Medication Reconciliation Process Failures: a Study of Serious Events Reported by Pennsylvania Hospitals 48 Incorrect Patient Heights Amy Harper Et Al

Former insurance executive Wendell Potter PATIENT on the brokenness of U.S. healthcare Generally considered safe, minimally SAFETY invasive surgery still carries risks March 2021 | Vol. 3, No.1 A deep dive into outpatient medication safety

I think vaccines are:

Lifesaving

Unsafe I’m not sure Can we talk? LETTER From the Editor

Patient safety is one of those passions. It ment of outpatient electronic health re- is a common aspiration, a common vision cords to mitigate these risks. that brings healthcare workers, healthcare systems, and patients together. As Wendell Potter, well-known former-health- the year turns to spring, we look forward insurance-executive-turned-whistleblower, with renewed spirit to the challenges that shares his perspectives on the healthcare still lie ahead. With this March issue of Pa- insurance industry in the United States, tient Safety, we acknowledge Patient Safe- how the average person can sniff out ty Awareness Week with our colleagues propaganda, and how patients can take across the country, bringing you stories charge of their own care. You do not want and research to both inform and inspire. to miss this interview.

From our cover: Patient Safety’s managing Other features include original articles re- editor, Caitlyn Allen, sits down for some lated to medication reconciliation, medi- vaccine straight talk with JoAnn Adkins, cation errors related to patient height, infection preventionist; Dr. Lily Brown, and injuries from misplaced nasogastric psychologist; and Susan Senator, the tubes, and perspective pieces from both mother of an autistic child who provides a patients and providers related to patient glimpse into the vaccine-hesitant culture height, resilience, derogatory language and what made her change her mind. The in medical records, and the importance discovery of vaccines is one of the great- of being your own advocate. est contributions to mankind in the history of medicine. Being able to bridge the I hope that your passion for safer care gap between fact and fiction takes real and a brighter tomorrow continues to conversations, a real understanding of burn through these difficult times. If you have stories to share about your experi- Regina Hoffman, the fears and skepticism of patients and families, and mutual respect. ence as a clinician or as a patient through Editor-in-Chief this pandemic, please do not hesitate to Patient Safety Also, from our cover: While minimally submit a perspective piece to our edito- invasive surgery (like the laparoscopic rial team. Original research, quality improvement initiatives, and case studies he world as we knew it is only removal of a gall bladder) is generally are also welcome at any time. a distant memory. It has been a safe, it is not without risk. Patient Safe- year since everything changed. ty Analyst Lea Anne Gardner shares her This journal was designed for our authors Kids came home from school— findings in an original analysis of Penn- to freely share the important work they some are still home. Meetings moved sylvania Patient Safety Reporting System T do to improve patient safety, and for our from downtown conference rooms to data. She includes the types of surgeries readers to freely receive the information, overcrowded dining rooms. Toilet paper and related injuries, patient outcomes, strategies, and lessons learned to make almost became a black-market commod- and measures to minimize this some- the care they provide and receive safer. ity. Our healthcare workers and health- times-underappreciated risk. care systems were pushed beyond their Authors Dr. Karen Zimmer, Dr. David Stay safe and stay well! limits. A half million people died in the Classen, and Jessica Cole walk us through United States alone and millions more their approach to creating a new catego- recovered. Some initiatives were put on rization schema for medication errors hold and others fast-tracked through in the ambulatory setting. This work not what in “normal” times would take years. only helped them to better understand One thing is for certain, none of us are medication safety errors in this setting, the same. And although we are not the but also, perhaps even more importantly, same, there are passions that we contin- will inform future work in the improve- ue to share.

2 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 ABOUT PATIENT SAFETY Mark Jarrett, MD, MBA, MS, Northwell Health Matthew Keris, Esq., Marshall, Dennehey, Warner, Coleman & Goggin As the journal of the Patient Safety Authority, committed to the vision of Stephen Lawless, MD, Nemours Children’s Health System “safe healthcare for all patients,” Patient Safety (ISSN 2689-0143) is fully Michael Leonard, MD, Safe & Reliable Healthcare, LLC open access and highlights original research, advanced analytics, and hot James McClurken, MD, Doylestown Health, Lewis Katz School of Medicine at topics in healthcare. Temple University Patrick J. McDonnell, PharmD, Temple University School of Pharmacy The mission of this publication is to inform and advise clinicians, adminis- Dwight McKay, BSL, Lancaster Rehabilitation Hospital trators, and patients on preventing harm and improving safety, by providing Ferdinando L. Mirarchi, DO, UPMC Hamot evidence-based, original research; editorials addressing current and some- Dona Molyneaux, PhD, RN, Frances M. Maquire School of Nursing and Health times controversial topics; and analyses from one of the world’s largest Professions, Jefferson Health; Gwynedd Mercy University adverse event reporting databases. Gina H. Moore, BSN, RN, Christiana Care Health Services Rustin B. Morse, MD, Nationwide Children’s Hospital We invite you to submit manuscripts that align with our mission. We’re Adam Novak, MA, Michigan Health & Hospital Association particularly looking for well-written original research articles, reviews, Barbara Pelletreau, MPH, RN, Dignity Health commentaries, case studies, data analyses, quality improvement studies, Julia C. Prentice, MSPH, PhD, Betsy Lehman Center for Patient Safety or other manuscripts that will advance patient safety. Mitesh Rao, MD, MHS, Stanford School of Medicine Cheryl Richardson, MPH, RN, Penn State Milton S. Hershey Med. Center All articles are published under the Creative Commons Attribution – Lisa Rodebaugh, BS, Mercy Ministries Noncommercial license, unless otherwise noted. The current issue is Jeffrey Stone, PhD, Penn State University available at patientsafetyj.com. Jennifer Taylor, PhD, Drexel University The patient is central to everything we do. Patient Safety complies with Carlos Urrea, MD, Hillrom the Patients Included™ journal charter, which requires at least two patient Linda Waddell, MSN, RN, Wolff Center at UPMC members on the editorial board; regular publication of editorials, reviews, Eric Weitz, Esq., The Weitz Firm, LLC or research articles authored by patients; and peer review by patients. Margaret Wojnar, MD, MEd, Penn State College of Medicine Zane R. Wolf, PhD, RN, LaSalle University This publication is disseminated quarterly by email at no cost to the sub- scriber. To subscribe, go to patientsafetyj.com. ACKNOWLEDGMENTS

A special thanks to our reviewers: PRODUCTION STAFF Scott W. Cowan, MD, Jefferson Health Regina M. Hoffman, MBA, RN, Editor-in-Chief Dan Degnan, PharmD, MS, Purdue University College of Pharmacy Caitlyn Allen, MPH, Managing Editor Michael Gaunt, PharmD, ISMP Eugene Myers, BA, Associate Editor Kelly Gleason, PhD, RN, John Hopkins School of Nursing Jackie Peck, BS, Layout Editor Jennifer Hamm, BS Shawn Kepner, MS, Data Editor Rebecca Jones, MBA, RN, Patient Safety Authority Krista Soverino, BA, Graphics Kathleen Law, DNP, MS, RN, Riley Hospital for Children Javier F. Magrina, MD, Mayo Clinic Brittany H. WMcCann, PharmD, MEd, Pennsylvania Hospital EDITORIAL BOARD Dwight McKay Julia Prentice, PhD, Betsy Lehman Center for Patient Safety Joshua Atkins, MD, PhD, Hospital of the University of Pennsylvania Cheryl Richardson, MPH, RN, Penn State Health Milton S. Hershey Med Center Russell Baxley, MHA, Beaufort Memorial Hospital Lindsey Valentine, MD, Cleveland Clinic OBGYN & Women's Health Institute Mark E. Bruley, BS, ECRI Institute (retired) Michael A. Bruno, MD, Penn State Milton S. Hershey Medical Center Michael R. Cohen, MS, RPh, Institute for Safe Medication Practices Patient Safety Authority Daniel D. Degnan, PharmD, Purdue University School of Pharmacy 333 Market Street - Lobby Level Barbara Fain, JD, MPP, Betsy Lehman Center for Patient Safety Harrisburg, PA 17101 Kelly Gleason, PhD, RN, Johns Hopkins School of Nursing patientsafetyj.com Hitinder Gurm, MD, Michigan Medicine [email protected] Julia A. Haller, MD, Wills Eye Hospital 717.346.0469 Jennifer Hamm, BS, Fred Hamm, Inc. Ann Hendrich, PhD, RN, Building Age-Friendly Health Systems, John A. Hartford Foundation; formerly Ascension Healthcare

Together we save lives

Patient Safety I Vol. 3 No. 1 I March 2021 I 3 Contents March 2021 I Vol. 3, No.1

6 2 LETTER FROM THE EDITOR ENGAGING PATIENTS PERSPECTIVES

6 WE ALL WIN WHEN PATIENTS SPEAK UP LISA RODEBAUGH ET AL. A conversation with patient advocate Lisa Rodebaugh; executive director of the Patient Safety Authority (PSA), Regina Hoffman; and PSA director of Engagement, Caitlyn Allen.

34 OUR BEST SHOT: BRIDGING THE GAP IN THE VACCINE DIVIDE CAITLYN ALLEN Infection preventionist JoAnn Adkins; psychologist Dr. Lily Brown; and mother of a son with autism, Susan Senator share their insights into how vaccines work, how to recognize when anxiety may be clouding our judgment, and how both sides of the vaccine debate can finally have a real—and productive—conversation.

42 FALLING SHORT: ADVERSE EVENTS RELATED TO PATIENT HEIGHT CAITLYN ALLEN AND NEAL WIGGERMANN 10 Did you know almost 1 million Americans are at least 6'4"? Or MEDICATION RECONCILIATION that patients with a tall stature may experience a delay of care, pressure injury, or fall simply because of their height?

58 I’M OK: MY PERSPECTIVE ON RESILIENCE CHRISTOPHER MAMROL 2020 proved we may be stronger than we realized.

74 DEROGATORY LANGUAGE IN CHARTING: THE DOMINO EFFECT BETHANY DAVIS “Sticks and stones may break my bones, but derogatory language in my medical record may really hurt me.”

84 HANGING BY A THREAD CAITLYN ALLEN Healthcare in the United States: Who’s really pulling the strings? Wendell Potter, former health insurance executive turned whistleblower, shares how the American people are being misled, signs a news story is just well-concealed propaganda, and 42 how recognizing it as such is vital to making healthcare safer. EXTREME HEIGHTS ORIGINAL ARTICLES

10 PATIENT HARM RESULTING FROM MEDICATION RECONCILIATION PROCESS FAILURES: A STUDY OF SERIOUS EVENTS REPORTED BY PENNSYLVANIA HOSPITALS 48 INCORRECT PATIENT HEIGHTS AMY HARPER ET AL. An analysis of 93 serious events, identifying the most frequent transition-in-care area associated with these errors; the most common classes of drugs and stages of failure; and risk reduction strategies.

23 CATEGORIZATION OF MEDICATION SAFETY ERRORS IN AMBULATORY ELECTRONIC HEALTH RECORDS KAREN ZIMMER ET AL As more people are choosing outpatient care from fear of COVID-19, it’s more important than ever to understand the unique challenges facing ambulatory settings, including medication errors.

48 CHALLENGES WITH MEASUREMENT AND TRANSCRIPTION OF PATIENT HEIGHT: AN ANALYSIS OF PATIENT SAFETY EVENTS IN PENNSYLVANIA RELATED TO INACCURATE PATIENT HEIGHT ELIZABETH KUKIELKA 58 Quick, how much taller is 5'6" than 56"? While you’re doing RESILIENCE the math, ask yourself how many medications are dosed based on body surface area and what are the potentially fatal consequences if that calculation is performed using an incorrect height.

62 TROCAR-RELATED SAFETY EVENTS IN MINIMALLY INVASIVE SURGICAL PROCEDURES: RISKS FOR ORGAN AND VASCULAR COMPLICATIONS LEA ANNE GARDNER Generally considered safe, minimally invasive surgery still carries risks. Injuries from trocars, seemingly innocuous devices used during these procedures, can cause injury, resulting in conversion to open surgery, return to the OR, and readmissions.

79 DATA SNAPSHOT: MISPLACED NASOGASTRIC TUBES LEA ANNE GARDNER AND SUSAN WALLACE Learn the risks of misplaced NG tubes in less time than it takes to insert one. 74 DEROGATORY LANGUAGE How often have you treated patients with my condition?

How long should I wait to call your office about a new symptom?

What are the long- term effects of this condition?

Which internet sources can I trust?

6 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 WE ALL WIN WHEN PATIENTS SPEAK UP A conversation with patient advocate Lisa Rodebaugh, Patient Safety Authority (PSA) Executive Director Regina Hoffman, and PSA Director of Engagement Caitlyn Allen

By Lisa Rodebaugh, MEd, Regina Hoffman, MBA, RN◊ & Caitlyn Allen, MPH◆◊

Corresponding author ◊Patient Safety Authority †Disclosure: The authors declare that they have no relevant or material financial interests.

Patient Safety I Vol. 3 No. 1 I March 2021 I 7 My husband, Bill, and Patient engagement Lisa Rodebaugh I can rattle off our sto- Regina Hoffman is more than ensur- ries in two minutes or less—stories from having five children, ing systems and processes are designed considering the per- each with varying medical and developmental needs that spective of the patient. It isn’t about making sure patients presented far differently from what specialists typically see. sit on your quality council or help design your next waiting Each required those who knew them best asking questions room. It is much more foundational than that. It is about to get them the right help at the right time. We tell the stories empowering patients and their loved ones, like Lisa, to be often, to encourage those listening to always ask the ques- their own advocate. Empowering our patients and their fam- tions. We tell the stories often, because self-advocacy matters. ilies to be active partners in their own care is critical to the Sometimes it is the difference between life and death. advancement of patient safety in all settings.

My oldest son, now 21, was 6 months old and kept running How do we help empower patients and families to be advo- a low-grade fever. I diligently called the pediatrician when it cates and partners in their own care? began and called again when it did not go away after 10 days. The doctor was convinced there was nothing to worry about: We start by arming them with information and knowledge. most likely he just had two different colds, and there was no Education of our own health and the healthcare system needs way the fever would last long. to begin in childhood and continue all the way through adult- hood. Health illiteracy is a continuing problem, and we need Three weeks later and with my son still running a fever, I to start addressing it early. Lisa talks about asking questions called again insisting we be seen. They agreed to see us—after and tells us to never stop asking questions. She is absolutely giving me the “new mom chat,” an assurance that everything right, AND we need to ensure that our patients and their loved was probably fine and a request to not worry so much. ones know what to ask.

Thankfully, they sent us for lab work, because by this point, Patient education needs to begin long before we get sick. For my son was in septic shock. We spent the next few days in the example, campaigns around knowledge of life-threatening hospital. He recovered, and I was equipped with information conditions, like sepsis, are crucial to reducing complications and the confidence to never stop asking questions. and death. If patients know that the signs of sepsis include confusion, shortness of breath, high heart rate, extreme pain, One year later, my second child was born with a condition fever and chills, and sweaty skin, they may be more likely to called hypoplastic left heart syndrome (HLHS), a congenital ask their provider if it could be sepsis when they have those condition (something you’re born with) that affects blood flow symptoms. through the heart. His first of three open-heart surgeries happened when he was just 8 days old. Two months later, I noticed While we go to healthcare practitioners because they are his chest looked buff, like a little Arnold Schwarzenegger. The experts at diagnosing and treating illness, being knowledge- cardiologist humored me with a visit, only because we were able of our own bodies, knowing what is “normal” for us, and in the precarious time between our son’s surgeries. Again, I being comfortable enough to share important information got that “do not worry so much” chat. and ask questions helps ensure a successful outcome.

Immediately following an echocardiogram, our infant son was Patients being comfortable is the second critical part of rushed into the cardiac cath lab to unblock a critical opening empowerment. It is the responsibility of healthcare pro- between his heart chambers that had unexpectedly closed with viders to set the stage for conversation. There are a few key scar tissue. Once again, the life of one of my children was saved concepts that we can employ to raise the comfort level of by asking questions and being persistent. our patients. Open the conversation door. Ask open-ended questions. Ask if there is anything specific that is really Those are the most dramatic stories. However, there are troubling the patient. Listen more than you speak. Really countless more in my two decades of parenting that have hear what the person is saying or asking. Providers do not made me never regret following my gut—parental intuition is know their patients better than the patients know them- real and can be lifesaving. Advocating for my children forced selves. Show respect. Providers may not always agree with me to step outside my comfort zone. Often, I had to ques- the choice a patient makes. The provider is responsible for tion what someone with specific training in that specialty ensuring that the patient’s choice is a well-informed choice, was telling me. Often, I had to keep pressing because I knew not necessarily the same choice the provider would make. something was wrong. The third aspect of empowerment is personal accountability. In the 20 years since To Err is Human was published, we have Patient engagement, empowerment, and advocacy is a two- come a long way in having a patient’s or caregiver’s voice way street. Providers must create an environment that fosters validated, but this is an area where we can continue to grow these three terms, but society at large must take a more active and evolve. I try not to think what would have happened had approach in personal health. Lisa tells us that she never regrets I shrunk in my corner as a new mom, not knowing how very trusting her intuition. She would not allow herself to be dis- critical my observations were. missed. If your body is telling you something just isn’t right, then you must persist until you are heard. Be actively engaged Trust your instincts. Ask questions.

8 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 in discussions and decisions involving your own health. Ask About the Authors questions or have someone with you who can ask questions on your behalf. No one is going to be a better advocate for you Lisa Rodebaugh, wife and mother of five children, is a certified than you or a loved one. mental health coach, patient advocate, and co-author of the blog One Bite at a Time: a mother-son journey through autism.

Years ago, a family medicine Regina Hoffman is executive director of Pennsylvania’s Patient Caitlyn Allen provider held a group session Safety Authority (PSA), an independent agency charged with for patients newly diagnosed with diabetes, under the assumption taking steps to reduce medical errors by identifying problems that a group appointment would allow more face time with him and recommending solutions to promote patient safety, and than individual ones, and the patients could learn from each other editor-in-chief of Patient Safety, PSA’s peer-reviewed journal. by hearing questions they themselves may not think to ask. Caitlyn Allen ([email protected]) is director of Engagement for During each session, the patients would have their bloodwork drawn the Patient Safety Authority and managing editor for Patient and learn things like how to safely inject insulin. While most of the Safety, the PSA’s peer-reviewed journal. Before joining the PSA, patients were thriving under this model, one patient consistently had she was the project manager for Patient Safety at Jefferson poor blood sugar levels despite his insistence he was taking his insulin Health, where she also was the only nonphysician elected as instructed. Yet, session after session there was little change in his to serve on the House Staff Quality and Safety Leadership bloodwork and if his diabetes continued unchecked, he would face Council. Previously, Allen also was a project manager and significant consequences. patient safety officer for Wills Eye Hospital. The physician became increasingly frustrated and began contemplat- ing his options. If this patient wasn’t going to take this seriously, there This article is published under the Creative Commons Attribution- were plenty of people who would. NonCommercial license.

Before kicking him out of the program, the physician had one more idea. He asked the patient to bring his insulin to the next session so he could witness the injection—anticipating evidence the patient had in fact not been taking his medication.

The patient complied, inserted the syringe into the vial, and began to inject himself with the ease that comes from repetition. It was then the physician realized the patient had poor eyesight and could not see he wasn’t inserting the syringe deep enough into the vial to draw sufficient insulin.

The patient had never mentioned his failing vision, and the physician had never asked.

I heard this story almost a decade ago while still in grad school, and it has stayed with me ever since. The provider in the story—a great physician who dedicated his career to helping underserved communities—shared it to demonstrate that even when you think you’re doing everything right, there will always be circumstances you can’t anticipate.

How much better would healthcare be, how many fewer errors would occur, if we could help clinicians consider the incon- ceivable and at the same time empower patients to speak up, ask questions, and be vested members of their care team, just like Lisa.

Patient Safety I Vol. 3 No. 1 I March 2021 I 9 10 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Patient Harm Resulting From Medication Reconciliation Process Failures: A Study of Serious Events Reported by Pennsylvania Hospitals

By Amy Harper, PhD, RN◆◊, Elizabeth Kukielka, PharmD, MA, RPh◊ & Rebecca Jones, MBA, RN◊ DOI: 10.33940/data/2021.3.1

Corresponding author Patient Safety Authority Disclosure: The authors declare that they have no relevant or material financial interests.

Patient Safety I Vol. 3 No. 1 I March 2021 I 11 Introduction

The Joint Commission defines the medication reconciliation process as “…the process of comparing the medications a patient is taking (or should be taking) with newly ordered medications. The comparison addresses duplications, omissions, and interactions, and the need to continue current medications.”1 In addition to these formal processes that facilities use to reconcile home medications, providers for hospitalized patients also must informally consider changing patient conditions; current therapies, diagnoses, and labs; and many other factors when making changes to therapies to also prevent discrepancies. The American Pharmacists Association (APhA) considers a broader definition of edication reconciliation broadly defined includes medication reconciliation as both formal and informal processes that involve the comprehensive evaluation of a patient’s medi- the comprehensive evaluation of a patient’s medication cations during each transition of care and change regimen any time there is a change in therapy in an effort to Min therapy. The medication reconciliation process is complex, avoid medication errors such as omissions, duplications, dosing and studies have shown that up to 91% of medication reconcil- errors, or drug interactions, as well as to observe compliance and iation errors are clinically significant and 1–2% are serious or adherence patterns. This process should include a comparison potentially life-threatening. We queried the Pennsylvania Patient of the existing and previous medication regimens and should Safety Reporting System (PA-PSRS) and identified 93 serious occur at every transition of care in which new medications are events related to the medication reconciliation process reported ordered, existing orders are rewritten or adjusted, or if the patient between January 2015 and August 2020. Serious events related to has added nonprescription medications to [his or her] self-care.2 medication reconciliation were most common among patients 65 years or older (55.9%; 52 of 93). The majority of events (58.1%; 54 Thus, medication reconciliation broadly defined is an ongoing of 93) contributed to or resulted in temporary harm and required process that optimally occurs at each transition of care and change treatment or intervention. Permanent harm or death occurred as in therapy for the patient (e.g., outpatient physician visits, admis- a result of 3.3% (3 of 93) of the events. Admission/triage was the sion and discharge from hospitals and nursing homes, transitions most frequent transition of care associated with events (69.9%; 65 between hospital units, before and after procedures, and changes of 93). The most common stage of the medication reconciliation in therapy).3-5 Medication reconciliation involves a complex set of process at which failures most directly contributed to patient steps that require effective communication, documentation, and harm was order entry/transcription (41.9%; 39 of 93) and resulted patient and clinician participation.6 The process is also resource most frequently in wrong dose (n=21) or dose omission (n=13). intensive and poses numerous challenges for healthcare pro- Most events were discovered after the patient had a change in viders. For example, medication reconciliation requires desig- condition (76.3%; 71 of 93), and patients most often required nating clear roles and responsibilities; standardizing admission, readmission, hospitalization, emergency care, intensive care, transfer of care, and discharge procedures; and ensuring access or transfer to a higher level of care (58.0%; 54 of 93). Among 128 to accurate patient medication lists.5-11 Though many studies medications identified across all events, neurologic or psychiatric have examined the medication reconciliation process and some medications were the most common (39.1%; 50 of 128), and anti- have achieved localized success through interventions involving convulsants were the most common pharmacologic class among nurses, medical interns, pharmacists, and pharmacy technicians, neurologic or psychiatric medications (42.0%; 21 of 50). Based on breakdowns in this process continue to contribute to patient safety our findings, risk reduction strategies that may improve patient events each year.12-18 safety related to the medication reconciliation process include defined clinician roles for medication reconciliation, listing the Studies have shown that 50–67% of medication histories contained indication for each medication prescribed, and for facilities to errors, most frequently because they included medications that consider adding anticonvulsants to their processes for medica- the patient was no longer taking or because a medication was tions with a high risk for harm. omitted.14,15,17,19 Furthermore, studies have also demonstrated that up to 81% of medication histories for geriatric patients contained errors.18,20 These errors in medication histories can result in inac- Keywords: medication reconciliation, home medication, medication curate lists of patients’ home medications, incorrect inpatient errors, transitions of care, patient safety, anticonvulsant orders, and incorrect medication prescriptions at discharge.6,21

Many medication reconciliation errors have the potential for adverse effects.22 Errors that involve high-alert medications pose an increased risk for patient harm.22-25 Some studies have shown that up to 91% of medication reconciliation errors are clinically significant or had potential for harm, and 1–2% are serious or potentially life-threatening.17,18 Furthermore, errors in medication orders at discharge are likely underreported and also have the potential to be life-threatening.10,26,27

12 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 In this study, we examined serious events involving medication ● The event occurred in a hospital setting.

reconciliation that were submitted to the Pennsylvania Patient † * ● The event was identified as a Serious Event by the Safety Reporting System (PA-PSRS) in order to gain a better underreporting facility. standing of the ways in which patient safety may be compromised by problems and errors associated with the medication recon- ● The event involved the process of medication ciliation process. Based on our findings, we have identified risk reconciliation as defined by APhA (see introduction).2 reduction strategies that may have the most tangible impact on patient safety. We manually coded events to identify the following:

● Transition of care point that was associated with the Methods proximate cause for the patient harm described in the event report. We queried the PA-PSRS acute care database for events that were ● Stage of the medication reconciliation process at which submitted by Pennsylvania facilities from January 1, 2015, to August failures most directly contributed to the event. 31, 2020. We searched free-text fields (i.e., “Event Details,” “Event Recommendations,” “Event Comments,” “Event Sub Type Other,” ● Type of medication error that occurred as a result of the “MedERR source,” and “MedERR ContributingFactors”) for key- process failure. words relating to medication reconciliation, medication list, home ● Medications involved in the event report. medications, and home drugs along with truncations and alternate ● Circumstances of how the event was discovered or spellings for each, (e.g., “med rec,” “med req,” or “home meds”). confirmed. We included events in our analysis if they met all of the following ● Additional care or services required as a result of the criteria: event.

Figure 1: Harm Scores of Serious Events Related to Medication Reconciliation Submitted to PA-PSRS From January 1, 2015–August 31, 2020, N=93

G I 2.2% Harm scores are assigned by reporting facilities. Definitions 1.1% for each harm score are provided below.

E – An event occurred that contributed to or resulted in temporary harm and required treatment or intervention F – An event occurred that contributed to or resulted in temporary harm and required initial or prolonged hospitalization G – An event occurred that contributed to or resulted in permanent harm F 38.7% H – An event occurred that resulted in a near-death event E (e.g., required intensive care unit care or other intervention 58.1% necessary to sustain life) I – An event occurred that contributed to or resulted in death

Note: Due to rounding, the percentages do not add up to 100%.

*PA-PSRS is a secure, web-based system through which Pennsylvania hospitals, ambulatory surgical facilities, abortion facilities, and birthing centers submit reports of patient safety–related incidents and serious events in accordance with mandatory reporting laws outlined in the Medical Care Availability and Reduction of Error (MCARE) Act (Act 13 of 2002).28 All reports submitted through PA-PSRS are confidential and no information about individual facilities or providers is made public. †The MCARE Act (Act 13 of 2002)28 defines a serious event as “An event, occurrence or situation involving the clinical care of a patient in a medical facility that results in death or compromises patient safety and results in an unanticipated injury requiring the delivery of additional health care services to the patient.”

Patient Safety I Vol. 3 No. 1 I March 2021 I 13 When process failures involved more than one transition of care Reporting facilities also included the patient age and gender for point, we identified the transition of care point with the error each event. We grouped each of the 93 events by the following that most directly resulted in the patient harm described in the age groups: 0 through 5 years (young children), 6 through 18 event report. When events described multiple medications, cir- years (school-aged children), 19 through 35 years (young adults), cumstances of event discovery/confirmation, or additional care/ 36 through 64 years (middle-aged adults), and 65 years and older services, we independently included each of these in the analyses. (older adults). Patients 65 years and older most commonly were associated with serious events related to medication reconciliation (55.9%; 52 of 93). Events more often involved male patients (52.7%; Results 49 of 93) than female patients (47.3%; 44 of 93). In Figure 2, we present the percentage of events by age group and gender. Our initial query identified 10,100 events submitted to PA-PSRS from January 1, 2015, to August 31, 2020. After excluding events Qualitative Analysis reported as incidents (harm score A–D)§ events that were reported by facilities other than hospitals, and events that did not directly Transitions of Care relate to the process of medication reconciliation, we identified We identified the transition of care point that was associated 93 events that met inclusion criteria for further analysis. with the proximate cause for the event resulting in patient harm (Figure 3). Four events did not include enough information to Descriptive Analysis determine which transition of care was associated with the event (unknown). Most of the events in our study occurred at admission/ Figure 1 shows the breakdown of harm scores that reporting triage (69.9%; 65 of 93), and these often carried through the hos- facilities assigned to each serious event (n=93). Most events were pital admission (if admitted) and sometimes even to discharge or assigned a harm score of E (58.1%; 54 of 93) or F (38.7%; 36 of 93); transfer to another unit/facility. Another 16.1% (15 of 93) of events the remaining events were assigned a harm score of G (1.1%; 1 occurred during medication reconciliation at discharge. Events of 93) or I (2.2%; 2 of 93). None of the events in our analysis were related to discharge often went unrecognized until the patient assigned a harm score of H.

Figure 2: Percentage of Event Reports by Patient Age Group and Gender Involved in Serious Events Related to Medication Reconciliation Submitted to PA-PSRS From January 1, 2015–August 31, 2020, N=93

% of Events Female 30% 29.0% Male 26.9% 25%

20% 16.1% 15% 11.8% 10%

4.3% 4.3% 5% 3.2% 3.2% 0.0% 1.1% 0% 0–5 6–18 19–35 36–64 65+ Age Group (Years) Note: Due to rounding, percentages do not add up to 100%.

§The MCARE Act (Act 13 of 2002)28 defines an incident as “An event, occurrence or situation involving the clinical care of a patient in a medical facility which could have injured the patient but did not either cause an unanticipated injury or require the delivery of additional health care services to the patient. The term does not include a serious event.”

14 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Figure 3: Transition of Care Point Associated With the Proximate Cause for Serious Events Related to Medication Reconciliation Submitted to PA-PSRS From January 1, 2015–August 31, 2020, N=93

Point of Care

Admission/Triage 65

Discharge 15

Prior to/Directly Following a Procedure 7

Transfer to Another Unit/Facility 2

Unknown 4

0 10 20 30 40 50 60 70

Number of Events

Figure 4: Stage of the Medication Reconciliation Process at Which Process Failures Occurred and Medication Error Types Associated With Serious Events Involving Medication Reconciliation Submitted to PA-PSRS From January 1, 2015–August 31, 2020, N=93

Stage at Which the Medication Reconciliation Process Failure Occurred

Clinical Medication Order Entry/ Assessment/ Discharge Error Type Source Transcription Decision Orders Other Unknown Total Wrong Dose 7 21 2 4 1 - 35 Dose Omission 2 13 3 6 - 1 25

Wrong Medication 9 2 1 - 1 2 15

Duplicate Therapy - 2 6 - - - 8 Contraindication - - 3 - 2 - 5 Wrong Formulation - 1 - - 2 1 4 Unknown - - - - - 1 1 Total 18 39 15 10 6 5 93

Patient Safety I Vol. 3 No. 1 I March 2021 I 15 returned to the emergency room (ER) or was readmitted. The or illegible information provided by patients, family members, remaining events occurred either prior to or directly following a transferring facilities, old electronic records from a previous procedure or surgery (7.5%; 7 of 93), or during transfer to another admission, or using the wrong patient’s information. The result- unit or facility (2.2%; 2 of 93). ing medication error types were wrong dose (n=7), wrong medication (n=9), and dose omission (n=2). Process Failures and Medication Error Types Clinical assessments or decisions contributed to 16.1% (15 of We analyzed each event to determine the stage of the medication 93) of the events. In these events, the clinician either missed an reconciliation process at which failures most directly contributed important piece of clinical information or made a clinical deci- to the event (Figure 4). We also determined the type of medica- sion that resulted in a medication error that resulted in patient tion error that occurred as a result (Figure 4). Five events did harm. For example, events included decisions to stop a medica- not include enough information to determine the stage of the tion without tapering, to hold a medication without offering a medication reconciliation process that contributed to the patient therapeutic alternative, or to discontinue a medication without harm (unknown). stopping a related medication (e.g., discontinuing furosemide without holding/monitoring potassium). Other examples include Failures in the process of electronic order entry or transcrip- decisions to order new medications with similar actions as home tion of information to the documented home medication list medications, not assessing for when the last medication was taken occurred in 41.9% (39 of 93) of the events. These events included or medication patch was applied, or not assessing for known drug missed orders, decimal place errors, orders where the total daily allergies. Medication error types were duplicate therapy (n=6), dosage was confused for individual dosages, orders for the wrong contraindication (n=3), dose omission (n=3), wrong dose (n=2), type of insulin or dosing errors for U-500 insulin, or entry of and wrong medication (n=1). duplicate orders for the same medication. These resulted in the following medication error types: wrong dose (n=21), dose Process failures in medication reconciliation related to discharge omission (n=13), wrong medication (n=2), duplicate therapy orders contributed to 10.8% (10 of 93) of the events. Dosage errors, (n=2), and wrong formulation (n=1). omitted orders, and instructions that were unclear or that differed from the discharge medication list contributed to these types of Failures involving the source of information contributed to 19.4% events. The resulting medication error types were dose omission (18 of 93) of the events. These events involved incorrect, outdated, (n=6) and wrong dose (n=4).

Figure 5: Frequency of Medication Classes by System in Serious Events Related to Medication Reconciliation Submitted to PA-PSRS From January 1, 2015–August 31, 2020, N=128

Medication Class

Neurologic/Psychiatric 50

Cardiovascular 42

Endocrine 14

Electrolyte/Vitamin 6

Anti-Infective 5

Anti-Inflammatory 4

Immune-Modulating 3

Other 4

0 10 20 30 40 50 60

Number of Medications

16 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Several other process failures occurred during the medication rec- 4 additional medications were not named but were specified by onciliation process (6.5%; 6 of 93). These included communication a class, such as a cardiovascular medication or a nonsteroidal or procedural errors in which the physician was not made aware anti-inflammatory drug (NSAID), so we analyzed a total of 128 that a patient had taken a contraindicated medication prior to a medications. We classified each medication by system Figure( 5) procedure, the home medication list was incorrectly marked as and then by pharmacologic class. Neurologic or psychiatric med- updated, medications were ordered that were not on formulary or ications were the most common (39.1%; 50 of 128), followed by that required special instructions, or instructions for storage and cardiovascular medications (32.8%; 42 of 128), endocrine med- administration of home medications were not clear. Medication ications (10.9%; 14 of 128), electrolytes or vitamins (4.7%; 6 of error types that resulted were wrong formulation (n=2), contra- 128), anti-infectives (3.9%; 5 of 128), anti-inflammatories (3.1%; indication (n=2), wrong medication (n=1), and wrong dose (n=1). 4 of 128), immune-modulators (2.3%; 3 of 128), gastrointestinal medications (0.8%; 1 of 128), hepatic medications (0.8%; 1 of 128), Circumstances of Event Discovery or Confirmation respiratory medications (0.8%; 1 of 128), and urologic medications (0.8%; 1 of 128). We analyzed the events to determine the circumstances of how, when, or who was involved in discovering or confirming the event. The most common pharmacologic classes of neurologic or psychi- Many events included more than one indication that led to the atric medications were anticonvulsants (42.0%; 21 of 50), opioids event being discovered or confirmed. Six events did not include (14.0%; 7 of 50), benzodiazepines (12.0%; 6 of 50), antipsychotics enough information to determine what additional services were (10.0%; 5 of 50), muscle relaxants (10.0%; 5 of 50), and antide- required (unknown). Most events described that a change in pressants (6.0%; 3 of 50). Other pharmacologic classes with only patient condition was one of the first indications that a medication one medication were a dopamine agonist, a potassium channel reconciliation error had occurred (76.3%; 71 of 93). Chart/case blocker, and a mood stabilizer. Patients with errors in dosing reviews prompted by the physician team during changes to care of anticonvulsants or benzodiazepines most often experienced plans, a change in patient condition, or subsequent investigation seizures. Patients with errors in dosing of opioids experienced a following unexpected patient outcomes or review of medication range of symptoms, including hypoxia, respiratory distress, loss orders by the pharmacist during daily reviews were mentioned in of consciousness, lethargy, and somnolence. 26 events (28.0%; 26 of 93). The patient, family, caregiver, guardian, primary care provider, and/or patient’s community pharmacy The most common pharmacologic classes of cardiovascular med- were also described in events as either identifying or confirming ications were beta-blockers (21.4%; 9 of 42), diuretics (16.7%; 7 of that a medication discrepancy occurred (24.7%; 23 of 93). Some 42), anticoagulants (14.3%; 6 of 42), angiotensin-converting-en- events described discovery at or after discharge or transfer that zyme (ACE) inhibitors/angiotensin II receptor blockers (9.5%; 4 resulted in a return to the ER, readmission, or return to the pro- of 42), alpha agonists (9.5%; 4 of 42), calcium channel blockers cedure area (19.4%; 18 of 93). Some events also mentioned that (9.5%; 4 of 42), and antihyperlipidemics (7.1%; 3 of 42); other phar- abnormal lab values indicated the need for further assessment macologic classes with only one medication were alpha blockers, to determine whether a medication error had occurred (10.8%; antiarrhythmics, neprilysin inhibitors, and nitrates. Patients with 10 of 93). Several miscellaneous indications included electronic errors in dosing of ACE inhibitors, alpha agonists, alpha blockers, alerts related to critical lab values or during medication order beta blockers, and calcium channel blockers experienced changes entry/verification, dispensing, or administration; a billing griev- in heart rhythm, heart rate (e.g., bradycardia or tachycardia), ance (for unexpected charges); or excessive bleeding during a and blood pressure (e.g., hypotension). Errors in dosing of anti- procedure (3.2%; 3 of 93). coagulants resulted in bleeding, stroke, a deep vein thrombosis, and death. Additional Care or Services The most common pharmacologic class of endocrine medications We also identified additional care or services required as a result was insulin (71.4%; 10 of 14). Other classes were oral antidiabetic of each event. Many events described multiple additional services agents (21.4%; 3 of 14) and a hormone (7.1%; 1 of 14). Patients that were required to correct the error or support the patient. Nine with errors in dosing of insulin and oral antidiabetics experienced events did not include enough information to determine what hyperglycemia or hypoglycemia, depending on the nature of the additional services were required (unknown). Many of the events error. The most common class of electrolytes/vitamins was ele- indicated the need for readmission, hospitalization, emergency ments (83.3%; 5 of 6). Among 4 patients with errors in dosing of care, intensive care, or transfer to a higher level of care (58.0%; potassium, all experienced hyperkalemia as a result of the error. 54 of 93). Other additional care or services included additional The two classes of anti-infectives were antibiotics (60.0%; 3 of 5) medications for reversal or support (37.6%; 35 of 93); additional and antivirals (40.0%; 2 of 5). monitoring or testing (32.3%; 30 of 93); an increased length of stay (8.6%; 8 of 93); additional consults (6.5%; 6 of 93); the need The most common pharmacologic classes across all medications for additional oxygen, intubation, or ventilator support (6.5%; 6 (those involved in 5 or more events) were anticonvulsants (16.4%; of 93); a code or rapid response team (RRT) being called (4.3%; 4 21 of 128), insulin (7.8%; 10 of 128), beta blockers (7.0%; 9 of 128), of 93); and the need for rehabilitation in a nursing home or long- diuretics (5.5%; 7 of 128), opioids (5.5%; 7 of 128), anticoagulants term care (2.2%; 2 of 93). (4.7%; 6 of 128), benzodiazepines (4.7%; 6 of 128), antipsychotics (3.9%; 5 of 128), and muscle relaxants (3.9%; 5 of 128); the specific Medications medications in each of these pharmacologic classes is detailed in Figure 6. The most common single medications were levetirace- We reviewed all 93 events to identify the medication or medi- tam (5.5%; 7 of 128), baclofen (3.9%; 5 of 128), clonazepam (3.1%; cations involved; 92 events listed one or more medications. A 4 of 128), clonidine (3.1%; 4 of 128), and potassium (3.1%; 4 of 128). total of 124 specific medications were mentioned by name, and

Patient Safety I Vol. 3 No. 1 I March 2021 I 17 Figure 6: Most Common Pharmacologic Classesa in Serious Events Related to Medication Reconciliation Submitted to PA-PSRS From January 1, 2015–August 31, 2020, n=76

Insulin n=10

aMost common pharmacologic classes were defined as those with 5 or more occurrences.

18 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Table 1: Strategies to Reduce the Risk of Medication Reconciliation Errors

Consider use of a dedicated pharmacy role for medication reconciliation to assist with the following processes5,16-18,20,30-32 • Collecting medication histories • Calling community pharmacies, primary or specialty physicians, family members, or care home for clarifications • Reconciling medication discrepancies with providers during admission, care transitions, Defined Roles/ changes in care plan, transfers, and discharge Responsibilities • Educating providers and patients when medication questions arise

If a dedicated role is not possible for all patients due to resource constraints, consider use of a dedicated pharmacy role for the following high-risk patients or transitions of care5,6,11,16,19,20,22,27 • Elderly (age 65 and older) • Polypharmacy (as defined by the facility) • Specified transitions of care, such as admissions through the emergency department

Include the purpose or reason the patient is taking the medication on the home medication list Medication and throughout all documentation systems for medication orders, care planning, and discharge Indication planning33

Develop standardized processes to ensure all clinicians follow consistent procedures throughout the continuum of care2,5,10,23,29 • Standardized processes such as interview questions or prompts to collect accurate medication histories ■ Include prompts for use of over-the-counter medications, patches, herbals, and other drug or alcohol use ■ Include prompts to verify patient identification for communication regarding medications over the phone or paper lists that are exchanged Standardized • Standardized processes to provide discharge medication instructions Processes ■ Highlight medications that are new or have been stopped or changed ■ Develop processes to communicate medication changes to patient’s pharmacy ■ Develop a standardized format for medication orders, such as directions for taking, purpose of the medication, and expected side effects ■ Determine if prescriptions need to be written and whether patient is able to fill the prescription for all home medications ordered ■ Encourage patient to take this documentation to all follow-up physician appointments and to the community pharmacy ■ Add updated medications to patient portal information Review facility lists and processes for high-alert medications25,42 • Consider adding anticonvulsants to facility lists of medications that trigger additional alerts, monitoring, or laboratory testing Triggers • Identify triggers and bundled orders at the point of ordering to ensure appropriate and Alerts monitoring is ordered for the medications prescribed • Identify triggers and bundled orders in the electronic medical record to ensure laboratory or other monitoring results are routinely reviewed

Develop shared electronic medication lists40,41 • Develop technologies to assist sharing of live/current patient medication histories Information across care areas, with community pharmacies and primary care physicians, and between Technology hospital visits • Encourage and educate patients on the use of patient portals on mobile devices to reference and double-check their current medication list Discussion concerning the potential hazards of errors in dosing of anticonvulsants is limited. Two studies specifically in the pediatric popula- Medication reconciliation is complex and requires extensive tion have shown that up to 24% of pediatric patients with epilepsy resources, effective processes, clear and consistent communica- had errors in orders for anticonvulsants during transitions of care tion and documentation, and active participation by the patient resulting in an additional risk for seizures.38, 39 Thus, facilities should and healthcare providers. It also necessitates that these steps consider adding anticonvulsants to their facility lists of medications be implemented at every transition point and change in therapy that trigger additional alerts, monitoring, or laboratory testing to throughout the continuum of care. Medication reconciliation prevent errors from occurring or provide an early warning system errors have potential to result in patient harm. A 2012 study of to identify errors before they cause harm. medication errors related to medication reconciliation reported to PA-PSRS showed that 67.3% of events reached the patient, and We found that process failures related to obtaining accurate infor- of these events, 3.6% resulted in patient harm and 17.4% required mation from the source contributed to 19.4% of serious medi- monitoring to prevent harm.29 cation reconciliation events. In addition, in 24.7% of analyzed events, a family member, guardian, care home, primary care In Table 1, we present strategies to reduce the risk for medi- provider, or community pharmacy was able to help identify or cation reconciliation errors. Standardization and development confirm that a medication reconciliation event had occurred. of clear roles and responsibilities for medication reconciliation Thus, standarized processes for validating and communicating can help to ensure accuracy and prevent errors that can lead to medication histories during transitions of care, such as standard- 2, 5-11 patient harm. Researchers have described decreased read- ized medication history interview questions and standardized missions, fewer returns to the ER, and up to a 66% reduction in discharge processes, can help prevent medication reconciliation medication discrepancies when using standardized processes events.6, 8 In addition, electronic medical records systems that can and a dedicated pharmacist or pharmacy technician to collect communicate home medications across computer systems or medication histories and coordinate the medication reconcilia- through patient portals on mobile devices, which can be accessed 5,16-18,20,30-32 tion process. by patients at physician offices, the pharmacy, during hospital admission, or discharge, can also help to improve medication Our analysis indicated that most of the medication reconciliation histories and prevent medication reconciliation errors.40, 41 errors occurred during admission/triage (69.9%), and another 16.1% occurred during discharge. The literature suggests that because medication reconciliation can be resource intensive, Limitations interventions focused on a specific transition point (such as in the Despite mandatory event-reporting laws in Pennsylvania, our data emergency department, on admission, or at the time of discharge) are subject to the limitations of self-reporting. Portions of this can help reduce medication reconciliation errors.5,11,16,19,22,27 analysis were limited by the amount and quality of information Furthermore, 55.9% of the events in our study occurred in those provided in the free-text and optional data fields. It is also import- 65 years and older. The literature also suggests a focus on specific ant to note that medication reconciliation events—especially those high-risk groups (such as the elderly, those taking many home that occur at discharge—may not be immediately realized, and medications, or those with comorbidities) for more intensive therefore, not reported. Thus, the number of serious events and medication reconciliation interventions.6,11,20 severity of the outcomes may be more substantial than those captured through PA-PSRS reporting. Knowing why a patient takes a particular medication can help to prevent many types of medication reconciliation errors by helping a clinician to better understand how the patient’s clinical Notes history and diagnoses correlate with the medication list. The Joint Commission already recommends including the indication for a This analysis was exempted from review by the Advarra medication along with the patient’s diagnosis during continuous Institutional Review Board. care planning.33 Thus, ensuring that the medication indication is included on medication orders, home medication lists, patient dis- Conclusion charge sheets, and pharmacy prescriptions can further enhance clinician critical thinking to prevent medication errors as well Medication reconciliation continues to pose significant challenges as enhance the patient’s understanding of why they are taking for patient safety due to the complex set of processes involved particular medications. throughout the continuum of patient care. In our analysis of serious events, we identified process breakdowns across many In our analysis, we identified several pharmacologic classes that are transitions of care, with the most frequent involving the process currently included on the Institute for Safe Medication Practices of order entry/transcription and occurring during admission/ (ISMP) acute care high-alert medication list and that can result triage. Of particular note, we found that anticonvulsants were the in patient harm if used in error, such as opioids, selected beta most common pharmacologic class of medications involved in blockers, anticoagulants, and insulin.25,34-37 However, the anticon- serious events. Risk reduction strategies, such as defined roles for vulsant pharmacologic class was most predominant in our analysis medication reconciliation; listing reasons/indications for medica- of serious events (16.4%; 21 of 128 medications), and this class is tions; and consideration for adding anticonvulsants to facility pro- not currently included on the ISMP acute care high-alert medi- cesses for high-alert medications that trigger additional warnings, cation list.25 Although our analysis revealed that events involving monitoring, or laboratory testing, may help reduce patient harm anticonvulsants were common across the age spectrum, research associated with errors in the medication reconciliation process.

20 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 References Hospital: An Interactive Case-Based Session for Internal Medicine Residents. MedEdPORTAL. 2018;14:10770. Epub 1. National Safety Goals Effective July 2020 for the Hospital 2019/02/26. doi: 10.15766/mep_2374-8265.10770. PubMed PMID: Program [Online]. Oakbrook Terrace, IL: The Joint Commission; 30800970; PubMed Central PMCID: PMCPMC6342339. 2020 [cited 2020 October 18, 2020]. Available from: https://www. jointcommission.org/-/media/tjc/documents/standards/nation- 14. Holt K, Thompson A. Implementation of a Medication al-patient-safety-goals/2020/npsg_chapter_hap_jul2020.pdf. Reconciliation Process in an Internal Medicine Clinic at an Academic Medical Center. Pharmacy. 2018;6(2):26. doi: 10.3390/ 2. Steeb D, L; W. Improving Care Transitions: Optimizing pharmacy6020026. Medication Reconciliation. USA: American Pharmacists Association and American Society of Health-System 15. Berthe A, Fronteau C, Le Fur É, Morin C, Huon JF, Rouiller- Pharmacists; 2012. p. 1-15. Furic I, et al. Medication Reconciliation: A Tool to Prevent Adverse Drug Events in Geriatrics Medicine. Geriatr Psychol 3. Reconcile Medications at All Transition Points [Online]. Neuropsychiatr Vieil. 2017;15(1):19-24. Epub 2017/01/26. doi: Boston, MA: Institute for Healthcare Improvement; [cited 2020 10.1684/pnv.2016.0642. PubMed PMID: 28120773. October 18]. Available from: http://www.ihi.org/resources/ Pages/Changes/ReconcileMedicationsatAllTransitionPoints. 16. Mills PR, McGuffie AC. Formal Medicine Reconciliation aspx. Within the Emergency Department Reduces the Medication Error Rates for Emergency Admissions. Emerg Med J. 4. Reconcile Medications in Outpatient Settings [Online]. 2010;27(12):911-5. doi: 10.1136/emj.2009.082255. Boston, MA: Institute for Healthcare Improvement; [updated 2020; cited 2020 October 18]. Available 17. Karaoui LR, Chamoun N, Fakhir J, Abi Ghanem W, Droubi from: http://www.ihi.org/resources/Pages/Changes/ S, Diab Marzouk AR, et al. Impact of Pharmacy-Led Medication ReconcileMedicationsinOutpatientSettings.aspx. Reconciliation on Admission to Internal Medicine Service: Experience in Two Tertiary Care Teaching Hospitals. BMC 5. Mueller SK, Sponsler KC, Kripalani S, Schnipper JL. Health Serv Res. 2019;19(1). doi: 10.1186/s12913-019-4323-7. Hospital-Based Medication Reconciliation Practices: A Systematic Review. Arch Intern Med. 2012;172(14):1057-69. doi: 18. Mekonnen AB, McLachlan AJ, Brien J-AE. Pharmacy-Led 10.1001/archinternmed.2012.2246. Medication Reconciliation Programmes at Hospital Transitions: A Systematic Review and Meta-Analysis. J Clin Pharm Ther. 6. Midlöv P, Bahrani L, Seyfali M, Höglund P, Rickhag E, 2016;41(2):128-44. doi: 10.1111/jcpt.12364. Eriksson T. The Effect of Medication Reconciliation in Elderly Patients at Hospital Discharge. Int J Clin Pharm. 2012;34(1):113- 19. Tam VC, Knowles SR, Cornish PL, Fine N, Marchesano 9. doi: 10.1007/s11096-011-9599-6. R, Etchells EE. Frequency, Type and Clinical Importance of Medication History Errors at Admission to Hospital: A 7. Barnsteiner J. Medication Reconciliation. In: Hughes R, edi- Systematic Review. CMAJ. 2005;173(5):510-5. Epub 2005/09/01. tor. Patient Safety and Quality: An Evidence-Based Handbook doi: 10.1503/cmaj.045311. PubMed PMID: 16129874; PubMed for Nurses. Rockville, MD: Agency for Healthcare Research and Central PMCID: PMCPMC1188190. Quality 2008. 20. Cornu P, Steurbaut S, Leysen T, De Baere E, Ligneel C, 8. Munck LK, Hansen KR, Mølbak AG, Balle H, Kongsgren S. Mets T, et al. Effect of Medication Reconciliation at Hospital The Use of Shared Medication Record as Part of Medication Admission on Medication Discrepancies During Hospitalization Reconciliation at Hospital Admission is Feasible. Dan Med J. and at Discharge for Geriatric Patients. Ann Pharmacother. 2014;61(5). 2012;46(4):484-94. Epub 2012/03/15. doi: 10.1345/aph.1Q594. 9. Medication Reconciliation [Online]. Rockville, MD: Agency PubMed PMID: 22414793. for Healthcare Research and Quality; [cited 2020 October 19]. 21. da Silva BA, Krishnamurthy M. The Alarming Reality of Available from: https://digital.ahrq.gov/ahrq-funded-projects/ Medication Error: A Patient Case and Review of Pennsylvania emerging-lessons/medication-reconciliation. and National Data. J Community Hosp Intern Med Perspect. 10. Advancing Successful Care Transitions to Improve 2016;6(4):31758-. doi: 10.3402/jchimp.v6.31758. PubMed PMID: Outcomes [Online]. Philadelphia, PA: Society of Hospital 27609720. Medicine; [cited 2020 October 19]. Available from: https://www. 22. Pippins JR, Gandhi TK, Hamann C, Ndumele CD, Labonville hospitalmedicine.org/clinical-topics/care-transitions/. SA, Diedrichsen EK, et al. Classifying and Predicting Errors 11. Williams M, Budnitz T, Coleman E, Forth V, Greenwald J, of Inpatient Medication Reconciliation. J Gen Intern Med. Halasyamani L, et al. Boost: Better Outcomes for Older Adults 2008;23(9):1414-22. Epub 2008/06/20. doi: 10.1007/s11606-008- Through Safe Transitions [Online]. Philadelphia, PA: Society of 0687-9. PubMed PMID: 18563493; PubMed Central PMCID: Hospital Medicine; 2010 [cited 2020 October 19]. Available from: PMCPMC2518028. http://tools.hospitalmedicine.org/Implementation/Workbook_ 23. Allaudeen N, Vidyarthi A, Maselli J, Auerbach A. Redefining for_Improvement.pdf. Readmission Risk Factors for General Medicine Patients. J Hosp 12. Nassaralla CL, Naessens JM, Chaudhry R, Hansen MA, Med. 2011;6(2):54-60. doi: 10.1002/jhm.805. Scheitel SM. Implementation of a Medication Reconciliation 24. High-Alert Medications Require Heightened Vigilance Process in an Ambulatory Internal Medicine Clinic. Qual Saf [Internet]. Boston, MA: Institue for Heatlhcare Improvement; Health Care. 2007;16(2):90-4. doi: 10.1136/qshc.2006.021113. 2020 [updated 2020; cited 2020 November 28th]. Available from: 13. McShane M, Stark R. Medication Reconciliation in the http://www.ihi.org/resources/Pages/ImprovementStories/

Patient Safety I Vol. 3 No. 1 I March 2021 I 21 HighAlertMedsHeightenedVigilance.aspx#:~:text=The%20 38. Jones C, Kaffka J, Missanelli M, Dure L, Ness J, Funkhouser most%20common%20types%20of,a%20dangerously%20 E, et al. Seizure Occurrence Following Nonopitmal slow%20heart%20rate). Anticonvulsant Medication Management During the Transition 25. ISMP List of High-Alert Medications in Acute Care Settings Into the Hospital. J Child Neurol. 2012;28(10):1250-8. [Internet]. Horsham, PA: Institute for Safe Medication 39. Jones C, Missanelli M, Dure L, Funkhouser E, Kaffka J, Practices; [cited 2020 Nov 9]. Available from: https://forms.ismp. Kilgore M, et al. Anticonvulsant Medication Errors in Children org/Tools/institutionalhighAlert.asp. With Epilepsy During the Home-to-Hospital Transition. J Child 26. Institute of Medicine Committee on Quality of Health Care Neurol. 2012;28(3):314-20. in A. To Err is Human: Building a Safer Health System. Kohn 40. Dendere R, Slade C, Burton-Jones A, Sullivan C, Staib LT, Corrigan JM, Donaldson MS, editors. Washington (DC): A, Janda M. Patient Portals Facilitating Engagement With National Academies Press (US) Copyright 2000 by the National Inpatient Electronic Medical Records: A Systematic Review. J Academy of Sciences. All rights reserved.; 2000. Med Internet Res. 2019;21(4):e12779. doi: 10.2196/12779. 27. Khansa SA, Mukhtar A, Abduljawad M, Aseeri M. Impact 41. Fitzgerald RJ. Medication Errors: The Importance of an of Medication Reconciliation Upon Discharge on Reducing Accurate Drug History. Br J Clin Pharmacol. 2009;67(6):671-5. Medication Errors. J Pharmacovigil. 2016;04(06). doi: doi: 10.1111/j.1365-2125.2009.03424.x. PubMed PMID: 19594536. 10.4172/2329-6887.1000222. 42. Your High-Alert Medication List—Relatively Useless Without 28. Medical Care Availability and Reduction of Error (MCARE) Associated Risk-Reduction Strategies [Online]. Horsham, PA: Act, pub. L. No. 154 stat. 13 (2002). ISMP (Institute for Safe Medication Practices); [cited 2020 29. Gao T, Gaunt MJ. Breakdowns in the Medication 27 Jan]. Available from: https://www.ismp.org/resources/ Reconciliation Process. Pa Patient Saf Advis. 2013;10(4):125-36. your-high-alert-medication-list-relatively-useless-without-associated-risk-reduction. 30. Gillespie U, Alassaad A, Henrohn D, Garmo H, Hammarlund-Udenaes M, Toss H, et al. A Comprehensive Pharmacist Intervention to Reduce Morbidity in Patients 80 About the Authors Years or Older. Arch Intern Med. 2009;169(9):894. doi: 10.1001/ archinternmed.2009.71. Amy Harper ([email protected]) is an infection prevention analyst for the Patient Safety Authority. She has more than 25 years 31. Hammad EA, Bale A, Wright DJ, Bhattacharya D. Pharmacy of combined microbiology/nursing experience, including roles Led Medicine Reconciliation at Hospital: A Systematic Review in infection prevention, medical-surgical and critical care nurs- of Effects and Costs.Res Social Adm Pharm. 2017;13(2):300-12. ing, teaching, and virology research. She is board certified in doi: 10.1016/j.sapharm.2016.04.007. medical-surgical nursing (CMSRN) and in infection control and 32. Hammour KA, Farha RA, Basheti I. Hospital Pharmacy epidemiology (CIC), and is a member of the Academy of Medical- Medication Reconciliation Practice in Jordan: Perceptions Surgical Nurses (AMSN), The Society for Healthcare Epidemiology and Barriers. J Eval Clin Pract. 2016;22(6):936-41. doi: 10.1111/ of America (SHEA), and the Association for Professionals in jep.12565. Infection Control and Epidemiology (APIC). 33. Joint Commission: Performance Measurement FAQs: Elizabeth Kukielka is a patient safety analyst on the Data Science Continuing Care Plan-Discharge Medications Indications for and Research team at the Patient Safety Authority. Before joining Use [Internet]. Oakbrook Terrace, IL: The Joint Commission; the PSA, she was a promotional medical writer for numerous pub- 2020 [cited 2020 Dec 9]. Available from: https://manual.jointcom- lications, including Pharmacy Times and The American Journal of mission.org/Manual/Questions/UserQuestionId03Hbips100019. Managed Care. Kukielka also worked for a decade as a community 34. CRH N, Delgado J, HF G. Calcium and Beta Receptor pharmacist and pharmacy manager, with expertise in immuni- Antagonist Overdose: A Review and Update of Pharmacological zation delivery, diabetes management, medication therapy man- Principles and Management. Semin Respir Crit Care Med. agement, and pharmacy compounding. 2002;23(1):20-5. Rebecca Jones is director of Data Science and Research at the 35. Cousins D, Rosario C, Scarpello J. Insulin, Hospitals and Patient Safety Authority, where she also founded and serves as Harm: A Review of Patient Safety Incidents Reported to the director of the Center of Excellence for Improving Diagnosis. National Patient Safety Agency. Clin Med. 2011;11(1):28-30. doi: Her previous roles at the PSA include director of Innovation and 10.7861/clinmedicine.11-1-28. Strategic Partnerships and regional patient safety liaison. Before 36. Mc Donnell C. Opioid Medication Errors in Pediatric joining the PSA, Jones served in various roles leading patient Practice: Four Years’ Experience of Voluntary Safety Reporting. safety efforts and proactively managing risk in healthcare orga- Pain Res Manag. 2011;16(2):93-8. doi: 10.1155/2011/739359. nizations. She currently is chair of the Practice Committee of 37. Valentine D, Gaunt M, Grissinger M. Identifying Patient the Society to Improve Diagnosis in Medicine and serves on the Harm From Direct Oral Anticoagulants. Pa Patient Saf Advis. Advisory Committee of the Coalition to Improve Diagnosis. 2018;15(2). This article is published under the Creative Commons Attribution- NonCommercial license.

22 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Categorization of Medication Safety Errors in Ambulatory Electronic Health Records

By Karen Paul Zimmer, MD, MPH◆◊, David C. Classen, MD, MS† & Jessica M. Cole, BS† DOI: 10.33940/med/2021.3.2

Med Error

Corresponding author Jefferson Health †University of Utah Disclosure: The authors declare that they have no relevant or material financial interests.

Patient Safety I Vol. 3 No. 1 I March 2021 I 23 reventable medication errors continue to affect the quality and consistency in the delivery of care. While numerous studies on medication safety have been performed in the inpatient setting, a review of ambulatory patient safety by the American Medical Association found that medication safety errors were the most frequent safety problems in the outpatient arena. The leading cause of ambulatory safety problems, adverse drug Pevents (ADEs), are common, with estimates of more than 2 million ADEs each year in the ambulatory Medicare population alone, and these events are frequently preventable. We conducted an environmental scan that allowed us to create our own categorization schema of medication safety errors in electronic healthcare records (EHRs) found in the outpatient setting and observed which of these were additionally supported in the literature. This study combines data from the Collaborative Healthcare Patient Safety Organization (CHPSO), with several key articles in the area of medication errors in the EHR era.

Method: To best utilize the various EHR ambulatory medication events submitted into CHPSO’s database, we chose to create a framework to bucket the near misses or adverse events (AEs) submitted to the database. This newly created categorization scheme was based on our own drafted categorization labels of events, after a high- level review, and from two leading articles on physician order entry. Additionally, we conducted a literature review of computerized provider order entry (CPOE) medication errors in the ambulatory setting. Within the newly created categorization scheme, we organized the articles based on issues addressed so we could see areas that were supported by the literature and what still needed to be researched.

Results: We initially screened the CHPSO database for ambulatory safety events and found 25,417 events. Based on those events, an initial review was completed, and 19,242 events were found in the “Medication or Other Substance” and “Other” categories, in which the EHR appeared to have been a potential contributing factor. This review identified a subset of 2,236 events that were then reviewed. One hundred events were randomly selected for further review to identify common categories. The most common categories in which errors occurred were orders in order sets and plans (n=12) and orders crossing or not crossing encounters (n=12), incorrect order placed on correct patient (n=10), orders missing (n=8), standing orders (n=8), manual data entry errors (n=6), and future orders (n=6).

Conclusion: There were several common themes seen in this analysis of ambulatory medication safety errors related to the EHR. Common among them were incorrect orders consisting of examples such as dose errors or ordering the wrong medication. The manual data entry errors consisted of height or weight being entered incorrectly or entering the wrong diagnostic codes. Lastly, different sources of medication safety information demonstrate a diversity of errors in ambulatory medication safety. This confirms the importance of considering more than one source when attempting to comprehensively describe ambulatory medication safety errors.

Introduction

Understanding the Problem Almost every study of inpatient safety has shown that medication errors remain the most common safety issues that patients experience; this has been fairly consistent in studies over most of the last 40 years.1 Similarly, a review of ambulatory patient safety by the American Medical Association found that medication safety errors were the most frequent safety problems in the outpatient arena.1-4 The leading cause of ambulatory safety problems, adverse drug events (ADEs), are common, with estimates of more than 2 million ADEs each year in the Medicare population alone, and these events are frequently preventable.5 Studies in private clinic settings suggest that patients in younger populations still experience ADEs about 25% of the time.6 The frequency and severity of medication safety errors across Improvement, Brigham and Women’s Hospital, and CHPSO, cre- the continuum of care and their relationship to errors in medi- ated and piloted the Ambulatory Electronic Health Record (EHR) cation prescribing has been a driving force for the adoption of Evaluation Tool, which has been shaped by the ambulatory med- electronic health record systems (EHRs), in both the inpatient and ication safety analysis presented in this paper. ambulatory patient care settings. Over the last decade, because of the Centers for Medicare & Medicaid Services (CMS) Meaningful As we began developing this new ambulatory EHR safety tool, we Use Incentive Program, there has been a remarkable adoption conducted this study to help inform our work. Specifically, we of EHRs.7 Use of EHR with computerized provider order entry reviewed deidentified data from CHPSO’s database of medication (CPOE) has resulted in a marked increase in the number of med- safety EHR errors in the outpatient setting. We conducted an ication prescriptions ordered electronically; it is estimated that environmental literature scan that allowed us to create our own in 2017 this number exceeded 80%.8,9 It is abundantly clear that compilated categorization schema of EHR errors found in the ambulatory EHRs with CPOE can have a medication safety impact. outpatient setting and observed which of these were additionally A recent study investigated whether physicians who meet the supported in the literature. meaningful use stage 2 threshold for e-prescribing (greater than 50% of prescriptions are e-prescribed) have lower rates of ADEs among their diabetic patients. That study found that e-prescrib- Methods ing to Medicare beneficiaries with diabetes was associated with reduced risk of hospital or emergency department visits for hypo- Collaborative Healthcare Patient Safety Organization (CHPSO) glycemia or ADEs related to antidiabetic medications.10 and CHPSO Database To better understand which types of medication safety errors are Numerous other studies have shown that EHRs with both CPOE encountered in the outpatient setting, we examined event reports and advanced functionality, such as decision support, can improve from CHPSO’s database to help inform the project of what errors medication safety, although it is not clear that these benefits are are encountered in the outpatient setting. Our report outlines the widely realized in the broad scope of actual use of these systems.1,7 types of errors encountered in the outpatient setting and which While this was first observed in the inpatient setting, this has also areas were supported by the literature. been observed in ambulatory EHR systems as well. In a study of prescribing errors in the leading commercial ambulatory CHPSO collects reports of events, near misses, and unsafe con- EHR vendor systems, 1 in 10 computer-generated prescriptions ditions from its member organizations. These reports are volun- included at least one error, of which a third had the potential for tarily generated by staff in the organizations and then voluntarily harm. The number, type, and severity of errors varied by EHR reported to CHPSO. Organizations may report some or all events vendor system used and suggests that some systems may be better and may selectively report in some categories and not in others. 11 at preventing medication errors than others. Most organizations report all event categories. CHPSO maintains the reports in a database in a standardized schema termed the From a patient’s perspective, these medication safety prescribing “Common Formats.” The Common Formats were developed by problems are often opaque and the assumption by patients and the Agency for Healthcare Research and Quality (AHRQ) in con- families is that our modern health systems, with all the automa- junction with the National Quality Forum (NQF). tion in place, prescribe medications safely and reliably. No recent story has shaken that confidence more than a Chicago Tribune As provider organizations often use nonstandard schema for their article in 2016, about how reporters were able get prescriptions reports, CHPSO provides a mapping service from the provider’s for two drugs (with lethal drug interactions) filled together at taxonomy to the Common Formats. CHPSO’s database records a 12 almost half the pharmacies in the Chicago area. Ironically this null value when the provider does not maintain a correspond- vulnerability had been previously demonstrated with a flight sim- ing field or corresponding answer for that field. Consequently, ulator of unsafe orders initially developed by the Institute for Safe the database has many missing fields. However, the text fields Medication Practices (ISMP) and sent to hundreds of pharmacies represent the richest value in terms of understanding risks in for testing; it revealed in 1999 and again in 2005 that most serious healthcare, so the structure fields are predominantly used for medication safety errors were missed by the operational elec- preliminary filtering prior to textual analysis. tronic pharmacy systems. Additionally, CHPSO has developed a mapping from words in the Based on the ISMP work outlined above, a simulation tool was cre- reports to the RxNorm ingredient names to normalize the drug ated over almost a 10-year period by researchers at the University data. For example, references in different reports to Tylenol and of Utah and Brigham and Women’s Hospital that found medication acetaminophen will be treated the same by using the mapping. safety vulnerabilities in the actual operation of inpatient EHR sys- This allows accurate tabulation of drug usage in events. tems with CPOE.13-16 This tool was eventually used to help develop the EHR Flight Simulator, which nearly 2,000 U.S. hospitals have The U.S. National Library of Medicine maintains RxNorm as part been using every year to improve their medication safety by opti- of their Unified Medical Language System (UMLS); RxNorm pro- mizing their EHRs. vides normalized names for drugs with linkages to other commonly used medication taxonomies. Because of the success of the inpatient tool, the Gordon and Betty Moore Foundation funded an ambulatory version of the tool. The Figure 1 delineates the way the events were selected. University of Utah, in partnership with the Institute for Healthcare

Patient Safety I Vol. 3 No. 1 I March 2021 I 25 Figure 1. Event Selection Process

25,417 Outpatient Events

19,242 “Medication or Other Substance” and “Other/ Uncategorized” events

2,236 Potential CPOE events*

100 randomly selected cases for review

*Exclusions, e.g., allergic reaction while on their standard medication within “Other” excluded falls and pressure injuries, for example, that were miscategorized

We developed a new ambulatory medication categorization scheme in three basic steps:

1. To best utilize the various EHR ambulatory events submitted into CHPSO’s database (as described above), we chose to create a framework to bucket the near misses or adverse events submitted to the database. This newly created categorization scheme was based on three sources:

● Our own draft categorization labels of events after a high-level review ● Computerized Prescriber Order Entry Medication Safety (CPOEMS): Uncovering and Learning From Errors17 ● “Computerized Physician Order Entry–Related Medication Errors: Analysis of Reported Errors and Vulnerability Testing of Current Systems"18

2. We then coded the three schemes and consolidated them based on duplicates either in wording or exact entries. The original list can be seen in Online Supplement Appendix Tables A1–A3. 3. We conducted a literature review of CPOE medication errors in the ambulatory setting. Within the newly created categorization scheme, we organized the articles based on errors addressed. Method for searching PubMed: ● Medication error/ or (medication error or medication errors or prescribing error or ● The search used a combination of subject prescribing errors or prescription error or headings and keywords for the following concepts: prescription errors or drug use error or drug use “ambulatory care,” “computerized provider order errors). entry,” and “medication errors.” The full search strategy is as follows: Results ((("Ambulatory Care"[Mesh] OR "ambulatory care"[tab] OR outpatient[tab] OR "urgent care"[tab] A total of 25,417 events were found in CHPSO’s database, explicitly OR clinic[tab])) AND ("Medical Order Entry marked “Outpatient care area,” and then were selected for review. Systems"[Mesh] OR "Computerized physician order These events are summarized in Table 1. entry"[tab] OR CPOE[tab] OR "Medical Order Entry System"[tab] OR "Computerized Provider Order From the initial 25,417 events a further review was done of the Entry"[tab])) AND ("Medication Errors"[Mesh] OR first two categories, representing a total of 19,242 events. Of these "medication error"[tab] OR "medication errors"[tab] 19,242 subset events we identified a group of 2,236 events that OR "prescribing error"[tab] OR "prescribing were related to EHRs. errors"[tab] OR "prescription error"[tab] OR "prescription errors"[tab] OR "drug use error"[tab] Characteristics of the 2,236 Events OR "drug use errors"[tab]) These 2,236 events were reported from a broad sampling of ● Then limited articles to those published since healthcare delivery organizations. When the outpatient facility January 1, 2013. was affiliated with a hospital, the bed size of the organization was split evenly between being fewer than 200 (51%) and more Method for searching Embase through the Ovid platform: than 200 beds (49%). Half of the reports came from ambulatory practices affiliated with academic medical centers. The majority The MeSH headings were translated to Emtree headings (Embase’s (67%) of these practices that reported events are in a large central controlled vocabulary) and used the same keywords. The Embase metro area and about half of the reporting organizations are gov- search is as follows, again limiting to articles published since ernment owned (51%). The next largest owned group identified January 1, 2013: their practice as being private, nonprofit (39%). Most of these events were characterized as an incident (58%). An ● Ambulatory care/ or (ambulatory care or outpatient incident is a patient safety event defined as one that reached the or urgent care or clinic). patient, whether harm occurred or not. The near miss events com- ● Physician order entry system/ or (Computerized prised 23% of the events, and 7% of the events were considered physician order entry or CPOE or Medical Order an unsafe condition. Most of harm reported from these events Entry System or Computerized Provider Order Entry). was low. Fifty-two percent reported no harm and 14% mild harm.

Table 1. Event Types with Outpatient Care Area, Location Initially selected for Review (n=25,417)

Type of Event Count

Other/Uncategorized 13,319 Medication or Other Substance 5,923 Fall 3,635 Surgery or Anesthesia (Includes Invasive Procedure) 816 Blood or Blood Product 632 Device or Medical/Surgical Supply, Including Health Information Technology (HIT) 508 Pressure Ulcer 338 Healthcare-Associated Infection 211 Perinatal 34 Venous Thromboembolism 1

Patient Safety I Vol. 3 No. 1 I March 2021 I 27 One-third of the reports did not categorize the severity of harm. Pharmacopeia MEDMARX reporting ystem was made and a taxonomy was developed for CPOE-related errors. Each error was Of the roughly half (47%) of events that recorded the age of the evaluated for what went wrong and why. The process and visual patient, the patient population that was most greatly affected was compilation of these three resources can be noted in Table 3. adults aged 18–64 (28%), followed by mature adults aged 65–74 (9%) and older adults aged 75–84 years (5%). All the various categorization schemes we considered are summarized in Tables A4–A9 within Online Supplement Appendix There are numerous structured fields that were not consistently A. The final categorization scheme we developed is listed inTable completed, which makes the results inconclusive. For example, 3, which includes appropriate references that were used in the the medication administration phase in which the event occurred creation of this final categorization. was not reported on 97% of the events. Of the remaining 3%, the greatest was administering (n=29), followed by prescribing (n=15) and then transcribing (n=10). Discussion

We then sampled 100 randomly selected events from the Medication safety errors continue to be the most common safety 2,236 events outlined above to help inform the categorization problem for patients in both the hospital and ambulatory patient 1 scheme. care settings. In high-risk patient populations, such as those with chronic diseases, who are often receiving multiple medications, These 100 events were reviewed to identify common categories. medication-related safety problems appear to be a primary risk Table 2 has all of the identified categories and Figure 2 highlights factor not only in posthospital discharge safety errors, but also the most common categories where errors occurred. in hospital admissions.3 To address these ongoing medication safety errors, policy makers have incentivized the adoption of There were consistent errors seen in the order sets, standing EHRs across the continuum of healthcare. These systems can help orders, and even future orders: a change in the dose or regimen us understand not only the epidemiology of medication safety from the routine protocol was often missed by the staff, incorrect errors, but also their prevention through the broad adoption of orders consisted of examples such as a dose error or ordering the electronic prescribing of medications. Currently, most medication wrong medication, and the manual data entry errors consisted of prescriptions are written using these EHR systems. While the height or weight being entered incorrectly or entering the wrong epidemiology of inpatient medication safety errors is reason- diagnostic codes. ably well defined, ambulatory medication safety errors are not. We then integrated all the above with three other resources: the Inpatient EHRs have been tailored to prevent common inpatient U.S. Food and Drug Administration’s Computerized Prescriber Order medication safety errors—ambulatory EHR systems have yet to Entry Medication Safety published by Brigham and Women’s Hospital be. This project has contributed to a better understanding of the (see Online Supplement Appendix A), and additional resources ambulatory medication safety errors that could be addressed by by Schiff et al. that provided two sets of codes.18 In Schiff’s work, ambulatory EHRs. a comprehensive review of medication errors reported to U.S.

Table 2. All Types of Errors Encountered in CHPSO’s Database (n=100)

▪ Administration error ▪ Orders canceled by the system ▪ Auto select/default ▪ Orders crossing/not crossing encounters ▪ Copy/paste of orders ▪ Orders in order sets/plans ▪ Expired orders ▪ Orders placed on wrong patient ▪ Future orders ▪ Patient portal as notification method ▪ Health information technology (HIT) and medication reconciliation ▪ Releasing orders ▪ Improper timing of medication ▪ Results accessioned or resulted to wrong patient ▪ Incorrect orders placed on correct patient ▪ Results accessioned or resulted to wrong provider ▪ Issues with hybrid system (some paper, some CPOE) ▪ Rx for weight-based medication (weight correct) ▪ Laterality discrepancies in orders ▪ Standing orders ▪ Manual data entry errors ▪ Verbal orders not changed in the system ▪ Missing order ▪ Wrong encounter, correct patient

28 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Figure 2. Most Common Categories Where Errors Occurred

Categories

Orders Crossing/ Not Crossing Encounters 12

Orders in Order Sets and Plans 12

Incorrect Orders Placed 10 on Correct Patient

Standing Orders 8

Orders Missing 8

Future Orders 6

Manual Data Entry Errors 6

0 2 4 6 8 10 12 14

Number of Errors

In this project we identify common ambulatory medication safety errors, using CHPSO. The most common errors noted in CHPSO’s database were errors within orders in the order sets, orders not crossing encounters, and an incorrect order placed on a correct patient. Separately, there is a need for alerting a physician of changes. For example, with the order sets, healthcare personnel often missed a change in dose or regimen from these routine orders. The reasons are unclear but could be due to provider expectations, and so the need for an alert to any change is essential. Additionally, if orders are not crossing over, it is difficult to recognize errors of omission.

Our new categorization scheme of ambulatory medication safety errors, which we have outlined in this paper, has allowed us to inform more effective electronic prescribing in ambulatory EHRs and to prevent ambulatory medication safety errors. Medication safety errors are guided by the actual errors that cause medication harm in the outpatient setting of care. Our next step is to build a tool that helps guide clinical decision support in operational ambulatory EHR systems to enable significant improvement in ambulatory safety. Our new categorization scheme is already guiding the development of that approach.

This scheme was built using data from our investigation and preexisting clas- sification schemes that were well developed and published; we believe this approach strengthens its generalizability. It can be used by ambulatory patient safety researchers and vendors that track and report ambulatory medication safety errors, as well as to improve medication safety in EHR systems.

Notes

This project was funded by the Gordon and Betty Moore Foundation, and we acknowledge the help of Rory Jaffe, MD. Table 3. Final Categorization Scheme

Category Subcategory List Information flow or communication between providers and pharmacists19-22 · Medication Reconciliation issues · Processing Orders ▪ Temporarily “on hold” ▪ Future orders ▪ Standing orders ▪ Releasing orders ▪ Orders in order sets/protocol issues ▪ Order not processed/delayed for various reasons (i.e., clarification) ▪ No confirmation of successful order transmission ▪ Wrong drug processed (ordered correctly) ▪ Telephone/verbal order issues (i.e., verbal orders not changed/entered system) ▪ Pharmacy order entry problems/issues ▪ Medication discontinuation issues ▪ Initial vs continuing order issue ▪ Duplicate order: same exact drug ▪ Failure to transmit medication discontinuation orders ▪ Expired orders ▪ Orders accidentally canceled by the system ▪ Routing issue/orders not crossing encounters ▪ Issues with hybrid system ▪ Orders not accessible to subsequent providers ▪ Results Accessioned to wrong provider ▪ Medication administration record (eMAR/MAR) issues ▪ Patient identification issues ▪ Ordered/entered for wrong patient ▪ Results accessioned or result to wrong patient ▪ Wrong encounter, correct patient Drug Name Display Issues22-24 · Character limitations · Truncation of medication names · Truncation of medication attributes (e.g., dosage form) · Items in dropdown lists not initially visible · Variation in display · Brand vs. generic names are displayed · Variable use of commercial drug data compendia vs. local customization · How medication lists are organized · Modifying medication field names (with descriptors, indications, or modifiers)

30 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Category Subcategory List Composing or Entering Drug Regimen/Sig19,22,25-33 · “Auto-complete” for drug names and sigs · Inability to order desired dose, form, strength, or quantity · Drug dictionary/out-of-date drug information in CPOE · Incorrect orders placed on correct patient · Ordered wrong formulation/dosage form/quantity selected · Ordered wrong drug · Wrong schedule/time entered · Omitted/Missed drug · Extra dose · Auto select/default · Manual data entry errors (such as use of abbreviations, transcription error, inappropriate use of units or inaccurate weight used) · Copy/paste of orders (i.e., inaccurate weight used for weight-based medication) · Comment field and special instructions issue Composing or Entering Labs or Studies20 · Laterality discrepancies in orders · Comment field and special instructions issue Clinical Decision Support34-36 · Inconsistency in application and alert firing · Interoperability: multiple systems within organizations, sites, and systems (e.g., inpatient to outpatient) · Incorrect settings (e.g., wrong dose limits) End User Knowledge of Computer Issues37-42 · Lack of computer training/system knowledge/inexperienced end user · Lack of clinical knowledge · Alert ignored/overridden Patient Communication Issues26 · Missing or incorrect patient instructions · Incorrect information on patient portal

References 5. Gurwitz JH, Field TS, Harrold LR et al. Incidence and Preventability of Adverse Drug Events Among Older Persons in 1. Institute of Medicine. Health IT and Patient Safety: Building the Ambulatory Setting. JAMA. 2003 Mar 5;289(9):1107-16. Safer Systems for Better Care. Washington, DC: National Academies Press, 2012. 6. Gandhi TK, Weingart SN, Borus J et al. Adverse Drug Events in Ambulatory Care. N Engl J Med. 2003; 348:1556-64. 2. James JT. A New, Evidence-Based Estimate of Patient Harms Associated With Hospital Care. J Patient Saf. 2013;9:122–128. 7. Office of the National Coordinator for Health IT (ONC). Improving Medication Safety in Ambulatory Care: The Role 3. Forster AJ, Murff HJ, Peterson JF, Gandhi TK, Bates DW. The E-Prescribing. 2015 Contract Number HHSP23320095651WC Incidence and Severity of Adverse Events Affecting Patients After Discharge From the Hospital.Ann Intern Med. 2003; 8. Blumenthal D. Stimulating the Adoption of Health 138(3):161-7. Information Technology. N Engl J Med. 2009;360(15):1477-9. 4. Lorincz CY, Drazen E, Sokol PE, et al. Research in Ambulatory 9. Office of the National Coordinator for Health IT (ONC). Patient Safety 2000–2010: A 10-Year Review. Chicago, IL: Health IT Dashboard. https://dashboard.healthit.gov, Accessed American Medical Association; 2011. August 12, 2017. Patient Safety I Vol. 3 No. 1 I March 2021 I 31 10. Powers C, Gabriel MH, Encinosa W, Mostashari F, Bynum J. Meaningful Use Stage 2 E-Prescribing Threshold and Adverse Drug Events in the Medicare Part D Population with Diabetes. J Am Med Inform Assoc. 2015. doi: 10.1093/jamia/ocv036. 11. Schiff GD, Amato MG, Eguale T, et al.Computerised Physician Order Entry- Related Medication Errors: Analysis of Reported Errors and Vulnerability Testing of Current Systems. BMJ Qual Saf. 2015;24(4):264-271. doi: 10.1136/bmjqs-2014-003555. 12. Roe S, Long R, King K. Pharmacies Miss Half of Dangerous Drug Combinations. Dec 15, 2017 Chicago Tribune. https://www.chicagotribune.com/investigations/ ct-drug-interactions-pharmacy-met-20161214-story.html 13. Kilbridge PM, Welebob EM, Classen DC. Development of the Leapfrog Methodology for Evaluating Hospital Implemented Inpatient Computerized Physician Order Entry Systems. Qual Saf Health Care. 2006;15(2):81-4. 14. Metzger J, Welebob E, Bates DW, Lipsitz S, Classen DC. Mixed Results in the Safety Performance of Computerized Physician Order Entry. Health Aff (Millwood). 2010;29(4):655-63. 15. Jha AK, Orav EJ, Ridgway AB, Zheng J, Epstein AM. Does the Leapfrog Program Help Identify High-Quality Hospitals? Jt Comm J Qual Patient Saf. 2008;34(6):318-25. 16. Chaparro JD, Classen DC, Danforth M, Stockwell DC, Longhurst CA. National Trends in Safety Performance of Electronic Health Record Systems in Children’s Hospitals. J Am Med Inform Assoc. 2016 Sep 16. 17. Brigham and Women’s Hospital, Harvard Medical School, Partners HealthCare. (2015). Computerized Prescriber Order Entry Medication Safety (CPOEMS): Uncovering and Learning From Errors. Washington, DC: US FDA. 18. Schiff GD, Amato MG, Eguale T, Boehne JJ, Wright A, Koppel R, Seger AC. (2015). Computerised Physician Order Entry-Related Medication Errors: Analysis of Reported Errors and Vulnerability Testing of Current Systems. BMJ Qual Saf, 24(4), 264–271. 19. Aita M, Belvedere O, De Carlo E, Deroma L, De Pauli F, Gurrieri L, Gianpiero F. (2013). Chemotherapy Presribing Errors: An Observational Study on the Role of Information Technology and Computerized Order Entry Systems. BMC Health Services Research, 13, 522. 20. Nanji KC, Rothschild JM, Boehne JJ, Keohane CA, Ash JS, & Poon EG (2014). Unrealized Potential and Residual Consequences of Electronic Prescribing on Pharmacy Workflow in the Outpatient Pharmacy. J Am Med Inform Assoc, 21(3), 481–486. 21. Priya K, Joy N, Thottumkal AV, Warrier AR, Krishna SG, & Joseph N. (2017). Impact of Electronic Prescription Audit Process to Reduce Outpatient Medication Errors. Indian J Pharm Sci, 79(6), 1017-1022. 22. Magrabi F, Liaw ST, Arachi D, Runciman W, Coiera E, & Kidd MR. (2016). Identifying Patient Safety Problems Associated With Information Technology in General Practice: An Analysis of Incident Reports. BMJ Qual Saf, 25(11), 870-880. 23. Quist AJ, Hickman TT, Amato MG, Volk LA, Salazar A, Robertson A, Schiff GD. (2017). Analysis of Variations in the Display of Drug Names in Computerized Prescriber-Order-Entry Systems. Am J Health-Syst Pharm, 74(7), 499-509. 24. Quist AJ, Robertson A, Thach TT, Volk LA, Wright A, Phansalkar S, Schiff GD. (2014). Examining the Potential for CPOE System Design and Functionality to Contribute to Medication Errors. J Gen Intern Med, S92. 25. Nanji KC, Rothschild JM, Salzberg C, Keohane CA, Zigmont K, Devita J, Poon EG. (2011). Errors Associated With Outpatient Computerized Prescribing Systems. J Am Med Inform Assoc, 18(6), 767-773. 26. Dhavle AA, Rupp MT, Sow M, & Lengkong V. (2015). A Continuous Quality Improvement Initiative for Electronic Prescribing in Ambulatory Care. Am J Med Qual, 30(6), 598-600. 27. Eguale T, Amato M, Slight SP, Seger AC, Whitney DL, Bates DW, & Schiff GD. (2014). Where Do Current Computerized Physician Order Entries (CPOE) Stand in Averting/Facilitating Medication Errors in the United States and Canada? J Gen Intern Med, S256. 28. Hou J-Y, Cheng K-J, Bai K-J, Chen H-Y, Wu W-H, Lin Y-M, Safety Alerts on Prescribing of Potentially Inappropriate & Wu M-TM. (2013). The Effect of a Computerized Pediatric Medications for Older Veterans in an Ambulatory Care Setting. Dosing Decision Support System on Pediatric Dosing Errors. J J Pharm Pract, 30(1), 82-88. Food Drug Anal, 21(3), 286-291. 29. Chaparro JD, Classen DC, Danforth M, Stockwell DC, & About the Authors Longhurst CA. (2017). National Trends in Safety Performance of Electronic Health Record Systems in Children’s Hospitals. J Am Karen Paul Zimmer ([email protected]), a pediatrician Med Inform Assoc, 24(2), 268-274 board certified in informatics, consults in the areas of health infor- 30. Kukreti V, Cosby R, Cheung A, & Lankshear S. (2014, mation technology (HIT), patient safety, and quality for nonprofit August). Computerized Prescriber Order Entry in the organizations, HIT companies, EHR vendors, and government Outpatient Oncology Setting: From Evidence to Meaningful agencies. She has over 25 years of combined experience with a Use. Curr Oncol, 21(4), e604-e612. focus on process improvement and strategic planning. Dr. Zimmer 31. Linsky A, & Simon SR. (2013). Medication Discrepancies was formerly the medical director for the ECRI Institute Patient in Integrated Electronic Health Records. BMJ Qual Saf, 22(2), Safety Organization (PSO), as well as the medical director for the 103-109. Patient Safety, Risk, and Quality Group. Her past works extend from 32. Sethurama U, Kannikeswaran N, Murray KP, Zidan MA, implementation of IT and Quality Improvement (QI) programs at & Chamberlain JM. (2015). Prescription Errors Before and Johns Hopkins Hospital to leading government-sponsored HIT After Introduction of Electronic Medication Alert System in collaboratives. She is also an expert in evaluation processes, as she a Pediatric Emergency Department. Acad Emerg Med, 22(6), designed and studied a formal educational evaluation for pediatric 714-719. residents at Johns Hopkins, where she held the rank of adjunct assistant professor in the Department of Pediatrics. Earlier in her 33. Abramson, EL, Malhotra, S, Fischer, K, Edwards, A, Pfoh, career, she was involved in policy on the Clinton Health Care Task ER., Osorio, SN, Kaushal, R. (2011). Transitioning Between Force. She completed her term as a member of the National Quality Electronic Health Records: Effects on Ambulatory Prescribing Safety. J Gen Intern Med, 26(8), 868-874. Forum (NQF) Committee: HIT Patient Safety Measures Expert Panel. She is on the medical staff at Nemours/Alfred I. duPont 34. Slight S, Eguale T, Amato M, Seger A, Whitney D, Bates D, & Hospital for Children, an associate professor in the Department Schiff G. (2014). Understanding the Vulnerabilities of Electronic of Pediatrics at Thomas Jefferson University, and an instructor at Prescribing Systems for Patient Safety. Int J Pharm Pract, 22(2), the Jefferson College of Population Health. S68. 35. Ranji SR, Rennke S, & Wachter RM. (2014). Computerised David C. Classen is a professor of Medicine at the University of Utah, Provider Order Entry Combined With Clinical Decision Support a consultant in Infectious Diseases and Clinical Epidemiology at the Systems to Improve Medication Safety: A Narrative Review. BMJ University of Utah School of Medicine in Salt Lake City, and the chief Qual Saf, 23(9), 773-780. medical information officer (CMIO) at Pascal Metrics. Dr. Classen 36. Abramson EL, & Bates DW. (2012). Ambulatory Prescribing developed the medication safety programs at Intermountain Errors Among Community-Based Providers in Two States. J Am Healthcare, served as chair of the Intermountain Healthcare Med Inform Assoc, 19(4), 644-48. Clinical Quality Committee for Drug Use and Evaluation, and was also the initial developer of patient safety research at Intermountain 37. Beeler PE, Orav EJ, Seger DL, Dykes PC, & Bates DW. (2016). Healthcare. In addition he developed, implemented, and eval- Provider Variation in Responses to Warnings: Do the Same uated a computerized physician order entry (CPOE) program at Providers Run Stop Signs Repeatedly? J Am Med Inform Assoc, LDS Hospital that significantly improved the safety of medication 23(e1), e93–e98. use. He was a member of the Institute of Medicine Committee 38. Cho I, Slight SP, Nanji KC, Seger DL, Maniam N, Dykes PC, & (IOM) that developed the National Quality Report on Health Care Bates DW. (2014). Understanding Physicians’ Behavior Toward Delivery and was also a member of the IOM Committee on Patient Alerts About Nephrotoxic Medications in Outpatients: A Cross- Safety Data Standards. He was recently a member of the IOM Sectional Analysis. BMC Nephrol, 15(1), 200. Committee on Health Information Technology and Patient Safety. 39. Cho, I, Slight SP, Nanji KC, Seger DL, Maniam N, Fiskio JM, Dr. Classen is an advisor to the Leapfrog Group and has developed Bates DW. (2015). The Effect of Provider Characteristics on the and implemented the CPOE/EHR flight simulator for the Agency Responses to Medication-Related Decision Support Alerts. Int J for Healthcare Research and Quality (AHRQ) and the NQF. Med Inform, 84(9), 630-639. Jessica M. Cole is a regulatory research coordinator (project man- 40. Czock D, Konias M, Seidling HM, Kaitschmidt J, Schwenger V, Zeier M, & Haefeli WE. (2015). Tailoring of Alerts ager) of the University of Utah Department of Internal Medicine Substantially Reduces the Alert Burden in Computerized and Division of Epidemiology. She began working at the University Clinical Decision Support for Drugs That Should Be Avoided of Utah in 2009, supporting various roles within the School of in Patients With Renal Disease. J Am Med Inform Assoc, 22(4), Medicine. In 2017, she started her career path in research as a 881–887. study coordinator and in two short years advanced from managing single-site projects to multisite projects. She currently sup- 41. Duke JD, Li X, & Dexter P. (2013). Adherence to Drug-Drug ports projects within the University of Utah and the Veterans Interaction Alerts in High-Risk Patients: A Trial of Context- Administration (VA) Salt Lake City Health Care System. Enhanced Alerting. J Am Med Inform Assoc, 20(3), 494–498. 42. Vanderman AJ, Moss JM, Bryan III WE, Sloane R, Jackson This article is published under the Creative Commons Attribution- GL, & Hastings SN. (2017). Evaluating the Impact of Medication NonCommercial license.

Patient Safety I Vol. 3 No. 1 I March 2021 I 33 w Our Best Shot: Bridging the Gap in the Vaccine Divide Infection preventionist, JoAnn Adkins; psychologist, Dr. Lily Brown; and mother of a son with autism, Susan Senator, share their insights into how vaccines work, how to recognize when anxiety may be clouding our judgment, and how both sides of the vaccine debate can finally have a real—and productive—conversation.

By Caitlyn Allen, MPH◊ DOI:10.33940/interview/2021.3.3

34 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 ◊Patient Safety Authority Disclosure: The author declares that they have no relevant or material financial interests. w

JoAnn Adkins, BSN, RN, is a senior infection preventionist with the Researchers look at which viruses circulating around the globe Patient Safety Authority (PSA). She is a fellow of the Association are most common at that time. And that’s what they base the for Professionals in Infection Control and Epidemiology (APIC), components of the flu vaccine on for that season—what they are and in 2019, she was a recipient of the APIC Heroes of Infection seeing elsewhere in the world. Prevention award. Prior to joining the PSA, she served as manager of Infection Prevention and Control for several years at a 315-bed JoAnn, you’re a mother and have been on the other side of this, acute care hospital. watching your young children be vaccinated. That must’ve been really difficult seeing how upset they were. Caitlyn Allen: How do vaccines work? It was very difficult, but I knew in my mind it was the right thing to JoAnn Adkins: Well, you need to understand how your immune do because the diseases are awful. However, when you’re handing system works. When a germ, be it a virus or bacteria, invades your over your 6-month-old, and they’re looking at you with those big, body, they attack cells and start to multiply, and that’s called the brown, trusting eyes and just crying because it hurts, it breaks infection. That infection is what causes the illness or the disease. your heart. One of the things I never did as the mother was be the The first time your body encounters that germ, it takes several days one restraining them. I talked to my child. I rubbed their legs. I for your body to make antibodies to fight that infection. But once held their hand, but I made the office bring in someone else to that infection is over, your immune system keeps what we call restrain them. So mom was not the bad guy holding them down memory cells and remembers what it learned to fight that infection to have that shot given. and protect against that disease. That’s how vaccines work. When we introduce a portion of that germ or a dead virus, such as That’s good advice to pass along to other parents who are strug- the flu shot, your body immediately recognizes it as an “invader” gling with this decision, to make it more tolerable. What advice and starts to produce antibodies. Because we’re administering would you give a parent who’s on the fence about whether or either weakened virus or dead virus, it’s not going to cause the not to vaccinate their children? actual disease, but it will cause your immune system to work so Please, please, please vaccinate your children. Vaccines are one that when you actually encounter that disease, your body recog- of the most important breakthroughs of modern medicine. It is nizes that foreign invader or that germ, and immediately starts much easier to prevent a disease than to treat that disease after it to produce the antibodies to fight it. occurs. The diseases we have made vaccines for have really serious complications. A lot of them can lead to death. For instance, When you say the weakened or dead form of the disease is measles can lead to encephalitis, which is an inflammation to your injected in your body and your body starts to remember it, brain, can lead to blindness, can lead to pneumonia. Diphtheria— does that mean you can catch that disease? which we do not see anymore, thank goodness, due to vaccina- No. When somebody gets a flu shot, many times you will hear tion—releases a toxin that forms this very thick membrane in the people say, “Oh, the flu shot gave me the flu.” Well, we’re injecting mouth, nose, and throat and causes your airway to be obstructed; dead virus. It’s not going to give you the flu. But what it will do many people died from choking. It also caused nerve damage and is kickstart your immune system, and you will have some side paralysis and kidney failure. We are lucky. We have had vaccines effects. You may have soreness at the injection site. You may have for many years, so we did not see the horror of these diseases, but some muscle aches. You may run a low-grade fever. All of that is it’s really important to make sure those diseases do not come back. your immune system working very, very rapidly to start producing antibodies to this foreign invader. It’s really amazing, and we’re so fortunate that nobody knows anybody who has diphtheria anymore, or polio, or any of these The primary benefit of those antibodies, specifically for the flu horrible diseases. But it seems like what’s become vaccines’ shot, are that if you do catch influenza, it’s going to be a lot less worst selling point is its own effectiveness—because people are severe than if you hadn’t gotten your shot? no longer around to say, “My uncle went blind from the measles. My neighbor was in a wheelchair because of polio.” Correct. Even if the flu shot is not a perfect match. We do our best to try to determine which viruses of flu will be circulating, because Right. If you talk to an elderly person—my mother, if she were there are many. The quadrivalent flu shot that most people receive alive today would be 98. She was one of my strongest vaccina- contains four different flu viruses: two types of influenza A and tion advocates because she remembered the horror of polio and two types of influenza B. Even as the flu, being a virus, continues the people being paralyzed or on the iron lungs because they to mutate throughout the season and may not be an exact match couldn’t breathe. My father had whooping cough, and he said to that shot that you received, your immune system will still rec- that he thought his lungs were being coughed up. They could ognize “influenza” if you’ve been exposed and start to produce remember people dying of measles or going blind from measles antibodies. You may still get sick, but your severity will be much, or even smallpox. They were very pro-vaccine. As we’ve lost that much less than if you weren’t vaccinated. generation, we’ve lost a lot of that, because there are not the people to tell the story anymore. What did you mean when you said “we do our best” when deter- mining which strains of the flu to include in the vaccine? Another common concern among parents who are vaccine hesitant is a preference to delay vaccine administration, rather than If you remember back in biology class, bacteria have DNA, and following the schedule that’s been outlined by the American when they replicate, it’s an exact copy. So it’s much easier to make Academy of Pediatrics; they feel like there are too many vac- a vaccine for that. Influenza, like most viruses, has RNA. Viruses cines given too quickly. Can you talk a little bit about why delay- replicate very, very rapidly, and each replication mutates slightly. ing vaccines may not be the safest course of action?

Patient Safety I Vol. 3 No. 1 I March 2021 I 35 The purpose of that immunization schedule is to protect infants COVID and influenza are both very serious respiratory viruses that and children by giving them immunity very early in their life- are circulating at the same time. There is a very effective flu vaccine time before they’re exposed to life-threatening diseases. That available for anyone. Even when the influenza vaccine is not a perfect whole schedule was based on a child’s immune system and how match, it still provides protection to decrease the severity of illness it responds to vaccines at various ages, and how likely that baby if you should get the flu. The COVID vaccine is starting to be adminis to be exposed to a particular disease at that time. The schedule istered according to risk, but it is not yet available to everyone. I ensures that infants are protected against several potentially seri- encourage you to get vaccinated when the COVID vaccine is available. ous diseases at the time they’re most likely to encounter them, or in some cases, like rubella, vaccines are vital to the health of the community. Most people who catch rubella will have mild Lily Brown, PhD, is an assistant professor of Psychology in Psychiatry symptoms, but the disease can cause serious birth defects, like and director at the Center for the Treatment and Study of Anxiety at the blindness; deafness; cataracts; and intellectual disability, if a University of Pennsylvania. In graduate school, Dr. Brown researched pregnant woman is exposed. To protect these mothers and their mechanisms of change in fear conditioning and extinction, with a children, we need to vaccinate. focus on posttraumatic stress disorder (PTSD), in the Anxiety and Depression Research Center. Dr. Brown’s current research focuses on Delaying vaccines because you don’t want your child to get them treatment development and implementation to mitigate suicide risk so frequently or at the same time could leave that child vulnerable in patients with anxiety disorders and PTSD. She specializes in cogni- to disease when they will have the most serious complications. tive-behavior therapy for anxiety disorders, including Exposure and Especially with infants, because they have a developing immune Response prevention (EX/RP) for obsessive compulsive disorder, and system. Sometimes it’s important because it takes more than one has been trained in the treatment of borderline personality disorder. dose to develop adequate immunity to protect that infant or that child—so they get boosters. We as adults get boosters also with the Caitlyn Allen: A common misperception about vaccine-hesitants tetanus vaccine. It’s only good for 10 years and then we need to is they are ignorant or uneducated, though many people may be reboostered over time. All that plays into why it’s important to not realize that most folks in this demographic are actually very follow the vaccination schedule. You don’t want to put your child at intelligent, educated individuals. And what’s fueling much of risk because they have not developed adequate immunity because this movement is actually fear: fear for their own safety and for they haven’t completed their series of vaccine. that of their children. What can we help people do to overcome this fear? Dr. Lily Brown: I think that the fear that comes from potential medical complications for ourselves or our children, as it relates to any kind of medical procedure—whether that’s a vaccine or a surgery—can often be overwhelming for people. Anytime we’re feeling anxious, it can lead to problematic behaviors. In my field, we talk a lot about anxiety-driven behavior. Often when we’re anxious, When you’re handing over your the common behavioral response is to avoid the thing that makes 6-month-old, and they’re looking at us feel anxious. That can include avoiding getting information, avoiding talking about a subject altogether, or avoiding getting a you with those“ big brown trusting vaccine. When folks are struggling with questions about, “Should I eyes and just crying because it hurts, get a vaccine for myself or my family?” it’s helpful to first consider it breaks your heart. But I knew in the choices that a person is making that are really fear-driven. What we know about anxiety is the more that you avoid things that my mind it was the right thing to do make you anxious, the worse your anxiety tends to get. And the because the diseases are awful. problem is that it can often be harder in the long run to get new learning about whatever it is that you’re nervous about if you keep avoiding it. So I really encourage people to first observe the impact that anxiety is having on their choices, whether that’s choices about getting evidence-based information from a trusted medical What advice would you give to clinicians when discussing vac- professional, or some other choice as it relates to vaccines. And cines with patients who are vaccine hesitant? I’d encourage you to practice doing the opposite of avoidance, Be transparent. Provide the education; lack of education leads to approaching these conversations and approaching them with fear. If we can provide the education and discuss the benefits of an open mind, knowing that in most cases you would still have the vaccine and the risks if that person should develop the disease, every right to choose what feels safe to you and to your family. I think that would be most important, especially when it comes to But going into these conversations open to hearing information parents with children. We need to build that trust, and that’s done is the important first step in this battle. through building relationships and through education. Those parents need to realize that it goes beyond the few minutes of pain Are there certain signs that somebody can use to try and deci- that child experiences with a vaccine to what the risks are if they pher whether their decisions are coming from a place of anxiety should develop pertussis, or diphtheria, or one of those diseases. or if they’re coming from a calm, rational thought? The best advice I have for people who are just trying to build Why is it more important now than ever in the middle of the more insight about their emotional experience is to first start with COVID-19 pandemic to make sure that you’re fully vaccinated? thinking about a recent example of a situation in which you felt

36 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 The moment you become a parent, the world seems like a really scary, dangerous place, because suddenly“ I’m faced with the pressure of caring for this vulnerable little being that is totally helpless. That can instigate a lot of anxiety in people who never struggled with that before.

anxious and writing down what were three core components of dangerous place, because suddenly I’m faced with the pressure of your experience in that situation. First, what was going through caring for this very vulnerable little being that is totally helpless. your mind? (We sometimes refer to this as “cognitions.”) What That can instigate a lot of anxiety in people who never struggled were the behavior urges that you had in that moment? And what with that before. For a lot of people, as a parent, folks want to was going on in your body? In fact, you can do this exercise of do what’s in the best interest of their child, but sometimes what mapping out these three pieces of an emotion related to any appears to be in the best interest of their child looks different in high-intensity feeling. the short run compared to in the long run. Some part of the fear that we experience sometimes translates For example, looking at a 2- or a 4-month-old baby getting their into other emotions, like anger, irritation, shame, or guilt. If you’re vaccines, it’s extraordinarily painful to watch. It can feel in that interested in building more awareness about anxiety, a great first moment like, “I want to do whatever I can to be the best par- step is to write it down. There’s research to suggest that writing ent. How could I possibly be a good parent if I’m putting my kid down your emotional experience can paradoxically lessen the through this pain or this anguish?”—when it might be the best intensity of the emotion—writing about unpleasant emotions makes thing that you can do for your child to inoculate them against you feel them less. So it’s a helpful strategy, both to build insight, vicious viruses that could have a tremendous impact on their but also to get some greater control over the emotion. health or the health of those around them. So often we make decisions for anxiety management that are focused on short-term What would make an otherwise rational person distrust science? problem-solving that ignore long-term solutions. That risk is even higher when we’re talking about making decisions to manage Right now, with some of the things that are being discussed, pol- anxiety about our caregiving for our children. itics or otherwise, a lot of people are concerned about who to trust and where to go for reputable information. We’re in strange When you mentioned about anxiety around caregiving for your times in terms of our groupthinking about science: what science children, that’s something that’s very personal for you because is trustworthy and what science is not trustworthy. We’re also at you have a new baby that you just had to do this with. a period in our history where access to information that appears evidence-based can be confusing for consumers to unpack what is Yes, that’s true. Since I was an undergraduate, I’ve been learn- legitimate science that was generated based on a hypothesis that’s ing a great deal about some of the pseudoscientific studies that falsifiable, versus pseudoscience. Oftentimes there are outlets of resulted in a lot of controversy around the perceived safety of journals that appear to be credible, but that in reality publish a vaccines. So this is something that I’ve been steeped in, and yet lot of pseudoscientific literature that aren’t, for instance, subject as a parent, seeing my daughter go through her vaccines... Her to peer review where other scientists can poke holes in the logic 6-month vaccines were the worst. She was totally unraveled by of a report to prove whether the theory holds. them, it’s awful to watch. And yet, I’m going to do it every time my This is one of the reasons why, on average, I tend to refer peo- primary care doctor tells me to do it, because again: In the short ple to world consensus guidelines, things like the World Health run, I would feel better to not make her go through that, but in Organization, the Centers for Disease Control and Prevention, the the long run, the risk to her and to her generation far outweighs National Institutes of Health. There are lists of scientific journals any sort of short-term anxiety that comes along with putting her that are reputable and are widely cited by credible scientists. If through that pain again. people are concerned about how credible a certain claim is in this environment, you’re not alone in feeling that way, but this is one You mentioned the differences between short-term and long- of the reasons why it’s helpful to turn to trusted experts. And for term payoff. Vaccines have become their own worst enemy most of us, that expert could be our primary care doctor. because they’ve been so effective, and now we don’t know people who have had these diseases anymore, where they used to be so Does anxiety present differently in individuals making deci- prevalent. Is that a part of the human psyche that factors here? sions for themselves versus for their children? Is a person likely Definitely. We are very impacted by emotional stories. For to be more anxious or less anxious about making decisions for instance, we think about the relative safety of flying in a plane their children? versus driving in a car. And we know statistically that driving in It really depends a great deal on the person. But for a lot of par- a car is much more dangerous on average than flying in a plane. ents, the responsibility that caregiving for children necessitates Car crashes kill way more people every year than plane crashes. creates an enormous amount of anxiety. The moment that they And yet there are more people afraid to fly in planes than are become parents, my goodness, the world seems like a really scary, afraid to drive. Why is this? Well, part of the reason is the image of a plane crash is so horrific, it’s so terrifying for people to think

Patient Safety I Vol. 3 No. 1 I March 2021 I 37 about that, that we have a quick, biased decision-making process in our brain that often makes inaccurate decisions on the basis of highly emotional content. And so, I think that it’s true, without having had the emotional Bedside Tips impact of seeing a child with polio, for instance, it’s hard for peo- How to discuss vaccines with a ple to empathize or to predict how that disease would impact them vaccine-hesitant patient or their family. Vaccines having been so effective in the long run has an enormous impact on strategies for ensuring high compliance, because ultimately human nature suggests that we want to Be empathetic. Parents are trying to make make decisions where we can get immediate gratification. And the best decisions they so if there was a disease, like the COVID virus, where I could give can for their child. While you a vaccine and promise that you were inoculated against it vaccines have been proven tomorrow, I think most of us, if we knew that it was safe, would jump out and do that. one of the effective public health measures, there But if instead I were to tell you that the vaccine that you’re going have been instances of to get is going to prevent a virus that you’ve never heard of, or vaccine-related injuries, that you can’t imagine what the impact of it would be, there’s not allergic reactions, etc.1 a lot of gratification that comes with making that choice. Often when we don’t have these emotional examples, sometimes people struggle to see the rationale around why would it be worth it to Be proactive. Patients are more likely get a vaccine. to get vaccinated when appointments have been What advice would you give a clinician to have an open conver- prescheduled for them sation with parents about vaccines? versus asking them to It’s challenging trying to convince someone to come around to schedule an appointment an evidence-based practice; there’s a whole field focused on this themselves.2 idea, implementation science. We have this practice that we know works. How do we get clinicians to use it, or in this case, how do we get patients to be willing to use it? And there’s a lot of factors Be upfront about Recognize the parental that are at play here. Certainly, appealing to people’s knowledge right to make decisions and providing them with education can be part of it. Often that’s the greater good. for their children. our first thought, is if someone inaccurately perceives a vaccine However, acknowledge to be harmful, why don’t we just educate them about the fallacy the consequences of their of that thinking? We know that that’s not the solution though. At decisions like prohibiting least it’s not the full solution because by and large, that’s what folks unimmunized children from have been trying to do in the face of the anti-vaccine movement 1 for a long time. attending public school. So that’s a piece of it, but there are other pieces that we need to draw on as well, like how many people with children your age are choosing to vaccinate their child. If it’s 80% of parents, Be transparent. Encourage parents to ask that’s a huge majority. Is it possible that 80% of people are mak- questions but ask they ing a horrible decision? It’s possible. But it’s also possible that approach each conversation they’re doing this based on the best science that’s available. So with an open mind.1 using normative comparisons can be helpful. The other piece, though, that’s helpful sometimes is having a local champion, a local cheerleader who resonates with people in that demographic, of a similar background, of a similar mindset, who similarly cares Be partners. Look for opportunities to about their children. Drawing on these local champions can help engage vaccine-hesitant to enhance uptake of any kind of evidence-based practice that we parents in broader want to implement. discussions. Involvement upstream may help Could the healthcare system stand for an image makeover? address concerns and What could be done to increase patients’ trust of it? make them feel more Definitely. I think the healthcare system requires a major make- comfortable at the time over, not just of their image, but of many things. There are huge of vaccination.1 systemic biases and discriminatory practices that have been operating for a long time that have an impact on the choices that Sources: patients make and the trust that they have in their doctors. That’s not only true just in terms of the history of healthcare decisions, 1. Reich JA. Calling the Shots: Why Parents Reject Vaccines. New York: New York University Press; 2018. 2. American Public Health Association. COVID-19 Vaccines: The Realities of the Next Steps. Webinar. December, 2020. 38 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 but also it’s true in terms of research practices, that there is a huge history of bias and discrimination. It’s important to keep in mind that when someone is skeptical or uncomfortable about My sister [a pediatrician] said that the healthcare system, it might be coming from a place of some important experiences that they’ve had in the past that have con- I could consider spacing out the tributed to their skepticism. vaccines more, but“ there’s a reason When I was working on a study looking at suicide prevention they cluster these vaccines. And among people who were HIV-positive, I learned some of the experiences that some of my folks had with things like psychiatric spacing out the doses was almost hospitalization were damaging in terms of their willingness to just a made-up solution. disclose certain pieces of information. I think the best way to make over the healthcare system is to start with making over the way in which we do research, by being very inclusive of people from all backgrounds, by making sure that What changed that made you rethink your position on vaccines there’s representation from members of the community through- causing autism? out the research process. That way, the entire community can Well, a lot of professionals began disproving the Andrew Wakefield rally around the data we get as something they’re confident in. It’s stuff, so there was some pushback in the medical community. My really easy to wag our fingers and to say, “Oh, how silly is it that own neurologist couldn’t really say definitively that there wasn’t someone doesn’t trust the science that comes out of a particular something in the environment that was maybe causing this institution?” without having in the back of your mind these his- increase in autism, but she never said a vaccine, or thimerosal, tories of injustices. On the other hand, I think that there’s a lot of or mercury. When it came time for Ben, my youngest, to get vac- opportunity for making progress and building trust with the com- cinated, I turned to my sister who’s a pediatrician and said, “What munity that will likely have an impact on enhancing compliance do you think of all this stuff? I don’t want Ben to have autism, but with recommendations from the healthcare system. I don’t want him to be sick and get any of these diseases.” My sister said that I could consider spacing out the vaccines more, Susan Senator is an author, journalist, and advocate for families but she also explained there’s a reason that they cluster these touched by autism. She and her husband are parents to three boys, vaccines. “It produces a more robust response in the baby’s body. including her oldest, Nat, who was diagnosed with autism as a toddler. By and large, we need that kind of response.” And spacing out the She’s given speeches at places like Harvard Medical School and the New doses was almost just a made-up solution. “There’s no evidence York State Autism Society, and in 2006, she was invited to attend a state that that’s okay to do.” Nevertheless, I did it because I was afraid, dinner at the White House in honor of the Special Olympics. I was still kind of ignorant about the whole thing, so we spread out Ben’s vaccines, but ultimately, he got all of them. By the time Caitlyn Allen: Tell me about your relationship with vaccines. he was done, I was starting to feel pretty sure that vaccines were a good thing and not what caused the autism. Susan Senator: My relationship with vaccines started when my oldest son, Nat, was about 3; this was also around when he was It’s got to be difficult as a mother when you have this young diagnosed. I had been talking about autism with a friend whose son child with a condition, and you think, “Oh no. What if I caused also has it, and she was telling me that there was this theory going that?” Spacing out the vaccines felt like a happy middle ground. around. It was a new theory that there were gastroenterological issues causing autistic-like symptoms in people. Now people really Right. Also, at that point, Ben was born six years later, around believe that, but back then it was kind of new. That was around 1998. By then, I knew a lot of people in the [anti-vax] community. 1992. She said that she had tried putting her son on a gluten-free I had a close friend who was anti-vax and just full of information and casein-free diet to try to mitigate these kinds of features. and alternative medicine suggestions. At the time it was a huge decision, a big ordeal to go gluten-free. The theory was that the vaccine was blowing out the immune system and causing this condition, making it so her son’s body I tried that a couple of times because of what she said. Even though could not absorb certain things like gluten and casein. And that I already knew that vaccines didn’t cause autism, I still had these was causing the autismlike symptoms. It sounded plausible to me, leftover feelings. A lot of it was because of the people I was sur- and then I happened to come across Andrew Wakefield’s paper, rounded by. So many other moms believed the vaccines were which hadn’t been disproven at the time. Intuitively it started to harmful and that there were things you could do to ameliorate make a lot of sense to me that this really could be the reason. I autism once there was this injury. I really had a lot of mixed feel- could just imagine the chemical or the vaccine itself messing with ings, as you can see. my son’s immunology and causing this. When I thought about that, I felt terrible. “I shouldn’t have gotten Granted autism is a spectrum, and everyone’s experiences are him the shots, because that means that my giving permission for unique, what has it been like for you and your family? the shots caused this.” That was a crisis time for me. My second It’s been life-changing for me and for my family to have some- child had already received his shots, and he didn’t have autism. one like Nat in our lives. He’s an adult now and has been living We knew that almost immediately. It was when my third child in group homes since the age of 17. But now because of COVID, came along six years later that the question came up again for us. he’s back living with us. We feel it’s not as safe in the group home

Patient Safety I Vol. 3 No. 1 I March 2021 I 39 right now, so we’re back with him full-time. Now I’m so different This parent had done so much research that she figured out what from how I was way back when I was a young mom. I was 27 when her son needed, and he’s doing better. There is that kind of story he was born, and I just didn’t know what was going on. He was where you have a parent who’s able to do that research and has different from what I expected from day one in very subtle ways. the time and energy and smarts to do it. But for the most part, I I just had a bad feeling for like two-and-a-half years, and no one mean, I just would blank out when people started to tell me, “Oh, else in my circle or family saw what I was talking about until one but you have to look into that. You have to go back and have them very difficult event. test the urine [or the hair or whatever].” At a family gathering, Nat would not come into the room. He Also, these tests cost a lot of money and nobody covers them. But stayed at the door crying the whole time. Suddenly everyone in there’s always a parent who’s got more superpowers than you. the family noticed that something was amiss. Soon after that, my There’s always another parent who’s able to fly across the country sister, gave me a book. It was the American Academy of Pediatrics and hire that doctor and bring them to live with them. There’s just Birth to 5 Years Old [Caring for Your Baby and Young Child: Birth always something more you can do according to the community, to Age 5, Bantam 2009]. In it, there were all these lists of what unfortunately. I still get links sent to me about things I could try. your child should be doing developmentally. I could just list all Just today, I said to someone, “He’s having a little bit of a difficult the things that he was not doing, and I brought the list to our day; he’s anxious.” They replied, “Do you have a fidget spinner for pediatrician, and she referred us immediately to Mass General him?” I’m thinking, “That is so not him. Just because he’s autistic for an evaluation. That’s what I consider the first period in our doesn’t mean he’s going to like a fidget spinner.” There’s a lot of lives with Nat. advice, a lot of people, well-meaning people, but they don’t know The next period was the years of learning about autism and start- what you know. ing to develop strategies. Then getting to be an older family—a It’s like Alice in Wonderland with the rabbit hole, you just go more-together family—where we could do things and trying not deeper and deeper, and there really is no end to it. to make autism be the driving influence in the family. Our goal There was a period when he was 5 or 6, where we were looking for was to be five people with equal needs. So, whether I succeeded everything, looking to try as much as we could. We were desperate or not, you’d have to ask my other two sons. to get him on a list for people who could try secretin. By the time It’s been life-changing. I just find that I rejoice at the smallest a potential slot came up for Nat, they were already disproving things that I see that Nat can do and the growth he’s having now that. That made me start feeling a little bit more skeptical. What at 31—learning more about social engagement and wanting to I learned from that is, do the reading. Talk to trusted people. Talk converse. There’s all this growth now, and that gives me so much to doctors. And then you have to put all that together and figure hope. I’ve learned lots of things, like never say never, and just try it out, and it’s hard. again; if something failed in the early days, try again. On top of having a child to take care of. Right. This is nothing to fool around with, the vaccinations. I really have come to see that it’s necessary, and we’re starting to see the return of measles because people are so misinformed about herd immunity or they’re not even thinking about people who are immunologically suppressed. I feel like we are really in danger of going back to the dark ages if you have people who are You start to feel like afraid of all this stuff. everything caused autism. “ Tell me more about the vaccine-hesitant community. Many of us are friends. It’s part of the big parent network on Facebook, pretty much, and Twitter. Even before the COVID iso- lation period, a lot of us were stuck more inside our houses than What did it feel like as a new mom having a child diagnosed we’d like to be, because our kids are a lot to handle. I know a lot with autism? Society puts so much blame on parents, especially of these other moms, so I know what they’re thinking to some mothers, if something is wrong with their child or if they’re degree. What’s it like? There’s a lot of anger. There is fear, I sup- sick. “If you had just done something differently, Nat wouldn’t pose, but what I see more of is this kind of anger and mistrust have autism.” of the institutions in general, the government, and Big Pharma. At the time when I felt that kind of blame, it was because my son It’s very easy to convince yourself that these large organizations wasn’t on the diet, so he wasn’t gluten-free, or when we tried it, are not out to protect you, they’re out for money. There’s a lot of we didn’t try it for long enough. There’s always someone who will cynicism too. say, “Well, actually they need to test the hair to make sure, to see if there’s an excess of copper.” There are parents who just go so How far does the mistrust of “Big Pharma” go? Does it typically deep into chemistry or pseudochemistry that it’s hard to combat extend to all medications or is that reserved more just for the that kind of message coming at you. Sometimes they’re right. vaccines? For my third book, I interviewed a parent whose son developed I think it would cover antibiotics too. There was a moment where catatonia, which they think happens with people with autism I was afraid to give my youngest penicillin. He was 8 days old and sometimes as they get older, some kind of movement disorder. had a slight fever, so we took him to the emergency room. They

40 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 wanted to give him antibiotics and do a spinal tap. It was past pragmatic, are from the parents or from the family experience. midnight, and I called my sister and asked what I should do. She That’s a really good point. Doctors, like your sister, have spent an said, “Let them do it.” incredible amount of time in training and residency, and have a Not everybody can do that: call a doctor at that point for a second wealth of knowledge. They’re experts in medicine. On the other opinion. That’s just to show you how scared I was, even though I hand, you, as Nat’s mom, you’re his constant. You’ve been there knew that antibiotics were safe. They’ve been around for a long and seen him every single day throughout his life. time and they’re very effective, and course you need to take them Right. I also think there’s some amount of reluctance on the part if you have something like strep or whatever it was that Benji had. of a pediatrician to give a parent bad news. A parent that you’ve There probably are medications that people are worried about. been with since the kid was a baby—you have a relationship. I I think there was a study at some point that said there was some loved my pediatrician, but she missed it. She missed this, so ulti- indication that antibiotics given at a very early age had a rela- mately we did move on to another one. But on the other hand, my tionship to autism. There were other studies going around, or sister picked up on something even before she was a pediatrician. rumors, that when the mother’s pregnant, taking Prozac could She was just in training on her psych rotation when we got the have some kind of adverse effect like autism later on, and Pitocin diagnosis. I called her and I said, “Do you think they’re right?” She given during labor. There’s a lot of things that circulate to the point paused for a minute, and then she said, “Yeah, I do.” You need the where you start to feel like everything caused autism. pediatricians to be the first responders to this. They’re the first ones that you take your child to, so they need to be really up to That’s sort of what it sounds like. speed, but they also need to be really empathic and able to deliver news without also delivering hopelessness. Yeah, these days it’s even harder to wade through all the information than when I was a young mom. Really it is understandable What advice would you give a physician if they have a patient that people would have a lot of questions and doubts. who might be hesitant about vaccines? How would you encourage them to have a conversation with them? Absolutely. I would definitely have them try to rustle up all of the studies that As an English professor, I tell my students go to the .gov websites they can and show the parents direct evidence that the vaccines or the .edu. You need to reinforce with them where the expertise are good and necessary. I think there are practices now that won’t is. You can’t raise a generation of people who are just completely even let patients in who don’t want the vaccines. I think they have cynical and skeptical about all the vetted knowledge and insti- to do that. As much as I love my friends who don’t believe in it, tutions that have been around forever. You really have to push I believe in it, and I don’t want to see a world that goes back to back on that. having measles and polio, and now COVID. I want people to get the vaccine so that this particular pandemic can go away, because It’s important for people to think critically, but it’s another to they really work. The doctors need to not take the tone of, “I’m the constantly be on the lookout for conspiracy theories. expert, listen to me.” It’s better if they’re empathic but informed Right. It’s also about not losing the trust to begin with. Again, the and full of evidence. pediatricians have to be so alert to the messages they’re giving, nonverbally as well as verbally, the things that they’re conveying, It’s curious where the lines get drawn for people who distrust and they need to give the parent respect and treat the parent like the science or their doctors about vaccines, but they’ll defer to the expert that they are, or else you’re really going to foster more their doctor if their child has chicken pox or a stomachache. of this distrust. Yeah. It’s just very hard to know. People will say, “Well, I don’t want to put those chemicals or those terrible germs, the bacteria—I How far does the mistrust go? What about other areas of don’t want to actually put that into my baby.” But on the other medicine? hand, the vitamins that you buy, they’re full of chemicals too. Nutrition. There’s a feeling out there that there’s a lot more that Everything’s a chemical and there’s lots of good bacteria. There’s can be done with diet, and with vitamins, and other kinds of sub- just so much, but to me it’s the conspiracy thinking today. It’s very stances: curcumin and blue algae. There’s been all different things hard to overcome that. over the years. There is this general skepticism that most of us Probably the only thing that would help, aside from the doctor have, and I still do, about our doctors, which is that they don’t showing the evidence and having the best studies on hand, some- really know what we know. The parents are the experts. We do thing irrefutable, is bringing those parents to the table. If there’s feel like we’re the experts. That starts from the beginning when going to be any other research, if somehow you could get a few your pediatrician doesn’t believe that your child has autism. We of the big vaccine-hesistant folks as a part of the research so that waited until Nat was almost 3 to get his evaluation, even though they’re heard, and that they can actually directly have input and I saw problems. That kind of thing. then see the facts, that might be a way to do it. It always helps to If the diagnosis is missed or delayed, and children miss out on bring the opposition together. That’s how both sides learn from early intervention: That’s the kind of thing that can start making each other. you feel a little isolated and skeptical about what your doctor says. I still feel like I’m the expert on Nat. When my mom says to me, The views of the interviewees expressed in this article are their own and “Is there any anyone you can ask about a given problem?” I say, “I do not necessarily reflect the official position or policies of the Patient can ask my husband.” That’s really how it still seems to me, that Safety Authority. the day-by-day experiences and the things that we’ve learned, the

This article is published under the Creative Commons Attribution- NonCommercial license. Patient Safety I Vol. 3 No. 1 I March 2021 I 41 WARNING

Patients taller than this height may experience: * Delays in care * Falls * Ill-fitting equipment * Pressure injuries Falling short: Adverse Events Related to Patient Height

WARNING By Caitlyn Allen, MPH◆◊ & Neal Wiggermann, PhD† DOI: 10.33940/bedside/2021.3.4

easuring a patient’s height is a routine part of a healthcare encounter. But once completed, how often is this information used? For most of us who fall within 95% of the mean population height, Mthis metric is rarely discussed, but what happens when it is over- looked? And what about those on the outer tails of the bell curve of population distribution?

Almost 1 million (909,222) adults in the United States are at least 6'4",1,2 more than the entire population of South Dakota (884,659).3 Conversely, an estimated 30,000 Americans have a form of dwarfism, typically defined as an adult height no taller than 4'10".4,5 See Figure 1. However, despite this prevalence, the healthcare system struggles to provide consistent, adequate care for patients with extreme heights.

“Being tall is not a disease, but it is a consideration,” shared one 6'3" woman, who descended from a long line of “Scottish Vikings.”* “I am one of the short ones at family reunions,” she continued; her father tops 6'8". Experiences like hers prompted this investigation into the medical mishaps and trauma that have befallen those who are not “average-sized.”

Corresponding author Patient Safety Authority †Hillrom Disclosure: The authors declare that they have no relevant or material financial interests. Patient Safety I Vol. 3 No. 1 I March 2021 I 43 One story detailed a man who had his hip need of a knee replacement. Though her replaced a few years ago, the operation for surgery went well, and she only needed which proved to be the least of his wor- to spend one night in the hospital, she ries. As part of her routine, his wife called was eager to leave. A nurse wheeled in a ahead to request a “tall-person” bed. The Being tall is bedside commode and told her one of the nurse assured her one would be available. “ requirements to go home was the ability When he went to preop, she called up to to self-toilet. After a few unsuccessful the unit where he would spend the night not a disease, attempts, the nurse reluctantly informed and confirmed the bed would be ready. but it is a her she would need to go to a long-term care facility to recover. The woman Following a successful surgery, he was consideration. explained that the standard commode transported to his room. He did not need was too short, even with the toilet seat to climb into the bed to know that it was riser, and requested a taller model. Several standard-sized and unable to accommo- hours later, the nurse was able to locate a date his 6'6" frame. The nurse apologized commode that adjusted to 25", well above for the oversight and began the surpris- the night contorted and without much the standard Americans with Disabilities ingly arduous process to locate a larger rest. His wife expressed frustration and Act–compliant 19". She was sent home model. Not wanting to impose, and in need disbelief at how difficult it was to find an where she enjoyed a full recovery.* of rest, the man quietly crawled into the adequate bed in Los Angeles. “The Lakers bed to get some sleep. play here. We have tall people.”* Hearing these stories prompted a look at Pennsylvania Patient Safety Reporting Twenty-one hours later and still no prog- Another patient, an otherwise-healthy 6'4" System (PA-PSRS)** data to see if there ress, the man and his new hip had spent tall woman in her 60s, found herself in were similar events where an adverse

Figure 1. Cumulative Percent Distribution of Population by Height and Sex: 2007 to 2008

Men Women 20–29 30–39 40–49 50–59 60–69 70–79 Height 20–29 30–39 40–49 50–59 60–69 70–79 – – – – – – 4'10" – 1.7* – 1* – 3.3* – – – – – – 4'11" 2.6* 3.1 1.6* 2.1 3.6* 8.7 – – – – – – 5' 5.7 6.0 5.0 8.0 9.0 16.0 – – – – 0.4* – 5'1" 12.3 11.6 10.8 16.7 14.7 26.0 – – – – – – 5'2" 20.8 19.7 19.8 23.3 23.4 36.9 – 3.1* 1.9* – 2.3* – 5'3" 30.4 31.3 30.8 36.3 38.4 51.9 Each cell represents the 3.7 4.4* 3.8 4.3* 4.4 5.8 5'4" 43.5 46.6 46.0 50.7 52.8 69.9 cumulative % 7.2 6.7 5.6 7.6 7.8 12.8 5'5" 54.1 61.2 58.0 68.4 66.6 82.8 of people at a 11.6 13.1 9.8 12.2 14.7 23.0 5'6" 72.4 74.0 72.2 79.7 83.3 89.3 given height by age. For 20.6 19.6 19.4 18.6 23.7 35.1 5'7" 82.3 84.9 83.0 88.4 93.3 95.4 instance, 51.9% 33.1 32.2 30.3 30.3 37.7 47.7 5'8" 90.3 91.8 91.2 95.2 97.0 98.4 of women aged 70-79 are 5’3” 42.2 45.4 40.4 41.2 50.2 60.3 5'9" 94.1 96.1 94.7 97.3 97.8 99.6 or shorter. 58.6 58.1 54.4 54.3 65.2 75.2 5'10" 97.6 98.9 97.8 98.9 99.6 99.6 69.9% of 70.7 69.4 69.6 70.0 75.0 85.8 5'11" 99.6 98.9 99.4 100.0 99.8 100.0 women in that age range are 79.9 78.5 79.1 81.2 84.3 91.0 6' 100.0 99.4 99.5 100.0 99.9 100.0 5’4” or shorter. 89.0 89.0 87.4 91.6 93.6 94.9 6'1" 100.0 99.9 99.5 100.0 99.9 100.0 94.1 94.0 92.5 93.7 97.8 98.6 6'2" 100.0 100.0 99.5 100.0 100.0 100.0 98.3 95.8 97.7 96.6 99.9 100.0 6'3" 100.0 100.0 99.5 100.0 100.0 100.0 100.0 97.6 99.0 99.5 100.0 100.0 6'4" 100.0 100.0 99.5 100.0 100.0 100.0 100.0 99.4 99.4 99.6 100.0 100.0 6'5" 100.0 100.0 100.0 100.0 100.0 100.0 100.0 99.5 99.9 100.0 100.0 100.0 6'6" 100.0 100.0 100.0 100.0 100.0 100.0

Note: “–” Represents zero or figure too small to meet statistical standards of reliability of a derived figure. Values with an asterisk do not meet standard for reliability or precision. Source: U.S. National Center for Health Statistics, unpublished data, .

*Anonymous patient interview **PA-PSRS is a secure, web-based system through which Pennsylvania hospitals, ambulatory surgical facilities, abortion facilities, and birthing centers submit reports of patient safety–related incidents and serious events in accordance with mandatory reporting laws outlined in the Medical Care Availability and Reduction of Error (MCARE) Act (Act 13 of 2002).16 All reports submitted through PA-PSRS are confidential and no information about individual facilities or providers is made public.

44 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 event was related to the patient’s height. The following are examples of events in Figure 2. Hospital Bed Hinge Points which the patient was too tall: 84 in (7’0” or 231.36 cm) Operating room table was modified to fit patient who was 6'10" and 410 lbs. Used two bed extensions and small pole attached for stability. Hydraulics were very slow to function throughout procedure. Table was modi- Taller than Average fied to fit height with questions of safety based on manufacturer specifications, but operating room manager told team to “continue and everything will be fine,” because operating Average room table can hold up to 1,000 lbs.

Patient was 6'5" and in need of a bed extender. Bed extender ordered, but none were avail- Shorter than Average able, so patient was put on a waiting list. Patient's wife at bedside reported that a bed extender was also ordered yesterday. Relayed Head information that one is not yet available to Iliac Crest Trochanter Knee wife. Patient's legs are bent, and bottoms of feet are pressing against footboard, possibly contributing to skin breakdown.

The following are examples of event reports in which the patient was too short: of the patient. The patient had a cushion in Patient Height and Hospital The patient, a 4'11" woman, was found on place on the recliner. Due to the patient's Equipment the floor in the middle of the night ringing her short stature and decreased strength, she was call bell. She was too short to climb in and out struggling to scoot herself back properly into Many factors are considered when design- of her bed on her own and required assistance the chair causing her to slide on the cushion ing hospital equipment, such as function- from the nursing staff several times during the rather than successfully lift her buttocks to ality, durability, and cost. Common prac- day. She needed to use the restroom, and not adjust herself back in the chair. Due to her tice for designing medical equipment wanting to impose upon them once again, sliding on the cushion, the cushion folded generally accommodates up to 95% of she attempted to exit her bed by herself. Her under her. Patient was wearing shorts. While the population, while potentially exclud- legs slipped out from under her, causing her the patient did not experience skin injury ing the most extreme variances in human to fall and hit her head. from event, the potential was there. The cush- attributes.6 In the United States, the fifth ion makes an already high surface higher for percentile female and 95th percentile Patient was attempting to get into bed, she patients who are small in stature and/or those male heights are 59 inches (4'11") and 74 slid against nursing assistant and was low- who have impaired mobility. inches (6'2"), respectively.7 ered to the floor. Patient of short stature 4'11" and was too short for the bed height. Patient Even when the necessity of height-related unable to stand independently while nursing precautions is recognized, adverse events Inpatient Beds assistant offered to get a step stool prior to can still occur. Specialized beds are available to increase getting into the bed. comfort and range of motion for taller A patient, 6'6", was admitted and given patients, and there are options to mod- Patient has been calling for assistance to get to an order for a bed extender. He was weak ify typical beds, such as bed extenders and from bedside commode. Tonight, patient from chemotherapy, and rather than forc- that can be added to the base or longer attempted to stand without assistance and ing him to stand, his nurse decided to weigh mattresses. fell. Patient denies injury and acknowledges him with a scale that connected to his hos- that she knew she was supposed to wait but pital bed. Unfortunately, when she went to However, both extenders and longer mat- did not. Patient and nurse also report that the tare the weight of the equipment, she for- tress are an added expense, so facilities bedside commode is too tall for the patient. got to zero out the additional weight of the may keep a limited stock on hand, if at Nurse suggested shorter bedside commode. extender. The patient’s weight was recorded all. The ability to monitor this inventory We found one on the unit that had adjustable as being an additional 12 kg heavier than in real time, such as through radio-fre- height capability and replaced the current he was. Thankfully, the error was caught by quency identification (RFID) technology, bedside commode. a pharmacist who was reviewing his order also requires additional resources. More for a weight-based medication. Without this financially challenged hospitals may then This therapist transferred the patient from the intervention, the patient would have received be forced to track the bed’s location manu- wheelchair to the recliner upon the request an incorrect dosage of a critical therapeutic. ally. Consider the difficulty of trying to find

Patient Safety I Vol. 3 No. 1 I March 2021 I 45 one “tall-person” bed from an inventory of 500. And then consider the implications if that bed had been previously assigned to Points to consider: another patient, because it was the only one available.

Another important consideration for comfort while lying in a hospital bed is frame cm articulation, or where the hinge points occur (Figure 2). If the hinge points of the bed are misaligned with the body, the patient may experience discomfort or slide down in bed.8 Although patients may technically fit within the bed, the Measure height and weight in Confirm the patient’s measure- movement of their body segments may not metric units for each patient at ments before prescribing any match the movements of the bedframe.8 every healthcare encounter.12,13 medication that is based on height and/or weight.12,13 Imaging Equipment Patient size is a frequent concern with imaging studies, though these conversations typically involve weight limits, girth, or body mass index (BMI). However, the patient’s height is an equally important consideration. The consequences of being unable to accommodate taller patients include prolonged exposure to radiation; Consider patient height regarding delays of care; and the removal of metal staffing:14 Implement protocols to source dental fillings when patients are placed in • When possible, assign taller specialized equipment when direct, or near direct, contact with mag- staff to assist taller patients extremely short or tall patients nets due to their size.9 Some newer models visit the facility. Ensure all hospital of imaging equipment no longer indicate • Teach staff how to safely equipment is appropriately sized height restrictions.9 handle taller or shorter patients to prevent a fall for this patient and available when needed, including but not limited to: • Use ambulation equipment Safe Patient Handling and Mobility such as fall arrest vests or • Bed (SPHM) pants • Stretcher SPHM equipment is used to reposi- tion dependent patients in bed, trans- • Wheelchair fer patients, hold limbs, and ambulate • Bedside commode or toilet patients who are regaining their ability • Mobile lift and slings to walk. The slings used to lift and mobi- lize patients come in many sizes, but • Shower (or a reasonable hospital units often only stock those that Test compatibility of specialized alternative) equipment with other devices (e.g., are most used. To accommodate patients • Blood pressure cuffs who are significantly taller or shorter each model of bed is tested to confirm 14 than average, hospitals should have the the use of bed extenders will not • Imaging equipment larger and smaller sizes available, possi- inadvertently shut off the bed alarm). bly even consider pediatric sizes for the shortest adult patients. Higher capacity lifts and slings may also be needed for heavier patients.10 When using mobile lifts to ambulate patients, the arm is adjustable based on patient height; however, like hospital beds, the lifts may also have a maximum height capacity causing insufficient head clearance for the tall- Confirm equipment or supplies Complete thorough skin checks, est patients.11 Facilities may respond by required postdischarge are including “height-related hot spots,” padding the sling bar to protect the head available, e.g., custom compression e.g., the bottom of the patient’s foot of very tall patients and should consider socks when standard sizes will not fit. rubbing against the footboard.15 installing ceiling lifts for ambulation which provide more vertical clearance for the tallest patients. For Consideration Totals and Components of Change: Practices website. https://www.ismp. 2010-2019. U.S. Census Bureau website. org/guidelines/standard-order-sets. It is imperative for healthcare facilities to https://www.census.gov/data/tables/ Published January 12, 2010. Accessed accurately capture patient height during time-series/demo/popest/2010s-state-to- January 8, 2021. every healthcare encounter. Patient height tal.html#par_textimage_1574439295. should be a factor when developing a plan 13. NIHR Southampton Biomedical Updated December 30, 2019. Accessed Research Centre. Procedure for of care, and hospitals may consider creat- September 29, 2020. ing a checklist for treating those who are Measuring Adult Height. National taller or shorter than average—just as they 4. Mayo Clinic Staff. Dwarfism. Mayo Institute for Health Research. https:// would for a patient who is a fall risk or has Clinic website. https://www.mayoclinic. www.uhs.nhs.uk/Media/Southampton- diabetes. org/diseases-conditions/dwarfism/ Clinical-Research/Procedures/ symptoms-causes/syc-20371969#:~:- BRCProcedures/Procedure-for-adult- This checklist should include a process to text=Dwarfism%20is%20short%20stat- height.pdf. Updated June 2014. Accessed ensure patients are measured accurately ure%20that,4%20feet%20(122%20cm). January 29, 2021. upon admission to determine whether spe- Published August 17, 2018. Accessed 14. Matz M, Celona J, Martin M, cialized equipment, e.g., a bed extender, is January 29, 2021. McCoskey K, Nelson G. Patient Handling required. Facilities should also consider a 5. Understanding Dwarfism. Basic and Mobility Assessments Second Edition. procedure to obtain height-related equip- Facts. Understanding Dwarfism website. The Facility Guidelines Institute. https:// ment quickly, e.g., locating and installing http://www.udprogram.com/basic-facts/. www.fgiguidelines.org/wp-content/ bed extenders when an extremely tall 2013. Accessed July 14, 2020. uploads/2019/10/FGI-Patient-Handling- patient arrives without notice. Extreme 6. Center for Devices and Radiological and-Mobility-Assessments_191008.pdf. patient height should also be considered 2019. Accessed January 8, 2021. by relevant clinical groups such as SPHM Health. Hospital Bed System or patient safety teams, and in safety meet- Dimensional and Assessment Guidance 15. Fletcher J. Articulated Bed Frames ings such as fall prevention huddles. to Reduce Entrapment. U.S. Department and Heel Ulcer Prevalence. Wound of Health and Human Services, Food and Essentials. 2015;10(1):8-13. https://www. A height-related protocol should include Drug Administration website. https:// woundsinternational.com/uploads/ safeguards to prevent harm caused by www.fda.gov/media/71460/download. resources/content_11574.pdf. Accessed height-related equipment whenever they Updated March 10, 2006. Accessed July January 8, 2021. are in use, for example, explicit guide- 14, 2020. 16. Medical Care Availability and lines to tare the weight of the bed and 7. Centers for Disease Control and Reduction of Error (MCARE) Act, Pub. L. bed extenders when weighing a patient Prevention. National health and No. 154 Stat. 13 (2002). in bed. Failure to do so may be difficult Nutrition Examination Survey 2015-2016 to identify because of the relatively small, Examination Data. CDC website. https:// About the Authors but critical, discrepancy. wwwn.cdc.gov/nchs/nhanes/search/data- page.aspx. Accessed November 25, 2020. Caitlyn Allen ([email protected]) is direc- Acknowledgments 8. Kotowski, S. E., Davis, K. G., tor of Engagement for the Patient Safety Wiggermann, N., & Williamson, Authority and managing editor for The authors would like to thank Michael R. (2013). Quantification of Patient Patient Safety, the PSA’s peer-reviewed Bruno, MD; Daniel Degnan, PharmD; Migration in Bed: Catalyst to Improve journal. Before joining the PSA, she was Carlos Urrea, MD; and Elizabeth Kukielka, Hospital Bed Design to Reduce Shear the project manager for Patient Safety at PharmD, MA, RPh, for their guidance and and Friction Forces and Nurses’ Injuries. Jefferson Health, where she also was the contributions. Hum Factors, 55(1), 36-47. only nonphysician elected to serve on the House Staff Quality and Safety Leadership 9. In a conversation with radiologist M. Council. Previously, Allen also was a proj- Bruno, MD, in July 2020. References ect manager and patient safety officer for 10. Cohen, M. H., FAIA, F., Nelson, Wills Eye Hospital. 1. Howden, LM, Meyer, JA. Age and G. G., Green, D. A., & Borden, C. M. Sex Composition: 2010. U.S. Census (2010). Patient Handling and Movement Neal Wiggermann is a research scientist Bureau website. https://www.census. Assessments: A White Paper. The Facility at Hillrom specializing in human factors gov/prod/cen2010/briefs/c2010br-03.pdf. Guidelines Institute. and ergonomics. He manages a lab that Published May 2011. Accessed July 14, performs scientific research and product 11. U.S. Food and Drug Administration. 2020. testing to inform the design of medical Patient Lifts Safety Guide. U.S. Food and devices, with the ultimate goal of reduc- 2. U.S. Census Bureau. Statistical Drug Administration website. https:// ing injuries in caregivers and improving Abstract of the United States: 2011. U.S. www.fda.gov/files/medical%20devices/ patient outcomes. Wiggermann received Census Bureau website. https://www2. published/Patient-Lifts-Safety-Guide.pdf. his PhD in industrial and operations engi- census.gov/library/publications/2010/ Accessed July 14, 2020. neering from the University of Michigan. compendia/statab/130ed/tables/11s0205. 12. Institute for Safe Medication pdf. Accessed July 14, 2020. Practices. Guidelines for Standard This article is published under the Creative 3. U.S. Census Bureau. State Population Order Sets. Institute for Safe Medication Commons Attribution-NonCommercial license.

Patient Safety I Vol. 3 No. 1 I March 2021 I 47 5’6”

◊Patient Safety Authority Disclosure: The author declares that they have no relevant or material financial interests.

48 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Challenges with Measurement and Transcription of Patient Height:

An Analysis of Patient Safety Events in Pennsylvania Related to Inaccurate Patient Height

By Elizabeth Kukielka, PharmD, MA, RPh◊ DOI: 10.33940/data/2021.3.5

Patient’s Weight:

Patient’s Height:

Calculated Dose based on height and weight: n accurate patient height is necessary to calculate certain patient care, and possible risk reduction strategies to decrease measurements (e.g., body surface area [BSA]) and lab the potential for these events. values (e.g., creatinine clearance [CrCl]), which may be needed to assess renal, cardiac, and lung function and to Methods Acalculate accurate medication doses. We queried the Pennsylvania Patient Safety Reporting System (PA-PSRS) and identified 679 event PA-PSRS has been collecting reports of patient safety events since reports related to an inaccurate patient height. All events were May 2004, and as a result it is one of the largest databases of its classified by the reporting facility as incidents, meaning that the kind in the world. Event reports include responses to both struc- patient did not sustain an unanticipated injury or require the tured fields (e.g., event date, patient age, patient sex, care area, delivery of additional healthcare services. The most common care facility type) and free-text narrative fields (e.g., event detail, which area group where an event occurred was outpatient/clinic (35.8%; allows the reporter to describe the details of the event in their 243 of 679). Events were categorized as being related to an error in own words). The information supplied in free-text narrative fields transcription (72.5%; 492 of 679) or measurement (7.4%; 50 of 679), is at the discretion of the reporter, so the depth and detail of the and the remainder were categorized as etiology of error unclear information varies from one event report to the next. (20.2%; 137 of 679). The most common transcription errors were the use of the wrong unit of measurement, the transposition of On October 5, 2020, we queried the entire acute care dataset another measurement with height, and typographical errors. in PA-PSRS to identify event reports related to an inaccurate Inaccurate patient heights most often led to errors in calculation patient height with an event date on or before August 31, 2020. of medication doses or laboratory values. The most common med- We employed the following search strategy: ication class involved in a dosing error was cancer chemotherapy. ● In order to ensure accuracy of patient height measurements, Event Detail, Event Comments, or Event Recommendation field contained one of the following patients should be measured at the beginning of every healthcare keywords or phrases: “wrong height,” “wrong pt height,” encounter, units of measurement should be consistent from mea- “wrong patient height,” “inaccurate height,” “inaccurate surement to transcription into the electronic medical record, and pt height,” “inaccurate patient height,” “incorrect height,” estimated patient height should never be relied upon or recorded. “incorrect pt height,” or “incorrect patient height.”

Keywords: patient height, measurement, transcription, medication ● Event Detail, Event Comments, or Event Recommendation error, electronic medical record, patient safety, medication safety field contained the keyword “height” and one of the following keywords: “switch,” “swap,” or “transcri.” ● Event Detail, Event Comments, or Event Recommendation Introduction field contained the keyword “height,” the root or abbreviation for centimeter (i.e., “cm” or “centim”), and Ideally, every patient should be measured at the beginning of the root or abbreviation for inches (i.e., “in” or “inch”) or every healthcare encounter*, whether at a routine check-up, an feet (i.e., “ft” or “feet”). emergency department visit, or prior to a procedure or surgery, to ensure baseline anthropometric (e.g., weight and height) and ● Event Type was specified as “medication error,” and Event vital (e.g., body temperature, heart rate, blood pressure, and respi- Detail, Event Comments, or Event Recommendation field ratory rate) information is accurate and up to date. An accurate contained the root or abbreviation for centimeter (i.e., patient height in particular is necessary to calculate certain mea- “cm” or “centim”) or the root or abbreviation for inches surements (e.g., body mass index [BMI] and body surface area (i.e., “in” or “inch”) or feet (i.e., “ft” or “feet”). [BSA]) and lab values (e.g., creatinine clearance [CrCl]), which may ● Event Subtype was specified as “other” and the associated be needed to assess renal function and to calculate accurate med- free-text response field contained the keyword “height.” ication doses.1,2 Inaccurate height measurements may negatively impact patient safety by causing treatment delays, medication We reviewed each event report to ensure it involved an inaccu- dosing errors, and inaccurate assessments of nutritional status.1, 2 rate patient height. We excluded any event report that was not related to an inaccurate patient height, such as those in which During a recent analysis of patient safety events related to height was mentioned but the patient safety event was related to extreme patient height submitted to the Pennsylvania Patient an inaccurate weight. Safety Reporting System (PA-PSRS)†, we observed event reports that detailed patient safety events involving an inaccurate patient We performed a descriptive analysis to evaluate trends among height. Because the etiology of and outcomes associated with an information specified by the reporting facility, including patient inaccurate patient height are distinct from those associated with age and sex, facility type, care area, harm score, and event type an extreme patient height, we undertook a separate analysis of and subtype(s). Concerning the care area field, it should be noted patient safety events related to inaccurate patient heights submit- that although PA-PSRS does specify that this should be the loca- ted to PA-PSRS. The objective of this analysis was to gain a better tion where the event occurred, there are times when it is clear understanding of factors that may contribute to an inaccurate that the reporter has listed the location where the event was patient height being recorded into a patient’s medical record, discovered. For example, a patient’s height may be transcribed incorrectly during triage in the emergency department, but the the ways in which a wrong patient height may negatively impact error may not be discovered until the pharmacist verifies the

*A healthcare encounter is defined in this article as a meeting between a patient and a healthcare provider in order to evaluate the health status of a patient or to deliver healthcare services. †PA-PSRS is a secure, web-based system through which Pennsylvania hospitals, ambulatory surgical facilities, abortion facilities, and birthing centers submit reports of patient safety–related incidents and serious events in accordance with mandatory reporting laws outlined in the Medical Care Availability and Reduction of Error (MCARE) Act (Act 13 of 2002).3 All reports submitted through PA-PSRS are confidential and no information about individual facilities or providers is made public.

50 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 patient’s medication orders, and the care area may be listed as analysis because they were not related to an inaccurate patient the pharmacy. For this reason, we did not analyze any potential height. The final dataset included the remaining 679 event reports relationship between care area group and other factors. submitted by 81 facilities.

We manually reviewed and coded each event report for the fol- Descriptive Analysis lowing (if specified): Event reports more often involved female patients (54.6%; 371 ● Whether the event was related to an error in of 679) than male patients (45.4%; 308 of 679). Event reports also measurement or an error in transcription. indicated that patients ranged in age from 1 day to 99 years, with a median patient age of 46 years (25th percentile=12 years; 75th ● The result of an inaccurate height, such as a wrong dose percentile=65 years). of medication or an inaccurate calculated measurement, laboratory value, or test result. Most event reports were submitted by an acute care hospital ● The specific medication, measurement, laboratory value, (87.0%; 591 of 679) or a children’s hospital (12.5%; 85 of 679), and or test result affected. the remaining event reports were submitted by a long-term acute care facility (n=2) and an ambulatory surgery facility (n=1). Of note, ● The medication class for each medication involved. approximately two-fifths of event reports were submitted by a single facility with event dates in 2019 and 2020, and we have highlighted All coding, reviews, and analyses were performed by a patient any place where this might have affected the data throughout the safety analyst at the Patient Safety Authority (PSA). results. While at times the PSA publishes articles highlighting the safety culture or improvement work of a facility, this was out of Results scope for this study, so we have refrained from any additional discussion related to the reporting culture at this facility. The query returned 820 event reports from the PA-PSRS database that occurred from the inception of the database in May 2004 The most common care area groups were outpatient/clinic (35.8%; through August 31, 2020. We excluded 141 event reports from the 243 of 679), medical/surgical unit (10.0%; 68 of 679), pediatric unit

Figure 1. Frequency of Event Reports Involving an Inaccurate Patient Height by Care Area Group, N=679

Care Area Group

Outpatient/Clinic 243 Medical/Surgical Unit 68 Pediatric Unit 51 Emergency Department 44 Specialty Unit 29 Pediatric Intensive Care Unit 29 Intensive Care Unit 28 Pharmacy Department 17 Laboratory 14 Imaging or Diagnostic 14 Rehabilitation Unit 13 Intermediate Unit 11 Surgical Services 9 Obstetrics and Gynecology 6 Psychiatric Unit 5 Other 98

0 50 100 150 200 250

Note: “Other” includes administration, admission/registration, labor and delivery, neonatal Number of Event Reports intensive care unit, nursery, outpatient observation, radiation oncology, and respiratory. Patient Safety I Vol. 3 No. 1 I March 2021 I 51 (7.5%; 51 of 679), and emergency department (6.5%; 44 of 679); injury or require the delivery of additional healthcare services.3 care area groups for all event reports are summarized in Figure 1. Harm scores for each event report are summarized in Figure 2; the About 70% of event reports from an outpatient/clinic were sub- most common harm scores were B2 (46.1%; 313 of 679), which is an mitted by the single facility identified as the largest reporter. event that did not reach the patient as a result of the intervention of a healthcare provider, and C (34.3%; 233 of 679), which is an Event reports were most often classified by the reporting facility event that reached the patient but did not cause harm or require as an error related to a procedure, treatment, or test (52.1%; 354 increased monitoring to prevent harm.3 About 60% of event reports of 679) or as a medication error (20.6%; 140 of 679); the remaining assigned a harm score of B2 were submitted by the single facility event reports were classified as a complication of a procedure, identified as the largest reporter. treatment, or test (1.3%; 9 of 679); equipment, supplies, or devices (0.6%; 4 of 679); or other/miscellaneous (25.3%; 172 of 679). The Qualitative Analysis vast majority of event reports were classified under the event We analyzed each event report to determine the type of error that subtype “other (specify)” (81.4%; 553 of 679), which allowed the had occurred, with particular attention to free-text fields (i.e., reporting facility to describe the event in their own terms. More Event Detail, Event Comments, Event Recommendation, and than 80% of event reports classified as an error related to a pro- Event Subtype - Other), and these are summarized in Figure 3. cedure, treatment, or test were submitted by the single facility identified as the largest reporter. Event reports that specified that the patient’s height had been transcribed, recorded, entered, or documented incorrectly (or other All event reports were classified by the reporting facility as inci- similar language) were categorized as transcription errors (72.5%; dents, meaning that the patient did not sustain an unanticipated 492 of 679). Event reports that specified that the patient had been

Figure 2. Frequency of Event Reports Involving an Inaccurate Patient Height by Harm Score, N=679

Number of Event Reports 350

313 Events that reached the patient 300

250 233

200

150

100

59 57 50 17

0 A B1 B2 C D

Harm Scores3 Harm Score A – Circumstances that could cause adverse events (e.g., look-alike medications, confusing equipment, etc.) B1 – An event occurred but it did not reach the individual (“near miss” or “close call”) because of chance alone B2 – An event occurred but it did not reach the individual (“near miss” or “close call”) because of active recovery efforts by caregivers C – An event occurred that reached the individual but did not cause harm and did not require increased monitoring (an error of omission such as a missed medication dose that does reach the individual) D – An event occurred that required monitoring to confirm that it resulted in no harm and/or required intervention to prevent harm

52 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Figure 3. Frequency of Event Reports Related to Inaccurate Patient Height by Type of Transcription Error, n=492

Type of Transcription Error

Wrong Units or Unit Conversion 136

Measurements Transposed 132

Typographical Error 52

Wrong Patient 12

Not Specified 160

0 20 40 60 80 100 120 140 160 180

Number of Event Reports

measured incorrectly or explicitly stated that the patient had not circumference was entered into the height field in the patient been measured at all were categorized as measurement errors record); and typographical errors (e.g., a height measurement of (7.4%; 50 of 679). The remainder of event reports were categorized 134 centimeters was entered into the patient record as 13.4 centi- as etiology of error unclear (20.2%; 137 of 679). This category meters or 124 centimeters). Event reports in which measurements included event reports that specified that the patient height field were transposed were submitted almost entirely by the single had been left blank (49.6%; 68 of 137) or that conflicting heights facility identified as the largest reporter. had been recorded in the patient record (16.1%; 22 of 137), and event reports that lacked sufficient detail to determine whether Measurement errors occurred less frequently and included event there had been a transcription error, a measurement error, or reports in which the patient height was not measured (58.0%; both (34.3%; 47 of 137). Distribution of event reports across these 29 of 50), height measurement equipment (e.g., an electronic categories was similar when the single facility identified as the scale with a stadiometer) malfunctioned (12.0%; 6 of 50), or the largest reporter was excluded. clinician employed the wrong technique (e.g., the patient was measured with shoes on) when measuring the patient (12.0%; Transcription errors leading to an inaccurate patient height 6 of 50). The following are examples of event reports coded as contributed to nearly three-quarters of event reports (see measurement errors: Figure 3). The following are examples of event reports coded as transcription errors: Patient was seen in oncologist's office for initiation of care. Per the patient, height was not measured at initial office visit. Height was entered in EMR [electronic medical record] as 35 Accurate height was measured upon arrival to chemotherapy cm instead of 135 cm. Error was identified and corrected in the infusion center for first treatment. Height measured in the infu- EMR by the nurse before it reached the patient. sion center was 5 cm less than the height documented in the office. Medication doses were calculated based on the incor- Height was documented as 65 cm and weight as 210 kg. Called rect height documented at oncologist’s office rather than the bedside nurse to verify, believe this should have been 65 inches patient's actual height. Call was made to oncologist’s office to and 210 lbs. fix the discrepancy, and a new order was obtained.

The most common mistakes were the use of the wrong unit of Patient height measured without shoes as 63 inches. Height that measurement (e.g., the patient was measured at 62 inches, but was entered in EMR at earlier visit was 69 inches, and this was the height was entered into the patient record as 62 centimeters, used to calculate chemotherapy doses. New height of 63 inches or the patient was measured at 5 feet 2 inches, but the height was was verified by two staff members. Chemotherapy doses had to entered into the patient record as 52 inches); the transposition of be recalculated based on new height. another measurement with height (e.g., patient weight or head

Patient Safety I Vol. 3 No. 1 I March 2021 I 53 Among measurement errors in which the patient height was not Calculated measurements, lab values, and test results were affected measured, the event report indicated that a placeholder such as 0 or in about one-third of events (30.6%; 64 of 209) that included detail 1 was entered into patient record without the patient actually being about the result associated with the inaccurate patient height. measured (34.5%; 10 of 29), the height was copied from an earlier The most common inaccurate calculated measurement was BMI admission (27.6%; 8 of 29), or the patient or caregiver reported the (25.0%; 16 of 64); other inaccurate calculated measurements height or the clinician estimated the height (27.6%; 8 of 29). included BSA and ideal body weight (IBW). The most common inaccurate calculated lab value was CrCl (43.8%; 28 of 64); other Among event reports for which the etiology of the error was inaccurate calculated lab values included blood volume, glomeru- unclear, either the height field was left completely blank (49.6%; lar filtration rate, and a vancomycin trough. Inaccurate calculated 68 of 137) or two or more conflicting heights were entered into test results included an echocardiogram, an electrocardiogram, the patient record (16.1%; 22 of 137). The following are examples and a pulmonary function test. of event reports coded as etiology of error unclear:

Patient was admitted through the emergency department yes- Discussion terday, but no height was entered into the EMR. Pharmacist is unable to calculate creatinine clearance for medication dosing To our knowledge, this is the first study of its kind to examine inac- and adjustments. curate patient heights in the context of patient safety events. Much of the existing literature and research on the topic of inaccurate Patient was seen in oncologist’s office and height was docu- patient heights focuses specifically on measurement errors. Some mented there as 4'11". I reverified the patient’s height today at measurement errors that were observed in both our study and in the cancer center as 5'1", and a second nurse verified this height. the literature include the use of improper measurement equipment Height and weight are very important because they are used by or techniques (e.g., measuring a patient with shoes on) or reliance the pharmacy to calculate chemotherapy doses. on an estimated patient height provided by a patient, caregiver, or clinician. Numerous studies have shown that patient height esti- In cases in which the height field was left blank, it was possible mates among inpatients in particular are not reliable, emphasizing that the patient had not been measured at all, or that the patient the importance of measuring patient height at each encounter.4-7 had been measured but the measurement had not been recorded. Additionally, although most events in our analysis were not explic- The remaining event reports (34.3%; 47 of 137) did not include itly the result of measurement errors, some events categorized as any information beyond that an inaccurate height was listed in transcription errors or etiology of error unclear may have involved the patient record. errors in measurement, such as events in which the wrong units were transcribed, the height field was left blank, or conflicting About one-third of event reports (30.8%; 209 of 679) included detail measurements were recorded. For example, when a clinician about the result associated with the inaccurate patient height. measures a patient, the units of measurement on the ruler would Two-thirds of those event reports (67.9%; 142 of 209) indicated be in centimeters or inches, while a patient would usually report that the inaccurate patient height led to an inaccurate medication their own height in feet and inches; so a transcription error in dosing. One hundred event reports involving an inaccurate med- which feet and inches were transcribed as inches might also have ication dose included more detail about the type of wrong dose, been an event in which the patient height was not measured. and underdoses (57.0%; 57 of 100) were observed more often than overdoses (43.0%; 43 of 100). Errors explicitly related to transcription of inaccurate patient heights were observed in nearly three-quarters of event reports A total of 142 event reports specified that a medication had been in our study. An evaluation of computer entry by nonprescribers dosed incorrectly as a result of an inaccurate patient height; some revealed that inconsistent expression of height and weight data of these events did not include any additional details about the in the EMR contributed to medication errors.8 In another study, medication involved, while others listed specific medications or researchers analyzed over 200,000 weight and height values to medication classes. The most common medication class was cancer develop a method for reducing the problems of transcription and chemotherapy (64.1%; 91 of 142), which included antineoplastic recording errors for height and weight; they observed that patterns agents, targeted therapies (e.g., monoclonal antibodies), and other in the data indicated that many outliers for individual patients were anticancer agents. Other medication classes were anti-infectives, not true outliers, but rather could be attributed to unit conversion which included antibiotics, antivirals, and antifungals (14.1%; 20 of errors, which we observed in 40.4% of event reports involving tran- 142); anticoagulants (11.3%; 16 of 142); and other classes (4.9%; 7 scription errors.9 Considering this research, it follows that frequent of 142), such as corticosteroids, antiemetics, anesthetics, and total measurement and recording of patient height may serve as a double parenteral nutrition. The remaining event reports (5.6%; 8 of 142) check and highlight errors in the patient record. did not specify a medication or medication class. Wrong medication doses were the most common result of an inac- Some of these event reports listed multiple medications; for example, curate patient height observed in our study, accounting for more one event named two antibiotics that were incorrectly dosed, and than two-thirds of event reports in which a result was specified. another event listed three different medications within the cancer The classes of medications most frequently involved in medi- chemotherapy class. Across 142 event reports that indicated that a cation errors related to inaccurate patient heights in our study medication was dosed incorrectly, 125 specific medications were were cancer chemotherapy, anti-infectives, and anticoagulants. mentioned by name (see Table 1). The medications that were most Although none of the events in our study resulted in an unan- often dosed incorrectly across all event reports were heparin (9.6%; ticipated injury requiring the delivery of additional healthcare 12 of 125), vancomycin (9.6%; 12 of 125), cyclophosphamide (8.0%; services, cancer chemotherapy and anticoagulants are consid- 10 of 125), and rituximab (8.0%; 10 of 125). ered high-alert medications in the acute care setting, meaning

54 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Table 1. Medications Dosed Incorrectly as a Result of an Inaccurate Patient Height, n=125

No. of Medication Classa Medication Name Occurrencesb azacitadine 1 bevacizumab 1 busulfan 1 carfilzomib 1 cetuximab 2 cisplatin 3 cyclophosphamide 10 Cancer cytarabine 2 Chemotherapyc daptomycin 1 decitabine 2 doxorubicin 7 etoposide 1 fluorouracil 5 gemcitabine 7 irinotecan 5 leucovorin 4 melphalan 1 methotrexate 5 oxaliplatin 4 paclitaxel 6 pemetrexed 2 rituximab 10 vinblastine 1 vincristine 2 ampicillin/sulbactam 2 ceftriaxone 1 Anti-Infective gentamicin 1 Antibiotics Agents piperacillin/tazobactam 1 tobramycin 1 aMedication classes are listed vancomycin 12 in order of frequency, and individual medications are Antifungals caspofungin 1 listed in alphabetical order ganciclovir 1 within each class for ease of Antivirals reference. Occurrences in oseltamivir 1 which only a medication class was indicated are not included enoxaparin 4 in this table. Anticoagulants heparin 12 bSome event reports listed multiple medications by name.

Corticosteroids hydrocortisone 3 c Other Cancer chemotherapy includes Antiemetics promethazine 1 antineoplastics, targeted therapies such as monoclonal Total 125 antibodies, and other anticancer agents.

Patient Safety I Vol. 3 No. 1 I March 2021 I 55 they bear a heightened risk of causing significant patient harm medications are often administered in the setting of an outpatient when they are used in error.10 Wrong medication doses could oncology clinic, this further reinforces the importance of accurate also have been the downstream result of events in which there patient heights in the outpatient setting. were errors in the calculation of BSA or CrCl if those errors were not identified and corrected, because these values are used in Overall, our findings reinforce the importance of measuring patient the calculation of dosing for numerous medications. Aside from height at the beginning of every healthcare encounter, because medication dosing, patient height is also important to accurately routine measurement of height ensures an accurate measurement assess renal, cardiac, and lung function, as well as nutritional is available for healthcare providers and serves as a double check status, and to set appropriate ventilation tidal volumes.1,2,11 to identify inaccurate measurements in the patient record.9 Best practices for patient height measurement based on the currently Although our study looked specifically at reports submitted by available evidence in the medical literature are summarized in hospitals, the most common care area group where events took Table 2.1,2,8,11-13 Facility leadership should ensure that policies and place was outpatient/clinic, which emphasizes that accurate equipment support best practices that have been put in place, patient heights are a concern in both the inpatient and outpa- and that any changes in policy are effectively communicated to tient setting. Additionally, cancer chemotherapy was the most all involved healthcare providers and support staff that practice common medication class observed in our study, and since these in that facility.

Table 2: Best Practices for Patient Height Measurement in the Healthcare Setting1,2,8,11-13

Healthcare providers should measure patient height at the beginning of every healthcare When should encounter or transfer to a new facility (e.g., from a nursing home to a hospital). Additionally, patient height be patient height should be reassessed whenever it could impact the course of care, such as measured? when dosing medications based on BSA or calculating tidal volumes for ventilation. Frequent measurement also provides a double check to ensure accuracy.

Standing: Patients who are mobile should be measured using a wall-mounted stadiometer, which is a device consisting of a vertical ruler with a sliding hori- zontal arm adjusted to rest on the top of the head. The patient should stand upright on a firm surface with shoes removed, feet together, looking straight ahead, with shoulders, buttocks, and heels touching the wall.

How should patient height be Supine: measured? Patients who are nonmobile should be measured while supine from the vertex of the head to the heel using a flexible tape measure. A more accurate method for measuring height while supine is the bookend method (BEM). For this method, the mattress must be laid flat and the head pillow removed. One BEM board (positioned upright 90° from the bed) is placed under the feet with the heels touching the board. The second BEM board (same positioning) is placed under the head with the head touching the board. An inflexible tape measure is used to measure between the BEM boards.

What units The electronic medical record (EMR) should record height in centimeters, and patients should be should be used measured in centimeters to eliminate errors in conversation between units. If the EMR records for patient height height in inches, patients should be measured in inches for the present time, but ultimately, the measurement? EMR should be converted to record in centimeters.

56 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Limitations 5. Bloomfield R, Steel E, MacLennan G, Noble DW. Accuracy of Weight and Height Estimation in an Intensive Care Unit: Despite mandatory event-reporting laws in Pennsylvania, our Implications for Clinical Practice and Research. Crit Care data are subject to the limitations of self-reporting. Because a Med. 2006;34(8):2153-7. Epub 2006/06/10. doi: 10.1097/01. standard taxonomy for reporting patient safety events related to CCM.0000229145.04482.93. PubMed PMID: 16763505. an inaccurate patient height does not exist, we may have missed 6. Leary TS, Milner QJ, Niblett DJ. The Accuracy of the relevant event reports with our query. In addition, because the Estimation of Body Weight and Height in the Intensive Care details included in each event report are left up to the discretion Unit. Eur J Anaesthesiol. 2000;17(11):698-703. Epub 2000/10/13. of the reporter, information was missing or incomplete in some doi: 10.1046/j.1365-2346.2000.00751.x. PubMed PMID: 11029569. reports, including specific details about what may have contributed to the event or the impact of an inaccurate patient height 7. Hendershot KM, Robinson L, Roland J, Vaziri K, Rizzo AG, on clinical care. Fakhry SM. Estimated Height, Weight, and Body Mass Index: Implications for Research and Patient Safety. J Am Coll Surg. 2006;203(6):887-93. Epub 2006/11/23. doi: 10.1016/j.jamcoll- Conclusion surg.2006.08.018. PubMed PMID: 17116557. 8. Santell JP, Kowiatek JG, Weber RJ, Hicks RW, Sirio CA. Our study highlights the many potential problems that arise Medication Errors Resulting From Computer Entry by during the measurement and transcription of patient height in the Nonprescribers. Am J Health Syst Pharm. 2009;66(9):843-53. hospital setting. Errors were more common in the transcription of Epub 2009/04/24. doi: 10.2146/ajhp080208. PubMed PMID: patient height, related largely to mixing up of either units or mea- 19386948. surements. The most frequently observed result of an inaccurate patient height was a wrong medication dose, and the most com- 9. Goodloe R, Farber-Eger E, Boston J, Crawford DC, Bush WS. mon medications involved were cancer chemotherapy, anti-in- Reducing Clinical Noise for Body Mass Index Measures Due to fective agents, and anticoagulants. In order to ensure accuracy of Unit and Transcription Errors in the Electronic Health Record. patient height measurements, patients should be measured at the AMIA Jt Summits Transl Sci Proc. 2017:102-11. PubMed Central beginning of every healthcare encounter, units of measurement PMCID: PMC5543370. should be consistent from measurement to transcription into the 10. ISMP List of High-Alert Medications in Acute Care EMR, and estimated patient height should never be relied upon or Settings: Institute for Safe Medication Practices; 2018 recorded. In the future, measurement equipment that interfaces [updated August 23, 2018; accessed November 28, 2020]. directly with the EMR could eliminate errors that occur between Available from: https://www.ismp.org/recommendations/ measurement and transcription of patient height. In addition, high-alert-medications-acute-list. future quality improvement studies at one or more healthcare 11. Freitag E, Edgecombe G, Baldwin I, Cottier B, Heland M. facilities may help to determine the impact of other solutions on Determination of Body Weight and Height Measurement for the measurement and transcription of patient height. Critically Ill patients Admitted to the Intensive Care Unit: A Quality Improvement Project. Aust Crit Care. 2010;23(4):197- Notes 207. Epub 2010/06/26. doi: 10.1016/j.aucc.2010.04.003. PubMed PMID: 20576445. This analysis was exempted from review by the Advarra 12. Venkataraman R, Ranganathan L, Nirmal V, Kameshwaran Institutional Review Board. J, Sheela CV, Renuka MV, et al. Height Measurement in the Critically Ill Patient: A Tall Order in the Critical Care Unit. Indian J Crit Care Med. 2015;19(11):665-8. Epub 2016/01/06. doi: References 10.4103/0972-5229.169342. PubMed PMID: 26730118; PubMed Central PMCID: PMCPMC4687176. 1. Byers D, France NE, Kuiper B. Measuring Height and Weight: From Research to Policy. Nursing. 2014;44(6):19-21. 13. Frid H, Adolfsson ET, Rosenblad A, Nydahl M. Agreement Epub 2014/05/21. doi: 10.1097/01.NURSE.0000444726.83265.c6. Between Different Methods of Measuring Height in Elderly PubMed PMID: 24841601. Patients. J Hum Nutr Diet. 2013;26(5):504-11. Epub 2013/01/09. doi: 10.1111/jhn.12031. PubMed PMID: 23294051. 2. Tipton PH, Aigner MJ, Finto D, Haislet JA, Pehl L, Sanford P, et al. Consider the Accuracy of Height and Weight Measurements. Nursing. 2012;42(5):50-2. Epub 2012/04/26. About the Author doi: 10.1097/01.NURSE.0000413627.20475.a0. PubMed PMID: 22531077. Elizabeth Kukielka ([email protected]) is a patient safety analyst on the Data Science and Research team at the Patient Safety 3. Medical Care Availability and Reduction of Error (MCARE) Authority. Before joining the PSA, she was a promotional medical Act, Pub. L. No. 154 Stat. 13 (2002). Available from: https://www. writer for numerous publications, including Pharmacy Times and health.pa.gov/topics/Documents/Laws%20and%20Regulations/ The American Journal of Managed Care. Kukielka also worked for Act%2013%20of%202002.pdf. a decade as a community pharmacist and pharmacy manager, 4. Beghetto MG, Fink J, Luft VC, de Mello ED. Estimates of with expertise in immunization delivery, diabetes management, Body Height in Adult Inpatients. Clin Nutr. 2006;25(3):438-43. medication therapy management, and pharmacy compounding. Epub 2005/12/27. doi: 10.1016/j.clnu.2005.11.005. PubMed PMID: 16376463. This article is published under the Creative Commons Attribution- NonCommercial license.

Patient Safety I Vol. 3 No. 1 I March 2021 I 57 I’m Ok ay My Perspective on Resilience

Christopher Mamrol, BSN, RN◊ “Chris, the Patient Safety Authority would like you to write a perspective piece on resilience.”

“Are you sure you called the right person?”

If pressed, I would say that I have resilience. I think everyone that works in healthcare needs some degree of resilience. I have thrived in high-stress situations in my life. I have learned from my past experiences. Upon reflection, I felt good about sharing my thoughts on resilience by the time I sat down to dinner with my wife and daughters.

“Chris, how was your day at work?”

“It was good. I was asked to write a perspective about resilience.”

“Are you sure they were talking to you?”

This was not as upsetting as it could be. Not just because my wife had the same initial reaction I did, but because resilience does not mean being okay all the time. Resilience is defined as “the ability to adapt successfully in the face of trauma, adversity, tragedy, or significant threat.”1 Resilience does not mean that a person does not experience difficulties or stress; in fact, resilience requires a crisis or some form of distress. All of us have been in those situations, such as when responding to a code or trying to get a wireless printer to work. Most of the time we bounce back from them, although sometimes that is easier for one reason or another. Perhaps paradoxically, I’ve tended to bounce back much quicker following high-stakes code situations than after dealing with technology problems.

However, sometimes we do not bounce back. Without resilience, we experience things like burnout, sleep difficulties, and depression. I have been burned-out. I have had difficulty sleeping. I have been depressed. I know what it is like to not feel resilient, and maybe that is why I was asked to write this piece. I have been not okay for periods of time. There is a nearly endless source of studies exploring the prevalence and effects of burnout in various healthcare professions and specialties. While I acknowledge that burnout can be harmful, almost everyone will have moments when they lack resilience. It is okay to not be okay sometimes.

That being the case, it is still more common to have resilience than not.2 I can think of far more situations in which I was able to bounce back and adjust to my experiences. I have been okay during crises and periods of stress. As I write this in the midst of a global pandemic, I remind myself of those times. While the COVID-19 pandemic has increased the conversation around resilience, it should not diminish how resilience impacts normal, everyday life. Resilience is always important. Sometimes, I have experienced situations that should have left me feeling

◊Patient Safety Authority Disclosure: The author declares that they have no relevant or material financial interests.

Patient Safety I Vol. 3 No. 1 I March 2021 I 59 had returned to the unit and was carrying on as if nothing had happened. I told her that she wasn’t expected to finish her shift, and that we could cover for her. “Why? I’m okay.” In the end, we came to a mutual agreement to at least assign her to another “ area of the unit. To address self-awareness, During the debriefing with administration that afternoon, we watched the camera footage and discussed the event. As I shared identify areas of irrational my observations and thoughts, I felt physically ill. When I was done, I stepped outside to get some air. When I went to return to thinking and use those to the debriefing, I overheard from just outside the room, “Imagine adopt a more balanced and how bad that could have been if Chris wasn’t there. His instincts probably saved her life.” When I told my wife the story that night, realistic thinking pattern. she showered praise, but I just felt a flood of anger and shame Accept that change is a part that I still do not fully understand. I do not tell this story very often because it makes me uncom- of life and look back at who or fortable. I did not feel like a hero then, and I do not feel like a hero when I think about it now. I think about how the psychiatric what was helpful in previous technician responded to the situation and I am still in awe. There times of distress. is no question that in this situation, she displayed more resilience than I did. Why? How?

There is definitely an innate aspect to resilience. Most of us know at least one person who is able, or at least appears able, to shrug disturbed but did not. During almost every employee evaluation off any stressor. I think some people are born more resilient than in my career, my tendency towards humor has been mentioned. others. However, some of us are born with traits that might be Joking during a crisis is not necessarily the result of being cold- good for our evolution as a species, such as constantly worrying hearted or oblivious. Coping skills are important and can man- about the future, anticipating bad outcomes, and remembering ifest in unusual ways and at unusual times. It is okay to be okay bad things, but these are not conducive to resilience or prolonged sometimes too. happiness.3 However, there are behaviors we can learn to help During my time working as a nurse on an inpatient behavioral build resilience. health unit, I certainly witnessed numerous crises and, unfortu- Resilience involves behaviors, thoughts, and actions that anyone nately, my share of violent incidents. However, when I think about can learn and develop. There are several traits that have been resilience, one particular instance stands out in my mind. During shown to be associated with higher levels of resilience.4 a shift, we had a patient under close observation due to his history of violent behavior. He called a staff member, one of our psychiatric technicians, into his room, complaining that his roommate ● Being able to reset easily following setbacks was acting inappropriately. As the staff member stepped in, I felt ● Being empathetic and compassionate a tingle on my neck, and I moved down the hall to give myself a ● Not wasting time worrying about what people think better angle to see into the room. As she began addressing the about you roommate, she was grabbed from behind by her lanyard. I saw it ● Maintaining healthy relationships happening and immediately rushed into the room, with another staff member on my heels. ● Not bowing to peer pressure ● Focusing time and energy on changing things you have The patient had grabbed her lanyard on opposite sides of the control over breakaway clip, but I was able to get my fingers under the fab- ric in an effort to protect her airway. A struggle ensued, during ● Learning from mistakes, post-traumatic growth which I looked into the staff member’s face as she struggled to ● Building and maintaining self-confidence breathe and I tried my best to pull the lanyard away from her neck. ● Maintaining perspective and staying present in the Eventually, with additional help, the lanyard was freed from the moment patient’s hands and the situation was brought under control. The staff member involved was relieved and I began documenting the There are several tools and strategies for building those traits. We in event. I was shaken up, partly due to the natural adrenaline rush healthcare often have found success looking towards other indus- of any physical altercation, but also because I kept thinking about tries or professions for improvement ideas. This can once again be how much worse the situation could have turned out. true when addressing the issue of resilience. Law enforcement and military personnel deal with life-or-death situations on a regular After about 30 minutes, I went back onto the unit to check on the basis. Instituting practices such as critical incident stress debriefing patient and touch base with the staff about covering the psychi- could have an impact in reducing the occurrences of post-traumatic atric technician’s remaining duties. I was shocked to see that she stress disorders in healthcare workers.5 In addition, widespread

60 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 industry efforts to improve organizational awareness, similar References to the Officer Safety and Wellness Group for law enforcement officers, could have a positive impact in helping organizations 1. Russo SJ, Murrough JW, Han MH, Charney DS, promote an environment supportive of resilience, including peer Nestler EJ. Neurobiology of Resilience. Nat Neurosci. support and outreach efforts. 2012; 15(11):1475-1484. doi:10.1038/nn.3234 2. Zolli, A, Healy, AM. Resilience: Why Things In order to build these traits among healthcare professionals spe- Bounce Back. New York: Free Press, 2012. cifically, the American Psychiatric Association has identified several areas on which to focus on: self-awareness, attention, physical 3. Fredrickson, B. Positivity. 2009. Harmony. R self-care, mental self-care, and cultivating positive emotions.6 4. Robertson, HD, Elliott, AM, Burton, C, Iversen, L, Murchie, P, Porteous, T, Matheson, To address self-awareness, identify areas of irrational thinking C. Resilience of Primary Healthcare and use those to adopt a more balanced and realistic thinking Professionals: A Systematic Review. Br J Gen pattern. Accept that change is a part of life and look back at Pract. 2016; 66 (647): e423-e433. DOI: 10.3399/ E who or what was helpful in previous times of distress. Formal bjgp16X685261 interventions of cognitive-behavioral therapy have been shown to have success in building resiliency at minimal cost.7 5. Sooda A, Prasad K, Schroeder D, Varkey P (2011) Stress Management and Resilience Mindfulness is a key tool for training attention, specifically Training Among Department of Medicine S present-moment awareness that is receptive, accepting, kindly, Faculty: A Pilot Randomised Clinical Trial. J Gen and appreciative.8 Mindful journaling, yoga, and other spiritual Intern Med. 26(8):858–861. practices like prayer or meditation can help. Building positive 6. The Road to Resilience. American interpersonal connections that restore hope and focus on posi- Psychological Association. 2012. http://www. I tive aspects of life and things to be grateful for is vital. apa.org/helpcenter/road-resilience.aspx. Physical self-care can be practiced by eating a healthy diet, get- 7. Guille C, Zhao Z, Krystal J, et al.: Web- ting ample sleep, and hydrating, in addition to including some based Cognitive Behavioral Therapy form of physical activity in your daily routine. Also, be sure to Intervention for the Prevention participate in activities and hobbies you enjoy. Often, as health- of Suicidal Ideation in Medical L care workers we leave work thinking about healthcare issues, Interns: A Randomized Clinical then we hear about health issues on the news, and then we watch Trial. JAMA Psychiatry. 2015; medical drama shows. Make sure you take some time away from 72(12):1192–1198. healthcare for outside interests. 8. Shapiro SL, Schwartz I GE, Bonner G: Effects of Mental self-care is just as important as physical needs. Prioritize Mindfulness-Based Stress relationships with empathetic and understanding people, both Reduction on Medical and within and outside your professional peer group. Develop real- Premedical Students. J Behav istic goals and stay motivated by doing something that gives Med. 1998; 21(6):581–599. E you a sense of accomplishment and purpose every day. Most importantly, seek help when you need it. Help is available in many different forms, and we owe it to ourselves to utilize those About the Author resources when necessary. Christopher Mamrol ([email protected]) n Finally, do what is possible to cultivate positive emotions. An is a senior patient safety liaison with the optimistic outlook empowers you to expect that good things will Patient Safety Authority for the Southeast happen to you. Go against our evolutionary urges, and visualize region of Pennsylvania. Prior to joining PSA, what you want, rather than worry about what you fear. Christopher worked at Montgomery County Emergency Service Inc. serving in multiple c A lot of material related to resilience is available. While I view roles, including as a psychiatric technician, reg- this as a very good thing, it can be overwhelming. My perspec- istered nurse, risk manager/patient safety officer, tive, provided here either intentionally or because there was a performance improvement director, and safety mix-up about whom to ask, is that it does not matter what your and quality systems director. Christopher is a resilience looks like. Journaling may help you bounce back fol- member of the Delta Epsilon Iota Academic e lowing trauma, or it may not help at all. You may need to sur- Honor Society, the American Psychiatric round yourself with loved ones or may prefer to be alone and Nurses Association, and the American Society think things through. Resilience is not a one-size-fits-all trait, of Professionals in Patient Safety. Christopher for either challenges or individuals. There are many strategies also has a Lean Six Sigma Black Belt certifica- for managing stress, and some will work while others do not. tion through Villanova University and is a Certified Sometimes we are okay. Sometimes we are not okay. The most Professional in Patient Safety. important thing is that whatever we are doing, it is the best we can do at the moment—and that is okay. This article is published under the Creative Commons Attribution-NonCommercial license.

Patient Safety I Vol. 3 No. 1 I March 2021 I 61 Risks for Organ and Vascular Complications

62 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 By Lea Anne Gardner, PhD, RN◊ DOI: 10.33940/data/2021.3.6

trocar is a hollow device used during minimally invasive surgery that serves as an entry port for optical scopes and surgical equipment. Insertion of this device into the body is determined using anatomical landmarks taking into consideration the patient’s history and physical attributes, e.g., scars or abdominal size. Insertion of the first trocar is the timeA of highest risk of injury. Intestinal and vascular injuries are two potentially life-threatening injuries that can occur. A retrospective review of trocar-related events submitted to the Pennsylvania Patient Safety Reporting System (PA-PSRS) between January 1, 2014, and June 30, 2020, identified 268 events. Internal organ and vascular injuries accounted for 81.0% of events; trocar site skin integrity injuries, bleeding/hemorrhage, and hernias accounted for 17.2% of events; and vasovagal responses accounted for 1.9% of events. Internal injuries occurring during the initial insertion of the trocar, Veress needle, or incision in preparation for a trocar insertion was reported in 64.5% of events. Adhesions were identified in 13.5% of internal injury events. Many internal injury events identified a single injury; however, in 17 instances patients sustained two trocar-related injuries. Conversion to open surgery, return to the operating room during the same admission, postoperative intensive care unit admission, ambulatory surgical facility transfer to a hospital, readmission, postdischarge return to the operating room, and death are outcomes identified in the event reports.

Keywords: patient safety, trocar, Veress needle, laparoscopy, robotic surgery, laceration, perforation

◊Patient Safety Authority Disclosure: The author declares that they have no relevant or material financial interests.

Patient Safety I Vol. 3 No. 1 I March 2021 I 63 Compared to open surgical procedures, minimally invasive reports. Manual review of these reports determined surgery (e.g., laparoscopic, thoracic, or arthroscopic) results inclusion in the final data set. in reduced recovery time, decreased scarring, and shortened ● 1 A manual review of the remaining reports identified length of hospital stay. These procedures begin with the sur- additional trocar-related reports not captured geon inserting one or more trocars, i.e., devices with hollow in the keyword search. The exclusion criteria tubes that serve as entry ports into the body for optical scopes 2,3 applied during this review comprised the following and surgical equipment. conditions: equipment issues such as defective trocars, broken specimen bags, and sterilization There are different methods used when placing the initial issues such as biofilm or tissue found on trocars; trocar into the body: lost surgical specimens that mentioned a trocar; aborted procedures due to contraindications 1. Inserting a Veress needle to create space such as administration of anticoagulants prior to between the abdominal wall and the organs (i.e., surgery; ophthalmologic procedures that used pneumoperitoneum) using carbon dioxide gas (i.e., trocars; and reports that contained the word “trocar” insufflation) to reduce the chances of the trocar within another term, such as electrocardiogram, touching the organs. retrocardiac, and dextrocardia. All non-trocar- 2. Inserting a sharp-bladed trocar directly into the body related reports were excluded. without a pneumoperitoneum. 3. Creating an opening into the body via a surgical incision to insert a blunt-edged (i.e., Hasson) trocar. Results 4. Inserting an optical trocar that provides views of the Demographic Analysis layers of the abdominal wall and organs on a monitor The data query identified 810 reports. Review and analysis of as cuts are made when entering the abdominal cavity.2,4,5 the report descriptions narrowed the final dataset to 268 trocar-related reports. Women accounted for 79.9% (214 of 268) Most minimally invasive surgical procedures are completed of the reports and men accounted for 20.1% (54 of 268). The without an iatrogenic injury (i.e., injury caused during medical median patient age was 51.5 years, interquartile range is 39–65 treatment).6 However, insertion of the initial trocar is the time years old (25th percentile and 75th percentile). The majority of when the highest risk of injury usually occurs.2,4-9 Two common, these events occurred in a hospital (95.1%; 255 of 268); the potentially life-threatening injuries that usually occur during remaining 4.9% (13 of 268) occurred in an ASF. initial trocar insertion are vascular and intestinal injuries.7,9-13 These injuries are the leading causes of death during laparo- Qualitative Analysis scopic procedures.2,4,9 An analysis of Pennsylvania Patient Safety Reporting System (PA-PSRS)a event reports, hereafter Surgical Procedures referred to as “reports,” was performed to find out what types All 268 reports were surgical procedures. In 69.8% (187 of and frequency of trocar-related safety events have occurred 268) of the reports, the surgical procedure was identified. in minimally invasive surgeries within Pennsylvania hospitals Gynecological procedures accounted for 47.6% (89 of 187), and ambulatory surgical facilities (ASFs). followed by cholecystectomies 17.1% (32 of 187) and hernia repairs 9.6% (18 of 187). In total, we identified 12 categories of surgical procedures. See Figure 1.

Trocar-Related Categories Methods We identified three groups of safety events surrounding the use of trocars and Veress needles: internal organ and vascular We queried PA-PSRSF free-text data fields using keywords “trocar,” injuries 81.0% (217 of 268), hereafter referred to as “internal “trochar,” “laceration,” “Hasson,” “Seldinger technique,” “laparos- c copy,” “obturator,” “Veress,” and “Veress needle” to identify mini- injuries;” trocar site injuries 17.2% (46 of 268); and vasovagal mally invasive surgical, trocar-related safety events that occurred responses 1.9% (5 of 268). between January 1, 2014, and June 30, 2020. Internal Injuries A two-step process refined the initial data set. Many PA-PSRS report descriptions identified circumstances with internal injuries based on when they happened; 62.7% b ● A second keyword search of the free-text data fields (136 of 217) occurred during the initial insertion of the trocar narrowed the number of potentially appropriate

aPA–PSRS is a secure, web-based system through which Pennsylvania hospitals, ambulatory surgical facilities, abortion facilities, and birthing centers submit reports of patient safety–related incidents and serious events in accordance with mandatory reporting laws outlines in the Medical Care Availability and Reduction of Error (MCARE) Act (Act 13 of 2002).13 All reports submitted through PA-PSRS are confidential and no information about individual facilities or providers is made public. bKeywords for the second search: "perforation", "laceration", "injury", "perforated", "puncture", "injure", "nicked", "bleed", "hemorrhage", "pierce", "damage", and "hematoma". cVasovagal response, or syncope, is a medical term for a sudden temporary drop in heart rate and blood pressure, causing fainting or passing out. It occurs when the vagus nerve, the part of your nervous system that regulates heart rate and blood pressure, responds to an event, such as extreme emotional duress.14,15

64 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Figure 1. Minimally Invasive Surgical Procedures With Trocar-Related Injuries, n=187*†

Minimally Invasive Surgical Procedures

Gynecological 89 Cholecystectomy 32 Hernia Repair 18 Appendectomy 10 Urinary - TURP/Cystotomy 8 Gastric Intestinal/Esophageal 8 Device Insertion/Removal/Revision 7 Roux-en-Y; Gastric Sleeve 7 Nephrectomy/Adrenalectomy 3 Pancreatectomy/Splenectomy 2 Thoracic/Breast 2 Orthopedic 1

0 10 20 30 40 50 60 70 80 90 Number of PA-PSRS Reports *There were 81 reports with no procedure specified. †Device insertion/removal/revision category are comprised of events involving abdominal ports, pleural drainage catheters, peritoneal dialysis catheters, ventral peritoneal shunts, and central lines.

Figure 2. Trocar-Related Internal Injury Circumstances, n=217

4 (1.8%) Injury Occurred During Incision for Trocar Insertion

136 (62.7%) Injury Occurred During Initial Insertion of 77 (35.5%) Trocar/Veress Needle Circumstances Surrounding Trocar Injury Not Specified or Veress needle, 1.8% (4 of 217) occurred when the incision used Patient for laparoscopic hysterectomy. Vital sign changes noted in preparation for the trocar insertion was made, and 35.5% (77 immediately after the surgeon placed the trocar through the of 217) of the reports did not specify the circumstances when the umbilicus. Procedure immediately converted to open incision trocar injuries occurred. See Figure 2. due to massive bleeding from a punctured aorta. Two additional surgeons were consulted for repair and found a small bowel In 92.2% (200 of 217) of internal injury reports, one injury per patient injury as well. Patient was transferred to ICU post-operatively. was noted;d however, 7.8% (17 of 217) described patients who sustained two internal injuries. In total, we identified 234 internal inju- This analysis led to identifying the types of internal injuries. ries across 217 reports. Below are examples of report descriptions. Intestinal injuries occurred most frequently (40.2%; 94 of 234), During an exploratory laparoscopy procedure, while placing followed by bladder and kidney injuries (17.1%; 40 of 234), and the trocar, two enterotomies [i.e., bowel injuries] were made to vascular injuries (12.8%; 30 of 234). the small bowel. The trocars were removed, and the procedure Bladder perforations accounted for 97.5% (39 of 40) of bladder and was converted to open to repair the enterotomies. kidney injuries. In 61.5% (24 of 39) of patients with a bladder injury Patient has history of abdominal surgeries. During a lapa- a foley catheter was used to allow the bladder to heal. The remain- roscopy for multiple gastrointestinal procedures, trocar place- ing 38.5% (15 of 39) of reports did not state whether a foley catheter ment went into the bowel. Multiple small bowel enterotomies was used. More than half (57.5%; 23 of 40) of these injuries occurred occurred. There was no place where the bowel was not adhered during the initial insertion of the trocar. The remaining 42.5% (17 to the abdominal wall. of 40) of reports did not indicate when the injuries occurred.

d PA-PSRS does not provide patient identifiers. When we use the word patient, we are referring to the person receiving care described in the event report. There is no way for us to confirm each patient is unique. eTotal does not add up to 100% due to rounding.

Figure 3. Number and Location of Trocar-Related Internal Injuries, n=234

Anatomical Location of Injuries

Intestinal Injury 94 Bladder/Kidney 40 Vascular 30 Unspecified Injury 17 Stomach 12 Liver 9 Mesentery/Omentum 9 Retroperitoneal 7 Abdominal Wall/Diaphragm 3 Gallbladder 3 Lung/Heart/Breast 3 Uterus 3 Pancreas 2 Spleen 2 0 10 20 30 40 50 60 70 80 90 100 Number of PA-PSRS Reports Figure 4. Surgical Procedures and the Associated Internal Injuries

Minimally Invasive Injured Internal Organs Surgical Procedures* or Blood Vessels

Appendectomy Bladder/Kidney

Cholecystectomy

Device Insertion/Removal/Revision Intestine Gastric Intestinal/Esophageal

Liver

Gynecological Mesentery/Omentum

Stomach

Unspecified

Hernia Repair Uterus

Vascular Urinary

*This diagram shows paired links between the surgical procedures and the affected organ or blood vessels. The links represent three or more reports in each category shown. The thickness of the links is proportional to the number of reports underlying that specific relationship. Links representing one or two reports between the columns are not presented here. For example, urinary procedures had fewer than three reports in the vascular and bladder/kidney injury categories.

The 30 vascular injuries were described in 29 reports. One patient procedure, additional care provided during the same admission, sustained two vascular injuries. An analysis revealed injuries to additional care postdischarge, or death. Many reports noted more the aorta occurred 33.3% (10 of 30) of the time, iliac artery 20.0% than one outcome. See Figure 5. (6 of 30), epigastric artery or vein 13.3% (4 of 30), mesenteric artery or vein 13.3% (4 of 30), vena cava 3.3% (1 of 30), and arte- In 69.1% (67 of 97) of reports that described the decision to con- rial bleeding site not specified 16.7% (5 of 30).e Eighteen of the 29 vert the surgery to an open procedure, intestinal injuries were vascular events (62.1%) reported an injury happening during the the most frequent reason (52.2%; 35 of 67) followed by vascular initial insertion of a trocar or Veress needle. injuries (19.4%; 13 of 67).

Figure 3 displays the number and location of all internal injuries. In 6.5% of the reports (14 of 97), patients required ICU care postoperatively. Fifty percent (7 of 14) of these patients experienced a To understand the link between the surgical procedures and the vascular injury. The two reported deaths also occurred in patients affected internal injuries, a Sankey diagram was created. The links with a vascular injury. between the two columns are based on three or more reports in each category shown. See Figure 4. Six of 13 (46.2%) reports submitted by ASFs identified patients who were transferred to a hospital for additional surgery, evaluation, and monitoring. Outcomes Surgical outcomes were identified in 44.7% (97 of 217) of the Nearly three-quarters (69.2%; 9 of 13) of the patients that returned internal injury reports. They were grouped by changes in the to the hospital postdischarge returned to the operating room (OR).

Patient Safety I Vol. 3 No. 1 I March 2021 I 67 Figure 5. Internal Injury Outcomes, n=97 Reports*

Surgical Outcomes

Conversion to Open Surgery 67 Return to OR During Same Hospitalization 16 Scheduled Procedure Aborted 15 Postoperative ICU Admission 14 Return to Hospital Postdischarge 13 Return to OR Postdischarge 9 Transfer from ASF to Hospital 6 Deaths 2

0 10 20 30 40 50 60 70 Number of PA-PSRS Reports

* The 141 outcomes were described in 97 reports. Many patients experienced more than one of these outcomes.

Adhesions played a role in patient outcomes. They were noted in Patient had laparoscopic cholecystectomy [i.e., gallbladder 16.1% (35 of 217) of the reports, and in 40% (14 of 35) of these reports, removal]. Patient returned to the emergency department after the description acknowledged a history of adhesions or expectation discharge with abdominal pain and low-grade fever. Imaging of possible adhesions based on the patient’s prior surgical history. showed a small bowel obstruction due to an incarcerated umbil- The remaining reports do not identify the presence of adhesions a ical hernia, [i.e., bowel trapped in the abdominal wall] through priori and/or did not state the patient’s past surgical history. the trocar site. Patient returned to the operating room where it was found the bowel had returned to the abdominal cavity. Intestinal injuries accounted for 65.7% (23 of 34) of these patient The connective tissue around the trocar site was closed and injuries and 69.5% (10 of 23) had their procedure converted to an patient was admitted. open surgery. Four of the 34 (11.8%) patients returned to the hospital and the OR for a bowel perforation and/or resection. Patient had laparoscopic appendectomy and was discharged home the same day. Four days after the procedure, patient Trocar Site Injuries presented to the emergency room with nausea, vomiting, and Trocar site injuries included skin integrity issues, infections, abdominal pain. Patient returned to surgery to repair a trocar bleeding/hematomas, and hernias. Skin integrity issues accounted site hernia. The bowel was incarcerated but still viable. for 56.5% (26 of 46) of the trocar site reports. Many of these reports identified minor injuries such as abrasions, bruising, skin tears, Outcomes lacerations, and a potential burn. Trocar site outcomes were described in 43.5% (20 of 46) of these reports. They were grouped by additional care provided during Trocar site bleeding and hematomas were grouped together and the same admission or return to the hospital postdischarge. accounted for 26.1% (12 of 46) of these reports. Eleven of these 12 patients returned to the OR to address the underlying cause Returning to the OR during the same admission due to trocar site for the bleeding or hematoma. Trocar site hernias occurred in 7 bleeding or hematomas was described in 50% (10 of 20) of the reports patients (15.2%). One patient with hernia had an adhesion. Below with outcome descriptions. One report described a patient with tro- are report descriptions. car site bleeding that received ICU care postoperatively. Nine reports described patients returning to the hospital postdischarge of which Patient had laparoscopy. In the post-acute care unit, a nurse dis- 77.8% (7 of 9) went back into surgery. All 7 patients had trocar site covered a hematoma developing at an abdominal trocar site. The hernias. Two trocar site skin infections were reported. One patient patient returned to the operating room for evacuation of the fluid. returned to the hospital due to a trocar site infection.

Patient returned to the hospital two weeks after robotic prostate Vasovagal Responses surgery with complaint of fluid leaking from the abdominal incision. Patient returned to the operating room to close the Five reports described vasovagal responses to the insertion of the leaking trocar site. trocar. Cardiac arrest occurred in 60.0% (3 of 5) and unstable vital

68 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 signs, e.g., diaphoresis, unresponsive (noncardiac arrest), and Trocar Site Injuries respiratory arrest (i.e., stopped breathing) occurred in 40.0% (2 Trocar site injuries identified in this analysis included bleeding/ of 5) patients. All five reports described the patients were success- hematomas, hernia, and skin injuries/infections. Trocar site bleed- fully resuscitated and/or stabilized. One patient was observed in ing and hematomas may not be noticed immediately. A couple the ICU overnight. All five procedures were aborted. of reports described vital sign changes and drops in hemoglobin and hematocrit postoperatively that signaled further investiga- Patient scheduled for minimally invasive surgical procedure. tion. Issues such as fascia bleeding and possible arterial bleeders During placement of the trocar, the patient’s heart rate dropped were identified, and all but one case required a return to the OR to then went asystolic [i.e., heart stopped beating]. CPR was suc- address the hematoma and control the bleeding. cessfully performed, and the procedure was aborted. Evaluation post-operatively showed this was likely a vasovagal response from Another serious issue is postoperative trocar site hernias. They 23-26 24,25 the trocar placement. may be asymptomatic or require emergency surgery. L e ft unattended they can result in incarcerated hernias (i.e., bowel is 27 After inserting the trocar, the patient stopped breathing. Chest trapped in the abdominal wall), strangulated hernias (i.e., trapped 28 27,29 x-ray showed no evidence of pneumothorax, (i.e., collapsed bowel and blood is cut off to the bowel), bowel obstructions, 30 lung). The procedure was aborted, and patient was evaluated and necrotic bowels. One study identified umbilical incisional post-operatively. hernias occurred three to 36 months (mean, 8.8 months) after the operation.28 The literature surrounding potential hernia development postoperatively addresses the impact of trocar size.3,23-25,27,31- Discussion 34 Hernias are more likely to occur when a 10 mm or larger trocar is used,23,24,27,31,33,35yet, they can form when 5 mm trocars are Most patients have no complications following minimally invasive used.23,24,27,31,34,35 Body size also plays a role in hernia formation. surgery. However, there are times when injuries may exist and go Obese patients are at higher risk of hernias because of increased undetected until a later time postdischarge.6 This situation can intra-abdominal pressure.27 lead to health emergencies. Skin injuries ranged from red marks and skin tears to lacerations Our data shows intestinal injuries were the most frequently and abrasions. The report descriptions indicated that these were reported injury with some requiring readmissions. Delays in minor injuries. The 2 infections were addressed with antibiotics. diagnosing intestinal injuries are a recognized issue that can lead to life-threatening illness and death.2, 3, 11 We have reports of Vasovagal Responses patients with intestinal injuries returning to the hospital and OR postdischarge requiring additional care. Cardiac arrests and reduced pulmonary compliance can occur with the insufflation of carbon dioxide (CO2) in the peritoneum.36, 37 The two next most frequent injuries, bladder and vascular, pro- There is a paucity of information regarding patients who have vided information that deviated from the literature. Urinary a vasovagal response with the insertion of the trocar. Yong et. al. injuries have been identified as more likely to occur during the identified vagal responses as a potential cause for cardiac arrest insertion of a second trocar.3 Our reports described close to 60% due to rapid peritoneal distension with CO2 insufflation and identified cases of pelvic organ manipulation leading to severe vagal of bladder injuries occurred during the insertion of the initial 37 trocar. Regarding vascular injuries, they usually occur during the responses. One case study identified what was believed to be a case initial insertion of a trocar or Veress needle.7, 9 Our data shows of a severe vagal reaction and asystolic cardiac arrest triggered by manipulation of the patient’s gallbladder.38 While this report was almost two-thirds of the vascular injuries occurred during the not associated with a trocar insertion, it was also believed not to initial insertion. be associated with CO2 insufflation in the peritoneum. Our reports Two patient characteristics that can affect laparoscopic surgical described situations of cardiac arrest and unstable vital signs as outcomes and internal injuries are abdominal adhesions and body occurring when the trocar was inserted. There was no mention size. Abdominal adhesions are bands of scarlike tissue that form of CO2 insufflation. Reports with a cardiac arrest when CO2 was between two or more organs or between organs and the abdom- inserted into a patient’s abdomen were excluded from this data set. inal wall.16 They are common and often develop after abdominal surgery.16 Adhesions and scarring can result in organs such as the General Safety Measures to Reduce the Risks of Trocar- intestine adhering to the abdominal wall or other organs, increas- Related Safety Events ing the likelihood of the trocar/Veress needle perforating or nicking The general safety measures listed below provide ways to iden- an organ during insertion.7, 9, 16 tify, address, and mitigate circumstances that can increase the Body size is an important consideration during minimally invasive likelihood of injuries. One statement found in the literature sug- 2 procedures4-8, 17-21 because anatomical landmarks of the aorta to the gested selecting patients at low risk for complications. This goal umbilicus vary with body size.18, 20-22 Thin patients are at higher risk is hard to achieve. Surgeons are faced with patients who present for complications such as vascular or intestinal injuries during the with comorbidities (i.e., patients with one or more chronic health 39-41 initial trocar insertion due to a small amount of space between their conditions) that raise the likelihood of injuries. abdominal wall, organs, and major vessels.4, 5, 20 Patients who are obese, Obesity is a good example. Between 1999 and 2018, the prevalence depending on their abdominal girth and weight, have challenges due of obesity increased from 30.5% to 42.4% and the prevalence of to the size of their panniculus, (i.e., excess fatty tissue in the lower severe obesity increased from 4.7% to 9.2%.42 This population has abdomen), and the hospital’s or ASF’s ability to have the appropriately challenges with the size of their abdominal girth along with an sized equipment to successfully carry out the surgery.18,19,21

Patient Safety I Vol. 3 No. 1 I March 2021 I 69 increased risk for many serious health conditions such as hyper- ● Patient education safety measures: tension (i.e., high blood pressure), diabetes, heart disease, gall- ○ Encourage patients to: bladder disease, sleep apnea, and stroke.43 Even with the higher risk of injury, it has been identified that patients who are obese ■ Ask their surgeon about alternative procedures2 benefit from minimally invasive surgery compared to undergoing ■ Carefully read all preoperative and postoperative 18,21 laparotomy (i.e., open surgery). materials2 Incorporating these safety measures can assist the physician in ■ Seek clarification of any unclear or confusing weighing the risks and benefits of performing minimally invasive instructions2 surgery versus open surgery and making the appropriate recom- ■ Seek medical attention immediately if signs or mendation of care for the patient. symptoms of complications develop2 ● Screen patients for history of surgical procedures; ○ Provide patients with information about signs and medical comorbidities (e.g., respiratory function in symptoms of any unrecognized trocar injury after patients who are obese, obstructive sleep apnea, or the procedure, such as trocar site hernia2,7,49,51 history of smoking) and conditions that increase the ○ Offer clear explanations to the patient in the likelihood of intra-abdominal adhesions (e.g., prior open written consent and information given regarding or laparoscopic surgical procedures, endometriosis, pelvic the risks and potential complications, including inflammatory disease, or inflammatory bowel disease); 3, 4, 7, 21 serious risks associated with laparoscopic surgery and history of adhesions and possibility of conversion to laparotomy if ● Consider patients with more than two prior abdominal clinical circumstances dictate or repairs to bowel, procedures at higher risk of having an inadvertent bladder, or blood vessels are necessary2,4,6,7,21,49,51 enterotomy12 ○ Counsel patients about individual risk related to ● Identify high-risk patients (e.g., patients with prior open their body mass index6,7 or laparoscopic surgery) and tailor consent appropriately17 ● Evaluate patient’s body type prior to beginning the There are a couple of evidence-based suggestions about closure procedure5, 7 of trocar sites and hernias.

● Prior to surgery, determine if the size of available surgical ● Fascial closing is recommended when using 10 mm or equipment (e.g., Veress needle or trocars), along with larger trocars3,23,52 standard equipment (e.g., blood pressure cuffs or stretchers) are the right size for patients who are obese19,21,44,45 ● There are mixed results regarding closing the fascia around the trocar site of a 5 mm trocar versus leaving it ● Consider strategies to reduce chances of injury during open23,24,27,35,52 trocar insertion, e.g., Palmer’s point for patients who are either thin or morbidly obese 5,6,8,18,20,21 and patients at risk 2,5,6,20 of adhesions Limitations ● Evaluate alternative entry sites and techniques based on history, body size, and procedure2,7 The narrative section of event reports submitted to PA-PSRS are provided by healthcare staff as an open-ended story. The infor- ● Seek consultation from experienced colleagues regarding mation provided in these report descriptions varies, thereby lim- questions about addressing concerns when caring for iting certain information and conclusions. For example, reports high-risk patients prior to starting the procedure12 can include long descriptions about the procedure, injury, and ● Investigate preoperative ultrasound, which has been shown circumstances surrounding the injury, or may be a short sen- to help identify the presence and location of adhesions in tence with very little detail. Patient weight is also unknown, lim- high-risk patients prior to laparoscopic surgery46-49 iting conclusions about body size in relation to a vascular injury. ● Maintain a low threshold for imaging or investigation in Infection reporting requirements also vary. Pennsylvania hos- the postoperative period when there is a high suspicion pitals report infections to the Centers for Disease Control and of injury17 Prevention’s National Healthcare Safety Network, whereas ASFs report infections to PA-PSRS. Therefore, the total number of trocar ● Arrange at minimum to have a general surgeon on site infections is unknown. The paucity of data on patients who standby in case of emergency patient complications. had similar procedures coupled with possible underreporting of When possible, have specialty surgeons (e.g., vascular trocar injuries prevents calculating complication rates. Finally, 2,17 and/or gastrointestinal) on standby information about complications arising from these injuries in the ● Be vigilant in observing for injuries and upon future, such as delays in identification of an intestinal injury, may recognition of injury immediately apply appropriate be reported but there is no direct link of future events with earlier surgical management; check and recheck suspected reports. The only way a future event is captured and linked with a areas of injury2,12 prior surgery is when the reporter provides detailed information such as the initial surgical date, procedure performed, type of ● Report adverse events involving the use of trocars trocar, and high-risk situations. through the Food and Drug Administration’s voluntary adverse event reporting program50 and, for Pennsylvania facilities, report adverse events to PA-PSRS.

70 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Conclusions 7. Alkatout I. Complications of Laparoscopy in Connection with Entry Techniques. J Gynecol Surg. 2017;33(3):81-91. doi: No surgical procedure is without risks. Our analysis revealed intes- 10.1089/gyn.2016.0111. tinal injuries and complications were the most frequent patient safety event associated with laparoscopic trocar insertions. 8. Thepsuwan J HK, Wilamarta M, Adlan AS, Manvelyan V, Urinary and vascular injuries were the next two most frequent Lee CL,. Principles of Safe Abdominal Entry in Laparoscopic patient safety events, respectively. The general safety measures Gynecologic Surgery. Gynecol Minim Invasive Ther. 2013;2:105-9. presented in this article offer risk reduction strategies for sur- 9. Belena JM NM. Postoperative Complications of geons to consider when encountering patient conditions that Laparoscopic Surgery. Int J Clin Anesthesiol. 2014;2(3):1034-9. can increase the chance of an injury. One risk reduction strat- 10. Jacobson MT OS, Milki A, Nezhat C,. Laparoscopic Control egy we identified involves educating patients about their proce- of a Leaking Inferior Mesenteric Vessel Secondary to Trocar dure and engaging them in a discussion with their surgeon. This Injury. JSLS. 2002;6(4):389-91. PubMed Central PMCID: interaction provides patients an opportunity to express prefer- PMCPMC3043438. ences about their care and making informed decisions.53,54 It has been shown that engaged, knowledgeable patients have better 11. LeBlanc KA EM, Corder III JM,. Enterotomy and Mortality health outcomes and better care experiences.54,55 Surgeons can Rates of Laparoscopic Incisional and Ventral Hernia Repair: use patient information gained preoperatively to weigh the risks A Review of the Literature. JSLS. 2007;11(4):408-14. PubMed and benefits of performing minimally invasive surgery versus Central PMCID: PMCPMC3015847. open surgery and make the appropriate recommendation of care 12. Binenbaum S GM. Inadvertent Enterotomy in Minimally for their patients. Suggestions for future research on this topic Invasive Abdominal Surgery. JSLS. 2006;10:336-40. include updated complication rates, as current available rates date back to the 1990s and 2000s. Updated complication rates can 13. Karadag MA CK, Demir A, Bagcioglu M, Kocaaslan R, help identify trends based on current evidence-based practices. Kadioglu TC,. Gastrointestingal Complications of Laparoscopic/ Another area of research is the impact of patient education and Robot-Assisted Urologic Surgery and a Review of the LIterature. engagement on patient choices and outcomes. J Clin Med Res. 2015;7(4):203-10. doi: http://dx.doi.org/10.14740/ jocmr2090w. 14. Mayo Clinic. Vasovagal Syncope 2018. Available Notes from: https://www.mayoclinic.org/diseases-conditions/ vasovagal-syncope/symptoms-causes/syc-20350527. This analysis was exempted from review by the Advarra Institu- tional Review Board. 15. Cedars Sinai. Vasovagal Syncope 2020. Available from: https://www.cedars-sinai.org/health-library/diseases-and-conditions/v/vasovagal-syncope.html. References 16. National Institute of Diabetes and Digestive and Kidney 1. American Institute of Minimally Invasive Surgery. Diseases. Abdominal Adhesions 2019 [cited 2020 27 August]. Benefits of Minimally Invasive Surgery 2020 [cited Available from: https://www.niddk.nih.gov/health-information/ 2020 28 August]. Available from: https://aimis.org/ digestive-diseases/abdominal-adhesions. benefits-of-minimally-invasive-surgery/. 17. O'Donovan OP VA. Complications of Laparoscopic Surgery. 2. Fuller J SW, Ashar B, Corrado J,. Laparoscopic Trocar Obstet Gynaecol Reprod Med. 2017;27(7):213-7. Injuries: A Report from a U.S. Food and Drug Administration 18. Biswas N HP. Review Surgical Risk From Obesity in (FDA) Center for Devices and Radiological Health (CDRH) Gynaecology. Obstet Gynecol. 2011;13:87-91. doi: 10.1576/ Systematic Technology Assessment of Medical Products toag.13.2.87.27651. (STAMP) Committee: FDA Safety Communication. In: FDA, 19. American College of Obstetricians and Gynecologists. editor. Silver Sping, MD2003. Gynecologic Surgery in the Obese Woman: Committee Opinion. 3. Hindman NM KS, Parikh MS,. Common Postoperative 2015. p. 1-5. Findings Unique to Laparoscopic Surgery. RadioGraphics. 20. la Chapelle CF BW, Rademaker BMP, van Barneveld TA, 2014;34(1):119-38. doi: https://doi.org/10.1148/rg.341125181. Jansen FW,. A Multidisciplinary Evidence-Based Guidelines 4. Krishnakumar S TP. Entry Complications in Laparascopic for Minimally Invasive Surgery: Part 1: Entry Techniques Surgery. J Gynecol Endosc Surg. 2009;1(1):4-11. doi: 10.4103/0974- and the Pneumoperitoneum. Gynecol Surg. 2012;9(3):281- 1216.51902. PubMed Central PMCID: PMCPMC3304260. 82. doi: 10.1007/s10397-012-0731-y. PubMed Central PMCID: 5. Djokovic D GJ, Thomas V, Maher P, Ternamian A, Vilos PMCPMC3401300. G, L:oddo A, Reich H, Downes E, Rachman IA, Clevin 21. Scheib SA TIE, Green IC, Fader AN,. Laparoscopy in the L, Abrao MS, Keckstein G, Stark M, van Herendael B. Â· Morbidly Obese: Physiologic Considerations and Surgical Principles of Safe Laparoscopic Entry. Eur J Obstet Gynecol Techniques to Optimize Success. J Minim Invasive Gynecol. Reprod Biol. 2016;201:179-88. doi: http://dx.doi.org/10.1016/j. 2014;21:182-95. doi: http://dx.doi.org/10.1016/j.jmig.2013.09.009. ejogrb.2016.03.040. 22. Hurd WW BR, DeLancey JOL, Pearl ML,. The Relationship 6. Royal College of Obstetricians and Gynaecologists. of the Umbilicus to the Aortic Bifurcation: Implications for Preventing Entry-Related Gynaecological Laparoscopic Laparoscopic Technique. Obstet Gynecol. 1992;80:48-51. Injuries: Green-top Guideline No. 49. 2008. 23. Dincel P BF, Goksu M,. Causes of Asymptomatic Trocar

Patient Safety I Vol. 3 No. 1 I March 2021 I 71 Site Hernia: How Can it be Prevented? North Clin Istanbul. 38. Valentin MD TN, Dolgin C,. Recurrent Asystolic Cardiac 2015;2(3):210-14. doi: 10.14744/nci.2015.50479. Arrest and Laparoscopic Cholecystectomy: A Case Report and 24. Gogacz M KA, Winkler I, Adamiak A, REchberger T, Phillipe Review of the Literature. JSLS. 2004;8:65-8. K,. Suture or Not to Suture? Trocar-Site Bowel Herniation as 39. Centers for Disease Control and Prevention. Chronic a Rare Complication after Trocar Placement in Laparoscopic Diseases in America: US Department of Health and Human Surgery: Case Report and Review. Clin Surg. 2018;3. Services; 2019 [cited 2020 16 September]. Available from: 25. Karampinis I LE, Hetjens S, Vassilev G, Galata C, Reissfelder https://www.cdc.gov/chronicdisease/resources/infographic/ C, Otto M,. Trocar Site HERnias after Bariatric Laparoscopic chronic-diseases.htm. Surgery (HERBALS): A Prospective Cohort Study. Obes Surg. 40. National Council on Aging. Health Aging. Washington DC: 2020;30:1820-26. National Council on Aging; 2014. p. 2. 26. Rossi A MD, Witte S, LynSue J, Haluck RS, Rogers AM,. 41. Akinyemiju T JM, Moore JX, Pisu M,. Disparities in the An Expanded Retrospective Review of Trocar Site Hernias in Prevalence of Comorbidities Among US Adults by State Laparoscopic Gastric Bypass Patients. J Laparoendosc Adv Surg Medicaid Expansion Status. Prev Med. 2016;88:196-202. doi: Tech. 2017;27(6):633-5. doi: 10.1089/lap.2016.0483. http://dx.doi.org/10.1016/j.ypmed.2016.04.009. 27. Tonouchi H OY, Kobayashi M, Kusunoki M,. Trocar Site 42. Centers for Disease Control and Prevention. Adult Obesity Hernia. ARch Surg. 2004;139:1248-56. Facts 2020 [16 September 2020]. Available from: https://www. 28. Coda A BM, Ferri F, Mattio R, Ramelini G, Poma A, Quaglino cdc.gov/obesity/data/adult.html. F, Filippa C, Alberto B,. Incisional Hernia and Fascial Defect 43. Centers for Disease Control and Prevention. Adult Obesity Following Laparoscopic Surgery. Surg Laparosc Endosc Percutan Causes and Consequences 2020 [16 September 2020]. Available Tech. 2000;10(1):34-8. from: https://www.cdc.gov/obesity/adult/causes.html. 29. Phillips E SD, Towfigh S,. Working Port-Site Hernias: To 44. Gardner LA GC. Class III Obese Patients: Is Your Hospital Close or Not to Close? Does it Matter in the Obese? Bariatric Equipped to Address Their Needs? PA Patient Saf Advis. Times. 2011;8(6):24-30. 2013;10(1):11-8. 30. Delmonaco P CR, La Mura F, Morelli U, Migliaccio C, 45. Kukielka E. How Safety Is Compromised When Hospital Napolitano V, Trastulli S, Farinella E, Giuliani D, Desol A, Equipment Is a Poor Fit for Patients Who are Obese. Patient Saf. Milani D, Di Patrizi MS, Spizzirri A, Bravetti M, Sciannameo V, 2020;2(1):49-54. doi: 10.33940/data/2020.3.4. Avenia N, Francesco S, . Trocar Site Hernia After Laparoscopic 46. Tu FF LG, Hartmann KE, Steege JF,. Preoperative Ultrasound Colectomy: A Case Report and LIterature Review. Int Sch Res to Predict Infraumbilical Adhesions: A Study of Diagnostic Notices. 2011:1-7. doi: 10.5402/2011/725601. Accuracy. Am J Obstet Gynecol. 2005;192:74-9. doi: 10.1016/j. 31. Karthik S AA, Shibumon MM, Pai MV,. Analysis of ajog.2004.07.034. Laparoscopic Port Site Complications: A Descriptive Study. J 47. Minaker S MC, Hayashi A,. Can General Surgeons Evaluate Min Access Surg. 2013;9(2):59-64. Visceral Slide with Transabdonimal Ultrasound to Predict Safe 32. Kang DI WS, Lee DH, Kim IY,. Incidence of Port-Site Hernias Sites for Primary Laparoscopic Port Placement? A Prospective After Robot-Assisted Radical Prostatectomy with the Fascial Study of Sonographically Naive Operators at a Tertiary Center. Closure of Only the Midline 12-mm Port Site. J Endourol. Am J Surg. 2015;209:804-9. 2012;26(7):848-51. doi: 10.1089/end.2011.0630. 48. Kothari SN FL, Lambert PJ, Mathiason MA,. Use of 33. Nofai MN YA, Hamdan FF, Oudat AH,. Characteristics Transabdominal Ultrasoundto Identify Intraabdominal of Trocar Site Hernia after Laparoscopic Cholecystectomy. Adhesions Prior to Laparoscopy: A Prospective Blinded Study. 10. 2020:2868. doi: https://www.nature.com/articles/ Am J Surg. 2006;192:843-7. doi: 10.1016/j.amjsurg.2006.08.055. s41598-020-59721-w. 49. Society of American Gastrointestingal and Endoscopic 34. Chorti A AS, Michalopoulos A, Papavramidis TS,. Richter's Surgeons (SAGES). Diagnostic Laparoscopy Patient Information Hernia in a 5-mm Trocar Site. SAGE Open Med Case Rep. From Sages. 2015. 2019;7:1-4. doi: DOI: 10.1177/2050313X18823413. 50. U.S. Food and Drug Administration. MedWatch: The FDA 35. Yamamoto M ML, Zaritsky E,. Laparoscopic 5-mm Trocar Safety Information and Adverse Event Reporting Program 2020 Site Herniation and Literature Review. JSLS. 2011;15:122-6. doi: [cited 2020 15 Oct]. Available from: https://www.fda.gov/safety/ DOI: 10.4293/108680811X13022985131697. medwatch-fda-safety-information-and-adverse-event-report- 36. Atkinson TM GG, Togioka BM, Jones DB, Cigarroa JE,. ing-program. Cardiovascular and Ventilatory Consequences of Laparoscopic 51. Royal College of Obstetricians and Gynaecologists. Surgery. Circulation. 2017;135:700-10. doi: 10.1161/ Diagnostic Laparoscopy: Consent Advice No. 2. 2008. CIRCULATIONAHA.116.023262. 52. la Chapelle CF BW, Bongers MY, van Barneveld TA, 37. Yong J HP, Runciman WB, Coventry BJ,. Bradycardia as Jansen FW,. A Multidisciplinary Evidence-Based Guidelines an Early Warning Sign for Cardiac Arrest During Routine for Minially Invasive Surgery: Part 2 - Laparoscopic Port Laparoscopic Surgery. J Qual Health Care. 2015;27(6):473-78. doi: Instruments, Trocar Site Closure, and Electrosurgical 10.1093/intqhc/mzv077. Techniques. Gynecol Surg. 2013;10:11-23. doi: 10.1007/ s10397-012-0776-y.

72 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 53. Wolters Kluwer. Engaged Patients are Healthier Patients. Wolters Kluwer LLC; 2019. 54. Agency for Healthcare Research and Quality. Strategy 61: Shared Decisionmaking. Rockville, MD: Agency for Healthcare Research and Quality; 2020. 55. Hibbard J GJ. What the Evidence Shows About Patient Activation: Better Health Outcomes and Care Experiences; Fewer Data on Costs. Health Aff. 2013;32(2):207-14. doi: 10.1377/ hlthaff.2012.1061.

About the Author

Lea Anne Gardner ([email protected]) is a patient safety analyst with the Patient Safety Authority. She has more than 30 years of research experience in patient safety, performance improvement, cardiology, and behavioral health, and a diversified work experience, including roles as a national director of quality improvement, administrative director of population health and an institutional review board, a coronary intensive care nurse, a clinical research nurse, and a nurse research/leadership preceptor. Gardner is also a Certified Professional in Patient Safety (CPPS).

This article is published under the Creative Commons Attribution- NonCommercial license.

Patient Safety I Vol. 3 No. 1 I March 2021 I 73 74 II PatientSafetyJ.comPatientSafetyJ.com II Vol.Vol. 33 No.No. 11 II MarchMarch 20212021 Derogatory Language in Charting: The Domino Effect

By Bethany Davis, MA, LSWAIC◊ DOI: 10.33940/culture/2021.3.7

“Dr. Smith” is meeting with a patient in Room 3 who has been admitted for abdominal pain and nausea. He suspects the patient may have appendicitis and will need her appendix removed. Dr. Smith begins to chart, “Ms. Dorian is a 34-year-old female; has a history of abusing alcohol; and is here to be consulted for an appendectomy upon admission for nausea, vomiting, and abdominal pain.”

The Room 3 nurse sits down at the computer to review Dr. Smith’s note on her patient. “Hmmm, maybe a consult for substance use would be helpful for this patient,” she thinks. The nurse puts in an electronic consult request for a social worker to meet with the patient in Room 3.

A few hours later, a unit social worker knocks on the door of Room 3. “Hello, I’m here to conduct an assessment for substance use and discuss alcohol-use resources.” The patient is holding her abdomen while she looks with wide eyes at the social worker. “I have been sober for 15 years. Why are you here?”

No matter your background and expertise in healthcare, it is still possible for you to be impacted by stigmatizing language through reviewing a chart. Medical documentation is used to coordinate and communicate care plans from provider to provider and with members of a patient’s care team. In reality, a patient’s history and plan of care is not the only thing that can be passed through. An important theme of this article is exposing the fact that bias can be passed from provider to provider through stigmatizing language in clinical documentation.

Is Stigmatizing Language Actually a Patient Safety Issue? Yes, indeed, it is. Creating assumptions based on documentation is dangerous for many reasons. When we make assumptions that a patient knows what they don’t know, this leaves room for error and places the patient at risk. Let me explain further. For example, labeling a patient as a frequent flyer can be a barrier to appropriate follow-up care, by means of the assumption that a patient will be back, impacting staff’s efforts in creating a sustainable care plan for the patient to follow through with. If we act on our bias, believing a patient will be back, documenting the patient will be back, we assume we don’t need to take further steps at preventing them to come back. This is the cycle that can be perpetuated by stigmatizing language and its effect on bias and further treatment of patients.

◊Virginia Mason Hospital and Seattle Medical Center Disclosure: The author declares that they have no relevant or material financial interests. Patient Safety I Vol. 3 No. 1 I March 2021 I 75 A further example is provided by a research study at Johns Hopkins School of Medicine published in 2018, which was conducted to observe the impact of bias in documentation on medical provider bias. In order to assess pain management and attitudes toward patients, the study provided two different What does stigmatizing chart notes describing the same hypothetical patient: one note with derogatory language, the other note with neutral lan- language look like? guage. When staff reviewed stigmatizing language regarding the hypothetical 28-year-old patient with sickle cell disease, the effects were significant. The study revealed that with the chart notes exposing stigmatizing language to staff, there was less comfort in managing the patient’s pain and more negative attitudes toward the patient based on surveys completed by medical providers.

When there are different modes of care and engagement with patients, this most definitely is an issue of patient safety.

Interestingly, the study also showed that after reviewing stig- Patient refuses Not tolerating matizing language, experienced medical providers showed more of a change in their attitude toward the patient than less experienced providers did. Substance abuser Substance use Further studies have shown that medical professionals’ attitudes have impacted patients’ feelings of empowerment and Individual that Sexual assault action-oriented tasks. This makes it even more essential for experienced healthcare workers to increase an inner awareness around victim sexual assault personal bias in regard to the patient’s safety and overall care, no matter your length of time in the medical field or your expertise. Patient with Schizophrenic schizophrenia “No one really reads my notes, does it actually matter?” Testing negative Clean or dirty It can be helpful to visualize documentation as a domino effect or testing positive in this regard. You knock the domino down whenever you chart bias, misinformation, or criminalizing language, which Neonatal symptoms of in essence knocks your domino down, and then the next, and Addicted baby the next, and so on and so forth. withdrawal Look around: Everyone is a domino, including medical assis- tants, nurses, physical and occupational therapists, phleboto- Person with substance mists, lab technicians, X-ray technicians, pharmacists. When Opioid addict use disorder, specifically you chart, it is not just you reading the chart, it is every health- opioid use care worker accessing the information you provide.

Hospital slang What Do We Do? Steps to Take: The Person such as Before the Diagnosis “frequent flyer” Using person-first language in clinical documentation allows for patients to be seen and described as human beings. Centering the patient in the documentation versus clinical behaviors, diagnosis, or medical histories is critical.

Some providers find it helpful to include space in documentation to list a patient’s hobbies, meaningful relationships, and desire for medical interventions or care. For example, “John is hopeful that engaging in physical therapy will increase his chances of rejoining his bowling league this fall.” This honors

76 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 the patient and can serve as a kind reminder to staff thatthe judgement or a mindset of punishment. Of note, studies have individual you are working with is a person, not just your patient. shown that bias and judgement have substantially increased with terms such as “abuser” versus “patient with a history of substance Substance Use and Mental Health use” in clinical documentation. You can be a long-standing physician or equipped mental health professional and still have habits Mental health diagnoses give us information that can impact a in documentation that may unintentionally impact another pro- care plan but are not meant as adjectives or labels to describe a vider’s mindset before they walk into the room. Lax, derogatory person. More often than not, mental health diagnoses have been language in clinical documentation is detrimental to the patient used as adjectives to describe patients in medical documenta- you have been charting on. tion. For example: “depressed patient” vs. “patient with history of depression.” This is faulty in the sense that this takes away from When we use awareness in charting, we are not only being accu- the person-first language approach, which can aid to risk of bias rate, but also honestly advocating for patients and their authentic upon reading documentation in a chart. selves, with a person-first approach. Patients can have multiple diagnoses listed in documentation, and when a patient is admitted, that does not mean every diagnosis Reflection Questions for Charting listed is the reason for an admission or care. This is the same for substance use and mental health: Those two markers should not Grab that cup of coffee, take a breath. As you sit at the computer, deter from the possibility that patients can have a wide range of ready to document your encounter with a patient, here are some health concerns outside of those two factors. Information about helpful questions to check in with yourself: mental health and substance use can be useful with treatment plans and preventative care, but it may not necessarily need to be ● What am I trying to say about this patient? the guiding clinical markers of care. This inappropriate guidance ● of care can impact clinical documentation, drifting away from Is this information relevant to what the patient is being why a patient is seeking care in the first place. admitted and seen for? ● Why am I saying it like this? What are my intentions here It is vital that we increase our awareness around language in at a clinical level? medical documentation, for just as the impact of any diagnosis ● Is this symptom or behavior I am documenting based is great, so too is the impact of bias surrounding that diagnosis on evidence given by the patient or personal witness, or and how we respond. Instead of criminalizing the patient and could it be feelings I am having? personalizing the individual to the concern, we remember that people have problems, but are not the problems. Oftentimes, ● Is there any piece of information I am leaving out? Why a patient may exhibit behaviors as a symptom of what they might I be leaving it out? What is coming up for me? are being seen for. Recognizing this, it is our responsibility as healthcare workers to acknowledge our bias in order to prevent perpetuated mistreatment and misunderstanding of a patient, Empowering Patients through Electronic Health regardless of their behaviors. Make no mistake, you can hold a Record Access gentle awareness to your bias and also adhere to appropriate boundaries when it is necessary and request staff support when Further data has shown that giving patients education and elec- safety is being questioned. tronic health record (EHR) access—where they can review chart notes from their providers and the listed diagnosis—helps ensure When healthcare workers can see a person before the diagno- ongoing consistency of care and adequate treatment. An open sis, it assists in adhering to treating patients and not providing line of access for a patient to their health information not only

Patients can have multiple diagnoses listed in documentation, and when a patient is admitted, that does not mean every diagnosis listed is the reason for an admission or care. This is the same for substance use and mental health: Those two markers “should not deter from the possibility that patients can have a wide range of health concerns outside of those two factors.

Patient Safety I Vol. 3 No. 1 I March 2021 I 77 prevents misinformation from being shared with medical teams, Professionals Towards Patients With Substance Use Disorders but also is a substantial and sometimes underrated opportunity and Its Consequences for Healthcare Delivery: Systematic for patients to be engaged in their care, their responsibility to Review, Drug Alcohol Depend, Volume 131, Issues 1–2, themselves, and their health, and an opportunity for them to 2013,https://doi.org/10.1016/j.drugalcdep.2013.02.018. continue asking questions and giving information as it pertains 3. Words Matter: Terms to Use and Avoid When Talking About to their diagnosis and care plan. If a patient is unaware of lab Addiction. NIDAMED https://www.drugabuse.gov/sites/default/ results, diagnoses, or a treatment plan they may be unsure of files/nidamed_words_matter.pdf what to look for. 4. Jensen ME, Pease EA, Lambert K, et al. Championing Encouraging patients to get connected to your hospital system’s Person-First Language: A Call to Psychiatric Mental Health patient record through a computer portal system is vital in main- Nurses. J Am Psychiatr Nurses Assoc. 2013;19(3):146-151. taining patient-care team communication and to allow for open doi:10.1177/1078390313489729 dialogue around diagnosis, treatment plans, lab work, and med- 5. Rose D, Thornicroft G, Pinfold V, Kassam A. 250 Labels Used ications prescribed. to Stigmatise People With Mental Illness. BMC Health Serv Res. We know that words are powerful, and we know that information 2007;7:97. Published 2007 Jun 28. doi:10.1186/1472-6963-7-97 matters. How we give and filter that information is up to those 6. John F. Kelly PhD, Richard Saitz MD & Sarah Wakeman MD who are holding it and sharing it on behalf of the patient for the (2016) Language, Substance Use Disorders, and Policy: The sake of their well-being and safety. Using information to open Need to Reach Consensus on an “Addiction-ary”, Alcohol Treat Q, access rather than limit access to healthcare depends on how we 34:1, 116-123, DOI: 10.1080/07347324.2016.1113103 choose to document diagnoses, histories, behaviors, and results. 7. Kelly, J. F., Dow, S. J., & Westerhoff, C. (2010). Does our As healthcare workers, you have an opportunity to build bridges Choice of Substance-Related Terms Influence Perceptions instead of roadblocks just by the words you use. We see results of Treatment Need? An Empirical Investigation With Two when we are in tune with how our actions and bias can come into Commonly Used Terms. J Drug Issues, 40(4), 805–818. play with documentation and interpreting that documentation. doi:10.1177/002204261004000403 This is your chance—don’t miss it. Make your words count. 8. Ashford RD, Brown AM, McDaniel J, Curtis B. Biased Labels: The Johns Hopkins study mentioned earlier also revealed implica- An Experimental Study of Language and Stigma Among tions that bias in a chart can be maintained throughout the life of Individuals in Recovery and Health Professionals. Subst Use a patient, which is frankly very concerning. In a patient’s lifetime, Misuse. 2019;54(8):1376-1384. doi:10.1080/10826084.2019.15812 every visit, check-up, lab work, procedure, and admission can be 21 impacted by the language that each staff member uses in docu- 9. Verheij RA, Curcin V, Delaney BC, McGilchrist MM. Possible mentation. The sense of permanency in medical documentation Sources of Bias in Primary Care Electronic Health Record Data and the impact of documentation on patient care mandates the Use and Reuse. J Med Internet Res. 2018;20(5):e185. Published healthcare community to take documentation seriously. 2018 May 29. doi:10.2196/jmir.9134 Overall, if there is less patient engagement and less empathy 10. Leonieke C. van Boekel, Evelien P.M. Brouwers, Jaap because of stigma and misinterpretation through derogatory van Weeghel, Henk F.L. Garretsen, Stigma Among Health language, this is a problem of patient safety, period. This is a Professionals Towards Patients With Substance Use Disorders great opportunity for the medical community to build awareness and Its Consequences for Healthcare Delivery: Systematic around systemic gaps and the impacts of documentation—both the Review, Drug Alcohol Depend, Volume 131, Issues 1–2, impact of documentation on bias and of bias on patient treatment. 2013, (http://www.sciencedirect.com/science/article/pii/ S0376871613000677) Any room for misinterpretation of information needs to be evaluated. Further Resources Any room for assumptions made due to stigmatizing language needs to be rewritten. Webcasts: “The Power of Language and Portrayals: What We Hear, What We See” https://www.samhsa.gov/power-language-portrayals “Speech has power. Words do not fade. What starts out as a Mental Health Guide to Reporting sound, ends in a deed.” -Abraham Joshua Herschel https://www.cartercenter.org/resources/pdfs/health/mental_ health/2015-journalism-resource-guide-on-behavioral-health.pdf References About the Author 1. P Goddu A, O’Conor KJ, Lanzkron S, et al. Do Words Matter? Stigmatizing Language and the Transmission of Bias in the Bethany Davis ([email protected]) is a Medical Record [published correction appears in J Gen Intern social worker and has a private practice in Seattle, WA. She enjoys Med. 2019 Jan;34(1):164]. J Gen Intern Med. 2018;33(5):685-691. writing and going on walks with her niece and nephew. doi:10.1007/s11606-017-4289-2 2. Leonieke C. van Boekel, Evelien P.M. Brouwers, Jaap This article is published under the Creative Commons Attribution- van Weeghel, Henk F.L. Garretsen, Stigma Among Health NonCommercial license.

78 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 By Lea Anne Gardner, PhD, RN◆◊ & Susan Wallace, MPH◊ DOI: 10.33940/data/2021.3.8

Corresponding author Patient Safety Authority Disclosure: The authors declare that they have no relevant or material financial interests.

PatientPatient SafetySafety II Vol. 3 No. 1 II March 2021 II 79 asogastric Methods The Pennsylvania Patient Safety Reporting System and orogastric (PA-PSRS)‡ was queried using a three-step iterative pro- tubes, herein cess to search for reports of NGT events that occurred collectively between January 1, 2017, and June 30, 2020. ● The initial query used keywords from a prior referred to as nasoga- 2 N NGT feeding data snapshot. Manual review of stric tubes (NGT), are randomly selected reports from the initial query inserted into a patient’s refined keywords and exclusion criteria. ● A second query was performed using the nasal or oral cavity to refined set of keywords# and exclusion criteria.¶ administer feedings or ● A manual review was performed to identify the final dataset. Reports identifying NGT dislodged, medications or remove pulled out, or present were excluded. stomach contents.1 Tube misplacement is Results The initial data query identified 2,216 reports. This num- a known complication ber was reduced to 1,346 repots after rerunning the data that can occur during query with the refined keywords and exclusion criteria. 2,3 We identified 197 reports that met inclusion criteria; 196 insertion. This NGT were nasogastric and 1 was orogastric. Figure 1 shows misplacement data the number of reports by quarter. snapshot provides Acute care hospitals submitted 82.7% (163 of 197) of the reports, followed by children’s hospitals at 11.7% (23 of updated information. 197) and long-term acute care facilities at 5.6% (11 of 197).

Figure 2 displays patient age for these reports. Keywords: patient safety, enteral feeding tubes, nasogastric tubes, Location of NGT misplacements by function were iden- pneumothorax tified. SeeFigure 3.

Given the severity of misplaced NGTs in the lung noted in Figure 3, an analysis of these reports was performed. Of the 139 lung reports, 62.6% (87) provided sufficient information about iatrogenic injuries or symptoms and interventions. In 96.6% (84 of 87) of these reports, pneumothoraces, i.e., air leaking into the space between the lung and chest wall, possibly causing the lung to col- lapse, were noted and most of these reports identified treatments. See Figure 4.

‡PA-PSRS is a secure, web-based system through which Pennsylvania hospitals, ambulatory surgical facilities, abortion facilities, and birthing centers submit reports of patient safety–related incidents and serious events in accordance with mandatory reporting laws outlines in the Medical Care Availability and Reduction of Error (MCARE) Act (Act 13 of 2002). All reports submitted through PA-PSRS are confidential and no information about individual facilities or providers is made public. #The keywords: “Dobhoff,” “DHT,” “duotube,” “Covidien,” “Keofeed,” “Keofed,” “Corflo,” “Cortrak,” “Corpak,” “Salem Sump,” “NG tube” and “naso*,” “NGT,” “ND tube,” “enteral,” “small bore,” “feeding tube,” and “orogastric tube” were combined with a second set of keywords: “position,” “placement,” “pneumo,” “lung,” “esoph*,” “bronch*” “thorax,” “KUB” (kidney, urinary, bladder X-ray), “AXR” (abdominal X-ray), “chest X-ray,” “malposition,” “migrated,” “misplaced,” “replace,” “pleural,” “perforation,” and “perforated.” The asterisk represents a “wildcard.” This syntax searches for words with the letters and any ending. ¶Exclusion criteria encompassed the following PA-PSRS event categories: skin integrity; falls; transfusions; adverse drug reactions; patient self-harm; radiology, such as wrong patient, not completed, or not ordered; laboratory; surgical, such as consent missing, foreign body, procedures cancelled; respiratory care; complications of procedures/tests/treatments such as unplanned transfer to the operating room, stroke; maternal; emergency department; and healthcare-associated infections.

80 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Figure 1. PA-PSRS Misplaced Nasogastric Tube (NGT) Reports, n=197

Number of Identified NGT Reports 25

21 20 20 19 18 17 16 16 15 13 12 12 11 11 10 8

5 3

0 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 Q3 Q4 Q1 Q2 2017 2018 2019 2020

Year and Quarter

Figure 2. PA-PSRS Patient Age Distribution of Misplaced Nasogastric Tubes, n=97

Number of Identified NGT Reports 60

51 50 46

30 27

21 20 15 13 11 10 9 4 0 0 <1 1-19 20-29 30-39 40-49 50-59 60-69 70-79 80-89 90-99

Patient Age (Years) This data snapshot shows the most prevalent issue with About the Authors misplaced NGT is pneumothorax, with the majority of cases treated by inserting a chest tube to remove air from the Lea Anne Gardner ([email protected]) is a patient safety chest cavity. analyst with the Patient Safety Authority. She has more than 30 years of research experience in patient safety, performance improvement, cardiology, and behavioral health, and References a diversified work experience, including roles as a national director of quality improvement, administrative director 1. Price G, Shuss SS. The Ins and Outs of NG Tubes. Nurs of population health and an institutional review board, a Made Incred Easy. 2016;14(5):52-4. coronary intensive care nurse, a clinical research nurse, 2. Wallace S. Data Snapshot: Complications Linked to and a nurse research/leadership preceptor. Gardner is also Iatrogenic Enteral Feeding Tube Misplacements. PA a Certified Professional in Patient Safety (CPPS). Patient Saf Advis. 2017;14(4):1-5. Susan Wallace is a senior patient safety liaison with the 3. Wallace S. Training Suggested When Changing Patient Safety Authority (PSA) and a core team lead for the Brands of Enteral Feeding Tubes. PA Patient Saf Advis. PSA’s Center of Excellence for Improving Diagnosis. She is 2014;11(2):78-81. a Certified Professional in Healthcare Risk Management (CPHRM), and a Certified Professional in Patient Safety (CPPS).

Figure 3. PA-PSRS Misplaced Nasogastric Tubes by Location and Function, n = 197

Number of Identified NGT Reports

120 Enteral Feeding 111 Indication of Use Not Specified Gastric Decompression 100

21 20

9 9 7 7 6 6 4 3 3 3 1 1 0 1 1 0 2 0 0 2 0 0 0 Lung Esophagus Coiled in Stomach Duodenum Chest & Nose Location Not Mouth/ Abdominal Identified Throat Cavity

Location of Misplaced NGTs

82 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Figure 4. PA-PSRS Misplaced Nasogastric Tube Pneumothoraces and Treatments, n=84* § †

Tube Size Treatment and Purpose

Chest Tube Small Bore/Enteral Feeding Tube

Monitoring/ Noninvasive Treatment^ Large Bore/ Gastric Decompression No Treatment Needed

Not Specified Unable to Determine

*Gastric decompression indicates empty stomach contents. §Eighteen reports of patients diagnosed with a pneumothorax had respiratory and/or cardiac symptoms. There were 3 reports of intu- bations and 1 report of a patient transferred to an intensive care unit. While the number of patients transferred to the intensive care unit is low, 70.2% (59 of 84) of reports were from patients in an intensive care unit. ^Ongoing monitoring refers to chest X-rays taken to evaluate the size of the pneumothorax. Noninvasive treatment refers to additional oxygen. †This diagram shows color-coded lines, i.e., links, between the size/purpose of the nasogastric tube and the type of treatment for a pneumothorax. The thickness of the lines is proportional to the number of reports underlying that specific relationship. The labeled boxes, when possible, are intended to show the proportionality of the reports. Regarding the four labeled boxes with height greater than the aggregate thickness of the adjacent lines, the height was extended for readability and therefore does not reflect the number of reports.

This article is published under the Creative Commons Attribution- NonCommercial license.

Patient Safety I Vol. 3 No. 1 I March 2021 I 83 84 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 Hanging

by a

Thread◊ by Caitlyn Allen, MPH DOI: 10.33940/interview/2021.3.9

Healthcare in the United States: Who’s really pulling the strings? Wendell Potter, former health insurance executive turned whistleblower, shares how the American people are being misled, signs a news story is just well-concealed propaganda, and how recognizing it as such is vital to making healthcare safer.

Caitlyn Allen: You came into the public eye in 2010 with the release The catalyst for wanting to leave the insurance industry was an of your book, Deadly Spin, which was a behind-the-scenes exposé incident that occurred in 2007 when you witnessed a makeshift on the influence that insurance companies have on healthcare in healthcare clinic at a fairgrounds in Virginia. What happened? the United States. Can you tell me more about that? In July 2007, I had gone back home to Tennessee to visit family, I Wendell Potter: Throughout my career, I remember just saying grew up there. And while I was there, I picked up my hometown to myself, I could write a book, but it never occurred to me that newspaper: the Kingsport Times-News. I read on the front page I would actually be able to do that. And it was not until after I about something that they were calling a “healthcare expedition” testified before Congress, and that was when I really came on the that was being held a few miles away. It was at a county fairground scene as a critic of the U.S. system of health insurance, that I had in a small community called Wise in Wise County, Virginia. The that opportunity. I testified before a Senate hearing in June 2009. article said that people were expected to drive from hundreds of And I had several invitations for temp television interviews after miles away to get care that they otherwise couldn’t afford. I never that, including the Bill Moyers Journal, which was seen by a few heard of this event or the organization that was putting it on, but agents in New York. this was the ninth year. And the article said people drove as far And I started getting calls from agents in New York, asking if I away as Ohio, South Carolina, and Georgia to this little hamlet in would be interested in writing a book. And that was what led to the coal mining section of southwest Virginia. Deadly Spin. I decided that this was an opportunity to go beyond I went there out of curiosity. At the time, I was writing a white that Senate testimony to pull the curtains back on how health paper for my company that would be used by the industry to try insurance companies in the United States really operate and also to get people to believe certain things about the uninsured in this how they spend enormous sums of money to influence public country. And the whole objective, as it turned out, was to try to get opinion and public policy and elections. people to think that being uninsured wasn’t such a big deal, and people in this country were largely uninsured by choice. I drove

◊Patient Safety Authority Disclosure: The author declares that they have no relevant or material financial interests. my dad’s car up to the fairground. And when I walked through precisely what I was doing in my job and being paid significantly the fairground gates, I was just stunned. It was almost like I had more than I was being paid as a journalist. And I had to own up to walked out of the U.S. into some refugee camp or MASH [Mobile the fact that I was a corporate propagandist and was misleading Army Surgical Hospital] unit in some war-torn country because the American public to protect profits. people were there by the hundreds. They were queued up in lines that stretched out of sight. And I noticed that some of those lines Propaganda is so effective because people don’t know that led to barns and animal stalls. they’re being influenced. As you’ve been watching the news, This was the county fairground, and people were really getting what are some of the more pernicious talking points regarding care in barns. Volunteers had cleaned up the animal stalls, put cur- healthcare coverage that you’ve heard? tains over them to provide some means of privacy, but that’s what The talking points are the same as they always have been. I witnessed. And it just shook me to my core. I couldn’t imagine Sometimes a few details will be changed, but typically, the talking that this was happening in my country and just a few miles from points, the soundbites, are the same. You want people to fear where I grew up. I knew immediately that people in those lines change. And our communications campaigns, our propaganda could be people I grew up with. It could have been neighbors, campaigns, were designed to instill fear, uncertainty, and doubt in even relatives of mine. And I realized I could have been one of the minds of Americans to get them to fear change; to be uncertain them had I not been fortunate, had I not been lucky to get a good that what’s being proposed in the way of reform would maybe not education and, ultimately, a good job. be in their best interest; and to doubt that those who are proposing I also knew that I had to take some responsibility for what I was it are on their side. Frankly, they’re fearmongering campaigns, seeing because my job in the industry at Cigna was to perpetuate and they’re very effective. They appeal to people’s emotions. And our current system, as dysfunctional as it is. And, of course, I was the talking points, there aren’t terribly many of them, but you expected to obscure the dysfunctionality of our healthcare system want people to believe that what’s being proposed is “socialized and try to get people to think that health insurance companies are medicine,” that if this reform is passed and enacted that we’ll be essential, that we have the best healthcare system in the world. I going down the slippery slope towards socialism. had all the evidence right in front of me that that was obviously not And we—my peers and I across the industry—knew from research the case. I realized when I was looking at those people—who, by that we had done working with linguists and other communica- the way, many of them were soaking wet because it had rained that tions experts and from polling, that people in this country just morning and they weren’t about to lose their place in line, so they have a fear and a resistance to anything that they are told might be were standing in the rain—I realized that what I was doing for a living socialism. It’s one that works, even though what’s being proposed was the exact opposite of what I tried to do when I was a journalist. is nowhere near socialism. You can get people to believe that just I made a commitment that day that I would have to figure out some by saying it and getting people who folks trust to say it, whether way other than that, other than what I was doing, to earn a living. they’re politicians or preachers or community leaders. It’s a very I was a newspaper reporter in my first career in Tennessee, in potent thing that’s used. Another is that what’s being proposed is Nashville and Memphis, and then later in Washington, where I a “government takeover of healthcare,” or what’s being proposed was lucky to be able to cover the White House and the Congress is government-run healthcare, regardless of what the facts are. and Supreme Court for Scripps-Howard newspapers. I had a That’s what you want people to believe. And the industry has been longer career in corporate public relations. But, as a reporter, I extraordinarily successful. always tried to be honest, as objective as I could be, never to know- The industry works through front groups. The current one is ingly mislead people or to leave out pertinent details. But that’s called the Partnership for America’s Health Care Future. They’re always operated out of a PR [public relations] firm, typically a PR firm in Washington, that churns out these talking points and misleading statements and press releases and advertisements. It’s an extraordinarily effective campaign. But, again, you don’t have to have a lot of original thinking, you just have to repeat those talking points, those soundbites, over and over and over again. And the The talking points are the industry has been doing it for decades, and it’s had its effect.

same as they always have I remember during the Affordable Care Act protests people “ holding up signs instructing the government to “keep [their] been. Sometimes a few hands off my Medicare.” It’s true. I spoke at a town hall in New Jersey in September 2009. details are different, but It was during the debate on what became the Affordable Care typically, the soundbites Act. It was a very contentious debate, and I saw people who were opposed to it with signs, “Get your government hands off my are the same. You want Medicare.” It was just unbelievable how people are so easily misled. We’re all more gullible than we think. people to fear change. Absolutely. Are there common red flags that we can look for to help discern whether a news story is based in fact, or if it’s likely well-disguised propaganda?

86 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 If you don’t see or hear the source of the information, be skeptical. If an organization is mentioned and you’ve never heard of it, Be Your Own Advocate: be skeptical. Even if you’ve heard of it, still be skeptical because a lot of these organizations, like the Partnership for America’s How to Tell if a News Story is Real— Health Care Future, people may have heard about that, but they or Just Designed to Manipulate You haven’t taken the time to understand that’s a front group, that it is funded by industry money. Actually, by our money. Our money that we pay in premiums when we get healthcare. Part of our money goes to finance these campaigns. And, again, even if it’s someone that you recognize, step back and think, “What is the 1. Check to see if the motive here? What does this person have to gain from this?” It information has a source.1,2 takes some research. It’s not easy. But, certainly, the red flags are Anyone can write “Studies terms like “socialized medicine,” “government-run healthcare,” “government takeover” of healthcare. Anything that, as you read show...” without citing who did them, you can tell that your emotions are being manipulated, the research. that’s really important.

One of the monumental shifts from the Affordable Care Act was enacting protections for people with preexisting conditions. 2. If the source of the For those who may not remember, can you remind us what care was like before that was enacted and how it’s since changed? information is listed, It’s a good question because a lot of people either don’t know, research it! Look for: don’t remember, or have forgotten what it was like, but insurance companies had the power to declare people uninsurable. A lot of • Funding: Can you tell how/ those folks in those long lines in Wise County, Virginia, were in where the organization gets that boat. They had preexisting conditions, and insurance com- its money? panies had the power, which they exercised all the time, to reject someone’s application for coverage because of some illness in the • History: How long has the group been around? past, some condition that they had. And if they were willing to sell • URL: Unusual domain names typically indicate coverage to that person, they would be able to charge them a lot 1 more money than someone who didn’t have that condition. And untruthful or intentionally misleading news. it was a big contributing factor to the rising number of uninsured in this country. By the way, that was a fact that I was expected to obscure in that white paper I mentioned. I was certainly not mentioning that a 3. Look for other sources to big reason why so many Americans were uninsured was because they either couldn’t get or couldn’t afford coverage because of a verify a story and consider preexisting condition. And that, frankly, applies to most of us in using fact-checking websites this country or ultimately will. We had, by the time the Affordable like Snopes or PolitiFact.2 Care Act was passed, almost 50 million people in this country who didn’t have health insurance. And a lot of them who did were paying far more than they should just because they’d been sick in the past or had a condition. 4. If you feel true anger after The Affordable Care Act changed that. Insurance companies can no longer blackball people because of a preexisting condition. reading a story, it may have And there were hundreds [of conditions] that they used in the been written specifically to past to either deny coverage or to charge people more. They can make you feel that way and no longer do that. And they can’t charge people more for coverage because of a preexisting condition. It was a horrible, horrible may have used misleading or situation. An untold number of people died because they couldn’t false information.1 Try to find get the care they needed, because they couldn’t get insurance. other sources that discuss the topic. Playing devil’s advocate for a moment: What are the benefits to patients under our current system? For instance, does a private system spurn more innovation or provide more choices than a more regulated system? We do have some of the best doctors and some of the best hospitals Sources: and medical facilities in the world, but they’re off limits to so many people, even with insurance, because of how insurance compa- 1. https://docs.google.com/document/d/10eA5-mCZLSS4MQY5Q Gb5ewC3VAL6pLkT53V_81ZyitM/edit nies create discrete networks of providers. That has been one of 2. https://libguides.ucmerced.edu/news/fact-check

Patient Safety I Vol. 3 No. 1 I March 2021 I 87 Everything I’ve mentioned here is uniquely American. You don’t have these kinds“ of nightmare scenarios for so many people in other countries. They just don’t exist.

the “innovations” in our system. Even if people have insurance, Regarding these “innovations” you mentioned, one that seems there are several barriers that insurance companies have erected most troubling is that you could go to the emergency room of a to make it more difficult and more expensive for people to get the hospital that’s in-network, but you only find out later that your care that they really need. One “innovation” is the use of prior ER doctor was out-of-network. authorization. In other words, doctors have to ask permission. They Yes. That’s a uniquely American dilemma. Everything I’ve men- need authorization from an insurance company before proceeding tioned here is uniquely American. You don’t have this kind of with a treatment, in many cases, even prescribing a medication. nightmare scenario for so many people in other countries. They And these prior authorization requirements, frankly, have even just don’t exist. There’s no need for them. Americans grew up gotten more aggressive since the Affordable Care Act was passed. with this system, so we can easily be misled into thinking that Another “innovation” is high-deductible plans. We had them before it’s worse somewhere else, but we have the most expensive, the the Affordable Care Act was passed, but they weren’t as prevalent most unfair, inequitable healthcare system on the planet. When as they are now. And now, a very high percentage of Americans are you talk about innovation ... obviously, there have been innova- in plans with such high deductibles that they’re foregoing the care tions—what I would refer to as the provider side—on treatments that they need because they don’t have enough money to meet their and development of new medications, new prescription drugs. deductible, so they’re not picking up their prescriptions. They’re But, again, a lot of those innovations, which are valuable, are off not going to the doctor when they need to, which of course has limits to millions and millions of Americans because they can’t repercussions on their health, both short- and long-term. afford it, or the facility is not in-network or the insurance company Another so-called “innovation” from the insurance industry are won’t approve a prescription for a medication your doctor says you these narrow networks that I referenced. Every health plan has need. So, yes, there’s innovation, but only, in many cases, for a few. its own set of doctors and hospitals that you can use. And if you go out of network, you likely will get little if any coverage for And it’s not like we have the health outcomes to support hav- seeing that provider or going to that facility. So, yes, we’ve had ing such an expensive, inequitable system. Americans perform some innovation, but in my view, on the insurance side of it, the worse in almost every outcome you can think of. So what are we innovations have been detrimental. getting for all this money that we’re paying? Now, you mentioned the word “choice,” and some people will You’re exactly right. The Commonwealth Fund assesses the per- point to that and say, “Well, Americans have more choice of health formance of health systems. They’ve done this for a long time. I plans,” for example. That’s not the choice that Americans really think they look at about 11 countries. And almost every metric value. What they value most is a choice of doctors and hospitals. that they look at, we’re at or near the bottom. And we are bewildered by the choices that we often have of health insurance companies. If you buy an insurance policy through the Do you think healthcare itself would be safer if more people exchanges in this country, it’s hard to figure out which one is the had access to it? best deal for you. And almost all of them have high deductibles. I absolutely think healthcare would be safer. There’s a lot more So, yes, there has been innovation. Another thing about choice, that needs to be done beyond getting people the access to care by the way, is that the opponents of reform often use that word. that they need. Comprehensive reform could certainly address I wrote a New York Times op-ed about that earlier this year. They patient safety in ways that we have not. say that we want to have choice and that reform will take choice We don’t have a national system. Health insurance, and healthcare away from us. But when you stop and think about it, we—partic- delivery, is largely regulated at the state level, so it varies from ularly those of us who get coverage through our employers—don’t state to state. But back to safety; one way of looking at it is that we have the choice of health insurance carrier. Our employer makes don’t really have an adequate safety net in this country. GoFundMe that decision. We might have two or three options from the same and these pop-up medical clinics are not a real safety net. And insurance company, but more often than not, they all have high many people in many states are not eligible for Medicaid, because deductibles. And it is just a matter of guessing and gambling on some states have not expanded the Medicaid program. We have a how healthy we are going to be over the next year. lot of work to do improving the so-called safety net in this country.

88 I PatientSafetyJ.com I Vol. 3 No. 1 I March 2021 What advice would you offer to a patient who is having difficulty of people didn’t even have that for their care this year. I’m sure with their insurance provider? For instance, if somebody has it’ll ramp back up. When they’re able to have these expeditions been denied a procedure or a treatment that their physicians again, I’m sure that the numbers will be even higher than they recommended, what should they do? have been. But, I would encourage executives to step out of their You should appeal the denial right off the bat. Work with your environment, to really talk to people, even their own employees. physician and encourage or ask your physician to appeal that They might be astonished to understand that even employees of denial. Sometimes that could be done as a matter of routine, but insurance companies, those at the low end of the pay scale, often you need to make sure that your doctor is appealing a denial. don’t have insurance. They can’t afford to take up the offer. Yes, a In many cases, the denials are overturned. And in many cases, lot of people who work for insurance companies are uninsured. insurance companies just automatically routinely deny requests They’re very low pay. It’s scandalous and outrageous. because they know that a lot of people just won’t bother to seek an appeal. Appeal that. It’s a legal right that people have. There’s a Back to the squeaky wheel idea, if the executives aren’t going process that can be lengthy, but it has to be expedited to a certain to come to the people, then maybe we can encourage more of extent because, in many cases, people need treatment right away. the people to come to the executives. Don’t waste time. Be your own advocate, but also try to find some- I think that’s a good idea. Interestingly, I had a Twitter thread a one else to help. Clearly, if you’re sick and you’re needing care, few months ago, and someone responded to one of my tweets with your ability to be your own advocate is impaired, so it’s important the photograph and the address of the CEO of UnitedHealthcare, to try to find someone else, a family member or someone else, which is the largest health insurance company in the country. And who can help advocate for you to help navigate the system and I got a call right away from the head of communications at United be a pain in the butt. Call the insurance companies. Call your asking if I could somehow remove that comment. I couldn’t, but doctor. Just make sure that you’re giving it a good old try and that got their attention. So, yes. And someone had also put some even reach out to your state legislator, your member of Congress, placards in the lawn around the corporate headquarters as well, your senator, the media. I know from years of experience in the too. That is something that is important for advocates to do for insurance industry that when that happens, your case gets ele- people to understand. You can get the attention of executives if vated within the insurance company. It becomes what they refer you do things like that. That’s what they care about. to as a “high-profile case.” And that gets special consideration. But my ultimate advice is to be a squeaky wheel. Don’t give up. With everything that we’ve just discussed, are you optimistic or Don’t assume that a denial is the final decision. pessimistic about the future of healthcare in the United States? I’m optimistic. If I weren’t, I don’t think I would still be doing Maybe if there are enough squeaky wheels, then maybe that’ll what I’m doing. Who knows? I could be tilting at windmills, but be enough to drive some change. I know it’s possible to change the system. One of the organiza- We need a lot more squeaky wheels. We need people to advocate tions I lead is called the Center for Health and Democracy. And not only for themselves but for a better system. We need to be the word “democracy” is there intentionally. And here’s why: really outraged at what’s going on. And that’s one of the things We need to reduce the power and influence of my old industry I’m trying to do is to get people to understand this system is so and other big entities, for their big corporations and their trade broken. We should be outraged. groups have enormous power to influence how we think, how our public officials are elected. They influence elections in ways that There’s a divide between health insurance executives and every- we don’t really understand, and they influence public policy. We day folks. They’re not seeing the impact their decisions are hav- need to attack that. Money and politics in this country is a big, ing on real people. Would that help? And if so, realistically, how big problem. could we change that? Anyhow, the point is I am optimistic. I think it’s going to be a I think it would help. The reason I changed my life and my career challenge because of the power and influence of the industry’s was allowing myself to be in a position where I saw something I best interests. We need to curb that power in many different ways, otherwise wouldn’t see. Most executives don’t do that. And they and we need to keep advocating for reform. steer far away from an experience like that. I often talk about my There is a lot more support for sweeping reform than a lot of folks former CEO who had such privilege, so many perks of the job. A might realize, so that gives me a lot of hope. driver would pick him up in the morning, bring him to the building. He would go to his office in a private elevator. If he needed to Wendell Potter is a former health insurance executive who had a “crisis go out of town, and I traveled with him quite a bit, the driver would of conscience,” becoming disgusted with America’s broken healthcare take us to the airport. We would take a private plane to wherever system after years of helping the industry mislead Americans. Potter we were going. When you live a life like that, you can be so far walked away from his job at Cigna, one of the country’s largest health removed from the way most Americans live. And you don’t have an insurers, in 2008 and became a vocal critic of the insurance industry understanding really of the consequences of the actions that you and a leading advocate of reform. and your fellow executives make, the decisions that they make.

I don’t know how you can change that. I wish that they would go *The views and opinions expressed here are those of the participants and may not to Wise County, Virginia, to see, to this day, how people line up reflect the opinions or positions of the Patient Safety Authority or the Commonwealth by the hundreds to get care in barns and animals stalls. That has of Pennsylvania. not ended. The Affordable Care Act did not end that. The pandemic put a pause on it. They couldn’t do that this year, so a lot This article is published under the Creative Commons Attribution- NonCommercial license. Send us your manuscripts!

Some journals get more than 16,000 submissions every year—thousands more than any publication could ever print. That means good papers won’t get published, just because there aren’t enough slots. In fact, most manuscripts probably won’t even get a fair read.

Our editors consider every submission carefully— so your manuscript will never be rejected because we didn’t have a chance to read it.

Submit today! patientsafetyj.com

333 Market Street - Lobby Level Harrisburg, PA 17101 patientsafetyj.com 717.346.0469