| 1.ObjectiveTo explore the clinical application and effect of the improved triage intelligence triage system,to analyze each evaluation index in the system by empirical research method,and to further improve the evaluation index system of emergency diagnosis.2.MethodsThis study adopts the empirical research method,which is divided into two parts and three stages: the first part,the present situation of the clinical application of the improved intelligent emergency triage system;the second part is the evaluation of the effect of the improved intelligent emergency triage system.The first stage,we used the descriptive research method to understand the status quo of clinical application of improved intelligent emergency triage system,4401 emergency department patients from September 1,2016 to September 30 were choosed as this research object.The first study stage applied this system to analysis the distribution of emergency triage levels,time distribution,the time points of the change in the amount of visits and the distribution of the frequcency of triage.The second stage included 100 emergency cases who were choosed to evaluat the sensitivity,specificity and the clinical effect of the system.The third stage included 475 emergency critical cases who were selected by convenient sampling method to record the chief complaints,to eavaluate indicators and to adjust triage level of this group.All the cases are re-triage by three different interception methods: Routine method(group A),"Routine method+MEWS"(group B)and "Routine method+ MEWS+NRS"(group C).The results of re-triage were compared with the adjusted triage.Analyzed the difference between the three evaluation methods,calculated the accuracy and missed rate of the three methods' grade.To expand the sample size and reduce bias,the number of cases has been extended to 500.According to the order of the three intercepting indexes mentioned above,the index groups were included in logistic regression analysis.By Logistic regression equation to establish three model: Model ?,Model ? and Model ?,to analyze the correlation between each index and the level,and to compare the goodness of fit of each model.3.Results3.1 The grades and time distribution of the improved intelligent emergency triage system: A total of 4401 patients were triaged and the triage frequency was 4728.Among them the grade distribution was as follows: grade ? has 31 patients accounting for the total number of 0.66%,grade?has 319(6.75%),grade ? has 239(5.05%),grade ? has 4139(87.54%).The time distribution of all grades shows: The first wave peak appeared in the four levels of the diagnosis are all at 8:00~12:00 in the morning;The second wave peak appeared at night,but at all grades,evening peaks are slightly different.The evening peaks of ? and ? grade are all at 18:00 ~ 20:00;? grade' evening peak would appear until 18:00 0:00;Level IV'evening peak mainly around 20:00.The number of the patients which were triaged twice or above was 269,which accounted for 6.11% of the total.3.2 Reliability,sensitivity and specificity of the improved intelligent emergency triage system: The consistency test of triage nurses showed that kappa= 0.972.The reliability test between the system grading and standard grading showed that Cronbach's ?=0.82.The specificity and the predictive value of this system for emergency patients was 94.67% and 81.82%.The predictive value for non-emergency patients was 91.03%..There were 7% patients were been over triaged and 8% patients were been insufficient triaged.Severe abdominal pain,chest distress,fever and other symptoms are always occurred in bouth excessive grading or lack of triage.3.3 The comparison of evaluation model of the improved intelligent emergency triage system:3.3.1 The rate of accuracy: The rate of accuracy of Group C(85.47%,82.53%)were higher than Group A(82.11%,80.21%)and Group B(70.95%,77.05%)(P<0.05).3.3.2 The rate of the missed detection: The rate of the missed detection of critically ill patients,the potentially critical patients,and nonemergency patients are respectively 34.78%,31.61%,and 31.61%.They are all lower than the other two methods.The results of the three methods were statistically different(P < 0.05).3.3.3 The Comparison of model goodness of fit: Based on the multi-Logistics model analysis:The Comparison of model goodness of fit shows that the model ?(OR=0.465)better than modle ?(OR=0.387)and model ?(OR=0.405).Age,respiratory and NRS are all have significant correlation for I and IV,for grade ? age(OR=1.053,95%CI 1.026~1.082),respiratory(OR=1.504,95%CI 1.211~1.868),NRS(OR=1.372,95%CI 1.156~1.628);for grade ? age(OR=1.037,95% CI1.021~ 1.053),respiratory(OR=1.373,95% CI1.142~1.651),NRS(OR=1.395,95%CI 1.266~ 1.538).4.Conclusion4.1 The improved intelligent emergency triage system have realized automatic,fast data entry,statistics and intelligent analysis.The data of the first emergency medical information were shared,Dynamic monitoring condition changes and dynamic traceability were also realized.This system make clinic treatment process for rational optimization of emergency resource allocation and provides an objective basis for implementation of flexible scheduling realize the significance of the auxiliary decisionmaking triage.4.2 The improved intelligence triage system make the efficiency of the triage work been progressed,the aided recognition becoming more objectively,faster and accurately,it can assist in the decision-making for nurses in the case of limited time and emergency resources.4.3 The Unite assessment method combines the advantages of this three methods:the simple and rapid of routine evaluation method,the stronger warning and good prediction function of MEWS scoring method,the physiological and psychological assessmen from NRS.It can better reflect the severity of condition,reduce the missd detection of critically ill patients.The constructed evaluation index model has a good fit of goodness of fit and provides scientific reference for the further improvement of the evaluation criteria.4.4 Through this research,the concept of intelligent emergency triage classification management of emergency triage have been introduced to the emergency triage nursing management with potential triage tool,which laid a foundation for further development of the auxiliary decision making tool.We hope that this research can provide clinical reference and data support to the follow-up development of the triage system in develop the scoring algorithm,the classification of the diagnostic accuracy rate.Then make the more accurate data entry procedures triage tool which have batter performance,availability,and upgrade on adaptability to better in the service of the nursing workers,ensure the safety of the patient. |
PDF Full Text Request