Professional Documents
Culture Documents
This is the seventh article in a series from the Arizona State University College of Nursing and Health Innovation’s
Center for the Advancement of Evidence-Based Practice. Evidence-based practice (EBP) is a problem-solving approach
to the delivery of health care that integrates the best evidence from studies and patient care data with clinician exper-
tise and patient preferences and values. When delivered in a context of caring and in a supportive organizational
culture, the highest quality of care and best patient outcomes can be achieved.
The purpose of this series is to give nurses the knowledge and skills they need to implement EBP consistently, one
step at a time. Articles will appear every two months to allow you time to incorporate information as you work toward
implementing EBP at your institution. Also, we’ve scheduled “Chat with the Authors” calls every few months to provide
a direct line to the experts to help you resolve questions. See details below.
I
n September’s evidence- they created earlier in this process of an article, their impressions
based practice (EBP) article, when they found and filled in the of it, as well as any tips—such as
Rebecca R., our hypothetical essential elements of the 15 stud- what worked in calling an RRT—
staff nurse, Carlos A., her hospi- ies and projects (see “Critical Ap- that could be used later when
tal’s expert EBP mentor, and Chen praisal of the Evidence: Part I,” they write up their ideas for ini-
M., Rebecca’s nurse colleague, ra- July). Now each takes a stack of tiating an RRT at their hospital, if
pidly critically appraised the 15 the “keeper” studies and system- the evidence points in that direc-
articles they found to answer their atically begins adding to the table tion. Chen remarks that although
clinical question—“In hospital- any remaining data that best re- she thought their initial table con-
ized adults (P), how does a rapid flect the study elements pertain- tained a lot of information, this
response team (I) compared with ing to the group’s clinical question final version is more thorough by
no rapid response team (C) affect (see Table 1; for the entire table far. She appreciates the opportu-
the number of cardiac arrests (O) with all 15 articles, go to http:// nity to go back and confirm her
and unplanned admissions to the links.lww.com/AJN/A17). They original understanding of the
ICU (O) during a three-month had agreed that a “Notes” sec- study essentials.
period (T)?”—and determined tion within the “Appraisal: Worth The team members discuss the
that they were all “keepers.” The to Practice” column would be a evolving patterns as they complete
team now begins the process of good place to record the nuances the table. The three systematic
evaluation and synthesis of the
articles to see what the evidence
says about initiating a rapid re-
Need Help with Evidence-Based Practice? Chat with
sponse team (RRT) in their hos-
pital. Carlos reminds them that the Authors on November 16!
O
evaluation and synthesis are syn- n November 16 at 3 PM EST, join the “Chat with the Au-
ergistic processes and don’t neces- thors” call. It’s your chance to get personal consultation from
sarily happen one after the other. the experts! Dial-in early! U.S. and Canada, dial 1-800-947-5134
Nevertheless, to help them learn, (International, dial 001-574-941-6964). When prompted, enter
he will guide them through the code 121028#.
EBP process one step at a time.
Go to www.ajnonline.com and click on “Podcasts” and then
STARTING THE EVALUATION on “Conversations” to listen to our interview with Ellen Fineout-
Rebecca, Carlos, and Chen begin Overholt and Bernadette Mazurek Melnyk.
to work with the evaluation table
Chan PS, et al. None SR N = 18 out of IV: RRT RRT: was the • Fre- 13/16 studies Weaknesses:
Arch Intern Med Purpose: effect of 143 potential DV1: HMR MD involved? quency reporting team • Potential missed evi-
2010;170(1): RRT on HMR and studies (including DNR, • Relative structure dence with exclusion
18-26 CR excluding DNR, HMR: overall risk of all studies except
• Searched 5 Setting: acute not treated in hospital deaths 7/11 adult those with control
databases from care hospitals; ICU, no HMR (see definition) and 4/5 peds groups
▼
“grey literature” DV2: CR CR: cardio nificant reduc- limited to medical meet-
from MD confer- Average no. and/or pulmo tion in CR ings
ences beds: NR nary arrest; • Only included HMR and
• Included only cardiac arrest CR: CR outcomes
1) RCTs and Attrition: NR calls • In adults, • No cost data
21%–48%
ajnonline.com
McGaughey J, None SR (Cochrane N = 2 studies IV: RRT HMR: OR OR of Aus- Weaknesses:
et al. Cochrane review) DV1: HMR Australia: tralian study, • Didn’t include full body
Database Syst Acute care set- overall hospital 0.98 (95% CI, of evidence
Rev 2007;3: Purpose: effect of tings in Australia mortality with- 0.83–1.16) • Conflicting results of
ajn@wolterskluwer.com
CD005529 RRT on HMR and the UK out DNR retained studies, but no
• Searched 6 OR of UK study, discussion of the impact
databases from Attrition: NR UK: Simplified 0.52 (95% CI, of lower-level evidence
1990–2006 Acute Physiol- 0.32–0.85) • Recommendation “need
• Excluded all but ogy more research”
2 RCTs Score (SAPS) II
death probabil- Conclusion:
ity estimate • Inconclusive
Winters BD, None SR N = 8 studies IV: RRT HMR: overall Risk ratio HMR: Strengths:
et al. Crit Purpose: effect of DV1: HMR death rate •O
bserva- • Provides comparison
Care Med RRT on HMR and Average no. DV2: CR tional studies, across studies for
2007;35(5): CR beds: 500 CR: no. of in- risk ratio for Study lengths (range,
1238-43 • Searched 3 hospital arrests RRT on HMR, 4–82 months)
databases from Attrition: NR 0.87 (95% Sample size (range,
1990 –2005 CI, 0.73– 2,183–199,024)
• Included only 1.04) Criteria for RRT initia-
studies with a •C
luster RCTs, tion (common: respira-
control group risk ratio for tory rate, heart rate,
RRT on HMR, blood pressure, mental
0.76 (95% status change; not all
CI, 0.39– studies, but notewor-
1.48) thy: oxygen saturation,
“worry”)
CR: • Includes ideas about
•O
bserva- future evidence gen-
tional studies, eration (conducting
risk ratio for research)— finding out
RRT on CR, what we don’t know
▼
CI, 0.56– Conclusion:
0.92) • Some support for RRT,
•C
luster RCTs, but not reliable enough
risk ratio for to recommend as stan-
RRT on CR, dard of care
CI = confidence interval; CR = cardiopulmonary arrest or code rates; DNR = do not resuscitate; DV = dependent variable; HMR = hospital-wide mortality rates; ICU = intensive care
unit; IV = independent variable; MD = medical doctor; NR = not reported; OR = odds ratio; Peds = pediatrics; RCT = randomized controlled trial; RR = relative risk; RRT = rapid
response team; SR = systematic review; UK = United Kingdom
45
reviews, which are higher-level Chen in their efforts to appraise as well as a good number of jour-
evidence, seem to have an inher- the MERIT study and comments nals have encouraged their use.
ent bias in that they included only on how well they’re putting the When they review the actual
studies with control groups. In pieces of the evidence puzzle to guidelines, the team notices that
general, these studies weren’t in gether. The nurses are excited they seem to be focused on re-
favor of initiating an RRT. Carlos that they’re able to use their new search; for example, they require
asks Rebecca and Chen whether, knowledge to shed light on the a research question and refer to
now that they’ve appraised all the study. They discuss with Carlos the study of an intervention,
evidence about RRTs, they’re con how the interpretation of the whereas EBP projects have PICOT
fident in their decision to include MERIT study has perhaps con questions and apply evidence to
all the studies and projects (in tributed to a misunderstanding practice. The team discusses that
cluding the lower-level evidence) of the impact of RRTs. these guidelines can be confusing
among the “keepers.” The nurses Comparing the evidence. As to the clinicians authoring the re-
reply with an emphatic affirma- the team enters the lower-level evi ports on their projects. In addition,
tive! They tell Carlos that the proj dence into the evaluation table, they note that there’s no mention
ects and descriptive studies were they note that it’s challenging to of the synthesis of the body of
what brought the issue to life for compare the project reports with evidence that should drive an
them. They realize that the higher- studies that have clearly described evidence-based project. While the
level evidence is somewhat in methodology, measurement, anal SQUIRE Guidelines are a step in
conflict with the lower-level evi- ysis, and findings. Chen remarks the right direction for the future,
dence, but they’re most interested that she wishes researchers and Carlos, Rebecca, and Chen con-
in the conclusions that can be clinicians would write study and clude that, for now, they’ll need
drawn from considering the entire project reports similarly. Although to learn to read these studies as
body of evidence. each of the studies has a process they find them—looking care-
Rebecca and Chen admit they or method determining how it was fully for the details that inform
have issues with the systematic conducted, as well as how out- their clinical question.
reviews, all of which include the comes were measured, data were Once the data have been en-
MERIT study.1-4 In particular, they analyzed, and results interpreted, tered into the table, Carlos sug-
discuss how the authors of the comparing the studies as they’re gests that they take each column,
systematic reviews made sure to currently written adds another one by one, and note the similari-
report the MERIT study’s finding layer of complexity to the eval ties and differences across the
that the RRT had no effect, but uation. Carlos says that while it studies and projects. After they’ve
didn’t emphasize the MERIT study would be great to have studies briefly looked over the columns,
authors’ discussion about how and projects written in a similar for he asks the team which ones they
their study methods may have mat so they’re easier to compare, think they should focus on to an-
influenced the reliability of the that’s unlikely to happen. But he swer their question. Rebecca and
findings (for more, see “Critical tells the team not to lose all hope, Chen choose “Design/Method,”
Appraisal of the Evidence: Part as a format has been developed “Sample/Setting,” “Findings,” and
II,” September). Carlos says that for reporting quality improve- “Appraisal: Worth to Practice”
this is an excellent observation. ment initiatives called the SQUIRE (see Table 1) as the initial ones
He also reminds the team that Guidelines; however, they aren’t to consider. Carlos agrees that
clinicians may read a systematic ideal. The team looks up the guide these are the columns in which
review for the conclusion and lines online (www.squire-statement. they’re most likely to find the
never consider the original stud- org) and finds that the Institute most pertinent information for
ies. He encourages Rebecca and for Healthcare Improvement (IHI) their synthesis.
Adapted with permission from Melnyk BM, Fineout-Overholt E, editors. Evidence-based practice in nursing and healthcare: a guide to best practice.
2nd ed. Philadelphia: Wolters Kluwer Health / Lippincott Williams and Wilkins; 2010.
1 = Chan PS, et al. (2010); 2 = McGaughey J, et al.; 3 = Winters BD, et al.; 4 = Hillman K, et al.; 5 = Sharek PJ, et al.; 6 = Chan PS, et al.
(2009); 7 = DeVita MA, et al.; 8 = Mailey J, et al.; 9 = Dacey MJ, et al.; 10 = McFarlan SJ, Hensley S.; 11 = Offner PJ, et al.; 12 = Bertaut Y,
et al.; 13 = Benson L, et al.; 14 = Hatler C, et al.; 15 = Bader MK, et al.
ajn@wolterskluwer.com AJN ▼ November 2010 ▼ Vol. 110, No. 11 47
Table 3: Effect of the Rapid Response Team on Outcomes
1a 2a 3a 4a 5a 6a 7 8 9 10 11 12 13 14 15
HMR b NE c b NR NE c NE b, d
adult
b
peds
CRO NE NE NE NE c b
NE NE b c b c NE c c
CR b NE b NE b c NE NE NE NE b NE NE
peds
and
adult
UICUA NE NE NE NE NE NE NE b c NE NE NE b
1 = Chan PS, et al. (2010); 2 = McGaughey J, et al.; 3 = Winters BD, et al.; 4 = Hillman K, et al.; 5 = Sharek PJ, et al.;
6 = Chan PS, et al. (2009); 7 = DeVita MA, et al.; 8 = Mailey J, et al.; 9 = Dacey MJ, et al.; 10 = McFarlan SJ, Hensley S.;
11 = Offner PJ, et al.; 12 = Bertaut Y, et al.; 13 = Benson L, et al.; 14 = Hatler C, et al.; 15 = Bader MK, et al.
CR = cardiopulmonary arrest or code rates; CRO = code rates outside the ICU; HMR = hospital-wide mortality rates;
NE = not evaluated; NR = not reported; UICUA = unplanned ICU admissions
a
higher-level evidence; b statistically significant findings; c statistical significance not reported; d non-ICU mortality was
reduced
The EBP team can tell from having level-VI evidence, a study what leads clinicians to act in con
reading the evidence that research and a project, had statistically fidence and apply the evidence (or
ers consider the impact of an RRT significant (less likely to occur by not) to their practice and expect
on hospital-wide mortality rates chance, P < 0.05) reductions in similar findings (outcomes). In
(HMR) as the more important HMR, which increases the reli- terms of making a decision about
outcome; however, the group re ability of the results. whether or not to initiate an RRT,
mains unconvinced that this out- Chen asks, since four level-VI Carlos says that their evidence
come is the best for evaluating reports documented that an RRT stacks up: first, the MERIT study’s
the purpose of an RRT, which, reduces HMR, should they put results are questionable because
according to the IHI, is early in more confidence in findings that of problems with the study meth-
tervention in patients who are occur more than once? Carlos re- ods, and this affects the reliability
unstable or at risk for cardiac or plies that it’s not the number of of the three systematic reviews as
respiratory arrest.16 That said, of studies or projects that determines well as the MERIT study itself;
the 11 studies and projects that the reliability of their findings, but second, the reasonably conducted
evaluated mortality, more than the uniformity and quality of their lower-level studies/projects, with
half found that an RRT reduced it. methods. He recites something he their statistically significant find-
Carlos reminds the group that heard in his Expert EBP Mentor ings, are persuasive. Therefore,
four of those six articles are level-VI program that helped to clarify the team begins to consider the
evidence and that some weren’t the concept of making decisions possibility that initiating an RRT
research. The findings produced based on the evidence: the level may reduce code rates outside the
at this level of evidence are typi- of the evidence (the design) plus ICU (CRO) and may impact non-
cally less reliable than those at the quality of the evidence (the ICU mortality; both are outcomes
higher levels of evidence; how- validity of the methods) equals the they would like to address. The
ever, Carlos notes that two articles strength of the evidence, which is evidence doesn’t provide equally
Respiratory distress Airway threatened RR < 10 or RR < 8 or > 30 RR < 8 or > 28 RR < 10 or > 30
(breaths/min) Respiratory arrest > 30
Unexplained dys- New-onset difficulty Shortness of breath
RR < 5 or > 36
pnea breathing
Change in mental Change in LOC ND Unexplained change Sudden decrease Decreased LOC
status Decrease in Glasgow in LOC with normal
Coma Scale of blood glucose
> 2 points
Tachycardia (beats/ >140 > 130 Unexplained > 130 > 120 > 130
min) for 15 min
Bradycardia (beats/ < 40 < 60 Unexplained < 50 < 40 < 40
min) for 15 min
Blood pressure SBP < 90 SBP < 90 or > Hypotension (unex- SBP > 200 or < 90 SBP < 90
(mmHg) 180 plained)
Concern/worry Serious concern NE Nurse concern about Nurse concern • Uncontrolled pain
about patient about a patient who overall deterioration • Failure to respond to
doesn’t fit the above in patients’ condi- treatment
criteria tion without any of • Unable to obtain prompt
the above criteria assistance for unstable
(p. 2077) patient
4 = Hillman K, et al.; 8 = Mailey J, et al.; 9 = Dacey MJ, et al.; 13 = Benson L, et al.; 15 = Bader MK, et al.
cc = cubic centimeters; CIWA = Clinical Institute Withdrawal Assessment; hr = hour; LOC = level of consciousness; min = minute; mmHg = millimeters
of mercury; ND = not defined; NE = not evaluated; RR = respiratory rate; SBP = systolic blood pressure; SIRS = systemic inflammatory response
syndrome; SpO2= arterial oxygen saturation; UOP = urine output
ajn@wolterskluwer.com AJN ▼ November 2010 ▼ Vol. 110, No. 11 49
Rebecca raises a question that’s of excitement about their project, that an RRT is a valuable inter-
been on her mind. She reminds that their colleagues across all vention to initiate. They decide
them that in the “Appraisal: Worth disciplines have been eager to hear to take the criteria for activating
to Practice” column, teaching was the results of their review of the an RRT from several successful
identified as an important factor evidence. In addition, Carlos says studies/projects and put them
in initiating an RRT and expresses that many resources in their hos- into a synthesis table to better
concern that their hospital is not pital will be available to help them see their major similarities (see
an academic medical center. Chen get started with their project and Table 44, 8, 9, 13, 15). From this com-
reminds her that even though reminds them of their hospital bined list, they choose the criteria
theirs is not a designated teaching administrators’ commitment to for initiating an RRT consult that
hospital with residents on staff support the team. they’ll use in their project (see
24 hours a day, it has a culture Table 5). The team also begins
of teaching that should enhance ACTING ON THE EVIDENCE discussing the ideal make up for
the success of an RRT. She adds As they consider the synthesis their RRT. Again, they go back to
that she’s already hearing a buzz of the evidence, the team agrees the evaluation table and look
Respiratory distress RR < 10 or > 30 breaths/min or unexplained dyspnea or new-onset difficulty breathing
or shortness of breath
Cardiovascular
Sepsis
cc = cubic centimeters; hr = hours; HR = heart rate; LOC = level of consciousness; min = minute; mmHg = millimeters of
mercury; RR = respiratory rate; SBP = systolic blood pressure; SpO2 = arterial oxygen saturation; Temp = temperature;
UOP = urine output; WBC = white blood count
ajn@wolterskluwer.com AJN ▼ November 2010 ▼ Vol. 110, No. 11 51