Attitudes and behavioral intentions regarding managed care: a comparison of academic and community physicians.

Physicians' attitudes toward managed care and the impact of these attitudes on behaviors that affect patient care are important factors in managed care reform. In addition, the attitudes of academic physicians may influence their willingness to reform medical education in an effort to prepare students to practice under managed care. Although it is a conventional opinion that the academic health center and its academic physicians are antagonistic toward managed care, there has not been a direct comparison of the attitudes of these physicians to those of practicing community physicians. We used a self-administered questionnaire to assess attitudes toward managed care and behavioral intentions regarding practices related to managed care; a sample of academic physicians (n = 129) was compared with a sample of community physicians (n = 307). Community physicians were less negative in their evaluations of the quality of care in a managed care environment, but no differences were identified between the two groups with regard to the cost-effectiveness, inevitability, or need to adapt to managed care. Academic specialists were more likely than academic primary care physicians to rate managed care as something to which they needed to adapt. Community physicians were less likely to report a willingness to change their referral patterns. Aggregating across practice type, we also uncovered systematic differences between primary care and specialist physicians. The data suggest that opinions about quality and cost-containment in managed care are significant correlates of intentions to change practice behaviors.

Response to dopamine vs norepinephrine in tricyclic antidepressant-induced hypotension.

OBJECTIVE: To compare the efficacy of dopamine (DA) with that of norepinephrine (NE) in the treatment of refractory hypotension caused by tricyclic antidepressants (TCAs). METHODS: A retrospective analysis was used to compare blood pressure responses in concurrent case series of patients treated with vasopressors for TCA-associated hypotension. The patients were adults (aged > or = 16 years) treated at 2 urban teaching hospitals from 1983 to 1994. All patients were diagnosed as having TCA ingestion (positive serum toxicologic assay), were hypotensive [systolic blood pressures (SBPs) < or = 90 mm Hg], and required vasopressor therapy. The patients were grouped by initial vasopressor treatment (i.e., NE vs DA). RESULTS: There were 26 hypotensive adult patients who met study criteria. All 26 patients remained hypotensive after initial treatment of their TCA-associated hypotension with crystalloid infusion and alkalinization. The NE and DA groups were similar in age, sex, and proportion of patients with single and co-drug ingestion. The NE group, however, had a significantly lower average SBP at study entry (56 vs 74 mm Hg, p = 0.04). Nine of 15 (60%) patients responded to DA (5-10 micrograms/kg/min) and 11/11 (100%) patients responded to NE (5-53 micrograms/min). The difference in response rates was statistically significant (p = 0.02). Six patients in whom DA (max rate 10-50 micrograms/min/kg) failed to raise the SBP subsequently responded to NE (max rate 5-74 micrograms/ min) when this drug was later used. One patient receiving NE (12 micrograms/min) developed ventricular ectopy, successfully treated with lidocaine. There were no ischemic complications from either NE or DA. Patients who failed DA therapy tended to be older (39 vs 30 years, p = 0.08), to be more hypotensive at study entry (64 vs 81 mm Hg, p = 0.008), and to remain hypotensive even at higher doses of DA (24 vs 7 micrograms/min/kg, p = 0.002). CONCLUSION: While this was not a prospective randomized study, NE appears advantageous over DA as the first-line vasopressor agent for those patients who require vasopressor support in the setting of TCA-induced hypotension that is refractory to IV fluid and serum alkalinization.

Physicians' attitudes toward managed care: assessment and potential effects on practice behaviors.

This study was designed to identify the key components of physicians' attitudes toward managed care and develop a tool to assess these components. We developed a questionnaire based on physicians' reactions to managed care, as reflected in the published literature. We mailed this questionnaire to a sample of 753 community physicians in the greater Sacramento area. A factor analysis of these data (n = 315) identified five unifactorial scales, which we labeled managed care quality, need to adapt to managed care, cost-containment effectiveness of managed care, personal knowledge of managed care, and inevitability of managed care. Physicians were most negative about the quality of managed care and most in agreement about the need to adapt to it. Correlations among these five scales, while statistically significant, were modest in size, suggesting that these physicians were quite discriminating in their evaluations. In comparison with medical/surgical specialists, primary care physicians rated the quality of managed care, their knowledge of it, and the inevitability of a national transition to managed care more positively. These measures predicted the physicians' intentions to alter their medical behaviors to comply with managed care practices.

High-dose versus standard-dose epinephrine treatment of cardiac arrest after failure of standard therapy.

STUDY OBJECTIVE: To assess the efficacy of high-dose epinephrine (HDE) compared with standard-dose epinephrine (SDE) in emergency department patients in cardiac arrest after SDE failed to improve asystole or ventricular fibrillation. DESIGN: Prospective, multicenter, blinded, controlled trial. SETTING: Eight academic center emergency departments. PATIENTS: One hundred forty patients treated for cardiac arrest. MEASUREMENTS AND MAIN RESULTS: Primary outcomes were either improvement in cardiac rhythm or return of spontaneous circulation (ROSC). Of the 140 patients enrolled, 78 received HDE and 62 received SDE. Of the 34 patients with ventricular fibrillation, 3 were resuscitated with HDE and 2 with SDE (p = 0.60). Of those with asystole, ROSC occurred in 12 of HDE and 5 of SDE recipients (p = 0.11). No patient had return of significant neurologic function and none survived to hospital discharge. Overall, there was no advantage to HDE after failure of SDE. CONCLUSION: Our results are similar to those of controlled clinical trials comparing HDE with SDE in cardiac arrest.

Randomized, prospective trial of fenoldopam vs sodium nitroprusside in the treatment of acute severe hypertension. Fenoldopam Study Group.

OBJECTIVE: To compare the safeties and efficacies of IV fenoldopam (FNP) vs sodium nitroprusside (NTP) in severe acute hypertension. METHODS: A prospective, randomized, open-label, multicenter international trial, at 24 academic medical centers, was conducted. The participants were adult patients (21-80 years of age) who had supine diastolic blood pressures (DBPs) > or = 120 mm Hg, were capable of written informed consent, and did not have selected exclusion criteria. The subjects were randomized to either FNP or NTP therapy; DBP was titrated to 95-110 mm Hg, or a maximum reduction of 40 mm Hg for very high pressures. Infusions were maintained for at least six hours, then the patients were weaned off the IV therapy and oral medication was started. Measurements included BP, heart rate, and duration of study drug infusion and frequency of side effects or complications. RESULTS: A total of 183 patients (90 FNP, 93 NTP) were enrolled. Fifteen patients from each arm were excluded from efficacy analysis due to protocol violation. There was no significant difference in baseline characteristics. The two antihypertensive agents were equivalent in controlling and maintaining DBP. Systolic blood pressure (SBP) was reduced to a slightly greater degree for the NTP-treated patients during the initial (0.5-1-hr) study period, and both SBP and DBP were reduced more for the FNP-treated patients in the subset receiving infusions during the 12-24-hour period. The adverse effect profiles of the drugs were similar, as were the times to achieve target pressure, with no clinically relevant difference. CONCLUSIONS: For patients who had acute severe hypertension, FNP and NTP were equivalent in terms of efficacy and acute adverse events. Because of a unique mechanism of action, FNP may have advantages in selected subsets of patients. Further studies may be indicated in patient populations with pure "hypertensive emergencies."

Recommendations for off-label use of intravenously administered immunoglobulin preparations. University Hospital Consortium Expert Panel for Off-Label Use of Polyvalent Intravenously Administered Immunoglobulin Preparations.

OBJECTIVE: To summarize consensus recommendations for off-label uses of standard intravenous immunoglobulin (IVIG), as developed by a University Hospital Consortium (UHC) Expert Panel. These findings are intended to help guide clinicians in the appropriate and efficient use of IVIG. PARTICIPANTS: The UHC-sponsored panel included eight physicians (board certified in critical care, hematology, immunology, neurology, oncology, pediatrics, or rheumatology) and two hospital pharmacists. EVIDENCE: MEDLINE and EMBASE were searched to identify all English-language review articles (n = 201) and original reports (n = 1904) on IVIG (human use only, excluding editorials, letters, and comments) published between January 1982 and March 1994. Relevant original reports (250) and review articles (87) were evaluated by the first author (T.A.R.). Extracted data included laboratory and clinical findings, objective measures, or clinical impressions. The evidence quality was graded by study design according to the US Preventive Services Task Force. CONSENSUS PROCESS: Before the panel meeting, a draft literature review and recommendations were produced by one of the authors (T.A.R.). The recommendations herein represent consensus (100% agreement) based on the published evidence. CONCLUSIONS: The UHC Expert Panel made specific recommendations for 53 off-label indications and the following general recommendations: (1) Usually IVIG is indicated only if standard approaches have failed, become intolerable, or are contraindicated; (2) IVIG products should be considered therapeutically equivalent and interchangeable; (3) interproduct pharmaceutical differences should be considered with the patient's clinical and physiological status when selecting an IVIG product; and (4) currently, IVIG manufacturers cannot guarantee freedom from viral contamination in the finished product.

Identifying toxicity risk early after antidepressant overdose.

Despite the risk of life-threatening toxicities, care of most patients after antidepressant overdose (ADO) does not require the use of critical care resources. The use of emergency department (ED) clinical findings to identify ADO patients who subsequently manifested toxicity was evaluated prospectively. ADO risk assessment (ADORA) criteria included development of QRS interval > 0.10 seconds, arrhythmias, altered mental status, seizures, respiratory depression, or hypotension. Sixty-seven ADO patients were identified on presentation to a single ED and classified as low risk (LR, absence of criteria) or high risk (HR, presence of one or more criteria) based on development of criteria within 6 hours of ingestion (or ED presentation if ingestion time was not established). This system demonstrated 100% sensitivity in identifying study patients who developed significant toxicity problems. None of the 28 LR patients and 13 of 39 HR patients had subsequent complications (P < .01). No single clinical finding permitted risk classification. ADORA should identify patients who do not require further monitoring or other aggressive medical management for ADO.

Treatment of gram-negative bacteremia and septic shock with HA-1A human monoclonal antibody against endotoxin. A randomized, double-blind, placebo-controlled trial. The HA-1A Sepsis Study Group.

BACKGROUND: HA-1A is a human monoclonal IgM antibody that binds specifically to the lipid A domain of endotoxin and prevents death in laboratory animals with gram-negative bacteremia and endotoxemia. METHODS: To evaluate the efficacy and safety of HA-1A, we conducted a randomized, double-blind trial in patients with sepsis and a presumed diagnosis of gram-negative infection. The patients received either a single 100-mg intravenous dose of HA-1A (in 3.5 g of albumin) or placebo (3.5 g of albumin). Other interventions, including the administration of antibiotics and fluids, were not affected by the study protocol. RESULTS: Of 543 patients with sepsis who were treated, 200 (37 percent) had gram-negative bacteremia as proved by blood culture. For the patients with gram-negative bacteremia followed to death or day 28, there were 45 deaths among the 92 recipients of placebo (49 percent) and 32 deaths among the 105 recipients of HA-1A (30 percent; P = 0.014). For the patients with gram-negative bacteremia and shock at entry, there were 27 deaths among the 47 recipients of placebo (57 percent) and 18 deaths among the 54 recipients of HA-1A (33 percent; P = 0.017). Analyses that stratified according to the severity of illness at entry showed improved survival with HA-1A treatment in both severely ill and less severely ill patients. Of the 196 patients with gram-negative bacteremia who were followed to hospital discharge or death, 45 of the 93 given placebo (48 percent) were discharged alive, as compared with 65 of the 103 treated with HA-1A (63 percent; P = 0.038). No benefit of treatment with HA-1A was demonstrated in the 343 patients with sepsis who did not prove to have gram-negative bacteremia. For all 543 patients with sepsis who were treated, the mortality rate was 43 percent among the recipients of placebo and 39 percent among those given HA-1A (P = 0.24). All patients tolerated HA-1A well, and no anti-HA-1A antibodies were detected. CONCLUSIONS: HA-1A is safe and effective for the treatment of patients with sepsis and gram-negative bacteremia.

Initial evaluation of human monoclonal anti-lipid A antibody (HA-1A) in patients with sepsis syndrome.

HA-1A, a human monoclonal immunoglobulin M antibody that binds specifically to the lipid A domain of endotoxin, was administered to septic patients to evaluate the safety, pharmacokinetics, and immunogenicity of the antibody. Thirty-four patients received a single infusion of either 25 mg, 100 mg, or 250 mg, and were followed clinically for 14 to 21 days after treatment. HA-1A serum levels were measured before infusion and frequently after infusion with a radiometric assay. A one-compartment pharmacokinetic model was fit to the measured serum levels, and accurately described the changes in HA-1A level over time in each dose group (r2 = .99). The mean +/- SEM apparent volume of distribution of HA-1A was 48.5 +/- 4.5 ml/kg, and the mean serum clearance was 2.8 +/- 0.4 ml/kg.h. The mean serum half-life of HA-1A was 15.9 +/- 1.5 h. The mean serum level one hour after a 100-mg dose was 33.2 +/- 2.4 micrograms/ml, and the mean concentration 24 h later was 9.1 +/- 1.6 micrograms/ml. The dose administered and presence of Gram-negative bacterial infection did not significantly influence the volume of distribution or serum clearance. No adverse reactions to HA-1A were observed, and no antibodies against HA-1A were detected in any patient. These data indicate that the pharmacokinetics of HA-1A are well described by a one-compartment pharmacokinetic model, and that HA-1A is safe and nonimmunogenic in patients with sepsis.

Antiulcer therapy: an exercise in formulary management.

The purpose of formulary management is to contain costs while preserving quality care. A pharmacy cost review can identify key areas where limiting the agents used could lower costs. H2-receptor antagonists represent such agents. At UC Davis Medical Center, H2-receptor antagonists were found to be one of the highest-cost drug groups in the hospital. Seventy percent of H2-receptor antagonist use was for intravenous prophylaxis of stress ulcer. After a committee review of available data, H2-receptor antagonists were considered therapeutically equivalent. Cimetidine was chosen as the primary H2-receptor antagonist because of an estimated 36% cost savings. Inpatient H2-receptor antagonist use changed from 90% ranitidine prior to the decision to 90% cimetidine after the decision. A monitoring program revealed that there were similar numbers of elevated drug levels for theophylline, phenytoin, and lidocaine before and after the programmatic change. There was no difference in numbers of patients with theophylline toxicity on cimetidine or ranitidine. The incidence of bleeding from stress ulcer was 1.4% in the year following the change to cimetidine use. With aggressive review and utilization of scientific data, cost-saving H2-receptor antagonist restriction decisions can be made that do not compromise quality care.

Effective measures for reducing blood loss from diagnostic laboratory tests in intensive care unit patients.

We studied ICU patient blood loss as a result of diagnostic testing (DBL) and the effect of two measures to reduce it. A policy of using small volumes (pediatric phlebotomy tubes, reduced syringe volumes) for the most frequent laboratory tests was implemented in our medical ICU. We prospectively studied 151 patients admitted during two consecutive 10-wk periods. During period 2, DBL was displayed on each ICU flow sheet. The DBL/day (43.6 +/- 3 [SEM] ml) was significantly lower (62.6 +/- 4 ml) than it would have been if standard volume tubes had been used. This represented an average savings of 33%. During period 1 (n = 81), eight (10%) patients with no diagnosis involving blood loss had a decrease in Hct and received transfusion. DBL was significantly higher (316 +/- 81 vs. 168 +/- 18 ml, p less than .001) for these patients and represented an average of 17% of transfusion requirements. During period 2 (n = 70), such transfusion requirements were significantly reduced (only one of 70, p less than .001), as were tests ordered/day (7.8 +/- 0.5 vs. 9.5 +/- 0.6, p less than .05). We conclude that DBL is a major health problem for the ICU patient. Routine use of small specimen volumes in this setting is warranted. Recording DBL for use in physician decision-making also significantly impacts this problem and should be considered an important part of the ICU database.

Clostridium difficile in the intensive care unit: management problems and prevention issues.

Several patients hospitalized in our 12-bed medical ICU were found to have Clostridium difficile associated colitis. Stool cultures of all patients identified eight cases (three culture positive, and five culture and cytotoxin positive), seven of which were geographically and temporarily clustered within a 2-wk period. At least one patient appeared to contract the disease after hospitalization and in the absence of antibiotic therapy or other known major risk factors. The outbreak highlights the problem of C. difficile in the ICU. We believe that a heightened awareness of the multiple risk factors and preventive measures, along with consideration of possible nosocomial transmission, will be necessary to prevent or arrest future clusters of cases in the ICU. The diagnosis of one patient with this infection in a unit should prompt a review of all other patients within the unit.

Selective left endobronchial suctioning in the intubated patient.

Suctioning of secretions from the left endobronchial tree is frequently necessary but often difficult in intubated patients. We examined the effectiveness of a catheter designed expressly for this purpose. Special curved tip (Bronchitrac-L) suction catheters were fitted with thin, radiopaque tubing to facilitate x-ray visualization. Eight-one attempts at left endobronchial placement were made on 74 stable adult intensive care unit patients. The suction catheter was inserted into the oral endotracheal tube or tracheostomy tube just prior to an x-ray filming of the chest. In 15 of 66 patients, the tip of the oral endotracheal tube was too distal (less than 2 cm above the carina) to allow proper functioning of the catheter. Patients with a properly positioned oral endotracheal tube were analyzed separately and showed 56 percent of the catheters went to the left bronchus. When the head was turned to the left prior to placement, successful left bronchus placement occurred in 65 percent. When the catheter was placed through a tracheostomy tube, 100 percent went into the left bronchus (n = 15). There were no catheter-induced complications in this study. The curved tip catheter is an effective means of suctioning the left bronchial tree in patients with tracheostomy tubes. Its reliability in patients with oral endotracheal tubes is reduced but more effective than current methods.

Superiority of activated charcoal alone compared with ipecac and activated charcoal in the treatment of acute toxic ingestions.

A prospective, randomized clinical trial compared the clinical effectiveness of syrup of ipecac and activated charcoal to that of activated charcoal alone in the treatment of acute toxic ingestions. Two hundred adult patients with mild to moderate oral overdoses were entered into the trial. Patients receiving only activated charcoal were discharged from the emergency department in significantly (P less than or equal to .05) less time than those receiving both syrup of ipecac and activated charcoal (6.0 +/- 0.3 vs 6.8 +/- 0.2 hours, respectively). The percentage of patients requiring nonpsychiatric hospitalizations was not significantly different between the two groups (11.2% vs 14.0%, respectively). For the hospitalized patients, the length of time spent in the ICU and in the hospital was not statistically different between the two groups. A complication rate of 5.4% was found with the ipecac and activated charcoal treatment compared with a 0.9% complication rate in the activated charcoal group (P less than or equal to .05). Three episodes of aspiration pneumonitis occurred after administration of ipecac and activated charcoal, while no episodes of aspiration were noted after treatment with only activated charcoal. Together, these data are consistent with the recommendation that ED treatment with activated charcoal alone be the gastrointestinal decontamination procedure of choice for the routine mildly-to-moderately orally poisoned adult patient.

Categorization of hospital emergency services--developing valid criteria.

In a pilot project sponsored by the California State Emergency Medical Services Authority, validated, verifiable criteria were developed for the vertical categorization of hospital emergency services in 11 different groupings of medical and surgical emergencies. We describe the development of an assessment process and categorization criteria to identify the most appropriate receiving facility for interfacility transfer and, in selected instances, field triage of patients with different levels of severity of illness or injury. We propose that this facility assessment project be used in the critical care planning process for the eventual vertical categorization of hospital emergency services in California and as a template for similar projects in other states.

Interactions of aminophylline and three benzodiazepine compounds with amygdala-kindled seizures in rats.

Fully amygdala-kindled rats were used to study the effect of diazepam, RO 15-1788, and CGS-8216 on aminophylline-induced prolongation of elicited kindled afterdischarges. Similar proportional reductions in seizure afterdischarge durations were seen with diazepam and with RO 15-1788 after aminophylline although the absolute length of the afterdischarge durations were increased significantly with both drugs after aminophylline. The partial agonist effect of the benzodiazepine antagonist RO 15-1788 was demonstrated before and after aminophylline pretreatment. However, no specific interaction (pro- or anticonvulsant) was demonstrated with the benzodiazepine antagonist CGS-8216 before or after aminophylline pretreatment. Together these data do not support the theory that the prolongation of elicited kindled seizures by the methylxanthine, aminophylline, is through a specific benzodiazepine receptor mechanism.