| BackgroundMalnutrition is a public health problem that has been troubling clinicians for long time,because the clinical problems brought by malnutrition have a profound impact on the long-term clinical outcome of patients.In recent years,clinicians,especially in emergency and critical care medical,have attached great importance to nutrition support,aiming at providing patients with refined,individualized and appropriate clinical nutrition support therapy plans.The pathophysiological status of critically ill patients is complex,and a single treatment regimen cannot improve the clinical outcome,but appropriate nutritional support therapy can help patients improve their condition.At present,the nutritional evaluation methods for critically ill patients are limited and complex,and the accuracy and timeliness cannot be guaranteed.Many clinical studies have shown that it is necessary to establish the corresponding nutritional evaluation methods according to the pathophysiological characteristics of critically ill patients.ObjectiveObjectiveThe most appropriate way to assess a patient's nutritional status and deliver nutrition support in clinical practices,especially for critically ill patients in intensive care units,is widely debated.Patients in emergency departments(EDs)often exhibit signs and symptoms of malnutrition.Only a few studies have assessed the nutritional status of ED patients,and have revealed the limited applicability of the existing clinical approaches to assess the nutritional status of these patients.Against this backdrop,the present study aimed to:Investigate how ED nurses and physicians perceive nutrition support therapy by distributing online questionnaires to specific groups of medical practitioners nationwide,and offer practical and feasible solutions to nutrition support education and publicity based on the existing problems.Discuss the possibility of establishing an easy-to-use nutritional assessment or scoring system for ED patients using the results of their nutritional assessment and ultrasound rectus femoris(RF)muscle measurement.MethodPart 1Online questionnaires were distributed to ED nurses and physicians throughout the country to gauge their understanding of nutrition support therapy and self-evaluation in clinical practice.Questionnaires can be valuable resources for decision-makers to improve education programs for ED nurses and physicians who require knowledge about nutrition.Qualitative data were subject to textual analysis performed by two researchers,and their findings were compared side-by-side.All discrepancies were discussed until a consensus was reached,and questionable data were reviewed to ensure reliability.Typical answers to open questions were quoted to improve the reliability of the qualitative data represented by n(%).Data were analyzed using SPSS version 26.0.Enumeration data were expressed as mean±standard deviation(x ± s)and compared among groups using an independent samples t test.Measurement data were represented by n(%)and compared among groups using chi-squared test or Fisher's exact test.The significance level a was set at 0.05.Part 2:Sixty-six patients admitted by the Emergency Room of the Emergency Department of Peking Union Medical College Hospital were invited to participate in this study between July and December 2019.During their stay in the Emergency Department,their conditions were under close observation and relevant data were acquired within a week.This study includes the following patient information:gender,age,body mass index(BMI),albumin(ALB),scores for the Acute Physiology and Chronic Health Evaluation(APACHE II)and the Nutrition Risk Screening 2002(NRS-2002),RF muscle thickness,primary diagnosis of disease(s),administration of nutrition support therapy and clinical treatment,and clinical prognosis(whether a patient survives after discharge).Nutrition risk screening was performed on patients who had to stay for over 72 h,and the ED nurses and physicians responsible for the screening were not involved in deciding nutrition support therapies for these patients;bedside ultrasound was conducted by a senior resident to measure the RF thickness.All necessary data were collected for statistical analysis:(1)the feature selection approach for machine learning,namely the best subsets regression,was applied to the determination of possible predictors and the building of an appropriate predictive model;(2)univariate and multivariate analyses were performed on the patient information to identify those factors that may affect the patients' treatment outcomes;following that,all variables were subject to correlation and multicollinearity analyses and a logistic regression model was built using the glm(),with the best set of input variables obtained from the preceding analyses as the independent variables(X)and the clinical prognosis as the dependent variable(Y);?,SE,Wald 2,and p-value were displayed using the summary(),while the odds ratio(OR)and the 95%confidence interval(95%CI)were determined by the exp().The variance inflation factor(VIF)was calculated using the vif()to assess the multicollinearity of the influencing factors of diagnosis and prognosis.(1)Discrimination:A receiver operating characteristic(ROC)curve was drawn using the plot.roc()and the area under the curve(AUC)was determined,as well as the best critical value(i.e.,the cutoff point)of predicting probability;the AUC values obtained from the two models were compared using the roc.test();(2)Calibration:A calibration curve was drawn using the val.prob();(3)Fitting:The Hosmer-Lemeshow(HL)test was performed using the gof().ResultsPart 1From December 1,2019 to February 20,2020,we collected 320 questionnaires from 12 provinces,cities,and autonomous administrative regions,including 300 complete questionnaires.Among the 300 ED nurses and physicians,100(33.3%)were attending physicians and 55(18.3%)were supervisor nurses.It was found that most ED nurses and physicians,particularly attending physicians,considered it necessary to provide nutrition support for ED patients.When patients had to stay for>72 h,58.33%of the ED nurses and physicians provided the patient with nutrition support and 78.6%of junior-level physicians had little knowledge of nutrition risk screening.Respondents largely agreed that patients who remained in ED observation units should be provided with nutrition risk screening,but found it difficult to implement the screening for the following reasons:78.33%of respondents thought that there was not sufficient time for nutrition risk screening because of ED patient flow;70%expressed that the prevailing assessment methods were too complicated to use in a short time and some laboratory tests were not available;and 28.33%considered it unnecessary to perform nutrition risk screening in EDs as it was the task of specialist physicians.Only 10%of the participants were familiar with the major types of malnutrition,and senior-level ED nurses and physicians outperformed those at lower levels in terms of grasping relevant concepts of malnutrition.A large number of the ED nurses and physicians could specify relevant diseases and symptoms entailing a high risk of malnutrition,including chronic infections(95%,n=285),acute pancreatitis(88.33%,n=265),hyperthyroidism(88.33%,n=265),chronic obstructive pulmonary disease(76.66%,n=230),and other diseases(e.g.,chronic cardiac insufficiency,chronic renal insufficiency).All participants had varied degrees of knowledge of nutrition risk screening or clinical nutrition and showed a lack of clinical nutrition knowledge,and there was no significant difference among ED nurses and physicians at different levels(x=3.99±2.34,p=0.057).All respondents,regardless of their position,believed that nutrition risk screening was closely related to their practice and expressed an interest in learning about clinical nutrition knowledge,especially the attending physicians(x=6.25 ± 2.23,p<0.001).Part 2Since July 1,2019 to December 31,2019,Peking union medical college hospital emergency of 1172 patients,of whom 751 ruled out because of emergency is not more than 48 hours,100 is in a state of evil fluid to pledge(late),70 patients were excluded by under 18,45 patients with pregnancy,40 patients except for leg or muscle skin infections,and eventually into the group of 66 patients,including 40 male and female,26,the average age of 68.Based on the NUTRIC score,patients with a NUTRIC score?5 were defined as high nutritional risk group and conversely low nutritional risk group.The comparison of basic clinical data between the low-nutrition risk group and the high-risk group showed no statistically significant difference between the two groups(P>,0.05).The differences in age,ALB,APACHE II,SOFA and nrs-2002 were statistically significant(P<0.05),while the differences in other indicators were not statistically significant(P>,0.05).Age,ALB,APACHE II and SOFA scores were included in logistic regression analysis,and the results showed that APACHE ? scores were statistically significant in multivariate logistic analysis(P<0.05).In other words,the higher the APACHE II score,the higher the possibility of high nutritional risk in patients(OR=1.659).Wilcoxon rank sum test was used between the two groups at each time point of muscle thickness,and the difference in the thickness of the rectus femoris muscle on the second day(d2)after admission was statistically significant(P<0.05).There was no statistically significant difference in the thickness of the rectus femoris on the day of admission(d0)and the day after admission(d3)(P>0.05).The generalized estimate equation(GEE)between the time points before and after the low-risk and high-risk treatment was compared,and the difference was statistically significant(P<0.05).In addition,pantwise comparison was conducted after the event:the changes in muscle thickness of low-risk and high-risk muscles were d0 greater than d2 and d3,respectively,and d2 greater than d3,which were statistically significant.In other words,the thickness of the rectus femoris muscle decreased gradually in both groups regardless of the nutritional risk,but no significant statistical difference was observed between the two groups(P>,0.05).All the variables into the most comprehensive subset regression,based on the bayesian information criterion(BIC)the number of the optimal combination,BIC model,the optimal score most hours,using adjr2 biggest=0.75 at the same time,the optimal combination for the six factors,the optimal combination is age,gender,mechanical ventilation or not,nutritional support,muscle thickness change value,SOFA score and APACHEII scores is the best combination forecast model is set up.Variables found in the optimal subset regression analysis were included in the Logistic regression model,and the results showed as follows.The regression equation was obtained according to the results of the multivariate regression analysis:P=1/(1+e^(y))Y=30.89383*mechanical ventilation+45.43792*muscle thickness change value+apache*5.1587-age*0.8713994-12.65335*gender+28.43666*nutritional support-221.9343 P is the probability of high nutritional risk.According to the calibration diagram of the established regression model(hosmer-lemeshow test P=0.119),it is proved that the calibration test has passed.According to the established regression model,the 2 value of GOF test is 1.675,and the P value of GOF test is 0.989,P=0.119,indicating a good fitting between the predicted value and the actual observed data.Finally,a more intuitive presentation of the nutrition scoring system is made by using the Alignment Diagram,also known as the Nomogram.ConclusionsPart 1In the present study,differences observed among ED nurses and physicians at different levels indicate that educational programs designed for ED nurses and physicians,especially those at the junior level,require additional training to improve knowledge of clinical nutrition.As more educational programs and activities regarding clinical nutrition are developed,ED nurses and physicians are expected to provide quality nutrition support therapy.In general,the available literature and data obtained in the present study indicate that malnutrition is associated with poor prognosis.Identifying risk factors likely to affect patients' predictable post-discharge clinical outcomes may facilitate the administration of pertinent interventions.Further studies supporting the feasibility of our scoring model may help ED nurses and physicians to assess a patient's nutritional status and perform nutrition risk screening in a highly convenient way,thereby providing ED patients with active intervention.Malnutrition is likely to be improved and rectified over time if enhanced nutrition support is available after discharge.Part 2In this study,we found that the thickness of rectus femoris muscle changed significantly in patients with high nutritional risk.In this study,a nutrition risk scoring system was established in accordance with the actual situation of emergency clinic,so as to facilitate emergency medical staff to identify patients at risk of malnutrition more easily. |
PDF Full Text Request