reference for the comprehensive development of transitional care services and the formulation of targeted intervention measures during the transfer from intensive care unit to a general ward in China.
As a complex critical care space, the ICU is fraught with conflict. Patients, family members, and clinicians may be in conflict with one another around goals of care, injury or disease management, as well as end-of-life issues. Thoracic injury, like other trauma, occurs suddenly, depriving patients’ or family members’ time to prepare for decision-making. Management is often complex and can establish inter-team conflict, especially regarding polytrauma care. Conflict management aims to recognize and defuse triggers before conflict arises, resolve conflict upon emergence, and pursue a resolution acceptable to all disputants. Consultative services including Palliative Care Medicine and Clinical Ethics Consultation may aid in relational dynamics and care continuity. Intractable conflict may require a specifically trained conflict mediator to reach resolution.KeywordsConflictMediationTrustEnd-of-lifePalliative care medicineClinical ethics consultationSurrogateHealthcare teams
Objectives:
Shared decision making is endorsed by critical care organizations; however, there remains confusion about what shared decision making is, when it should be used, and approaches to promote partnerships in treatment decisions. The purpose of this statement is to define shared decision making, recommend when shared decision making should be used, identify the range of ethically acceptable decision-making models, and present important communication skills.
Design:
The American College of Critical Care Medicine and American Thoracic Society Ethics Committees reviewed empirical research and normative analyses published in peer-reviewed journals to generate recommendations. Recommendations approved by consensus of the full Ethics Committees of American College of Critical Care Medicine and American Thoracic Society were included in the statement.
Main results:
Six recommendations were endorsed: 1) Definition: Shared decision making is a collaborative process that allows patients, or their surrogates, and clinicians to make healthcare decisions together, taking into account the best scientific evidence available, as well as the patient's values, goals, and preferences. 2) Clinicians should engage in a shared decision making process to define overall goals of care (including decisions regarding limiting or withdrawing life-prolonging interventions) and when making major treatment decisions that may be affected by personal values, goals, and preferences. 3) Clinicians should use as their "default" approach a shared decision making process that includes three main elements: information exchange, deliberation, and making a treatment decision. 4) A wide range of decision-making approaches are ethically supportable, including patient- or surrogate-directed and clinician-directed models. Clinicians should tailor the decision-making process based on the preferences of the patient or surrogate. 5) Clinicians should be trained in communication skills. 6) Research is needed to evaluate decision-making strategies.
Conclusions:
Patient and surrogate preferences for decision-making roles regarding value-laden choices range from preferring to exercise significant authority to ceding such authority to providers. Clinicians should adapt the decision-making model to the needs and preferences of the patient or surrogate.
CLINICAL SCENARIO You are back where we put you in the previous article1 on diagnostic tests in this series on how to use the medical literature: in the library studying an article that will guide you in interpreting ventilation-perfusion (V/Q) lung scans. Using the criteria in Table 1, you have decided that the Prospective Investigation of Pulmonary Diagnosis (PIOPED) study2 will provide you with valid information. Just then, another physician comes looking for an article to help with the interpretation of V/Q scanning. Her patient is a 28-year-old man whose acute onset of shortness of breath and vague chest pain began shortly after completing a 10-hour auto trip. He experienced several episodes of similar discomfort in the past, but none this severe, and is very apprehensive about his symptoms. After a normal physical examination, electrocardiogram and chest radiograph, and blood gas measurements that show a Pco2 of
Patient reported outcome measures (PROMs) movement has largely been driven by the agenda of researchers or service payers and has failed to focus effectively on improving the quality of care from the patient’s perspective. We use two examples to show how the use of PROMs in everyday practice has the potential to narrow the gap between the clinician’s and patient’s view of clinical reality and help tailor treatment plans to meet the patient’s preferences and needs.
Little is known about functional trajectories of older persons in the year before and after admission to the intensive care unit (ICU) or how pre-ICU functional trajectories affect post-ICU functional trajectories and death.
To characterize functional trajectories in the year before and after ICU admission and to evaluate the associations among pre-ICU functional trajectories and post-ICU functional trajectories, short-term mortality, and long-term mortality.
Prospective cohort study of 754 community-dwelling persons 70 years or older, conducted between March 23, 1998, and December 31, 2012, in greater New Haven, Connecticut. The analytic sample included 291 participants who had at least 1 admission to an ICU through December 2011.
Functional trajectories in the year before and after an ICU admission based on 13 basic, instrumental, and mobility activities. Additional outcomes included short-term (30 day) and long-term (1 year) mortality.
The mean (SD) age of participants was 83.7 (5.5) years. Three distinct pre-ICU functional trajectories identified were minimal disability (29.6%), mild to moderate disability (44.0%), and severe disability (26.5%). Seventy participants (24.1%) experienced early death, defined as death in the hospital (50 participants [17.2%]) or death after hospital discharge but within 30 days of admission (20 participants [6.9%]). Among the remaining 221 participants, 3 distinct post-ICU functional trajectories identified were minimal disability (20.8%), mild to moderate disability (28.1%), and severe disability (51.1%). More than half of the participants (53.4%) experienced functional decline or early death after critical illness. The pre-ICU functional trajectories of mild to moderate disability and severe disability were associated with more than double (adjusted hazard ratio [HR], 2.41; 95% CI, 1.29-4.50) and triple (adjusted HR, 3.84; 95% CI, 1.84-8.03) the risk of death within 1 year of ICU admission, respectively. Other factors associated with 1-year mortality included ICU length of stay (adjusted HR, 1.03; 95% CI, 1.00-1.05), mechanical ventilation (adjusted HR, 2.89; 95% CI, 1.91-4.37), and shock (adjusted HR, 2.68; 95% CI, 1.63-4.38).
Among older persons with critical illness, more than half died within 1 month or experienced significant functional decline over the following year, with particularly poor outcomes in those who had high levels of premorbid disability. These results may help to inform discussions about prognosis and goals of care before and during critical illness.
There have been few detailed, in-person interviews and examinations to obtain follow-up data on 5-year outcomes among survivors of the acute respiratory distress syndrome (ARDS).
We evaluated 109 survivors of ARDS at 3, 6, and 12 months and at 2, 3, 4, and 5 years after discharge from the intensive care unit. At each visit, patients were interviewed and examined; underwent pulmonary-function tests, the 6-minute walk test, resting and exercise oximetry, chest imaging, and a quality-of-life evaluation; and reported their use of health care services.
At 5 years, the median 6-minute walk distance was 436 m (76% of predicted distance) and the Physical Component Score on the Medical Outcomes Study 36-Item Short-Form Health Survey was 41 (mean norm score matched for age and sex, 50). With respect to this score, younger patients had a greater rate of recovery than older patients, but neither group returned to normal predicted levels of physical function at 5 years. Pulmonary function was normal to near-normal. A constellation of other physical and psychological problems developed or persisted in patients and family caregivers for up to 5 years. Patients with more coexisting illnesses incurred greater 5-year costs.
Exercise limitation, physical and psychological sequelae, decreased physical quality of life, and increased costs and use of health care services are important legacies of severe lung injury.
Cognitive impairment and functional disability are major determinants of caregiving needs and societal health care costs. Although the incidence of severe sepsis is high and increasing, the magnitude of patients' long-term cognitive and functional limitations after sepsis is unknown.
To determine the change in cognitive impairment and physical functioning among patients who survive severe sepsis, controlling for their presepsis functioning.
A prospective cohort involving 1194 patients with 1520 hospitalizations for severe sepsis drawn from the Health and Retirement Study, a nationally representative survey of US residents (1998-2006). A total of 9223 respondents had a baseline cognitive and functional assessment and had linked Medicare claims; 516 survived severe sepsis and 4517 survived a nonsepsis hospitalization to at least 1 follow-up survey and are included in the analysis.
Personal interviews were conducted with respondents or proxies using validated surveys to assess the presence of cognitive impairment and to determine the number of activities of daily living (ADLs) and instrumental ADLs (IADLs) for which patients needed assistance.
Survivors' mean age at hospitalization was 76.9 years. The prevalence of moderate to severe cognitive impairment increased 10.6 percentage points among patients who survived severe sepsis, an odds ratio (OR) of 3.34 (95% confidence interval [CI], 1.53-7.25) in multivariable regression. Likewise, a high rate of new functional limitations was seen following sepsis: in those with no limits before sepsis, a mean 1.57 new limitations (95% CI, 0.99-2.15); and for those with mild to moderate limitations before sepsis, a mean of 1.50 new limitations (95% CI, 0.87-2.12). In contrast, nonsepsis general hospitalizations were associated with no change in moderate to severe cognitive impairment (OR, 1.15; 95% CI, 0.80-1.67; P for difference vs sepsis = .01) and with the development of fewer new limitations (mean among those with no limits before hospitalization, 0.48; 95% CI, 0.39-0.57; P for difference vs sepsis <.001 and mean among those with mild to moderate limits, 0.43; 95% CI, 0.23-0.63; P for difference = .001). The declines in cognitive and physical function persisted for at least 8 years.
Severe sepsis in this older population was independently associated with substantial and persistent new cognitive impairment and functional disability among survivors. The magnitude of these new deficits was large, likely resulting in a pivotal downturn in patients' ability to live independently.
Growing numbers of critically ill patients receive prolonged mechanical ventilation. Little is known about the patterns of care as patients transition from acute care hospitals to postacute care facilities or about the associated resource utilization.
To describe 1-year trajectories of care and resource utilization for patients receiving prolonged mechanical ventilation.
1-year prospective cohort study.
5 intensive care units at Duke University Medical Center, Durham, North Carolina.
126 patients receiving prolonged mechanical ventilation (defined as ventilation for >or=4 days with tracheostomy placement or ventilation for >or=21 days without tracheostomy), as well as their 126 surrogates and 54 intensive care unit physicians, enrolled consecutively over 1 year.
Patients and surrogates were interviewed in the hospital, as well as 3 and 12 months after discharge, to determine patient survival, functional status, and facility type and duration of postdischarge care. Physicians were interviewed in the hospital to elicit prognoses. Institutional billing records were used to assign costs for acute care, outpatient care, and interfacility transportation. Medicare claims data were used to assign costs for postacute care.
103 (82%) hospital survivors had 457 separate transitions in postdischarge care location (median, 4 transitions [interquartile range, 3 to 5 transitions]), including 68 patients (67%) who were readmitted at least once. Patients spent an average of 74% (95% CI, 68% to 80%) of all days alive in a hospital or postacute care facility or receiving home health care. At 1 year, 11 patients (9%) had a good outcome (alive with no functional dependency), 33 (26%) had a fair outcome (alive with moderate dependency), and 82 (65%) had a poor outcome (either alive with complete functional dependency [4 patients; 21%] or dead [56 patients; 44%]). Patients with poor outcomes were older, had more comorbid conditions, and were more frequently discharged to a postacute care facility than patients with either fair or good outcomes (P < 0.05 for all). The mean cost per patient was 285,467), and total cohort cost was 3.5 million per independently functioning survivor at 1 year.
The results of this single-center study may not be applicable to other centers.
Patients receiving prolonged mechanical ventilation have multiple transitions of care, resulting in substantial health care costs and persistent, profound disability. The optimism of surrogate decision makers should be balanced by discussions of these outcomes when considering a course of prolonged life support.
None.
The objective of this study was to refine the APACHE (Acute Physiology, Age, Chronic Health Evaluation) methodology in order to more accurately predict hospital mortality risk for critically ill hospitalized adults. We prospectively collected data on 17,440 unselected adult medical/surgical intensive care unit (ICU) admissions at 40 US hospitals (14 volunteer tertiary-care institutions and 26 hospitals randomly chosen to represent intensive care services nationwide). We analyzed the relationship between the patient's likelihood of surviving to hospital discharge and the following predictive variables: major medical and surgical disease categories, acute physiologic abnormalities, age, preexisting functional limitations, major comorbidities, and treatment location immediately prior to ICU admission. The APACHE III prognostic system consists of two options: (1) an APACHE III score, which can provide initial risk stratification for severely ill hospitalized patients within independently defined patient groups; and (2) an APACHE III predictive equation, which uses APACHE III score and reference data on major disease categories and treatment location immediately prior to ICU admission to provide risk estimates for hospital mortality for individual ICU patients. A five-point increase in APACHE III score (range, 0 to 299) is independently associated with a statistically significant increase in the relative risk of hospital death (odds ratio, 1.10 to 1.78) within each of 78 major medical and surgical disease categories. The overall predictive accuracy of the first-day APACHE III equation was such that, within 24 h of ICU admission, 95 percent of ICU admissions could be given a risk estimate for hospital death that was within 3 percent of that actually observed (r2 = 0.41; receiver operating characteristic = 0.90). Recording changes in the APACHE III score on each subsequent day of ICU therapy provided daily updates in these risk estimates. When applied across the individual ICUs, the first-day APACHE III equation accounted for the majority of variation in observed death rates (r2 = 0.90, p less than 0.0001).
The questions patients are asked about their preferences with regard to life-sustaining treatment usually focus on specific interventions, but the outcomes of treatment and their likelihood affect patients' preferences.
We administered a questionnaire about treatment preferences to 226 persons who were 60 years of age or older and who had a limited life expectancy due to cancer, congestive heart failure, or chronic obstructive pulmonary disease. The study participants were asked whether they would want to receive a given treatment, first when the outcome was known with certainty and then with different likelihoods of an adverse outcome. The outcome without treatment was specified as death from the underlying disease.
The burden of treatment (i.e., the length of the hospital stay, extent of testing, and invasiveness of interventions), the outcome, and the likelihood of the outcome all influenced treatment preferences. For a low-burden treatment with the restoration of current health, 98.7 percent of participants said they would choose to receive the treatment (rather than not receive it and die), but 11.2 percent of these participants would not choose the treatment if it had a high burden. If the outcome was survival but with severe functional impairment or cognitive impairment, 74.4 percent and 88.8 percent of these participants, respectively, would not choose treatment. The number of participants who said they would choose treatment declined as the likelihood of an adverse outcome increased, with fewer participants choosing treatment when the possible outcome was functional or cognitive impairment than when it was death. Preferences did not differ according to the primary diagnosis.
Advance care planning should take into account patients' attitudes toward the burden of treatment, the possible outcomes, and their likelihood. The likelihood of adverse functional and cognitive outcomes of treatment requires explicit consideration.
This is a review of the Health Utilities Index (HUI) multi-attribute health-status classification systems, and single- and multi-attribute utility scoring systems. HUI refers to both HUI Mark 2 (HUI2) and HUI Mark 3 (HUI3) instruments. The classification systems provide compact but comprehensive frameworks within which to describe health status. The multi-attribute utility functions provide all the information required to calculate single-summary scores of health-related quality of life (HRQL) for each health state defined by the classification systems. The use of HUI in clinical studies for a wide variety of conditions in a large number of countries is illustrated. HUI provides comprehensive, reliable, responsive and valid measures of health status and HRQL for subjects in clinical studies. Utility scores of overall HRQL for patients are also used in cost-utility and cost-effectiveness analyses. Population norm data are available from numerous large general population surveys. The widespread use of HUI facilitates the interpretation of results and permits comparisons of disease and treatment outcomes, and comparisons of long-term sequelae at the local, national and international levels.
Physical function is an important measure of success of many medical and surgical interventions. Ability to adjust for comorbid disease is essential in health services research and epidemiologic studies. Current indices have primarily been developed with mortality as the outcome, and are not sensitive enough when the outcome is physical function. The objective of this study was to develop a self-administered Functional Comorbidity Index with physical function as the outcome.
The index was developed using two databases: a cross-sectional, simple random sample of 9,423 Canadian adults and a sample of 28,349 US adults seeking treatment for spine ailments. The primary outcome measure was the SF-36 physical function (PF) subscale.
The Functional Comorbidity Index, an 18-item list of diagnoses, showed stronger association with physical function (model R(2) = 0.29) compared with the Charlson (model R(2) = 0.18), and Kaplan-Feinstein (model R(2) = 0.07) indices. The Functional Comorbidity Index correctly classified patients into high and low function, in 77% of cases.
This new index contains diagnoses such as arthritis not found on indices used to predict mortality, and the FCI explained more variance in PF scores compared to indices designed to predict mortality.
Rationale:
Intensive care unit (ICU) patients' expected post-discharge outcomes are rarely discussed in family meetings despite this information being centrally important to patients and their families.
Objectives:
To characterize intensivist-identified barriers and facilitators to discussing postdischarge outcomes with surrogates of ICU patients.
Methods:
Qualitative study conducted via one-on-one, semi-structured telephone interviews with 23 intensivists from 20 hospitals with accreditation council for graduate medical education accredited critical care medicine programs in 16 states. A limited application of grounded theory methods were used to code transcribed interviews and identify themes and illustrative quotes.
Measurements and main results:
Intensivists reported tension between their professional responsibility to discuss likely functional outcomes versus uncertainty about their ability to predict those outcomes in an individual patient. They cited three main barriers as limiting their ability to conduct conversations about post-discharge outcomes with ICU surrogates: 1) incorrectly optimistic expectations for recovery among ICU surrogates, 2) having little or no contact with their patients after ICU discharge, and 3) minimal confidence applying existing outcomes research to individual patients. Despite these barriers, experience talking to ICU surrogates, seeing ICU survivors in the out-patient setting, and trusted research on functional outcomes were identified as important facilitators to discussing likely patient outcomes with surrogates. Intensivists generally welcomed questions from surrogates about post-discharge outcomes as opportunities to initiate conversations about prognosis and patient values.
Conclusions:
In this sample of intensivists from 20 academic hospitals, experience conducting conversations with surrogates and interactions with ICU survivors as out-patients were identified as facilitating discussion of expected post-discharge outcomes while optimistic surrogate expectations and prognostic uncertainty were barriers. There was tension between self-perceived ability to prognosticate and belief in a professional obligation to discuss patient outcomes.
Importance
Misperceptions about prognosis by individuals making decisions for incapacitated critically ill patients (surrogates) are common and often attributed to poor comprehension of medical information.
Objective
To determine the prevalence of and factors related to physician-surrogate discordance about prognosis in intensive care units (ICUs).
Design, Setting, and Participants
Mixed-methods study comprising quantitative surveys and qualitative interviews conducted in 4 ICUs at a major US medical center involving surrogate decision makers and physicians caring for patients at high risk of death from January 4, 2005, to July 10, 2009.
Main Outcomes and Measures
Discordance about prognosis, defined as a difference between a physician’s and a surrogate’s prognostic estimates of at least 20%; misunderstandings by surrogates (defined as any difference between a physician’s prognostic estimate and a surrogate’s best guess of that estimate); differences in belief (any difference between a surrogate’s actual estimate and their best guess of the physician’s estimate).
Results
Two hundred twenty-nine surrogate decision makers (median age, 47 [interquartile range {IQR}, 35-56] years; 68% women) and 99 physicians were involved in the care of 174 critically ill patients (median age, 60 [IQR, 47-74] years; 44% women). Physician-surrogate discordance about prognosis occurred in 122 of 229 instances (53%; 95% CI, 46.8%-59.7%). In 65 instances (28%), discordance was related to both misunderstandings by surrogates and differences in belief about the patient’s prognosis; 38 (17%) were related to misunderstandings by surrogates only; 7 (3%) were related to differences in belief only; and data were missing for 12. Seventy-five patients (43%) died. Surrogates’ prognostic estimates were much more accurate than chance alone, but physicians’ prognostic estimates were statistically significantly more accurate than surrogates’ (C statistic, 0.83 vs 0.74; absolute difference, 0.094; 95% CI, 0.024-0.163; P = .008). Among 71 surrogates interviewed who had beliefs about the prognosis that were more optimistic than that of the physician, the most common reasons for optimism were a need to maintain hope to benefit the patient (n = 34), a belief that the patient had unique strengths unknown to the physician (n = 24), and religious belief (n = 19).
Conclusions and Relevance
Among critically ill patients, discordant expectations about prognosis were common between patients’ physicians and surrogate decision makers and were related to misunderstandings by surrogates about physicians’ assessments of patients’ prognoses and differences in beliefs about patients’ prognoses.
Rationale:
Disability risk groups and one-year outcome after ≥ 7 days of mechanical ventilation in medical / surgical ICU patients are unknown and may inform education, prognostication, rehabilitation and study design.
Objectives:
To stratify patients for post-ICU disability and recovery to one year after critical illness.
Methods:
We evaluated a multi-center cohort of 391 medical / surgical ICU patients who received ≥ 1 week of MV at 7 days, 3, 6, and 12 months after ICU discharge. Disability risk groups were identified using recursive partitioning modeling.
Results:
The 7-day post-ICU Functional Independence Measure (FIM) determined the recovery trajectory to one-year after ICU discharge and was an independent risk factor for 1-year mortality. The 7-day post-ICU FIM was predicted by age and ICU length of stay (LOS). By 2 weeks of MV, ICU patients could be stratified into four disability groups characterized by increasing risk for post ICU disability, ICU and post ICU healthcare utilization and disposition. Patients < 42 years with ICU length of stay < 2 weeks had the best function and fewest deaths at one year compared to patients > 66 years with ICU length of stay > 2 weeks who sustained the worst disability and 40% 1-year mortality. Depressive symptoms (17%) and posttraumatic stress disorder (18%) persisted at 1 year.
Conclusions:
ICU survivors of ≥ 1 week of mechanical ventilation may be stratified into 4 disability groups based on age and ICU LOS. These groups determine 1-year recovery and healthcare utilization and are independent of admitting diagnosis and illness severity.
Background:
Growing numbers of critically ill patients receive prolonged mechanical ventilation. Little is known about the patterns of care as patients transition from acute care hospitals to postacute care facilities or about the associated resource utilization.
Objective:
To describe 1-year trajectories of care and resource utilization for patients receiving prolonged mechanical ventilation.
Design:
1-year prospective cohort study.
Setting:
5 intensive care units at Duke University Medical Center, Durham, North Carolina.
Participants:
126 patients receiving prolonged mechanical ventilation (defined as ventilation for >or=4 days with tracheostomy placement or ventilation for >or=21 days without tracheostomy), as well as their 126 surrogates and 54 intensive care unit physicians, enrolled consecutively over 1 year.
Measurements:
Patients and surrogates were interviewed in the hospital, as well as 3 and 12 months after discharge, to determine patient survival, functional status, and facility type and duration of postdischarge care. Physicians were interviewed in the hospital to elicit prognoses. Institutional billing records were used to assign costs for acute care, outpatient care, and interfacility transportation. Medicare claims data were used to assign costs for postacute care.
Results:
103 (82%) hospital survivors had 457 separate transitions in postdischarge care location (median, 4 transitions [interquartile range, 3 to 5 transitions]), including 68 patients (67%) who were readmitted at least once. Patients spent an average of 74% (95% CI, 68% to 80%) of all days alive in a hospital or postacute care facility or receiving home health care. At 1 year, 11 patients (9%) had a good outcome (alive with no functional dependency), 33 (26%) had a fair outcome (alive with moderate dependency), and 82 (65%) had a poor outcome (either alive with complete functional dependency [4 patients; 21%] or dead [56 patients; 44%]). Patients with poor outcomes were older, had more comorbid conditions, and were more frequently discharged to a postacute care facility than patients with either fair or good outcomes (P < 0.05 for all). The mean cost per patient was 285,467), and total cohort cost was 3.5 million per independently functioning survivor at 1 year.
Limitation:
The results of this single-center study may not be applicable to other centers.
Conclusion:
Patients receiving prolonged mechanical ventilation have multiple transitions of care, resulting in substantial health care costs and persistent, profound disability. The optimism of surrogate decision makers should be balanced by discussions of these outcomes when considering a course of prolonged life support.
Primary funding source:
None.
Background
Survivors of critical illness often have a prolonged and disabling form of cognitive impairment that remains inadequately characterized. Methods
We enrolled adults with respiratory failure or shock in the medical or surgical intensive care unit (ICU), evaluated them for in-hospital delirium, and assessed global cognition and executive function 3 and 12 months after discharge with the use of the Repeatable Battery for the Assessment of Neuropsychological Status (population age-adjusted mean [SD] score, 10015, with lower values indicating worse global cognition) and the Trail Making Test, Part B (population age-, sex-, and education-adjusted mean score, 5010, with lower scores indicating worse executive function). Associations of the duration of delirium and the use of sedative or analgesic agents with the outcomes were assessed with the use of linear regression, with adjustment for potential confounders. ResultsOf the 821 patients enrolled, 6% had cognitive impairment at baseline, and delirium developed in 74% during the hospital stay. At 3 months, 40% of the patients had global cognition scores that were 1.5 SD below the population means (similar to scores for patients with moderate traumatic brain injury), and 26% had scores 2 SD below the population means (similar to scores for patients with mild Alzheimer's disease). Deficits occurred in both older and younger patients and persisted, with 34% and 24% of all patients with assessments at 12 months that were similar to scores for patients with moderate traumatic brain injury and scores for patients with mild Alzheimer's disease, respectively. A longer duration of delirium was independently associated with worse global cognition at 3 and 12 months (P=0.001 and P=0.04, respectively) and worse executive function at 3 and 12 months (P=0.004 and P=0.007, respectively). Use of sedative or analgesic medications was not consistently associated with cognitive impairment at 3 and 12 months. Conclusions
Patients in medical and surgical ICUs are at high risk for long-term cognitive impairment. A longer duration of delirium in the hospital was associated with worse global cognition and executive function scores at 3 and 12 months. (Funded by the National Institutes of Health and others; BRAIN-ICU ClinicalTrials.gov number, NCT00392795.) In this study, patients treated in ICUs were at high risk for new cognitive impairment during 12 months of follow-up, with 24% of patients having deficits similar in severity to those in Alzheimer's disease. A longer duration of delirium was associated with worse cognitive scores. Survivors of critical illness frequently have a prolonged and poorly understood form of cognitive dysfunction,(1)-(4) which is characterized by new deficits (or exacerbations of preexisting mild deficits) in global cognition or executive function. This long-term cognitive impairment after critical illness may be a growing public health problem, given the large number of acutely ill patients being treated in intensive care units (ICUs) globally.(5) Among older adults, cognitive decline is associated with institutionalization,(6) hospitalization,(7) and considerable annual societal costs.(8),(9) Yet little is known about the epidemiology of long-term cognitive impairment after critical illness. Delirium, a form of acute brain ...
Mobility limitations are common in older adults, affecting the physical, psychological, and social aspects of an older adult's life.
To identify mobility risk factors, screening tools, medical management, need for physical therapy, and efficacy of exercise interventions for older primary care patients with limited mobility.
Search of PubMed and PEDro from January 1985 to March 31, 2013, using the search terms mobility limitation, walking difficulty, and ambulatory difficulty to identify English-language, peer-reviewed systematic reviews, meta-analyses, and Cochrane reviews assessing mobility limitation and interventions in community-dwelling older adults. Articles not appearing in the search referenced by reviewed articles were also evaluated.
The most common risk factors for mobility impairment are older age, low physical activity, obesity, strength or balance impairment, and chronic diseases such as diabetes or arthritis. Several tools are available to assess mobility in the ambulatory setting. Referral to physical therapy is appropriate, because physical therapists can assess mobility limitations and devise curative or function-enhancing interventions. Relatively few studies support therapeutic exercise to improve mobility limitation. Strong evidence supports resistance and balance exercises for improving mobility-limiting physical weakness and balance disorders. Assessing a patient's physical environment and the patient's ability to adapt to it using mobility devices is critical.
Identification of older adults at risk for mobility limitation can be accomplished through routine screening in the ambulatory setting. Addressing functional deficits and environmental barriers with exercise and mobility devices can lead to improved function, safety, and quality of life for patients with mobility limitations.
Survivors of critical illness have a high rate of cognitive impairments that may persist years after hospital discharge. Data are lacking regarding whether cognitive screening tests administered at hospital discharge can be used to predict which critically ill patients are likely to have long-term cognitive sequelae.
This prospective study assessed whether two cognitive screening tests, the Mini-Mental State Examination (MMSE) and Mini-Cog, administered at hospital discharge, predict cognitive sequelae in survivors of critical illness 6 months after hospital discharge.
Seventy critically ill patients completed the MMSE and Mini-Cog just before hospital discharge. Of these 70 patients, 53 completed a neuropsychological battery 6 months after hospital discharge.
At hospital discharge, 45 patients (64%) had impaired performance on the MMSE (score < 27, mean = 24.4) and 32 (45%) on the Mini-Cog. Twenty-seven patients (39%) were impaired on both the MMSE and Mini-Cog, whereas only 20 patients (28%) had scores in the normal range on both tests. Cognitive sequelae occurred in 57% of survivors (30 of 53) at 6 months, with predominant dysfunction in the memory (38%) and executive (36%) domains. Logistic regression analyses showed that neither the MMSE nor the Mini-Cog predicted cognitive sequelae at 6 months.
A large number of critically ill survivors had cognitive impairments, as assessed by the MMSE and Mini-Cog, at hospital discharge. However, the MMSE and Mini-Cog scores did not predict long-term cognitive sequelae at 6-month follow-up and cannot be used as surrogate endpoints for long-term cognitive impairment.
Home return after critical care is very important not only to patients and families. To move back home, patients have to fulfill two conditions: survive, and have a relatively good functional status. In addition, home return could be considered a low-cost outcome because of the reduced permanent healthcare costs.
To determine the factors influencing the home-return probability of critically ill elderly patients 6 months after an intensive care unit (ICU) admission, we analyzed a cohort of patients aged 65 years or older admitted to an ICU. Demographic and social parameters, as well as admission diagnosis, underlying diseases, severity scores, ICU stay parameters, and complications were recorded. The final outcome was the place of stay (or death) 180 days after ICU admission.
Of 526 patients, 72% of the cohort and 93% of hospital survivors were able to return to their homes. Among the variables used in the multivariate logistic regression, advanced age, length of hospital stay before ICU admission, severity of acute illness, diagnosis category, and complications, as well as certain comorbidities, such as chronic heart failure or a neoplasia, were independently negatively associated with a home return.
Some interesting factors were identified in this single-center study. They could be considered for a multicenter study to build a universal prediction model for home return. Home return could be used for elderly patients as a surrogate for outcomes that are very important to the elderly but also to health politics.
We tested the accuracy of predictions of impending death for medical intensive care unit patients, offered daily by their professional medical caretakers.
For 560 medical intensive care unit patients, on each medical intensive care unit day, we asked their attending physicians, fellows, residents, and registered nurses one question: "Do you think this patient will die in the hospital or survive to be discharged?"
We obtained>6,000 predictions on 2018 medical intensive care unit patient days. Seventy-five percent of MICU patients who stayed≥4 days had discordant predictions; that is, at least one caretaker predicted survival, whereas others predicted death before discharge. Only 107 of 206 (52%) patients with a prediction of "death before discharge" actually died in hospital. This number rose to 66% (96 of 145) for patients with 1 day of corroborated (i.e., >1) prediction of "death," and to 84% (79 of 94) with at least 1 unanimous day of predictions of death. However, although positive predictive value rose with increasingly stringent prediction criteria, sensitivity fell so that the area under the receiver-operator characteristic curve did not differ for single, corroborated, or unanimous predictions of death. Subsets of older (>65 yrs) and ventilated medical intensive care unit patients revealed parallel findings.
1) Roughly half of all medical intensive care unit patients predicted to die in hospital survived to discharge nonetheless. 2) More highly corroborated predictions had better predictive value; although, approximately 15% of patients survived unexpectedly, even when predicted to die by all medical caretakers.
To compare prolonged mechanical ventilation decision-makers' expectations for long-term patient outcomes with prospectively observed outcomes and to characterize important elements of the surrogate-physician interaction surrounding prolonged mechanical ventilation provision. Prolonged mechanical ventilation provision is increasing markedly despite poor patient outcomes. Misunderstanding prognosis in the prolonged mechanical ventilation decision-making process could provide an explanation for this phenomenon.
Prospective observational cohort study.
Academic medical center.
A total of 126 patients receiving prolonged mechanical ventilation.
None.
Participants were interviewed at the time of tracheostomy placement about their expectations for 1-yr patient survival, functional status, and quality of life. These expectations were then compared with observed 1-yr outcomes measured with validated questionnaires. The 1-yr follow-up was 100%, with the exception of patient death or cognitive inability to complete interviews. At 1 yr, only 11 patients (9%) were alive and independent of major functional status limitations. Most surrogates reported high baseline expectations for 1-yr patient survival (n = 117, 93%), functional status (n = 90, 71%), and quality of life (n = 105, 83%). In contrast, fewer physicians described high expectations for survival (n = 54, 43%), functional status (n = 7, 6%), and quality of life (n = 5, 4%). Surrogate-physician pair concordance in expectations was poor (all kappa = <0.08), as was their accuracy in outcome prediction (range = 23%-44%). Just 33 surrogates (26%) reported that physicians discussed what to expect for patients' likely future survival, general health, and caregiving needs.
One-year patient outcomes for prolonged mechanical ventilation patients were significantly worse than expected by patients' surrogates and physicians. Lack of prognostication about outcomes, discordance between surrogates and physicians about potential outcomes, and surrogates' unreasonably optimistic expectations seem to be potentially modifiable deficiencies in surrogate-physician interactions.
Many physicians are reluctant to discuss a patient's prognosis when there is significant prognostic uncertainty.
We sought to understand surrogate decision makers' views regarding whether physicians should discuss prognosis in the face of uncertainty.
We conducted semi-structured interviews with 179 surrogates for 142 incapacitated patients at high risk of death in four intensive care units at an academic medical center. The interviews explored surrogates' attitudes about whether physicians should discuss prognosis when they cannot be certain their prognostic estimates are correct. We used constant comparative methods to analyze the transcripts. Validation methods included triangulation by multidisciplinary analysis and member checking.
Eighty-seven percent (155/179) of surrogates wanted physicians to discuss an uncertain prognosis. We identified five main reasons for this, including surrogates' belief that prognostic uncertainty is unavoidable, that physicians are their only source for prognostic information, and that discussing prognostic uncertainty leaves room for realistic hope, increases surrogates' trust in the physician, and signals a need to prepare for possible bereavement. Twelve percent (22/179) of surrogates felt that discussions about an uncertain prognosis should be avoided. The main explanation was that it is not worth the potential emotional distress if the prognostications are incorrect. Surrogates suggested that physicians should explicitly discuss uncertainty when prognosticating.
The majority of surrogates of patients that are critically ill want physicians to disclose their prognostic estimates even if they cannot be certain they are correct. This stems from surrogates' belief that prognostic uncertainty is simultaneously unavoidable and acceptable.
Methods of evaluating and comparing the performance of diagnostic tests are of increasing importance as new tests are developed and marketed. When a test is based on an observed variable that lies on a continuous or graded scale, an assessment of the overall value of the test can be made through the use of a receiver operating characteristic (ROC) curve. The curve is constructed by varying the cutpoint used to determine which values of the observed variable will be considered abnormal and then plotting the resulting sensitivities against the corresponding false positive rates. When two or more empirical curves are constructed based on tests performed on the same individuals, statistical analysis on differences between curves must take into account the correlated nature of the data. This paper presents a nonparametric approach to the analysis of areas under correlated ROC curves, by using the theory on generalized U-statistics to generate an estimated covariance matrix.
The Index of Independence in Activities of Daily Living (ADL), now in frequent use in rehabilitation settings, has application for prevention of disability and maintenance of rehabilitation gains in the aging person in all settings. Since the Index is sensitive to changes in meaningful self-care functions, uses well-defined criteria, and can be broadly taught to non-professionals, it has considerable practical value as a longitudinal measure of change and predictor of adaptive capacity in terms of community residences and congregate living facilities.
Contemporary textbooks of internal medicine give scant attention to the prognosis of diseases. Has this always been the case? If not, when and why did prognosis come to be de-emphasized? Using a highly regarded, standard medical textbook initially authored by William Osler, The Principles and Practice of Medicine, I performed qualitative and quantitative content analysis of entries regarding lobar pneumonia in selected editions published between 1892 and 1988, with special attention to the period between 1892 and 1947. I chose lobar pneumonia because it was a leading cause of death throughout this period and because it is recognizable across time, thus making it possible to follow the evolution in clinical thinking about prognosis while holding constant the diagnosis. I argue that two powerful forces converged to lead to the ellipsis of prognosis: (1) the emergence of effective therapy, and (2) a fundamental change in the cognitive basis of medicine. With respect to the former, I show that there is a complementary, inverse relationship between the clinical acts of prognostication and therapy; as one increases in salience in the management of a disease, the other decreases. With respect to the latter, I argue that the particular clinical facts deemed to be important about a patient's case have shifted over time, and I explore changes in the clinical and cognitive foundations of physicians' estimation of patients' prognoses-in particular, "symptoms" and "complications." I conclude that, concurrent with a shift in clinical thought from an individual-based to a diagnosis-based conceptualization of disease, prognosis came to be seen as intrinsic to diagnosis and therapy, and explicit attention to prognosis consequently diminished.
Likelihood ratios are one of the best measures of diagnostic accuracy, although they are seldom used, because interpreting them requires a calculator to convert back and forth between "probability" and "odds" of disease. This article describes a simpler method of interpreting likelihood ratios, one that avoids calculators, nomograms, and conversions to "odds" of disease. Several examples illustrate how the clinician can use this method to refine diagnostic decisions at the bedside.
In critically ill patients who are receiving mechanical ventilation, the factors associated with physicians' decisions to withdraw ventilation in anticipation of death are unclear. The objective of this study was to examine the clinical determinants that were associated with the withdrawal of mechanical ventilation.
We studied adults who were receiving mechanical ventilation in 15 intensive care units, recording base-line physiological characteristics, daily Multiple Organ Dysfunction Scores, the patient's decision-making ability, the type of life support administered, the use of do-not-resuscitate orders, the physician's prediction of the patient's status, and the physician's perceptions of the patient's preferences about the use of life support. We examined the relation between these factors and withdrawal of mechanical ventilation, using Cox proportional-hazards regression analysis.
Of 851 patients who were receiving mechanical ventilation, 539 (63.3 percent) were successfully weaned, 146 (17.2 percent) died while receiving mechanical ventilation, and 166 (19.5 percent) had mechanical ventilation withdrawn. The need for inotropes or vasopressors was associated with withdrawal of the ventilator (hazard ratio, 1.78; 95 percent confidence interval, 1.20 to 2.66; P=0.004), as were the physician's prediction that the patient's likelihood of survival in the intensive care unit was less than 10 percent (hazard ratio, 3.49; 95 percent confidence interval, 1.39 to 8.79; P=0.002), the physician's prediction that future cognitive function would be severely impaired (hazard ratio, 2.51; 95 percent confidence interval, 1.28 to 4.94; P=0.04), and the physician's perception that the patient did not want life support used (hazard ratio, 4.19; 95 percent confidence interval, 2.57 to 6.81; P<0.001).
Rather than age or the severity of the illness and organ dysfunction, the strongest determinants of the withdrawal of ventilation in critically ill patients were the physician's perception that the patient preferred not to use life support, the physician's predictions of a low likelihood of survival in the intensive care unit and a high likelihood of poor cognitive function, and the use of inotropes or vasopressors.
Items such as physical exam findings, radiographic interpretations, or other diagnostic tests often rely on some degree of subjective interpretation by observers. Studies that measure the agreement between two or more observers should include a statistic that takes into account the fact that observers will sometimes agree or disagree simply by chance. The kappa statistic (or kappa coefficient) is the most commonly used statistic for this purpose. A kappa of 1 indicates perfect agreement, whereas a kappa of 0 indicates agreement equivalent to chance. A limitation of kappa is that it is affected by the prevalence of the finding under observation. Methods to overcome this limitation have been described.
Guidelines for conducting studies and reading medical literature on diagnostic tests have been published: Requirements for the selection of cases and controls, and for ensuring a correct reference standard are now clarified. Our objective was to provide tables for sample size determination in this context.
In the usual situation, where the prevalence Prev of the disease of interest is <0.50, one first determines the minimal number Ncases of cases required to ensure a given precision of the sensitivity estimate. Computations are based on the binomial distribution, for user-specified type I and type II error levels. The minimal number N(controls) of controls is then derived so as to allow for representativeness of the study population, according to Ncontrols=Ncases [(1-Prev)/Prev].
Tables give the values of Ncases corresponding to expected sensitivities from 0.60 to 0.99, acceptable lower 95% confidence limits from 0.50 to 0.98, and 5% probability of the estimated lower confidence limit being lower than the acceptable level.
When designing diagnostic test studies, sample size calculations should be performed in order to guarantee the design accuracy.
Risk-prediction models offer potential advantages over physician predictions of outcomes in the intensive care unit (ICU). Our systematic review compared the accuracy of ICU physicians' and scoring system predictions of ICU or hospital mortality of critically ill adults.
MEDLINE (1966-2005), CINAHL (1982-2005), Ovid Healthstar (1975-2004), EMBASE (1980-2005), SciSearch (1980-2005), PsychLit (1985-2004), the Cochrane Library (Issue 1, 2005), PubMed "related articles," personal files, abstract proceedings, and reference lists.
We considered all studies that compared physician predictions of ICU or hospital survival of critically ill adults to an objective scoring system, computer model, or prediction rule. We excluded studies if they focused exclusively on the development or economic evaluation of a scoring system, computer model, or prediction rule.
We independently abstracted data and assessed study quality in duplicate. We determined summary receiver operating characteristic curves and areas under the summary receiver operating characteristic curves+/-se and summary diagnostic odds ratios.
We included 12 observational studies of moderate methodological quality. The area under the summary receiver operating characteristic curves for seven studies was 0.85+/-0.03 for physician predictions compared with 0.63+/-0.06 for scoring system predictions (p=.002). Physicians' summary diagnostic odds ratios derived from the area under the summary receiver operating characteristic curves were significantly higher (12.43; 95% confidence interval 5.47, 27.11) than scoring systems' summary diagnostic odds ratios (2.25; 95% confidence interval 0.78, 6.52, p=.001). Combined results of all 12 studies indicated that physicians predict mortality more accurately than do scoring systems: ratio of diagnostic odds ratios (95% confidence interval) 1.92 (1.19, 3.08) (p=.007).
Observational studies suggest that ICU physicians discriminate between survivors and nonsurvivors more accurately than do scoring systems in the first 24 hrs of ICU admission. The overall accuracy of both predictions of patient mortality was moderate, implying limited usefulness of outcome prediction in the first 24 hrs for clinical decision making.
To develop clinical practice guidelines for the support of the patient and family in the adult, pediatric, or neonatal patient-centered ICU.
A multidisciplinary task force of experts in critical care practice was convened from the membership of the American College of Critical Care Medicine (ACCM) and the Society of Critical Care Medicine (SCCM) to include representation from adult, pediatric, and neonatal intensive care units.
The task force members reviewed the published literature. The Cochrane library, Cinahl, and MedLine were queried for articles published between 1980 and 2003. Studies were scored according to Cochrane methodology. Where evidence did not exist or was of a low level, consensus was derived from expert opinion.
The topic was divided into subheadings: decision making, family coping, staff stress related to family interactions, cultural support, spiritual/religious support, family visitation, family presence on rounds, family presence at resuscitation, family environment of care, and palliative care. Each section was led by one task force member. Each section draft was reviewed by the group and debated until consensus was achieved. The draft document was reviewed by a committee of the Board of Regents of the ACCM. After steering committee approval, the draft was approved by the SCCM Council and was again subjected to peer review by this journal.
More than 300 related studies were reviewed. However, the level of evidence in most cases is at Cochrane level 4 or 5, indicating the need for further research. Forty-three recommendations are presented that include, but are not limited to, endorsement of a shared decision-making model, early and repeated care conferencing to reduce family stress and improve consistency in communication, honoring culturally appropriate requests for truth-telling and informed refusal, spiritual support, staff education and debriefing to minimize the impact of family interactions on staff health, family presence at both rounds and resuscitation, open flexible visitation, way-finding and family-friendly signage, and family support before, during, and after a death.
Prognostic information is important to the family members of incapacitated, critically ill patients, yet little is known about what prognostic information physicians provide. Our objectives were to determine the types of prognostic information provided to families of critically ill patients when making major end-of-life treatment decisions and to identify factors associated with more physician prognostication.
Multiple-center, cross-sectional study.
ICUs of four hospitals.
Thirty-five physicians, 51 patients, and 169 family members.
We audiotaped 51 physician-family conferences in which there were deliberations about major end-of-life treatment decisions at four hospitals in 2000-2002. Conferences were coded to identify the types of prognostic information provided by physicians. We used a mixed-effects regression model to identify factors associated with more prognostication by physicians.
The mean number of prognostic statements per conference was 9.4+/-6.4 (range 0-29). Eighty-six percent of conferences contained discussion of the patient's anticipated functional status or quality of life, compared with 63% in which the chances for survival were discussed (p=.01). There were significantly more statements about prognosis for functional outcomes per conference compared with statements about prognosis for survival (median 4 [interquartile range 2-8] vs. 1 [interquartile range 0-3]; p<.001). Increasing educational level of the family was independently associated with more prognostic statements by physicians (p<.001) as was the degree of physician-family conflict about withdrawing life support (p<.001) and the physician's race being white (p=.009).
Prognostication occurred frequently during physician-family deliberations about whether to forego life support, but physicians did not discuss the patient's prognosis for survival in more than one third of conferences. Less educated families received less information about prognosis. Future studies should address whether these observations partially explain the high prevalence of family misunderstandings about prognosis in intensive care units.
The performance of two binary diagnostic tests is traditionally compared by their respective sensitivities and specificities. Other measures to describe the performance of a binary diagnostic test are likelihood ratios, defined as the ratio between the likelihood of a diagnostic test result in a group of diseased patients and the likelihood of a diagnostic test result in a group of non-diseased patients. In this study, we propose a method, based on the log-transformation of the ratio of the likelihood ratios, to compare the likelihood ratios of two binary diagnostic tests in paired designs. We have deduced hypothesis tests to compare the likelihood ratios and we have carried out simulation experiments to study the power and the type I error of the hypothesis tests deduced. We have also deduced a joint hypothesis test to simultaneously compare the likelihood ratios. The procedure used has been extended to the situation in which more than two binary diagnostic tests are applied to the same sample, and the situation in which two diagnostic tests with multilevel results are compared.
Improving ICU family meetings: do the experts agree with the evidence?
- S Ahluwalia
- R A Mularski
- J Lendon
Ahluwalia S, Mularski RA, Lendon J, et al.
Improving ICU family meetings: do the experts
agree with the evidence? Am J Respir Crit Care Med.
2015;191:A1005.
Level of Care Study Investigators and the Canadian Critical Care Trials Group. Withdrawal of mechanical ventilation in anticipation of death in the intensive care unit
- D Cook
- G Rocker
- J Marshall
Cook D, Rocker G, Marshall J, et al; Level of Care
Study Investigators and the Canadian Critical Care
Trials Group. Withdrawal of mechanical ventilation
in anticipation of death in the intensive care unit.
N Engl J Med. 2003;349(12):1123-1132.
Intensivist-reported facilitators and barriers to discussing post-discharge outcomes Research Original Investigation Prediction of Survival and Functional Outcomes After ICU Admission with intensive care unit surrogates: a qualitative study
- A E Turnbull
- W E Davis
- D M Needham
- D B White
- M N Eakin
Turnbull AE, Davis WE, Needham DM, White
DB, Eakin MN. Intensivist-reported facilitators and
barriers to discussing post-discharge outcomes
Research Original Investigation
Prediction of Survival and Functional Outcomes After ICU Admission
with intensive care unit surrogates: a qualitative
study. Ann Am Thorac Soc. 2016;13(9):1546-1552.
Society of Critical Care Medicine. Clinical practice guidelines for support of the family in the patient-centered intensive care unit: American College of Critical Care Medicine Task Force
American College of Critical Care Medicine Task
Force 2004-2005, Society of Critical Care
Medicine. Clinical practice guidelines for support of
the family in the patient-centered intensive care
unit: American College of Critical Care Medicine
Task Force 2004-2005. Crit Care Med. 2007;35(2):
605-622.
- R H Fletcher
Fletcher RH, ed. Clinical Epidemiology:
The Essentials. Philadelphia, PA: Wolters Kluwer
Health/Lippincott Williams & Wilkins; 2014.
III: how to use an article about a diagnostic test, B: what are the results and will they help me in caring for my patients?
Evidence-Based Medicine Working Group. Users'
Guides to the Medical Literature, III: how to use an
article about a diagnostic test, B: what are the
results and will they help me in caring for my
patients? JAMA. 1994;271(9):703-707.
24. Nofuentes JA, Del Castillo JdeD. Comparison of
the likelihood ratios of two binary diagnostic tests
in paired designs. Stat Med. 2007;26(22):4179-4201.
Canadian Critical Care Trials Group. Functional disability 5 years after acute respiratory distress syndrome
Canadian Critical Care Trials Group. Functional
disability 5 years after acute respiratory distress
syndrome. N Engl J Med. 2011;364(14):1293-1304.
with intensive care unit surrogates: a qualitative study
with intensive care unit surrogates: a qualitative
study. Ann Am Thorac Soc. 2016;13(9):1546-1552.