A 62-year-old man with a history of malnutrition-related encephalopathy was admitted for possible aspiration pneumonia complicated by empyema and coagulopathy. During the hospitalization, he was uncooperative and exhibited signs of delirium. For a variety of reasons, he spent two weeks in the hospital with minimal oral intake and without receiving most of his oral medications, putting him at risk for complications and adverse outcomes.
Pandya C, Clarke T, Scarsella E, et al. J Oncol Pract. 2019;15:e480-e489.
Care transitions and handoffs represent a vulnerable time for patients, as failure to communicate important clinical information may occur with the potential for harm. In this pre–post study, researchers found that implementation of an electronic health record tool designed to improve the handoff between oncology clinic and infusion nurses was associated with a reduction in medication errors, shorter average patient waiting time, and better communication between nurses.
Herlihy M, Harcourt K, Fossa A, et al. Obstet Gynecol. 2019;134:128-137.
Prior research has shown that when patients have access to clinicians' notes, they may identify relevant safety concerns. In this study, 9550 obstetrics and gynecology patients were provided with access to their outpatient visit documentation. Almost 70% of eligible patients read one or more notes during the study period, but only 3.2% shared feedback through 232 electronic reports. Of patients who provided feedback, 27% identified errors in the documentation; provider reviewers determined that 75% of these could impact care.
Transferred to the emergency department from the transfusion center after becoming unresponsive and hypotensive, an elderly man with signs of sepsis is given incomplete and delayed antimicrobial coverage due to a history of penicillin allergy. Neither gram-negative nor anaerobic coverage were provided until several hours later, and the patient developed septic shock.
Seen in the emergency department, a man with insulin-dependent diabetes mellitus had not taken insulin for 3 days. His blood glucose levels were in the 800s with an anion-gap acidosis and positive beta hydroxybutyrate. While awaiting an ICU bed for treatment of diabetic ketoacidosis, the patient received fluids, an insulin drip was started, and blood glucose levels were monitored hourly. When lab results showed he was improving, the team decided to convert his insulin drip to subcutaneous long-acting insulin.
Driver BE, Scharber SK, Fagerstrom ET, et al. J Emerg Med. 2019;56:109-113.
This pre–post study examined the effect of an electronic health record alert that required physicians to respond "yes" or "no" regarding whether tests were pending at the time of discharge from the emergency department. Investigators found that physician responses were often inaccurate, and the proportion of discharged patients with tests pending increased following the intervention, contrary to intentions.
To transfer a man with possible sepsis to a hospital with subspecialty and critical care, a physician was unaware of a formal protocol and called a colleague at the academic medical center. The colleague secured a bed, and the patient was sent over. However, neither clinical data nor the details of the patient's current condition were transmitted to the hospital's transfer center, and the receiving physician booked a general ward bed rather than an ICU bed. When the patient arrived, his mentation was altered and breathing was rapid.
The operating room environment harbors particular patient safety hazards. Drawing from 1137 perioperative medication error reports submitted over a 1-year period, this analysis found that more than half of the recorded incidents reached the patient and the majority of those stemmed from communication breakdowns during transitions or handoffs. The authors provide recommendations to reduce risks of error, including using barcode medication administration, standardizing handoff procedures, and stocking prefilled syringes.
Craynon R, Hager DR, Reed M, et al. Am J Health Syst Pharm. 2018;75:1486-1492.
Pharmacists are expanding their reach as stewards of medication safety into the front line of care. This project report describes the pilot testing of pharmacist involvement in development and review of medication orders in the discharge workflow. A substantive percentage of medication problems were prevented due to pharmacist engagement.
Gupta A, Harrod M, Quinn M, et al. Diagnosis (Berl). 2018;5:151-156.
This direct observation study of hospitalist teams on rounds and conducting follow-up work examined the interaction between systems problems and cognitive errors in diagnosis. Researchers found that information gaps related to electronic health records, challenges with handoffs, and time constraints all contributed to difficulties in diagnostic cognition. The authors suggest considering both systems and cognitive challenges to diagnosis in order to promote safety.
Dalal AK, Schaffer A, Gershanik EF, et al. J Gen Intern Med. 2018;33:1043-1051.
Incomplete follow-up of tests pending at hospital discharge is a persistent patient safety issue. This cluster-randomized trial used medical record review to assess whether an automated notification of test results to discharging hospitalist physicians and receiving primary care physicians improved follow-up compared with usual care. The intervention was focused on actionable test results, which constituted less than 10% of all pending tests. Even with the intervention, only 60% of tests deemed actionable had any documented follow-up in the medical record, and there was no significant difference compared to usual care. The authors conclude that automated clinician notification does not constitute a sufficient intervention to optimize management of tests pending at discharge. Previous WebM&M commentaries explored problems related to tests pending at discharge and how organizations can improve follow-up of abnormal test results.
Bastawrous S, Carney B. J Digit Imaging. 2017;30:309-313.
Inadequate test result management is known to contribute to missed and delayed diagnosis. This Veterans Affairs study found that 0.17% of radiologic studies were not evaluated by radiologists. The study team identified several technical and process problems that contributed to these unread studies. They were able to address the issues to ensure all studies were read.
Meyer AND, Murphy DR, Al-Mutairi A, et al. J Gen Intern Med. 2017;32.
Trigger tools facilitate identification of adverse events. In this retrospective medical record review study, investigators found that an automated trigger successfully identified delayed follow-up of laboratory thyroid testing among patients with hypothyroidism, with a positive predictive value of 60%. The authors suggest that this trigger approach could be used to detect and ameliorate follow-up delays in real time.
Lo H-Y, Mullan PC, Lye C, et al. BMJ Qual Improv Rep. 2016;5.
Patient handoffs are vulnerable to errors of omission. This quality improvement project designed and implemented a checklist as a way to standardize the process of pediatric handoffs. The program found the tool to be effective in uncovering problems and physicians felt the checklist supported situational awareness and patient safety.
Administered antibiotics in the emergency department and rushed to the operating room for emergent cesarean delivery, a pregnant woman was found to have an infection of the amniotic sac. After delivery, she was transferred to the hospital floor without a continuation order for antibiotics. Within 24 hours, the inpatient team realized she had developed septic shock.
Murphy DR, Meyer AND, Bhise V, et al. Chest. 2016;150:613-20.
Insufficient follow-up of test results is a known contributor to missed and delayed diagnosis. This observational study used a trigger tool to detect diagnostic delays related to chest imaging follow-up. Investigators used an automated algorithm to identify chest imaging cases that potentially had a follow-up delay. A clinician then reviewed the medical records for a random sample of cases identified by the trigger tool and a reference set of cases involving patients with abnormal test results but no delays. They found that the trigger tool had 99% sensitivity and 38% specificity in detecting delays in follow-up of abnormal chest imaging. The authors suggest that this trigger tool may help identify patients at risk for diagnostic delay following abnormal chest imaging. A WebM&M commentary discussed delayed follow-up of a diagnostic test.
A woman with a history of chronic obstructive pulmonary disease underwent hip surgery and experienced shortness of breath postoperatively. A chest radiograph showed a pneumothorax, but the radiologist was unable to locate the first call physician to page about this critical finding.
Yadav S, Kazanji N, C NK, et al. J Am Med Inform Assoc. 2017;24:140-144.
Compared to paper charts, electronic health records offer safety benefits for physician documentation including better availability and legibility. However, electronic documentation introduces new concerns, such as copy-and-paste practices (which can perpetuate errors) and lack of diagnostic reasoning in electronic notes. This study compared physical exam documentation in initial physician progress notes before and after implementation of an electronic health record. Investigators found more inaccuracies in electronic notes, but more errors of omission in paper charts. Trainee physicians' documentation had fewer errors in both paper and electronic formats. The authors recommend that hospitals discourage copied notes and encourage accurate documentation at the time of the patient encounter. The importance of the physical examination itself was discussed in a PSNet interview with Dr. Abraham Verghese.
Despite the potential for electronic health record (EHR) systems to improve access to patient data, unintended consequences have emerged that can hinder information seeking. To highlight how EHRs can detract from patient–physician relationships, this commentary reveals insights from a physician who failed to notice a patient's respiratory failure and distress due to over-reliance on the EHR.
Please select your preferred way to submit a case. Note that even if you have an account, you can still choose to submit a case as a guest. And if you do choose to submit as a logged-in user, your name will not be publicly associated with the case. Learn more information here.