close

Вход

Забыли?

вход по аккаунту

?

j.jcjd.2018.08.192

код для вставкиСкачать
Accepted Manuscript
Title: Development and Implementation of a DKA Protocol for Adults with
Type 1 and Type 2 Diabetes at a Tertiary Care Multi-Campus Hospital
Author: Medina Mohamed, Angela Assal, Loree Boyle, Edmund Kwok,
Filomena De Sousa, Alan Karovitch, Janine Malcolm
PII:
DOI:
Reference:
S1499-2671(18)30108-4
https://doi.org/10.1016/j.jcjd.2018.08.192
JCJD 1081
To appear in:
Canadian Journal of Diabetes
Received date:
Revised date:
Accepted date:
23-3-2018
9-8-2018
10-8-2018
Please cite this article as: Medina Mohamed, Angela Assal, Loree Boyle, Edmund Kwok,
Filomena De Sousa, Alan Karovitch, Janine Malcolm, Development and Implementation of a
DKA Protocol for Adults with Type 1 and Type 2 Diabetes at a Tertiary Care Multi-Campus
Hospital, Canadian Journal of Diabetes (2018), https://doi.org/10.1016/j.jcjd.2018.08.192.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service
to our customers we are providing this early version of the manuscript. The manuscript will
undergo copyediting, typesetting, and review of the resulting proof before it is published in its
final form. Please note that during the production process errors may be discovered which could
affect the content, and all legal disclaimers that apply to the journal pertain.
1
2
3
4
Development and Implementation of a DKA Protocol for Adults
with Type 1 and Type 2 diabetes at a Tertiary Care Multi-Campus
Hospital
5
Medina Mohamed MD1, 2, Angela Assal MD, FRCPC1,4 , Loree Boyle BN1,2, MD, FRCPC, Edmund
Kwok MD, MHA, MSc, FRCPC1,3, Filomena De Sousa RN, BScN1,2, Alan Karovitch, MD, M.Ed
FRCPC1,2, Janine Malcolm MD, FRCPC1,2
6
7
8
9
10
11
1
The Ottawa Hospital, Ottawa, Canada; 2 Department of Medicine, University of Ottawa, 3Department of
Emergency Medicine, 4University of Toronto
12
Corresponding Author:
13
Janine Malcolm
14
Division of Endocrinology and Metabolism, University of Ottawa
15
1967 Riverside Drive
16
Ottawa, Ontario
17
K1H 7W9
18
613-738-8400 ext 81945
19
jamalcolm@toh.ca
20
21
Key Words: Diabetic Ketoacidosis, Protocol, Order Set, Quality Improvement
22
23
24
25
26
27
28
Key Messages:
1) Implementation of a pre-printed protocol for DKA management supported by ongoing userfeedback and continuous revisions resulted in standardized best practices and improved outcomes
for DKA care
2) A strong implementation plan and ongoing support for end users was critical for the successful
uptake of the DKA protocol.
29
Abstract: 229
30
Paper: 2233
31
Disclosures: None
1
Page 1 of 21
32
33
Abstract:
34
Introduction:
35
Diabetic Ketoacidosis (DKA) is associated with significant morbidity and mortality. Use of standardized
36
protocols for DKA management improves outcomes and is recommended in Diabetes Canada (DC)
37
Clinical Practice Guidelines (CPG). Audits of DKA care at our institution revealed inconsistent
38
management. We developed, piloted and evaluated a standardized DKA protocol adapted into pre-printed
39
order sets for use in the emergency department and acute monitoring area.
40
Methods:
41
The protocol was developed by an expert committee based on the DC CPG, literature review, and
42
environmental survey. A before and after analysis was used. Uptake of the DKA protocol and clinical
43
outcomes were monitored through statistical process control.
44
Results:
45
Patients admitted post-protocol (n=55, mean age 37.9 years (SD 17.5 years), 62% M, 85% type 1
46
diabetes) were compared to those admitted pre-protocol (n=55, mean age 43.3 years (SD 17.5), 53% M,
47
67.2% DM1). Post-implementation, 87% of patients were managed on the protocol. Post-protocol
48
ordering of appropriate laboratory investigations increased, appropriate IV fluid resuscitation improved,
49
continuation of IV insulin until AG closure increased, mean time to AG closure decreased and mean LOS
50
was reduced. Eighty-five percent of nurses and 74% of physicians surveyed felt the protocol improved
51
patient care while 75% of patients rated their DKA management as satisfactory.
52
Conclusions: Successful implementation of a standardized pre-printed protocol for DKA management
53
significantly improved best practices for DKA management and was valued by treating clinicians.
2
Page 2 of 21
54
Introduction:
55
Diabetic Ketoacidosis (DKA) is a life-threatening metabolic complication of diabetes mellitus associated
56
with significant morbidity and mortality1. In Canada, it results in 5000 to 10,000 hospitalizations annually
57
and carries a mortality rate of 4-10%.1
58
Over the last decade, several studies have shown benefit for the use of protocols and clinical pathways to
59
standardize and improve outcomes in DKA management.2,3,4 As such, their use has been recommended as
60
the preferred method of care delivery for DKA. 2,5 However, the presence of guidelines alone may not be
61
adequate to ensure optimal care6. An audit of the Joint British Diabetes Society consensus guideline for
62
DKA management indicated that although adherence to guidelines was high for the initial DKA
63
management, later management and monitoring was insufficient resulting in high rates of hypokalemia
64
and hypoglycemia.7
65
A review of DKA management at our institution revealed inconsistencies in DKA treatment. Out of 24
66
DKA cases reviewed in 2013 only 54% received appropriate fluid resuscitation, 79% had IV insulin
67
continued to the closure of the anion gap, and 67% received all of the recommended laboratory
68
investigations. A quality improvement initiative was undertaken. The objective was to develop,
69
implement and evaluate an evidenced based protocol adapted into pre-printed order sets for the
70
management of DKA. We aimed to achieve 80% compliance with the DKA protocol across our
71
institution by June 2016 and hypothesized that outcomes would improve with protocol use.
72
Methods:
73
Study Setting and Participants:
74
The Ottawa Hospital (TOH) is a multi-centre, bilingual academic teaching hospital in Ottawa, Ontario.
75
There are 1,122 beds and 172,445 emergency room visits per year. Patients with DKA are admitted
76
through the emergency room and managed in the acute monitoring area (AMA) of the internal medicine
77
ward or the ICU depending on severity of presentation at the two inpatient campuses, the Civic Campus
3
Page 3 of 21
78
and the General Campus. Paper orders are written by physicians that are then incorporated into the paper
79
and electronic medical record.
80
Consecutively admitted patients with DKA were identified over two time periods: September 2013 to
81
March 2014 (pre-intervention) and September 2015 to June 2016 (post-intervention). The diagnosis of
82
DKA required the presence of a metabolic acidosis (pH < 7.3), an elevated anion gap greater than 12 and
83
positive serum ketones. Only patients with DKA management initiated at our center were included.
84
Intervention Development and Implementation:
85
The development and implementation phase of the protocol was based on the Ottawa Model of Research
86
Utilization, a framework for adopting innovations that involves 6 steps (setting the stage, assessing
87
barriers and facilitators, selection and monitoring of knowledge translation strategies, monitoring of the
88
adoption, and evaluation of outcomes)8 and the Institute for Health Care Improvement Model for
89
Improvement. A multidisciplinary expert committee consisting of key stakeholders in internal medicine,
90
emergency medicine, endocrinology, nursing, and pharmacy developed a protocol based on a literature
91
review, the 2013 DC CPG on DKA management, consensus expert opinion, and stakeholder feedback.
92
The protocol was adapted into two standardized pre-printed order sets (Appendix 1), for use in the
93
emergency department or the intensive care unit (ICU) or acute monitoring area (AMA). End-user
94
feedback resulted in 2 modifications to the protocol to clarify potassium replacement orders, adjustment
95
to the insulin infusion protocol, and automatic notification to MD for reassessment if pH and AG
96
remained abnormal (Appendix 2).
97
An implementation strategy that included stakeholder engagement, discipline specific education sessions,
98
and an analysis of barriers and facilitators of uptake specific to local areas was used. The education
99
intervention was designed by a multidisciplinary committee composed of an endocrinologist, general
100
internist, advanced practice nurse and resident physician. Discipline specific case based workshops
101
targeted at physicians, resident physicians, nurses, and clerical staff delivered by opinion leaders were
4
Page 4 of 21
102
conducted two months prior to implementation. Feedback was obtained from protocol users through
103
questionnaires and emailed direct communication to project leads.
104
Email reminders to use the protocol occurred if clerical or nursing staff noted the protocol was not used.
105
Non-adopters were approached to discuss their barriers to implementation and coaching was provided as
106
needed.
107
Data Collection:
108
A trained data extractor (MM) abstracted data on patient demographics, medical history, diabetes type
109
and duration, biochemical parameters, medication use, capillary blood glucose measurements, insulin and
110
fluid ordered, fluid received and precipitants of DKA. Protocol use was collected and reviewed monthly.
111
Outcome Measures:
112
Table 1 outlines the clinical outcomes evaluated. Process measures and balance measures were also
113
evaluated. A target of 80% of all DKA orders written on the new DKA order forms was established. This
114
goal recognized that a small proportion of patients would require unique treatment regimen that was not
115
easily supported by the form.
116
Analysis:
117
Descriptive statistics were used to depict the baseline characteristics of patients. Nominal data were
118
analyzed using chi square tests, and continuous data was analyzed using independent T-tests. An α <0.05
119
was considered statistically significant.
120
A before and after analysis was used to evaluate the effect of a hospital wide implementation of a paper
121
based pre-printed DKA protocol on clinical outcomes and processes of care. Uptake of the DKA protocol
122
and clinical outcomes were monitored post-implementation through the use of statistical process control
123
through the development and use of Shewhart charts to track progress and stability of process change.
124
The Microsoft Excel and QI macros were used to analyze data.
5
Page 5 of 21
125
Patient and Caregiver Satisfaction:
126
The first 25 patients in the post-protocol group were contacted by telephone following their discharge to
127
complete a satisfaction survey on their DKA management. Nurses and physicians provided feedback
128
through an electronic survey during protocol implementation.
129
Ethics Review:
130
The protocol was reviewed by the Ottawa Hospital Research Ethics Board.
131
Results:
132
A total of 110 DKA admissions (55 pre-intervention and 55 post-intervention) were evaluated. Male
133
patients accounted for 52% of the pre-intervention and 62% post-intervention populations. Most patients
134
had a history of Type 1 diabetes (67% pre-intervention and 85% post-intervention). The most common
135
precipitants of DKA were infections and insulin omission. Mean admission glucose was significantly
136
higher (35 ± 15.8 mmol/L vs 28.8 ± 13.9 mmol/L, p=0.03) and mean admission pH was significantly
137
lower (7.12 ± 0.17 vs 7.19 ± 0.14, p=0.01) in the pre-intervention group. There was no difference
138
between the pre-intervention and post-intervention groups as to the setting (ICU vs ward) where the
139
protocol was used. One patient in each group was managed in the ICU, while the remainder were
140
managed on the ward.
141
Process outcomes:
142
Post-implementation, 87% of patients admitted with DKA were managed on the protocol (Figure 1).
143
Prior to implementation, 100% of endocrinology residents, attending endocrinologists, internal medicine
144
trainees and attending internists, and 85% of nursing staff were trained on the DKA protocol. In total,
145
125 attending physicians (80 emergency physicians, 25 general internists, 20 endocrinologists), 141
146
resident physicians (85 internal medicine, 6 endocrinology, and 50 emergency medicine trainees) were
147
trained on the protocol. Approximately 300 nurses (230 emergency room nurses and 70 acute monitoring
148
area nurses) received training. Training for physicians involved teaching sessions at departmental
6
Page 6 of 21
149
meetings, quality meetings, academic half day and rounds. Email communication of the protocol with
150
instructions was sent to all involved physicians and residents.
151
Nurses were educated through presentations at orientation, lunch and learns, in-services, education
152
updates, and emailing out the protocol with instructions. An email reminder to users occurred 5 months
153
after implementation when protocol use dropped (Figure 1). When use dropped again at 7 and 8 months
154
post-implementation, project leads contacted resident physicians who were noted to not be using the
155
protocol. Feedback was obtained about barriers to using the form and targeted support and education were
156
provided.
157
Clinical Outcomes:
158
Clinical variables improved significantly (Table 1 and Appendix 3). There were 2 patients in the pre-
159
protocol group who developed hypoglycemia (serum glucose < 4mmol/L) and none in the post-protocol
160
group. Hypokalemia (K+ < 3.3mmol/L) was significantly higher in the pre-protocol group. No patients in
161
either group had serious complications such as shock or thrombosis.
162
Balance Outcomes:
163
Consultation requests to the Endocrinology service and Diabetes Nurse Specialists increased by
164
approximately 30% with implementation of the protocol. Average length of stay (LOS) decreased from
165
4.4 to 3.0 days. During the same time period, the average LOS for acute admissions to general medicine
166
services without a diagnosis of diabetes decreased by 0.7 days while length of stay for patients with
167
diabetes admitted to general medical services had a decrease of 0.4 days. Seventy-five percent of patients
168
rated their DKA management with the protocol as satisfactory. Main patient concerns included the
169
frequency of blood work and pain with IV insertions. Eighty-five percent of nurses and 74% of
170
physicians surveyed felt that the protocol improved patient care.
171
Discussion:
7
Page 7 of 21
172
We successfully designed and implemented a standardized, evidence-based protocol for DKA
173
management at a multi-campus tertiary care centre that lead to safer practices and standardized care
174
regardless of the treating physician’s knowledge or experience in DKA management. Clinical outcomes
175
improved without an increase in either hypoglycemia or hypokalemia. The order set was highly
176
supported by treating clinicians and its use was associated with good patient satisfaction.
177
178
Recognizing the need to improve DKA management, national guidelines have advocated for the use of
179
treatment protocols2. Numerous studies have shown that treatment protocols improve outcomes in
180
hyperglycemic emergencies.9.10 Similar to our findings, Bull et al (2007) found protocol implementation
181
was associated with decreased length of stay, shorter time to anion gap closure without increased
182
hypoglycemia or hypokalemia.10 Hara et al (2013) examined the effect of a mandatory protocol for DKA
183
management in an academic medical centre and found protocol use lead to a shorter time to resolution of
184
acidosis, a decrease in LOS with no difference in hypoglycemia rates.11 Laliberte et al (2017) also found
185
improvement in compliance to DKA guidelines with a DKA order-set, however, they experienced a
186
higher rate of hypoglycemia in the protocol group.12 In contrast to our protocol, which contained a
187
detailed dynamic insulin dosing scale that did not require physician verification for dosage changes, their
188
protocol had limited suggestions for insulin infusion changes that required physician verification which
189
may have resulted in delays in insulin dose changes.
190
191
In order for the benefits of a protocol to be realized, they must be used. Evans et al (2014) found
192
improvement in best practices DKA management through the implementation of a DKA protocol but only
193
for those treated with full fidelity to the DKA protocol.13 A strength of our study is the high rate of usage
194
of our protocol within our institution which is in contrast to most published DKA protocols.
195
Optimal implementation requires a robust knowledge translation strategy, ongoing stakeholder
196
involvement, use of champions and tailoring the intervention to local barriers and facilitators 8, 15, 16 Our
8
Page 8 of 21
197
protocol was developed using a multidisciplinary approach relying on key evidence and a strong
198
implementation plan. The use of the Ottawa Model of Research Utilization was helpful in providing a
199
framework to assess the barriers and supports to implementation and promoted early engagement of the
200
opinion leaders, key stakeholders and adopters.8 The model also promoted real-time monitoring of the
201
intervention with frequent audits and feedback on protocol use. This was one of the most important
202
factors in promoting confidence and engagement of end users of the protocol.
203
Features of the protocol itself may have also contributed to uptake as it facilitated the management of
204
DKA for treating physicians. Suggestions for alternate fluid replacement regimens for patients with heart
205
failure and end-stage renal disease, potassium replacement recommendations and dosing adjustments to
206
insulin rates based on dynamic insulin dosing scales provided the treating clinician with easily accessible
207
evidenced based treatment recommendations.
208
Consultations to the Endocrinology Service and Diabetes Nurse Specialist service increased with the use
209
of our protocol which has also been found in other studies.17 The increased rate of consultations to nurse
210
specialists may be beneficial for improved long-term patient outcomes as their involvement in DKA
211
management has been shown to lower readmission rates through increased education and more frequent
212
follow up.18, 19
213
A major barrier to adoption of the protocol was concern that use of a standardized order set may have a
214
negative effect on the education of medical trainees. It was perceived that less active decision making is
215
required when using a standardized form. In a prospective cohort study that assessed education outcomes
216
in medical students who used manual order writing compared to a standardized order set, there was no
217
difference in total scores between students who used the order set versus those who did not.20 Although
218
education outcomes were not formally assessed in our study, the anecdotal feedback from attending
219
physicians on their perception of the educational impact of the form was positive.
220
Limitations include the retrospective nature of the chart review and relatively small sample size. Some
221
data may have been missing during the chart review due to inconsistent documentation. Patients in the
9
Page 9 of 21
222
pre-protocol group had seemingly more severe DKA. However, the differences were clinically small.
223
Finally, the intervention may have been prone to the Hawthorne effect; however, most corporate
224
interventions to improve glycemic control were already in place during the baseline chart review and did
225
not change during the protocol implementation.
226
Conclusion:
227
We demonstrated that implementation of a pre-printed protocol for DKA management supported by
228
ongoing user-feedback and continuous revisions resulted in standardized best practices, facilitated
229
improved outcomes for DKA care, and was highly supported by treating practitioners. A strong
230
implementation plan and ongoing support for end users was critical for the successful uptake. This
231
approach may be useful for other hospitals with a similar structure for DKA management.
232
Acknowledgements:
233
Joanne Colas: Performance Management at the Ottawa Hospital for data extraction
234
235
236
237
238
239
References:
1. Chiasson JL, Aris-Jilwan N, Belanger R, et al. Diagnosis and treatment of diabetic ketoacidosis
and the hyperglycemic hyperosmolar state. CMAJ. 2003; 168:885-866.
240
241
2. Goguen J, Gilbert J. Canadian Diabetes Association 2013 Clinical Practice Guidelines for the
242
Prevention and Management of Diabetes in Canada: Hyperglycemic emergencies in adults. Can J
243
Diabetes 2013;37(Suppl. 1): S72–6.
244
245
246
247
3. Thuzar M, Malabu U, Tisdell B, et al. Use of a standardized diabetic ketoacidosis management
protocol improved clinical outcomes. Diabetes Res Clin Practice 2014; 104: e8-e11.
4. Waller SL, Delaney S, Strachan MW. Does an integrated care pathway enhance the management
of diabetic ketoacidosis? Diabet Med 2007; 24:359–63.
10
Page 10 of 21
248
5. Kitabchi AE, Umpierrez GE, Murphy MB, Kreisberg RA. Hyperglycemic crises in adult patients
249
with diabetes: a consensus statement from the American Diabetes Association. Diabetes
250
Care. 2006;29(12):2739–2748.
251
252
253
6. Barth J, Misra S, Aakre K, Langlois M, Watine J, Twomey P , Oosterhuis et al. Why are clinical
practice guidelines not followed? Clin Chem Lab Med 2016; 54(7):1133-1139.
7. Castro W, Htike Z, Turner L, Higgins K. Management of diabetic ketoacidosis following
254
implementation of the JBDS guidelines: Where are we and where should we go? Br J Vasc Dis
255
2015; 15:11-16.
256
257
258
8. Graham ID, Logan J. Innovations in knowledge transfer and continuity of care. Canadian Journal
of Nursing Research. 2004; 36(2), 89-103.
9. Beik N, Anger KE, Forni AA, Bawa K, Szumita PM. Evaluation of an institution-
259
wide guideline for hyperglycemic emergencies at a tertiary academic medical center. Annals of
260
Pharmacotherapy 2013; 47(10) 1260 –1265
261
10. Bull SV, Douglas IS, Foster M, Albert RK. Mandatory protocol for treating adult patients with
262
diabetic ketoacidosis decreases intensive care unit and hospital lengths of stay: results of a
263
nonrandomized trial. Crit Care Med. 2007; 35:41-46.
264
265
266
11. Hara JS, Rahbar AJ, Jeffres MN, Izuora KE. Impact of a hyperglycemic crises protocol. Endocr
Pract. 2013 Nov-Dec;19(6):953-62.
12. Laliberte B, Yeung SYA, Gonzales JP. Impact of diabetic ketoacidosis management in the
267
medical intensive care unit after order set implementation. Int J Pharm Pract. 2017 Jun;25(3):238-
268
243.
269
13. Evans KJ, Thompson J, Spratt SE, Lien LF, Vorderstrasse A.
270
The implementation and evaluation of an evidence-based protocol to treat diabetic ketoacidosis:
271
a quality improvement study. Adv Emerg Nurs J. 2014 Apr-Jun;36(2):189-98.
272
273
14. Yu CH, Sun XH, Nisenbaum R, Halapy H,. Insulin order sets improve glycemic control and
process of care. Am J Medicine. 2012; 12:922-928.
11
Page 11 of 21
274
275
276
15. Singh RK, Perros P, Frier BM. Hospital management of diabetic ketoacidosis: are clinica
guidelines implemented effectively? Diabet Med. 1997; 14:482-486.
16. Gurses AP, Seidl KL, Vaidya V, et al. Systems ambiguity and guideline compliance: a qualitative
277
study of how intensive care units follow evidence-based guidelines to reduce healthcare-
278
associated infections. Qual Saf Health Care.2008;17:351-359.
279
280
281
282
283
284
285
17. Ilag LL, Kronick S, Ernst RD, et al. Impact of a critical pathway on inpatient management of
diabetic ketoacidosis. Diabetes Res Clin Pract. 2003; 62:23-32.
18. Devalia B. Adherance to protocol during the acute management of diabetic ketoacidosis:
would specialist involvement lead to better outcomes? Int J Clin Pract. 2010 Oct;64(11):1580-2.
19. Levetan CS, Passaro MD, Jablonski KA, Ratner RA. Effect of physician specialty on outcomes in
diabetic ketoacidosis. Diabetes Care 1999; 22(11): 1790–5.
20. Lee Y, Mourad O, Panisko D, Sargeant R, Cavalcanti RB. Evaluation of standardized doctor’s
286
orders as an educational tool for undergraduate medial students: a prospective cohort study.BMD
287
Medical Education 2013, 13:97.
288
289
290
291
292
12
Page 12 of 21
293
294
295
296
297
298
299
300
301
302
303
Appendix 1. Pre-printed DKA order set for use in the Emergency Departments of a multi-campus tertiary
care center
13
Page 13 of 21
304
305
Appendix 2: PDSA cycles
306
307
308
D
309
DATA
310
311
D
312
A
P
S
D
S
A
P
S
D
Cycle 3A: Adjusted orders
implemented
Cycle 2C: Orders modified based on
feedback to address minimum and
maximum IV insulin rates
S
P A
A
P
S
A
Cycle 2B: Adjusted orders implemented
P
D
Cycle 2A: Modified Orders to address potassium and fluid management
changes based on feedback
Cycle 1C: Identify barriers/changes required
Cycle 1B:Test Pre-Printed Orders on Pilot units
Cycle 1A: Develop DKA Insulin Orders
14
Page 14 of 21
313
314
Appendix 3: Clinical Outcomes:
A) Appropriate Fluid Management:
315
316
317
318
319
15
Page 15 of 21
320
321
322
323
324
325
B) Appropriate Laboratory Investigations:
326
327
328
16
Page 16 of 21
329
330
331
332
333
334
335
336
337
338
C) IV Insulin Continued to Anion Gap Closure
339
17
Page 17 of 21
340
341
342
343
18
Page 18 of 21
344
Figure 1: Protocol Uptake
345
346
347
348
349
350
351
352
19
Page 19 of 21
353
20
Page 20 of 21
354
Table 1: Clinical Outcomes for the Pre-implementation and Post-Implementation Groups
Parameter
Pre-Implementation
Post-Implementation
P value
Proportion with appropriate fluid
management
33.3%
93%
<0.01
Proportion with appropriate laboratory
investigations
60.0%
91.0%
<0.01
Proportion with IV insulin continued until
anion gap closure
76.0%
100%
<0.01
10.5
6.0
<0.01
Proportion with hypokalemia
18.1%
5.4%
0.04
Proportion with hypoglycemia
3.6%
0%
0.15
Mean time (hours) to anion gap closure
355
356
357
21
Page 21 of 21
Документ
Категория
Без категории
Просмотров
1
Размер файла
1 802 Кб
Теги
jcjd, 192, 2018
1/--страниц
Пожаловаться на содержимое документа