Evan M. Benjamin

Errors in a busy emergency department.

STUDY OBJECTIVE We describe errors occurring in a busy ED. METHODS This is a prospective, observational study of reported errors at an academic emergency department (ED) with 100000 annual visits. Trained personnel interviewed all ED staff with direct patient contact, during and at the end of every shift, by using standardized data sheets. RESULTS One thousand nine hundred thirty-five ED patients registered during the 7-day study period in the summer of 2001. Four hundred error reports were generated, identifying 346 nonduplicative errors (18 per 100 registered patients; 95% confidence interval [CI] 15.9 to 20.0). Forty percent of errors were reported by nurses, 25% by providers, 19% by clerical staff, 13% by technicians and orderlies, and 3% multiple reporters. Errors reported for every 100 hours worked were similar for all groups (5.5; 95% CI 5.2 to 5.9). Errors were categorized as 22% diagnostic studies, 16% administrative procedures, 16% pharmacotherapy, 13% documentation, 12% communication, 11% environmental, and 9% other. Patients involved in errors were more likely to be older (P <.0001) and more likely to have higher visit level intensity (P <.0001) than registered ED patients. Ninety-eight percent of errors did not have a significant adverse outcome. Seven errors (0.36 per 100 registered patients; 95% CI 0.14 to 0.72) were associated with an adverse outcome. CONCLUSION Reported errors occurred in almost every aspect of emergency care. Ninety-eight percent of errors in the ED do not result in adverse outcomes. System changes need to be implemented to reduce ED errors.

Quality of care for patients hospitalized for acute exacerbations of chronic obstructive pulmonary disease.

Context Evidence-based guidelines for the care of patients with chronic obstructive pulmonary disease (COPD) recommend explicit criteria for appropriate management of the disease. These guidelines identify tests and treatments of uncertain benefit as well as those that are potentially harmful. The degree of adherence to these guidelines is unknown. Contribution By surveying 360 hospitals, these investigators found that use of ideal care varied from 10% at some hospitals to greater than 60% at others. Cautions Administrative data were used instead of medical chart review to determine adherence to guidelines. Implications Despite well-accepted criteria for care of acute exacerbations of COPD, guideline adherence remains poor. The Editors Chronic obstructive pulmonary disease (COPD) affects approximately 16 million adults, accounts for more than

Association of Corticosteroid Dose and Route of Administration With Risk of Treatment Failure in Acute Exacerbation of Chronic Obstructive Pulmonary Disease

18 billion in annual health care costs, and is the fourth leading cause of death in the United States (1, 2). In 2002, approximately 620000 persons were hospitalized for acute exacerbation of COPD, making this 1 of the 10 leading causes of hospitalization among U.S. adults (3). In 1987, the American Thoracic Society became the first organization to produce clinical practice guidelines for the management of COPD (4). The number of standards has grown steadily since then, and various national and international organizations now produce guidelines (5-11). The American College of Physicians and the American College of Chest Physicians have coproduced evidence-based guidelines recommending that patients with acute exacerbations of COPD undergo a diagnostic evaluation that includes chest radiography and arterial blood gas analysis, followed by treatment with supplemental oxygen; anticholinergic bronchodilators; short-acting 2-agonists; systemic corticosteroids; antibiotics; and, in some circumstances, noninvasive positive-pressure ventilation. These guidelines identify spirometry, mucolytic agents, sputum examinations, methylxanthine bronchodilators, and chest physiotherapy to be of uncertain or no benefit, with the latter 2 treatments being potentially harmful (5). While the attention of policymakers, regulatory agencies, and the federal government has focused on measuring and improving quality of care for patients with pneumonia (12-17), remarkably little is known about the management of patients with acute exacerbations of COPD. General information about the quality of care for patients with COPD is lacking, and it is unknown whether regional differences in treatment exist or whether there is a positive relationship between hospital volume and quality of care for patients with COPD (such relationships are well documented for some surgical procedures and medical conditions [18]). Similarly, it is unclear whether disparities that have been observed between sexes and among ethnic groups across a wide variety of conditions and treatment settings are found in the management of patients with acute exacerbations of COPD (19). Consequently, we evaluated the quality of care provided to patients hospitalized for acute exacerbations of COPD by measuring adherence to current guideline recommendations and examining the impact of hospital and patient characteristics on composite measures of performance. Methods Setting and Participants We conducted a retrospective cohort study using data from 360 hospitals that participate in Perspective (Premier Inc., Charlotte, North Carolina), a database developed for measuring quality and health care utilization. Participating hospitals represent all regions of the United States, are predominantly small to medium-sized nonteaching facilities, and serve mostly urban patient populations. In addition to the information available in the standard hospital discharge file, Perspective contains a date-stamped log of all billed items (including medications and laboratory, diagnostic, and therapeutic services) at the individual patient level. Patients were included in our analysis if they were 40 years of age and older, had a principal diagnosis of COPD or a principal diagnosis of respiratory failure paired with a secondary diagnosis of COPD, and were discharged between 1 January 2001 and 31 December 2001. International Classification of Diseases, Ninth Revision, Clinical Modification (ICD-9-CM) codes were used to assess diagnostic information. Patients were excluded if they had a secondary diagnosis of pneumonia. The All Patient RefinedDiagnosis-Related Group classification system, version 15.0 (3M Corp., Minneapolis, Minnesota), was used to exclude patients if they were assigned to a diagnosis-related group other than COPD or one consistent with a hospitalization for COPD, such as respiratory failure. The institutional review board at Baystate Medical Center approved the study, and the need for written informed consent was waived. Data Elements In addition to age, sex, and ethnicity, we recorded the presence of congestive heart failure, valvular heart disease, pulmonary circulation disorders, peripheral vascular disorders, hypertension, paralysis and other neurologic disorders, diabetes, hypothyroidism, renal failure, liver disease, chronic peptic ulcer disease, HIV and AIDS, lymphoma, metastatic cancer, solid tumor without metastasis, rheumatoid arthritis and collagen vascular diseases, coagulation deficiency, obesity, weight loss, fluid and electrolyte disorders, blood loss anemia, deficiency anemias, alcohol abuse, drug abuse, psychoses, and depression. On the basis of work by Elixhauser and colleagues (20), we assessed comorbid conditions using Healthcare Cost and Utilization Project Comorbidity Software, version 3.1 (Agency for Healthcare Research and Quality, Rockville, Maryland). We obtained data regarding in-hospital mortality; length of stay; and disease-specific, pulmonary-specific, and overall readmission rates at 14 and 30 days from the Perspective discharge file. In addition to information related to the admission, we noted each hospitals bed size, annual number of admissions for acute exacerbations of COPD, teaching status, geographic region, and whether the institution served an urban or rural population. Adherence to Guideline Recommendations Using guidelines produced collaboratively by the American College of Physicians and the American College of Chest Physicians (5), we developed a set of performance measures that could be used to evaluate quality of care. On the basis of recommendations contained in these guidelines, we categorized the following diagnostic evaluations and treatments as beneficial: chest radiography, arterial blood gas analysis, supplemental oxygen, inhaled anticholinergic bronchodilators, inhaled short-acting 2-agonists, systemic corticosteroids, antibiotics, and noninvasive positive-pressure ventilation. Antibiotic regimens were classified as providing broad- or narrow-spectrum coverage. Narrow-spectrum coverage was defined as treatment with first-generation penicillins, first-generation cephalosporins, macrolides, tetracyclines, sulfonamides, vancomycin, or clindamycin. Broad-spectrum coverage was defined as treatment with second- or later-generation cephalosporins, antistaphylococcal penicillins, aminopenicillins, antipseudomonal penicillins, fluoroquinolones, carbapenems, monobactams, aminoglycosides, aztreonam, or combination therapy that included 2 or more narrow-spectrum agents. In addition, we categorized the following management strategies as having uncertain or no benefit and possibly causing harm: sputum examinations, acute spirometry, mucolytic agents, chest physiotherapy, and methylxanthine bronchodilators. Acute spirometry was defined as spirometry performed before the day of discharge. Adherence to these measures was assessed by using a combination of pharmacy billing data and records of other diagnostic and therapeutic services rendered during the hospitalization. Statistical Analysis Summary statistics at both the patient and the hospital level were constructed by using frequencies and proportions for categorical data and by using means, standard deviations, medians, and interquartile ranges for continuous variables. By applying the Institute for Healthcare Improvements (21) concept of the bundle, a collection of processes needed to effectively care for patients with a particular condition, we classified patients as receiving recommended care if they received all of the following diagnostic tests and treatments: chest radiography, supplemental oxygen, bronchodilator therapy, systemic corticosteroid therapy, and antibiotic treatment. The authors of the American College of Physicians and American College of Chest Physicians guideline viewed arterial blood gas analysis as helpful; however, given insufficient evidence demonstrating a benefit, the guideline stopped short of giving this test a full recommendation. Consequently, arterial blood gas analysis was not included in the recommended care bundle. Patients were identified as receiving nonrecommended care if they received any of the following: sputum examination, acute spirometry, therapy with methylxanthine bronchodilator or mucolytic agents, or chest physiotherapy. Patients were considered to have received ideal care if they received all 5 recommended care elements and none of the nonrecommended ones. We examined the association of patient age (differentiating patients who were 40 to 64 years of age, 65 to 74 years of age, and 75 years of age), sex, and ethnicity with the provision of recommended and ideal care by using chi-square statistics. The MantelHaenszel chi-square test was used to adjust for hospital, and hospital-adjusted relative risks for receiving recommended and ideal care were computed. In addition, KruskalWallis analysis of variance was used to determine if hospital region, teaching status, and the annual number of patients admitted with COPD were associated with hospital rates of delivery of recommended and ideal care. All analyses wer

Choosing The Best Hospital: The Limitations Of Public Quality Reporting

CONTEXT Systemic corticosteroids are beneficial for patients hospitalized with acute exacerbation of chronic obstructive pulmonary disease (COPD); however, their optimal dose and route of administration are uncertain. OBJECTIVE To compare the outcomes of patients treated with low doses of steroids administered orally to those treated with higher doses administered intravenously. DESIGN, SETTING, AND PATIENTS A pharmacoepidemiological cohort study conducted at 414 US hospitals involving patients admitted with acute exacerbation of COPD in 2006 and 2007 to a non-intensive care setting and who received systemic corticosteroids during the first 2 hospital days. MAIN OUTCOME MEASURES A composite measure of treatment failure, defined as the initiation of mechanical ventilation after the second hospital day, inpatient mortality, or readmission for acute exacerbation of COPD within 30 days of discharge. Length of stay and hospital costs. RESULTS Of 79,985 patients, 73,765 (92%) were initially treated with intravenous steroids, whereas 6220 (8%) received oral treatment. We found that 1.4% (95% confidence interval [CI], 1.3%-1.5%) of the intravenously and 1.0% (95% CI, 0.7%-1.2%) of the orally treated patients died during hospitalization, whereas 10.9% (95% CI, 10.7%-11.1%) of the intravenously and 10.3% (95% CI, 9.5%-11.0%) of the orally treated patients experienced the composite outcome. After multivariable adjustment, including the propensity for oral treatment, the risk of treatment failure among patients treated orally was not worse than for those treated intravenously (odds ratio [OR], 0.93; 95% CI, 0.84-1.02). In a propensity-matched analysis, the risk of treatment failure was significantly lower among orally treated patients (OR, 0.84; 95% CI, 0.75-0.95), as was length of stay and cost. Using an adaptation of the instrumental variable approach, increased rate of treatment with oral steroids was not associated with a change in the risk of treatment failure (OR for each 10% increase in hospital use of oral steroids, 1.00; 95% CI, 0.97-1.03). A total of 1356 (22%) patients initially treated with oral steroids were switched to intravenous therapy later in the hospitalization. CONCLUSION Among patients hospitalized for acute exacerbation of COPD low-dose steroids administered orally are not associated with worse outcomes than high-dose intravenous therapy.

Improving the care of patients with regard to chemotherapy-induced nausea and emesis: the effect of feedback to clinicians on adherence to antiemetic prescribing guidelines.

The call for accountability in health care quality has fueled the development of consumer-oriented Web sites that provide hospital ratings. Taking the consumer perspective, we compared five Web sites to assess the level of agreement in their rankings of local hospitals for four diagnoses. The sites assessed different measures of structure, process, and outcomes and did not use consistent patient definitions or reporting periods. Consequently, they failed to agree on hospital rankings within any diagnosis, even when using the same metric (such as mortality). In their current state, rating services appear likely to confuse, rather than inform, consumers.

Diabetes in Pregnancy in Zuni Indian Women: Prevalence and subsequent development of clinical diabetes after gestational diabetes

PURPOSE To evaluate the effect of performance and outcomes feedback on adherence to clinical practice guidelines regarding chemotherapy-induced nausea and emesis (CINE). METHODS Institutional CINE clinical practice guidelines were developed based on American Society of Clinical Oncology guidelines. Consecutive administrations of moderately/highly emetogenic chemotherapy were assessed for errors. Baseline statistical process control (SPC) charts were created and mean errors per administration were calculated. Prospective SPC charts were used to measure the effect of guideline development and distribution, a visiting lecturer, and ongoing feedback regarding compliance with guidelines employing SPC charts. Patients were surveyed regarding the extent and severity of CINE for 5 days postadministration. These outcomes were then shared with physicians. RESULTS Baseline compliance was poor (mean, 0.87 omissions per chemotherapy administration), largely because of inadequate adherence to recommendations for delayed CINE management. Most patients experienced delayed nausea, particularly on day 3 postchemotherapy. Physician prescribing performance did not undergo sustained improvement despite guideline development or distribution, a lecture by a visiting expert, or sharing of adherence data with clinicians. Once patient outcomes were shared, physicians accepted the need for compliance and instituted nurse practitioner antiemetic prescribing, with almost complete compliance and concurrent measurable reduction in day 3 nausea. SPC charts documented improvements in both outcomes. CONCLUSIONS SPC charts effectively monitor ongoing compliance and patient symptoms and represent appropriate outcome measurement and change facilitation tools. However, physician participation in guideline development and evidence of poor compliance alone did not improve prescribing performance. Only evidence of patient CINE experience coupled with noncompliance improved results.

Physician leadership: enhancing the career development of academic physician administrators and leaders.

Objective— To determine the prevalence of gestational diabetes mellitus in Zuni Indian women and the subsequent rate of diabetes among Zuni women with GDM. Research Design and Methods— A retrospective analysis of 809 deliveries over a 4-yr period among Zuni Indian women was conducted to determine the prevalence of GDM and diabetes antedating pregnancy. A prospective case-control study of 47 full-blooded Zuni Indian women with GDM and 47 control subjects was performed to determine the progression to clinical diabetes in women with a first-time diagnosis of GDM. Cases with GDM delivered during a defined 8-yr period. The control group of Zuni women delivered during the same time period but had plasma glucose values <7.8 mM on the 1-h glucose screening test. Cases with GDM and control subjects were matched for age, body mass index, gravidity, and length of follow-up. All women were re-evaluated for diabetes up to 9 yr after the index pregnancy. Results— Between 1987–1990, 116 cases of GDM and 8 cases of pre-existing diabetes were identified, giving a prevalence of maternal diabetes in pregnancy of 15.3%. At the time of follow-up, 14 of 47 (30%) women with GDM had developed diabetes after a mean of 4.8 yr compared with only 3 of 47 (6%) from the control group with an average of 5.5 yr follow-up. Conclusions— GDM is prevalent among Zuni Indians and is associated with an increased risk of diabetes. Glucose tolerance after GDM may deteriorate at a greater rate in Native Americans than in other populations.

Use of antihypertensive agents in the management of patients with acute ischemic stroke

As the health care environment grows more complex, there is greater opportunity for physician administrative and management leadership. Although physicians in general, and academic physicians in particular, view management as outside their purview, the increased importance of physician administrative leadership represents an opportunity for academic physicians interested in working at the interface of clinical medicine, health care, finance, and management. These physicians are called academic physician administrators and leaders (APALs). APALs are clinician–administrators whose academic contributions include both scholarly work related to their administrative duties and administrative leadership of academically important programs. However, existing academic career development infrastructure, such as academic promotions, is oriented toward traditional clinician–educator and clinician–researcher faculty. The APAL career path differs from traditional academic pathways because APALs require unique skills, different mentors, and a more expansive definition of academic productivity. This article describes how academic medical institutions could enhance the career development of academic physicians in administrative and leadership positions.

Voluntarily Reported Emergency Department Errors

Background: To protect the ischemic penumbra, guidelines have recommended against treating all but the severest elevations in blood pressure during acute ischemic stroke. Objective: To determine how often antihypertensive agents were used in routine clinical practice and whether this use was consistent with guideline recommendations. Methods: The records of patients discharged with ischemic stroke in 2000 at Baystate Medical Center in Springfield, MA, were reviewed. Adherence was evaluated by examining the use of antihypertensive agents in the context of daily blood pressure recordings during the first 4 days of hospitalization. Therapy was considered appropriate in the setting of severe hypertension (systolic blood pressure of >220 mm Hg or mean arterial blood pressure of >130 mm Hg) and potentially harmful in the setting of relative (systolic blood pressure of <120 mm Hg or mean arterial blood pressure of <85 mm Hg) or absolute (systolic blood pressure of <90 mm Hg or mean arterial blood pressure of <60 mm Hg) hypotension. Results: One hundred (65%) of the 154 ischemic stroke patients were treated with antihypertensive agents. Forty-two percent of those who had received therapy prior to admission had their regimen intensified, and 36% of previously untreated patients had therapy initiated. Sixteen (11%) patients had hypertension severe enough to warrant treatment upon arrival, and 34 (22%) had at least one episode of severe hypertension during the first 4 hospital days. Sixty-five (65%) patients developed relative hypotension on a day when antihypertensive agents were administered, and five (5%) developed absolute hypotension. Conclusions: Most patients with acute ischemic stroke are treated with antihypertensive agents despite the absence of severe hypertension. Although low blood pressure is common among treated patients, frank hypotension is unusual.

Public reporting of hospital quality: Recommendations to benefit patients and hospitals

Objective: To describe voluntarily reported errors in an Emergency Department using a reward-based system. Methods: Prospective, observational study in urban, academic Emergency Department with 107,000 annual visits. All Emergency Department staff with direct patient contact voluntarily reported medical errors on a standardized form. Staff received