The Impact of Implementing an Intensivist Model With Nighttime In-Hospital Nocturnist and Effect on ICU Outcomes.

BACKGROUND: Various intensivist staffing models have been suggested, but the long-term sustainability and outcomes vary and may not be sustained. We examined the impact of implementing a high-intensity intensivist coverage model with a nighttime in-house nocturnist (non-intensivist) and its effect on intensive care unit (ICU) outcomes. METHODS: We obtained historical control baseline data from 2007 to 2011 and compared the same data from 2011 to 2015. The Acute Physiological and Chronic Health Evaluation outcomes system was utilized to collect clinical, physiological, and outcome data on all adult patients in the medical ICU and to provide severity-adjusted outcome predictions. The model consists of a mandatory in-house daytime intensivist service that leads multidisciplinary rounds, and an in-house nighttime coverage is provided by nocturnist (nonintensivists) with current procedural skills in airways management, vascular access, and commitment to supervise house staff as needed. The intensivist continues to be available remotely at nighttime for house staff and consultation with the nocturnist. A backup intensivist is available for surge management. RESULTS: First year yielded improved throughput (2428 patients/year to 2627 then 2724 at fifth year). Case mix stable at 53.7 versus 55.2. The ICU length of stay decreased from 4.7 days (predicted 4.25 days) to 3.8 days (4.15) in first year; second year: 3.63 days (4.29 days); third year: 3.24 days (4.37), fourth year: 3.34 days (4.45), and fifth year: 3.61 days (4.42). Intensive care unit <24 hours readmission remained at 1%; >24 hours increased from 4% to 6%. Low-risk monitoring admissions remained at an average 17% (benchmark 17.18%). Intensive care unit mortality improved with standardized mortality ration averaging at 0.84. Resident satisfaction surveys improved. CONCLUSIONS: Implementing an intensivist service with nighttime nocturnist staffing in a high-intensity large teaching hospital is feasible and improved ICU outcomes in a sustained manner that persisted after the initial implementation phase. The model resulted in reduced and sustained observed-to-predicted length of ICU stay.

Do elderly patients fare well in the ICU?

BACKGROUND: A recent update of the Mortality Probability Model (MPM)-III found 14% of intensive care patients had age as their only MPM risk factor for hospital mortality. This subgroup had a low mortality rate (2% vs 14% overall), and pronounced differences were noted among elderly patients. This article is an expanded analysis of age-related mortality rates in patients in the ICU. METHODS: Project IMPACT data from 135 ICUs for 124,885 patients treated from 2001 to 2004 were analyzed. Patients were stratified as elective surgical, emergency/unscheduled surgical, and medical and then further stratified by age and whether additional MPM risk factors were present or absent. RESULTS: Mortality rose with advancing age within all patient categories. Elective surgical patients without other risk factors were the least likely to die at all ages (0.4% for patients aged 18-29 years to 9.2% for patients aged ≥ 90 years), whereas medical patients with one or more additional risk factors had the highest mortality rate (12.1% for patients aged 18-29 years to 36.0% for patients aged ≥ 90 years). In these two subsets, mortality rates approximately doubled in the elective surgical group among patients aged in their 70s (2.4%), 80s (4.3%), and 90s (9.2%) but rose less dramatically in the medical group (27.0%, 30.7%, and 36.0%, respectively). CONCLUSIONS: Although mortality increased with age, the risk differed significantly by patient subset, even among elderly patients, which may reflect a selection bias. Advanced age alone does not preclude successful surgical and ICU interventions, although the presence of serious comorbidities decreases the likelihood of survival to discharge for all age groups.

Can we identify patients at high risk for unplanned extubation? A large-scale multidisciplinary survey.

BACKGROUND: Unplanned extubation represents a threat to patient safety, and risk factors and prevention strategies for unplanned extubation have not been fully explored. OBJECTIVES: To define high-risk patients for unplanned extubation and determine clinicians' beliefs on perceived risks for unplanned extubation METHODS: With a Web-based survey instrument we surveyed critical care clinician members of the American Association for Respiratory Care, the American Association of Critical Care Nurses, and the Society of Critical Care Medicine. RESULTS: Surveys were completed by 1,976 clinicians, including 419 respiratory therapists, 870 critical care nurses, and 605 critical care physicians. The majority of respondents considered an outward migration of the endotracheal tube (by 3 cm, 2 cm if an air leak is present) to represent a risk for unplanned extubation. Respondents considered the following as high risk for unplanned extubation: absence of physical restraints (72% of respondents), a nurse/patient ratio of 1/3 (60%), trips out of the intensive care unit (59%), light sedation (43%), and bedside portable radiograph (29%). In addition, most respondents considered accidental removal of the nasogastric tube (71%) or tugging on the endotracheal tube (87%) by the patient to be risk factors for unplanned extubation. The rank order of the perceived risks was related to the respondents' primary discipline. CONCLUSIONS: We identified perceived risk factors and defined "near misses" for unplanned extubation. Our findings should inform strategies for prevention of unplanned extubation.

Subgroup mortality probability models: are they necessary for specialized intensive care units?

OBJECTIVES: To examine the sensitivity of the performance of the latest Mortality Probability Model at intensive care unit admission (MPM0-III) to case-mix variations and to determine how specialized models for these subgroups would affect intensive care unit performance assessment. MPM0-III is an important benchmarking tool for intensive care units in Project IMPACT. Overall, MPM0-III has excellent discrimination and calibration but its performance varies on six common patient subsets. DESIGN: A total of 124,171 patients in six subgroups (complex cardiovascular, trauma, elective surgery, medical, neurosurgery, and emergency surgery) were divided randomly into development (60%) and validation (40%) groups. A logistic regression model was developed to predict hospital mortality for each subgroup, using MPM0-III variables. Model performance was evaluated on the validation sets, using Hosmer-Lemeshow and receiver operating characteristic statistics. Intensive care unit standardized mortality ratios, using the subgroup models and MPM0-III, were compared. A sensitivity analysis was used to identify the occurrence of each subgroup associated with degraded MPM0-III performance. SETTING: One hundred thirty-five intensive care units at 98 hospitals participating in Project IMPACT between 2001 and 2004. ICUs with <100 patient records were excluded. PATIENTS: Consecutive intensive care unit patients in the Project IMPACT database who were eligible for MPM0-II scoring. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Hospital mortality and standardized mortality ratio values by intensive care unit. All six subgroup models had good performance on their validation sets. Intensive care unit standardized mortality ratios calculated with MPM0-III and the subgroup models were nearly identical, with MPM0-III identifying 33 of 135 as significant standardized mortality ratio outliers and the subgroup models identifying 35 of 135, with 33 overlapping. Sensitivity analysis indicated that MPM0-III calibration degraded substantially only when patient mix varied significantly from that of the data set on which MPM0-III was based. CONCLUSION: We recommend users of MPM make MPM0-III their primary model. Subgroup models may have utility when evaluating highly specialized intensive care units or in research on specific, homogeneous populations.

Prospective validation of the intensive care unit admission Mortality Probability Model (MPM0-III).

OBJECTIVE: To validate performance characteristics of the intensive care unit (ICU) admission mortality probability model, version III (MPM0-III) on Project IMPACT data submitted in 2004 and 2005. This data set was external from the MPM0-III developmental and internal validation data collected between 2001 and 2004. DESIGN: Retrospective analysis of clinical data collected concurrently with care. SETTING: One hundred three (103) adult ICUs in North America that voluntarily collect and submit data to Project IMPACT. SUBJECTS: A total of 55,459 patients who were eligible for MPM scoring (age >or=18; first ICU admission for hospitalization, excludes burns, coronary care, and cardiac surgical patients). INTERVENTIONS: None. MEASUREMENTS: Prevalence of MPM risk factors and their relationship to hospital mortality; calibration and discrimination of MPM0-III model applied to new data. MAIN RESULTS: Seventy-eight ICUs contributed data to both this study and the original development set. Fifty-six ICUs from the original MPM0-III study were replaced by 25 new ICUs in this external validation set. Patient characteristics (type of patient, risk factors, and resuscitation status) were similar to the original 2001-2004 cohort, except for slightly more patients on mechanical ventilation at admission (32% vs. 27%, p < 0.01) and the percentage of patients having no MPM0-III risk factors except age (11% vs. 14%, p < 0.01). Observed deaths were 7331 (13.2%) vs. 7456 predicted, yielding a standardized mortality ratio of 0.983, 95% CI (0.963-1.001). CONCLUSIONS: MPM0-III calibrates on a new population of 55,459 North American patients who include many patients from new ICUs, which helps confirm that the model is robust and was not overfitted to the development sample. Although Project IMPACT participants change over time, 2004-2005 patient risk factors and their relationship to hospital mortality have not significantly changed. The increase in mechanically ventilated patients and reduction in admissions with no risk factors are trends worth following.

Outcome prediction in critical care: the Mortality Probability Models.

PURPOSE OF REVIEW: The comparison of morbidity, mortality, and length-of-stay outcomes in patients receiving critical care requires adjustment based on their presenting illness. These adjustments are made with severity-of-illness models. These models must be periodically updated to reflect current medical practices. This article will review the history of the Mortality Probability Model (MPM), discuss why and how it was recently updated, and outline examples of MPM use. RECENT FINDINGS: All severity-of-illness models have limitations, especially if a unit's patient population becomes highly specialized. In these situations, customized models may provide better accuracy. The MPMs include those calculated at admission (MPM0) and additional models at 24, 48, and 72 h (MPM 24, MPM 48, and MPM 72). The model is now in its third iteration (MPM 0-III). Length of stay (LOS) and subgroup models have also been developed. SUMMARY: Understanding appropriate application of models such as MPM is important as transparency in healthcare drives demand for severity-adjusted outcomes data.

A revised method to assess intensive care unit clinical performance and resource utilization.

OBJECTIVE: In 1994, Rapoport et al. published a two-dimensional graphical tool for benchmarking intensive care units (ICUs) using a Mortality Probability Model (MPM0-II) to assess clinical performance and a Weighted Hospital Days scale (WHD-94) to assess resource utilization. MPM0-II and WHD-94 do not calibrate on contemporary data, giving users of the graph an inflated assessment of their ICU's performance. MPM0-II was recently updated (MPM0-III) but not the model for predicting resource utilization. The objective was to develop a new WHD model and revised Rapoport-Teres graph. DESIGN: Multicenter cohort study. SETTING: One hundred thirty-five ICUs in 98 hospitals participating in Project IMPACT. PATIENTS: Patients were 124,855 MPM0-II eligible Project IMPACT patients treated between March 2001 and June 2004. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: WHD was redefined as 4 units for the first day of each ICU stay, 2.5 units for each additional ICU day, and 1 unit for each non-ICU day after the first ICU discharge. Stepwise linear regression was used to construct a model to predict ICU-specific log average WHD from 39 candidate variables available in Project IMPACT. The updated WHD model has four independent variables: percent of patients dying in the hospital, percent of unscheduled surgical patients, percent of patients on mechanical ventilation within 1 hr of ICU admission, and percent discharged from the ICU to an external post-acute care facility. The first three variables increase average WHD and the last decreases it. The new model has good performance (R = 0.47) and, when combined with MPM0-II, provides a well-calibrated Rapoport-Teres graph. CONCLUSIONS: A new WHD model has been derived from a large, contemporary critical care database and, when used with MPM0-III, updates a popular method for benchmarking ICUs. Project IMPACT participants will likely perceive a decline in their ICU performance coordinates due to the recalibrated graph and should instead focus on their unit's performance relative to their peers.

Assessing contemporary intensive care unit outcome: an updated Mortality Probability Admission Model (MPM0-III).

OBJECTIVE: To update the Mortality Probability Model at intensive care unit (ICU) admission (MPM0-II) using contemporary data. DESIGN: Retrospective analysis of data from 124,855 patients admitted to 135 ICUs at 98 hospitals participating in Project IMPACT between 2001 and 2004. Independent variables considered were 15 MPM0-II variables, time before ICU admission, and code status. Univariate analysis and multivariate logistic regression were used to identify risk factors associated with hospital mortality. SETTING: One hundred thirty-five ICUs at 98 hospitals. PATIENTS: Patients in the Project IMPACT database eligible for MPM0-II scoring. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Hospital mortality rate in the current data set was 13.8% vs. 20.8% in the MPM0-II cohort. All MPM0-II variables remained associated with mortality. Clinical conditions with high relative risks in MPM0-II also had high relative risks in MPM0-III. Gastrointestinal bleeding is now associated with lower mortality risk. Two factors have been added to MPM0-III: "full code" resuscitation status at ICU admission, and "zero factor" (absence of all MPM0-II risk factors except age). Seven two-way interactions between MPM0-II variables and age were included and reflect the declining marginal contribution of acute and chronic medical conditions to mortality risk with increasing age. Lead time before ICU admission and pre-ICU location influenced individual outcomes but did not improve model discrimination or calibration. MPM0-III calibrates well by graphic comparison of actual vs. expected mortality, overall standardized mortality ratio (1.018; 95% confidence interval, 0.996-1.040) and a low Hosmer-Lemeshow goodness-of-fit statistic (11.62; p = .31). The area under the receiver operating characteristic curve was 0.823. CONCLUSIONS: MPM0-II risk factors remain relevant in predicting ICU outcome, but the 1993 model significantly overpredicts mortality in contemporary practice. With the advantage of a much larger sample size and the addition of new variables and interaction effects, MPM0-III provides more accurate comparisons of actual vs. expected ICU outcomes.

The value and limits of severity adjusted mortality for ICU patients.

Intensive care unit (ICU) data systems serve a variety of valuable functions for ICU directors, quality-of-care managers, safety officers and health service researchers. Although controversial, severity adjusted mortality cross-linked with resource measures may provide additional value when comparisons are made to similar types of ICUs. This article describes several options for improving the standardized mortality ratio. An essential ingredient for fostering wider application of ICU data systems is through automated information technology to assure high quality data input and to minimize the burden of manual data collection effort.

Length of stay data as a guide to hospital economic performance for ICU patients.

CONTEXT: Length of stay data are increasingly used to monitor ICU economic performance. How such material is presented greatly affects its utility. OBJECTIVE: To develop a weighted length of stay index and to estimate expected length of stay. To assess alternative ways to summarize weighted length of stay to evaluate ICU economic performance. DESIGN: Retrospective database study. SUBJECTS: Data for 751 ICU patients in 1998 at two hospitals used to develop weighted length of stay index. Data on 42,237 patients from 72 ICUs used as the basis of economic performance evaluation. MAIN OUTCOME MEASURES: Difference between actual and expected weighted length of stay, where expected weighted length of stay is based on patient clinical characteristics. RESULTS: Length of stay statistically explains approximately 85 to 90% of interpatient variation in hospital costs. The first ICU day is approximately four times as expensive, and other ICU days approximately 2.5 times as expensive, as non-ICU hospital days. In a regression model for weighted length of stay, patient clinical characteristics explain 26% of variation. ICU economic performance can be measured by excess weighted length of stay of a "typical" patient or by occurrence of long excess weighted lengths of stay. Although different summary measures of performance are highly correlated, choice of measure affects relative ranking of some ICUs' performance. CONCLUSION: Providers of statistical data on ICU economic performance should adjust length of stay for patient characteristics and provide multiple summary measures of the statistical distribution, including measures that address both the typical patient and outliers.

Early indicators of prolonged intensive care unit stay: impact of illness severity, physician staffing, and pre-intensive care unit length of stay.

OBJECTIVE: Scoring systems that predict mortality do not necessarily predict prolonged length of stay or costs in the intensive care unit (ICU). Knowledge of characteristics predicting prolonged ICU stay would be helpful, particularly if some factors could be modified. Such factors might include process of care, including active involvement of full-time ICU physicians and length of hospital stay before ICU admission. DESIGN: Demographic data, clinical diagnosis at ICU admission, Simplified Acute Physiology Score, and organizational characteristics were examined by logistic regression for their effect on ICU and hospital length of stay and weighted hospital days (WHD), a proxy for high cost of care. SETTING: A total of 34 ICUs at 27 hospitals participating in Project IMPACT during 1998. PATIENTS: A total of 10,900 critically ill medical, surgical, and trauma patients qualifying for Simplified Acute Physiology Score assessment. INTERVENTIONS: None. RESULTS: Overall, 9.8% of patients had excess WHD, but the percentage varied by diagnosis. Factors predicting high WHD include Simplified Acute Physiology Score survival probability, age of 40 to 80 yrs, presence of infection or mechanical ventilation 24 hrs after admission, male sex, emergency surgery, trauma, presence of critical care fellows, and prolonged pre-ICU hospital stay. Mechanical ventilation at 24 hrs predicts high WHD across diagnostic categories, with a relative risk of between 2.4 and 12.9. Factors protecting against high WHD include do-not-resuscitate order at admission, presence of coma 24 hrs after admission, and active involvement of full-time ICU physicians. CONCLUSIONS: Patients with high WHD, and thus high costs, can be identified early. Severity of illness only partially explains high WHD. Age is less important as a predictor of high WHD than presence of infection or ventilator dependency at 24 hrs. Both long ward stays before ICU admission and lack of full-time ICU physician involvement in care increase the probability of long ICU stays. These latter two factors are potentially modifiable and deserve prospective study.

Effects of severity of illness on resource use by survivors and nonsurvivors of severe sepsis at intensive care unit admission.

OBJECTIVE: To determine the relationship between severity of illness and length of stay for survivors and nonsurvivors of severe sepsis at intensive care unit admission. DESIGN: Observational study. SETTING: Fifty intensive care units participating in Project IMPACT submitted data during 1998-99. Trained personnel followed comprehensive operations to manually perform data collection. PATIENTS: We identified 2,434 patients with severe sepsis at intensive care unit admission by using clinical variables that followed the American College of Chest Physicians/Society of Critical Care Medicine Consensus Panel and evaluated them by using a general intensive care unit severity model customized for sepsis. The analyses included major diagnosis at intensive care unit admission (respiratory, infectious disease, and shock) and comparison to 19,046 patients without severe sepsis at intensive care unit admission. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: Sepsis patients were older, were more severely ill, had a higher mortality rate, and had more intensive care unit readmissions than did nonsepsis patients. Mean length of stay in the intensive care unit and hospital was longer for sepsis patients than for nonsepsis patients. Among sepsis patients, nonsurvivors had slightly longer mean intensive care unit length of stay than survivors (9.06 vs. 8.15 days, p =.03). For sepsis patients in the lower two quartiles of severity, mean intensive care unit length of stay was significantly higher for nonsurvivors than for survivors. Unlike intensive care unit length of stay, mean hospital length of stay was greater for survivors than for nonsurvivors (18.48 vs. 12.22 days, p <.001). In the upper two quartiles of severity, survivors had longer mean hospital stays (p <.001). For nonsurvivors, the sicker patients had shorter stays. CONCLUSIONS: Differences in length of stay between sepsis survivors and nonsurvivors were related to severity of illness. Thus, the potential economic effect of a new therapy for sepsis would depend, in part, on which particular patients, in terms of severity of illness, were enrolled. New therapies targeted to decrease mortality rate in patients with severe sepsis can potentially lead to the overall cost of care being neutral or increased depending on the severity levels of patients included in clinical trials.