Use of the interprofessional Awakening and Breathing Coordination, Delirium Monitoring and Management, and Early Mobility (ABCDE) bundle is recommended practice in intensive care, but its adoption remains limited.
To examine the relationship between intensive care unit provider attitudes regarding the ABCDE bundle and ABCDE bundle adherence.
A 1-time survey of 268 care providers in 10 intensive care units across the country who had worked at least 4 shifts per month to examine their attitudes toward workload burden, difficulty carrying out the bundle, perceived safety, confidence, and perceived strength of evidence. Logistic regression models were used to examine the relationship of unit-level provider attitudes with ABCDE bundle adherence in 101 patients, adjusted for patients’ age, severity of illness, and comorbidity.
For every unit increase in workload burden, adherence to the ABCDE bundle decreased 53% (odds ratio [OR], 0.47; 95% CI, 0.28–0.79; P = .004). Bundle difficulty (OR, 0.29; 95% CI, 0.08–1.07), perceived safety (OR, 0.51; 95% CI, 0.10–2.65), confidence (OR, 0.37, 95% CI, 0.10–1.35), and perceived strength of evidence (OR, 0.69; 95% CI, 0.14–3.35) were not associated with ABCDE bundle adherence. For every unit increase in perceived difficulty carrying out the bundle, adherence with early mobility was reduced 59% (OR, 0.41; 95% CI, 0.19–0.90; P = .03). In addition, ABCDE bundle adherence (ie, ventilator bundle) was less than DE bundle adherence (ie, ventilator-free bundle) (97% vs 72%, z = 5.47; P < .001).
Focusing interventions on workload burden and factors influencing bundle difficulty may facilitate ABCDE bundle adherence.
Intensive care unit (ICU) delirium and ICU-acquired weakness are common and serious public health problems. Duration of ICU delirium has been associated with reduced probability of survival after hospital discharge and long-term cognitive impairment.1–3 Likewise, ICU-acquired weakness is independently associated with postdischarge mortality and reduced physical functioning up to 5 years following critical illness.4,5 Attention is now turning to the long-term outcomes of ICU survivors and the role of critical care therapies on daily life. Interprofessional approaches are a solution for taking the complexity of critical care therapies and bundling them into organized, practical, and traceable procedures.6
Implementation of the interprofessional, evidence-based ABCDE bundle (Awakening and Breathing Coordination, Delirium assessment/management, and Early mobility) has resulted in reduced ventilator, delirium, and hospital days; increased ICU mobilization; and marked financial benefits.7–9 Implementation of the ABCDE bundle is not only endorsed by critical care societies (eg, Society of Critical Care Medicine, American Association of Critical-Care Nurses), but also by national quality improvement agencies (eg, Institute for Healthcare Improvement, Agency for Healthcare Research and Quality, and the Centers for Disease Control and Prevention) as a means to enhance the quality and safety of critical care. Despite endorsements and evidence for effectiveness of the ABCDE bundle, its uptake is limited.10–17 In a recent survey, 212 interprofessional Michigan ICU health care professionals reported that only 12% of them have implemented the ABCDE bundle despite a statewide quality improvement initiative.18
Factors affecting interprofessional ICU protocol implementation and adherence are poorly understood. Review of the literature suggests that provider attitudes (ie, prevailing tendencies and way of thinking) influence protocol implementation and adherence.19–21 For example, a study22 to evaluate factors influencing nurses administration of sedatives in patients receiving mechanical ventilation showed that nurses’ attitudes toward the efficacy of sedation were associated with reports of sedative administration. Nurses’ attitudes toward the mechanical ventilation experience were positively correlated with sedation practices (rs = 0.28; P < .01) and the intention to administer sedatives to all patients receiving mechanical ventilation (rs = 0.58; P < .01).22
The conceptual framework for interprofessional bundle implementation (Figure 1) illustrates earlier studies’ findings that organizational domains (such as policy and protocol factors [ie, accessibility, clarity, complexity], unit milieu [ie, coordination among disciplines], tasks [ie, autonomy and time demands], physical environment [ie, unit layout and access to supplies]) can have a direct influence on provider attitudes. However, the association of ICU provider attitudes with adherence to the ABCDE bundle is unknown. Therefore, the objective of this study was to examine the associations of ICU provider attitudes with ABCDE bundle adherence. Specific provider attitudes include (1) workload burden, (2) bundle difficulty, (3) perceived safety, (4) confidence in carrying out the bundle, and (5) perceived strength of evidence of the ABCDE bundle. We hypothesized that provider attitudes are associated with execution of the ABCDE bundle.20,23
This multicenter, prospective, cohort study was funded by an American Association of Critical-Care Nurses–Sigma Theta Tau International critical care grant. Approval was obtained from the institutional review board at each of the participating centers. Vanderbilt University was the coordinating center for the study. Recruitment was conducted within the medical and surgical ICUs in 6 participating centers: Baystate Medical Center (Springfield, Massachusetts), Vanderbilt University Hospital (Nashville, Tennessee), University Hospital San Antonio (San Antonio, Texas), University of Maryland Medical Center (Baltimore, Maryland), University of Michigan Health System (Ann Arbor, Michigan), and Harborview Medical Center (Seattle, Washington).
A total sample of 268 ICU health care professionals included registered nurses, advanced practice nurses, physical therapists, occupational therapists, respiratory therapists, pharmacists, and physicians providing care to patients (≥4 shifts/month) nested in eligible medical and surgical ICUs practicing the ABCDE bundle in participating hospitals. A waiver of documentation of informed consent was obtained for administration of the anonymous survey to ICU health care professionals. Patients included those with qualifying organ failure (ie, noninvasive ventilation, treatment for shock) enrolled in an ongoing clinical trial with daily tracking of ABCDE bundle adherence.
ABCDE bundle adherence was computed for the entire period of ventilator ICU days.
A 71-item, content-validated electronic ABCDE provider survey was used to collect data on the provider attitudes (content validity index, 0.96; P = .05, α = 0.95).23 The investigator conducted in-person meetings with the leaders of each unit and department at participating hospitals to describe survey distribution requirements. An electronic survey link was forwarded to the targeted ICU health care professionals by unit and departmental leaders. Survey participation was facilitated through the use of unit signage and recruitment postcards. Site-specific methods were employed to reach the target sample while, at the same time, reducing sampling error. Reminders were sent to ICU health care professionals at 4 and 8 weeks to maximize survey response rates.
Daily conduct of the ABCDE bundle was at the discretion of the ICU team and guided by a standardized protocol (Figure 2). The investigators had no role in performing ABCDE bundle components. Adherence was tracked via the ABCDE adherence checklist (Figure 3). The checklist was placed at the patient’s bedside and completed daily (ie, 24-hour calendar day) by the nurses and other health care professionals involved in completing ABCDE bundle components. ABCDE bundle adherence checklists were distributed, collected, and recorded daily by trained study staff. All study data were managed using Research Electronic Data Capture (REDCap) tools hosted at Vanderbilt University by either the study’s principal investigator (L.M.B.) or trained personnel.24
Variables and Measures
Provider attitudes are defined as the internal disposition of health care professionals to adhere to the ABCDE bundle. Provider attitudes were calculated from 10-point visual analog scale responses. Higher scores represented more positive views for all but workload burden and bundle difficulty attitudes. Five analyses of 2 individual items and 3 subscales of provider attitudes were run. The 3 provider attitude subscales, perceived safety of ABCDE bundle implementation (α = 0.73), confidence in performing the ABCDE bundle (α = 0.69), and perceived strength of evidence for the ABCDE bundle (α = 0.86), were used for ease of analysis. Workload burden and bundle difficulty were each analyzed as individual items because of the poor subscale reliability (α = 0.16). Averages for each of the provider attitude subscales (Table 1) and individual items were subsequently calculated and aggregated by unit.
ABCDE bundle adherence was defined as all 5 components (ABCDE bundle) requiring completion during ventilator ICU days and the delirium assessment/management and early mobility components (DE bundle) requiring completion during ventilator-free ICU days, as awakening and breathing trial coordination are not relevant for patients who are not undergoing mechanical ventilation. ABCDE bundle adherence was computed for the entire period of ventilator ICU days as [(days of ABCDE bundle adherence)/(total ventilator ICU days)]. DE bundle adherence was computed for the entire period of ventilator-free ICU days as [(days of DE adherence)/(total ventilator-free ICU days)]. Adherence to individual components on ventilator days (ABCDE bundle) and ventilator-free days (DE bundle) was computed separately using the same equation.
Statistical analyses were conducted using IBM SPSS version 23 (IBM Corp) and STATA, version 14 (StataCorp LLC). Graphical and descriptive statistical methods were used to summarize and evaluate data distributions. Frequency distributions were used to summarize nominal data. Continuous variable distributions for provider attitudes and adherence data were skewed; therefore, median and interquartile ranges were used to summarize those data. Provider attitude data were first aggregated at the unit level. Subsequently, those unit-level provider attitude scores were linked with the patient adherence records in the respective units. Logistic regression models were used to test the effects of unit level provider attitude values on ABCDE bundle adherence and select individual bundle components (ie, coordination and early mobility) while controlling for relevant patient characteristics (ie, age, Charlson Comorbidity Index, Acute Physiology and Chronic Health Evaluation [APACHE] II score, ventilator status). There were not enough cases of nonadherence to evaluate logistic regression in the remaining individual bundle components. To maintain the statistical power and variability in ABCDE adherence among assessments for the same patient, the standard errors in each model were adjusted for clustering of patients’ data.25 Statistical significance was determined as a P value less than .05.
A total of 101 patients were enrolled in the study: 70 enrolled in medical units (median, 11; range, 3–22) and 31 enrolled in surgical units (median, 10; range, 2–13). Patients were mostly white (88%) and male (58%), with a mean age of 54.6 years (Table 2). Patients were admitted to the ICUs for several different medical and surgical reasons, with the highest percentage admitted for management of sepsis and/or septic shock (38%). APACHE II scores for the sample population (mean, 27.7; SD, 9.3) indicated a high severity of illness, yet comorbidity was minimal per the Charlson Comorbidity Index (median, 1.5; IQR, 0–2.3).26,27
Provider Characteristics and Attitudes
The survey response rate was 25%. Surveys were excluded for the following reasons: wrong unit (n = 9) and lack of provider attitude data for analysis (n = 106). A total of 268 surveys were included in the analysis. The majority of participants were nurses (49%, n = 131) and physicians (24%, n = 64). The remainder of the sample included nurse practitioners (n = 6), occupational therapists (n = 7), pharmacists (n = 10), physical therapists (n = 20), and respiratory therapists (n = 30).
Participants reported a high perceived strength of evidence (median, 9.4; IQR, 8.3–9.9) for the ABCDE bundle. Most participants reported feeling confident (median, 8.6; IQR, 7.0–9.5) with ABCDE bundle implementation with a moderate perceived level of safety (median, 8.75; IQR, 7.7–9.6). Workload burden associated with the bundle was neutral (median, 5.2; IQR, 2.8–7.0), and participants tended to disagree with having difficulty carrying out the bundle (median, 4.0; IQR, 2.0–5.9).
ABCDE (Ventilator Days) and DE (Ventilator-Free Days) Bundle Adherence
Adherence was measured for 101 patients on a total of 752 ICU days (Figure 4). Variation in ABCDE bundle adherence (on ventilator days) was noted across units (range, 38%–85%). DE bundle adherence (on ventilator-free days) was less variable among units (range, 86%–100%). Overall bundle adherence was greater on ventilator-free (DE bundle) days than on ventilator (ABCDE bundle) days across all units (97% vs 72%, z = 5.47; P < .001). Overall ABCDE bundle adherence was lower in surgical units than in medical units, but that difference was not statistically significant (63% vs 75%, z = 1.89; P = .06). When bundle components were evaluated individually (Table 3), coordination (ie, breathing trial preceded by awakening trial, 89%) and early mobility (86%) had the lowest levels of adherence for patients receiving mechanical ventilation. Reasons for ABCDE bundle components not being completed are recorded in Table 4. The most common reason for awakening trials not being completed was respiratory instability (33.3%). The most common reason for breathing trials not being completed was positive end-expiratory pressure greater than 7.5 cm H2O (36.9%). The most common reason for early mobility not being completed was a score of -4 or -5 on the Richmond Agitation-Sedation Scale (23.8%). Reasons for not completing coordination and delirium assessment/management were not tracked.
Associations of Provider Attitudes and ABCDE Bundle Adherence
Results of the logistic regression analyses are presented in Table 5. After patients’ characteristics (ie, age, Charlson comorbidity index, APACHE II score, ventilator status) were controlled for, the likelihood of adherence to the ABCDE bundle decreased 53% for every unit increase in the response to workload burden (OR, 0.47; 95% CI, 0.28–0.79; P = .004). Provider attitudes of bundle difficulty, perceived safety, confidence, and perceived strength of evidence were not significantly associated with ABCDE bundle adherence.
Upon evaluation of individual bundle components, for every unit increase in the response to bundle difficulty, the likelihood of early mobility adherence decreased 59% (OR, 0.41; 95% CI, 0.19–0.90; P = .03). Provider attitudes of bundle difficulty, perceived safety, confidence, and perceived strength of evidence were not significantly associated with coordination adherence. Variation in awakening trial, breathing trial, and delirium assessment/management adherence was minimal; thus, odds ratios could not be calculated for these components.
Although previous investigations associated nurses’ attitudes with sedation practices, no one has linked interprofessional provider attitudes with ABCDE bundle adherence.22 A multisite, multidisciplinary study of ICU health care professionals was conducted to investigate whether provider attitudes are associated with ABCDE bundle adherence. We demonstrated statistically significant relationships between provider attitudes and ABCDE bundle adherence (Table 5). After select patient characteristics were adjusted for, the odds of ABCDE bundle adherence were 53% less with perceptions of high workload burden than with low workload burden. Additionally, adherence to early mobility was 59% less likely when reported difficulty with carrying out the bundle was high. Therefore, focusing interventions on reducing workload burden and simplifying task implementation may facilitate ABCDE bundle adherence.
Overall, adherence to the ABCDE bundle was 72% on ventilator days, when all bundle components are required, and 97% on ventilator-free days, when only delirium assessment/management and early mobility components are required (P < .001). Upon evaluation of previous work, we were unable to find reports of full ABCDE bundle adherence on ventilator days or DE bundle adherence on ventilator-free days for comparison. Various studies report adherence to individual ABCDE bundle components as follows: 71% to 100% for awakening trials,8,19,28,29 67% to 100% for breathing trials,19,28,29 87% for coordination,29 46% to 92% for delirium assessment/management,8,19,21,28 and 82% for early mobility.21,28 Differing definitions of adherence across studies make comparisons difficult. For example, Balas et al19 reported breathing trial adherence as patients receiving a breathing trial at least once during the ICU stay while Klompas et al29 reported breathing trial adherence as the percentage of days with a breathing trial done when indicated. In spite of that, our individual component adherence results are consistent with the previous reports of ABCDE bundle implementation.
The adherence results found in this investigation support previous findings suggesting that bundle complexity influences adherence.19,20 Those components of the ABCDE bundle that require the most coordination across disciplines (ie, coordination of awakening and breathing trials and early mobility) have the lowest levels of adherence on ventilator days: 89% for coordination of awakening and breathing trials and 86% for early mobility. Awakening trial (97%), breathing trial (96%), and delirium (100%) bundle components, which are essentially single-discipline activities, had higher rates of adherence on ventilator days. This finding is further evidenced by increased adherence to early mobility (98%) on ventilator-free days when mobilization may not necessitate the presence of a respiratory, physical, or occupational therapist for execution.
Understanding the particular provider attitudes associated with ABCDE bundle adherence provides a basis for devising interventions to improve implementation. Guided by the conceptual framework, we suggest intervening in those organizational domains that influence perceived ease of completion (ie, workload burden and difficulty carrying out the bundle), which in turn may lead to improved adherence to the overall ABCDE bundle and those bundle components that require the most coordination across disciplines (ie, awakening and breathing trial coordination and early mobility). We had found that policy and protocol factors, unit milieu, and access to supplies and equipment are organizational domains most closely associated with difficulty carrying out the bundle.23 Thus, for example, access to supplies and equipment can be improved by keeping necessary ABCDE bundle supplies (eg, ambulatory bag, ventilator extension tubing, gait belt) and equipment (eg, walker, high-back chair, oxygen tank, lift) in the patient’s room and/or geographically convenient supply rooms so as to maximize nurses’ efficiency of movement. Specific strategies that target both policy and protocol factors as well as unit milieu include the development of standardized protocols (eg, checklists, daily goal sheets), structured rounding processes (eg, interprofessional rounds), and interprofessional training (eg, simulation training, core competencies).30
We hypothesized that provider attitudes regarding perceived safety, confidence, and perceived strength of evidence would be associated with ABCDE bundle adherence, but the findings were not statistically significant. The internal consistency of both our perceived safety (α = 0.73) and confidence (α = 0.69) subscales may not have been reliable enough to make associations and most likely require further refinement before future attempts to evaluate relationships. Further investigation of perceived safety and confidence with refined subscales is necessary to elucidate whether a relationship with ABCDE adherence is present. Further refinement of the perceived strength of evidence subscale items may also be required to ensure that the appropriate constructs are being captured. A second potential explanation for the null findings is the study’s small sample size. Researchers in studies with larger sample sizes may be more able to identify relationships between provider attitudes and ABCDE bundle adherence.
Strengths of the current study include interprofessional input for provider attitudes regarding the ABCDE bundle and a statistical analysis that allowed us to control for covariates. Still, some limitations must be addressed. First, we applied the ABCDE bundle framework as originally described by Vasilevskis et al.31 At this time, the bundle was described as an evolving framework open to new strategies being included. Since its original publication, the bundle has now developed into the ABCDEF bundle to include family engagement and recommendations from recent guidelines.6,32 Next, ABCDE adherence data were not collected on every patient in the ICU. Bedside providers were encouraged to perform ABCDE bundle components daily; thus, adherence data for this study most likely result in an overestimate of the actual unit adherence. Finally, nonresponse bias may be a concern because of the low survey response rate. The assistance of ICU leaders was solicited for guidance on the best methods to achieve survey response goals, but it is possible that only ICU health care providers with strong opinions for or against the ABCDE bundle participated in the survey.
Focusing interventions on reducing workload burden and simplifying task implementation may facilitate ABCDE bundle adherence.
The ABCDE bundle is recommended practice in critical care, but evidence suggests that utilization is low and implementation varies. In this study, adherence to the ABCDE bundle was influenced by workload burden of the bundle. Secondary analysis demonstrated adherence to early mobility was influenced by perceived difficulty with carrying out the bundle. Focusing on interventions to address workload burden and difficulty with carrying out the bundle may optimize implementation. Consider use of checklists, daily goal sheets, and interprofessional training in addition to evaluating geographic convenience of ABCDE bundle supplies and equipment as interventions to optimize implementation. Future research requires refinement of provider attitude subscales, which will allow us to further investigate relationships with ABCDE bundle adherence using larger sample sizes for ICU providers, patients, and units. A prospective study is indicated to determine if interventions to influence provider attitudes regarding workload and simplicity of task implementation result in improved ABCDE bundle adherence.
Thank you to the MIND-USA and MENDS2 study investigators and coordinators as well as the nurses, pharmacists, physicians, physical therapists, respiratory therapists, and occupational therapists who participated. All were integral to the success of this investigation!
This research project was supported by the AACN-Sigma Theta Tau Critical Care grant (20170) and the Vanderbilt Institute for Clinical and Translational Research (UL1 TR000445 from the National Center for Advancing Translational Sciences/National Institutes of Health [NIH]). Dr Vasilevskis is supported by the NIH (K23AG040157). Drs Ely and Pandharipande are supported by the NIH (AG027472, HL111111, AG035117, AG034257) and the Veterans Affairs Clinical Science Research and Development Service. Dr Ely is supported by the Geriatric Research, Education and Clinical Center.
Now that you’ve read the article, create or contribute to an online discussion on this topic. Visit www.ajcconline.org and click “Submit a response” in either the full-text or PDF view of the article.
To purchase electronic or print reprints, contact American Association of Critical-Care Nurses, 101 Columbia, Aliso Viejo, CA 92656. Phone, (800) 899-1712 or (949) 362-2050 (ext 532); fax, (949) 362-2049; e-mail, firstname.lastname@example.org.