Most Monitored Patients Donʼt Benefit from Monitoring

Like most EDs, ours is experiencing substantial problems with gridlock. Too many patients for the staff to manage, inability to move patients to beds on the floors, patients in the halls, lots of patients with flu-like illness who actually think we can do something for them — everyone knows the scenario. If we could fix a single problem, the one most likely to result in the most favorable impact, it would be the rapid admission of patients to the inpatient areas. We often find ourselves holding patients who are designated to go to monitored units because beds are not available. The fact is that most patients on these monitored units get no benefit from their monitoring. To compound the matter, there are generally no objective criteria used by the hospital and medical staff to determine whether a patient is a candidate for being on a monitored unit — if the primary physician wants a patient there, then that's where he goes. It's clear at our hospital and I dare say at most others that the ratio of patients on monitored units to those on general medical floors is askew. A lot more patients should be on general medical floors, and a lot less should be monitored (assuming that every patient cannot be monitored). If this were to occur, monitored beds would be more likely to be available for patients who might truly benefit from them, and ED gridlock might be lessened. As an aside, why can't we send the patients who are waiting in our halls to be admitted to the halls on the floors — our hallways are no more special than theirs. What makes it acceptable to have patients in the ED hallways but not in the floor hallways? The following study is one of two of which I'm aware that clearly demonstrates that the vast majority of patients get no benefit from being monitored. This being the case, it would seem that hospitals and their medical staffs would try to limit the use of monitoring when beds are tight through the development of criteria that could be used to determine which patients were appropriate for placement on a monitored unit. Continuous Electrocardiographic Monitoring and Cardiac Arrest Outcomes in 8,932 Telemetry Ward Patients Schull MJ, et al Acad Emerg Med 2000;7:647 METHODS: In this study from the University of Toronto, the authors did a structured semi-explicit chart review of cardiac arrest occurring among 8,932 patients admitted to a non-CCU cardiac telemetry ward during a five-year period (1994–1999). The decision to admit a patient to the telemetry unit was at the discretion of the managing physician, and not subject to prespecified criteria or protocols. RESULTS: Cardiac arrest occurred in 20 patients (0.2%), representing about one arrest for every 500 patients on telemetry monitoring. Telemetry was actually in use at the time of the arrest in 16 of these 20 patients, and the arrest onset was recognized as a result of a monitoring abnormality in only nine cases. Only three patients survived to hospital discharge after cardiac arrest, all of whom had been on a cardiac monitor at the time of the arrest (although the arrest was felt to be heralded clearly by a monitoring abnormality in only one case). Among a subgroup of 4,305 patients with an admission diagnosis of chest pain of undetermined cause, angina, unstable angina or “rule out MI,” only one developed a (fatal) cardiac arrest (0.02%). CONCLUSIONS: Cardiac arrest was very rare among patients admitted to a non-CCU telemetry ward, and continuous EKG monitoring in these patients often failed to herald the arrest or influence survival. These findings suggest that a more selective approach to continuous EKG monitoring may be safe and more efficient. Decreasing Length of Stay Another way to increase bed availability is to decrease patients' lengths of stay. This will not only free up beds for other patients who may be waiting in the ED, but will also increase net revenues if the patients are being paid for on a DRG-type basis. The following article nicely points out, however, that money saved by earlier discharge of DRG-type patients will not necessarily result in proportionate savings. This is because hospital costs are not the same for each day a patient is hospitalized. By far, the greatest costs to a hospital for the care of a patient are in the initial days. The last days are relatively inexpensive. Length of Stay Has Minimal Impact on the Cost of Hospital Admission Taheri PA, et al J Am Coll Surg 2000;191:123 BACKGROUND: Although efforts to reduce the length of stay (LOS) for hospitalized patients are believed to result in substantial decreases in health care costs, there is some evidence to suggest that the marginal cost of one additional hospital day may be relatively minimal. METHODS: The authors from the University of Michigan Medical Center examined costs of care for 12,365 patients who survived to discharge after a hospital stay of four or more days during fiscal year 1998. The analysis considered variable direct costs associated with the care of an individual patient on a given day (including nursing costs), fixed direct costs associated with a specific department but not a particular patient, and indirect costs incurred entirely outside of individual departments. RESULTS: The mean and median LOS were 10.5 days and seven days, respectively. The last full day before discharge accounted for 9.5 percent of the mean LOS but only 2.4 percent of total costs of care. Mean variable direct costs on the last hospital day were $396 for patients having major surgery (1.5% of total costs) and $432 for those not having major surgery (3.4% of average total costs). In a separate analysis of 665 patients discharged from the Level I trauma center, the median variable direct cost on the first day of the hospital stay was more than four times higher than that of the last hospital day, and hospital overhead (fixed direct and indirect costs) accounted for 58 percent of the mean total cost per patient. CONCLUSIONS: This study suggests that not all hospital days are economically equivalent, and that efforts to reduce the LOS by one day will only minimally influence health care costs. Interventions to reduce costs early during the hospital stay may be more productive. Another Infection Not Benefited by Blood Cultures Most of the time that blood cultures are obtained in the ED, they are a waste of time and money. Blood cultures rarely change therapy in pyelonephritis because nothing is found on blood cultures that is not found on urine cultures. Blood cultures in source-unknown pediatric fever in otherwise well children also are a tremendous waste of resources because the incidence of pediatric bacteremia is now probably about 1.5 percent given the effectiveness of the H. influenzae vaccine and the fact that the vast majority of the bacteremia that remains (pneumococcal) clears by itself or is cleared by making the ubiquitous diagnosis of “otitis media,” probably the most overdiagnosed disorder in all of medicine. With the widespread use of pediatric pneumococcal vaccine, this percentage of febrile children with bacteremia will likely approach zero. In pneumonia, blood cultures also are generally a total waste of time given that pneumonia is supposed to be empirically treated based on the likely pathogens to be causing it. The next paper shows the nonvalue of blood cultures in cellulitis. Bottom line: Although blood cultures in sick patients with source-unknown infections may be appropriate, in most settings, despite long-standing knee-jerk tradition, blood cultures are a waste of time and money because they rarely change treatment, and when they result in a change of treatment, it is often unclear whether the change mattered. Cost-Effectiveness of Blood Cultures for Adult Patients with Cellulitis Perl B, et al Clin Infect Dis 1999;29:1483 METHODS: The authors of this Israeli study did an implicit chart review regarding the use and utility of blood cultures in 757 adults hospitalized with non-facial cellulitis. RESULTS: Blood specimens were obtained for culture in 73 percent of these patients, and yielded a pathogen in only two percent of cases. A contaminant (defined by clinical outcome) was isolated in 3.6 percent. At a cost of $50 per negative culture and $100 per positive culture (including sensitivity testing), the total cost of culturing was more than $36,000. The 11 pathogens isolated were group G Streptococcus (5/11 cases), group A Streptococcus (3/11), S. aureus (1/11), Vibrio vulnificus (1/11), and Morganella morganii (1/11). Blood culture results prompted a change from cefazolin to penicillin in eight patients. Because cellulitis is often empirically treated with a first-generation cephalosporin, empiric therapy would have been appropriate in the nine patients with Gram positive bacteremia. The two patients with Gram negative bacteremia had unusual clinical characteristics that the authors feel would likely have prompted additional work-up. CONCLUSIONS: Most (non-facial) cellulitis in adults is caused by group A Streptococcus or S. aureus, and is usually empirically treated with agents directed at these organisms. Blood cultures are often performed in such patients at substantial cost, but their yield is very low, and their impact on clinical management is marginal. Parenteral vs. Oral NSAIDs Those of you who have read this column in the past are aware of my ongoing soap box campaign against the widespread use of parenteral NSAIDs that are no more effective than their oral counterparts (except perhaps in vomiting patients). One of the reasons cited for giving an injection for pain is that patients will more likely perceive the IM medication to be stronger and more effective than an oral medication. Although I intuit there may be some medical literature to support this position, it certainly is not unanimous in this regard. The following study clearly demonstrated that the route of administration of a pain medication did not result in differing perceptions of its effectiveness. Patients' Perceptions of Route of Nonsteroidal Anti-Inflammatory Drug Administration and its Effect on Analgesia NA, et al Acad Emerg Med 2001;7:857 BACKGROUND: Although several studies comparing oral and parenteral NSAIDs found no difference in their analgesic effects, many health care providers believe patients respond better to parenteral analgesics in general because of a placebo effect associated with “stronger” parenteral treatment. METHODS: This randomized, double-blind prospective study from the University of Pittsburgh compared the effects of oral with intramuscular (IM) administration of a “placebo NSAID” in 64 adults presenting to the ED with acute musculoskeletal pain due to minor trauma. All of the patients were pretreated with an orange-flavored drink that contained 800 mg of ibuprofen. The patients were told to drink this liquid because of a “need for hydration.” The patients received a placebo, either as a tablet or as an IM injection. They then scored their pain on a 100mm visual analogue scale (VAS) at baseline and serially for two hours after treatment. Neither the health care providers nor the patients were aware of the content of the drink, the tablet, or the injection. RESULTS: The mean VAS score decreased from 58 at baseline to 29 at 120 minutes in the injection group, and from 60 to 31 in the oral treatment group. There were no differences between the two groups at any of the scoring intervals. CONCLUSIONS: In this study, there were no measurable differences in the effect of a placebo added to an active analgesic, regardless of the route of administration of that placebo. Avoid Routine Creatinine Testing with IV Contrast Here's an interesting paper that addresses a practice that appears to be routine at some hospitals, although fortunately our hospital is not among them. You would think that if it were so important that entire hospitals adopt this practice, it would be the standard of care in the community, although obviously it is not. The study addresses the necessity to measure creatinine levels routinely prior to infusion of IV contrast in radiologic procedures, in this case CTs. It nicely demonstrated that a few simple questions could determine the vast majority of at-risk patients. Are Screening Serum Creatinine Levels Necessary Prior to Outpatient CT Examinations? Tippins RB, et al Radiology 2000;216:481 BACKGROUND: Because IV radiographic contrast material can impair renal function, serum creatinine levels are often routinely measured prior to performance of IV contrast-enhanced imaging studies. METHODS: The authors from the Emory Clinic in Atlanta retrospectively reviewed the records of 2,034 outpatients having measurement of serum creatinine prior to IV contrast-enhanced CT scanning. RESULTS: The creatinine was in the normal range (less than 1.5mg/dl) in 92 percent of the patients, borderline (1.5–1.9mg/dl) in five percent, and abnormal (2.0mg/dl or higher) in three percent. Risk factors for contrast nephrotoxicity in the 102 patients with borderline serum creatinine and the 66 with abnormal creatinine levels were compared with risk factor profiles in a random sample of 240 patients with serum creatinine in the normal range. Because there was no linear correlation between the serum creatinine level and patient age, age was excluded as a risk factor. Nearly all of the patients with abnormal serum creatinine (97%) had one or more risk factors for contrast nephrotoxicity, most commonly a history of renal insufficiency (94%). Diabetes (insulin- or noninsulin-dependent) was present in 18 percent of the patients with abnormal serum creatinine. Only two patients with serum creatinine in the abnormal range (specifically, 2.0 and 2.2mg/dl) had no identifiable risk factors. CONCLUSIONS: The authors suggest that questioning patients who are scheduled for contrast-enhanced imaging studies regarding risk factors for contrast-induced nephrotoxicity, with particular attention to a history of renal disease or diabetes, would identify essentially all patients who would require a baseline measurement of serum creatinine, and that selective screening based on the presence or absence of risk factors could be associated with substantial cost savings. Threats Successful in Changing Nursing Behavior Finally, the last paper demonstrates that nurses can be coerced to do things they don't want to do by threatening to fire them. Unfortunately, there is a growing checklist of items that various well-intended organizations have forced upon us in the emergency department in their attempt to improve the health and safety of the community, independent of the patient's reason for coming to the ED. The list is growing longer and longer — measure head circumference, tell parents to put their children in car seats properly, don't smoke, don't have unsafe sex, don't take drugs, be immunized, avoid fatty foods, tell us if loved ones mistreat you, and oh, yes, and have a nice day. Effect of an Administrative Intervention on Rates of Screening for Domestic Violence in an Urban Emergency Department Larkin GL, et al Am J Pub Health 2000;90:1444 BACKGROUND: The JCAHO has mandated interventions to enhance early identification of victims of domestic violence, but adherence to universal screening in various health care settings has been found to be suboptimal. METHODS: The authors from the University of Pittsburgh examined the effect of a disciplinary intervention on adherence to screening for domestic violence by ED nurses. It was mandated that nurses use a six-item Domestic Safety Assessment to screen all female ED patients 18 or older for domestic violence. When it was found that compliance was only 21 percent, a four-tiered disciplinary intervention was introduced. A screening rate below 90 percent was first addressed with counseling of the noncompliant nurse. A second offense prompted additional counseling and documentation in the employee's record, while a third offense prompted written warning of termination, and a fourth offense resulted in termination of employment. Documentation of screening was examined for 1,638 patient records for visits occurring before the disciplinary intervention and 1,617 post-intervention records. RESULTS: Completion of the Domestic Safety Assessment increased from 29 percent before the intervention to 73 percent after the intervention. Of 40 full-time ED nurses, 54 percent received verbal counseling for a first offense, 20 percent received second-tier counseling, and 10 percent received written warning of termination for a third offense. No nurse was terminated. CONCLUSIONS: In this study, a disciplinary intervention improved rates of screening for domestic violence among women presenting to the ED. The study did not examine the effects of this policy on employee morale or attempt to verify the accuracy of screening.

Récupéré en direct depuis OpenAlex et désinversé. Les résumés ne sont pas conservés dans cette base de données : les index inversés représentent 8,6 Go des 9,3 Go de texte de la base, et le serveur dispose de 13 Go libres.