- Email: [email protected]
What is the most effective way to communicate results after endoscopy?
ORIGINAL ARTICLE: Clinical Endoscopy
What is the most effective way to communicate results after endoscopy? David T. Rubin, MD, Alex Ulitsky, MD, Jason Poston, MD, Ryan Day, AB, Dezheng Huo, PhD Chicago, Illinois, USA
Background: The growing demand for endoscopy associated with colorectal cancer screening has resulted in busier endoscopy units and an increase in the practice of open-access endoscopy, in which patients are referred for procedures without prior consultation by the gastroenterologist, all of which may result in less-informed patients. Objective: We sought to determine whether providing patients with a written copy of their standard endoscopy report at the conclusion of their procedure enhanced recall of the findings and recommendations. Design: Eighty consecutive outpatients who presented to 3 endoscopists were randomized to receive the results of their upper or lower endoscopy via standard verbal report (VR) or by standard VR followed by receipt of a computer-generated endoscopy report (VR þ WR) from the Olympus ImageManager report generator. The endoscopist communicated the VR after a standard postprocedure recovery period of 30 to 60 minutes and routinely discussed all findings and recommendations as mentioned in the WR. The endoscopist was blinded as to whether the patient subsequently received the WR. Recall of the endoscopic procedure was assessed by using a piloted 11-question survey instrument to be filled out 3 days after the procedure. Results were calculated by using the Fisher exact and Wilcoxon rank sum tests. Patients: Referral for endoscopy from University of Chicago physicians. Results: Seventy-eight of 80 patients (98%) approached about the study agreed to participate. The response rate was 77%. Patients in the VR þ WR group overall had a greater composite score than patients in the VR group (8.9/10 vs 7.7/10, P ! .01). Patients in the VR þ WR group were also significantly more likely to recall the recommendations for therapy or follow-up (72% vs 42%, P ! .01) and the name of the endoscopist (97% vs 74%, P ! .05). Limitations: Patients with an education beyond the 10th-grade level were not formally accessed in this study. Because of this, we could not evaluate whether differences in educational attainment affected patient understanding of endoscopy procedure details and findings. Conclusions: A computer-generated endoscopy report (WR) significantly improved patient recall of endoscopic procedure information compared with a VR alone. Despite this, patients were unable to recall 28% of recommendations. Additional study to determine if such enhanced physician-patient communication improves patient satisfaction or follow-up, and whether more specific patient-directed results further improve recall needs to occur. (Gastrointest Endosc 2007;66:108-12.)
Copyright ª 2007 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2006.12.056
because of advances in endoscopic therapies. Because the demand for endoscopic procedures has, in many areas, exceeded the available supply of endoscopists, there has been a resulting increase in open-access endoscopy (OAE), in which endoscopic procedures are performed without prior office consultation with the performing endoscopist. OAE has increased further because of a lack of Medicare and third-party reimbursement for office consultation before screening colonoscopies.
108 GASTROINTESTINAL ENDOSCOPY Volume 66, No. 1 : 2007
www.giejournal.org
The last several years have seen a significant increase in the demand for endoscopic procedures. This has been, in large part, because of Medicare and other third-party payer reimbursement for screening colonoscopy but also
Rubin et al
The increased workload in the endoscopy units, the use of amnesia-inducing sedation, and the proliferation of OAE may have resulted in patients who are less knowledgeable about the procedures they underwent and the implications of those procedures, as has been suggested by other investigators.1 Previous work demonstrated that written instructions in addition to verbal ones may enhance patient understanding and compliance with treatment recommendations and follow-up,2-5 but 1 study found no benefit to the addition of written instructions.6 This study of physician-patient communication compared patient recollection of endoscopic indications, results, and recommendations with and without a postprocedure written endoscopic report supplementing verbal communication. We sought to determine patient retention of information communicated after outpatient endoscopic procedures and to evaluate the effectiveness of a standard computer-generated endoscopy report (WR) in enhancing this patient knowledge.
PATIENTS AND METHODS Study design and selection of study participants This was a randomized, double-blind prospective study, in which the endoscopist was blinded to the study group assignment of each patient, and the patients were blinded to the purpose of this study. The institutional review board of the University of Chicago approved this study. Eighty consecutive, literate outpatients (having at least a 10th-grade education) who presented to 3 endoscopists over a 1-month period at the University of Chicago Gastrointestinal Procedure Unit were eligible and were asked to participate in the study. Patients who agreed to participate were randomized into 2 groups. One group received the results of their endoscopy via a standard verbal report (VR) from the performing endoscopist, whereas the other group received the same VR followed by a WR (VR þ WR) from the Olympus ImageManager (Olympus America Corp, Melville, NY) report generator. The WR is a standard, menu-driven endoscopy report with the option of customizing text. It routinely contains the name of the referring physician, the endoscopist, and the nurses; indications for the procedure; medications administered; findings; impressions; and recommendations. For this study, the participating endoscopists were instructed to write the impressions and recommendations in layperson’s terms, defined as understandable at a 10th-grade level, and the quality of these reports was reviewed by the study coordinator to assure consistent information and level of comprehension. The endoscopist communicated the VR to each of the subjects after a standard postprocedure 30- to 60-minute recovery period and routinely discussed all findings and recommendations as mentioned in the WR, but the www.giejournal.org
What is the most effective way to communicate results after endoscopy?
Capsule Summary What is already known on this topic d
Because of endoscopists’ increased workload, use of amnesia-inducing sedation, and the proliferation of open-access endoscopy, patients undergoing endoscopy may be less knowledgeable about the procedure and its implications.
What this study adds to our knowledge d
In a survey of 78 patients undergoing endoscopy, a computer-generated report significantly improved patient recall of information compared to a verbal report alone.
endoscopist did not know whether the patient was to receive the WR. Per our procedure unit protocol, family members or friends who provided the patient a ride home were present for the VR discussion. The research coordinator used a computer-derived random-number generator to determine whether a patient was to receive the supplemental WR or not and, if appropriate, provided it to the patient after the endoscopist left the recovery area. All subjects received a written survey to assess their understanding of the procedure they had just undergone and were asked to complete it 3 days after the procedure and return it in a postage-paid envelope. This survey was designed to access facts about the type of procedure being performed, types of physicians, and findings and recommendations. The survey also included an assessment of whether the patient understood the level of training of the physician who performed the endoscopic procedure, because we believed that this information would be useful in future reporting mechanisms and informed consent. Each patient also received a telephone call from the research assistant as a reminder to fill out the survey. The survey questions can be found in Table 1.
Analysis The returned surveys were graded, by using the official endoscopy report, by an investigator who was blinded to the subject’s study group assignment. The primary ‘‘composite score’’ for each patient was the number of correct survey responses of 10 (question no. 1 was not counted in the analysis, because it was not meant to assess recall), and the secondary ‘‘score’’ was the number of correct answers to individual survey items. A subgroup analysis of patients who were referred as open access was performed. The statistical analysis was performed by using the Student t test, the c2 test, the Fisher exact test, and the 2-sample Wilcoxon rank sum test as appropriate. To control for potential imbalance in sedation medication, linear regression models were used to compare ‘‘composite score’’ between the 2 study groups. Volume 66, No. 1 : 2007 GASTROINTESTINAL ENDOSCOPY 109
What is the most effective way to communicate results after endoscopy?
TABLE 1. Survey questions
Rubin et al
TABLE 3. Indications for the procedures (no. subjects)
Did you feel that the results of your endoscopy were communicated to you adequately?
Screening
VR alone
VR D WR
P value
19
27
.08
Which procedure was performed (colonoscopy, flexible sigmoidoscopy, EGD, barium enema, other, don’t know)
Diarrhea
1
3
.37
Who performed your procedure?
Constipation
1
1
1.00
What type of doctor performed your procedure? (attending physician, fellow, resident, student)
Abdominal pain
2
4
.02
10
3
.003
Who recommended that the procedure be performeddname of physician.
Anemia
3
1
.33
Inflammatory bowel disease
2
1
.5
Other
2
1
.18
What type of doctor recommended the procedure? (primary care physician, gastroenterologist, oncologist, surgeon, nurse practitioner, other)? What is the reason that the procedure was performed (cancer screening, diarrhea, constipation, low blood count, abdominal pain, inflammatory bowel disease, don’t know)?
GI bleeding
TABLE 4. Findings during procedures (% of all subjects)
Was a biopsy performed? What were the results of your procedure (normal exam, single polyp, multiple polyps, etc, don’t know)? What recommendations were made based upon your procedure (no follow-up, clinic visit, different diagnostic test, surgery, repeat same procedure)? If recommended to repeat your procedure, when (not recommended, 6 months, 1 year, 3 years, 5 years, O5 years)?
TABLE 2. Study population
VR alone
VR D WR
P value
Normal examination
15
20
.22
Hemorrhoids
45
35
.17
Diverticular disease
30
33
.2
Single polyp
35
23
.14
Multiple polyps
13
23
.2
Cancer
0
3
1.00
Inflammatory bowel disease
3
3
.4
20
17
.2
Other VR alone
VR D WR
Total no. subjects consented
39
39
No. OAE subjects consented
32
30
TABLE 5. Recommendations, all subjects (% of subjects) Returned surveys (all subjects), no. (%) Age (standard deviation), y Female sex, no. (%)
31 (80)
VR alone
VR D WR
P value
0
0
d
Follow-up biopsy
55
55
.22
Follow-up in clinic
35
58
.04
Repeat procedure
65
60
.2
New/change medication
23
10
.18
Referral
13
23
.29
29 (74) None
58.5 (13.5) 20 (65)
57.5 (16.7) 21 (72)
RESULTS Study population Of the 80 consecutive patients approached about participation in the study, 78 (97.5%) agreed to participate; 60 of 78 (77%) returned their surveys. The study population was randomized into 2 groups, which were equally matched with respect to age, sex, whether or not this was an OAE, and survey return rate (Table 2).
with the exception of patients in the VR þ WR group being referred more often for bleeding indications and abdominal pain, and, similarly, this group being recommended for follow-up in the clinic more often (Tables 3 to 5).
Indications, findings, and recommendations
Survey results
The indications for, findings during, and recommendations after the procedures were similar in both groups,
In the overall analysis, the VR þ WR group had a statistically higher composite score, as well as better recall of
110 GASTROINTESTINAL ENDOSCOPY Volume 66, No. 1 : 2007
www.giejournal.org
Rubin et al
What is the most effective way to communicate results after endoscopy?
TABLE 6. Questionnaire results (% correct of 10 questions) All subjects VR alone
OAE subjects
VR D WR
P value
VR alone
VR D WR
P value .6
1
Were results communicated in adequate fashion?
97
86
.187
96
91
2
Which of the following procedures was performed?
90
100
.238
92
100
.489
3
Who performed your procedure?
74
97
.026
67
95
.023
4
What type of doctor performed your procedure?
81
90
.47
75
86
.463
5
Who recommended that the procedure be performed?
87
97
.35
83
95
.349
6
What type of doctor recommended?
84
90
.7
79
86
.702
7
What was the reason for the procedure?
88
98
.056
85
100
.025
8
Was a biopsy performed?
74
93
.081
71
95
.049
9
What were the results?
67
75
.429
63
76
.286
10
What recommendations were made?
42
72
.003
36
68
.005
11
When is a repeat procedure recommended?
84
86
83
82
Composite score
7.7
recommendations and the endoscopist’s name than the VR group (Table 5). In a subgroup analysis of only the OAE subjects, the VR þ WR group also had a significantly higher composite score than the VR group, 8.8 of 10 correct responses versus 7.3 of 10, P !.01. The subjects who were OAE and received VR þ WR were significantly more likely to recall the indications for their procedure (100% vs 85%, P ! .05), recommendations for therapy or follow-up (68% vs 36%, P ! .01), whether a biopsy was performed (95% vs 71%, P ! .05), and the name of the endoscopist (95% vs 67%, P ! .05) (Table 6). These differences were not noted in the non-OAE group, although there was insufficient power to assess this difference (there were only 14 patients in the non-OAE group).
8.9
1 .002
7.3
8.8
1.00 .002
In this randomized and blinded study, we examined the effects of WRs on patients’ understanding of their GI procedure, both in the OAE setting, as well as the traditional gastroenterologist-referred practice. We found that receiving the WR in addition to the VR from the endoscopist significantly increased the overall number of correct responses on the postprocedure survey, indicating that a WR enhanced the subjects’ overall understanding of the procedure they underwent and its implications in their care. The significant differences in the VR versus VR þ WR groups were also present in the OAE subset alone.
There are several studies that also demonstrated the effectiveness of written instructions. Raynor et al4 showed that 83% of the patients who received computer-generated reminder charts with their drug regimens upon hospital discharge were able to correctly describe their drug regimens, as opposed to 47% of those who did not receive the charts, with a significant increase in compliance scores also reported in the treatment group. Crichton et al5 also reported the highest recall of drug-regimen information in patients who received both verbal and written instructions, as opposed to verbal instructions alone. There have also been studies that demonstrate a positive effect of written instructions on compliance. For example, in a peripheral vascular disease (PVD) screening study using the ‘‘Legs for Life’’ program, founded by the Society of Interventional Radiology, it was found that providing printed instruction cards to patients at moderate to high risk of PVD increased their compliance with follow-up care.3 Blinder et al2 found that the addition of simple written instructions to verbal ones after oral surgery significantly increased patient compliance with the postoperative regimen of care. Although our study and these other findings are encouraging and support the utility of providing patients with endoscopy reports, in our overall analysis, the patients in the VR þ WR group could correctly identify postprocedure recommendations only 72% of the time and the results 75% of the time. These numbers were surprising to us, because the patients in the VR þ WR group presumably
www.giejournal.org
Volume 66, No. 1 : 2007 GASTROINTESTINAL ENDOSCOPY 111
DISCUSSION
What is the most effective way to communicate results after endoscopy?
Rubin et al
had a copy of their endoscopy report available when answering the survey questions. They were also concerning, considering that, for the patients, the results and recommendations are the most important outcomes of GI studies. There are several studies that demonstrate the ineffectiveness of written instructions in increasing patient recall or compliance. Isaacman et al7 found that when parents of children treated for otitis media and discharged from an emergency department received written discharge instructions in addition to the standard verbal ones, it did not significantly improve their knowledge of the child’s illness or recall of instructions, although the overall satisfaction with the visit may have been improved. Zahr et al6 found that providing mothers of pediatric patients with written instructions in addition to verbal ones did not improve their adherence to dietary and follow-up recommendations stated in the instructions. Why might WRs not be uniformly effective in enhancing patient understanding and compliance? One possible explanation is that the patients are simply not reading them. In a study of veterans at a Veterans Affairs hospital, Hayes and Buffum8 found that a substantial portion of the patients did not remember to read their postendoscopy instructions, despite being given a colorful wrist band stating ‘‘GI Procedure: Read Instructions’’ to wear after their study. In that study, as in our GI laboratory, patients undergoing procedures are sedated with benzodiazepines, most commonly midazolam, which has a well-known effect of inducing dense, although temporary, anterograde amnesia for explicit memories. Although in our study, a 30- to 60-minute recovery period was used before verbal instructions were provided, there exists evidence that the memory effects of midazolam may last 90 or more minutes, thus making it difficult for patients to recall verbal instructions or, in some cases, to remember to read written ones.9 Another problem may be related to the possibility of a gap between the literacy and educational levels of the patients and the reading level of the written material. Jolly10 found that a patient’s ability to understand written instructions was correlated to his or her educational level and was not related to age, whereas Spandorfer11 brought up the possibility of a difference in the reading level of printed instructions and the average reading level of the patients, thus making them more reliant on verbal instructions. Although, in our study, the WRs were done at a 10th-grade reading level, it is possible that some patients, although literate, could not fully comprehend the material. Even though all patients who participated in this study had obtained at least a 10th-grade education, we cannot exclude the possibility that differences in educational attainment between study patients may have affected their understanding of endoscopic results, because additional educational attainment was not formally accessed in this study. The presence of family and/or friends during the postprocedure discussion likely did not influence our results, although this was not formally assessed.
This study did not access the amount of communication between the unsedated family or friend and the patient, which may be a critical link in conveying information to recovering patients. The patients who received a WR did not convey any additional questions or concerns about the report or their endoscopic procedure. This suggests that either the WR was an effective aid in communicating information to patients or that the patients did not understand the report enough to ask questions. In summary, we found that recall of endoscopic findings and recommendations was enhanced if a written endoscopy report is provided in addition to a verbal one, especially in the OAE environment. The fact that this was still inadequate for some information and some patients urges us to develop better methods for communicating with our patients in these settings.
112 GASTROINTESTINAL ENDOSCOPY Volume 66, No. 1 : 2007
www.giejournal.org
DISCLOSURE None of the authors have any disclosures to make. REFERENCES 1. Staff DM, Saelian K, Rochling F. Does open access endoscopy close the door to an adequately informed patient? Gastrointest Endosc 2000;52: 212-5. 2. Blinder D, Rotenberg L, Peleg M. Patient compliance to instructions after oral surgical procedures. Int J Oral Maxillofac Surg 2001;30:216-9. 3. Savader SJ, Ehrman KO, Porter DJ. The Legs For Life screening for peripheral vascular disease: results of a prospective study designed to improve patient compliance with physician recommendations. J Vasc Interv Radiol 2001;12:1149-55. 4. Raynor DK, Booth TG, Blenkinsopp A. Effects of computer generated reminder charts on patients’ compliance with drug regimens. Br Med J 1993;306:1158-61. 5. Crichton EF, Smith DL, Demanuele F. Patient recall of medication information. Drug Intell Clin Pharm 1978;12:591-9. 6. Zahr LK, Yazigi A, Armenian H. The effect of education and written material on compliance of pediatric clients. Int J Nurs Stud 1989;26: 213-20. 7. Isaacman DJ, Purvis K, Gyuro J. Standardized instructions: do they improve communication of discharge information from the emergency department? Pediatrics 2002;89:1204-8. 8. Hayes A, Buffum M. Educating patients after conscious sedation for gastrointestinal procedures. Gastroenterol Nurs 2001;24:54-7. 9. Langlois S, Kreeft JH, Chouinard G. Midazolam: kinetics and effects on memory, sensorium and haemodynamics. Br J Clin Pharmacol 1987;23: 273-8. 10. Jolly BT. Discharge instructions. Ann Emerg Med 1995;26:443-6. 11. Spandorfer JM. Patient comprehension of written discharge instructions. Ann Emerg Med 1996;25:71-4.
Received February 16, 2006. Accepted December 26, 2006. Current affiliations: Departments of Medicine (D.T.R., A.U., J.P., R.D.) and Health Studies (D.H.), University of Chicago, Chicago, Illinois, USA. Presented at the American College of Gastroenterology Annual Scientific Meeting, October 2002, Seattle, Washington (Am J Gastroenterol 2002; 97[Suppl]:886). Reprint requests: David T. Rubin, MD, Department of Medicine, University of Chicago, 5841 S Maryland Ave, MC 4080, Chicago, IL 60637.
What is the most effective way to communicate results after endoscopy? David T. Rubin, MD, Alex Ulitsky, MD, Jason Poston, MD, Ryan Day, AB, Dezheng Huo, PhD Chicago, Illinois, USA
Background: The growing demand for endoscopy associated with colorectal cancer screening has resulted in busier endoscopy units and an increase in the practice of open-access endoscopy, in which patients are referred for procedures without prior consultation by the gastroenterologist, all of which may result in less-informed patients. Objective: We sought to determine whether providing patients with a written copy of their standard endoscopy report at the conclusion of their procedure enhanced recall of the findings and recommendations. Design: Eighty consecutive outpatients who presented to 3 endoscopists were randomized to receive the results of their upper or lower endoscopy via standard verbal report (VR) or by standard VR followed by receipt of a computer-generated endoscopy report (VR þ WR) from the Olympus ImageManager report generator. The endoscopist communicated the VR after a standard postprocedure recovery period of 30 to 60 minutes and routinely discussed all findings and recommendations as mentioned in the WR. The endoscopist was blinded as to whether the patient subsequently received the WR. Recall of the endoscopic procedure was assessed by using a piloted 11-question survey instrument to be filled out 3 days after the procedure. Results were calculated by using the Fisher exact and Wilcoxon rank sum tests. Patients: Referral for endoscopy from University of Chicago physicians. Results: Seventy-eight of 80 patients (98%) approached about the study agreed to participate. The response rate was 77%. Patients in the VR þ WR group overall had a greater composite score than patients in the VR group (8.9/10 vs 7.7/10, P ! .01). Patients in the VR þ WR group were also significantly more likely to recall the recommendations for therapy or follow-up (72% vs 42%, P ! .01) and the name of the endoscopist (97% vs 74%, P ! .05). Limitations: Patients with an education beyond the 10th-grade level were not formally accessed in this study. Because of this, we could not evaluate whether differences in educational attainment affected patient understanding of endoscopy procedure details and findings. Conclusions: A computer-generated endoscopy report (WR) significantly improved patient recall of endoscopic procedure information compared with a VR alone. Despite this, patients were unable to recall 28% of recommendations. Additional study to determine if such enhanced physician-patient communication improves patient satisfaction or follow-up, and whether more specific patient-directed results further improve recall needs to occur. (Gastrointest Endosc 2007;66:108-12.)
Copyright ª 2007 by the American Society for Gastrointestinal Endoscopy 0016-5107/$32.00 doi:10.1016/j.gie.2006.12.056
because of advances in endoscopic therapies. Because the demand for endoscopic procedures has, in many areas, exceeded the available supply of endoscopists, there has been a resulting increase in open-access endoscopy (OAE), in which endoscopic procedures are performed without prior office consultation with the performing endoscopist. OAE has increased further because of a lack of Medicare and third-party reimbursement for office consultation before screening colonoscopies.
108 GASTROINTESTINAL ENDOSCOPY Volume 66, No. 1 : 2007
www.giejournal.org
The last several years have seen a significant increase in the demand for endoscopic procedures. This has been, in large part, because of Medicare and other third-party payer reimbursement for screening colonoscopy but also
Rubin et al
The increased workload in the endoscopy units, the use of amnesia-inducing sedation, and the proliferation of OAE may have resulted in patients who are less knowledgeable about the procedures they underwent and the implications of those procedures, as has been suggested by other investigators.1 Previous work demonstrated that written instructions in addition to verbal ones may enhance patient understanding and compliance with treatment recommendations and follow-up,2-5 but 1 study found no benefit to the addition of written instructions.6 This study of physician-patient communication compared patient recollection of endoscopic indications, results, and recommendations with and without a postprocedure written endoscopic report supplementing verbal communication. We sought to determine patient retention of information communicated after outpatient endoscopic procedures and to evaluate the effectiveness of a standard computer-generated endoscopy report (WR) in enhancing this patient knowledge.
PATIENTS AND METHODS Study design and selection of study participants This was a randomized, double-blind prospective study, in which the endoscopist was blinded to the study group assignment of each patient, and the patients were blinded to the purpose of this study. The institutional review board of the University of Chicago approved this study. Eighty consecutive, literate outpatients (having at least a 10th-grade education) who presented to 3 endoscopists over a 1-month period at the University of Chicago Gastrointestinal Procedure Unit were eligible and were asked to participate in the study. Patients who agreed to participate were randomized into 2 groups. One group received the results of their endoscopy via a standard verbal report (VR) from the performing endoscopist, whereas the other group received the same VR followed by a WR (VR þ WR) from the Olympus ImageManager (Olympus America Corp, Melville, NY) report generator. The WR is a standard, menu-driven endoscopy report with the option of customizing text. It routinely contains the name of the referring physician, the endoscopist, and the nurses; indications for the procedure; medications administered; findings; impressions; and recommendations. For this study, the participating endoscopists were instructed to write the impressions and recommendations in layperson’s terms, defined as understandable at a 10th-grade level, and the quality of these reports was reviewed by the study coordinator to assure consistent information and level of comprehension. The endoscopist communicated the VR to each of the subjects after a standard postprocedure 30- to 60-minute recovery period and routinely discussed all findings and recommendations as mentioned in the WR, but the www.giejournal.org
What is the most effective way to communicate results after endoscopy?
Capsule Summary What is already known on this topic d
Because of endoscopists’ increased workload, use of amnesia-inducing sedation, and the proliferation of open-access endoscopy, patients undergoing endoscopy may be less knowledgeable about the procedure and its implications.
What this study adds to our knowledge d
In a survey of 78 patients undergoing endoscopy, a computer-generated report significantly improved patient recall of information compared to a verbal report alone.
endoscopist did not know whether the patient was to receive the WR. Per our procedure unit protocol, family members or friends who provided the patient a ride home were present for the VR discussion. The research coordinator used a computer-derived random-number generator to determine whether a patient was to receive the supplemental WR or not and, if appropriate, provided it to the patient after the endoscopist left the recovery area. All subjects received a written survey to assess their understanding of the procedure they had just undergone and were asked to complete it 3 days after the procedure and return it in a postage-paid envelope. This survey was designed to access facts about the type of procedure being performed, types of physicians, and findings and recommendations. The survey also included an assessment of whether the patient understood the level of training of the physician who performed the endoscopic procedure, because we believed that this information would be useful in future reporting mechanisms and informed consent. Each patient also received a telephone call from the research assistant as a reminder to fill out the survey. The survey questions can be found in Table 1.
Analysis The returned surveys were graded, by using the official endoscopy report, by an investigator who was blinded to the subject’s study group assignment. The primary ‘‘composite score’’ for each patient was the number of correct survey responses of 10 (question no. 1 was not counted in the analysis, because it was not meant to assess recall), and the secondary ‘‘score’’ was the number of correct answers to individual survey items. A subgroup analysis of patients who were referred as open access was performed. The statistical analysis was performed by using the Student t test, the c2 test, the Fisher exact test, and the 2-sample Wilcoxon rank sum test as appropriate. To control for potential imbalance in sedation medication, linear regression models were used to compare ‘‘composite score’’ between the 2 study groups. Volume 66, No. 1 : 2007 GASTROINTESTINAL ENDOSCOPY 109
What is the most effective way to communicate results after endoscopy?
TABLE 1. Survey questions
Rubin et al
TABLE 3. Indications for the procedures (no. subjects)
Did you feel that the results of your endoscopy were communicated to you adequately?
Screening
VR alone
VR D WR
P value
19
27
.08
Which procedure was performed (colonoscopy, flexible sigmoidoscopy, EGD, barium enema, other, don’t know)
Diarrhea
1
3
.37
Who performed your procedure?
Constipation
1
1
1.00
What type of doctor performed your procedure? (attending physician, fellow, resident, student)
Abdominal pain
2
4
.02
10
3
.003
Who recommended that the procedure be performeddname of physician.
Anemia
3
1
.33
Inflammatory bowel disease
2
1
.5
Other
2
1
.18
What type of doctor recommended the procedure? (primary care physician, gastroenterologist, oncologist, surgeon, nurse practitioner, other)? What is the reason that the procedure was performed (cancer screening, diarrhea, constipation, low blood count, abdominal pain, inflammatory bowel disease, don’t know)?
GI bleeding
TABLE 4. Findings during procedures (% of all subjects)
Was a biopsy performed? What were the results of your procedure (normal exam, single polyp, multiple polyps, etc, don’t know)? What recommendations were made based upon your procedure (no follow-up, clinic visit, different diagnostic test, surgery, repeat same procedure)? If recommended to repeat your procedure, when (not recommended, 6 months, 1 year, 3 years, 5 years, O5 years)?
TABLE 2. Study population
VR alone
VR D WR
P value
Normal examination
15
20
.22
Hemorrhoids
45
35
.17
Diverticular disease
30
33
.2
Single polyp
35
23
.14
Multiple polyps
13
23
.2
Cancer
0
3
1.00
Inflammatory bowel disease
3
3
.4
20
17
.2
Other VR alone
VR D WR
Total no. subjects consented
39
39
No. OAE subjects consented
32
30
TABLE 5. Recommendations, all subjects (% of subjects) Returned surveys (all subjects), no. (%) Age (standard deviation), y Female sex, no. (%)
31 (80)
VR alone
VR D WR
P value
0
0
d
Follow-up biopsy
55
55
.22
Follow-up in clinic
35
58
.04
Repeat procedure
65
60
.2
New/change medication
23
10
.18
Referral
13
23
.29
29 (74) None
58.5 (13.5) 20 (65)
57.5 (16.7) 21 (72)
RESULTS Study population Of the 80 consecutive patients approached about participation in the study, 78 (97.5%) agreed to participate; 60 of 78 (77%) returned their surveys. The study population was randomized into 2 groups, which were equally matched with respect to age, sex, whether or not this was an OAE, and survey return rate (Table 2).
with the exception of patients in the VR þ WR group being referred more often for bleeding indications and abdominal pain, and, similarly, this group being recommended for follow-up in the clinic more often (Tables 3 to 5).
Indications, findings, and recommendations
Survey results
The indications for, findings during, and recommendations after the procedures were similar in both groups,
In the overall analysis, the VR þ WR group had a statistically higher composite score, as well as better recall of
110 GASTROINTESTINAL ENDOSCOPY Volume 66, No. 1 : 2007
www.giejournal.org
Rubin et al
What is the most effective way to communicate results after endoscopy?
TABLE 6. Questionnaire results (% correct of 10 questions) All subjects VR alone
OAE subjects
VR D WR
P value
VR alone
VR D WR
P value .6
1
Were results communicated in adequate fashion?
97
86
.187
96
91
2
Which of the following procedures was performed?
90
100
.238
92
100
.489
3
Who performed your procedure?
74
97
.026
67
95
.023
4
What type of doctor performed your procedure?
81
90
.47
75
86
.463
5
Who recommended that the procedure be performed?
87
97
.35
83
95
.349
6
What type of doctor recommended?
84
90
.7
79
86
.702
7
What was the reason for the procedure?
88
98
.056
85
100
.025
8
Was a biopsy performed?
74
93
.081
71
95
.049
9
What were the results?
67
75
.429
63
76
.286
10
What recommendations were made?
42
72
.003
36
68
.005
11
When is a repeat procedure recommended?
84
86
83
82
Composite score
7.7
recommendations and the endoscopist’s name than the VR group (Table 5). In a subgroup analysis of only the OAE subjects, the VR þ WR group also had a significantly higher composite score than the VR group, 8.8 of 10 correct responses versus 7.3 of 10, P !.01. The subjects who were OAE and received VR þ WR were significantly more likely to recall the indications for their procedure (100% vs 85%, P ! .05), recommendations for therapy or follow-up (68% vs 36%, P ! .01), whether a biopsy was performed (95% vs 71%, P ! .05), and the name of the endoscopist (95% vs 67%, P ! .05) (Table 6). These differences were not noted in the non-OAE group, although there was insufficient power to assess this difference (there were only 14 patients in the non-OAE group).
8.9
1 .002
7.3
8.8
1.00 .002
In this randomized and blinded study, we examined the effects of WRs on patients’ understanding of their GI procedure, both in the OAE setting, as well as the traditional gastroenterologist-referred practice. We found that receiving the WR in addition to the VR from the endoscopist significantly increased the overall number of correct responses on the postprocedure survey, indicating that a WR enhanced the subjects’ overall understanding of the procedure they underwent and its implications in their care. The significant differences in the VR versus VR þ WR groups were also present in the OAE subset alone.
There are several studies that also demonstrated the effectiveness of written instructions. Raynor et al4 showed that 83% of the patients who received computer-generated reminder charts with their drug regimens upon hospital discharge were able to correctly describe their drug regimens, as opposed to 47% of those who did not receive the charts, with a significant increase in compliance scores also reported in the treatment group. Crichton et al5 also reported the highest recall of drug-regimen information in patients who received both verbal and written instructions, as opposed to verbal instructions alone. There have also been studies that demonstrate a positive effect of written instructions on compliance. For example, in a peripheral vascular disease (PVD) screening study using the ‘‘Legs for Life’’ program, founded by the Society of Interventional Radiology, it was found that providing printed instruction cards to patients at moderate to high risk of PVD increased their compliance with follow-up care.3 Blinder et al2 found that the addition of simple written instructions to verbal ones after oral surgery significantly increased patient compliance with the postoperative regimen of care. Although our study and these other findings are encouraging and support the utility of providing patients with endoscopy reports, in our overall analysis, the patients in the VR þ WR group could correctly identify postprocedure recommendations only 72% of the time and the results 75% of the time. These numbers were surprising to us, because the patients in the VR þ WR group presumably
www.giejournal.org
Volume 66, No. 1 : 2007 GASTROINTESTINAL ENDOSCOPY 111
DISCUSSION
What is the most effective way to communicate results after endoscopy?
Rubin et al
had a copy of their endoscopy report available when answering the survey questions. They were also concerning, considering that, for the patients, the results and recommendations are the most important outcomes of GI studies. There are several studies that demonstrate the ineffectiveness of written instructions in increasing patient recall or compliance. Isaacman et al7 found that when parents of children treated for otitis media and discharged from an emergency department received written discharge instructions in addition to the standard verbal ones, it did not significantly improve their knowledge of the child’s illness or recall of instructions, although the overall satisfaction with the visit may have been improved. Zahr et al6 found that providing mothers of pediatric patients with written instructions in addition to verbal ones did not improve their adherence to dietary and follow-up recommendations stated in the instructions. Why might WRs not be uniformly effective in enhancing patient understanding and compliance? One possible explanation is that the patients are simply not reading them. In a study of veterans at a Veterans Affairs hospital, Hayes and Buffum8 found that a substantial portion of the patients did not remember to read their postendoscopy instructions, despite being given a colorful wrist band stating ‘‘GI Procedure: Read Instructions’’ to wear after their study. In that study, as in our GI laboratory, patients undergoing procedures are sedated with benzodiazepines, most commonly midazolam, which has a well-known effect of inducing dense, although temporary, anterograde amnesia for explicit memories. Although in our study, a 30- to 60-minute recovery period was used before verbal instructions were provided, there exists evidence that the memory effects of midazolam may last 90 or more minutes, thus making it difficult for patients to recall verbal instructions or, in some cases, to remember to read written ones.9 Another problem may be related to the possibility of a gap between the literacy and educational levels of the patients and the reading level of the written material. Jolly10 found that a patient’s ability to understand written instructions was correlated to his or her educational level and was not related to age, whereas Spandorfer11 brought up the possibility of a difference in the reading level of printed instructions and the average reading level of the patients, thus making them more reliant on verbal instructions. Although, in our study, the WRs were done at a 10th-grade reading level, it is possible that some patients, although literate, could not fully comprehend the material. Even though all patients who participated in this study had obtained at least a 10th-grade education, we cannot exclude the possibility that differences in educational attainment between study patients may have affected their understanding of endoscopic results, because additional educational attainment was not formally accessed in this study. The presence of family and/or friends during the postprocedure discussion likely did not influence our results, although this was not formally assessed.
This study did not access the amount of communication between the unsedated family or friend and the patient, which may be a critical link in conveying information to recovering patients. The patients who received a WR did not convey any additional questions or concerns about the report or their endoscopic procedure. This suggests that either the WR was an effective aid in communicating information to patients or that the patients did not understand the report enough to ask questions. In summary, we found that recall of endoscopic findings and recommendations was enhanced if a written endoscopy report is provided in addition to a verbal one, especially in the OAE environment. The fact that this was still inadequate for some information and some patients urges us to develop better methods for communicating with our patients in these settings.
112 GASTROINTESTINAL ENDOSCOPY Volume 66, No. 1 : 2007
www.giejournal.org
DISCLOSURE None of the authors have any disclosures to make. REFERENCES 1. Staff DM, Saelian K, Rochling F. Does open access endoscopy close the door to an adequately informed patient? Gastrointest Endosc 2000;52: 212-5. 2. Blinder D, Rotenberg L, Peleg M. Patient compliance to instructions after oral surgical procedures. Int J Oral Maxillofac Surg 2001;30:216-9. 3. Savader SJ, Ehrman KO, Porter DJ. The Legs For Life screening for peripheral vascular disease: results of a prospective study designed to improve patient compliance with physician recommendations. J Vasc Interv Radiol 2001;12:1149-55. 4. Raynor DK, Booth TG, Blenkinsopp A. Effects of computer generated reminder charts on patients’ compliance with drug regimens. Br Med J 1993;306:1158-61. 5. Crichton EF, Smith DL, Demanuele F. Patient recall of medication information. Drug Intell Clin Pharm 1978;12:591-9. 6. Zahr LK, Yazigi A, Armenian H. The effect of education and written material on compliance of pediatric clients. Int J Nurs Stud 1989;26: 213-20. 7. Isaacman DJ, Purvis K, Gyuro J. Standardized instructions: do they improve communication of discharge information from the emergency department? Pediatrics 2002;89:1204-8. 8. Hayes A, Buffum M. Educating patients after conscious sedation for gastrointestinal procedures. Gastroenterol Nurs 2001;24:54-7. 9. Langlois S, Kreeft JH, Chouinard G. Midazolam: kinetics and effects on memory, sensorium and haemodynamics. Br J Clin Pharmacol 1987;23: 273-8. 10. Jolly BT. Discharge instructions. Ann Emerg Med 1995;26:443-6. 11. Spandorfer JM. Patient comprehension of written discharge instructions. Ann Emerg Med 1996;25:71-4.
Received February 16, 2006. Accepted December 26, 2006. Current affiliations: Departments of Medicine (D.T.R., A.U., J.P., R.D.) and Health Studies (D.H.), University of Chicago, Chicago, Illinois, USA. Presented at the American College of Gastroenterology Annual Scientific Meeting, October 2002, Seattle, Washington (Am J Gastroenterol 2002; 97[Suppl]:886). Reprint requests: David T. Rubin, MD, Department of Medicine, University of Chicago, 5841 S Maryland Ave, MC 4080, Chicago, IL 60637.