Patient Education and Counseling 98 (2015) 257–262
Contents lists available at ScienceDirect
Patient Education and Counseling
journal homepage: www.elsevier.com/locate/pateducou
Short communication
Mind the (knowledge) gap: The effect of a communication instrument
on emergency department patients’ comprehension of and satisfaction
with care
Stefanie Simmons a,*, Brian Sharp b, Jennifer Fowler a, Hope Fowkes c, Patricia Paz-Arabo c,
Mary Kate Dilt-Skaggs d, Bonita Singal a, Thomas Carter e
a
St. Joseph Mercy Hospital, Department of Emergency Medicine, Ann Arbor, USA
University of Wisconsin Hospital, Division of Emergency Medicine, Madison, USA
c
St. Mary Mercy Hospital, Department of Emergency Medicine, Livonia, USA
d
MSN Southern Ohio Medical Center, Department of Emergency Medicine, Portsmouth, USA
e
DO Southern Ohio Medical Center, Department of Emergency Medicine, Portsmouth, USA
b
A R T I C L E I N F O
A B S T R A C T
Article history:
Received 15 November 2013
Received in revised form 25 September 2014
Accepted 19 October 2014
Objectives: We developed a communication instrument to be used in the Emergency Department (ED)
and hypothesized that use of this guide would increase patient comprehension of and satisfaction
with care.
Methods: This multi-site trial enrolled 643 patients in treatment and control groups. Comprehension of
care was assessed by chart review and satisfaction measured via validated survey.
Results: Use of the instrument was not associated with improvements in patient knowledge about their
care, with a mean of 4.6 (95% CI: 4.8–5.8) comprehension defects in the control group and 4.4 (95% CI:
3.9–4.9) in the treatment group. There was no significant effect on patient satisfaction 76.4% versus
76.9%, p = 0.34. Elderly patients in both groups were found to have 1.1 (p < 0.01) more knowledge gaps
than younger patients.
Conclusion: Patients frequently misunderstand medical care in the ED. Comprehension decreases
with increasing age. An isolated communication instrument does not improve satisfaction with or
understanding of the care received.
Practice implications: Providing a structured place for providers and patients to record details of care
does not seem to improve satisfaction with or comprehension of care. Interventions that focus on
communication skills and face time with patients may prove more effective.
ß 2014 Elsevier Ireland Ltd. All rights reserved.
Keywords:
Patient education
Patient–HCP communication
Patient satisfaction
1. Introduction
Effective communication is a critical component of efficient
and safe healthcare and has been demonstrated to correlate with
improved Emergency Department (ED) follow-up, patient compliance and satisfaction [1–4]. The fast pace, high acuity and
interruptions in the ED make communication challenging, leading
to poor patient comprehension of care [5–10]. Multiple approaches
to optimize patient comprehension have been evaluated with
mixed success [3,11–13].
In a previous single site pilot study, we evaluated whether
a customizable written instrument was effective to improve
communication and comprehension [14]. The initial study
demonstrated a non-statistically significant trend toward improved comprehension. This multisite study was undertaken to
evaluate the tool in a larger, more diverse sample of patients.
The hypothesis was that a sufficiently powered sample would
show a significant improvement in patients’ comprehension and
satisfaction with communication.
2. Methods
* Corresponding author at: St. Joseph Mercy Health System, 530 East Huron River
Drive, PO Box 995, Ann Arbor, MI 48106, USA.
Tel.: +1 734 712 3962; fax: +1 734 712 5178.
E-mail addresses: [email protected], [email protected]
(S. Simmons).
http://dx.doi.org/10.1016/j.pec.2014.10.020
0738-3991/ß 2014 Elsevier Ireland Ltd. All rights reserved.
2.1. Study design
The study was conducted at four teaching hospitals in Michigan
and Ohio staffed by Emergency Physicians Medical Group (EPMG),
S. Simmons et al. / Patient Education and Counseling 98 (2015) 257–262
258
an ED staffing and management company. The annual ED census
at these hospitals ranges from 42,000 to 90,000 patient visits
per year.
The ED team was trained on the use of our instrument prior to
the study (see Fig. 1). Study design was a two phase analysis of
patient satisfaction and comprehension with the first 317 patients
enrolled without the use of the instrument (Group 1), and the
last 326 enrolled with use of the instrument (Group 2). The data
analysts and chart reviewers were blinded to which group the
patient belonged.
The instrument was developed in collaboration with focus
groups of patients and was intended for use throughout the
patients’ ED stay. Providers were instructed to record patient data
in the provided fields (see Fig. 2). After the care episode, patients
provided demographic data, satisfaction with communication and
comprehension of care.
2.3. Measure of comprehension
We measured comprehension directly by using the patient
comprehension survey created for this study by the investigators (Fig. 3). It consists of a structured set of questions that
cues the patient to recall the events of their visit, including
diagnoses, treatment given, follow-up recommendations and
reasons to return to the ED. If the patient was admitted to the
hospital, they were instead asked to state reason(s) for hospital
admission.
Two experienced ED nurses compared the comprehension
surveys with a chart review. They rated patient comprehension
by judging whether the patient fully understood (concordant),
partially understood (partially concordant), or did not understand
(discordant). Objective criteria were used when possible, for
example in the fields recording recommended follow up timing
and diagnosis.
2.2. Measure of communication
2.4. Statistical methods
To measure for satisfaction with communication, we used the
Communication Assessment Tool-Team (CAT-T), a validated
instrument to assess patient satisfaction with communication
and satisfaction with the medical team [15,16]. The CAT-T is scored
by the number of items rated as ‘‘excellent’’, given as a percentage
of total items.
My InformER
Today’s Date:____________________
My Room Number: ____________________
My Doctor or Physician’s Assistant:
___________________________________________
My Nurses: ________________________________________________________________________
My Allergies: ______________________________________________________________________
My Plan for Today’s Visit:
We tested the hypothesis that those in the intervention group
have higher total CAT-T scores than those in the control group
using a linear mixed model with hospital included as a random
effect. Since the CAT-T score is highly skewed toward higher
numbers, a transformation was required to comply with the
linearity assumptions of the model. Age, race and gender were
included as covariates.
Whether the communication instrument improved patient
comprehension was tested using linear mixed models. The
dependent variable in one model was the number of discordances;
and in the other model the number of discordances and partial
concordances. The dependent variables were not transformed
because the residuals complied with the model assumptions. The
independent variable was group and we included age, race, gender
and whether the patient was admitted, in the model to control for
possible confounding.
3. Results
Blood tests: _____________________________________________________________________
Urine tests: _____________________________________________________________________
X-rays: __________________________________________________________________________
CT scans: _______________________________________________________________________
Ultrasounds: ___________________________________________________________________
EKGs or other tests: ___________________________________________________________
Medications:
1._______________________________________________________
2._______________________________________________________
3._______________________________________________________
4._______________________________________________________
5._______________________________________________________
3.1. Demographics and enrollment retention
Six hundred and forty-three patients were enrolled in the study
(Tables 1 and 2). There were no significant differences in age,
ethnicity or sex between groups. There were some differences in
patient demographics between sites as shown in Table 3.
3.2. Results of satisfaction survey
Do you think I’ll go home or be admitted to the hospital? _________________________
There was no statistically significant difference between groups
in total CAT-T scores. The median of the total score was 76.4 in the
control group and 76.9 in the intervention group, p = 0.43 (Table 4).
The median of the percent of excellent responses, calculated as
the number of ‘‘excellent’’ responses out of 16 total questions
multiplied by 100, was 81.3% in both groups. The scores on the
CAT-T found in our study are similar to those found during
the validation and subsequent studies of the CAT-T (76.3%
excellent) [15].
My other questions are:
3.3. Results of comprehension testing
Can I eat food when I’m here? ________________________________________________________
Can you estimate how long will I be in the ER? _____________________________________
1. __________________________________________________________________
2. __________________________________________________________________
Fig. 1. PEC_My_InformER.
The intervention had no effect on the total number of
discordances or the total number of discordances plus partial
concordances. The solution for fixed effects from multivariable
mixed regression models is shown in Table 4. The estimated
mean number of discordances was 4.6 (95% CI: 4.1–5.1) for the
control group and 4.4 (95% CI: 3.9–4.9) for the intervention
S. Simmons et al. / Patient Education and Counseling 98 (2015) 257–262
259
Fig. 2. My_Informer_Used_log.
group, p = 0.36. The mean number of discordances plus
partial concordances was 5.3 (95% CI: 4.8–5.8) for the
control group and 5.0 (95% CI: 4.5–5.5) for intervention group,
p = 0.15.
Age, race and whether the patient was admitted were
associated with comprehension level (Table 5). Patients who were
admitted had fewer mean discordances, 3.8 versus 5.2, but they
had fewer questions to answer. Extremely elderly patients (85 or
older) had 1.6 more discordances than patients 24 or younger,
p = 0.01; and 1.7 more discordances than patients 24 to 44,
p < 0.01. Elderly patients (age 65 to 84) had 1.0 more discordances
than those 24 or younger, p < 0.05; and 1.1 more than those
25 to 44, p = <0.01. Caucasians had one less discordance than
African Americans, p < 0.01.
4. Discussion and conclusions
4.1. Discussion
Our instrument failed to show any effect on comprehension of
medical care or satisfaction with communication. We did find
substantial discordances between the medical record and patient’s
recall of ED care, consistent with previous studies [6,20].
One of our sites, which has a heavy nursing presence in the ED
involved in research, used the instrument to structure nursing
education for the patient. There was strong nursing enthusiasm
to continue using the technique after the study.
We found that elderly patients had a higher number of
misunderstandings of their care, and that the effect increased
S. Simmons et al. / Patient Education and Counseling 98 (2015) 257–262
260
Paent Comprehension Assessment
1. What were the reasons for the symptoms that brought you to the ER (Your
ER diagnosis)? Name up to three.
1._______________________________________________________________
2. _______________________________________________________________
3. ________________________________________________________________
2. Did you receive any medicaons in the ER?
Yes or No
If Yes, please name one (try to remember the specific name):
________________________________________
3. Did you receive any xrays, CTs or ultrasounds while in the ER?
Yes or No
If Yes, please name one (try to remember the specific name):
__________________________________________
4. Did you receive any blood, urine or other body fluid tests while in the ER?
Yes or No
If Yes, please name one (try to remember the specific test name):
___________________________________________
Fig. 3. PEC_Comprehension_tool.
Table 1
Enrollment flow.
p
Site 1
Site 2
Site 3
Site 4
with increasing age. This is not surprising and is likely due to
increasing complexity of care, physical difficulties with communication, and declining cognitive powers [17,18].
Patients met all inclusion criteria (n = )
Patients met enrollment criteria but were
not enrolled into study (n=)
296
98
351
153
254
93
116
30
4.2. Limitations
75
7
6
10
198
108
5
19
21
198
49
4
2
38
161
21
0
1
30
86
Breakdown of reasons for non-enrollment
‘‘Didn’t feel up to it’’ (n=)
Discharged before completion (n=)
Questionable mental status (n=)
Other (n=)
Patients enrolled after exclusions (n=)
A sequential enrollment design can cause temporal trend bias. A
randomized contemporaneous design was not chosen because of
concern about contamination of the control group with techniques
learned by providers using the instrument for the intervention
group. The method we used to evaluate comprehension does not
address patient comprehension in the sense that it means
Table 2
Percent rated ‘‘excellent’’ for each item of the CAT-T.
Questions
(1) Greeted me in a way that made me feel comfortable
(2) Treated me with respect
(3) Showed interest in my ideas about my health
(4) Understood my main health concerns
(5) Paid attention to me (looked at me, listened carefully)
(6) Let me talk without interruptions
(7) Gave me as much information as I wanted
(8) Talked in terms I could understand
(9) Checked to be sure I understood everything
(10) Encouraged me to ask questions
(11) Involved me in decisions as much as I wanted
(12) Discussed next steps, including any follow-up plans
(13) Showed care and concern
(14) Spent the right amount of time with me
(15) Front desk treated me with respect
(16) Rate the care provided by your medical team?
*
From multivariable hierarchical models.
Control N = 317
Intervention N = 326
Rated excellent frequency (%)
Rated excellent frequency (%)
190
213
196
200
219
233
207
223
208
180
188
203
219
194
206
196
189 (58.0)
220 (67.5)
183(56.1)
195 (59.8)
214 (65.6)
212 (65.0)
210 (64.4)
225 (69.0)
209 (64.1)
176 (54.0)
185 (56.8)
205 (62.9)
217 (66.6)
193 (59.2)
199 (61.0)
196 (60.1)
(59.9)
(67.2)
(61.8)
(63.1)
(69.1)
(70.4)
(65.3)
(70.4)
(65.6)
(56.8)
(59.3)
(64.0)
(69.1)
(61.2)
(65.0)
(61.8)
p-Value*
0.64
0.82
0.13
0.40
0.32
0.15
0.84
0.67
0.69
0.38
0.35
0.68
0.54
0.59
0.30
0.74
S. Simmons et al. / Patient Education and Counseling 98 (2015) 257–262
261
Table 3
Patient demographic variables by site.
Patient demographics
Site 1 frequency (%)
Age category
24
25–44
45–64
65–84
85
Race
Black or African American
White or Caucasian
Other
Gender
Male
25
43
61
49
12
Site 2 frequency (%)
(13.2)
(22.6)
(32.1)
(25.8)
(6.3)
20
59
56
47
14
Site 3 frequency (%)
(10.2)
(30.1)
(28.6)
(24.0)
(7.1)
22
58
57
21
1
(13.8)
(36.5)
(35.9)
(13.2)
(0.7)
Site 4 frequency (%)
9
32
27
13
4
(10.9)
(37.7)
(31.7)
(15.3)
(4.7)
31 (16.3)
146 (76.8)
13 (6.8)
26 (13.3)
156 (79.6)
14 (4.1)
2 (1.3)
149 (93.7)
9 (5.0)
19 (22.4)
58 (68.2)
8 (9.4)
70 (36.8)
74 (37.8)
53 (33.3)
28 (32.9)
Table 4
Solution for fixed effects from the multivariable mixed effects models.
Variable
Total discordances
Standard error
Estimate
Intercept
Intervention
Control
Age category
24
25–44
45–64
65–84
85
Race
Black or African American
White or Caucasian
Other
Gender
Male
Female
Admitted
Yes
No
Total discordances plus partial discordances
t-Value
Estimate
Standard error
t-Value
p-Value
0.5694
0.1826
8.65
0.91
0.0003
0.3627
5.2387
0.2627
0.5639
0.1825
9.29
1.44
0.0026
0.1506
1.6411
1.7180
1.2672
0.6565
0
0.5078
0.4573
0.4439
0.4536
3.23
3.76
2.85
1.45
0.0013
0.0002
0.0045
0.1483
1.3944
1.4861
1.0800
0.4947
0.5074
0.4569
0.4436
0.4533
2.75
3.25
2.43
1.09
0.0062
0.0012
0.0152
0.2756
0.1292
0.8481
0.4354
0.3606
0.30
2.35
0.7667
0.0190
0.2687
0.6889
0.4351
0.3604
0.62
1.91
0.5372
0.0564
0.1294
0.1904
0.68
0.4970
0.2680
0.1903
1.41
0.1595
1.4573
0.2148
6.78
<0.0001
1.3817
0.2145
6.44
<0.0001
0
0
Table 5
Estimated difference in least square mean total discrepancies from the multivariable mixed effects model.
Comparisons
Difference in LS means (95% CI)*
p-Value*
Intervention versus control
Age categories
24 versus 25–44
24 versus 45–64
24 versus 65–84
24 versus 85
25–44 versus 45–64
25–44 versus 65–84
25–44 versus 85
45–64 versus 65–84
45–64 versus 85
65–84 versus 85
Race categories
Black versus White
Black versus other
White versus other
Male versus female
Admit versus discharge
0.17 ( 0.19 to 0.53)
0.36
*
p-Value
4.9225
0.1663
0
0.08
0.37
0.98
1.64
0.45
1.06
1.72
0.61
1.27
0.66
(
(
(
(
(
(
(
(
(
(
0.77
1.24
1.96
3.03
1.10
1.83
2.97
1.33
2.48
1.90
to
to
to
to
to
to
to
to
to
to
0.93)
0.49)
0.01)
0.25)
0.20)
0.29)
0.47)
0.10)
0.05)
0.58)
0.99
0.76
<0.05
0.01
0.31
<0.01
<0.01
0.13
0.04
0.60
0.98
0.13
0.85
0.13
1.46
(0.30 to 1.65)
( 0.89 to 1.15)
( 1.70 to 0.00)
( 0.25 to 0.50)
(1.04 to 1.88)
<0.01
0.95
0.05
0.50
<0.01
With Tukey adjustment for multiple comparisons except for intervention.
understanding all the implications of having a specific test or
procedure but rather is a test of the patient’s ability to recall these
details.
4.3. Conclusions
Patients frequently misunderstand medical care in the ED
and their post care instructions. A standardized, portable
record that the patient and provider write on during the visit
was not shown to be an adequate method of increasing patient
knowledge about the care during their visit or expected post-ED
care. Future research involving written communication guides
should integrate their use with established systematic, structured communication strategies which may include evaluating
the role of a written tool in guiding bedside sign-out or hourly
rounding.
4.4. Practice implications
(1) Elderly patients have decreased comprehension of care when
compared to younger patients. In this particularly complex
and potentially high-risk group, this finding supports the use
of increasing communication resources when treating and
ultimately discharging elderly patients.
(2) Providing a structured place for providers and patients to
record details of care does not seem to improve satisfaction
with or comprehension of care and should not be regarded
as a sufficient intervention. When used as part of a structured
interaction, such as nurse rounding or physician sign-out, a
written record may prove more effective.
S. Simmons et al. / Patient Education and Counseling 98 (2015) 257–262
262
Funding
Funding for study provided by Blue Cross Blue Shield of
Michigan Foundation (Grant 1773.11).
Conflict of interest statement
Authors have no financial conflicts of interest to report.
IRB
Study approved by IRB for all participating institutions.
Acknowledgements
Thanks to Emergency Physician Medical Group (EPMG) for
providing initial research network establishment opportunities.
Though the four hospitals in our study group belong to different
health care systems, the Emergency Departments at the hospitals
are staffed by EPMG, which facilitated formation of our research
network.
References
[1] Beckman HB, Frankel RM. The effect of physician behavior on the collection of
data. Ann Intern Med 2001;134:1079–85.
[2] Bjorvell H, Stief J. Patients’ perceptions of the health care received in an ED.
Ann Emerg Med 1991;20:734–8.
[3] Mayer T, Cates R, Mastorovich M, Royalty D. Emergency department patient
satisfaction: customer service training improves patient satisfaction and
ratings of physician and nurse skill. J Healthc Manag 1998;43:427–40.
[4] Tamblyn R, Abrahamowicz M, Dauphinee D. Influence of physicians’ management and communication ability on patients’ persistence with antihypertensive medication. Arch Intern Med 2010;170:1064–72.
[5] Crane J. Patient comprehension of doctor–patient communication on discharge from the emergency department. J Emerg Med 1997;1915:1–7.
[6] Engel K, Heisler M, Smith DM, Robinson CH, Forman JH, Ubel PA. Patient
comprehension of emergency department care and instructions: are patients
aware of when they do not understand? Ann Emerg Med 2009;53:454–61.
[7] Logan P, Schwab R. Patient understanding of emergency department discharge
instructions. South Med J 1996;89:770–4.
[8] Roberts J. ED discharge instructions: another chance to help patients and
prove your worth. Emerg Med News 2006;28:17–20.
[9] Spandorfer J, Karras D, Hughes L, Caputo C. Comprehension of discharge
instructions by patients in an urban emergency department. Ann Emerg
Med 1995;25:71–4.
[10] Vashi A, Rhodes K. Sign right here and you’re good to go: a content analysis of
audiotaped emergency department discharge instructions. Ann Emerg Med
2011;51:315–22.
[11] Eisenberg E, Murphy A, Sutcliffe K, Wears R, Schenkel S, Perry S, et al.
Communication in emergency medicine: implications for patient safety.
Commun Monogr 2005;72:390–413.
[12] Boudreaux E, Cruz B, Baumann B. The use of performance improvement
methods to enhance emergency department patient satisfaction in the United
States: a critical review of the literature and suggestions for future research.
Acad Emerg Med 2006;3:795–802.
[13] Tran TP, Schulte WP, Muelleman RL, Wadman MC. Provision of clinically based
information improves patients’ perceived length of stay and satisfaction with
EP. Am J Emerg Med 2002;20:506–9.
[14] Simmons SA, Sharp B, Fowler J, Singal B. Implementation of a novel communication tool and its effect on patient comprehension of care and satisfaction.
Emerg Med J 2013;30:363–70.
[15] Mercer L, Tanbe P, Pang PS, Gisondi MA, Courtney DM, Engel KG, et al. Patient
perspectives on communication with the medical team: pilot study using the
communication assessment tool-team (CAT-T). Patient Educ Couns 2008;73:
220–3.
[16] Engel K, Buckley B, Adams J, et al. Patient understanding of emergency
department discharge instructions: where are knowledge deficits greatest.
Acad Emerg Med 2012;19:E1035–44.
[17] Kalyanaraman M, McQueen D, Wiggins M, et al. Impact of Nurse Integrated
Rounds on Self-Reported Comprehension, Attitudes, and Practices of Nurses
and Resident Physicians in a Pediatric Intensive Care Unit. J Intensive Care Med
2013;29(5):285–91. http://dx.doi.org/10.1177/0885066613486612.
[18] Ferranti D, Makoul G, Forth VE, Rauworth J, Lee J, Williams MV. Assessing
patient perceptions of hospitalist communication skills using the communication assessment tool (CAT). J Hosp Med 2010;9:522–7.
[20] Royall D, Cordes J, Polk M. Executive control and the comprehension of medical
information by elderly retirees. Exp Aging Res 1997;4:301–31.