- Email: [email protected]
Multiprofessional survey of protocol use in the intensive care unit
Journal of Critical Care (2012) 27, 738.e9–738.e17
Multiprofessional survey of protocol use in the intensive care unit Jaclyn M. LeBlanc PharmD, BCPS a,⁎, Sandra L. Kane-Gill PharmD, MSc, FCCM, FCCP b , Anne S. Pohlman MSN, CCRN c , Daniel L. Herr MD d a
Department of Pharmacy, Saint John Regional Hospital, Saint John, NB E2E 4L2 Pharmacy and Therapeutics, Clinical Translational Sciences and Critical Care Medicine, Schools of Pharmacy and Medicine, University of Pittsburgh and Medication Safety Officer, Department of Pharmacy, UPMC, Pittsburgh, PA c Critical Care Clinical Research, University of Chicago, Chicago, IL d University of Maryland, Baltimore, MD b
Keywords: Critical care; Protocol; Multidisciplinary
Abstract Purpose: To date, there has been no large multicenter, multiprofessional evaluation of protocol and guideline use in the intensive care unit (ICU). The primary purpose of this study was to describe national availability, development, implementation, and assessment of protocols in ICUs. A secondary objective was to compare perceived utility by ease of use, patient safety, cost containment, and compliance of protocols between nurses, physicians, and pharmacists. Materials and Methods: The survey was developed and tested for validity by 15 clinicians who identified additional domains of interest. An additional 15 clinicians of the 3 different professions evaluated the survey for relevancy and appropriateness of responses. Three survey experts evaluated survey construction. The survey was uploaded to a Web survey tool and pilot tested for clarity and ease of completion. Results: The overall response rate for the survey was 18.1% (n = 614). Popular methods of education for protocol implementation included staff meetings (85.3%) and unit-specific in-services (77.7%). Protocols were most often updated when new information was available (40.8%) or every 12 months (17.9%). The most common limitation to development and implementation was limited personnel resources (24.5%) and physicians not wanting to use them (21.3%), respectively. Clinicians indicated that protocols made their job easier and improved cost containment some or most of the time. Sepsis protocols were identified as most useful in promoting patient outcomes by all 3 professions. Conclusions: The types of protocols available appear to be those assisting with management of high-alert medications. Overcoming the perceived barriers of protocol use within ICUs requires personnel for development and physician support. A better protocol review process may be necessary to assure optimal content, desired outcomes, and consistency with Institute for Safe Medication Practices guidelines. © 2012 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Tel: +1 506 649 2875. E-mail address: [email protected] (J.M. LeBlanc). 0883-9441/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2012.07.012
738.e10
1. Background Many hospitals have instituted evidence-based, drugspecific protocols to create uniformity of medication orders to decrease variability among prescribers, promote safe practices, and potentially improve patient outcomes throughout a variety of patient care areas. The intensive care unit (ICU) is an environment where medication errors and adverse drug events are more frequent and severe, so protocols aimed to improve safe medication use are appropriate [1,2]. Protocol use in the ICU has improved patient outcomes, reduced length of patient stays, and reduced costs [3-8]. In fact, the use of standardized order sets is recommended by regulatory bodies such as the Institute for Safe Medication Practices (ISMP) [9]. To date, there has been no large multicenter, multiprofessional evaluation of protocol and guideline use in the ICU. One study did survey ICU physicians about protocols, but it contained limited questions to protocol availability and did not explore development, implementation, and assessment [10]. Obtaining this information is important to understand the impetus for protocol development, how implementation occurs, and the frequency of assessments for updates or adherence. These insights would aid in recommending best practices for protocol use. In addition, an appreciation for barriers to protocol adherence with possible recommendations for improvement would be obtained. The primary purpose of this study was to describe national availability, development, implementation, and assessment of protocols in ICUs. A secondary objective was to compare perceived utility by ease of use, patient safety, cost containment, and compliance of protocols between nurses, physicians, and pharmacists.
2. Methods 2.1. Survey development A thorough approach to survey development was followed similar to a previous publication [11]. The survey was developed by the investigators to obtain information addressing the study's primary and secondary objectives. Questions were derived addressing hospital demographics; protocols available in the respondent's hospital; implementation and developments of these protocols; and perceived usefulness of these protocols with regards to ease of use, patient safety, and cost containment. A literature search was conducted in MEDLINE to develop a comprehensive list of existing protocols. References to more than 70 protocols were identified in the MEDLINE search, and questions were developed pertaining to these. The literature search also revealed confusion surrounding the definition of a protocol; thus, a clear definition was provided in the current survey stating a protocol is as
J.M. LeBlanc et al. follows: “an explicit framework for the process of or algorithmic approach to care whereby members of the health care team (physician, nurse, or pharmacist) can follow precise steps of practice. The word protocol is also being used in place of the word order set, clinical pathway, or bundle”. In addition, a sentence was included to ensure that participants realized that protocol was not meant to refer to those used in the realm of research. The response format consisted of tick boxes and open-ended questions, depending on the type of question.
2.2. Survey testing To test the validity of the survey created by the investigators, 5 critical care physicians, 5 pharmacists, and 5 nurses external to the investigative team were surveyed to identify additional domains of interest. These clinicians were sampled until redundancy. Subsequently, the investigators reviewed all questions and reduced the number to a goal of 30, identifying the relevant questions that met the overall objectives and provided educational information to the audience. An additional 15 clinicians of the 3 different professions evaluated the survey for relevancy and appropriateness of responses. Modifications were made to the survey in response to this review. Lastly, 3 survey experts evaluated survey construction. Further survey modifications were made in response to the expert review. The survey was uploaded to a Web survey tool and pilot tested for clarity and ease of completion by 9 clinicians of various professions.
2.3. Survey administration The study was approved by Society of Critical Care Medicine (SCCM) Pharmacy and Clinical Pharmacology Advisory Board and then reviewed by members of the SCCM internal medicine and nursing sections. After institutional review board approval, the proposal was submitted to the SCCM Research Committee and the SCCM Executive Committee for approval to be sent to SCCM members. The survey invitation was e-mailed to members of the following SCCM sections: Pharmacy and Clinical Pharmacology, Nursing, and Internal Medicine. Reminders were sent via email to the members at 1 and 2.5 months after the initial notification of the survey. The mailings were facilitated by SCCM. A statement within the survey communicated that completion of the questionnaire conveyed the respondents' consent to participate in the study.
2.4. Data analysis All data were exported directly from the Web survey tool and analyzed with SPSS 16.0 (SPSS, Inc, Chicago, Ill). A complete survey was defined as 80% completion and provision of the respondent's profession. The survey was anonymous with a voluntary request for the submission of an
Multiprofessional survey of protocol use in the ICU Table 1
738.e11
Demographics of all participants (n = 551)
Type of hospital (all n = 544, hospital n = 329)
Approximate no. of licensed hospital beds (all n = 546, hospital n = 332)
Type of ICU (all n = 547, hospital n = 332)
Patient population in the ICU (all n = 543, hospital n = 333)
No. of licensed ICU beds (all n = 549, hospital n = 332)
Community nonteaching Community teaching Private for-profit Private not-for-profit University teaching Veterans Affairs/military/government 1-199 200-399 400-599 600-799 800-999 ≥1000 Cardiology (including cardiology, cardiothoracic with or without surgery, coronary care) Medical Medical/Surgical Neonatal/Pediatric Surgical/Trauma Mixed Other Adult Pediatric Both Unknown 5-10 11-20 21-30 31-40 41-50 51-60 N60
institution name with the intention of removing duplicate institutions for subgroup analyses. Descriptive statistics were used to analyze data, in addition to χ2 test of proportions and analysis of variance to compare responses by profession. The Kruskal-Wallis and Bonnferroni post hoc tests were used to analyze differences between professions. A P value of .05 was considered significant.
All participants (%)
Individual hospitals (%)
20.2 36.0 1.5 5.0 33.6 3.7 11.4 30.8 30.2 15.9 7.7 4.0 6.4
21.6 39.2 1.5 4.3 30.4 3.0 11.7 31.3 31.0 14.2 7.5 4.2 5.7
19.7 53.6 1.6 15.2 3.3 0.2 92.0 1.5 4.7 0.4 7.8 47.4 25.1 11.2 3.1 1.5 2.2
17.8 54.2 1.8 17.5 2.7 0.3 92.0 1.8 4.7 1.5 8.1 50.1 25.5 9.3 3.9 0.9 1.2
clinicians (87.5%) in the form of physicians (42.6%), nurses (30.3%), and pharmacists (27.0%). The response rate per profession was 24.8% nursing, 19.9% pharmacist, and 11.9% physician. Of these, 435 provided their hospital name for subanalysis, and those respondents represented 333 different ICUs.
3.2. Demographics
3. Results 3.1. Response rate A total of 3400 survey invitations were e-mailed, with an overall response rate of 614 (18.1%). Of these, 55 were excluded for completion of less than 80% of the survey (n = 52) or not providing their profession (n = 3). Also, 7 respondents indicated their profession as “other” and were thus excluded from the analysis. This left 551 (16.2%) responses for analysis representing predominantly US
The data set represented primarily US practitioners and distribution across different ICU types. The ICU types (n = 542) represented included 28.2% open (any attending physician with hospital admitting privileges can be the physician of record and direct ICU care), 32.8% closed (only an intensivist is the physician of record for all ICU patients), and 38.9% transitional (shared, comanaged care between ICU staff and private physician; an intensivist director, trainees, and intensivist team are present as locally available). Respondents encompassed a wide range of clinical experience indicating that they had been a licensed
738.e12 Table 2
J.M. LeBlanc et al. Protocols used by respondent hospitals (n = 333)
Category
Specific protocols
%
Cardiac
ACS Atrial fibrillation Chest pain (independent of ACS, managing only pain) CHF management Coronary artery bypass grafting postgraft management Hypertension management Hypotension management Therapeutic hypothermia Activated protein C (independent of a sepsis bundle or protocol) Catheter-related blood stream infections Community-acquired pneumonia Febrile neutropenia Fungal infection prevention Meningitis Nosocomial pneumonia (either on or off ventilator) Sepsis (including goal directed therapy) Ventilator-associated pneumonia Heparin Anemia/Hemoglobin management Low-molecular-weight heparin Warfarin Factor VII Heparin-induced thrombocytopenia— argatroban Heparin-induced thrombocytopenia— bivalirudin Heparin-induced thrombocytopenia— lepirudin Thrombocytopenia management Venous thromboembolism prophylaxis Benzodiazepine tapering Delirium (not including sedation) Drug withdrawal Ethanol withdrawal Intracranial pressure management/ intracerebral hemorrhage Sedation Status epilepticus Stroke Sleep disturbance Transient ischemic attack Traumatic brain injury Subarachnoid hemorrhage Spinal cord injury medication management Adrenal insufficiency management Cosyntropin/cotrosyn adrenal stimulation test Diabetic ketoacidosis Potassium
58.3 17.4 35.7
Infectious diseases
Hematology
CNS
Metabolic
Table 2 (continued) Category
38.7 44.4 12.6 15.6 53.8 74.2
Respiratory
34.5
Gastrointestinal
58.0 13.5 5.4 6.0 33.9 68.8 47.7 83.8 16.2 44.4 34.8 18.6 36.9 10.2 18.0
Medications
Specific protocols
%
Phosphate Magnesium Calcium Sodium Combined electrolytes Hyperglycemia—IV insulin Hyperglycemia—SQ insulin Hyperosmolar hyperglycemic nonketotic coma ARDS Bronchodilators Status asthmaticus Bowel regime Gastric motility Hepatic encephalopathy Nutritional support—enteral Nutritional support—parenteral Pancreatitis Stress ulcer prophylaxis Upper GI bleed Aminoglycosides Contrast nephropathy prevention Desensitization for drug allergies Eptifibatide Neuromuscular blockers Opioid taping protocol Pain management Phenytoin Abciximab Vancomycin
40.2 49.2 27.9 19.2 29.1 88.3 57.7 19.5 47.1 27.9 11.1 28.8 21.9 6.3 52.9 49.2 6.0 71.5 12.0 39.0 41.4 16.2 32.4 43.8 6.9 47.7 7.8 19.2 38.4
ACS, acute coronary syndrome; CHF indicates congestive heart failure; CNS, central nervous system; IV, intravenous; SQ, subcutaneous; ARDS, acute respiratory distress syndrome; GI, gastrointestinal.
4.5 69.4 12.3 25.2 14.7 47.7 25.5 69.1 9.6 52.6 1.5 11.7 21.3 19.8 18.0 13.8 24.9 47.1 62.5
clinician for 19.6 ± 10.0 years (range, 2-43 years) and had been working as a clinician in the ICU for 16.6 ± 9.6 years (range, 0-41 years). Other demographics are shown in Table 1. The types of ICUs represented primarily comprised adult medical and surgical patients, with little representation of pediatric or neonatal patients. There was a fairly even distribution of hospital size, with more than 70% of respondents practicing in ICUs with a bed size between 11 and 30 beds.
3.3. Types of protocols available nationally The most common types of protocols are shown in Table 2 for the 333 respondents from different hospitals. Clinicians were also asked about ICU admission order sets, and 72.4% of these hospitals had them. These order sets most often included deep vein thrombosis prophylaxis (67.0%), stress ulcer prophylaxis (66.1%), sedation (49.2%), pain management (46.8%), breathing treatments (41.7%), and bowel regimens (36.9%),
Multiprofessional survey of protocol use in the ICU Table 3
738.e13
Top 5 limitations of development and implementation of protocols
Development (n = 550)
%
Implementation (n = 550)
%
Limited personnel resources to develop Lack of support from physicians Difficulty with broadly educating Lack of support from hospital administration Lack of monetary resources
24.5 23.6 18.4 4.7 3.6
Physicians do not like to use Clinicians unaware Lack of resources to implement Lack of time to implement Lack of staffing to implement
21.3 20.9 19.6 10.0 9.1
3.4. Protocol development methods Sixty-nine percent of respondents indicated that protocols offered by national organizations were either very or extremely important/valuable in their institutions' development of protocols. The initiation of protocol development at the respondents' institution was spurred by the need for a standard approach to care (73.2%), the critical care team (69.3%), someone personally interested in development (59.9%), publication of national protocols (59.2%), and national safety committee recommendations (58.1%). Perceived limitations of protocol development are listed in Table 3, although 8.7% of respondents indicated that there were no limitations. Final approval of the protocol before implementation was most often granted by hospital administration/medical executive committee (33.6%), followed by multidisciplinary critical care committee (19.0%) and ICU administrator teams (medical director, nursing management, etc; 12.0%).
3.5. Protocol implementation strategies More than 85% of respondents indicated that they adopted national protocols when available some, most, or all of the time. Popular methods of education for protocol implemen-
didn't know
28.1
13.3
other
16.6
7.1 0
when new info is available
40.8
0.2 1.1
> every 4 years
4.9
every 2-3 years
12.4 14.6 17.9
every 12 months every 6 months
4.0
every 3 months
9.1 14.2
1.1
never
12.2
2.2 0
10
Assessed for Adherence
20
30
40
50
Assessed for Updates
Fig. 1 Percent of respondents reporting timelines of assessment of protocols.
tation included staff meetings (85.3%), unit-specific inservices (77.7%), e-mails (54.4%), road shows (48.5%), and newsletters (41.4%). Limitations of implementing protocols are shown in Table 3. Adequate training before implementation was felt to be adequate some or most of the time (84.0%). Automatic initiation of protocols according to specified criteria and not requiring a physician's order occurred most or all of the time according to 18.1% of respondents, with 56.2% reporting that this happened with none or only a few protocols.
3.6. Protocol assessment approaches Protocols were most often updated when new information was available (40.8%) or every 12 months (17.9%; Fig. 1). Fig. 1 also shows timelines of when respondents reported that protocols were assessed for adherence.
3.7. Perceived benefit of protocol use Respondents were asked to evaluate how effective protocols were in improving patient outcomes, improving patient safety, and decreasing costs. Results are shown in Table 4. Clinicians indicated that the most important advantages in using protocols were efficacy/improved patient outcomes (55.9%), consistency in ordering (27.1%), and patient safety (13.7). Most common disadvantages of protocols were cited as patient care not being specific to the patient (34.4%), causing confusion among clinicians (28.2%), and lack of accountability of staff (20.2%). Reported advantages and disadvantages did not differ significantly between the 3 professions. Barriers to protocol use included physicians not knowing protocols existed (52.8%) or not agreeing with the protocol content (53.0%), not enough training for staff (54.1%), and complexity of the protocol (49.0%). In the opinion section of the survey, 26.2% of respondents indicated that they had encountered a situation where they believed a protocol had led to the harm of the patient; this included 14.7% of physicians, 6.9% of pharmacists, and 4.5% of nurses answering the survey. More than 90% of clinicians also indicated that they believed that it was necessary to deviate from protocols to serve the best interests of the patient either a little or some of the time. Sepsis protocols were identified as most useful in promoting patient outcomes by all 3 professions. Nurses
738.e14 Table 4
J.M. LeBlanc et al. Percent perceived efficacy of protocols
Never A little of the time Some of the time Most of the time All of the time a b
Improve patient outcomes (n = 549) a
Improve patient safety (n = 548) b
Decrease costs (n = 551) b
0.5 3.3 20.8 62.7 12.8
0.2 1.3 17.5 63.7 17.3
0.4 6.2 37.3 46.1 10.1
P b .05 via Bonferroni post hoc analysis for nurse vs physician or pharmacist. P b .05 via Bonferroni post hoc analysis for nurse vs physician.
Table 5
Opinions regarding the outcomes of protocols
Parameter a
Make your job easier (nurse n = 167, pharmacist n = 149, physician n = 233)
Improve cost containment (nurse n = 166, pharmacist n = 149, physician n = 234)
Provides consistency in ordering a (nurse n = 167, pharmacist n = 149, physician n = 235)
Improve patient outcomes b (nurse n = 166, pharmacist n = 149, physician n = 230)
Help prevent medication errors (nurse n = 165, pharmacist n = 149, physician n = 233)
Prevent adverse drug reactions (nurse n = 166, pharmacist n = 149, physician n = 234)
Prevent patient harm c (nurse n = 167, pharmacist n = 148, physician n = 234)
Causes confusion in patient care (nurse n = 165, pharmacist n = 149, physician n = 230)
a b c
Likert scale
Nurse
Pharmacist
Physician
Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time
– 2.4 17.4 59.9 20.4 0.6 3.6 31.9 51.8 12.0 0.6 – 7.8 73.1 18.6 – 0.6 9.0 68.7 21.7 1.2 3.6 26.1 55.8 13.3 2.4 12.0 32.5 46.4 6.6 0.6 2.4 18.0 70.7 8.4 12.1 47.9 30.3 7.3 2.4
– 2.0 22.1 61.7 14.1 – 1.3 36.9 55.7 6.0 – – 6.7 79.2 14.1 – 0.7 16.8 71.1 11.4 – 2.0 26.2 61.1 10.7 1.3 8.1 39.6 47.0 4.0 – 3.4 26.4 63.5 6.8 12.1 46.3 35.6 6.0 –
1.3 4.3 24.0 55.8 14.6 0.4 6.4 40.6 44.0 8.5 0.4 1.3 13.2 72.8 12.3 0.4 5.7 26.5 54.8 12.6 1.3 4.7 30.0 55.8 8.2 3.0 16.7 41.5 32.1 6.8 1.3 6.0 36.6 48.7 7.3 7.0 51.7 32.2 8.3 0.9
P b .05 by post hoc Bonferroni for nurse vs physician. P b .05 by post hoc Bonferroni for each group vs the other. P b .05 by post hoc Bonferroni for physician vs nurse or pharmacist.
P .024
.045
.015
b.001
.089
.023
b.001
.584
Multiprofessional survey of protocol use in the ICU and physicians indicated that sedation protocols were most useful in promoting patient safety, whereas pharmacists most commonly reported heparin protocols. Protocols commonly providing cost containment differed among all 3 professions: ventilator-associated pneumonia (nurses), activated protein C (pharmacists), and sedation protocols (physicians). Respondents' opinions about outcomes of protocol use differed significantly between professions in regard to making their job easier, improving patient outcomes, providing consistency in care, and preventing patient harm (Table 5).
4. Discussion Protocol use has substantially increased over the past decade. Patient outcomes associated with protocol use have typically demonstrated positive results [8,12-17]. This is the first study to provide an understanding on development, implementation, assessment, and perception of utility of protocols among interdisciplinary, critical care clinicians. This information will be useful for institutions in the process of developing protocols and for those institutions encouraging compliance with existing protocols. The most common order sets encompassed deep vein thrombosis and stress ulcer prophylaxis, potentially a result of the Center for Medicare and Medicaid Services requirement to measure provision of these therapies as a quality indicator or the inclusion of these therapies in sepsis bundles demonstrated to improve patient outcomes. The most common protocols included heparin, activated protein C, venous thromboembolism prophylaxis and sedation. These may be the most common for very different reasons. Heparin is a high-risk medication that has received much press over the last several years and represents a drug often protocolized for ease of administration and titration by nursing staff. Activated protein C represents a high-cost medication that institutions have protocolized to manage costs. The need for an activated protein C protocol is projected to decline with the recent removal of Xigris (Eli Lilly and Company) form the market. Sedation protocols have become very popular, with published evidence showing improved outcomes with nurse- and pharmacist-driven protocols [12,18,19]. The most common reason for development of a protocol was to standardize the approach to care. National protocol or safety committee publications also helped to spur the development of protocols, with almost 70% of respondents indicating that protocols offered by national organizations were important for their own institutions' protocol development. National protocols were adopted at least some of the time when available. The most common limitation to protocol development was a lack of support, either from physicians or hospital administration. This represents an opportunity to provide evidence-based data to the nonsupporters demonstrating that the use of protocols improves patient outcomes and reduces costs to gain support. In addition to development, many institutions struggled with
738.e15 the implementation of protocols. Barriers to this were cited by respondents, including lack of physician buy-in and knowledge of protocol existence. These are consistent with previous surveys that have shown lack of physician awareness, familiarity, agreement, and outcome expectancy to be barriers to protocol adherence [20]. In a multidisciplinary survey, Tanios et al [21] found the top 3 barriers to be lack of physician order, lack of nursing acceptance, and more control being preferred than a protocol offers, which did not differ in order among the 3 health professional groups. Deficiencies in resources to implement protocols, in the form of staffing, time, or other, were also commonly cited in the present survey. Another prominent limitation cited included clinicians being unaware of protocols. This may be linked to inadequate education about protocols in use. Less than 50% of respondents indicated that staff received adequate training before implementation at least most of the time. Respondents indicated that approval of protocols was usually required from a hospital-based administration committee or a multidisciplinary critical care committee. The ISMP standards for order set implementation recommend that approval come from an interdisciplinary committee composed of allied health care professional who might “use, carry out, or maintain the order set” [9]. Given these recommendations, perhaps more emphasis should be placed on having an ICU committee comprised of the allied health stakeholders involved. Protocols were updated when new information was available in more than 40% of responses. This number was anticipated to be larger because it would be expected that new information should be incorporated into a protocol if it was deemed necessary. As above, the ISMP guidelines are not clear as to when to assess the protocols; however, having a timeline and designated person in place to champion ongoing review of updated literature pertaining to the protocol would be beneficial. This is a recommended approach to the success of quality improvement programs [22]. All 3 professions indicated that the most important advantage of protocols was improved patient outcomes. Protocol implementation has been demonstrated in the literature to improve mortality and morbidity outcomes. Although 21.3% of respondent institutions had a traumatic brain injury protocol in place, reports from single institutional implementation demonstrated a reduction in both hospital and ICU mortality [14,15]. Sedation protocols have been shown to decrease the duration of mechanical ventilation and length of stay in hospital [5]. Protocols managing transfusion management, sepsis resuscitation, and ventilator-associated pneumonia have also shown improved outcomes [8,16,17]. Sepsis protocols were identified as the most frequently used protocol to improve patient outcomes by all 3 professions. This is supported by a recent analysis showing that incorporating “sepsis bundles” improved survival and antibiotic use among the heterogeneous studies [23].
738.e16 More than 80% of clinicians indicated that they believed that protocols improved patient safety most or all of the time. Although many clinicians agreed that improved patient safety was an advantage of protocol use, more than 25% of respondents indicated that they had encountered a situation where a protocol led to patient harm. Most clinicians also indicated that it was necessary to deviate from protocols to serve the best interests of the patient some times. This may identify with the concept of “cookbook medicine” discussed previously. The most commonly identified protocols to promote patient safety were sedation protocols among nurses and physicians. Interestingly, the most commonly identified protocol by pharmacists was heparin. This may be accounted for in differences of roles and focuses of the different health care professionals. Although the choices were different among professions, the actual percentages of the most common protocol chosen were low, representing much variability within professions, as well. More than 95% of clinicians felt that protocols made their job easier at least some of the time. For nurses, protocols have been reported to increase autonomous clinical decisions, allowing them to act without having to follow up with a physician [24]. Almost 90% of respondents indicated they felt that protocols provided consistency in ordering most or all of the time. Afessa et al [13] demonstrated that care variability was decreased with institution of multiple evidence-based guidelines for staff. Most respondents indicated that protocols aided in preventing medication errors and adverse drug reactions at least some of the time. Also, protocols were felt to prevent patient harm by most respondents. Cost containment has and continues to be a major impetus for the establishment of protocols. All 3 professions chose a different protocol in being the one most likely to provide cost containment, with the most common being activated protein C chosen by pharmacists. Almost 75% of institutions had a protocol for the use of activated protein C. Respondents believed that protocols decreased costs most or all of the time, and the literature supports this assertion. Jackson et al [5] found a reduction in costs during a systematic review of sedation protocols, ranging from 22% to 94%. Although some protocols do not appear to directly decrease costs, they are considered cost-effective, which should also be a consideration of the institution. The most common disadvantage of protocols was the lack of specificity of care to the individual patient. Meade and Ely [7] commented that protocols should not be viewed as cookbook medicine with regard to respiratory weaning protocols and should not be so rigid as to compromise patient safety and comfort. More than 20% of respondents in a recent survey of more than 1300 predominantly nursing clinicians agreed that practice “guidelines are too cookbook and prescriptive” [25]. Another study also found that nurses were afraid that protocols “were taking the thinking out of nursing” [26]. Within the current survey, greater than 25% of
J.M. LeBlanc et al. respondents also indicated that protocols cause confusion among clinicians.
4.1. Limitations Limitations of this study include the low response rate, particularly from the physician subset of data. Because the survey was disseminated through the SCCM organization, we were unable to access the list of nonresponders to ensure that they did not differ from the survey responders. The responses are also mainly representative of US clinicians practicing with an adult population and may not be generalizable to other types of practice. Although definitions and examples were used liberally in the questions, misinterpretation of questions or potential answers may have affected the results. Also, we relied on participants' responses in sharing accurate data and did not have institutions submit their actual protocols.
5. Conclusions The types of protocols available appear to be those assisting with management of high-alert medications. Designated personnel evaluating the new literature to determine the need for protocol updates may be useful to the institution. Also, quality improvement processes that evaluate compliance and reasons for noncompliance specific to the institution may be useful. Overcoming the perceived barriers of protocol use within ICUs requires personnel for development and physician support. A better protocol review process may be necessary to assure optimal content, desired outcomes, and consistency with ISMP guidelines. Clinician opinions regarding the outcomes of protocols differed between professions, but overall clinicians believe that protocols improve patient outcomes.
References [1] Cullen DJ, Sweitzer BJ, Bates DW, et al. Preventable adverse drug events in hospitalized patients: a comparative study of intensive care and general care units. Crit Care Med 1997;25:1289-97. [2] Kane-Gill SL, Kowiatek JG, Weber RJ. A comparison of voluntarily reported medication errors in intensive care and general care units. Qual Saf Health Care 2010;19:55-9. [3] Hammond JJ. Protocols and guidelines in critical care: development and implementation. Curr Opin Crit Care 2001;7:464-8. [4] Holcomb BW, Wheeler AP, Ely EW. New ways to reduce unnecessary variation and improve outcomes in the intensive care unit. Curr Opin Crit Care 2001;7:304-11. [5] Jackson DL, Proudfoot CW, Cann KF, et al. A systematic review of the impact of sedation practice in the ICU on resource use, costs and patient safety. Crit Care 2010;14:R59. [6] Talmor D, Greenberg D, Howell MD, et al. The costs and costeffectiveness of an integrated sepsis treatment protocol. Crit Care Med 2008;36:1168-74.
Multiprofessional survey of protocol use in the ICU [7] Meade MO, Ely EW. Protocols to improve the care of critically ill pediatric and adult patients. JAMA 2002;288:2601-3. [8] Nicasio AM, Eagye KJ, Kuti EL, et al. Length of stay and hospital costs associated with a pharmacodynamic-based clinical pathway for empiric antibiotic choice for ventilator-associated pneumonia. Pharmacotherapy 2010;30:453-62. [9] ISMP'S guidelines for standard order sets. Institute of Safe Medication Practices; 2010. Available at: http://www.ismp.org/Tools/guidelines/ StandardOrderSets.asp. Accessed August 10, 2012. [10] Prasad M, Christie JD, Bellamy SL, et al. The availability of clinical protocols in US teaching intensive care units. J Crit Care 2010;25: 610-9. [11] Mehta S, Burry L, Fischer S, et al. Canadian survey of the use of sedatives, analgesics, and neuromuscular blocking agents in critically ill patients. Crit Care Med 2006;34:374-80. [12] Quenot JP, Ladoire S, Devoucoux F, et al. Effect of a nurseimplemented sedation protocol on the incidence of ventilatorassociated pneumonia. Crit Care Med 2007;35:2031-6. [13] Afessa B, Gajic O, Keegan MT, et al. Impact of introducing multiple evidence-based clinical practice protocols in a medical intensive care unit: a retrospective cohort study. BMC Emerg Med 2007;7:10. [14] Arabi YM, Haddad S, Tamim HM, et al. Mortality reduction after implementing a clinical practice guidelines–based management protocol for severe traumatic brain injury. J Crit Care 2010;25: 190-5. [15] Clayton TJ, Nelson RJ, Manara AR. Reduction in mortality from severe head injury following introduction of a protocol for intensive care management. Br J Anaesth 2004;93:761-7. [16] Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care.
738.e17
[17]
[18]
[19]
[20]
[21]
[22]
[23] [24] [25]
[26]
Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340:409-17. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368-77. Brook AD, Ahrens TS, Schaiff R, et al. Effect of a nursingimplemented sedation protocol on the duration of mechanical ventilation. Crit Care Med 1999;27:2609-15. Marshall J, Finn CA, Theodore AC. Impact of a clinical pharmacist– enforced intensive care unit sedation protocol on duration of mechanical ventilation and hospital stay. Crit Care Med 2008;36:427-33. Cabana MD, Rand CS, Powe NR, et al. Why don't physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999;282:1458-65. Tanios MA, de Wit M, Epstein SK, et al. Perceived barriers to the use of sedation protocols and daily sedation interruption: a multidisciplinary survey. J Crit Care 2009;24:66-73. Curtis JR, Cook DJ, Wall RJ, et al. Intensive care unit quality improvement: a “how-to” guide for the interdisciplinary team. Crit Care Med 2006;34:211-8. Barochia A, Cui X, Vitberg D, et al. Bundled care for septic shock: an analysis of clinical trials. Crit Care Med 2010;38:668-78. Manias E, Street A. Legitimation of nurses' knowledge through policies and protocols in clinical practice. J Adv Nurs 2000;32:1467-75. Pogorzelska M, Larson EL. Assessment of attitudes of intensive care unit staff toward clinical practice guidelines. Dimens Crit Care Nurs 2008;27:30-8. Flynn AV, Sinclair M. Exploring the relationship between nursing protocols and nursing practice in an Irish intensive care unit. Int J Nurs Pract 2005;11:142-9.
Multiprofessional survey of protocol use in the intensive care unit Jaclyn M. LeBlanc PharmD, BCPS a,⁎, Sandra L. Kane-Gill PharmD, MSc, FCCM, FCCP b , Anne S. Pohlman MSN, CCRN c , Daniel L. Herr MD d a
Department of Pharmacy, Saint John Regional Hospital, Saint John, NB E2E 4L2 Pharmacy and Therapeutics, Clinical Translational Sciences and Critical Care Medicine, Schools of Pharmacy and Medicine, University of Pittsburgh and Medication Safety Officer, Department of Pharmacy, UPMC, Pittsburgh, PA c Critical Care Clinical Research, University of Chicago, Chicago, IL d University of Maryland, Baltimore, MD b
Keywords: Critical care; Protocol; Multidisciplinary
Abstract Purpose: To date, there has been no large multicenter, multiprofessional evaluation of protocol and guideline use in the intensive care unit (ICU). The primary purpose of this study was to describe national availability, development, implementation, and assessment of protocols in ICUs. A secondary objective was to compare perceived utility by ease of use, patient safety, cost containment, and compliance of protocols between nurses, physicians, and pharmacists. Materials and Methods: The survey was developed and tested for validity by 15 clinicians who identified additional domains of interest. An additional 15 clinicians of the 3 different professions evaluated the survey for relevancy and appropriateness of responses. Three survey experts evaluated survey construction. The survey was uploaded to a Web survey tool and pilot tested for clarity and ease of completion. Results: The overall response rate for the survey was 18.1% (n = 614). Popular methods of education for protocol implementation included staff meetings (85.3%) and unit-specific in-services (77.7%). Protocols were most often updated when new information was available (40.8%) or every 12 months (17.9%). The most common limitation to development and implementation was limited personnel resources (24.5%) and physicians not wanting to use them (21.3%), respectively. Clinicians indicated that protocols made their job easier and improved cost containment some or most of the time. Sepsis protocols were identified as most useful in promoting patient outcomes by all 3 professions. Conclusions: The types of protocols available appear to be those assisting with management of high-alert medications. Overcoming the perceived barriers of protocol use within ICUs requires personnel for development and physician support. A better protocol review process may be necessary to assure optimal content, desired outcomes, and consistency with Institute for Safe Medication Practices guidelines. © 2012 Elsevier Inc. All rights reserved.
⁎ Corresponding author. Tel: +1 506 649 2875. E-mail address: [email protected] (J.M. LeBlanc). 0883-9441/$ – see front matter © 2012 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.jcrc.2012.07.012
738.e10
1. Background Many hospitals have instituted evidence-based, drugspecific protocols to create uniformity of medication orders to decrease variability among prescribers, promote safe practices, and potentially improve patient outcomes throughout a variety of patient care areas. The intensive care unit (ICU) is an environment where medication errors and adverse drug events are more frequent and severe, so protocols aimed to improve safe medication use are appropriate [1,2]. Protocol use in the ICU has improved patient outcomes, reduced length of patient stays, and reduced costs [3-8]. In fact, the use of standardized order sets is recommended by regulatory bodies such as the Institute for Safe Medication Practices (ISMP) [9]. To date, there has been no large multicenter, multiprofessional evaluation of protocol and guideline use in the ICU. One study did survey ICU physicians about protocols, but it contained limited questions to protocol availability and did not explore development, implementation, and assessment [10]. Obtaining this information is important to understand the impetus for protocol development, how implementation occurs, and the frequency of assessments for updates or adherence. These insights would aid in recommending best practices for protocol use. In addition, an appreciation for barriers to protocol adherence with possible recommendations for improvement would be obtained. The primary purpose of this study was to describe national availability, development, implementation, and assessment of protocols in ICUs. A secondary objective was to compare perceived utility by ease of use, patient safety, cost containment, and compliance of protocols between nurses, physicians, and pharmacists.
2. Methods 2.1. Survey development A thorough approach to survey development was followed similar to a previous publication [11]. The survey was developed by the investigators to obtain information addressing the study's primary and secondary objectives. Questions were derived addressing hospital demographics; protocols available in the respondent's hospital; implementation and developments of these protocols; and perceived usefulness of these protocols with regards to ease of use, patient safety, and cost containment. A literature search was conducted in MEDLINE to develop a comprehensive list of existing protocols. References to more than 70 protocols were identified in the MEDLINE search, and questions were developed pertaining to these. The literature search also revealed confusion surrounding the definition of a protocol; thus, a clear definition was provided in the current survey stating a protocol is as
J.M. LeBlanc et al. follows: “an explicit framework for the process of or algorithmic approach to care whereby members of the health care team (physician, nurse, or pharmacist) can follow precise steps of practice. The word protocol is also being used in place of the word order set, clinical pathway, or bundle”. In addition, a sentence was included to ensure that participants realized that protocol was not meant to refer to those used in the realm of research. The response format consisted of tick boxes and open-ended questions, depending on the type of question.
2.2. Survey testing To test the validity of the survey created by the investigators, 5 critical care physicians, 5 pharmacists, and 5 nurses external to the investigative team were surveyed to identify additional domains of interest. These clinicians were sampled until redundancy. Subsequently, the investigators reviewed all questions and reduced the number to a goal of 30, identifying the relevant questions that met the overall objectives and provided educational information to the audience. An additional 15 clinicians of the 3 different professions evaluated the survey for relevancy and appropriateness of responses. Modifications were made to the survey in response to this review. Lastly, 3 survey experts evaluated survey construction. Further survey modifications were made in response to the expert review. The survey was uploaded to a Web survey tool and pilot tested for clarity and ease of completion by 9 clinicians of various professions.
2.3. Survey administration The study was approved by Society of Critical Care Medicine (SCCM) Pharmacy and Clinical Pharmacology Advisory Board and then reviewed by members of the SCCM internal medicine and nursing sections. After institutional review board approval, the proposal was submitted to the SCCM Research Committee and the SCCM Executive Committee for approval to be sent to SCCM members. The survey invitation was e-mailed to members of the following SCCM sections: Pharmacy and Clinical Pharmacology, Nursing, and Internal Medicine. Reminders were sent via email to the members at 1 and 2.5 months after the initial notification of the survey. The mailings were facilitated by SCCM. A statement within the survey communicated that completion of the questionnaire conveyed the respondents' consent to participate in the study.
2.4. Data analysis All data were exported directly from the Web survey tool and analyzed with SPSS 16.0 (SPSS, Inc, Chicago, Ill). A complete survey was defined as 80% completion and provision of the respondent's profession. The survey was anonymous with a voluntary request for the submission of an
Multiprofessional survey of protocol use in the ICU Table 1
738.e11
Demographics of all participants (n = 551)
Type of hospital (all n = 544, hospital n = 329)
Approximate no. of licensed hospital beds (all n = 546, hospital n = 332)
Type of ICU (all n = 547, hospital n = 332)
Patient population in the ICU (all n = 543, hospital n = 333)
No. of licensed ICU beds (all n = 549, hospital n = 332)
Community nonteaching Community teaching Private for-profit Private not-for-profit University teaching Veterans Affairs/military/government 1-199 200-399 400-599 600-799 800-999 ≥1000 Cardiology (including cardiology, cardiothoracic with or without surgery, coronary care) Medical Medical/Surgical Neonatal/Pediatric Surgical/Trauma Mixed Other Adult Pediatric Both Unknown 5-10 11-20 21-30 31-40 41-50 51-60 N60
institution name with the intention of removing duplicate institutions for subgroup analyses. Descriptive statistics were used to analyze data, in addition to χ2 test of proportions and analysis of variance to compare responses by profession. The Kruskal-Wallis and Bonnferroni post hoc tests were used to analyze differences between professions. A P value of .05 was considered significant.
All participants (%)
Individual hospitals (%)
20.2 36.0 1.5 5.0 33.6 3.7 11.4 30.8 30.2 15.9 7.7 4.0 6.4
21.6 39.2 1.5 4.3 30.4 3.0 11.7 31.3 31.0 14.2 7.5 4.2 5.7
19.7 53.6 1.6 15.2 3.3 0.2 92.0 1.5 4.7 0.4 7.8 47.4 25.1 11.2 3.1 1.5 2.2
17.8 54.2 1.8 17.5 2.7 0.3 92.0 1.8 4.7 1.5 8.1 50.1 25.5 9.3 3.9 0.9 1.2
clinicians (87.5%) in the form of physicians (42.6%), nurses (30.3%), and pharmacists (27.0%). The response rate per profession was 24.8% nursing, 19.9% pharmacist, and 11.9% physician. Of these, 435 provided their hospital name for subanalysis, and those respondents represented 333 different ICUs.
3.2. Demographics
3. Results 3.1. Response rate A total of 3400 survey invitations were e-mailed, with an overall response rate of 614 (18.1%). Of these, 55 were excluded for completion of less than 80% of the survey (n = 52) or not providing their profession (n = 3). Also, 7 respondents indicated their profession as “other” and were thus excluded from the analysis. This left 551 (16.2%) responses for analysis representing predominantly US
The data set represented primarily US practitioners and distribution across different ICU types. The ICU types (n = 542) represented included 28.2% open (any attending physician with hospital admitting privileges can be the physician of record and direct ICU care), 32.8% closed (only an intensivist is the physician of record for all ICU patients), and 38.9% transitional (shared, comanaged care between ICU staff and private physician; an intensivist director, trainees, and intensivist team are present as locally available). Respondents encompassed a wide range of clinical experience indicating that they had been a licensed
738.e12 Table 2
J.M. LeBlanc et al. Protocols used by respondent hospitals (n = 333)
Category
Specific protocols
%
Cardiac
ACS Atrial fibrillation Chest pain (independent of ACS, managing only pain) CHF management Coronary artery bypass grafting postgraft management Hypertension management Hypotension management Therapeutic hypothermia Activated protein C (independent of a sepsis bundle or protocol) Catheter-related blood stream infections Community-acquired pneumonia Febrile neutropenia Fungal infection prevention Meningitis Nosocomial pneumonia (either on or off ventilator) Sepsis (including goal directed therapy) Ventilator-associated pneumonia Heparin Anemia/Hemoglobin management Low-molecular-weight heparin Warfarin Factor VII Heparin-induced thrombocytopenia— argatroban Heparin-induced thrombocytopenia— bivalirudin Heparin-induced thrombocytopenia— lepirudin Thrombocytopenia management Venous thromboembolism prophylaxis Benzodiazepine tapering Delirium (not including sedation) Drug withdrawal Ethanol withdrawal Intracranial pressure management/ intracerebral hemorrhage Sedation Status epilepticus Stroke Sleep disturbance Transient ischemic attack Traumatic brain injury Subarachnoid hemorrhage Spinal cord injury medication management Adrenal insufficiency management Cosyntropin/cotrosyn adrenal stimulation test Diabetic ketoacidosis Potassium
58.3 17.4 35.7
Infectious diseases
Hematology
CNS
Metabolic
Table 2 (continued) Category
38.7 44.4 12.6 15.6 53.8 74.2
Respiratory
34.5
Gastrointestinal
58.0 13.5 5.4 6.0 33.9 68.8 47.7 83.8 16.2 44.4 34.8 18.6 36.9 10.2 18.0
Medications
Specific protocols
%
Phosphate Magnesium Calcium Sodium Combined electrolytes Hyperglycemia—IV insulin Hyperglycemia—SQ insulin Hyperosmolar hyperglycemic nonketotic coma ARDS Bronchodilators Status asthmaticus Bowel regime Gastric motility Hepatic encephalopathy Nutritional support—enteral Nutritional support—parenteral Pancreatitis Stress ulcer prophylaxis Upper GI bleed Aminoglycosides Contrast nephropathy prevention Desensitization for drug allergies Eptifibatide Neuromuscular blockers Opioid taping protocol Pain management Phenytoin Abciximab Vancomycin
40.2 49.2 27.9 19.2 29.1 88.3 57.7 19.5 47.1 27.9 11.1 28.8 21.9 6.3 52.9 49.2 6.0 71.5 12.0 39.0 41.4 16.2 32.4 43.8 6.9 47.7 7.8 19.2 38.4
ACS, acute coronary syndrome; CHF indicates congestive heart failure; CNS, central nervous system; IV, intravenous; SQ, subcutaneous; ARDS, acute respiratory distress syndrome; GI, gastrointestinal.
4.5 69.4 12.3 25.2 14.7 47.7 25.5 69.1 9.6 52.6 1.5 11.7 21.3 19.8 18.0 13.8 24.9 47.1 62.5
clinician for 19.6 ± 10.0 years (range, 2-43 years) and had been working as a clinician in the ICU for 16.6 ± 9.6 years (range, 0-41 years). Other demographics are shown in Table 1. The types of ICUs represented primarily comprised adult medical and surgical patients, with little representation of pediatric or neonatal patients. There was a fairly even distribution of hospital size, with more than 70% of respondents practicing in ICUs with a bed size between 11 and 30 beds.
3.3. Types of protocols available nationally The most common types of protocols are shown in Table 2 for the 333 respondents from different hospitals. Clinicians were also asked about ICU admission order sets, and 72.4% of these hospitals had them. These order sets most often included deep vein thrombosis prophylaxis (67.0%), stress ulcer prophylaxis (66.1%), sedation (49.2%), pain management (46.8%), breathing treatments (41.7%), and bowel regimens (36.9%),
Multiprofessional survey of protocol use in the ICU Table 3
738.e13
Top 5 limitations of development and implementation of protocols
Development (n = 550)
%
Implementation (n = 550)
%
Limited personnel resources to develop Lack of support from physicians Difficulty with broadly educating Lack of support from hospital administration Lack of monetary resources
24.5 23.6 18.4 4.7 3.6
Physicians do not like to use Clinicians unaware Lack of resources to implement Lack of time to implement Lack of staffing to implement
21.3 20.9 19.6 10.0 9.1
3.4. Protocol development methods Sixty-nine percent of respondents indicated that protocols offered by national organizations were either very or extremely important/valuable in their institutions' development of protocols. The initiation of protocol development at the respondents' institution was spurred by the need for a standard approach to care (73.2%), the critical care team (69.3%), someone personally interested in development (59.9%), publication of national protocols (59.2%), and national safety committee recommendations (58.1%). Perceived limitations of protocol development are listed in Table 3, although 8.7% of respondents indicated that there were no limitations. Final approval of the protocol before implementation was most often granted by hospital administration/medical executive committee (33.6%), followed by multidisciplinary critical care committee (19.0%) and ICU administrator teams (medical director, nursing management, etc; 12.0%).
3.5. Protocol implementation strategies More than 85% of respondents indicated that they adopted national protocols when available some, most, or all of the time. Popular methods of education for protocol implemen-
didn't know
28.1
13.3
other
16.6
7.1 0
when new info is available
40.8
0.2 1.1
> every 4 years
4.9
every 2-3 years
12.4 14.6 17.9
every 12 months every 6 months
4.0
every 3 months
9.1 14.2
1.1
never
12.2
2.2 0
10
Assessed for Adherence
20
30
40
50
Assessed for Updates
Fig. 1 Percent of respondents reporting timelines of assessment of protocols.
tation included staff meetings (85.3%), unit-specific inservices (77.7%), e-mails (54.4%), road shows (48.5%), and newsletters (41.4%). Limitations of implementing protocols are shown in Table 3. Adequate training before implementation was felt to be adequate some or most of the time (84.0%). Automatic initiation of protocols according to specified criteria and not requiring a physician's order occurred most or all of the time according to 18.1% of respondents, with 56.2% reporting that this happened with none or only a few protocols.
3.6. Protocol assessment approaches Protocols were most often updated when new information was available (40.8%) or every 12 months (17.9%; Fig. 1). Fig. 1 also shows timelines of when respondents reported that protocols were assessed for adherence.
3.7. Perceived benefit of protocol use Respondents were asked to evaluate how effective protocols were in improving patient outcomes, improving patient safety, and decreasing costs. Results are shown in Table 4. Clinicians indicated that the most important advantages in using protocols were efficacy/improved patient outcomes (55.9%), consistency in ordering (27.1%), and patient safety (13.7). Most common disadvantages of protocols were cited as patient care not being specific to the patient (34.4%), causing confusion among clinicians (28.2%), and lack of accountability of staff (20.2%). Reported advantages and disadvantages did not differ significantly between the 3 professions. Barriers to protocol use included physicians not knowing protocols existed (52.8%) or not agreeing with the protocol content (53.0%), not enough training for staff (54.1%), and complexity of the protocol (49.0%). In the opinion section of the survey, 26.2% of respondents indicated that they had encountered a situation where they believed a protocol had led to the harm of the patient; this included 14.7% of physicians, 6.9% of pharmacists, and 4.5% of nurses answering the survey. More than 90% of clinicians also indicated that they believed that it was necessary to deviate from protocols to serve the best interests of the patient either a little or some of the time. Sepsis protocols were identified as most useful in promoting patient outcomes by all 3 professions. Nurses
738.e14 Table 4
J.M. LeBlanc et al. Percent perceived efficacy of protocols
Never A little of the time Some of the time Most of the time All of the time a b
Improve patient outcomes (n = 549) a
Improve patient safety (n = 548) b
Decrease costs (n = 551) b
0.5 3.3 20.8 62.7 12.8
0.2 1.3 17.5 63.7 17.3
0.4 6.2 37.3 46.1 10.1
P b .05 via Bonferroni post hoc analysis for nurse vs physician or pharmacist. P b .05 via Bonferroni post hoc analysis for nurse vs physician.
Table 5
Opinions regarding the outcomes of protocols
Parameter a
Make your job easier (nurse n = 167, pharmacist n = 149, physician n = 233)
Improve cost containment (nurse n = 166, pharmacist n = 149, physician n = 234)
Provides consistency in ordering a (nurse n = 167, pharmacist n = 149, physician n = 235)
Improve patient outcomes b (nurse n = 166, pharmacist n = 149, physician n = 230)
Help prevent medication errors (nurse n = 165, pharmacist n = 149, physician n = 233)
Prevent adverse drug reactions (nurse n = 166, pharmacist n = 149, physician n = 234)
Prevent patient harm c (nurse n = 167, pharmacist n = 148, physician n = 234)
Causes confusion in patient care (nurse n = 165, pharmacist n = 149, physician n = 230)
a b c
Likert scale
Nurse
Pharmacist
Physician
Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time Never A little of the time Some of the time Most of the time All of the time
– 2.4 17.4 59.9 20.4 0.6 3.6 31.9 51.8 12.0 0.6 – 7.8 73.1 18.6 – 0.6 9.0 68.7 21.7 1.2 3.6 26.1 55.8 13.3 2.4 12.0 32.5 46.4 6.6 0.6 2.4 18.0 70.7 8.4 12.1 47.9 30.3 7.3 2.4
– 2.0 22.1 61.7 14.1 – 1.3 36.9 55.7 6.0 – – 6.7 79.2 14.1 – 0.7 16.8 71.1 11.4 – 2.0 26.2 61.1 10.7 1.3 8.1 39.6 47.0 4.0 – 3.4 26.4 63.5 6.8 12.1 46.3 35.6 6.0 –
1.3 4.3 24.0 55.8 14.6 0.4 6.4 40.6 44.0 8.5 0.4 1.3 13.2 72.8 12.3 0.4 5.7 26.5 54.8 12.6 1.3 4.7 30.0 55.8 8.2 3.0 16.7 41.5 32.1 6.8 1.3 6.0 36.6 48.7 7.3 7.0 51.7 32.2 8.3 0.9
P b .05 by post hoc Bonferroni for nurse vs physician. P b .05 by post hoc Bonferroni for each group vs the other. P b .05 by post hoc Bonferroni for physician vs nurse or pharmacist.
P .024
.045
.015
b.001
.089
.023
b.001
.584
Multiprofessional survey of protocol use in the ICU and physicians indicated that sedation protocols were most useful in promoting patient safety, whereas pharmacists most commonly reported heparin protocols. Protocols commonly providing cost containment differed among all 3 professions: ventilator-associated pneumonia (nurses), activated protein C (pharmacists), and sedation protocols (physicians). Respondents' opinions about outcomes of protocol use differed significantly between professions in regard to making their job easier, improving patient outcomes, providing consistency in care, and preventing patient harm (Table 5).
4. Discussion Protocol use has substantially increased over the past decade. Patient outcomes associated with protocol use have typically demonstrated positive results [8,12-17]. This is the first study to provide an understanding on development, implementation, assessment, and perception of utility of protocols among interdisciplinary, critical care clinicians. This information will be useful for institutions in the process of developing protocols and for those institutions encouraging compliance with existing protocols. The most common order sets encompassed deep vein thrombosis and stress ulcer prophylaxis, potentially a result of the Center for Medicare and Medicaid Services requirement to measure provision of these therapies as a quality indicator or the inclusion of these therapies in sepsis bundles demonstrated to improve patient outcomes. The most common protocols included heparin, activated protein C, venous thromboembolism prophylaxis and sedation. These may be the most common for very different reasons. Heparin is a high-risk medication that has received much press over the last several years and represents a drug often protocolized for ease of administration and titration by nursing staff. Activated protein C represents a high-cost medication that institutions have protocolized to manage costs. The need for an activated protein C protocol is projected to decline with the recent removal of Xigris (Eli Lilly and Company) form the market. Sedation protocols have become very popular, with published evidence showing improved outcomes with nurse- and pharmacist-driven protocols [12,18,19]. The most common reason for development of a protocol was to standardize the approach to care. National protocol or safety committee publications also helped to spur the development of protocols, with almost 70% of respondents indicating that protocols offered by national organizations were important for their own institutions' protocol development. National protocols were adopted at least some of the time when available. The most common limitation to protocol development was a lack of support, either from physicians or hospital administration. This represents an opportunity to provide evidence-based data to the nonsupporters demonstrating that the use of protocols improves patient outcomes and reduces costs to gain support. In addition to development, many institutions struggled with
738.e15 the implementation of protocols. Barriers to this were cited by respondents, including lack of physician buy-in and knowledge of protocol existence. These are consistent with previous surveys that have shown lack of physician awareness, familiarity, agreement, and outcome expectancy to be barriers to protocol adherence [20]. In a multidisciplinary survey, Tanios et al [21] found the top 3 barriers to be lack of physician order, lack of nursing acceptance, and more control being preferred than a protocol offers, which did not differ in order among the 3 health professional groups. Deficiencies in resources to implement protocols, in the form of staffing, time, or other, were also commonly cited in the present survey. Another prominent limitation cited included clinicians being unaware of protocols. This may be linked to inadequate education about protocols in use. Less than 50% of respondents indicated that staff received adequate training before implementation at least most of the time. Respondents indicated that approval of protocols was usually required from a hospital-based administration committee or a multidisciplinary critical care committee. The ISMP standards for order set implementation recommend that approval come from an interdisciplinary committee composed of allied health care professional who might “use, carry out, or maintain the order set” [9]. Given these recommendations, perhaps more emphasis should be placed on having an ICU committee comprised of the allied health stakeholders involved. Protocols were updated when new information was available in more than 40% of responses. This number was anticipated to be larger because it would be expected that new information should be incorporated into a protocol if it was deemed necessary. As above, the ISMP guidelines are not clear as to when to assess the protocols; however, having a timeline and designated person in place to champion ongoing review of updated literature pertaining to the protocol would be beneficial. This is a recommended approach to the success of quality improvement programs [22]. All 3 professions indicated that the most important advantage of protocols was improved patient outcomes. Protocol implementation has been demonstrated in the literature to improve mortality and morbidity outcomes. Although 21.3% of respondent institutions had a traumatic brain injury protocol in place, reports from single institutional implementation demonstrated a reduction in both hospital and ICU mortality [14,15]. Sedation protocols have been shown to decrease the duration of mechanical ventilation and length of stay in hospital [5]. Protocols managing transfusion management, sepsis resuscitation, and ventilator-associated pneumonia have also shown improved outcomes [8,16,17]. Sepsis protocols were identified as the most frequently used protocol to improve patient outcomes by all 3 professions. This is supported by a recent analysis showing that incorporating “sepsis bundles” improved survival and antibiotic use among the heterogeneous studies [23].
738.e16 More than 80% of clinicians indicated that they believed that protocols improved patient safety most or all of the time. Although many clinicians agreed that improved patient safety was an advantage of protocol use, more than 25% of respondents indicated that they had encountered a situation where a protocol led to patient harm. Most clinicians also indicated that it was necessary to deviate from protocols to serve the best interests of the patient some times. This may identify with the concept of “cookbook medicine” discussed previously. The most commonly identified protocols to promote patient safety were sedation protocols among nurses and physicians. Interestingly, the most commonly identified protocol by pharmacists was heparin. This may be accounted for in differences of roles and focuses of the different health care professionals. Although the choices were different among professions, the actual percentages of the most common protocol chosen were low, representing much variability within professions, as well. More than 95% of clinicians felt that protocols made their job easier at least some of the time. For nurses, protocols have been reported to increase autonomous clinical decisions, allowing them to act without having to follow up with a physician [24]. Almost 90% of respondents indicated they felt that protocols provided consistency in ordering most or all of the time. Afessa et al [13] demonstrated that care variability was decreased with institution of multiple evidence-based guidelines for staff. Most respondents indicated that protocols aided in preventing medication errors and adverse drug reactions at least some of the time. Also, protocols were felt to prevent patient harm by most respondents. Cost containment has and continues to be a major impetus for the establishment of protocols. All 3 professions chose a different protocol in being the one most likely to provide cost containment, with the most common being activated protein C chosen by pharmacists. Almost 75% of institutions had a protocol for the use of activated protein C. Respondents believed that protocols decreased costs most or all of the time, and the literature supports this assertion. Jackson et al [5] found a reduction in costs during a systematic review of sedation protocols, ranging from 22% to 94%. Although some protocols do not appear to directly decrease costs, they are considered cost-effective, which should also be a consideration of the institution. The most common disadvantage of protocols was the lack of specificity of care to the individual patient. Meade and Ely [7] commented that protocols should not be viewed as cookbook medicine with regard to respiratory weaning protocols and should not be so rigid as to compromise patient safety and comfort. More than 20% of respondents in a recent survey of more than 1300 predominantly nursing clinicians agreed that practice “guidelines are too cookbook and prescriptive” [25]. Another study also found that nurses were afraid that protocols “were taking the thinking out of nursing” [26]. Within the current survey, greater than 25% of
J.M. LeBlanc et al. respondents also indicated that protocols cause confusion among clinicians.
4.1. Limitations Limitations of this study include the low response rate, particularly from the physician subset of data. Because the survey was disseminated through the SCCM organization, we were unable to access the list of nonresponders to ensure that they did not differ from the survey responders. The responses are also mainly representative of US clinicians practicing with an adult population and may not be generalizable to other types of practice. Although definitions and examples were used liberally in the questions, misinterpretation of questions or potential answers may have affected the results. Also, we relied on participants' responses in sharing accurate data and did not have institutions submit their actual protocols.
5. Conclusions The types of protocols available appear to be those assisting with management of high-alert medications. Designated personnel evaluating the new literature to determine the need for protocol updates may be useful to the institution. Also, quality improvement processes that evaluate compliance and reasons for noncompliance specific to the institution may be useful. Overcoming the perceived barriers of protocol use within ICUs requires personnel for development and physician support. A better protocol review process may be necessary to assure optimal content, desired outcomes, and consistency with ISMP guidelines. Clinician opinions regarding the outcomes of protocols differed between professions, but overall clinicians believe that protocols improve patient outcomes.
References [1] Cullen DJ, Sweitzer BJ, Bates DW, et al. Preventable adverse drug events in hospitalized patients: a comparative study of intensive care and general care units. Crit Care Med 1997;25:1289-97. [2] Kane-Gill SL, Kowiatek JG, Weber RJ. A comparison of voluntarily reported medication errors in intensive care and general care units. Qual Saf Health Care 2010;19:55-9. [3] Hammond JJ. Protocols and guidelines in critical care: development and implementation. Curr Opin Crit Care 2001;7:464-8. [4] Holcomb BW, Wheeler AP, Ely EW. New ways to reduce unnecessary variation and improve outcomes in the intensive care unit. Curr Opin Crit Care 2001;7:304-11. [5] Jackson DL, Proudfoot CW, Cann KF, et al. A systematic review of the impact of sedation practice in the ICU on resource use, costs and patient safety. Crit Care 2010;14:R59. [6] Talmor D, Greenberg D, Howell MD, et al. The costs and costeffectiveness of an integrated sepsis treatment protocol. Crit Care Med 2008;36:1168-74.
Multiprofessional survey of protocol use in the ICU [7] Meade MO, Ely EW. Protocols to improve the care of critically ill pediatric and adult patients. JAMA 2002;288:2601-3. [8] Nicasio AM, Eagye KJ, Kuti EL, et al. Length of stay and hospital costs associated with a pharmacodynamic-based clinical pathway for empiric antibiotic choice for ventilator-associated pneumonia. Pharmacotherapy 2010;30:453-62. [9] ISMP'S guidelines for standard order sets. Institute of Safe Medication Practices; 2010. Available at: http://www.ismp.org/Tools/guidelines/ StandardOrderSets.asp. Accessed August 10, 2012. [10] Prasad M, Christie JD, Bellamy SL, et al. The availability of clinical protocols in US teaching intensive care units. J Crit Care 2010;25: 610-9. [11] Mehta S, Burry L, Fischer S, et al. Canadian survey of the use of sedatives, analgesics, and neuromuscular blocking agents in critically ill patients. Crit Care Med 2006;34:374-80. [12] Quenot JP, Ladoire S, Devoucoux F, et al. Effect of a nurseimplemented sedation protocol on the incidence of ventilatorassociated pneumonia. Crit Care Med 2007;35:2031-6. [13] Afessa B, Gajic O, Keegan MT, et al. Impact of introducing multiple evidence-based clinical practice protocols in a medical intensive care unit: a retrospective cohort study. BMC Emerg Med 2007;7:10. [14] Arabi YM, Haddad S, Tamim HM, et al. Mortality reduction after implementing a clinical practice guidelines–based management protocol for severe traumatic brain injury. J Crit Care 2010;25: 190-5. [15] Clayton TJ, Nelson RJ, Manara AR. Reduction in mortality from severe head injury following introduction of a protocol for intensive care management. Br J Anaesth 2004;93:761-7. [16] Hebert PC, Wells G, Blajchman MA, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care.
738.e17
[17]
[18]
[19]
[20]
[21]
[22]
[23] [24] [25]
[26]
Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. N Engl J Med 1999;340:409-17. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of severe sepsis and septic shock. N Engl J Med 2001;345:1368-77. Brook AD, Ahrens TS, Schaiff R, et al. Effect of a nursingimplemented sedation protocol on the duration of mechanical ventilation. Crit Care Med 1999;27:2609-15. Marshall J, Finn CA, Theodore AC. Impact of a clinical pharmacist– enforced intensive care unit sedation protocol on duration of mechanical ventilation and hospital stay. Crit Care Med 2008;36:427-33. Cabana MD, Rand CS, Powe NR, et al. Why don't physicians follow clinical practice guidelines? A framework for improvement. JAMA 1999;282:1458-65. Tanios MA, de Wit M, Epstein SK, et al. Perceived barriers to the use of sedation protocols and daily sedation interruption: a multidisciplinary survey. J Crit Care 2009;24:66-73. Curtis JR, Cook DJ, Wall RJ, et al. Intensive care unit quality improvement: a “how-to” guide for the interdisciplinary team. Crit Care Med 2006;34:211-8. Barochia A, Cui X, Vitberg D, et al. Bundled care for septic shock: an analysis of clinical trials. Crit Care Med 2010;38:668-78. Manias E, Street A. Legitimation of nurses' knowledge through policies and protocols in clinical practice. J Adv Nurs 2000;32:1467-75. Pogorzelska M, Larson EL. Assessment of attitudes of intensive care unit staff toward clinical practice guidelines. Dimens Crit Care Nurs 2008;27:30-8. Flynn AV, Sinclair M. Exploring the relationship between nursing protocols and nursing practice in an Irish intensive care unit. Int J Nurs Pract 2005;11:142-9.