| Objective:Feeding intolerance is a main reason for critically ill patients who can not reach their target nutrient intake during enteral nutrition,leading to a poor nutrition and clinical outcomes.Therefore,the early prevention of feeding intolerance has a great significance in promoting effective implementation of enteral nutrition and Reducing the incidence of feeding intolerance.In order to help clinical medical staff screen the risk of feeding intolerance in the early phrase,we firstly analyses the likely risk factors of feeding intolerance,on the basis of what builds a tool that assess enteral feeding intolerance risk with a series of scientific scale development principles then,which can provide evidence for target measures,reduce the incidence of feeding intolerance,and promot the quality and level of enteral nutrition implementation further.Methods:There are four parts in this study:?Building the items pool.We refined the risk factors of feeding intolerance with the methods of literature review,semi-structured interview and group discussion,coming to being the items pool of scale.?Combining objective and subjective method in the item selection.We adopted the Delphi Method as the subjective method,in which we selected 30 experts from 16 hospitals in 6 provinces and 3 municipalities nationwide.Besides,we analysed the clinic data of 180 critically ill patients who received enteral nutrition with the "Enteral feeding intolerance risk assessment scale initial scoring criteria for critically ill patients",which developed accorrding to the results of Delphi Method.And objectively selected the items with frequency analytic approach and discrimination analysis approach.?Developing the sale.Based on the outcomes above,we confirming the item weight by factor analysis method,scale threshold by receiver operating characteristic curve,and the severity score of scale by percentiles method,through which finally developped a whole enteral feeding intolerance risk assessment scale.?Evaluating the reliability and validity of the scale.We verified the reliability and validity preliminary by testing the raters reliability,face validity and construct validity.Results:1.The forming of the items pool of enteral feeding intolerance risk factors.We refined 27 risk factors related to feeding intolerace in the literature review and 26 in the semi-structured interview,and finally confirmed 23 risk factors through group discussion on the basis of the results above.2.The selection of enteral feeding intolerance risk factors.2.1 The results of Delphi Method.The expert authority of two rounds of expert consultation were 0.85 and 0.9,the valid response rate were 90.9%and 100%,both effective questionnaire ratio were 100%,and each Kendall's concordance coefficient has statistically significant difference(P<0.01);During the first round,we removed 12 items,and added 9 items,and modified 2 items;During the second round,we removed 1 items and modified 1 items.2.2 Selecting items based on clinic data.There are 9 items' occurrence frequency less than 0.3 and 8 items'P-value less than 0.05 by discrimination.After synthesizing the two results,we finally removed these items below,"PaO2 less than 60mmHg","PH value less than 7.35","severe malnutrition" and"head of the bed less than 30 degree";and kept 19 items,including "age greater than or equal to 70","long-term bed","in one day after operation ","Abdominal operation","long-term fasting or total parenteral nutrition","acute or severe pancreatitis","gastrointestinal diseases or injury","Multiple Organ Dysfunction Syndrome","Acute Physiology and Chronic Health Evaluation II score greater than or equal to 20","Glasgow score Less than or Equal to 8","injury severity score greater than or equal to 25","serum albumin less than or equal to 25","serum potassium concentration less than or equal to 3.5","blood glucose concentration greater than or equal to 11.0 ","acute gastrointestional injury grade","mechanical ventilation","Calm analgesic","antacids"and "systemic or intestional infection".3.Confirmination of the item weight,and threshold and severity score of the scale.3.1 Results of items' weight.Factor analysis showed that KMO=0.540>0.5,?2=383.648,df=171,P=0.000<0.01,there were 8 common factors' eigenvalues greater than 1,and the accumulative contribution rate is 63.279%.each item's normalization weights is "age greater than or equal to 70(0.03307)","long-term bed(0.096398)","in one day after operation(0.034944)","Abdominal operation(0.061642)","long-term fasting or total parenteral nutrition(0.093002)","acute or severe pancreatitis(0.048374)","gastrointestinal diseases or injury(0.061642)","Multiple Organ Dysfunction Syndrome(0.026722)","Acute Physiology and Chronic Health Evaluation II score greater than or equal to 20(0.069271)","Glasgow score Less than or Equal to 8(0.055779)","injury severity score greater than or equal to 25(0.147704)","serum albumin less than or equal to 25(0.038004)","serum potassium concentration less than or equal to 3.5(0.050089)","blood glucose concentration greater than or equal to 11.0(0.056608)","acute gastrointestional injury grade(0.056666)","mechanical ventilation(0.057874)","Calm analgesic(0.043942)"," antacids(0.037408)"and "systemic or intestional infection(0.07774)".3.2 Confirmation of the threshold of scale.In this study,the area under the ROC curve is 0.95,P=0.000<0.01,the maximum value of Youden Index is 0.793,and the Sensitivity and specificity are 89.3%and 80%.We finally confirmed the threshold of scale at 17,which meaned when the evaluation score greater than 17,we would regarded it risky.3.3 Confirmation of the severity score of scale.The severity score of scale was divided into three level.First level,the score ranges between 17 and 21;Second level,the score ranges between 22 and 32;And the third level,the score exceeds 33.4.Evaluating of the reliability and validity of enteral feeding intolerance risk assessment scale.In the raters reliability,the spearman rank correlation is 0.814,P=0.000<0.01;The face validity showed that all the items reflected the measure concept of feeding intolerance risk;And in the factor anaysis,more than 84%of the items' maximum loads coefficients is greater than 0.4.Conclusion:The scale we developped has better sensitivity,specificity,reliability and validity. |
PDF Full Text Request