论文部分内容阅读
背景马尔堡心脏分数(MHS)是一项简单、有效、稳健的决策尺度,能够帮助全科医生为胸痛患者排除冠心病的可能。目的探讨MHS是否能够帮助全科医生进行临床决策。设计和场所选取全科诊疗中832例胸痛患者数据进行诊断准确性的比较研究。方法利用MHS制定诊断策略:单独以MHS为基础进行诊断、使用MHS进行分诊测试、利用MHS帮助全科医生进行临床诊断。将上述结果的准确性与无协助全科医生诊断结果的准确性进行比较。结果无协助全科医生诊断结果的灵敏度和特异度分别为82.9%〔95%CI(72.4%,89.9%)〕和61.0%〔95%CI(56.7%,65.2%)〕。与之比较,单独以MHS为基础进行诊断的灵敏度更高〔相差8.5%,95%CI(-2.4%,19.6%)〕,特异度相近〔相差-0.4%,95%CI(-5.3%,4.5%)〕;使用MHS进行分诊测试的灵敏度相近〔相差-1.5%,95%CI=-9.8%,7.0%)〕,特异度更高〔相差11.6%,95%CI(7.8%,15.4)〕;利用MHS帮助全科医生进行临床诊断的灵敏度和特异度更高〔灵敏度相差8.0%,95%CI(-6.9%,23.0%);特异度相差5.8%,95%CI(-1.6%,13.2%)〕。结论利用MHS进行初步分诊有助于促进全科医疗中冠心病的临床诊断。
Background The Marlborough Heart Score (MHS) is a simple, effective and robust decision-making scale that can help GPs rule out the possibility of coronary heart disease in patients with chest pain. Objective To investigate whether MHS can help general practitioners make clinical decisions. Design and location of general practitioners selected 832 cases of chest pain in patients with diagnostic accuracy of the comparative study. Methods MHS was used to develop diagnostic strategies: MHS-based diagnosis alone, MHS-based triage testing, and MHS to assist general practitioners in clinical diagnosis. The accuracy of these results was compared with the accuracy of non-assisted GP diagnosis. The sensitivity and specificity of the results for non-GP diagnosis were 82.9% (95% CI, 72.4%, 89.9%) and 61.0% (95% CI, 56.7%, 65.2%, respectively). In contrast, the sensitivity based on MHS alone was higher (8.5% versus 95% CI (-2.4%, 19.6%)] and similar in specificity (-0.4%, 95% CI, -5.3% 4.5%)]; the sensitivity of the triage test using MHS was similar (-1.5% difference, 95% CI -9.8%, 7.0%)], with higher specificity (11.6%, 95% CI, 7.8%, 15.4 )]. The sensitivity and specificity of using MHS to help general practitioners in clinical diagnosis were higher (sensitivity difference 8.0%, 95% CI (-6.9%, 23.0%); specificity difference 5.8%, 95% CI , 13.2%)]. Conclusions The initial triage of MHS can help to promote the clinical diagnosis of coronary heart disease in general practice.