Knowledge In Translation Journal of Parenteral and

Enteral Nutrition
Volume 34 Number 6

Understanding Adherence to November 2010 616-624

© 2010 American Society for

Guidelines in the Intensive Care Unit:

Parenteral and Enteral Nutrition
10.1177/0148607110361904
http://jpen.sagepub.com

Development of a Comprehensive hosted at

http://online.sagepub.com

Framework
Naomi E. Cahill, RD, MSc1,2; Jeanette Suurdt, RN, MSc3;
Hélène Ouellette-Kuntz, RN, MSc1; and
Daren K. Heyland, MD, MSc, FRCPC1,2,4
Financial disclosure: The study was unfunded. NEC received personnel support from the Canadian Institutes for Health Research
(CIHR) Training Program in Digestive Diseases and currently holds a CIHR Fellowship in Knowledge Translation.

Background: Clinical practice guidelines (CPGs) have been hailed the developed framework were characteristics of the CPGs, the
as a useful method of translating evidence into practice. Several implementation process, institutional factors, provider intent, and
CPGs have been published that provide recommendations for the clinical condition of the patient. These key themes encapsu-
feeding patients in the intensive care unit (ICU). Despite a rigor- late numerous itemized factors that contribute to guideline adher-
ous development process and active dissemination of these guide- ence either as barriers or enablers. Conclusions: Adherence to
lines, their impact on nutrition practice has been modest. The nutrition CPGs is determined by a complex interaction of multi-
purpose of this study was to develop a comprehensive framework ple factors that act as barriers or enablers. The comprehensive
for understanding adherence to nutrition CPGs in the critical framework for adherence to CPGs in the ICU attempts to eluci-
care setting. Methods: Multiple case studies were completed at 4 date this process and provides a useful template for future
Canadian ICUs. Semistructured interviews were conducted with research. Future quality improvement initiatives should assess
7 key informants at each ICU site who were asked about their local barriers to change and design interventions to overcome
perceptions and attitudes toward guidelines in general and the these barriers. (JPEN J Parenter Enteral Nutr. 2010;34:616-624)
Canadian Critical Care Nutrition CPGs specifically. Interview
transcripts and related documents were analyzed qualitatively Keywords:   nutrition therapy; critical care; guideline
using a framework approach. Results: The 5 key components of implementation; guideline adherence; knowledge translation

E
vidence-based practice has been described as the making.3 Clinical practice guidelines (CPGs) are “system-
“conscientious, explicit, and judicious use of cur- atically developed statements to assist practitioner and
rent best evidence in making decisions about the patient decisions about appropriate health care for spe-
care of individual patients.”1 Busy critical care practition- cific clinical circumstances”4 and have been hailed as a
ers are faced with the challenge of managing the rapid method for facilitating this translation of evidence into
proliferation of new information. Systematic literature practice. Several CPGs have been published that provide
reviews that critically evaluate and integrate evidence are recommendations regarding the delivery of nutrition ther-
a useful method for remaining updated2 but fail to pro- apy in the intensive care unit (ICU).5-7 These guidelines
vide a method for applying these data in everyday decision aim to assist practitioners in making appropriate deci-
sions with regard to feeding their critically ill patients.
From the 1Department of Community Health and Epidemiology, Knowledge translation or Implementation science is
Queen’s University, Kingston, Ontario, Canada; 2Clinical
Evaluation Research Unit, Kingston General Hospital, Kingston, the scientific study of methods to promote the systematic
Ontario, Canada; 3School of Nursing, Queen’s University, uptake of clinical research findings and other evidence-
Kingston, Ontario, Canada; and 4Department of Medicine, based practices into routine practice and is an emerging
Queen’s University, Kingston, Ontario, Canada. area of interest. The few studies of non-nutrition guide-
Received for publication June 2, 2009; accepted for publication line implementation that have been conducted in the
August 30, 2009. ICU have demonstrated that these strategies have the
Address correspondence to: Daren K. Heyland, MD, MSc, potential to improve the processes8 and outcomes9 and
FRCPC, Kingston General Hospital, 76 Stuart St, Kingston, reduce the costs10 of caring for critically ill patients.
Ontario, Canada, K7L 2V7; e-mail: dkh2@queensu.ca. However, the 3 cluster randomized controlled trials (RCTs)

616
Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014
 Framework for Understanding Guideline Adherence / Jones et al   617

Table 1.   Case Study Site Characteristics

Characteristic Site 1 Site 2 Site 3 Site 4
Hospital type Community Community Academic Community
Hospital size (beds) 306 589 532 400
ICU typea Open Closed Closed Closed
ICU size (beds) 10 26 20 8
% FTE dietitian 0.5 1.0 0.2 0.5

FTE, full-time equivalent; ICU, intensive care unit.

a
Closed refers to an ICU structure in which patients are under the care of an intensivist.

focusing on the implementation of nutrition guidelines of the Canadian Critical Care Nutrition CPGs.11 ICUs
have had mixed results.11-13 To date, our understanding of with different macro-organizational characteristics were
the specific elements of the guidelines—the educational selected to ensure representation from academic and
intervention, the individuals involved, and the study set- community hospitals, open and closed ICUs, rural and
ting that determine the success or failure of implementa- urban locations, small and large units, and medical and
tion strategies—is limited. Identifying the barriers and surgical patient case mix, thus capturing maximum varia-
enablers to guideline implementation and subsequent tion and enabling full exploration of the study question
adherence may aid in designing more effective interven- (Table 1). Semistructured interviews were conducted with
tions by tailoring implementation strategies to local cir- 7 key informants at each site (ICU medical director, nurse
cumstances. manager, clinical nurse educator, dietitian, 2 physicians,
The objective of this study was to develop a comprehen- and 1 bedside staff nurse). These key informants were
sive framework for understanding the barriers and enablers selected to represent different roles, responsibilities, and
to adherence of CPGs in the critical care setting. Although levels of clinical experience within the ICU.
we expect the framework to be applicable to guidelines in During the interview, the key informants were asked
general in the ICU, we used the Canadian Critical Care about their perceptions and attitudes to CPGs and factors
Nutrition CPGs as an illustrative example of an ICU guide- that act as barriers or enablers to adherence to recommen-
line to enable us to probe both general and specific issues dations of guidelines in general and the Canadian Critical
related to guideline adherence. The Canadian Critical Care Care Nutrition CPGs specifically. The interview guide was
Nutrition guidelines, published in 2003 and most recently informed by review of the literature and peer feedback
updated in 2009, consist of 17 recommendations pertaining obtained through a piloting process. Data were also col-
to nutrition practices in the ICU.5,14 lected on the demographic characteristics of each key
informant. In addition, the content of documents pertain-
ing to the local implementation of guidelines, such as
Materials and Methods policy manuals, preprinted orders, bedside protocols, and
algorithms were obtained. Interview transcripts and rele-
Multiple Case Studies vant documents were analyzed qualitatively using a frame-
work approach.17 This strategy was particularly well suited
Multiple case studies were conducted at 4 ICU sites across to our objective, as it focused the analysis on an existing
Canada. The case study is a research strategy that focuses knowledge–attitude–behavior framework developed by
on understanding the dynamics present within specific set- Cabana et al18 in a review of 76 studies that assessed the
tings—in this instance, ICUs—and to date has been most potential barriers to physician adherence to CPGs. Other
commonly used in the fields of social sciences and busi- investigators have also used this framework to guide the
ness management. A case study design was adopted to development of a structured questionnaire, focus group,
provide an in-depth evaluation of the study objective and and key informant interview questions,19-22 and we believe
to explore less tangible aspects of guideline adherence, that its utilization as a starting point for our analysis was
thus identifying causal influences and interaction effects a pertinent and efficient methodology. Data analysis was
that are difficult to measure using conventional statistical managed by QSR NVivo 7 (QSR International Pty Inc.,
techniques.15 As such, case studies are particularly helpful Victoria, Australia).
in generating hypotheses and theories in emerging fields of
inquiry such as guideline implementation. The methods
Revision of the Framework
and results of the multiple case studies have been described
previously and are summarized briefly here.16 Following analysis of the first case ICU site, the results were
Case study sites were selected by purposeful sampling applied to the knowledge–attitude–behavior framework by
from ICU sites that were randomized to the active arm in the investigator (NEC), and revisions to the framework were
a cluster RCT comparing active to passive dissemination made. This process was repeated independently by a second

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014

618   Journal of Parenteral and Enteral Nutrition / Vol. 34, No. 6, November 2010

CLINICAL
PRACTICE
ADHERENCE
GUIDELINE

Guideline Patient Characteristics

Characteristics

Implementation Process Institutional Provider Intent

Characteristics

Provider Characteristics
Hospital and ICU Structure

Knowledge Attitudes
Hospital Processes

Resources Outcome
Familiarity Agreement
Expectancy

ICU Culture
Awareness Motivation Self-efficacy

Ovals = Theme; Boxes = Factors; Italics = New themes/factors.

Figure 1.   Framework for adherence to clinical practice guidelines in the intensive care unit (ICU).

researcher (JS), with high agreement achieved. Two addi- from publication of guidelines to the final behavior of
tional experienced researchers were consulted to ensure adherence to guidelines,23 and this primary relationship is
accuracy of the revisions and to resolve any discrepancies indicated by shaded boxes. The characteristics of the
(DKH and HOK). Following completion of analyses for each CPGs, implementation process, and institutional, indi-
of the ICU sites, the framework was again revised by NEC, vidual provider, and patient characteristics were identi-
with consultation and feedback from the other 3 investiga- fied as key themes associated with adherence, as visualized
tors. Finally, key informants were sent the revised framework by the oval shapes. These 5 key themes encapsulate
for further comment and feedback. numerous factors that contribute to adherence through
Institutional ethics approval was obtained for this acting as a barrier or enabler for this behavior, which are
study from the Health Sciences Research Ethics Board at represented in the boxes. Italicized text is used to indicate
Queen’s University and all participating sites that required new themes and factors that had not previously been
approval. Written informed consent was obtained from included in the Cabana et al18 knowledge–attitude–
each key informant before being interviewed. behavior framework; a dashed-lined box (---------) repre-
sents how provider characteristics modify individuals’
knowledge, attitudes, and intended behavior; and shaded
Results text indicates factors specific to the Canadian Critical
Care Nutrition CPGs. It was felt that the scope of these
factors could not be reflected by the schema alone, and
Revision of the Framework
therefore additional explanatory tables were developed to
The developed framework for describing adherence to support interpretation of the schema (Table 2). All barri-
CPGs in the ICU is shown in Figure 1. The structure of ers and enablers reported by key informants were included
the framework reflects the temporal sequence that occurs in the explanatory tables.

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014

 Framework for Understanding Guideline Adherence / Jones et al   619

Table 2.   Framework for Adherence to Nutrition Clinical Practice Guidelines in the Intensive Care Unit
Theme Factor Barrier Enabler
CPG characteristics Outdated Evidence based
Vague or complex statements Respected developer
User-friendly format
Action oriented
Implementation process Lack of availability of entire ICU Multiple approaches
team to attend meetings, Support of clinical educator and
educational sessions etc medical director
No dedicated individual willing to Tailored to specific needs of
“champion” the guidelines individual
Time commitment to develop and Reminders (eg, checklist)
implement educational strategies Protocols
Restricted access to computers Academic detailing (eg, one-on-
Displacement of posters and one education)
pamphlets over time Educational sessions
Audit and feedback
E-mail/Web-based tools
Opinion leader (eg, the dietitian)
Institutional characteristics
Hospital and ICU Community hospital Large hospital and/or ICU
structure Open structure Closed structure
Rural location Critical care residency program
Small hospital and/or ICU Dedicated, stable workforce
Lack of geographical consolidation
Hospital processes Long, slow administrative process Support for evidence-based
Disconnect between priorities of practice
management and clinical Efficient, flexible administrative
personnel process
Organizational constraints on
practice
Resources for Shortage of staff Adequate resources available (eg,
implementation Limited budget level of staffing, equipment,
Lack of appropriate equipment/ budget for implementation
materials activities)
Lack of access to specialist services New, contemporary facilities
Prevailing culture of No cohesive multidisciplinary team Established multidisciplinary
ICU structure team
No multidisciplinary daily rounds Leadership support
Unresolved conflict or Collaborative decision making
disagreements among ICU team Patient-centered approach
members Formal/informal mentorship
Reliance on written communication Group learning
(eg, Cardex, paper notes) Respect for expertise of each ICU
Leadership not physically present team member
on unit Innovation, embracing change
Poor communication Informal, open communication
Positive work environment
Provider intent
Provider characteristics
Professional roles Circle of influence in nursing staff Attending physician responsible
and allied healthcare for patient care
professionals (eg, dietitian) Nurse manager and ICU medical
dependent on support of director accountable for
physician and leadership team management of ICU
Critical care expertise Junior, novice staff Senior, experienced staff
Locum or casual staff Full time in ICU

(continued)

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014

620   Journal of Parenteral and Enteral Nutrition / Vol. 34, No. 6, November 2010

Table 2. (continued)
Theme Factor Barrier Enabler
Educational
background
Clinical training >10 years Recent graduate
Reliance on expert opinion Training in evidence-based
medicine and critical appraisal
Training in large academic
institution

Personality Type B personalitya Type A personalityb

Uncooperative Team player
Laggard/skeptic Innovator/early adopter
Knowledge
Familiarity CPGs infrequently used because of Part of daily routine
rare clinical condition or narrow Visibility/access to CPGs on ICU
case mix
Awareness Numerous conflicting CPGs on Effective implementation process
same topic
Information overload
Time required to remain updated
Poor dissemination
Attitudes
Outcome expectancy Experience of adverse event from Belief that best for patient
following guideline Positive experience from following
guideline
Self-efficacy Labor intensive Recommendation simple and
Complex procedure quick to perform
Limited circle of influence Procedure frequently performed
successfully
Possession of skills and training
to perform procedure
Procedure within usual scope of
practice
Motivation Inertia of previous practice, Shared team goal to optimize
especially among experienced, patient care
older staff Enforcement or incentive to
Physician resistance, especially perform task
locums, surgeons, and non-ICU
physicians
High cost/work burden associated
with following the guideline
Agreement Paucity of evidence supporting Buy-in of attending physicians
recommendation Buy-in of all members of the ICU
Lack of generalizability to critical team
care and/or specific patient Understanding rationale behind
groups recommendation
Patient characteristics Poor prognosis Medically stable
Other priorities of care Functioning gastrointestinal tract
Unstable clinical condition/
contraindication
Surgical patients
Reconciliation with family
preferences

Shaded portions pertain to the Canadian Critical Care Nutrition Clinical Practice Guidelines. CPG, clinical practice guideline;
ICU, intensive care unit.
a
Type B personality = patient, relaxed, and easygoing.
b
Type A personality = high achieving, competitive, driven, time conscious, and unable to relax.

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014

 Framework for Understanding Guideline Adherence / Jones et al   621

Guideline Characteristics we used the term provider intent instead of physician behav-
ior, as we felt that provider better represented the multidis-
The characteristics of the CPGs were identified as an
ciplinary nature of our study, and intent recognizes that
external factor affecting guideline adherence by key infor-
intention does not always lead to the behavior of adhering to
mants in our study. CPGs supported by high-level evi-
a guideline recommendation.
dence, presented in a user-friendly format, or developed
These 6 factors affecting provider intent to adhere to
by respected researchers enabled adherence. CPGs con-
guidelines were consistently reported by each key informant.
sisting of complex statements that were difficult to inter-
However, the influence of each of these factors differed by
pret or were based on outdated or weak evidence impeded
profession (physicians vs nurses vs dietitians), educational
adherence.
background, critical care experience, and personality. Lack
of agreement, a poor outcome expectancy, and paucity of
Implementation Process evidence supporting the recommendations were stronger
barriers for physicians, whereas practical concerns such as
The educational strategy used to implement the guide-
expertise, workload, and patient comfort were deemed to be
lines emerged as a new theme from the key informant
more important for nursing staff. The familiarity and confi-
interviews. Our findings suggest that multifaceted educa-
dence gained through experience of working in the critical
tional strategies tailored to the specific needs of the ICU
care setting enhanced a provider’s ability to apply CPGs
and the individual practitioner may facilitate adherence
appropriately. In contrast, key informants believed that pro-
to guidelines. Strategies that act as a reminder or aid in
viders who had recently completed their clinical training
incorporating guidelines into routine practice, such as
were more competent at critically appraising the literature
checklists, preprinted orders, protocols, or daily rounds,
compared with providers who were trained before the emer-
were deemed to be successful in facilitating adherence.
gence of evidence-based medicine. Personality type was also
Presentation of the evidence and rationale supporting
found to mold one’s attitudes toward guidelines. Key inform-
recommendations of guidelines through educational
ants felt that the critical care setting attracts individuals who
workshops or one-on-one discussion with an opinion
are assertive and proactive and therefore more willing to
leader were also important enablers for key informants.
embrace change.
Positive reinforcement or identification of problem areas
through a system of audit and feedback was another suc-
cessful motivator. However, lack of adequate resources in Patient Characteristics
terms of time to plan, conduct, and for critical care pro-
viders to attend educational sessions was recognized as a Key informants across all sites and professions identified
barrier to effective guideline implementation. the characteristics of the individual patient as the main
barrier to adherence in scenarios where they agreed with
guidelines and had adequate resources, training, and sup-
Institutional Factors port to perform the recommendation. Key informants felt
Institutional factors related to both the hospital and the that guideline adherence was more difficult in patients
ICU were perceived to have a significant impact on with a poor prognosis or for whom there were other more
adherence to guidelines. Having a closed ICU structure urgent priorities of care.
(ie, ICUs in which patients are under the care of an
intensivist) and being a large teaching hospital enhanced Factors Specific to Adherence of the Canadian
adherence to guidelines, whereas resource constraints
Critical Care Nutrition CPGs
and a slow administrative process were highlighted as bar-
riers. A novel factor that was not present in the original Although all the identified factors remain relevant when
knowledge–attitude–behavior framework was the concept considering adherence to the Canadian Critical Care
that the culture of an organization (eg, leadership style, Nutrition CPGs, the shaded text in Table 2 highlights
decision-making processes, communication, teamwork) factors that specifically affect adherence to these guide-
influences adherence to guidelines. lines and consequently may not apply to other, nonnu-
trition guidelines. These predominantly refer to role of
the dietitian and his or her integration into the multidis-
Provider Intent
ciplinary team. Dietitians felt that commitments exter-
The 6 factors associated with physician adherence to guide- nal to the ICU hindered their ability to be present
lines identified in the original framework by Cabana et al18 during daily rounds and engage in face-to-face discus-
(ie, familiarity, awareness, outcome expectancy, self-efficacy, sions about feeding patients. Lack of time to remain
motivation, and agreement) were also identified as pertinent updated and to provide nutrition education to the ICU
factors by the key informants in our study. In our framework, team was also cited as a barrier to the effectiveness of

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014

622   Journal of Parenteral and Enteral Nutrition / Vol. 34, No. 6, November 2010

dietitians as opinion leaders for the nutrition guidelines. External Factors

Key informants also felt that adherence to the Canadian
Our framework recognizes the need for high-quality
Critical Care Nutrition CPGs was more difficult in sur-
CPGs that incorporate research evidence and local values
gical patients, who frequently experience gastric dys-
in a simple format. Guideline developers should be cogni-
motility and are therefore more difficult to feed
zant of the formatting and presentation of their CPGs and
compared with medical patients.
develop processes for implementing, updating, and evalu-
ating their guidelines. The Canadian Critical Care
Discussion Nutrition CPGs were produced using a systematic devel-
opment process that includes regular review and updates
Achieving adherence to guidelines is profoundly complex, of the recommendations. Thus, they meet international
being determined by the interplay of a number of factors standards for high-quality guidelines.26
that act as barriers or enablers to this behavior. The chal- Our framework infers that provider adherence does
lenges associated with guideline implementation may be not occur in a vacuum or sterile environment but is
greater for nutrition CPGs, as feeding critically ill patients modified by external factors. Hospitals, and within them,
is executed by multiple providers. In this study, we have ICUs, are complex organizational systems whose primary
developed a holistic framework for understanding guide- aim is to deliver clinical care to individual patients. In our
line adherence in the ICU setting. previous surveys of nutrition practices in ICUs, we have
suggested that characteristics of the hospital, ICU, and
patient are associated with performance.27,28
Revised Framework
As in our study, a recent qualitative study conducted
A large number and range of theories and frameworks in 3 Canadian ICUs identified that multifaceted imple-
proposed in the literature have aimed to understand mentation strategies such as education tailored to the
the mechanisms necessary for successful know- specific learning needs of the ICU team, bedside remind-
ledge translation. We chose to base our analysis on ers, and audit and feedback are important enablers of
Cabana et al’s knowledge–attitude–behavior frame- guideline adherence.29 In the same study, Sinuff et al29
work18 because this framework focuses on barriers to also observed that specific aspects of the organizational
guideline adherence, which corresponds to the specific culture (ie, the shared beliefs, attitudes, values, and
objective of our study. Our revised framework adds to norms of behavior among colleagues) were reported to
Cabana et al’s18 by illustrating that efforts to achieve affect adherence, such as informal and formal leadership
guideline adherence should not only focus on individ- support, nonhierarchical multidisciplinary team involve-
ual behavior change but are part of a continuum ment, and open communication. Previous studies have
whereby guideline development and implementation also demonstrated that specific aspects of ICU culture
strategies interact with the practitioner, the patient, such as interdisciplinary collaboration, effective commu-
and their surroundings. Another study also observed nication, and leadership support are associated with bet-
that the knowledge–attitude–behavior framework was ter quality of care.30-32 Manipulating these specific
not a comprehensive representation of all factors asso- organizational characteristics may be a strategy for
ciated with guideline adherence in the ICU. In their improving the utilization of guideline recommendations.
qualitative study evaluating barriers to guidelines on
the use of spine radiography for back pain, Espeland
Provider Intent
and Baerheim24 identified 4 barriers not originally
included in the framework—namely, lack of process We found that the factors that determined a provider’s
expectancy, emotional feelings, improper access to intent to adhere to guidelines were the same for all pro-
healthcare services, and pressure from other providers. fessions. However, the degree of influence of each factor
Patient preferences, lack of reimbursement, and mal- appeared to differ between physicians, nurses, and dieti-
practice liability were 3 factors identified by Cabana et tians. Ricart et al33 also described significant variability
al18 but not highlighted by key informants in our study. between physicians and nurses in reported adherence to
We believe that the preferences of family members sub- CPGs for preventing ventilator-associated pneumonia.
stitute for patient preferences within the ICU, and In their survey, nurses were more likely to identify
although not deemed a significant factor for adherence to patient discomfort and adverse events as barriers to
nutrition CPGs, family preferences have been found to be adherence of CPGs, whereas physicians were more
relevant for other guidelines such as end-of-life care.25 likely to report costs and disagreement with the interpre-
The absence of concern over reimbursement and mal- tation of trials as reasons for nonadherence. Similarly,
practice liability may be a reflection of differences in the Slomka et al34 reported differences in the values and
organization of the critical care service in the Canadian perceptions of physicians and nurses in their study of
and US healthcare systems and should be clarified in the use of CPGs for sedation and neuromuscular block-
future studies. ade in mechanically ventilated patients. In our study,
Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014
 Framework for Understanding Guideline Adherence / Jones et al   623

clinical training, critical care experience, and personality Conclusions

traits of the provider also appeared to modify a provider’s
intent to adhere to a recommendation. Interestingly, Our framework for understanding adherence to nutrition
there appears to be a contradiction in these observa- guidelines in the ICU provides a comprehensive descrip-
tions, whereby recent completion of clinical training and tion of potential barriers and enablers. Critical care provid-
greater time working in the ICU are both associated ers engaged in quality improvement initiatives should use
with higher adherence. It is likely that the former over- this framework to assess barriers and enablers in their local
comes the barrier of lack of knowledge, whereas the setting and design change strategies to address these fac-
latter addresses the enabler self-efficacy. This deserves tors. Increasing adherence to the nutrition CPGs through
further investigation. such a tailored approach may ultimately result in reduced
variation in nutrition practice, increased delivery of energy
Weaknesses and protein, and a potential impact on patient outcomes.

The main limitation of our study and resulting frame- Acknowledgments

work for adherence to guidelines in the ICU is that it
was conducted in Canadian ICUs, and the framework The authors thank Rupinder Dhaliwal, BS, RD, Project
may not be generalizable to other countries with differ- Leader at the Clinical Evaluation Research Unit, Kingston
ent healthcare systems. However, the case studies were General Hospital, Kingston, ON, Canada, for her signifi-
conducted in 4 ICUs with differing organizational cant contribution and guidance in selecting and recruit-
characteristics—a large sample size for this qualitative ing the 4 ICU sites. The authors are grateful to the key
methodology. In addition, the interviews focused on informants at the 4 ICU sites for their hospitality and
both guidelines in general and the Canadian Critical willingness to participate in this study.
Care Nutrition CPGs specifically. We feel that, together,
these aspects increase the applicability of the revised
framework. References
1. Sackett DL, Rosenberg WM, Gray JA, Haynes RB, Richardson
Strengths WS. Evidence based medicine: what it is and what it isn’t. BMJ.
1996;312:71-72.
A strength of our framework is that it is based on an exist- 2. Mulrow CD. Systematic reviews: rationale for systematic reviews.
ing conceptual framework by Cabana et al.18 The utiliza- BMJ. 1994;309:597-599.
tion of a qualitative paradigm to identify the barriers and 3. Hayward RS, Wilson MS, Tunis SR, Bass EB, Guyatt G.
Users’ guides to the medical literature. VIII. How to use clinical
enablers to guideline adherence in the ICU ensured an practice guidelines. Are the recommendations valid? JAMA. 1995;​
in-depth analysis of both the objective and less tangible 274:570-574.
aspects of this issue. The numerous steps employed to 4. Institute of Medicine. Clinical Practice Guidelines: Directions for a
systematically revise the framework provide support for New Program. Washington, DC: National Academy Press; 1990.
the trustworthiness of our findings and ensure that it is a 5. Heyland DK, Dhaliwal R, Drover JW, et al. Canadian clinical prac-
tice guidelines for nutrition support in mechanically ventilated,
comprehensive representation of barriers and facilitators critically ill adult patients. JPEN J Parenter Enteral Nutr.
associated with guideline adherence. In addition, the 2003;27:355-373.
external validity of our framework is supported by other 6. Kreymann KG, Berger MM, Deutz NEP, Hiesmeyer M. ESPEN
studies of barriers to various evidence-based practices in guidelines on enteral nutrition: intensive care. Clin Nutr. 2006;​
the ICU8,29,35-39 and a recent systematic review.40 This 25:210-223.
7. A.S.P.E.N. Board of Directors and the Clinical Guidelines Task
comprehensive review of barriers to optimal clinical prac- Force. Guidelines for the use of parenteral and enteral nutrition in
tices in a wide variety of settings adopted similar catego- adult and pediatric patients. JPEN J Parenter Enteral Nutr.
rization by grouping barriers as cognitive/behavioral, 2002;26(1 suppl):1SA-138SA.
attitudinal, professional, embedded in the guidelines, 8. Sinuff T, Cook D, Randall J, Allen CJ. Evaluation of a practice
patient, resources, and system/process.40 Rather than guideline for noninvasive positive-pressure ventilation for acute
respiratory failure. Chest. 2003;123:2062-2073.
being a weakness, the considerable overlap among factors 9. Burns SM, Earven S, Fisher C, et al. Implementation of an insti-
identified in our study and those identified in other stud- tutional program to improve clinical and financial outcomes of
ies conducted in other settings adds validity to our find- mechanically ventilated patients: one-year outcomes and lessons
ings and supports the assertion that our framework is an learned. Crit Care Med. 2003;31:2752-2763.
accurate and comprehensive representation of factors 10. Pitimana-aree S, Forrest D, Brown G, Anis A, Wang XH, Dodek P.
Implementation of a clinical practice guideline for stress ulcer
associated with guideline adherence in the ICU, recog- prophylaxis increases appropriateness and decreases cost of care.
nizing the contribution of CPG characteristics, the imple- Intensive Care Med. 1998;24:217-223.
mentation process, institutional factors, and patient 11. Jain M, Heyland DK, Dhaliwal R, et al. Dissemination of the
eligibility, in addition to the influence of the intended Canadian clinical practice guidelines for nutrition support: results
behavior of the provider. of a cluster randomized trial. Crit Care Med. 2006;34:2362-2399.

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014

624   Journal of Parenteral and Enteral Nutrition / Vol. 34, No. 6, November 2010

12. Doig GS, Simpson F, Finfer S, et al. Effect of evidence-based feed- 27. Heyland DK, Dhaliwal R, Day A, Jain M, Drover JW. Validation of the
ing guidelines on mortality of critically ill patients: a cluster rand- Canadian clinical practice guidelines for nutrition support in
omized trial. JAMA. 2008;300:2731-2741. mechanically ventilated, critically ill adult patients: results of a pro-
13. Martin CM, Doig GS, Heyland DK, Morrison T, Sibbald WJ; spective observational study. Crit Care Med. 2004;32:2260-2266.
Southwestern Ontario Critical Care Research Network. Multi- 28. Jones NE, Dhaliwal R, Day AG, Ouellette-Kuntz H, Heyland DK.
centre, cluster-randomized clinical trial of algorithms for critical Factors predicting adherence to the Canadian Clinical Practice
care enteral and parenteral therapy (ACCEPT). Can Med Assoc J. Guidelines for nutrition support in mechanically ventilated, criti-
2004;170:197-204. cally ill adult patients. J Crit Care. 2008;23:301-307.
14. Critical Care Nutrition. www.criticalcarenutrition.com. Accessed 29. Sinuff T, Cook D, Giacomini M, Heyland D, Dodek P. Facilitating
February 5, 2009. clinician adherence to guidelines in the intensive care unit: a mul-
15. Yin RK. Case Study Research: Design and Methods. 3rd ed. ticenter, qualitative study. Crit Care Med. 2007;35:2083-2089.
Thousand Oaks, CA: Sage; 2003. 30. Davies HT, Nutley SM, Mannion R. Organisational culture and
16. Jones NE, Suurdt J, Ouellette-Kuntz H, Heyland DK. quality of health care. Qual Health Care. 2000;9:111-119.
Implementation of the Canadian Clinical Practice Guidelines for 31. Baggs JG, Schmitt MA, Mushlin AI, et al. Association between
Nutrition Support: a multiple case study of barriers and enablers. nurse-physician collaboration and patient outcomes in three inten-
Nutr Clin Pract. 2007;22:449-457. sive care units. Crit Care Med. 1999;27:1991-1998.
17. Pope C, Ziebland S, Mays N. Analysing qualitative data. BMJ. 32. Shortell SM, Zimmerman JE, Rousseau DM, et al. The perform-
2000;320:114-116. ance of intensive care units: does good management make a differ-
18. Cabana MD, Rand CS, Powe NR, et al. Why don’t physicians fol- ence? Med Care. 1994;32:508-525.
low clinical practice guidelines? A framework for improvement. 33. Ricart M, Lorente C, Diaz E, Kollef MH, Rello J. Nursing adher-
JAMA. 1999;282:1458-1465. ence with evidence-based guidelines for preventing ventilator-
19. Cabana MD, Kim C. Physician adherence to preventive cardiology associated pneumonia. Crit Care Med. 2003;31:2693-2696.
guidelines for women. Womens Health Issues. 2003;13:142-149. 34. Slomka J, Hoffman-Hogg L, Mion LC, Bair N, Bobek MB, Arroliga
20. Ward MM, Vaughn TE, Uden-Holman T, Doebbeling BN, Clarke AC. Influence of clinicians’ values and perceptions on use of clini-
WR, Woolson RF. Physician knowledge, attitudes and practices cal practice guidelines for sedation and neuromuscular blockade in
regarding a widely implemented guideline. J Eval Clin Pract. patients receiving mechanical ventilation. Am J Crit Care.
2001;8:155-162. 2000;9:412-418.
21. Brand C, Landgren F, Hutchinson A, Jones C, MacGregor L, 35. Sinuff T, Kahnamoui K, Cook D, Giacomini M. Practice guidelines
Campbell D. Clinical practice guidelines: barriers to durability after as multipurpose tools: a qualitative study of noninvasive ventila-
effective early implementation. Intern Med J. 2005;35:162-169. tion. Crit Care Med. 2007;35:776-782.
22. Larson E. A tool to assess barriers to adherence to hand hygiene 36. Rubenfeld GD, Cooper C, Carter G, Thompson T, Hudson LD.
guidelines. Am J Infect Control. 2004;32:48-51. Barriers to providing lung-protective ventilation to patients with
23. Pathman DE, Konrad TR, Freed GL, Freeman VA, Koch GG. The acute lung injury. Crit Care Med. 2004;32:1289-1293.
awareness-to-adherence model of the steps to clinical guideline 37. Nelson JE, Angus DC, Weissfeld LA, et al. End-of-life care for the
compliance: the case of pediatric vaccine recommendations. Med critically ill: A national intensive care unit survey. Crit Care Med.
Care. 1996;34:873-889. 2006;34:2547-2553.
24. Espeland A, Baerheim A. Factors affecting general practitioners’ 38. Cook D, Mead MO, Hand LE, McMullin M. Toward understand-
decisions about plain radiography for back pain: implications for ing evidence uptake: semirecumbency for pneumonia prevention.
classification of guideline barriers—a qualitative study. BMC Crit Care Med. 2002;30:1472-1477.
Health Serv Res. 2003;3:8. 39. Sinuff T, Eva KW, Meade M, Dodek P, Heyland D, Cook D.
25. Truog RD, Cist AF, Brackett SE, et al. Recommendations for end- Clinical practice guidelines in the intensive care unit: a survey of
of-life care in the intensive care unit: the Ethics Committee of the Canadian clinicians’ attitudes. Can J Anaesth. 2007;54:728-736.
Society of Critical Care Medicine. Crit Care Med. 2001;29:2332-2348. 40. Cochrane LJ, Olson CA, Murray S, Dupuis M, Tooman T, Hayes
26. AGREE Collaboration. Development and validation of an interna- S. Gaps between knowing and doing: Understanding and assessing
tional appraisal instrument for assessing the quality of clinical practice the barriers to optimal health care. J Contin Educ Health Prof.
guidelines: the AGREE project. Qual Saf Health Care. 2003;12:18-23. 2007;27:94-102.

Downloaded from pen.sagepub.com at TEXAS SOUTHERN UNIVERSITY on December 14, 2014