| Summary: | 碩士 === 國立陽明大學 === 公共衛生研究所 === 98 === Objectives: Emergency department (ED) overcrowding has become an issue of concern since late 1980s. Recently, the impacts of ED overcrowding on the quality of healthcare are under even more scrutiny. However no study has ever evaluated the effects of ED overcrowding on the management of acute illnesses that develop among patients with chronic disease. We studied all patients who visited the ED of hospitals in the Taipei Sector of the Taiwan National Health Insurance Bureau due to concomitant stroke and pneumonia between 2005 and 2007 to better understand the impacts of ED overcrowding on the healthcare of such patients.
Methods: The study was a case-control study that included all patients who visited the ED of 17 hospitals in the Taipei Sector of the Taiwan National Health Insurance Bureau due to concomitant stroke and pneumonia between 2005 and 2007. Patients who died within 30 days after the ED visit (index date) were defined as cases, while those patients who survived more than 1 year after the index date were defined as controls. Cases and controls were then 1:2 matched on age and sex to study the association between ED overcrowding and death within 30 days after the index date. The secondary study aims of this study were the probability of admission to intensive care unit (ICU) and its related factors, and the medical expenses incurred from the ED visit.
In the analysis of death within 30 days after the index date or the probability of ICU admission, we first employed descriptive analysis and chi-square test. We then used conditional logistic regression analysis to estimate univariate and multivariate odds ratio of all variables. As for the analysis of average medical expenses per capita that incurred from the ED visit, we employed descriptive analysis, F test for uniavariate analysis, and generalized lineqar regression model for multivariate analysis.
Results: The main findings of this study were as follows:
(1). There was no statistical association between ED overcrowding and the risk of death among stroke patients complicated with pneumonia; however there was a dose-response relationship between the magnitude of ED overcrowding and the risk of death.
After controlling for patient’s characteristics (e.g. complication of limb paralysis, the presence of catastrophic illnesses, low household income), hospital’s characteristics (accreditated level of the hospital, and occupancy rate), time (season and calendar year of ED visit), ICU admission, classification of antibiotics, use of lipid-lowering drugs, and hospitalization in the prior 1 year, the odds ratio of death within 30 days after the index date for 2nd, 3rd, 4th, and 5th quintiles of ED overcrowding was 1.04, 95% CI 0.77~1.40，p= 0.80), 1.11 (95% CI 0.83~1.47，p= 0.50), 1.16 (95% CI 0.87~1.56，p= 0.31), and 1.32 (95% CI 0.99~1.76，p= 0.06) respectively, as compared with the 1st quintile, and there was a dose-response relationship
(2).Patient’s characteristics were the major risk factors of mortality.
Patients classified as ED triage level 1 had higher mortality rate as compared with patients classified as ED triage level 3 (OR 4.79 for all study patients; OR 4.19 for patients without paralys, OR 5.49 for patients with mild paralysis, OR 5.39 for patients with moderate paralysis, OR 5.51 for patients with severe paralysis; OR 4.83 for patients visiting medical centers, and OR 5.32 for patients visiting regional hospitals). Moreover, patients who had more than 3 hospitalizations in the previous year before the index date had higher mortality rate as compared to patients without prior hospitalization (OR 2.03 for all study patients; OR 2.23 for patients without paralysis, OR 1.95 for patients with severe paralysis; OR 2.00 for patients visiting medical centers, and OR 2.11 for patients visiting regional hospitals).
(3). Lipid-lowering drugs were associated with a lower risk of mortality, but were not associated with the risk of ICU admission.
Patients who ever used statins were associated with a lower risk of mortality (OR 0.73, 95% CI 0.57~0.92, p= 0.01 for all study patients; OR 0.56, 95 % CI 0.36~0.85，p= 0.01 for patients with severe paralysis; and OR 0.65, 95% CI 0.46~0.92，p< 0.01 for patients visiting medical centers). The use of statins however was not associated with the risk of ICU admission (OR 0.68, 95% CI 0.90~1.96，p= 0.16).
(4). Patient’s acuity and severity of patient’s illness were the risk factors of ICU admission.
After controlling for confounding variables, patients classified as ED Trige level 1 were 10.31 times more likely to be hospitalized to ICU, as compared with patients classified as ED Trige level 1. Moreover, patients with catastrophic illness were 1.65 times more likely to be hospitalized to ICU, and patients who received both 1st and 2nd generation antibiotios were 2.14 times more likely to have ICU admission, as compared with patients receiving 1st generation antibiotics. All of the aforementioned findings suggested that patient’s acuity and severity of patient’s illness were the risk factors of ICU admission. On the contrary, ED overcrowding and hospital’s occupancy rate were inversely associated with the risk of ICU admission.
(5). Severity of patient’s illness and diseases requiring higher cost of treatment were associated with more medical expenses.
Among the controls, patients classified as ED trige level 1 used 2.55 times the medical expenses than patients classified as ED triage level 3. Similarily, the patients who had catastrophic illness and patients with ICU admission used 2.31 times and 2.03 times the medical expenses than patients without catastrophic illness and patients without ICU admission, respectively. The above-noted findings showed that severity of patient’s illness was positively associated with the medical expenses after ED visit for pneumonia among stroke patients.
Conclusions: ED overcrowding may affect the mortality of patients. In this study, we found a borderline statistical significance and a dose-response relationship between ED overcrowding and mortality among patients with concomitant stroke and pneumonia. Moreover, the mortality of patients with both stroke and pneumonia was primarily associated with patient’s characteristics rather than the hospital’s characteristics, which indicated that the medial rights of vulnearable patients were not jeopardized. We also found that the use of statins might be associated with a lower risk of pneumonia related mortality, but did not protective against ICU admission. The study of the impacts of ED overcrowding on the quality of care among ED patients by using other diseases is warranted in the future.
|
|---|
