Abstract
Objective
Hospital-level variations in structure and process may affect clinical outcomes in intensive care units (ICUs). We sought to characterize the organizational structure, processes of care, use of protocols and standardized outcomes in a large sample of U.S. ICUs.
Design
We surveyed 69 ICUs about organization, size, volume, staffing, processes of care, use of protocols, and annual ICU mortality.
Setting
ICUs participating in the United States Critical Illness and Injury Trials Group Critical Illness Outcomes Study (USCIITG-CIOS).
Measurements and Main Results
We characterized structure and process variables across ICUs, investigated relationships between these variables and annual ICU mortality, and adjusted for illness severity using APACHE II. Ninety-four ICU directors were invited to participate in the study and 69 ICUs (73%) were enrolled, of which 25 (36%) were medical, 24 were surgical (35%) and 20 (29%) were of mixed type, and 64 (93%) were located in teaching hospitals with a median number of 5 trainees per ICU. Average annual ICU mortality was 10.8%, average APACHE II score was 19.3, 58% were closed units and 41% had a 24-hour in-house intensivist. In multivariable linear regression adjusted for APACHE II and multiple ICU structure and process factors, annual ICU mortality was lower in surgical ICUs than in medical ICUs (5.6% lower, 95% CI 2.4%–8.8%) or mixed ICUs (4.5% lower, 95% CI 0.4%–8.7%). We also found a lower annual ICU mortality among ICUs that had a daily plan of care review (5.8% lower, 95% CI 1.6%–10.0%) and a lower bed-to-nurse ratio (1.8% lower when the ratio decreased from 2:1 to 1.5:1; 95% CI 0.25%–3.4%). In contrast, 24-hour intensivist coverage (p=0.89) and closed ICU status (p=0.16) were not associated with a lower annual ICU mortality.
Conclusions
In a sample of 69 ICUs, a daily plan of care review and a lower bed-to-nurse ratio were both associated with a lower annual ICU mortality. In contrast to 24-hour intensivist staffing, improvement in team communication is a low-cost, process-targeted intervention strategy that may improve clinical outcomes in ICU patients.
Keywords: protocols, structure, process, ICU management, intensivist, ICU administration
INTRODUCTION
Caring for the critically ill is a resource intensive process that requires a specialized clinical team and real-time monitoring to provide life-sustaining interventions. In the United States, each day of intensive care costs on average 3500 US dollars and intensive care units (ICUs) account for 13% of hospital costs, 4% of the national health expenditures and approximately 1% of the GDP (1, 2). Clinical outcomes vary substantially between ICUs (5–8), which can be attributed to heterogeneous patient case-mix and differences in organizational structure and processes of care within an ICU. With the increasing demand and cost of critical care worldwide, it is important to understand the organizational characteristics and process of care variables associated with optimal outcomes and costs.
Recently it has been recognized that both organizational structure (i.e., the conditions under which patient care is provided) and processes of care (i.e., activities that constitute patient care) in an ICU directly influence clinical outcomes and are an important platform for care improvements. Structure-driven factors previously shown to be associated with clinical outcomes include the type of ICU (3, 4), hospital and ICU case volume (5–7), open or closed ICU format (8–10), 24-hour presence of an intensivist (11–13), nurse staffing (14–17) and staff workload (18). Several of these characteristics, are resource-intensive (e.g., 24-hour intensivist staffing or higher ratio nurse staffing) or immutable (e.g., case volume or type of ICU). Processes of care frequently involve disease-driven interventions, such as lung protective ventilation and conservative fluid management strategies for Acute Respiratory Distress Syndrome (19, 20), and early goal-directed therapy for sepsis (21). Process-driven interventions offer a potentially cost-effective alternative to reduce practice variation and improve clinical outcomes across ICUs. Previous investigations in single-center studies have demonstrated that the use of a standardized protocol vs. physician-directed assessment (i.e., usual care) of spontaneous breathing trials alone (22) or in combination with a spontaneous awakening trial (23) resulted in better patient outcomes. Similarly, another investigation in a single center-study found that use of a daily plan of care checklist was associated with 50% decrease in length of stay (24).
In this study, we sought to characterize the heterogeneity in organizational characteristics in a sample of 69 ICUs in the United States, to identify variations in processes of care and use of protocols; and, describe the ecological association between these variables and annual ICU mortality.
MATERIALS AND METHODS
Study design
The United States Critical Illness and Injury Trials Group Critical Illness Outcomes Study (USCIITG-CIOS has been described in detail elsewhere (25). Briefly, CIOS is a large, prospective, observational study of ICU patients. Participating investigators were selected from a range of institutions participating in USCIITG. CIOS was conducted in two stages. In the first stage, participating investigators were asked to complete a standardized questionnaire about physician and nurse staffing, number of beds and case volume of the hospital and study ICU, processes of care including rounding practices and use of protocols, and the most recent annual reported ICU mortality for the study ICU. After recruitment into the study but before enrollment of individual patient data, investigators were asked to first complete this standardized questionnaire regarding ICU structure and process. All investigators were given a manual of operations that provided definitions for each potential answer (See Online Supplement). This standardized questionnaire was also pilot tested at a couple of our facilities and independently validated by our study staff prior to use. Moreover, once completed, each standardized questionnaire was individually reviewed by the Principal Investigator and the responses were discussed with individual investigators over the telephone. We chose to use annual ICU mortality for this study because it is a highly reliable, standardized metric used at most hospitals. With the exception of APACHE II data, which is not a commonly available quality metric collected by hospitals or ICUs, we limited our analyses to data collected in the first stage of CIOS (i.e., structural questionnaire). We obtained the average Acute Physiology and Chronic Health Evaluation II (APACHE II) score for each ICU from individual patient data collected prospectively in approximately 100 patients per ICU. We instructed investigators to enroll all newly admitted patients on randomly chosen days, with 5 to 10 days between enrollments to allow for patient turnover (25). CIOS was approved by the ethics review boards of all participating institutions.
Definitions
We defined an intensivist as a physician who is board-eligible or board-certified in Critical Care Medicine, board-certified in Emergency Medicine and completed a Critical Care fellowship in an accredited program, or board-certified in Medicine, Anesthesiology, Pediatrics or Surgery who completed training prior to the availability of Critical Care Medicine certification and who provides at least six weeks of full-time ICU care annually since 1987. We defined an ICU as: open if any credentialed physician could admit and manage a patient in the unit, and semi-open if any credentialed physician could admit to the unit, but an automatic critical care consult occurred if certain parameters such as expected length of stay are exceeded. The type of ICU was: medical if it managed exclusively medical or cardiac patients; surgical if it managed only surgical, cardiothoracic, trauma, burn or neurosurgical patients; and, mixed if it managed both medical and surgical or neurology patients. We defined a trainee as either a resident or fellow. We categorized case volume as high or low according to the median number of annual ICU admissions, and bed capacity as high or low according to the median number of ICU beds. We considered an ICU as having a daily plan of care review if it either had a checklist for daily goals of care or if the attending physician reviewed the plan of care with the charge nurse. We defined rounds as multidisciplinary if it included at least two other clinical specialties beyond the ICU physician and nurse.
We defined a protocol as a precise and detailed plan for a regimen of therapy, which included a set of guiding rules initiated by a physician order or included as part of standing orders during admission. We grouped protocols into the following categories: pulmonary and ventilator management bundle (lung protective ventilation, weaning and ventilator-associated pneumonia prevention), infection control bundle (sepsis treatment, rapid antibiotic use, stress ulcer prevention, catheter placement, oral hygiene and ventilator-associated pneumonia prevention), nutrition bundle (nutrition and glucose control), thromboembolism management bundle (deep vein thrombosis prevention and venous thromboembolism), delirium management bundle (delirium assessment and delirium treatment), sedation management bundle (daily discontinuation of sedation or other sedation protocol), neuroprotective bundle (stroke treatment, acute brain injury, intracerebral hemorrhage and hypothermia after acute cardiac arrest) and transfusion management bundle (massive transfusion or transfusion restriction). We defined a protocol with a high requirement for a physician order if >75% of the ICUs in the study required a physician order to initiate a protocol.
Biostatistical methods
We summarized structure and process variables stratified by type of ICU. In unadjusted analysis, we compared proportions across types of ICU using Fisher exact tests, and compared continuous variables across types of ICU using one-way analysis of variance as appropriate. We used multivariable linear regression adjusted to determine structure and process factors associated with annual ICU mortality adjusted for the APACHE II at each center. Specifically, the multivariable linear regression model included the following variables: average APACHE II score, ICU type, case volume, bed capacity, 24-hour intensivist coverage, bed-to-nurse ratio, trainee-to-bed ratio, ICU organization (open vs. closed), computerized order entry, daily plan of care review, multidisciplinary rounding, and use of protocols guiding management of electrolytes, mobility, codes, neuroprotection, delirium, and transfusions. We selected these variables a priori based on biological plausibility and previous research. We did not include protocol bundles that were ubiquitously used across ICUs (>97%) such as pulmonary and ventilator management, infection control, nutrition and thromboembolism management protocols. For example, 68 centers (99%) had either the lung protective ventilation, weaning or ventilator-associated pneumonia prevention protocols, i.e., the pulmonary and ventilator management bundle. Due to the high number of covariates included, which can increase the risk of overfitting, we also conducted sensitivity analyses examining the association between each of the structure and process variables and annual ICU mortality, adjusted only for average APACHE II score and ICU type. Although all structure and process data were complete, 11 ICUs were missing average APACHE II scores. We therefore used multiple imputation analysis to perform multivariable linear regression, to a total of 20 imputations. In sensitivity analyses, the use of 50 and 100 imputations did not affect the results (See Online Supplement). Additional exploratory analyses demonstrated reasonable ranges for the imputed values. As a further sensitivity analysis, we conducted a multivariable linear regression in which we excluded the 11 ICUs without average APACHE II scores and found that the point estimates were similar (See Online Supplement). Finally, since a large percentage of centers only contributed data for 1 or 2 ICUs of a different type (the median number of ICUs per hospital center was 1), we assume that ICUs are independent and identically distributed units in our analyses. We conducted all statistical analyses in R (www.r-project.org) and STATA 12 (Stata Corp., College Station, USA).
RESULTS
Hospital characteristics and utilization
We approached 94 ICU directors in the United States and 69 (73%) agreed to participate in the study. All 69 participating investigators completed the structure and process form for their ICUs without any missing data. A total of 25 hospital centers contributed data for 1 ICU only, 10 centers contributed data for 2 ICUs, 5 centers contributed data of 3 ICUs, 1 center contributed data of 4 ICUs and 1 center contributed data of 5 ICUs. The median number of ICUs per hospital center was 1. We summarized hospital characteristics and demographics in Table 1. Briefly, 25 (36%) ICUs were medical, 24 (35%) were surgical and 20 (29%) were of mixed type. Average annual ICU mortality was 10.8% (median 9%); average APACHE II score was 19.3 (median 19.1); 64% of hospitals were private non-profit and 18% were non-federal public. 80% were located in urban centers; 93% had a resident training program and 83% had a critical care training program; 90% had electronic patient records and 81% had computerized patient order entry.
Table 1.
Hospital characteristics and utilization by Intensive Care Unit (ICU) type.
| All (n=69) | Medical (n=25) | Surgical (n=24) | Mixed (n=20) | P | |
|---|---|---|---|---|---|
| Annual ICU mortality, median (range) | 9.0 (1.5 – 38.6) | 14.8 (6.2 – 38.6) | 5.5 (1.5 – 22.0) | 8.4 (3.3 – 22.0) | <0.001 |
| Mean APACHE II, median (range) | 19.1 (9.7 – 29.2) | 20.2 (13.3 – 28.4) | 17.7 (9.7 – 24.8) | 20.6 (15.6 – 29.2) | 0.02 |
| Type of hospital, n (%) | |||||
| Private non-profit | 44 (64) | 18 (72) | 16 (67) | 10 (50) | 0.14 |
| Non-federal public | 18 (26) | 7 (28) | 6 (25) | 5 (25) | |
| Private for profit | 4 (6) | 0 (0) | 2 (8) | 2 (10) | |
| Federal | 3 (4) | 0 (0) | 0 (0) | 3 (15) | |
| Hospital location | |||||
| Urban (vs. Suburban) | 55 (80) | 22 (88) | 19 (79) | 14 (70) | 0.29 |
| Hospital characteristics, n (%) | |||||
| Teaching hospital | 64 (93) | 25 (100) | 24 (100) | 15 (75) | <0.01 |
| Critical care training program | 57 (83) | 24 (96) | 22 (92) | 11 (55) | <0.001 |
| Rapid response team | 66 (96) | 24 (96) | 22 (92) | 20 (100) | 0.63 |
| Electronic patient record | 62 (90) | 23 (92) | 22 (92) | 18 (90) | 1 |
| Computer physician order entry | 56 (81) | 21 (84) | 21 (88) | 14 (70) | 0.37 |
| Number of Beds, median (range) | |||||
| Adult beds at hospital | 611 (10 – 1300) | 618 (16 – 1300) | 718.5 (10 – 1113) | 364.5 (41 – 1113) | <0.01 |
| Adult ICU beds at hospital | 82 (4 – 184) | 84 (35 – 184) | 99.5 (35 – 180) | 41 (4 – 158) | <0.001 |
| Adult step down beds at hospital | 31 (0 – 539) | 36 (0 – 202) | 48 (0 – 539) | 15 (0 – 202) | 0.04 |
| Pediatric ICU beds at hospital | 48 (0 – 214) | 48 (0 – 122) | 50 (0 – 107) | 27 (0 – 214) | 0.77 |
| Beds in study ICU | 16 (4 – 44) | 16 (8 – 44) | 17.5 (10 – 36) | 19 (4 – 33) | 0.55 |
| Utilization, median (range) | |||||
| Annual ED visits | 65660 (0– 156519) | 75000 (29816 – 156519) | 65660 (29816 – 106000) | 54287.5 (0– 106000) | 0.02 |
| Annual hospital admissions | 34856 (1170 – 56330) | 35000 (19021 – 56330) | 37580 (19021 – 56330) | 23348 (1170 – 56330) | <0.001 |
| Annual study ICU admissions | 1299 (230 – 4556) | 1200 (581 – 4556) | 1119.5 (496 – 2500) | 1313 (230 – 3816) | 0.46 |
Abbreviations: Acute Physiology and Chronic Health Evaluation II (APACHE II), Emergency Department (ED).
Staffing, organization and rounding practices
We summarize staffing, organization and rounding practices in Table 2. Overall, 100% of participating ICUs had attending intensivists; 100% had a medical director and 99% had a nurse manager; 58% were closed units; 41% had 24-hour intensivist coverage. Mean bed-to-nurse ratio was 1.8:1 (median 1.7:1), mean number of critical care trainees was 5.9 (median 5, range 0 to 29), 4% had electronic ICU coverage. Multidisciplinary rounds were performed in 41% of units, but included palliative care in only 7% of ICUs. Nurses performed delirium assessment while charting in 67% of ICUs, and 87% conducted a daily plan of care review.
Table 2.
Staffing and organization by Intensive Care Unit (ICU) type.
| All (n=69) | Medical (n=25) | Surgical (n=24) | Mixed (n=20) | P | |
|---|---|---|---|---|---|
| ICU staffing | |||||
| Intensivist in ICU (%) | 69 (100%) | 25 (100%) | 24 (100%) | 20 (100%) | 1 |
| 24/7 intensivist (%) | 28 (41%) | 9 (36%) | 11 (46%) | 8 (40%) | 0.78 |
| Leapfrog compliant? | 57 (83%) | 19 (76%) | 22 (92%) | 16 (80%) | 0.37 |
| Median number of ICU fellows (range) | 1 (0 – 10) | 1 (0 – 4) | 1 (0 – 5) | 0.5 (0 – 10) | 0.95 |
| 24/7 ICU fellow | 18 (26%) | 3 (12%) | 11 (46%) | 4 (20%) | 0.03 |
| Median number of ICU residents (range) | 4 (0 – 16) | 6 (0 – 16) | 3.5 (0 – 9) | 3 (0 – 14) | 0.04 |
| Median number of RTs in ICU (range) | 2 (1 – 7) | 2 (1 – 7) | 1 (1 – 5) | 2 (1 – 3) | 0.38 |
| Median number of ICU nurses (range) | 10 (2 – 30) | 8 (4 – 30) | 9.5 (5 – 24) | 10 (2 – 30) | 0.92 |
| Median ratio of beds to nurses (range) | 1.7:1 (0.8:1 – 5.2:1) | 1.7:1 (0.8:1 – 4.4:1) | 1.7:1 (0.8:1 – 5.2:1) | 1.9:1 (0.9:1 – 2.7:1) | 0.70 |
| Median number of PAs (range) | 0 (0 – 4) | 0 (0 – 4) | 0 (0 – 4) | 0 (0 – 1) | 0.18 |
| Median number of NPs (range) | 0 (0 – 12) | 0 (0 – 5) | 0 (0 – 12) | 0 (0 – 3) | 0.38 |
| Charge nurse provides patient care? | 26 (38%) | 9 (36%) | 10 (42%) | 7 (35%) | 0.91 |
| ICU Organization, n (%) | |||||
| Closed units | 40 (58%) | 20 (80%) | 12 (50%) | 8 (40%) | 0.02 |
| Mandatory critical care consult in open or semi open units | 16 (55%) | 1 (20%) | 11 (92%) | 4 (33%) | <0.01 |
| Has medical director | 69 (100%) | 25 (100%) | 24 (100%) | 20 (100%) | 1 |
| Has nurse manager | 68 (99%) | 25 (100%) | 24 (100%) | 19 (95%) | 0.29 |
| Has clinical nurse specialist | 57 (83%) | 20 (80%) | 20 (83%) | 17 (85%) | 0.92 |
| Electronic ICU coverage | 3 (4%) | 2 (8%) | 1 (4%) | 0 (0%) | 0.77 |
| CRRT in ICU | 59 (86%) | 23 (92%) | 22 (92%) | 14 (70%) | 0.09 |
| Sustained Low Efficiency Dialysis | 38 (55%) | 11 (44%) | 13 (54%) | 14 (70%) | 0.21 |
| ICU rounding practices, n (%) | |||||
| Pharmacist on rounds | 63 (91%) | 25 (100%) | 21 (88%) | 17 (85%) | 0.10 |
| Respiratory therapist on rounds | 41 (59%) | 13 (52%) | 13 (54%) | 15 (75%) | 0.24 |
| Physical therapist on rounds | 7 (10%) | 1 (4%) | 1 (4%) | 5 (25%) | 0.04 |
| Social worker on rounds | 18 (26%) | 6 (24%) | 1 (4%) | 11 (55%) | <0.001 |
| Nutritionist on rounds | 28 (41%) | 5 (20%) | 9 (38%) | 14 (70%) | <0.01 |
| Palliative care on rounds | 5 (7%) | 1 (4%) | 1 (4%) | 3 (15%) | 0.43 |
| Delirium assessment by nursing | 45 (65%) | 20 (80%) | 14 (58%) | 11 (55%) | 0.14 |
| Daily goals of care sheet | 47 (68%) | 17 (68%) | 19 (79%) | 11 (55%) | 0.24 |
| Daily meeting between physician and charge nurse | 49 (71%) | 17 (68%) | 17 (71%) | 15 (75%) | 0.94 |
| Median number of protocols (range) | 19 (5 – 27) | 18 (6 – 27) | 19 (10 – 27) | 18.5 (5 – 27) | 0.26 |
Abbreviations: Respiratory Therapist (RT), Physician Assistant (PA), Nurse Practitioner (NP), Continuous renal replacement therapy (CRRT).
Protocols
The median number of protocols was 19 for all ICUs (Table 2) and 93% of the ICUs had 10 or more protocols in place. We summarized the availability of protocols and need for a physician order to initiate the protocol rules in Table 3. Pulmonary and ventilator management, infection control, nutrition and thromboembolism management protocols were ubiquitous with 97% to 99% coverage across ICUs. Neuroprotective and sedation management protocols were also very common with 84% and 86% coverage, respectively. In contrast, only 36% of ICUs had early mobility protocols, 55% had transfusion management protocols, 57% had acute coronary syndrome protocols, 62% had delirium management protocols and 81% had electrolyte protocols in place. When we examined individual protocols, only 48% used a rapid antibiotic protocol and 48% had a palliative care protocol, and these were consistently low across all types of ICUs. Most protocols had a high requirement (i.e., >75%) for a physician order (Table 3). Examples of specific protocols with a lower requirement for a physician order included such as catheter placement, ventilator associated pneumonia prevention, oral hygiene, daily interruption of sedation, both delirium assessment and treatment and Advanced Cardiovascular Life Support/Critical care code.
Table 3.
Types of protocols available and requirement of physician order by Intensive Care unit (ICU) type, n (%).
| Available | Physician Order required | |||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Protocol | All (n=69) |
Medical (n=25) |
Surgical (n=24) |
Mixed (n=20) |
P | All (n=69) |
Medical (n=25) |
Surgical (n=24) |
Mixed (n=20) |
P |
| Ventilator Management | ||||||||||
| Lung protective ventilation | 55 (80) | 18 (72) | 19 (79) | 18 (90) | 0.34 | 52 (95) | 18 (100) | 19 (100) | 15 (83) | 0.06 |
| Ventilator Weaning | 61 (88) | 23 (92) | 21 (88) | 17 (85) | 0.81 | 43 (70) | 15 (65) | 16 (76) | 12 (71) | 0.73 |
| Infection Control | ||||||||||
| Sepsis treatment | 46 (67) | 18 (72) | 16 (67) | 12 (60) | 0.73 | 42 (91) | 17 (94) | 14 (88) | 11 (92) | 0.82 |
| Rapid antibiotics | 33 (48) | 12 (48) | 12 (50) | 9 (45) | 0.95 | 30 (91) | 11 (92) | 10 (83) | 9 (100) | 0.76 |
| Stress ulcer prevention | 59 (86) | 20 (80) | 22 (92) | 17 (85) | 0.52 | 54 (92) | 19 (95) | 19 (86) | 16 (94) | 0.61 |
| Catheter placement | 53 (77) | 22 (88) | 21 (88) | 10 (50) | <0.01 | 30 (57) | 14 (64) | 9 (43) | 7 (70) | 0.30 |
| Oral hygiene | 63 (91) | 24 (96) | 23 (96) | 16 (80) | 0.17 | 24 (38) | 8 (33) | 11 (48) | 5 (31) | 0.52 |
| Ventilator associated pneumonia prevention | 66 (96) | 25 (100) | 24 (100) | 17 (85) | 0.02 | 35 (53) | 9 (36) | 14 (58) | 12 (71) | 0.08 |
| Nutrition | ||||||||||
| Nutrition | 46 (67) | 15 (60) | 18 (75) | 13 (65) | 0.53 | 41 (89) | 15 (100) | 16 (89) | 10 (77) | 0.17 |
| Glucose control | 66 (96) | 24 (96) | 24 (100) | 18 (90) | 0.28 | 61 (92) | 23 (96) | 21 (88) | 17 (94) | 0.62 |
| Electrolyte replacement | 56 (81) | 18 (72) | 23 (96) | 15 (75) | 0.06 | 54 (96) | 17 (94) | 22 (96) | 15 (100) | 1 |
| Venous Thromboembolism | ||||||||||
| Deep vein thrombosis prevention | 62 (90) | 22 (88) | 23 (96) | 17 (85) | 0.48 | 58 (94) | 21 (95) | 21 (91) | 16 (94) | 1 |
| Venous thromboembolism | 55 (80) | 17 (68) | 21 (88) | 17 (85) | 0.24 | 55 (100) | 17 (100) | 21 (100) | 17 (100) | 1 |
| Sedation Management | ||||||||||
| Daily interruption of sedation | 53 (77) | 20 (80) | 18 (75) | 15 (75) | 0.88 | 34 (76) | 9 (45) | 12 (67) | 13 (87) | 0.04 |
| Other sedation | 45 (65) | 18 (72) | 17 (71) | 10 (50) | 0.27 | 42 (93) | 17 (94) | 15 (88) | 10 (100) | 0.60 |
| Delirium assessment | 39 (57) | 18 (72) | 14 (58) | 7 (35) | 0.05 | 8 (21) | 3 (17) | 3 (21) | 2 (29) | 0.87 |
| Delirium treatment | 25 (36) | 13 (52) | 8 (33) | 4 (20) | 0.08 | 25 (36) | 13 (100) | 8 (100) | 4 (100) | 1 |
| Neuroprotective | ||||||||||
| Acute brain injury | 22 (32) | 5 (20) | 13 (54) | 4 (20) | 0.02 | 21 (95) | 4 (80) | 13 (100) | 4 (100) | 0.41 |
| Stroke treatment | 46 (67) | 15 (60) | 18 (72) | 13 (65) | 0.53 | 46 (100) | 15 (100) | 18 (100) | 13 (100) | 1 |
| Intracerebral hemorrhage | 23 (33) | 7 (28) | 11 (46) | 5 (25) | 0.29 | 23 (100) | 7 (100) | 11 | 5 (100) | 1 |
| Hypothermia after cardiac arrest | 55 (80) | 23 (92) | 18 (75) | 14 (70) | 0.13 | 55 (100) | 23 (100) | 18 (100) | 14 (100) | 1 |
| Early mobility | 25 (36) | 11 (44) | 9 (38) | 5 (25) | 0.42 | 15 (60) | 5 (45) | 5 (56) | 5 (100) | 0.13 |
| Transfusion | ||||||||||
| Transfusion restriction | 27 (39) | 10 (40) | 11 (46) | 6 (30) | 0.58 | 20 (75) | 8 (80) | 6 (55) | 6 (100) | 0.12 |
| Massive transfusion | 30 (43) | 9 (36) | 11 (46) | 10 (50) | 0.65 | 26 (87) | 7 (78) | 9 (82) | 10 (100) | 0.42 |
| Other | ||||||||||
| Palliative care | 33 (48) | 12 (48) | 9 (38) | 12 (60) | 0.35 | 30 (91) | 10 (83) | 8 (89) | 12 (100) | 0.47 |
| Acute Coronary Syndrome | 39 (57) | 13 (52) | 13 (54) | 13 (65) | 0.68 | 39 (100) | 13 (100) | 13 (100) | 13 (100) | 1 |
| Advanced Cardiovascular Life Support/Critical care code | 48 (70) | 16 (36) | 18 (33) | 14 (25) | 0.72 | 26 (54) | 7 (44) | 10 (56) | 9 (64) | 0.51 |
Structure and process factors associated with ICU mortality
In single variable analyses, both APACHE II score and type of ICU were associated with annual ICU mortality as were several structure and process variables (first column of Table 4). Several of these relationships, however, were likely confounded by either severity of illness, type of ICU or other factors, and most of the associations found in single variable analyses became non-significant in multivariable analyses (second and third columns of Table 4). In multivariable linear regression adjusted for average APACHE II score among other organizational characteristics, annual ICU mortality was lower in surgical ICUs than in medical and mixed ICUs (third column of Table 4). In addition, we found that the adjusted annual ICU mortality was lower among ICUs that had a daily plan of care review (5.8% lower than having no daily review, 95% CI 1.6%–10.0%) and a lower bed-to-nurse ratio (1.8% lower when the ratio decreased from 2:1 to 1.5:1; 95% CI 0.25%–3.4%). In contrast, 24-hour intensivist coverage and closed ICU status were not associated with a lower annual ICU mortality. In sensitivity analyses examining structure and process factors in models adjusting only for severity of illness and ICU type, results were similar to those found in the full multivariable model with a few exceptions: the association between bed-to-nurse ratio and mortality was less robust, and associations between electrolyte and mobility protocols and mortality became significant when adjusting only for average APACHE II and ICU type.
Table 4.
Single variable and multivariable linear regression analysis of annual ICU hospital mortality as a function of organizational structure and process factors.
| Single variable analysis | Adjusted for APACHE II score and type of ICU | Multivariable analysis | ||||
|---|---|---|---|---|---|---|
| % difference in annual ICU mortality (95% CI) |
P | % difference in annual ICU mortality (95% CI) |
P | % difference in annual ICU mortality (95% CI) |
P | |
| Severity of illness | ||||||
| APACHE II (per unit score) | 0.5 (0.01 to 1.1) | 0.05 | 0.3 (−0.3 to 0.8) | 0.35 | ||
| Type of ICU | ||||||
| Medical ICU | 3.0 (0.6 to 5.4) | 0.01 | 5.6 (2.4 to 8.8) | <0.001 | ||
| Mixed ICU | 0.8 (−1.2 to 2.8) | 0.44 | 4.6 (0.4 to 8.7) | 0.03 | ||
| Utilization | ||||||
| High case volume | 3.0 (1.5 to 4.6) | <0.001 | −0.6 (−3.4 to 2.2) | 0.65 | −2.0 (−4.9 to 0.9) | 0.17 |
| High bed capacity | 0.6 (−1.3 to 2.4) | 0.54 | −0.2 (−3.2 to 2.7) | 0.88 | 2.0 (−1.6 to 5.6) | 0.26 |
| Staffing | ||||||
| 24-hour intensivist | 5.6 (4.0 to 7.3) | <0.001 | −1.5 (−4.2 to 1.3) | 0.29 | 0.2 (−2.9 to 3.4) | 0.89 |
| Bed-to-nurse ratio (per 1:1 unit increase) | 2.4 (1.8 to 3.1) | <0.001 | 2.1 (−0.3 to 4.6) | 0.09 | 3.7 (0.5 to 6.8) | 0.02 |
| Trainee-to-bed ratio (per 1:1 unit increase) | 2.5 (−1.7 to 6.8) | 0.24 | 2.0 (−5.1 to 9.1) | 0.58 | 2.7 (−3.8 to 9.1) | 0.41 |
| ICU organization | ||||||
| Closed ICU | −2.0 (−3.6 to −0.4) | 0.02 | 2.3 (−0.5 to 5.3) | 0.11 | 2.0 (−0.8 to 4.8) | 0.16 |
| Computerized patient order entry | −4.7 (−6.5 to −2.9) | <0.001 | −0.5 (−4.2 to 3.2) | 0.80 | 0.5 (−3.1 to 4.1) | 0.79 |
| Rounding practices | ||||||
| Daily plan of care review | −9.6 (−11.3 to −7.7) | <0.001 | −6.2 (−10.0 to −2.5) | <0.01 | −5.8 (−10.0 to −1.6) | <0.01 |
| Multidisciplinary rounding | −3.6 (−5.2 to −2.0) | <0.001 | −0.8 (−3.9 to 2.4) | 0.93 | 0.4 (−2.8 to 3.6) | 0.82 |
| Protocols | ||||||
| Electrolyte protocol | −1.5 (−3.6 to 0.6) | 0.15 | −4.8 (−8.2 to −1.3) | <0.01 | −3.0 (−6.7 to 0.7) | 0.11 |
| Mobility protocol | 4.6 (2.9 to 6.3) | <0.001 | 3.3 (0.6 to 6.1) | 0.02 | 1.8 (−1.1 to 4.8) | 0.22 |
| Critical care code | −5.2 (−7.4 to −3.0) | <0.001 | −0.5 (−3.5 to 2.5) | 0.74 | 1.4 (−1.4 to 4.3) | 0.32 |
| Neuroprotective protocol | 9.3 (7.5 to 11.2) | <0.001 | 2.9 (−1.2 to 6.9) | 0.17 | 2.7 (−1.6 to 7.0) | 0.21 |
| Delirium management protocol | 0.3 (−1.3 to 1.9) | 0.35 | 1.2 (−1.8 to 4.2) | 0.42 | 1.8 (−1.3 to 4.9) | 0.26 |
| Transfusion protocol | 2.7 (1.2 to 4.3) | 0.001 | −0.4 (−3.2 to 2.4) | 0.78 | −1.2 (−4.7 to 2.2) | 0.48 |
DISCUSSION
Our study identified substantial heterogeneity in both ICU organizational structure and processes of care across 69 centers in the United States, most of which were located in teaching hospitals. While annual ICU mortality varied by type of ICU, the primary factors that were strongly associated with a lower ICU mortality were improved daily team communication strategies and a lower bed-to-nurse ratio. We did not find that 24-hour attending intensivist staffing (vs. fewer attending hours) or a closed ICU format (vs. open) was associated with a lower annual ICU mortality in the USCIITG-CIOS cohort.
We identified that daily team communication strategies was a process-driven factor associated with a lower ICU mortality in the 69 participating centers, after adjusting for case volume, utilization, severity of illness, type of ICU and other organizational factors. In our study, more than 25% of ICUs reported either not having a checklist for daily goals of care or not having a daily meeting between the attending intensivist and charge nurse to discuss the plan of care. This proportion may be higher in community hospitals; however, we do not have data to substantiate this statement as >92% of participating ICUs were in teaching hospitals. Improved team communication via a daily meeting or checklist is a modifiable process-driven factor that is simple, cost-effective and is associated with improved clinical outcomes (11, 26, 27).
The ratio of beds to nurses was a structure-driven factor that was associated with a lower ICU mortality in multivariable analysis. Our study is in agreement with findings of previous studies and systematic reviews, which identified that a higher number of nursing care hours or a higher relative number of nurses to patients (or beds) was related to improved clinical outcomes (14–17, 28). An increase in nursing staff is costly and may be challenging for some centers, especially those with limited resources (e.g., safety net and critical access hospitals). Moreover,, intensive care nursing requires a high level of qualifications and competencies (29) and is associated with a high rate of burnout, each of which contribute to nursing shortages to adequately cover the growing demand for critical care. Thus, while nurse staffing may be important to achieve optimal clinical outcomes, it is constrained by both financial and workforce limitations.
Our study did not confirm previous findings that the presence of a 24-hour intensivist was associated with improved clinical outcomes (11–13, 30). A large number of the previous studies, however, were single-center based, whereas our study summarizes information across 69 ICUs using a standardized survey. It seems likely that, since the majority of our ICUs were located in teaching hospitals with critical care trainees, this may obscure the impact of 24-hour intensivist staffing. Our findings were consistent with a recent cross-over trial of two intensivist staffing models in Manitoba which found no difference in hospital mortality with greater intensivist staffing (31) and with a recent retrospective study which found that addition of night-time intensivist coverage was not associated with a reduction in mortality in ICUs with high-intensity daytime staffing (32). A closed ICU format was also not associated with a lower annual ICU mortality, and counters previous studies (8–10, 33). One potential explanation for this difference is this ICU cohort was highly managed (100% of ICUs were covered by a medical director and 99% had a nurse manager) and included a large number of protocols, both likely decreasing practice variation across admitting, non-intensivist trained physicians.
Limitations of our study include a modest sample of participating ICUs in the United States, limited representation of community hospitals, use of standardized annual ICU mortality as the outcome, and voluntary participation by ICU directors. We did not collect any information in non-participating ICUs to evaluate the potential risk of a self-selection bias in our sample. Another potential shortcoming is that our findings may not be generalizable to all hospital ICUs, especially to those found in community hospitals. Our findings are based on a sample of hospitals participating in USCITTG, which is primarily composed of leading academic institutions and university hospitals from around the country. Several structure and process variables may be different in our sample of ICUs vs. those in community hospitals: such as the median number of ICU beds (84 beds), hospital locations (80% found in urban settings), a large number of available protocols (median of 19 per ICU), the ubiquitous presence of electronic patient records (90%), and a high number of median number of ICU trainees (5 trainees per ICU). Similar process and structure relationships still need to be studied in community hospitals which are markedly different in terms of staffing characteristics, protocol utilization, patient case-mix and ICU resource availability. Several of these factors may limit the generalizability of our findings. Moreover, our sample size may also affect both the effect size and significance of our findings; however, the association between improved daily team communication or bed-to-nurse ratio and annual ICU mortality were relatively robust in both single variable and multivariable analyses. While we based our analysis on reported annual ICU mortality, we believe that this reported value is quite robust as this is a common statistic measured and utilized by hospital units and it summarizes mortality using a denominator of thousands of patients seen over a one-year period. Thus, the use of annual reported ICU mortality may provide a more precise estimate of the actual mortality in any given ICU. Annual ICU mortality is a standardized metric, however, factors beyond ICU care such as hospital throughput and transfer practices can influence this metric. For example, the Critical Care Societies Collaborative has recommended that ICU mortality, even when risk-adjusted, not be used to evaluate ICU quality or performance. Instead measures such as 30-day mortality or duration on mechanical ventilation can more reliably indicate quality. There are other clinically-important endpoints such as ICU length-of-stay, ventilator-free days, case and severity adjusted cost of care that did not examine in this report but which could also be affected different structural and process of care variables. Although we had complete structure and process data for participating ICUs, we had to use multiple imputation analysis because 11 ICUs (16%) were missing average APACHE II scores. However, in sensitivity analyses, we found that the multiple imputation and case-deletion approaches yielded similar point estimates for the variables under study. We chose to report the results of multiple imputation analysis because it will result in less biased estimates under the assumption of missing at random than a case-deletion approach. The fact that APACHE II was not associated with ICU mortality in our multivariable analysis may be due to sample size, potential unmeasured confounders or lack of APACHE II calibration for contemporary ICU outcomes. It is also possible that the structural and process factors examined in this study were unaccounted in the original APACHE II derivation, and their inclusion in our study mitigated the predictive power of APACHE II. Finally, we did not collect the period of time in which APACHE II score were obtained at each center. Therefore, it is possible that variations in the data period collection may influence the average severity of illness measured across ICUs.
Our study also has significant strengths. We used a standardized questionnaire supported with a manual of operations that standardized the definitions of our structure and process variables, which were then individually checked for consistency by study staff. Second, our study directly obtained comprehensive structure and process information from multiple geographically and organizationally disparate ICUs, and it is one of the first aimed to investigate the effects of multiple process-driven factors on clinical outcomes independent of other organizational factors in the ICU and utilization such as case volume. Third, we also aimed to include a similar number of medical, surgical and mixed ICUs to capture the full spectrum of critical care services.
Increasing evidence points to the importance of organization, structure and process within an ICU in the management of critically ill patients (33, 34–40). Heterogeneity in structure and process can contribute to practice variation and, in turn, affect clinical outcomes across ICUs, and it provides a unique opportunity to implement structure-and-process driven interventions to test whether clinical outcomes can be improved. Our results suggest that heterogeneity in both structure-and-process and type of ICU (medical vs. surgical vs. mixed) may play an important confounding or effect-modifying role in the interpretation of multi-center observational studies in critical care. We recommend that future observational studies that involve multiple ICUs collect both organizational structure and process data for adjustment in multivariable analyses.
In summary, we found that better daily team communication and greater nurse staffing were associated with patient-related outcomes in ICUs; these ICU structure and process factors are amenable to intervention. In particular, improvement in team communication is a low-cost process-targeted intervention that may improve clinical outcomes in ICU patients.
Supplementary Material
Acknowledgments
Drs. Checkley, Brown, and Sevransky received support for article research from the National Institutes of Health.
Dr. Martin’s institution received grant support from FDA, Baxter Healthcare, and Abbott. Dr. Brown’s institution received grant support from NIH (K23GM094465). Drs. Martin and Girard’s institutions received grant support from NIH. Dr. Pastores’ institution received grant support from Spectral Diagnostics (is the principal investigator at MSKCC for an ongoing Phase 3 septic shock clinical trial) and Altor Biosciences (was the principal investigator at MSKCC for a phase 2 sepsis-induced ARDS clinical trial in 2011–2012). Dr. Patil’s institution received research grant support from NIH/NIAID and from Canyon Pharmaceuticals. Dr. Sevransky’s institution received grant support from NIGMS and Abott Labs.
Dr. Martin has board membership with Cumberland Pharmaceuticals and Pulsion Medical Systems (no money paid). Dr. O'Brien serves as Chairman of Board of Sepsis Alliance and is the vice-chair of Quality Improvement Committee of ACCP (neither provide any financial reimbursement).
Dr. Martin consulted for AstraZeneca and Agennix. Dr. Rice consulted for GlaxoSmithSkline, LLC and Asiva Pharma, LLC. Dr. Johnson consulted for Becton-Dickenson and Medimmune.
Dr. Girard lectured for Hospira (honoraria for non-promotional 78 presentation). Dr. Pastores lectured for the American Physician Institute and the International Multiprofessional critical care review course (Korea and Saudi Arabia). Dr. Pastores spoke at the Medical Grand Rounds Conferences for New York Downtown Hospital, Coney Island Hospital, and Pinnacle Healthcare System.
Dr. Girard received support for travel from Hospira. Dr. Patil received support for travel from the FTCTS-Care Foundation.
Dr. Johnson is employed by the American College of Surgeons and Banner Health. Dr. Patil has stock options with Microsoft, GE, Intel, Walmart, and Bank of America.
Sources of funding: William Checkley was supported by a Pathway to Independence Award (R00 HL096955) from the National Heart, Lung and Blood Institute, National Institutes of Health. Jonathan E Sevransky was supported in part by K23 GM071399. Greg S. Martin was supported in part by R01 FD003440,P50 AA013757 and UL1 TR000454.
USIITG-CIOS investigators
ARIZONA: University of Arizona Medical Center, Tucson, AZ, Terence O’Keeffe (PI), Coy Collins; Laurel Rokowski; CALIFORNIA: LA County-University of South California Hospital, Los Angeles, CA, Janice Liebler (PI), Ali Ahoui, Anahita Nersiseyan, Usman Shah, Hidenobu Shigemitsu, Nanditha Thaiyananthan; Stanford University Medical Center, Stanford, CA, Joe Hsu (PI), Lawrence Ho; VA Palo Alto Health Care System, Juliana Barr (PI); CONNECTICUT: Bridgeport Hospital, Bridgeport, CT; David Kaufman (PI) Yale University Hospital, New Haven, CT, Jonathan M. Siner (PI), Mark D. Siegel; GEORGIA: Emory University Hospital, Atlanta, GA, Greg S. Martin (PI), Craig Coopersmith, Micah Fisher,David Gutteridge, Mona Brown, SangLee, Apryl Smith; Emory University Midtown Hospital, Atlanta, GA, Greg S. Martin (PI), Kenneth Leeper, Mona Brown; Grady Memorial Hospital, Atlanta, GA ,Greg S. Martin (PI), Sushma Cribbs, Annette Esper, Mona Brown, David Gutteridge ;Emory University Hospital, Atlanta, GA, Greg S. Martin (PI), Craig Coopersmith, Sushma, Cribbs, Annette Esper, Micah Fisher ,David Gutteridge, Olufunmilayo Dosunmu; KANSAS: VA Medical Ctr., Wichita, Ka, Zubair Hassan(PI) Co-PI: Jing Liu Bart Ridder; ILLINOIS: Northwest Community Hospital, Arlington Heights, IL,Melanie Atkinson (PI), Aimee Draftz, Jackie Durgin, Yelena Rikhman, Jessica Scheckel, Mary Walthers; Saint Francis Hospital, Evanston, IL, Gerald Luger (PI), Carol Downer; University of Illinois Medical Center, Chicago, IL, Ruxana T. Sadikot (PI), Kamran Javaid, Daniel Rodgers, Vibhu Sharma; MARYLAND: Johns Hopkins University, Baltimore, MD, Jon Sevransky (PI), William Checkley, Romer Geocadin, David Murphy, Dale Needham, Adam Sapirstein, Steven Schwartz, Glenn Whitman, Brad Winters, Addisu Workneh, Sammy Zakaria; St. Agnes Hospital, Baltimore, MD, Anthony Martinez (PI), Fran Keith; University of Maryland Medical Center, Baltimore, MD, Steven Johnson (PI), Dan Herr, Giora Netzer Carl Shanholtz, Arabela Sampaio, JenniferTitus; NIH Clinical Center, Bethesda, MD; Michael Eberlein Suburban Hospital Bethesda, Bethesda, MD, Leo Rotello (PI), Jennifer Anderson; MASSACHUSETTS: Beth Israel Deaconess Medical Center, Boston, MA, Sajid Shahul (PI), Valerie Banner-Goodspeed, Michael Howell, Sabina Hunziker, VictoriaNielsen, Jennifer Stevens, Daniel Talmor; Brigham and Women’s Hospital, Boston, MA, Namrata Patil (PI), LisaChin, Michael Myers, Stanthia Ryan; MICHIGAN: St Joseph Mercy Health System, Ann Arbor Michigan Joseph Bander, (PI) University of Michigan Health Systems, Ann Arbor, MI, Pauline Park (PI), James Blum, VivekArora, Kristin Brierley, Jessica DeVito, Julie Harris, Elizabeth Jewell, Deborah Rohner; Kathleen To; MINNESOTA: Mayo Clinic Rochester, Rochester, MN, Brian W. Pickering (PI), Jyothsna Giru, Rahul Kashyap, Naman Trivedi; Mayo Clinic Rochester; MISSOURI: University ofMissouri-Columbia Hospital, Columbia, Missouri; KansasCity VA Hospital, Kansas City, MO, Timothy Dwyer (PI),Kyle Brownback; NEW JERSEY: University of Medicine and Dentistry of New Jersey, Newark, NJ, Steven Chang(PI), Zaza Cohen, Frank Italiano, Zeeshan Kahn, Amee Patrawalla; NEW MEXICO: Presbyterian Healthcare Services, Albequerque, NM, Denise Gonzales (PI), Paul Campbell; NEW YORK: Columbia University Medical Center, New York, NY, David Chong (PI), Matthew Baldwin ,Luke Benvenuto, Natalie Yip; Memorial Sloan Kettering Cancer Center, New York, NY; Steven M Pastores, University of Rochester Medical Center, Rochester, NY, Anthony Pietropaoli(PI), Kathleen Falkner, Timothy Bouck, Ann Marie Mattingly; NORTH CAROLINA: Wake Forest University Health Science, Durham, NC, Peter E. Morris (PI), Lori S. Flores; East Carolina University, Greenville, NC, Abid Butt (PI) , Mark Mazer, Kelly Jernigan; Cone Health, Greensboro,NC, Patrick Wright (PI), Sarah Groce, Jeanette McLean, Arshena Overton; OHIO: Cleveland Clinic, Cleveland, OH, Jorge A. Guzman (PI), Mohammed Abou El Fadl,Tonya Frederick, Gustavo-Cumbo-Nacheli, John Komara; The Ohio State Wexner University Medical Center, Columbus, OH, James M. O’Brien (PI), Naeem Ali, Matthew Exline; PENNSYLVANIA: Eastern Regional Medical Center Cancer Treatment Centers of America, Philadelphia, PA, Jeffrey Hoag (PI), Daniela Albu, Pat McLaughlin; Hahnemann University Hospital, Philadelphia, PA Jeffrey Hoag (PI); Emil Abramian, John Zeibeq.; Hospital of the University of Pennsylvania, Philadelphia, PA, Meeta Prasad (PI), Scott Zuick; TENNESSEE: Meharry Medical College Hospital, Nashville, TN, Richard D. Fremont (PI), Chinenye O. Emuwa , Victor C. Nwazue, Olufemi S. Owolabi; Vanderbilt University Medical Center, Nashville, TN, Bryan Cotton (PI), George Hart, Judy Jenkins; Vanderbilt University Medical Center, Nashville, TN, Todd W. Rice (PI), Timothy D. Girard, Margaret Hays, Susan Mogan; TEXAS: University of Texas-Houston Medical Center, Houston, TX; Imo P. Aisiku (PI) UTAH: Intermountain Medical Center, Murray, Utah, Samuel Brown (PI), Colin Grissom, Russ Miller III, Anita Austin, Heather Gallo, Naresh Kumar, David Murphy; VIRGINIA: Inova Health Systems, Falls Church, VA, Maryann Putman (PI), Joanne Ondrush.
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
Disclosures:
The remaining authors have disclosed that they do not have any potential conflicts of interest.
REFERENCES
- 1.Halpern NA, Pastores SM. Critical care medicine in the United States 2000–2005: an analysis of bed numbers, occupancy rates, payer mix, and costs. Crit Care Med. 2010;38(1):65–71. doi: 10.1097/CCM.0b013e3181b090d0. [DOI] [PubMed] [Google Scholar]
- 2.Chalfin DB, Cohen IL, Lanken PN. The economics and cost-effectiveness of critical care medicine. Intensive Care Med. 1995;21:952–961. doi: 10.1007/BF01712339. [DOI] [PubMed] [Google Scholar]
- 3.Halpern NA, Wang JK, Alicea M, Greenstein R. Critical care medicine: observations from the Department of Veterans Affairs' intensive care units. Crit Care Med. 1994;22:2008–2012. [PubMed] [Google Scholar]
- 4.Esteban A, Anzueto A, Frutos F, Alía I, Brochard L, Stewart TE, Benito S, Epstein SK, Apezteguía C, Nightingale P, Arroliga AC, Tobin MJ Mechanical Ventilation International Study Group. Characteristics and outcomes in adult patients receiving mechanical ventilation: a 28-day international study. JAMA. 2002;287:345–355. doi: 10.1001/jama.287.3.345. [DOI] [PubMed] [Google Scholar]
- 5.Halm EA, Lee C, Chassin MR. Is volume related to outcome in health care? A systematic review and methodologic critique of the literature. Ann Intern Med. 2002;137:511–520. doi: 10.7326/0003-4819-137-6-200209170-00012. [DOI] [PubMed] [Google Scholar]
- 6.Durairaj L, Torner JC, Chrischilles EA, Vaughan Sarrazin MS, Yankey J, Rosenthal GE. Hospital volume-outcome relationships among medical admissions to ICUs. Chest. 2005;128:1682–1689. doi: 10.1378/chest.128.3.1682. [DOI] [PubMed] [Google Scholar]
- 7.Kahn JM, Goss CH, Heagerty PJ, Kramer AA, O'Brien CR, Rubenfeld GD. Hospital volume and the outcomes of mechanical ventilation. N Engl J Med. 2006;355:41–50. doi: 10.1056/NEJMsa053993. [DOI] [PubMed] [Google Scholar]
- 8.Multz AS, Chalfin DB, Samson IM, Dantzker DR, Fein AM, Steinberg HN, Niederman MS, Scharf SM. A "closed" medical intensive care unit (MICU) improves resource utilization when compared with an "open" MICU. Am J Respir Crit Care Med. 1998;157:1468–1473. doi: 10.1164/ajrccm.157.5.9708039. [DOI] [PubMed] [Google Scholar]
- 9.Ghorra S, Reinert SE, Cioffi W, Buczko G, Simms HH. Analysis of the effect of conversion from open to closed surgical intensive care unit. Ann Surg. 1999;229:163–171. doi: 10.1097/00000658-199902000-00001. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Baldock G, Foley P, Brett S. The impact of organizational change on outcome in an intensive care unit in the United Kingdom. Intensive Care Med. 2001;27:865–872. doi: 10.1007/s001340100938. [DOI] [PubMed] [Google Scholar]
- 11.Pronovost PJ, Angus DC, Dorman T, Robinson KA, Dremsizov TT, Young TL. Physician staffing patterns and clinical outcomes in critically ill patients: a systematic review. JAMA. 2002;288:2151–2162. doi: 10.1001/jama.288.17.2151. [DOI] [PubMed] [Google Scholar]
- 12.Gajic O, Afessa B, Hanson AC, Krpata T, Yilmaz M, Mohamed SF, Rabatin JT, Evenson LK, Aksamit TR, Peters SG, Hubmayr RD, Wylam ME. Effect of 24-hour mandatory versus on-demand critical care specialist presence on quality of care and family and provider satisfaction in the intensive care unit of a teaching hospital. Crit Care Med. 2008;36:36–44. doi: 10.1097/01.CCM.0000297887.84347.85. [DOI] [PubMed] [Google Scholar]
- 13.Banerjee R, Naessens JM, Seferian EG, Gajic O, Moriarty JP, Johnson MG, Meltzer DO. Economic implications of nighttime attending intensivist coverage in a medical intensive care unit. Crit Care Med. 2011;39:1257–1262. doi: 10.1097/CCM.0b013e31820ee1df. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Aiken LH, Clarke SP, Sloane DM, Sochalski J, Silber JH. Hospital nurse staffing and patient mortality, nurse burnout, and job dissatisfaction. JAMA. 2002;288:1987–1993. doi: 10.1001/jama.288.16.1987. [DOI] [PubMed] [Google Scholar]
- 15.Kane RL, Shamliyan TA, Mueller C, Duval S, Wilt TJ. The association of registered nurse staffing levels and patient outcomes: systematic review and meta-analysis. Med Care. 2007;45:1195–1204. doi: 10.1097/MLR.0b013e3181468ca3. [DOI] [PubMed] [Google Scholar]
- 16.Penoyer DA. Nurse staffing and patient outcomes in critical care: a concise review. Crit Care Med. 2010;38:1521–1528. doi: 10.1097/CCM.0b013e3181e47888. [DOI] [PubMed] [Google Scholar]
- 17.Needleman J, Buerhaus P, Pankratz VS, Leibson CL, Stevens SR, Harris M. Nurse staffing and inpatient hospital mortality. N Engl J Med. 2011 Mar 17;364(11):1037–1045. doi: 10.1056/NEJMsa1001025. [DOI] [PubMed] [Google Scholar]
- 18.Tarnow-Mordi WO, Hau C, Warden A, Shearer AJ. Hospital mortality in relation to staff workload: a 4-year study in an adult intensive-care unit. Lancet. 2000;356:185–189. doi: 10.1016/s0140-6736(00)02478-8. [DOI] [PubMed] [Google Scholar]
- 19.The ARDS Network. Ventilation with lower tidal volumes as compared with traditional tidal volumes for acute lung injury and the ARDS. N Engl J Med. 2000;342:1301–1308. doi: 10.1056/NEJM200005043421801. [DOI] [PubMed] [Google Scholar]
- 20.National Heart, Lung, and Blood Institute Acute Respiratory Distress Syndrome (ARDS) Clinical Trials Network. Wiedemann HP, Wheeler AP, Bernard GR, Thompson BT, Hayden D, deBoisblanc B, Connors AF, Jr, Hite RD, Harabin AL. Comparison of two fluid-management strategies in acute lung injury. N Engl J Med. 2006;354:2564–2575. doi: 10.1056/NEJMoa062200. [DOI] [PubMed] [Google Scholar]
- 21.Rivers E, Nguyen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M Early Goal-Directed Therapy Collaborative Group. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med. 2001;345:1368–1377. doi: 10.1056/NEJMoa010307. [DOI] [PubMed] [Google Scholar]
- 22.Ely EW, Baker AM, Dunagan DP, Burke HL, Smith AC, Kelly PT, Johnson MM, Browder RW, Bowton DL, Haponik EF. Effect on the duration of mechanical ventilation of identifying patients capable of breathing spontaneously. N Engl J Med. 1996;335:1864–1869. doi: 10.1056/NEJM199612193352502. [DOI] [PubMed] [Google Scholar]
- 23.Girard TD, Kress JP, Fuchs BD, Thomason JW, Schweickert WD, Pun BT, Taichman DB, Dunn JG, Pohlman AS, Kinniry PA, Jackson JC, Canonico AE, Light RW, Shintani AK, Thompson JL, Gordon SM, Hall JB, Dittus RS, Bernard GR, Ely EW. Efficacy and safety of a paired sedation and ventilator weaning protocol for mechanically ventilated patients in intensive care (Awakening and Breathing Controlled trial): a randomised controlled trial. Lancet. 2008;371:126–134. doi: 10.1016/S0140-6736(08)60105-1. [DOI] [PubMed] [Google Scholar]
- 24.Pronovost P, Berenholtz S, Dorman T, Lipsett PA, Simmonds T, Haraden C. Improving communication in the ICU using daily goals. J Crit Care. 2003;18:71–75. doi: 10.1053/jcrc.2003.50008. [DOI] [PubMed] [Google Scholar]
- 25.Ali NA, Gutteridge D, Shahul S, Checkley W, Sevransky J, Martin GS. Critical illness outcome study: an observational study of protocols and mortality in intensive care units Open Access. Journal of Clinical Trials. 2011;3:55–65. doi: 10.2147/OAJCT.S24223. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Pronovost P, Needham D, Berenholtz S, Sinopoli D, Chu H, Cosgrove S, Sexton B, Hyzy R, Welsh R, Roth G, Bander J, Kepros J, Goeschel C. An intervention to decrease catheter-related bloodstream infections in the ICU. N Engl J Med. 2006;355:2725–2732. doi: 10.1056/NEJMoa061115. [DOI] [PubMed] [Google Scholar]
- 27.Hales BM, Pronovost PJ. The checklist-a tool for error management and performance improvement. J Crit Care. 2006;21(3):231–235. doi: 10.1016/j.jcrc.2006.06.002. [DOI] [PubMed] [Google Scholar]
- 28.Sasichay-Akkadechanunt T, Scalzi CC, Jawad AF. The relationship between nurse staffing and patient outcomes. J Nurs Adm. 2003;33:478–485. doi: 10.1097/00005110-200309000-00008. [DOI] [PubMed] [Google Scholar]
- 29.Aäri RL, Tarja S, Helena LK. Competence in intensive and critical care nursing: a literature review. Intensive Crit Care Nurs. 2008;24:78–89. doi: 10.1016/j.iccn.2007.11.006. [DOI] [PubMed] [Google Scholar]
- 30.Cooke CR, Watkins TR, Kahn JM, Treggiari MM, Caldwell E, Hudson LD, Rubenfeld GD. The effect of an intensive care unit staffing model on tidal volume in patients with acute lung injury. Crit Care. 2008;12(6):R134. doi: 10.1186/cc7105. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Garland A, Roberts D, Graff L. Twenty-four-hour intensivist presence: a pilot study of effects on intensive care unit patients, families, doctors, and nurses. Am J Respir Crit Care Med. 2012;185:738–743. doi: 10.1164/rccm.201109-1734OC. [DOI] [PubMed] [Google Scholar]
- 32.Wallace DJ, Angus DC, Barnato AE, Kramer AA, Kahn JM. Nighttime intensivist staffing and mortality among critically ill patients. N Engl J Med. 2012;366:2093–2101. doi: 10.1056/NEJMsa1201918. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33.Treggiari MM, Martin DP, Yanez ND, Caldwell E, Hudson LD, Rubenfeld GD. Effect of intensive care unit organizational model and structure on outcomes in patients with acute lung injury. Am J Respir Crit Care Med. 2007;176:685–690. doi: 10.1164/rccm.200701-165OC. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Barnato AE, Kahn JM, Rubenfeld GD, McCauley K, Fontaine D, Frassica JJ, Hubmayr R, Jacobi J, Brower RG, Chalfin D, Sibbald W, Asch DA, Kelley M, Angus DC. Prioritizing the organization and management of intensive care services in the United States: the PrOMIS Conference. Crit Care Med. 2007;35:1003–1011. doi: 10.1097/01.CCM.0000259535.06205.B4. [DOI] [PubMed] [Google Scholar]
- 35.Angus DC, Shorr AF, White A, Dremsizov TT, Schmitz RJ, Kelley MA Committee on Manpower for Pulmonary and Critical Care Societies (COMPACCS) Critical care delivery in the United States: distribution of services and compliance with Leapfrog recommendations. Crit Care Med. 2006;34:1016–1024. doi: 10.1097/01.CCM.0000206105.05626.15. [DOI] [PubMed] [Google Scholar]
- 36.Chang SY, Multz AS, Hall JB. Critical care organization. Crit Care Clin. 2005;21:43–53. doi: 10.1016/j.ccc.2004.07.003. [DOI] [PubMed] [Google Scholar]
- 37.Dodek PM, Keenan SP, Norena M, Martin C, Wong H. Structure, process, and outcome of all intensive care units within the province of British Columbia, Canada. J Intensive Care Med. 2010;25:149–155. doi: 10.1177/0885066609359267. [DOI] [PubMed] [Google Scholar]
- 38.Dodek PM, Wong H, Jaswal D, Heyland DK, Cook DJ, Rocker GM, Kutsogiannis DJ, Dale C, Fowler R, Ayas NT. Organizational and safety culture in Canadian intensive care units: relationship to size of intensive care unit and physician management model. J Crit Care. 2012;27:11–17. doi: 10.1016/j.jcrc.2011.07.078. [DOI] [PubMed] [Google Scholar]
- 39.Chang SY, Sevransky J, Martin GS. Protocols in the management of critical illness. Crit Care. 2011;16:306. doi: 10.1186/cc10578. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40.Pronovost PJ, Jenckes MW, Dorman T, Garrett E, Breslow MJ, Rosenfeld BA, Lipsett PA, Bass E. Organizational characteristics of intensive care units related to outcomes of abdominal aortic surgery. JAMA. 1999;281:1310–1317. doi: 10.1001/jama.281.14.1310. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.