Quality in Action
Improving Chemotherapy Ordering Process Jill N. Bryant-Bova
University of Texas Medical Branch, Galveston, TX
Abstract Purpose Chemotherapy is a high-risk medication and is the second most common cause of fatal medication errors. The ordering process can be unsafe and inefficient, putting patients at risk for medication errors. The aim of this project was to decrease the number of chemotherapy order forms with at least one deviation by 50% within 5 months.
Methods A multidisciplinary team identified causes for variance in form completion, deficits in knowledge of ordering processes, and acceptance of incomplete orders by the staff. The Plan, Do, Study, Act improvement methodology evaluated the chemotherapy ordering process and found different types of deviations on order forms. Interventions consisted of educating physicians on entering complete information on orders, instituting same-day laboratory work on the day of the physician’s visit, standardizing laboratory parameters, performing audits of the order forms manually, and educating nurses on not accepting orders with deviations.
Results All order forms were reviewed, and data were collected on different types of deviations. The following deviations were identified: laboratory test results were not being entered into the order form within 7 days, physicians were not providing their name and pager number, and the days of the chemotherapy cycle were missing from the order forms. Before the intervention, 70.1% of the chemotherapy order forms had at least one deviation. After 5 months of interventions, there was a reduction of 19% in the number of order forms with at least one deviation. Follow-up at 6 months and 1 year showed continuing reductions in the number of chemotherapy orders with at least one deviation.
Conclusion Improvement was a result of collaboration between interdisciplinary departments. The original goal was surpassed as a result of educating physicians and staff and standardizing the ordering process. If the number of deviations in chemotherapy order forms is decreased, oncology patients will receive safe, efficient, and quality care.
ASSOCIATED CONTENT Appendices DOI: 10.1200/JOP.2015. 007443 DOI: 10.1200/JOP.2015.007443; published online ahead of print at jop.ascopubs.org on January 5, 2016.
e248
INTRODUCTION Prescribing errors account for the majority of medication errors.1 As the complexity of the drug regimen increases, so does the possibility of errors. Chemotherapy is a high-risk medication, and there are several reasons for increased susceptibility to
errors within the intricate ordering and administration process, such as complex chemotherapy regimens and requirement for dose adjustments.2 Chemotherapy is the second most common cause of fatal medication errors.3 Even if there is no patient harm from the error, correcting the
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
error increases the workload for the staff, and extra effort is required to remedy the patient’s perception of his or her care as suboptimal.4 Deviations in the chemotherapy ordering process were identified in the outpatient infusion centers at the University of Texas Medical Branch (UTMB) in Galveston, TX. There was a lack of consistency in documenting the orders and administering the chemotherapy, which made it difficult to locate previous orders and keep track of the chemotherapy cycle. In addition, the electronic health record allowed chemotherapy orders to be added to the patient’s record simply by using Microsoft Word’s copy and paste function in a Word document that had no standardized format. This led to orders being incomplete or incorrect for the specific patient. Other issues of concern included bypassing the chemotherapy dose checking step at the prescriber level and difficulty updating the original order when changes were required. Extra time was needed to reconcile the errors on the chemotherapy orders. Improving the chemotherapy ordering process is geared toward improving the quality of patient care, increasing staff satisfaction through improved communication and coordination of care between clinicians, and enhancing the overall patient experience in the UTMB outpatient infusion centers. The specific aim of this project was to decrease the number of chemotherapy order forms that deviated from standard chemotherapy ordering policy by 50% by June 2013 (original project end date) within outpatient infusion centers. METHODS This quality improvement project was initiated to examine UTMB’s current process for ordering chemotherapy (Appendix Fig A1, online only). The process begins in the outpatient infusion center where physicians place an order for laboratory work and a patient-specific chemotherapy regimen. The nurses complete a checklist of questions in preparation for administering chemotherapy and submit the orders to the pharmacy. The pharmacy follows up with verification of the order and protocols, prepares the chemotherapy, and ultimately transfers the chemotherapy to the nurses for administration to the patient. The Plan, Do, Study, Act improvement methodology was used to evaluate the chemotherapy ordering process and to analyze types of deviations within the chemotherapy order forms. A fishbone diagram tool was used to organize the various causes and factors that contributed to deviations in the chemotherapy ordering process.
Focus for improvement was narrowed to the following issues: variances in completing the order form by physicians on the basis of their preferences, deficit in staff knowledge regarding acceptance of incomplete orders, and lack of knowledge of the chemotherapy ordering process (Appendix Fig A2, online only). All chemotherapy order forms were reviewed during the initial evaluation and again at the postproject evaluation. Fifteen different types of deviations were discovered on the chemotherapy order forms. A Pareto chart showed that 40% of the deviations were the result of laboratory values not being added to theformwithin7daysbeforestartingthechemotherapyinfusion. The second highest deviation (14%) was physician’s name and pager number not being provided on the chemotherapy order form. Thirteen additional types of deviations made up the remaining 46% (Appendix Fig A3, online only). Based on the deviations found, the following interventions were initiated: • Provide education to physicians who order chemotherapy regarding: • completing the order form and including all pertinent information (ie, name and pager number); • standardizing written protocols for chemotherapy cycles; • adding laboratory parameters on chemotherapy orders; • linking to journal articles for additional knowledge. • Institute same-day laboratory work on day of physician visit. • Develop standardized laboratory parameters with medical director. • Perform manual audits and provide feedback. • Educate nursing staff on the policy of not accepting orders with deviations. All interventions were introduced concurrently in the outpatient infusion center. Education for physicians was provided during grand rounds and in one-on-one discussions that included feedback on deficiencies discovered in the chart audit. Education for the nursing staff was also provided by oneon-one discussions and during staff meetings. RESULTS The initial (baseline) analysis showed that 70.1% of the forms had at least one deviation. After the interventions, there was a 19% relative decrease to 56.5% of the forms having at least one deviation. The key to improvement in ensuring that all information was included in the chemotherapy order form was
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e249
Bryant-Bova
the one-on-one education provided to the physicians. The process of educating both physicians and nursing staff continued after the project ended. A random sample of chemotherapy orders was reviewed during the follow-up period from October 2013 to March 2014. An updated Pareto chart from this review reflected an overall decrease in the number of types of deviations from 15 to seven (Appendix Fig A4, online only). Laboratory values not being reported within 7 days was still the major deviation at approximately 44%. The remaining 56% consisted of only six other types of deviations. Follow-up data showed that 34.5% of the forms had at least one deviation, a 51% relative decrease from the baseline, thus exceeding the initial goal to decrease the number of deviations by 50%. One year after follow-up, another review of random samples was conducted, which showed a slight increase in the number of deviations. Of the total number of charts reviewed, 50% of the chemotherapy orders had at least one deviation (Fig 1), representing a 29% relative decrease from the baseline. DISCUSSION The chemotherapy ordering process can be unsafe and inefficient, putting patients at risk for dosing and scheduling errors. There were many reasons for errors being made in the process of ordering and delivering chemotherapy. Within the limited time frame of the initial project, there was a minimal change in the number of orders with at least one deviation. Education
interventions for the physicians and staff continued after the project ended. Improvements consisted of collaboration between interdisciplinary departments to ensure that laboratory tests were completed during the appropriate time frame and that all relevant information was included on the order forms (eg, physician’s name and pager number). After 6 months, a follow-up review reflected a further decrease from the original 70.1% to 51% of the forms having a deviation, which surpassed the original goal of a 50% decrease. There was a slight increase after 1 year from follow-up, in which 50% of the order forms had at least one deviation, which is a relative decrease of 29% from the baseline. The main deviation was that one physician did not print his name after his signature. Since then, he has received one-on-one education. Despite this slight increase, education of physicians and staff continues, and the ordering process is being standardized. Physicians remain receptive to education and are cognizant of the deviations in their orders. They have made adjustments to their process of writing orders before submitting them to the nurses. Nurses became increasingly comfortable with speaking up when deviations were discovered in the chemotherapy orders, and they returned the order forms to the appropriate physician for correction before advancing to the next step in the process. The interventions continue to improve with completion of orders during the chemotherapy ordering process in the outpatient infusion centers.
80 70
Percent
60 50 40 30 20 10 0 January to February 2013 Total reviewed 127 Total with deviations 89
March to May 2013
October to March 2014
July to September 2015
207
29
36
117
10
18
FIG 1. Percentage of forms with deviations.
e250
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
The following are next steps in the improvement program: Implement interventions in the inpatient care setting. Modify the chemotherapy checklist to reflect an acceptable standardized order form. Eliminate distractions for the pharmacist or nurse while they are reviewing order forms. Accelerate the implementation of a computerized physician ordering system specific to chemotherapy (eg, BEACON). Request that the electronic health record provide separate tabs for chemotherapy orders and chemotherapy administration so that chemotherapy cycles can be easily tracked. With continued improvement in the chemotherapy ordering process, the number of deviations will decrease, thereby providing safe, efficient, and quality care for oncology patients. This project was an initiative for improving provision of efficient care throughout the cancer care continuum.
Acknowledgment I thank Tonya Jinkensen, OCH, CRNI, NE-BC, Rachel McKown, MHA, Beula Abraham, RPh, Nihal Abdulla, MD, and Lien Do, PharmD, members of the Clinical Safety and Effectiveness Team. Author’s Disclosures of Potential Conflicts of Interest Disclosures provided by the author are available with this article at jop.ascopubs.org. Corresponding author: Jill N. Bryant-Bova, University of Texas Medical Branch-Galveston, 301 University Blvd, Galveston, TX 77555-0155; e-mail: [email protected].
References 1. Cho E, Kim HJ, Kim GM, et al: Assessment of efficiency and safety of the comprehensive chemotherapy assistance program for ordering oncology medications. Int J Med Inform 82:504-513, 2013 2. Aita M, Belvedere O, De Carlo E, et al: Chemotherapy prescribing errors: An observational study on the role of information technology and computerized physician order entry systems. BMC Health Serv Res 13:522-528, 2013 3. Rachon F, Salles G, Sp¨ath HM, et al: Chemotherapeutic errors in hospitalised cancer patients: Attributable damage and extra costs. BMC Cancer 11:478, 2011 4. Garfield S, Reynolds M, Dermont L, et al: Measuring the severity of prescribing errors: A systematic review. Drug Saf 36:1151-1157, 2013
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e251
Bryant-Bova
AUTHOR’S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Improving Chemotherapy Ordering Process The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO’s conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml. Jill N. Bryant-Bova No relationship to disclose
e252
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
Appendix Patient seen in clinic Chemo administered to patient
Chemo is checked by pharmacist
Chemo is prepared by technician
Dose and drug is checked by two chemo RNs
Chemo is delivered to the clinic
Patient arrives for chemo treatment
Labs and chemo are ordered by physician
Patient is scheduled for chemo 3 days later
Wait for nurse to notify pharmacy of patient arrival
Infusion nurse checks order with pharmacy
Infusion nurse will complete the checklist
NO
Order is checked by second pharmacist
Are all questions answered? Generate label for medication bag YES Infusion nurse will fax orders to pharmacy Wait for nurse to refax or provide verbal orders
Pharmacy calls physician/nurse NO
Is the patient new?
YES Pharmacy pulls patient infusion folder Pharmacy checks protocols
YES
Is there an issue?
NO
Orders are entered into Pharmakon
Pharmacy creates new infusion folder
FIG A1. Process flowchart. Purple indicates the nursing flow and green indicates the pharmacy flow. Chemo, chemotherapy; Pharmakon, database.
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e253
Bryant-Bova
PROCESS
PEOPLE Acceptance of incomplete orders
Variances in completion of the form Under utilization of chemotherapy forms
Patient wait times Physician preferences
Knowledge deficit of staff
Differences in systems (Mosaic, Epic, Word)
Availability of physicians
Lack of awareness that it is available Time constraints
Differences in documentation Standardized flexible process needed Epic Doses, dates are not changed
Time constraints of staff Physician preferences
Use of Word instead of the form Outdated order sets
Lack of knowledge about the policy
Multiple deviations with chemotherapy ordering policy
No resources to update
Ability to copy and paste
Do not know of the policy
No hard stops Documentation variances
Not easy to find policy Not trained on policy
No order entry of chemotherapy Navigation to order cycles varies SYSTEMS
RESOURCES
EDUCATION
FIG A2. Fishbone diagram.
e254
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
PARETO CHART DATA No.
January 1 to February 28, 2013
Type of Deviation on Order Form
Cumulative Frequency
Frequency 42 15 11
Percentage
1 2 3
Lab values not recorded within 7 days Typed name and pager number not provided Missing day
4
Missing lab values on order form
9
77
73
5
Missing cycle
7
84
79
6 7
Missing BSA Missing start date
4 4
88 92
83 87
8
3
95
90
9
Multiple cycles on same order form; no specifications Wrong dose on order
3
98
92
10
Area under the curve formula not written out
2
100
94
11
Order to draw labs but no specific labs ordered
2
102
96
12 13 14
Missing protocol Missing allergies Duration of infusion not provided
1 1 1
103 104 105
97 98 99
15
Missing physician signature
1
106
100
Total
42 57 68
40 54 64
106
Deviations from Chemotherapy Ordering Process 50
100.0 90.0
42
40
80.0
35
70.0
30
60.0
25
50.0
20
40.0 15
15
30.0 11
9
10
20.0
7 4
5
4
3
3
2
2
10.0 1
1
1
1
m in g cy M cle Ar iss m M ea e in iss or g un i Or B d n de er SA g de ;n st rt rt ar o he o td s W p dr cu at ec ro aw rv e ifi ng e ca la fo d bs t rm io os n bu e ul s on tn a no o or tw sp de ec r rit ifi te cl n ab ou so t M Du rd iss ra er in tio ed g n M pr of iss ot in oc in fu M ol g sio iss al l in n er g no gi ph es tp ys ro ici vi an de d sig na tu re
rf
or
da
ed
de
in so
iss M
n
or
iss M
ot
g
id
da
ov
pr
7 in
la b g in iss M
sa
ith w
rn
d
be
de
um
or
rn
ec
ge
tr
pa
no
d
es
so
n
an
lu
e
cle
m
va b
cy
na
M
ul
tip
pe
le
d
La
y
0.0
ys
0
Ty
Percent
No. of Deviations
45
Type of Deviation FIG A3. (A) Pareto chart data. (B) Deviations from chemotherapy ordering process. BSA, body surface area.
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e255
Bryant-Bova
100.0
16 93.8%
14
87.5%
12
90.0 80.0
75.0%
70.0 10 8
62.5%
7
60.0 50.0
43.8%
40.0
6
Percent
No. of Deviations
81.3%
30.0 4 2
3
20.0 2 1
1
1
1
10.0 0.0
M
re L co ab iss rde va d lue in M w g ul ph ith s no tip ys in t le ici 7 d cy M an a cle iss Ar sig ys so ea in g na n un l a sa tu de bs re m rt on e he or o d M r c d e u Ty iss er r; rv pe no e in fo d g sp na rm B SA ec m ul ifi e a ca an no t io d tw ns pa rit M ge te iss rn n in o um ut g al be le rn rg ot ie s pr ov Or id M ed de iss rt in o g M dr d iss ay aw M in iss la g bs cy in W g cle bu ro st ng tn ar o td do sp at se ec e on Du ifi c o ra l r a d tio bs er n M or of iss de in in re fu g d sio pr n ot no oc ol tp ro vi de d
0
FIG A4. Follow-up Pareto chart for October 2013 to March 2014. BSA, body surface area.
e256
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving Chemotherapy Ordering Process Jill N. Bryant-Bova
University of Texas Medical Branch, Galveston, TX
Abstract Purpose Chemotherapy is a high-risk medication and is the second most common cause of fatal medication errors. The ordering process can be unsafe and inefficient, putting patients at risk for medication errors. The aim of this project was to decrease the number of chemotherapy order forms with at least one deviation by 50% within 5 months.
Methods A multidisciplinary team identified causes for variance in form completion, deficits in knowledge of ordering processes, and acceptance of incomplete orders by the staff. The Plan, Do, Study, Act improvement methodology evaluated the chemotherapy ordering process and found different types of deviations on order forms. Interventions consisted of educating physicians on entering complete information on orders, instituting same-day laboratory work on the day of the physician’s visit, standardizing laboratory parameters, performing audits of the order forms manually, and educating nurses on not accepting orders with deviations.
Results All order forms were reviewed, and data were collected on different types of deviations. The following deviations were identified: laboratory test results were not being entered into the order form within 7 days, physicians were not providing their name and pager number, and the days of the chemotherapy cycle were missing from the order forms. Before the intervention, 70.1% of the chemotherapy order forms had at least one deviation. After 5 months of interventions, there was a reduction of 19% in the number of order forms with at least one deviation. Follow-up at 6 months and 1 year showed continuing reductions in the number of chemotherapy orders with at least one deviation.
Conclusion Improvement was a result of collaboration between interdisciplinary departments. The original goal was surpassed as a result of educating physicians and staff and standardizing the ordering process. If the number of deviations in chemotherapy order forms is decreased, oncology patients will receive safe, efficient, and quality care.
ASSOCIATED CONTENT Appendices DOI: 10.1200/JOP.2015. 007443 DOI: 10.1200/JOP.2015.007443; published online ahead of print at jop.ascopubs.org on January 5, 2016.
e248
INTRODUCTION Prescribing errors account for the majority of medication errors.1 As the complexity of the drug regimen increases, so does the possibility of errors. Chemotherapy is a high-risk medication, and there are several reasons for increased susceptibility to
errors within the intricate ordering and administration process, such as complex chemotherapy regimens and requirement for dose adjustments.2 Chemotherapy is the second most common cause of fatal medication errors.3 Even if there is no patient harm from the error, correcting the
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
error increases the workload for the staff, and extra effort is required to remedy the patient’s perception of his or her care as suboptimal.4 Deviations in the chemotherapy ordering process were identified in the outpatient infusion centers at the University of Texas Medical Branch (UTMB) in Galveston, TX. There was a lack of consistency in documenting the orders and administering the chemotherapy, which made it difficult to locate previous orders and keep track of the chemotherapy cycle. In addition, the electronic health record allowed chemotherapy orders to be added to the patient’s record simply by using Microsoft Word’s copy and paste function in a Word document that had no standardized format. This led to orders being incomplete or incorrect for the specific patient. Other issues of concern included bypassing the chemotherapy dose checking step at the prescriber level and difficulty updating the original order when changes were required. Extra time was needed to reconcile the errors on the chemotherapy orders. Improving the chemotherapy ordering process is geared toward improving the quality of patient care, increasing staff satisfaction through improved communication and coordination of care between clinicians, and enhancing the overall patient experience in the UTMB outpatient infusion centers. The specific aim of this project was to decrease the number of chemotherapy order forms that deviated from standard chemotherapy ordering policy by 50% by June 2013 (original project end date) within outpatient infusion centers. METHODS This quality improvement project was initiated to examine UTMB’s current process for ordering chemotherapy (Appendix Fig A1, online only). The process begins in the outpatient infusion center where physicians place an order for laboratory work and a patient-specific chemotherapy regimen. The nurses complete a checklist of questions in preparation for administering chemotherapy and submit the orders to the pharmacy. The pharmacy follows up with verification of the order and protocols, prepares the chemotherapy, and ultimately transfers the chemotherapy to the nurses for administration to the patient. The Plan, Do, Study, Act improvement methodology was used to evaluate the chemotherapy ordering process and to analyze types of deviations within the chemotherapy order forms. A fishbone diagram tool was used to organize the various causes and factors that contributed to deviations in the chemotherapy ordering process.
Focus for improvement was narrowed to the following issues: variances in completing the order form by physicians on the basis of their preferences, deficit in staff knowledge regarding acceptance of incomplete orders, and lack of knowledge of the chemotherapy ordering process (Appendix Fig A2, online only). All chemotherapy order forms were reviewed during the initial evaluation and again at the postproject evaluation. Fifteen different types of deviations were discovered on the chemotherapy order forms. A Pareto chart showed that 40% of the deviations were the result of laboratory values not being added to theformwithin7daysbeforestartingthechemotherapyinfusion. The second highest deviation (14%) was physician’s name and pager number not being provided on the chemotherapy order form. Thirteen additional types of deviations made up the remaining 46% (Appendix Fig A3, online only). Based on the deviations found, the following interventions were initiated: • Provide education to physicians who order chemotherapy regarding: • completing the order form and including all pertinent information (ie, name and pager number); • standardizing written protocols for chemotherapy cycles; • adding laboratory parameters on chemotherapy orders; • linking to journal articles for additional knowledge. • Institute same-day laboratory work on day of physician visit. • Develop standardized laboratory parameters with medical director. • Perform manual audits and provide feedback. • Educate nursing staff on the policy of not accepting orders with deviations. All interventions were introduced concurrently in the outpatient infusion center. Education for physicians was provided during grand rounds and in one-on-one discussions that included feedback on deficiencies discovered in the chart audit. Education for the nursing staff was also provided by oneon-one discussions and during staff meetings. RESULTS The initial (baseline) analysis showed that 70.1% of the forms had at least one deviation. After the interventions, there was a 19% relative decrease to 56.5% of the forms having at least one deviation. The key to improvement in ensuring that all information was included in the chemotherapy order form was
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e249
Bryant-Bova
the one-on-one education provided to the physicians. The process of educating both physicians and nursing staff continued after the project ended. A random sample of chemotherapy orders was reviewed during the follow-up period from October 2013 to March 2014. An updated Pareto chart from this review reflected an overall decrease in the number of types of deviations from 15 to seven (Appendix Fig A4, online only). Laboratory values not being reported within 7 days was still the major deviation at approximately 44%. The remaining 56% consisted of only six other types of deviations. Follow-up data showed that 34.5% of the forms had at least one deviation, a 51% relative decrease from the baseline, thus exceeding the initial goal to decrease the number of deviations by 50%. One year after follow-up, another review of random samples was conducted, which showed a slight increase in the number of deviations. Of the total number of charts reviewed, 50% of the chemotherapy orders had at least one deviation (Fig 1), representing a 29% relative decrease from the baseline. DISCUSSION The chemotherapy ordering process can be unsafe and inefficient, putting patients at risk for dosing and scheduling errors. There were many reasons for errors being made in the process of ordering and delivering chemotherapy. Within the limited time frame of the initial project, there was a minimal change in the number of orders with at least one deviation. Education
interventions for the physicians and staff continued after the project ended. Improvements consisted of collaboration between interdisciplinary departments to ensure that laboratory tests were completed during the appropriate time frame and that all relevant information was included on the order forms (eg, physician’s name and pager number). After 6 months, a follow-up review reflected a further decrease from the original 70.1% to 51% of the forms having a deviation, which surpassed the original goal of a 50% decrease. There was a slight increase after 1 year from follow-up, in which 50% of the order forms had at least one deviation, which is a relative decrease of 29% from the baseline. The main deviation was that one physician did not print his name after his signature. Since then, he has received one-on-one education. Despite this slight increase, education of physicians and staff continues, and the ordering process is being standardized. Physicians remain receptive to education and are cognizant of the deviations in their orders. They have made adjustments to their process of writing orders before submitting them to the nurses. Nurses became increasingly comfortable with speaking up when deviations were discovered in the chemotherapy orders, and they returned the order forms to the appropriate physician for correction before advancing to the next step in the process. The interventions continue to improve with completion of orders during the chemotherapy ordering process in the outpatient infusion centers.
80 70
Percent
60 50 40 30 20 10 0 January to February 2013 Total reviewed 127 Total with deviations 89
March to May 2013
October to March 2014
July to September 2015
207
29
36
117
10
18
FIG 1. Percentage of forms with deviations.
e250
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
The following are next steps in the improvement program: Implement interventions in the inpatient care setting. Modify the chemotherapy checklist to reflect an acceptable standardized order form. Eliminate distractions for the pharmacist or nurse while they are reviewing order forms. Accelerate the implementation of a computerized physician ordering system specific to chemotherapy (eg, BEACON). Request that the electronic health record provide separate tabs for chemotherapy orders and chemotherapy administration so that chemotherapy cycles can be easily tracked. With continued improvement in the chemotherapy ordering process, the number of deviations will decrease, thereby providing safe, efficient, and quality care for oncology patients. This project was an initiative for improving provision of efficient care throughout the cancer care continuum.
Acknowledgment I thank Tonya Jinkensen, OCH, CRNI, NE-BC, Rachel McKown, MHA, Beula Abraham, RPh, Nihal Abdulla, MD, and Lien Do, PharmD, members of the Clinical Safety and Effectiveness Team. Author’s Disclosures of Potential Conflicts of Interest Disclosures provided by the author are available with this article at jop.ascopubs.org. Corresponding author: Jill N. Bryant-Bova, University of Texas Medical Branch-Galveston, 301 University Blvd, Galveston, TX 77555-0155; e-mail: [email protected].
References 1. Cho E, Kim HJ, Kim GM, et al: Assessment of efficiency and safety of the comprehensive chemotherapy assistance program for ordering oncology medications. Int J Med Inform 82:504-513, 2013 2. Aita M, Belvedere O, De Carlo E, et al: Chemotherapy prescribing errors: An observational study on the role of information technology and computerized physician order entry systems. BMC Health Serv Res 13:522-528, 2013 3. Rachon F, Salles G, Sp¨ath HM, et al: Chemotherapeutic errors in hospitalised cancer patients: Attributable damage and extra costs. BMC Cancer 11:478, 2011 4. Garfield S, Reynolds M, Dermont L, et al: Measuring the severity of prescribing errors: A systematic review. Drug Saf 36:1151-1157, 2013
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e251
Bryant-Bova
AUTHOR’S DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST Improving Chemotherapy Ordering Process The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO’s conflict of interest policy, please refer to www.asco.org/rwc or jop.ascopubs.org/site/misc/ifc.xhtml. Jill N. Bryant-Bova No relationship to disclose
e252
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
Appendix Patient seen in clinic Chemo administered to patient
Chemo is checked by pharmacist
Chemo is prepared by technician
Dose and drug is checked by two chemo RNs
Chemo is delivered to the clinic
Patient arrives for chemo treatment
Labs and chemo are ordered by physician
Patient is scheduled for chemo 3 days later
Wait for nurse to notify pharmacy of patient arrival
Infusion nurse checks order with pharmacy
Infusion nurse will complete the checklist
NO
Order is checked by second pharmacist
Are all questions answered? Generate label for medication bag YES Infusion nurse will fax orders to pharmacy Wait for nurse to refax or provide verbal orders
Pharmacy calls physician/nurse NO
Is the patient new?
YES Pharmacy pulls patient infusion folder Pharmacy checks protocols
YES
Is there an issue?
NO
Orders are entered into Pharmakon
Pharmacy creates new infusion folder
FIG A1. Process flowchart. Purple indicates the nursing flow and green indicates the pharmacy flow. Chemo, chemotherapy; Pharmakon, database.
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e253
Bryant-Bova
PROCESS
PEOPLE Acceptance of incomplete orders
Variances in completion of the form Under utilization of chemotherapy forms
Patient wait times Physician preferences
Knowledge deficit of staff
Differences in systems (Mosaic, Epic, Word)
Availability of physicians
Lack of awareness that it is available Time constraints
Differences in documentation Standardized flexible process needed Epic Doses, dates are not changed
Time constraints of staff Physician preferences
Use of Word instead of the form Outdated order sets
Lack of knowledge about the policy
Multiple deviations with chemotherapy ordering policy
No resources to update
Ability to copy and paste
Do not know of the policy
No hard stops Documentation variances
Not easy to find policy Not trained on policy
No order entry of chemotherapy Navigation to order cycles varies SYSTEMS
RESOURCES
EDUCATION
FIG A2. Fishbone diagram.
e254
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Improving the Chemotherapy Ordering Process
PARETO CHART DATA No.
January 1 to February 28, 2013
Type of Deviation on Order Form
Cumulative Frequency
Frequency 42 15 11
Percentage
1 2 3
Lab values not recorded within 7 days Typed name and pager number not provided Missing day
4
Missing lab values on order form
9
77
73
5
Missing cycle
7
84
79
6 7
Missing BSA Missing start date
4 4
88 92
83 87
8
3
95
90
9
Multiple cycles on same order form; no specifications Wrong dose on order
3
98
92
10
Area under the curve formula not written out
2
100
94
11
Order to draw labs but no specific labs ordered
2
102
96
12 13 14
Missing protocol Missing allergies Duration of infusion not provided
1 1 1
103 104 105
97 98 99
15
Missing physician signature
1
106
100
Total
42 57 68
40 54 64
106
Deviations from Chemotherapy Ordering Process 50
100.0 90.0
42
40
80.0
35
70.0
30
60.0
25
50.0
20
40.0 15
15
30.0 11
9
10
20.0
7 4
5
4
3
3
2
2
10.0 1
1
1
1
m in g cy M cle Ar iss m M ea e in iss or g un i Or B d n de er SA g de ;n st rt rt ar o he o td s W p dr cu at ec ro aw rv e ifi ng e ca la fo d bs t rm io os n bu e ul s on tn a no o or tw sp de ec r rit ifi te cl n ab ou so t M Du rd iss ra er in tio ed g n M pr of iss ot in oc in fu M ol g sio iss al l in n er g no gi ph es tp ys ro ici vi an de d sig na tu re
rf
or
da
ed
de
in so
iss M
n
or
iss M
ot
g
id
da
ov
pr
7 in
la b g in iss M
sa
ith w
rn
d
be
de
um
or
rn
ec
ge
tr
pa
no
d
es
so
n
an
lu
e
cle
m
va b
cy
na
M
ul
tip
pe
le
d
La
y
0.0
ys
0
Ty
Percent
No. of Deviations
45
Type of Deviation FIG A3. (A) Pareto chart data. (B) Deviations from chemotherapy ordering process. BSA, body surface area.
Volume 12 / Issue 2 / February 2016 n jop.ascopubs.org Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
Copyright © 2016 by American Society of Clinical Oncology
e255
Bryant-Bova
100.0
16 93.8%
14
87.5%
12
90.0 80.0
75.0%
70.0 10 8
62.5%
7
60.0 50.0
43.8%
40.0
6
Percent
No. of Deviations
81.3%
30.0 4 2
3
20.0 2 1
1
1
1
10.0 0.0
M
re L co ab iss rde va d lue in M w g ul ph ith s no tip ys in t le ici 7 d cy M an a cle iss Ar sig ys so ea in g na n un l a sa tu de bs re m rt on e he or o d M r c d e u Ty iss er r; rv pe no e in fo d g sp na rm B SA ec m ul ifi e a ca an no t io d tw ns pa rit M ge te iss rn n in o um ut g al be le rn rg ot ie s pr ov Or id M ed de iss rt in o g M dr d iss ay aw M in iss la g bs cy in W g cle bu ro st ng tn ar o td do sp at se ec e on Du ifi c o ra l r a d tio bs er n M or of iss de in in re fu g d sio pr n ot no oc ol tp ro vi de d
0
FIG A4. Follow-up Pareto chart for October 2013 to March 2014. BSA, body surface area.
e256
Volume 12 / Issue 2 / February 2016 n Journal of Oncology Practice Copyright © 2016 by American Society of Clinical Oncology Downloaded from ascopubs.org by Fudan University on November 28, 2016 from 061.129.042.030 Copyright © 2016 American Society of Clinical Oncology. All rights reserved.
