- Email: [email protected]
The effects of weather on pneumonia incidence
RESEARCH FORUM ABSTRACTS
accounting for 110,000 deaths and 500,000 hospitalizations annually. The forced expiratory volume in 1 second (FEV1) has been used for years as a reliable clinical outcome predictor of morbidity and mortality in patients with stable COPD, and several studies have shown that peak expiratory flow (PEF) values and FEV1 have good correlation during AECOPD. Recent reviews of the literature have not revealed a useful clinical predictor for the outcome risk or relapse risk in the acute setting. We proposed that a 35% improvement in PEF as a response to bronchodilator therapy can predict those patients with AECOPD who may be successfully discharged from the ED. Methods: Patients older than 50 years and with a history of COPD presenting to a large, academic ED with worsening dyspnea, increase in sputum volume, or increase in sputum purulence were prospectively enrolled by a physician, nurse, or licensed vocational nurse during a 7-month period. A chest radiograph and ECG were obtained on arrival. An initial PEF value was recorded (the best of 3 measurements). Two albuterol (2.5 mg) and Atrovent (0.5 mg) in 2 mL of saline nebulizer treatments were administered 10 to 20 minutes apart, followed by 50 mg of oral prednisone or 125 mg of intravenous Solu-Medrol and intravenous or oral antibiotics. A posttreatment PEF value was recorded (best of 3 measurements) 60 to 90 minutes after initiation of bronchodilator therapy. Additional therapy was administered at the physician’s discretion. Patients stable for discharge were given 50 mg of oral prednisone for 5 days, as well as an antibiotic chosen by the treating physician, and received telephone follow-up within 10 days of discharge. Those patients who were admitted, returned to the ED, or had an unscheduled primary care visit due to persistent or worsening symptoms were considered to be outpatient treatment failures. Results: During a 9-month period, 19 patients were enrolled. There was no significant difference between the first and last peak flow within the unsuccessfully treated group (P[.01) on the paired t test. There was a significant difference between the first and last peak flow within the successfully treated group (P=.008) on the paired t test. There was no significant difference in first peak flows between outcome groups (P[.01) on the independent samples t test. There was no significant difference in last peak flows between outcome groups (P[.01) on the independent samples t test. Conclusion: There is no statistical difference in the response to ED treatment of AECOPD by measuring PEF between patients successfully treated as outpatients and those who were admitted or failed outpatient therapy.
116
visits for pneumonia. Our best multivariate model isolated effects of sulfur dioxide and suspended small particles. Further study of more patients during longer periods may further clarify these relationships.
117
Pulmonary Embolism: Is It Truly Missed in the Emergency Department?
Henderson SO, Calder KK, Herbert M/Keck School of Medicine, Los Angeles, CA Study objectives: Emergency medicine and internal medicine literature states that pulmonary embolism is a common entity that is frequently misdiagnosed in the emergency department (ED) and often fatal when untreated. Definitive diagnosis entails costly radiographic studies, and it is likely that these studies continue to be overused, especially in low-risk patients. Our goals are to review current literature for statistics about the prevalence, misdiagnosis, and mortality of pulmonary embolism and to determine whether these were applicable to the ambulatory, hemodynamically stable patient presenting to the ED with signs and symptoms consistent with pulmonary embolism. Methods: We conducted an ancestral search for epidemiology statistics for pulmonary embolism, starting with review articles on pulmonary embolism in commonly referenced emergency medicine and internal medicine textbooks and peer-reviewed publications. Results: We located 20 articles for our search. These reported an incidence of pulmonary embolism ranging from 500,000 to 780,000 annually, with more than 400,000 missed diagnoses. There was general agreement on mortality for untreated disease, approximately 30% (range 18% to 35%). Either directly or indirectly, the most commonly referenced citation was dated 1975. This citation drew from government mortality statistics from the 1960s, inpatient and autopsy study data from the 1940s to 1960s, and author estimates to extrapolate: mortality rate, total incidence of disease, and number of cases that are misdiagnosed. There were no citations that assessed this information in the ED setting. Conclusion: Current epidemiology about pulmonary embolism may be applicable to inpatients or the recently deceased but is unlikely to provide an accurate representation of disease in the outpatient setting. Studies to provide incidence and outcome data for pulmonary embolism in the ambulatory ED patient are necessary.
The Effects of Weather on Pneumonia Incidence
Haspil-Corgan TA, Tewari M, Low RB, Gauff WR/University of Medicine and Dentistry of New Jersey, Newark, NJ Study objectives: We explore possible relationships between the incidences of pneumonia, weather, and air pollution. Methods: We identified all patients treated in our emergency department (ED) and diagnosed as having pneumonia from January 2003 to December 2003. We obtained weather data from the National Weather Service and pollution data from the Environmental Protection Agency. Data were analyzed with linear regression using SAS9 software. Results: During 2003, there were a total of 1,823 patients diagnosed with pneumonia in the ED. The number of visits per day was normally distributed, with a mean6SD of 5.062.3. Throughout the year, ambient air temperature was also normally distributed, with a mean of 11.8°C69.9°C; relative humidity was approximately normally distributed at 51.6%617.9%. The concentration of small particles (\2.5 mm) was not normally distributed, the 100th (maximum), 90th, 75th, 50th (median), 25th, and 10th percentiles are 59.3, 26.4, 19.2, 13.2, 9.9, and 8.0 ppm, respectively. Bivariate analysis showed statistically significant effects on pneumonia incidence of temperature (P\.0001), humidity (P\.0001), amount of precipitation (P\.0001), small particles (\2.5 mm; P\.0001), sulfur dioxide (SO2; P\.0001), oxides of nitrogen (NOx), nitrogen dioxide (NO2; P\.0001), carbon monoxide (CO; P\.0001), and ozone (O3; P\.0001) levels. The best multivariate model showed combined statistically significant effects of SO2 (P=.0009), small particles (P=.0539), temperature (P=.0004), and humidity (P\.0001). The pollutants were positively correlated with each other. Conclusion: Contrary to popular belief, the number of patients diagnosed with pneumonia increases as it gets warmer. Consistent with popular belief, the incidence of pneumonia increases along with increasing relative humidity. Studied individually, increasing levels of all air pollutants are associated with increasing ED
OCTOBER 2004
44:4
ANNALS OF EMERGENCY MEDICINE
118
Clinical Gestalt and the Diagnosis of Pulmonary Embolism: Does Experience Matter?
Kabrhel C, Camargo CA, Goldharber SZ/Massachusetts General Hospital, Boston, MA Study objectives: We seek to determine whether the accuracy of pretest assessment of the likelihood of pulmonary embolism was related to physician experience. We compared the accuracy of the subjective pretest probability assessment made by senior physicians (postgraduate year [PGY]-41) to that of interns (PGY-1) and residents (PGY-2 and PGY-3) working in the emergency department (ED) of a large teaching hospital. Methods: We performed a prospective observational study in an urban, academic ED with annual census of 48,000 patient visits. Eligible patients had at least 1 diagnostic test ordered to evaluate pulmonary embolism. The physician treating the patient was asked whether he or she considered pulmonary embolism the ‘‘most likely’’ diagnosis or whether an alternative diagnosis was ‘‘most likely.’’ This result was compared with the patient’s ultimate diagnosis. Physicians’ experience was categorized by the number of years of training since medical school graduation. Results: We enrolled 583 adults evaluated for pulmonary embolism in the ED. Pulmonary embolism was considered the most likely diagnosis for 26% of patients. Pulmonary embolism was diagnosed in 10% of patients. There was a trend toward increasing diagnostic accuracy with increasing experience, as demonstrated by the frequency of true positive assessments (17% versus 20% versus 25%), true negative assessments (89% versus 94% versus 96%), and likelihood ratio (1.49 versus 2.34 versus 3.33), respectively (P for trend .1). However, when compared with validated clinical prediction rules, no rule was identified with a likelihood ratio between those of the inexperienced and experienced physicians in our study. Conclusion: Clinical experience appears to be related to the ability to accurately determine the pretest probability of pulmonary embolism. However, this difference is unlikely to be clinically significant. The difference in accuracy between
S 3 7
accounting for 110,000 deaths and 500,000 hospitalizations annually. The forced expiratory volume in 1 second (FEV1) has been used for years as a reliable clinical outcome predictor of morbidity and mortality in patients with stable COPD, and several studies have shown that peak expiratory flow (PEF) values and FEV1 have good correlation during AECOPD. Recent reviews of the literature have not revealed a useful clinical predictor for the outcome risk or relapse risk in the acute setting. We proposed that a 35% improvement in PEF as a response to bronchodilator therapy can predict those patients with AECOPD who may be successfully discharged from the ED. Methods: Patients older than 50 years and with a history of COPD presenting to a large, academic ED with worsening dyspnea, increase in sputum volume, or increase in sputum purulence were prospectively enrolled by a physician, nurse, or licensed vocational nurse during a 7-month period. A chest radiograph and ECG were obtained on arrival. An initial PEF value was recorded (the best of 3 measurements). Two albuterol (2.5 mg) and Atrovent (0.5 mg) in 2 mL of saline nebulizer treatments were administered 10 to 20 minutes apart, followed by 50 mg of oral prednisone or 125 mg of intravenous Solu-Medrol and intravenous or oral antibiotics. A posttreatment PEF value was recorded (best of 3 measurements) 60 to 90 minutes after initiation of bronchodilator therapy. Additional therapy was administered at the physician’s discretion. Patients stable for discharge were given 50 mg of oral prednisone for 5 days, as well as an antibiotic chosen by the treating physician, and received telephone follow-up within 10 days of discharge. Those patients who were admitted, returned to the ED, or had an unscheduled primary care visit due to persistent or worsening symptoms were considered to be outpatient treatment failures. Results: During a 9-month period, 19 patients were enrolled. There was no significant difference between the first and last peak flow within the unsuccessfully treated group (P[.01) on the paired t test. There was a significant difference between the first and last peak flow within the successfully treated group (P=.008) on the paired t test. There was no significant difference in first peak flows between outcome groups (P[.01) on the independent samples t test. There was no significant difference in last peak flows between outcome groups (P[.01) on the independent samples t test. Conclusion: There is no statistical difference in the response to ED treatment of AECOPD by measuring PEF between patients successfully treated as outpatients and those who were admitted or failed outpatient therapy.
116
visits for pneumonia. Our best multivariate model isolated effects of sulfur dioxide and suspended small particles. Further study of more patients during longer periods may further clarify these relationships.
117
Pulmonary Embolism: Is It Truly Missed in the Emergency Department?
Henderson SO, Calder KK, Herbert M/Keck School of Medicine, Los Angeles, CA Study objectives: Emergency medicine and internal medicine literature states that pulmonary embolism is a common entity that is frequently misdiagnosed in the emergency department (ED) and often fatal when untreated. Definitive diagnosis entails costly radiographic studies, and it is likely that these studies continue to be overused, especially in low-risk patients. Our goals are to review current literature for statistics about the prevalence, misdiagnosis, and mortality of pulmonary embolism and to determine whether these were applicable to the ambulatory, hemodynamically stable patient presenting to the ED with signs and symptoms consistent with pulmonary embolism. Methods: We conducted an ancestral search for epidemiology statistics for pulmonary embolism, starting with review articles on pulmonary embolism in commonly referenced emergency medicine and internal medicine textbooks and peer-reviewed publications. Results: We located 20 articles for our search. These reported an incidence of pulmonary embolism ranging from 500,000 to 780,000 annually, with more than 400,000 missed diagnoses. There was general agreement on mortality for untreated disease, approximately 30% (range 18% to 35%). Either directly or indirectly, the most commonly referenced citation was dated 1975. This citation drew from government mortality statistics from the 1960s, inpatient and autopsy study data from the 1940s to 1960s, and author estimates to extrapolate: mortality rate, total incidence of disease, and number of cases that are misdiagnosed. There were no citations that assessed this information in the ED setting. Conclusion: Current epidemiology about pulmonary embolism may be applicable to inpatients or the recently deceased but is unlikely to provide an accurate representation of disease in the outpatient setting. Studies to provide incidence and outcome data for pulmonary embolism in the ambulatory ED patient are necessary.
The Effects of Weather on Pneumonia Incidence
Haspil-Corgan TA, Tewari M, Low RB, Gauff WR/University of Medicine and Dentistry of New Jersey, Newark, NJ Study objectives: We explore possible relationships between the incidences of pneumonia, weather, and air pollution. Methods: We identified all patients treated in our emergency department (ED) and diagnosed as having pneumonia from January 2003 to December 2003. We obtained weather data from the National Weather Service and pollution data from the Environmental Protection Agency. Data were analyzed with linear regression using SAS9 software. Results: During 2003, there were a total of 1,823 patients diagnosed with pneumonia in the ED. The number of visits per day was normally distributed, with a mean6SD of 5.062.3. Throughout the year, ambient air temperature was also normally distributed, with a mean of 11.8°C69.9°C; relative humidity was approximately normally distributed at 51.6%617.9%. The concentration of small particles (\2.5 mm) was not normally distributed, the 100th (maximum), 90th, 75th, 50th (median), 25th, and 10th percentiles are 59.3, 26.4, 19.2, 13.2, 9.9, and 8.0 ppm, respectively. Bivariate analysis showed statistically significant effects on pneumonia incidence of temperature (P\.0001), humidity (P\.0001), amount of precipitation (P\.0001), small particles (\2.5 mm; P\.0001), sulfur dioxide (SO2; P\.0001), oxides of nitrogen (NOx), nitrogen dioxide (NO2; P\.0001), carbon monoxide (CO; P\.0001), and ozone (O3; P\.0001) levels. The best multivariate model showed combined statistically significant effects of SO2 (P=.0009), small particles (P=.0539), temperature (P=.0004), and humidity (P\.0001). The pollutants were positively correlated with each other. Conclusion: Contrary to popular belief, the number of patients diagnosed with pneumonia increases as it gets warmer. Consistent with popular belief, the incidence of pneumonia increases along with increasing relative humidity. Studied individually, increasing levels of all air pollutants are associated with increasing ED
OCTOBER 2004
44:4
ANNALS OF EMERGENCY MEDICINE
118
Clinical Gestalt and the Diagnosis of Pulmonary Embolism: Does Experience Matter?
Kabrhel C, Camargo CA, Goldharber SZ/Massachusetts General Hospital, Boston, MA Study objectives: We seek to determine whether the accuracy of pretest assessment of the likelihood of pulmonary embolism was related to physician experience. We compared the accuracy of the subjective pretest probability assessment made by senior physicians (postgraduate year [PGY]-41) to that of interns (PGY-1) and residents (PGY-2 and PGY-3) working in the emergency department (ED) of a large teaching hospital. Methods: We performed a prospective observational study in an urban, academic ED with annual census of 48,000 patient visits. Eligible patients had at least 1 diagnostic test ordered to evaluate pulmonary embolism. The physician treating the patient was asked whether he or she considered pulmonary embolism the ‘‘most likely’’ diagnosis or whether an alternative diagnosis was ‘‘most likely.’’ This result was compared with the patient’s ultimate diagnosis. Physicians’ experience was categorized by the number of years of training since medical school graduation. Results: We enrolled 583 adults evaluated for pulmonary embolism in the ED. Pulmonary embolism was considered the most likely diagnosis for 26% of patients. Pulmonary embolism was diagnosed in 10% of patients. There was a trend toward increasing diagnostic accuracy with increasing experience, as demonstrated by the frequency of true positive assessments (17% versus 20% versus 25%), true negative assessments (89% versus 94% versus 96%), and likelihood ratio (1.49 versus 2.34 versus 3.33), respectively (P for trend .1). However, when compared with validated clinical prediction rules, no rule was identified with a likelihood ratio between those of the inexperienced and experienced physicians in our study. Conclusion: Clinical experience appears to be related to the ability to accurately determine the pretest probability of pulmonary embolism. However, this difference is unlikely to be clinically significant. The difference in accuracy between
S 3 7