Derivation of a clinical decision rule to guide the interhospital transfer of patients with blunt traumatic brain injury.

OBJECTIVE: To derive a clinical decision rule for people with traumatic brain injury (TBI) that enables early identification of patients requiring specialised trauma care. METHODS: We collected data from 1999 through 2003 on a retrospective cohort of consecutive people aged 18-65 years with a serious head injury (AIS > or =3), transported directly from the scene of injury, and evaluated in the ED. Information on 22 demographical, physiological, radiographic, and lab variables was collected. Resource based "high therapeutic intensity" measures occurring within 72 hours of ED arrival (the outcome measure) were identified a priori and included: neurosurgical intervention, exploratory laparotomy, intensive care interventions, or death. We used classification and regression tree analysis to derive and cross validate the decision rule. RESULTS: 504 consecutive trauma patients were identified as having a serious head injury: 246 (49%) required at least one of the HTI measures. Five ED variables (GCS, respiratory rate, age, temperature, and pulse rate) identified subjects requiring at least one of the HTI measures with 94% sensitivity (95% CI 91 to 97%) and 63% specificity (95% CI 57 to 69%) in the derivation sample, and 90% sensitivity and 55% specificity using cross validation. CONCLUSIONS: This decision rule identified among a cohort of head injured patients evaluated in the ED the majority of those who urgently required specialised trauma care. The rule will require prospective validation in injured people presenting to non-tertiary care hospitals before implementation can be recommended.

Influence of trauma system implementation on process of care delivered to seriously injured patients in rural trauma centers.

BACKGROUND: Statewide trauma systems are implemented by health care policy makers whose intent is to improve the process of care delivered to seriously injured patients. In Oregon, Advanced Trauma Life Support (ATLS) training was mandated for all physicians employed in the emergency department of trauma centers. The purpose of this study was to test the hypothesis that mandatory ATLS training favorably influenced processes of care. METHODS: Seriously injured patients treated at 9 rural Level 3 and Level 4 hospitals were studied before (PRE) and after (POST) implementation of Oregon's trauma system. The processes of care evaluated on the basis of chart review were 20 diagnostic and therapeutic interventions advocated in the ATLS course. A cumulative process score (CPS) between 0 and 1 was assigned on the basis of the processes of care delivered. A CPS of 1 indicated optimal process of care. RESULTS: Mean CPS for 506 PRE period patients (0.44 +/- 0.27) was significantly lower than the mean CPS for 512 POST period patients (0.57 +/- 0.27) with an unpaired t test (P <.001). For the subgroup with injury severity score of 16 to 34, the mean CPS of survivors (0.67 +/- 0.19) was significantly higher than the mean CPS of decedents (0.57 +/- 0.25). CONCLUSIONS: Process of care for seriously injured patients improved after categorization of rural trauma centers in Oregon. Evidence shows improved process of care may have benefitted patients with serious but survivable injuries. Measurement of process of care is an alternative to mortality analysis as an indication of the quality of care.

Mortality among seriously injured patients treated in remote rural trauma centers before and after implementation of a statewide trauma system.

BACKGROUND: Injury mortality in rural regions remains high with little evidence that trauma system implementation has benefited rural populations. OBJECTIVE: To evaluate risk-adjusted mortality in remote regions of Oregon before and after implementation of a statewide trauma system. RESEARCH DESIGN: A retrospective cohort study assessing injury mortality through 30 days after hospital discharge. SETTING: Nine rural Oregon hospitals serving counties with populations <18 persons per square mile. SUBJECTS: Severely injured patients presenting to four level-3 and five level-4 trauma hospitals 3 years before and 3 years after trauma system implementation. MEASURES: Interhospital transfer, hospital death, and demise within 30 days following hospital discharge. RESULTS: A total of 940 patients were analyzed. After trauma system implementation, patients presenting to level-4 hospitals were more likely transferred to level-2 facilities (P <0.001). Interhospital transfer times from level-3 hospitals lengthened significantly after system implementation (P <0.001). Overall mortality rates were higher in the postsystem period (8.3%) than the presystem period (6.7%), but not significantly. Controlling for covariates, no additional benefit to risk-adjusted mortality was associated with trauma system implementation. Additional deaths, occurring after trauma system implementation, included head-injured patients transferred from rural hospitals to nonlevel-1 trauma center hospitals. CONCLUSIONS: Increased injury survival after Oregon trauma system implementation, demonstrated in urban and statewide analyses, was not confirmed in remote regions of the state. Efforts to improve trauma systems in rural areas should focus on the processes of care for head-injured patients transferred to higher designation trauma centers.

Impact of community intervention to reduce patient delay time on use of reperfusion therapy for acute myocardial infarction: rapid early action for coronary treatment (REACT) trial. REACT Study Group.

BACKGROUND: Reperfusion therapy for acute myocardial infarction (AMI) is a time-dependent intervention that can reduce infarct-related morbidity and mortality. Out-of-hospital patient delay from symptom onset until emergency department (ED) presentation may reduce the expected benefit of reperfusion therapy. OBJECTIVE: To determine the impact of a community educational intervention to reduce patient delay time on the use of reperfusion therapy for AMI. METHODS: This was a randomized, controlled community-based trial to enhance patient recognition of AMI symptoms and encourage early ED presentation with resultant increased reperfusion therapy rates for AMI. The study took place in 44 hospitals in 20 pair-matched communities in five U.S. geographic regions. Eligible study subjects were non-institutionalized patients without chest injury (aged > or =30 years) who were admitted to participating hospitals and who received a hospital discharge diagnosis of AMI (ICD 410); n = 4,885. For outcome assessment, patients were excluded if they were without survival data (n = 402), enrolled in thrombolytic trials (n = 61), receiving reperfusion therapy >12 hours after ED arrival (n = 628), or missing symptom onset or reperfusion times (n = 781). The applied intervention was an educational program targeting community organizations and the general public, high-risk patients, and health professionals in target communities. The primary outcome was a change in the proportion of AMI patients receiving early reperfusion therapy (i.e., within one hour of ED arrival or within six hours of symptom onset). Trends in reperfusion therapy rates were determined after adjustment for patient demographics, presenting blood pressure, cardiac history, and insurance status. Four-month baseline was compared with the 18-month intervention period. RESULTS: Of 3,013 selected AMI patients, 40% received reperfusion therapy. Eighteen percent received therapy within one hour of ED arrival (46% of treated patients), and 32% within six hours of symptom onset (80% of treated patients). No significant difference in the trends in reperfusion therapy rates was attributable to the intervention, although increases in early reperfusion therapy rates were noted during the first six months of the intervention. A significant association of early reperfusion therapy use with ambulance use was identified. CONCLUSIONS: Community-wide educational efforts to enhance patient response to AMI symptoms may not translate into sustained changes in reperfusion practices. However, an increased odds for early reperfusion therapy use during the initiation of the intervention and the association of early therapy with ambulance use suggest that reperfusion therapy rates can be enhanced.

Patient satisfaction investigations and the emergency department: what does the literature say?

BACKGROUND: Patient satisfaction is an indicator of the quality of care provided by emergency department (ED) personnel. It is this perception of satisfaction that becomes the basis for future ED choice or the recommendation of a specific ED to other potential patients. OBJECTIVE: To perform an evidence-based literature review to: 1) characterize measures of "patient satisfaction"; 2) identify factors that have been associated with overall ED patient satisfaction; 3) critique the methods used to assess patient satisfaction in the literature; and 4) address how this information can be beneficial to those reading the satisfaction literature or designing a satisfaction survey instrument. METHODS: The MEDLINE database was searched for studies addressing ED patient satisfaction, from January 1976 through July 1999, using MeSH terms and a text word search. Bibliographies of manuscripts also were searched for additional relevant articles and each clinical study was used as a search criterion in Science Citation Index, from the date of publication through July 1999. RESULTS: Multiple measures have been used to evaluate overall patient satisfaction. Sixteen studies were found associating ED patient satisfaction with service and patient factors. Most studies are observational and of these, most are cross-sectional. Hence, cause-and-effect determination of factors responsible for patient satisfaction cannot be resolved using the current literature. CONCLUSIONS: Despite considerable methodologic variability, key themes (e.g., association of satisfaction with patient information, provider-patient interpersonal factors, and perceived waiting time) emerge from review of the ED patient satisfaction literature. To standardize future investigations, clinicians and investigators should use a common definition for the state of overall patient satisfaction, e.g., when the patient's own expectations for treatment and care are met (or exceeded). This common definition should be incorporated into the instrument used to measure overall ED patient satisfaction.

Association between prepayment systems and emergency medical services use among patients with acute chest discomfort syndrome. For the Rapid Early Action for Coronary Treatment (REACT) Study.

STUDY OBJECTIVE: Cost concerns may inhibit emergency medical services (EMS) use. Novel tax-based and subscription prepayment programs indemnify patients against the cost of EMS treatment and transport. We determine whether the presence of (or enrollment in) prepayment plans increase EMS use among patients with acute chest discomfort, particularly those residing in low-income areas, those lacking private insurance, or both. METHODS: This study uses a subset of baseline data from the REACT trial, a multicenter, randomized controlled community trial designed, in part, to increase EMS use. The sample includes 860 consecutive noninstitutionalized patients (>30 years old) presenting with nontraumatic chest discomfort to hospital emergency departments in 4 Oregon/Washington communities. The association between prepayment systems and EMS use was analyzed using multivariable logistic regression. RESULTS: Overall EMS use was 52% (n=445). Among EMS users, 338 (75%) were subsequently admitted to the hospital and 110 (25%) were released from the ED. Prepayment was not associated with increased EMS use in the overall patient sample. However, patients residing in low-income census block groups (median annual income <$30,000) were 2.6 times (95% confidence interval [CI] 1.4 to 4.8) more likely to use EMS when a prepayment system was available than when no system was present. No association was noted among higher-income block group residents. Among low-income block group residents lacking private insurance, prepayment systems were associated with 3.8 times (95% CI 1.2 to 13.4) greater EMS usage. CONCLUSION: Economic considerations may affect EMS system utilization among underinsured and low-income patients experiencing a cardiac event. Prepayment systems may increase EMS utilization among these groups.

Effect of a community intervention on patient delay and emergency medical service use in acute coronary heart disease: The Rapid Early Action for Coronary Treatment (REACT) Trial.

CONTEXT: Delayed access to medical care in patients with acute myocardial infarction (AMI) is common and increases myocardial damage and mortality. OBJECTIVE: To evaluate a community intervention to reduce patient delay from symptom onset to hospital presentation and increase emergency medical service (EMS) use. DESIGN AND SETTING: The Rapid Early Action for Coronary Treatment Trial, a randomized trial conducted from 1995 to 1997 in 20 US cities (10 matched pairs; population range, 55,777-238,912) in 10 states. PARTICIPANTS: A total of 59,944 adults aged 30 years or older presenting to hospital emergency departments (EDs) with chest pain, of whom 20,364 met the primary population criteria of suspected acute coronary heart disease on admission and were discharged with a coronary heart disease-related diagnosis. INTERVENTION: One city in each pair was randomly assigned to an 18-month intervention that targeted mass media, community organizations, and professional, public, and patient education to increase appropriate patient actions for AMI symptoms (primary population, n=10,563). The other city in each pair was randomly assigned to reference status (primary population, n=9801). MAIN OUTCOME MEASURES: Time from symptom onset to ED arrival and EMS use, compared between intervention and reference city pairs. RESULTS: General population surveys provided evidence of increased public awareness and knowledge of program messages. Patient delay from symptom onset to hospital arrival at baseline (median, 140 minutes) was identical in the intervention and reference communities. Delay time decreased in intervention communities by -4.7% per year (95% confidence interval [CI], -8.6% to -0.6%), but the change did not differ significantly from that observed in reference communities (-6. 8% per year; 95% CI, -14.5% to 1.6%; P=.54). EMS use by the primary study population increased significantly in intervention communities compared with reference communities, with a net effect of 20% (95% CI, 7%-34%; P<.005). Total numbers of ED presentations for chest pain and patients with chest pain discharged from the ED, as well as EMS use among patients with chest pain released from the ED, did not change significantly. CONCLUSIONS: In this study, despite an 18-month intervention, time from symptom onset to hospital arrival for patients with chest pain did not change differentially between groups, although increased appropriate EMS use occurred in intervention communities. New strategies are needed if delay time from symptom onset to hospital presentation is to be decreased further in patients with suspected AMI. JAMA. 2000;284:60-67

Prehospital delay in patients hospitalized with heart attack symptoms in the United States: the REACT trial. Rapid Early Action for Coronary Treatment (REACT) Study Group.

BACKGROUND: The use of thrombolytic therapy for patients with myocardial infarction has been limited by patient delay in seeking care. We sought to characterize prehospital delay in patients hospitalized for evaluation of heart attack symptoms. METHODS AND RESULTS: The Rapid Early Action for Coronary Treatment (REACT) is a multicenter, randomized community trial designed to reduce patient delay. At baseline, data were abstracted from the medical records of 3783 patients hospitalized for evaluation of heart attack symptoms in 20 communities. The median prehospital delay was 2.0 hours; 25% of patients delayed longer than 5.2 hours. In a multivariable analysis, delay time was longer among non-Hispanic blacks than among non-Hispanic whites, longer at older ages, longer among Medicaid-only recipients and shorter among Medicare recipients than among privately insured patients, and shorter among patients who used an ambulance. CONCLUSIONS: The observed pattern of differences is consistent with the contention that demographic, cultural, and/or socioeconomic barriers exist that impede rapid care seeking.

First-postgraduate-year resident clinical time use on three specialty rotations.

OBJECTIVE: To compare in-hospital time uses by first-postgraduate-year (PGY1) residents during rotations in emergency medicine (EM), internal medicine (IM), and surgery (S). This article reports the clinical components of residency time use. METHODS: A cross-sectional, observational study of the clinical activities of EM PGY1 residents was performed while the residents were on duty during the three specialty rotations. The activities were recorded by an observer using a log with predetermined categories for clinical activities. A time-blocked, convenience sample of resident shifts was observed for each service rotation. The sample was proportional to the total number of hours for which a PGY1 resident was expected to be in the hospital during a rotation on that service. No attempt was made to sample the same resident at all time periods or on all rotations. Proportions were compared by chi2; alpha = 0.0001. RESULTS: Twelve PGY1 residents were observed for a total of 166 hours on S, 156 hours on IM, and 120 hours on EM. These hourly amounts were representative of a typical two-week span of service on each rotation for the residents. On average, the residents spent 57% of their time on clinical or service-oriented activities. During EM and IM rotations, the residents spent most of their time performing clinical information gathering and engaging in case management and data synthesis (52% of total clinical effort). Within this category, residents on EM were more involved with case discussion and review of ancillary test results than on IM (34% vs 20% of time in this category). Conversely, proportionately less time in this category was devoted to documentation on the EM vs IM rotation (56% vs 80%; p < 0.0001). The greatest opportunity to perform procedures was on the S rotation (31% of total clinical time vs 6% for other specialties; p < 0.0001). CONCLUSION: Awareness of the clinical activities performed on PGY1 rotations can help residency directors anticipate educational needs to balance their residents' experience. Since 29% and 42% of total clinical time on PGY1 EM and IM rotations, respectively, is focused on documentation, efforts to enhance charting skills and efficiency are warranted. Also, efforts to enhance PGY1 procedural experience outside of the S rotation appear warranted.

Trauma system impact on admission site: a comparison of two states.

OBJECTIVE: To introduce a measure assessing statewide hospital admission patterns for trauma and evaluate its utility in identifying significant admission redistributions in neighboring states as a function of trauma system implementation. METHODS: A retrospective claims database analysis was performed of hospitalized trauma patients in Washington and Oregon. The site of hospitalization and diagnostic information were used to determine an admission rate difference for each hospital: the rate of index patient admissions minus the rate of non-index patient admissions. Mean admission rate differences for Level II and III versus other hospitals in each state were compared as a function of time. RESULTS: A bias in favor of admitting more serious injuries to Level II and III hospitals was seen with trauma system implementation in both states, but it was more pronounced in the Oregon trauma system. CONCLUSION: Assessment of admission rate differences is a helpful method of characterizing the evolution of statewide trauma systems.

Knowledge of heart attack symptoms in a population survey in the United States: The REACT Trial. Rapid Early Action for Coronary Treatment.

BACKGROUND: Greater use of thrombolysis for patients with myocardial infarction has been limited by patient delay in seeking care for heart attack symptoms. Deficiencies in knowledge of symptoms may contribute to delay and could be a target for intervention. We sought to characterize symptom knowledge. METHODS: Rapid Early Action for Coronary Treatment is a community trial designed to reduce this delay. At baseline, a random-digit dialed survey was conducted among 1294 adult respondents in the 20 study communities. Two open-ended questions were asked about heart attack symptom knowledge. RESULTS: Chest pain or discomfort was reported as a symptom by 89.7% of respondents and was thought to be the most important symptom by 56.6%. Knowledge of arm pain or numbness (67.3%), shortness of breath (50.8%), sweating (21.3%), and other heart attack symptoms was less common. The median number of correct symptoms reported was 3 (of 11). In a multivariable-adjusted model, significantly higher mean numbers of correct symptoms were reported by non-Hispanic whites than by other racial or ethnic groups, by middle-aged persons than by older and younger persons, by persons with higher socioeconomic status than by those with lower, and by persons with previous experience with heart attack than by those without. CONCLUSIONS: Knowledge of chest pain as an important heart attack symptom is high and relatively uniform; however, knowledge of the complex constellation of heart attack symptoms is deficient in the US population, especially in low socioeconomic and racial or ethnic minority groups. Efforts to reduce delay in seeking medical care among persons with heart attack symptoms should address these deficiencies in knowledge.

Assessment of chest pain onset and out-of-hospital delay using standardized interview questions: the REACT Pilot Study. Rapid Early Action for Coronary Treatment (REACT) Study Group.

OBJECTIVE: To determine the consistency of responses to a standardized 2-part "key" question (Key-Q) about acute symptom onset in patients presenting with chest pain when measured using alternative questions (Qs) about symptom perception and decisions to seek treatment. METHODS: A structured patient interview was performed at 3 university teaching hospitals and 1 community hospital. Convenience samples of adult patients presenting to these EDs with chest pain were asked specific questions related to their symptoms and recognition of illness. Information obtained included the 2-part Key-Q: "What are the symptoms that brought you here today?" and "When did those symptoms start?" The alternative Qs (in order of use) were as follows: Q1 = "When did your very first symptom or sensation begin?"; Q2 = "When did your symptoms lead you to think something was wrong or that you were ill?"; Q3 = "When did your symptoms become serious enough for you to seek medical care?"; and Q4 = "When did you actually call 9-1-1/emergency medical services (EMS) or go to the hospital?" The documented ED arrival time, demographic variables, and whether the patient arrived by ambulance were obtained from the medical record. Patients also were queried regarding potential barriers to seeking medical care and their cardiac risk factors. RESULTS: Of the 135 patients surveyed, 9 were unsure of the date and time of symptom onset. For the 126 patients with analyzable data, the mean (+/- SD) patient age was 62 +/- 16 years, and 59% were male. The general sequence of events reported from acute symptom onset until hospital care was Q1/Key-Q-->Q2-->Q3-->Q4-->ED arrival. The median differences and interquartile ranges (IQRs) in minutes between Q times and the Key-Q response were: Q1 = 0 (0-0); Q2 = 30 (0-210); Q3 = 140 (30-720); Q4 = 265 (90-1,215); and ED arrival = 340 (120-1,230). The interval from the Key-Q response until calling 9-1-1/EMS or going to the hospital was shorter for those who used an ambulance and for those who did not consult a physician first. The interval from the Key-Q response until considering symptoms to be serious was shorter for those with a family history of heart disease, but longer for non-white patients. CONCLUSION: The Key-Q elicited a response recalled near the time of first symptoms and generally before the patient had concluded something was "wrong or that he or she was ill." Measurement of the out-of-hospital delay in chest pain patients using the Key-Q appears promising.

Resident educational time study: a tale of three specialties.

OBJECTIVE: To compare amounts of in-hospital time use by PGY1 residents during rotations in emergency medicine (EM), internal medicine (IM), and surgery. This article reports the general study methodology and focuses on the educational aspects of residency time use. METHODS: A cross-sectional, observational study of the activities of EM PGY1 residents was performed while the residents were on duty during the 3 specialty rotations. The activities were recorded by an observer using a log with predetermined categories for clinical/service, educational, and personal areas. A time-blocked, convenience sample of resident shifts was observed for each service rotation. The sample was proportional to the total number of hours for which a PGY1 resident was expected to be in the hospital during a rotation on that service. No attempt was made to sample the same resident at all time periods or on all rotations. RESULTS: Twelve PGY1 residents were observed for a total of 166 hours on surgery, 156 hours on IM, and 120 hours on EM. These hourly amounts were representative of a typical 2-week span of service on each rotation for the residents. On average, the residents spent 57% of their time on clinical or service-oriented activities, 24% on educational activities, and 19% on personal activities. The proportions of time devoted to the 3 major areas were similar for the 3 rotations. In all 3 rotations, the largest proportion of time was spent on patient-focused education (81% to 92% of total educational time). Only 2% to 11% of educational time was devoted to self-education. Within the patient-focused education category, proportionately less resident time with faculty occurred on the surgery rotation than on the EM and IM rotations (18% vs 30% and 27%, respectively). CONCLUSION: The general breakdowns of clinical/service, educational, and personal time use by PGY1 residents are proportionately similar for the 3 service rotations. Patient-focused education is the primary mode of education for all services. In-hospital, self-education time is limited. Clinical teaching is largely by nonfaculty. The educational implications of these findings are discussed.

Rapid early action for coronary treatment: rationale, design, and baseline characteristics. REACT Research Group.

OBJECTIVE: Early reperfusion for acute myocardial infarction (AMI) can reduce morbidity and mortality, yet there is often delay in accessing medical care after symptom onset. This report describes the design and baseline characteristics of the Rapid Early Action for Coronary Treatment (REACT) community trial, which is testing community intervention to reduce delay. METHODS: Twenty U.S. communities were pair-matched and randomly assigned within pairs to intervention or comparison. Four months of baseline data collection was followed by an 18-month intervention of community organization and public, patient, and health professional education. Primary cases were community residents seen in the ED with chest pain, admitted with suspected acute cardiac ischemia, and discharged with a diagnosis related to coronary heart disease. The primary outcome was delay time from symptom onset to ED arrival. Secondary outcomes included delay time in patients with MI/unstable angina, hospital case-fatality rate and length of stay, receipt of reperfusion, and ED/emergency medical services utilization. Impact on public and patient knowledge, attitudes, and intentions was measured by telephone interviews. Characteristics of communities and cases and comparability of paired communities at baseline were assessed. RESULTS: Baseline cases are 46% female, 14% minorities, and 73% aged > or =55 years, and paired communities have similar demographics characteristics. Median delay time (available for 72% of cases) is 2.3 hours and does not vary between treatment conditions (p > 0.86). CONCLUSIONS: REACT communities approximate the demographic distribution of the United States and there is baseline comparability between the intervention and comparison groups. The REACT trial will provide valuable information for community educational programs to reduce patient delay for AMI symptoms.

Adequacy of hospital discharge status as a measure of outcome among injured patients.

CONTEXT: Crude mortality rates at the time of hospital discharge are commonly used to assess the quality of care provided to patients hospitalized following trauma. OBJECTIVES: To evaluate the adequacy of hospital death rates as an outcome measure following trauma and to determine the influence of noninjury illness as a cause of hospital death and the frequency of postdischarge death. DESIGN: Retrospective cohort analyses using hospital discharge data for injured patients cross-linked to death certificate data that provided 1 year of follow-up for all patients discharged alive. PATIENTS: A total of 90048 injured patients admitted to all acute care hospitals in the state of Washington from 1991 through 1993 and discharged with at least 1 diagnosis coded in the International Classification of Diseases, Ninth Revision, Clinical Modification to indicate trauma. MAIN OUTCOME MEASURES: Death in the hospital and death within 30 days of hospital discharge. RESULTS: Among 1912 injured patients with in-hospital deaths, 825 death certificates (43%) listed a noninjury cause of death. The overall mortality rate at hospital discharge was 21.2 per 100000 hospitalized injured patients, and was 12.1 per 100000 for trauma deaths and 9.1 per 100000 for those designated as nontrauma deaths. Patients with trauma-related death designations were younger (mean age, 51.5 years vs 77.9 years), had shorter lengths of stay (median stay, 2 days vs 5 days), and sustained more severe injures (P<.001). Including the 1273 deaths that occurred within 30 days of hospital discharge increased rates for trauma-designated deaths to 14.1 per 100000 and increased rates for nontrauma-designated deaths to 21.3 per 100000. CONCLUSIONS: Hospital discharge death rates are incomplete measures of death frequency for injured patients. Designation of the cause of death, especially among older, hospitalized, injured patients often reflects preexisting medical conditions. Adequate assessment of mortality following trauma requires measurement of the frequency of death following hospital discharge.