| Objective To translate the Edinburgh Feeding Evaluation in Dementia scale(EdFED) into Simplified Chinese and evaluate its reliability and validity. To evaluate the level of feeding difficulty in people with dementia, investigate the relative factors and further explore the influential factors, such as cognitive function, symptoms of psychosis, depression and agitation.Methods Convenience sampling method was used to collect 104 samples of demented people who met the inclusion criteria from two long-term care centers and one hospital in Jinan. Eight instruments were applied for data collection:the simplified Chinese version of EdFED, Activity of Daily Living Scale-Barthel Index(ADL-BI), Mini-Mental State Examination(MMSE), Global Deterioration Scale(GDS), California dementia behavioral questionnaire(CDBQ), Nutritional Risk Screening 2002(NRS-2002), Clinical Nursing Swallowing Assessment Tool(CNSAT) and the general information questionaire. All data were statistically analyzed by software package SPSS 16.0 for windows, and statistical methods such as Content Validity Index(CVI), Cronbach'a coefficient, correlation analysis, T-Test, analysis of variance, factor analysis, convergent and discriminate validity and multiple linear regression analysis were used. All data of EdFED were also analyzed by software package MSP(Mokken Scaling Procedure)5.0 for windows to explore whether the EdFED scale or certain items could retain as a Mokken scale.Results 1. The results of item analysis showed that the item-scale correlation coefficients were more than 0.40 and significant(r=0.50-0.87, P<0.001), and the Critical Ratio(CR) of all items were more than 0.30 and significant(CR=6.90～19.70, P<0.001). The results of reliability analysis revealed that the Cronbach'a was 0.90(P<0.01), the intra-rater reliability was 0.89(P<0.001) and the inter-rater reliability was 0.81(P<0.001). The results of validity analysis showed that the facial validity was good and the CVI was 1.00.3 components were extracted, the same as that of the original EdFED, and the total percentage of variance explained was 81.55%. Significant and positive correlation(r=0.35, P<0.01)was found between EdFED and NRS-2002, and insignificant correlation(r=0.20, P>0.05) between EdFED and CNSAT, the same as the results of convergent and discriminate validity of the original EdFED. The results of MSP analysis revealed that item 5～10 were retained as a Mokken scale, exactly the same as that of the original EdFED study. The overall Scalability coefficient(H) was 0.83, the individual items'H was between 0.78-0.87, and the Reliability p was 0.95. According to the Mean values, the retained 6 items were ranked as 5,7,8,10,6 and 9. The results were also the same as that of original study, except for the ranking of item 6.2. The average EdFED score of different kinds of demented people was between 1.92-9.67, and the overall EdFED score was 6.79±5.60, indicating mild to moderate level of feeding difficulty in people with dementia.3. The results of correlation analysis revealed that correlative factors of feeding difficulty were Body Mass Index, activity of daily living, cognitive function, duration of dementia, severity of dementia and behavioral and psychological symptoms, including symptoms of psychosis, depression and agitation. The results of multiple linear regression analysis demonstrated that feeding difficulty in people with dementia was influenced by such factors as cognitive function and symptoms of psychosis, depression and agitation.Conclusions1. The Simplified Chinese version of the Edinburgh Feeding Evaluation in Dementia scale is reliable and valid in measuring feeding difficulty in Chinese people with dementia and can be a useful tool for scientific research and clinical practice.2. People with dementia commonly suffer from different levels of feeding difficulty.3. The influential factors of feeding difficulty in people with dementia include cognitive function and symptoms of psychosis, depression and agitation. |
PDF Full Text Request