SEARCH

SEARCH BY CITATION

Keywords:

  • China;
  • feeding difficulty;
  • Mokken scaling;
  • nursing;
  • psychometrics;
  • reliability;
  • the Edinburgh Feeding Evaluation in Dementia scale;
  • validity

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Aims and objectives.  To translate the Edinburgh Feeding Evaluation in Dementia scale (EdFED) into simplified Chinese and to comprehensively evaluate its reliability and validity.

Background.  The EdFED, the only validated instrument at present for assessing feeding difficulty in older people with dementia, is available in the original English and traditional Chinese versions, but not available in simplified Chinese. The traditional Chinese version may not be applicable in Mainland China because of linguistic and cultural differences.

Design.  Survey.

Methods.  The scale was translated into simplified Chinese by the cross-culture translation method, and 102 participants with dementia were assessed. Data were collected by comprehensive methods and analysed by correlation, Mokken scaling and exploratory factor analysis.

Results.  Reliability and validity were demonstrated for the scale, and a strong and reliable Mokken scale was formed by six items. A three-factor structure was illustrated by exploratory factor analysis, and construct validity was further demonstrated by good convergent and discriminant validity.

Conclusions.  The simplified Chinese version shows good reliability and validity and can be applicable to measure feeding difficulty in people with dementia in Mainland China and other Chinese cultural groups. More work is required on Mokken scaling, and a confirmatory factor analysis is needed to confirm the three-factor structure.

Relevance to clinical practice.  The validation of Ch-EdFED has provided a validated instrument for measuring feeding difficulty in people with dementia in Chinese culture; thus, early recognition of feeding difficulty in older people with dementia can be achieved and proper interventions could be designed. Moreover, with the availability of the three different validated versions of the EdFED, research into cross-cultural comparisons could be conducted.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

As the trend of ageing populations continues, the number of older people is increasing with the concomitant rise in the number of people with dementia. The statistics of the latest national demographic survey reveals that the prevalence of dementia among people aged 60 and above has reached 4·0%; that is, up to 6·69 million Chinese people aged 60 and above now suffer from dementia (National Bureau of Statistics of China 2010). Owing to dementia-related mental and cognitive impairments and physical disabilities, people with dementia are increasingly inclined to suffer from difficulties in feeding, such as eating inappropriate amounts of food, refusing to eat when definitely not satiated, failing to recognise or use utensils correctly and difficulties with chewing or swallowing (Watson 1993). People with dementia suffering from feeding difficulty for a long time may experience negative outcomes, such as malnutrition, weight loss and dehydration, detrimentally affecting their physical health and thus decreasing quality of life (Watson 1997). Therefore, early recognition of and intervention for feeding difficulty in people with dementia by healthcare providers (family members, nursing staff, etc) are of fundamental importance to help to decrease the occurrence of such negative effects and to increase quality of individual life. Thus, a reliable and valid instrument for assessing feeding difficulty in people with dementia is required.

Background

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

The overseas study of development of instruments for evaluating feeding difficulty in people with dementia has been extensively conducted since the early 1990s, in contrast to little research or investigation in China. For example, several related instruments are available overseas for research or clinical practice, including Edinburgh Feeding Evaluation in Dementia scale (EdFED) and Feeding Abilities Assessment and Feeding Behavior Inventory (Aselage 2010). Among these, the EdFED has been considered as the only validated instrument for evaluating feeding difficulty in older people with dementia (Stockdell & Amella 2008, Aselage 2010).

The EdFED, developed by Watson (1994a,b), was reported to have good validity and reliability in long-term care settings (Watson & Deary 1994, 1997, Watson 1996). The scale consists of 11 items, including behaviours of feeding difficulty (e.g. refusing to eat, turning head away while being fed, spitting out food), nursing intervention (e.g. supervision, physical help) and patient passivity during meals (e.g. spillage, leftover). It was reported to be able to assess not only the current level of feeding difficulty, but also the change of level of feeding difficulty before and after nursing interventions. Considered as a valid instrument for nursing practice, it has now been widely used in English-speaking countries, such as England and USA (Amella et al. 2007a,b). It was introduced into Taiwan in 2003 (Lin & Chang 2003), and the traditional Chinese version of EdFED (C-EdFED-Q) was also extensively tested in clinical settings on aspects of adequacy of individual items, construct validity and external validity. However, inter- and intra-rater reliability and convergent and discriminant validity were not established (Lin et al. 2008).

While there is a clear need for a valid instrument to assess feeding difficulty in people with dementia in Mainland China, the C-EdFED-Q cannot be used directly because of geographical, linguistic and cultural differences between Taiwan and Mainland China. Instead of adapting the C-EdFED-Q directly to Mainland China, extensive psychometric testing was required and this is reported here.

Methods

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Design

A survey design was applied, and a series of tools, including Demographic Information Questionnaire, Global Deterioration Scale, Ch-EdFED, the Nutritional Risk Screening 2002 and the Clinical Nursing Swallowing Assessment Tool (CNSAT), were used to collect data through direct observation, informal interview and review of the medical records.

Cross-cultural translation

Systematic translation process

Consent was obtained from Roger Watson, the original designer, for the translation of the EdFED. Lee’s systematic translation process was applied for cross-cultural translation (Lee et al. 2009). First, a native Chinese bilingual doctoral nursing graduate made the forward translation of the EdFED into simplified Chinese. Another qualified bilingual translator who was also a native Chinese doctoral nursing graduate performed a backward translation of the simplified Chinese version into English. Then, the two translators compared the consistency of meaning between the original EdFED and the backward translation and concluded that the main discrepancies available were verb choices and tense. Modification of the forward translation was accordingly made to eliminate all discrepancies. Lastly, a third qualified translator made a second backward translation of the modified forward translation into English. Comparison between the original EdFED and the second back-translated version was made again, and no discrepancy was found. Finally, the simplified Chinese version of EdFED (Ch-EdFED) was obtained, as is shown in Table 1.

Table 1. Simplified Chinese version of EdFED (Ch-EdFED)Thumbnail image of
Committee review

A panel of five experts in rehabilitative, gerontological, psychiatric or ageing health areas was established to examine the items’ fluency, readability and comprehensibility and to evaluate the content validity of Ch-EdFED by grading the relevance of the 11 items on a four-point Likert scale (1 = not relevant; 2 = somewhat relevant; 3 = quite relevant; 4 = highly relevant). A content validity index (CVI) was calculated by summing the percentage of agreement of all items that were rated ‘3’ or ‘4’ by all the experts (Polit & Beck 2006).

Pretest study

Participants (n = 20) were selected in a long-term care centre and evaluated using the Ch-EdFED by two onsite health care nursing staff. Inclusion criteria were that participants should (1) meet the diagnostic criteria of dementia of ICD-10 and DSM-IV; (2) be supported with oral feeding, with or without assistance; (3) be conscious and able to cooperate; and (4) be able to give informed consent or have it provided by proxy by an appropriate family member. Demographic data, including age and sex, were also collected. The 20 participants had an average age of 77·1 years (SD 10·19), ranging from 49–91 years, and 50·0% were female.

The Ch-EdFED was reported to have good understandability and readability. The item-to-total correlation coefficients ranged from 0·38–0·79 (p < 0·001), revealing high correlation with the overall scale. Cronbach’s α for all 11 items of Ch-EdFED was 0·87, demonstrating good internal consistency. These results suggested that a study of the Ch-EdFED in a larger scale of participants was worthwhile.

Participants

Participants (n = 102) who met the inclusion criteria were selected from two long-term care centres in Jinan, Shandong Province, China. Inclusion criteria were the same as in the pretest study. Watson’s previous study was conducted originally with older people with dementia who were aged 65 and above. By reviewing the literature from which the items of the EdFED originated, it is found that the majority of 11 items originated from feeding behaviours of not only those with dementia who were aged 65 years and above, but also those aged below 65 years (Miller 1971, Davies & Snaith 1980, Norberg et al. 1980a,b, Athlin et al. 1989). The present study is intended to be conducted on people with dementia of all ages, including those below 65 years of age.

Data collection

The study was conducted during June–December, 2011. Two research assistants were trained by the investigator for data collection. Contents of training included a basic knowledge of dementia and feeding difficulty, introduction and application of tools involved and basic skills for communicating with participants, family members and nursing staff. Once the training was complete, the two assistants began to collect data independently in each centre.

Data were collected by methods of direct mealtime observation; informal interview with participants, nursing staff or family members; and review of the medical records. Observation of mealtime behaviours of participants was performed for the consecutive lunch and dinner of the same day before evaluation. Demographic data and dementia-related data were collected by personal interviews and record reviews. Twenty of the 102 participants were selected by random sampling and evaluated by the two research assistants, respectively, for evaluation of inter-rater reliability. After the original evaluation by the Ch-EdFED scale, 20 participants were also randomly selected among the 102 participants and re-evaluated 14–15 days later by the same research assistant to assess intra-rater reliability. In this study, a total of five tools are applied as follows.

Demographic Information Questionnaire

The questionnaire was designed to collect demographic and dementia-related information of the participants, including data of age, sex, education, dementia pattern and duration of dementia.

Global Deterioration Scale

Global Deterioration Scale (GDS) was used to evaluate the severity of dementia. The GDS had seven grades, with 1–3 graded as mild dementia, 4–5 graded as moderate dementia and 6–7 graded as severe dementia (Reisberg et al. 1982). The Chinese version of GDS has been validated and has now been widely used in institutional and clinical settings.

The simplified Chinese version of EdFED

The Ch-EdFED, consisting of 11 items, was used to assess the level of feeding difficulty in people with dementia (Watson 1994a,b).

The Nutritional Risk Screening 2002

The Nutritional Risk Screening 2002 (NRS-2002) was used here to measure nutritional risks of demented people (Kondrup et al. 2003). The Chinese version of NRS-2002 has excellent validity and reliability and has been widely used.

The Clinical Nursing Swallowing Assessment Tool

The Clinical Nursing Swallowing Assessment Tool was a Chinese instrument developed by Bao-Yan Huang and was used specifically by nurses to assess the swallowing function of patients (Huang et al. 2007a). It has been clinically applied and has good reliability and validity (Huang et al. 2007b).

Ethical permission

The study was approved by the Ethical Board of Shandong University School of Nursing. Written informed consent was signed by participants or their family members (e.g. spouse, adult children), and thus they were able to understand the aims and implications of the study.

Data analysis

Data were entered into SPSS for Windows, version 16·0 (SPSS Inc., Chicago, IL, USA) for statistical analysis. Data were analysed mainly by methods of correlation and exploratory factor analysis. Specifically, CVI was used to evaluate the content validity of the Ch-EdFED, which was demonstrated above, and Cronbach’s α to determine the internal consistency reliability. Data for total EdFED scores of 20 participants collected, respectively, by two research assistants were analysed by paired-samples t-test and then by Pearson’s correlation for inter-rater reliability. Similarly, data for total EdFED scores of 20 participants collected by the same research assistant twice with an interval of 14–15 days were also analysed by paired-samples t-test and Pearson’s correlation for intra-rater reliability.

Pearson coefficients were also applied to estimate the correlation of the Ch-EdFED with NRS-2002 and CNSAT, respectively, for its convergent and discriminant validity. Principal component analysis, a method of factor analysis (Watson & Thompson 2006), was applied to explore the factor structure of the Ch-EdFED, and analysis of rotation of factors to determine the loadings of variables on factors.

Data were imported into the Mokken Scaling Analysis for Polytomous Items (msp) for Windows version 5.0 software (iec ProGAMMA, Groningen, The Netherlands), and Mokken scaling was applied to explore whether the items of the Ch-EdFED formed a Mokken scale – a unidimensionally hierarchical scale – of which a score on any single item could reveal a certain level of ‘difficulty’ on specific measurements. The data analysis was as described in previous studies of the EdFED scale (Watson 1996, Lin et al.2008). In addition, the data were entered into the R statistical programme (van der Ark 2007) for analysis of invariant item ordering (IIO) – the extent to which all participants respond to all items in the same way (Ligtvoet et al. 2010) – within the Mokken scaled items.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

The 102 participants had an average age of 77·8 years (SD 10·35), with a range of 49–99 years, and 58·8% were female. The average duration of dementia was 41·2 months (SD, 31·5), with a range of 1–180 months; 60·8% were severely demented and 39·2% mildly or moderately demented, as is shown in Table 2.

Table 2. Demographic information of participants (n = 102)
ItemCategory n (%)
Age (years)≥6591 (89·2)
45–6411 (10·8)
SexMale42 (41·2)
Female60 (58·8)
Education (years)≤145 (44·1)
≤525 (24·5)
≤916 (15·7)
>916 (15·7)
Dementia patternAD51 (50·0)
VD31 (30·4)
Others20 (19·6)
Dementia severityMild10 (9·8)
Moderate30 (29·4)
Severe62 (60·8)

Item analysis

Table 3 shows most of the correlation coefficients were larger than 0·30, and the majority of individual items were significantly correlated with one another. The item-to-total correlation coefficients ranged from 0·47–0·87 (p < 0·001), indicating that all the items are measuring the same content as the overall scale.

Table 3. Correlation matrix of Ch-EdFED items (n = 102)
ItemCh-EdFED1234567891011
  1. **p < 0·001; *p ≤ 0·037.

Ch-EdFED1           
10·730**1          
20·725**0·947**1         
30·468**0·1610·1501        
40·520**0·216* (p = 0·029)0·1780·512**1       
50·773**0·362**0·357**0·262**0·310**1      
60·820**0·435**0·432**0·1900·340**0·807**1     
70·869**0·469**0·466**0·291**0·324**0·858**0·866**1    
80·830**0·453**0·449**0·335**0·386**0·718**0·886**0·824**1   
90·773**0·380**0·377**0·278**0·274**0·675**0·747**0·825**0·787**1  
100·785**0·474**0·471**0·275**0·207* (p = 0·037)0·707**0·743**0·811**0·725**0·716**1 
110·748**0·787**0·790**0·322**0·321**0·403**0·398**0·476**0·435**0·416**0·484**1

As is shown in Table 4, only six of 11 items were retained as a Mokken scale. A hierarchy of the six items was also demonstrated below in terms of feeding difficulty from a general refusing to eat to leaving mouth open allowing food to drop out, according to their mean values. Specifically, item 5 has the greatest mean value of 0·46, indicating the lowest level of feeding difficulty, while item 9 has the lowest mean value of 0·33, indicating the highest level of feeding difficulty.

Table 4. Mokken scaling of simplified Chinese version of EdFED (Ch-EdFED) (n = 102)Thumbnail image of

The scale and individual coefficients (H), indicating whether a scale is a strong Mokken scale, and the reliability coefficient (ρ), indicating whether a scale is a reliable Mokken scale, are also shown. On the basis that the overall scale and individual coefficients (H) are to be greater than 0·50 to indicate strong scalability and that the reliability coefficient (ρ) is to be >0·70 to indicate reliability, the six-item scale is demonstrated as a very strong, statistically significant and reliable Mokken scale. While IIO, the extent to which all individuals respond to the same set of items in the same way, was evident in the items retained in the Mokken scale, the accuracy of the IIO was very low as indicated by the low value of HT, which should be >0·40 for acceptable IIO.

Reliability and validity

In determining the internal consistency, Cronbach’s α for 11 items of the Ch-EdFED was 0·91. Both sets of data revealed no significant difference by paired-samples t-test (p = 0·690, p = 0·205), and Pearson’s coefficients were respectively 0·81 (p < 0·001) and 0·89 (p < 0·001). The expert panel rated all 11 items of the Ch-EdFED, and the ratings were all ‘3’ or ‘4’; that is, all items yielded 100% agreement among the five experts. Thus, CVI calculated for the scale’s content validity was 1·00. As to the convergent and discriminant validity of the Ch-EdFED, the correlation coefficients between the Ch-EdFED and NRS-2002 and CNSAT, respectively, were 0·37 (p < 0·001) and 0·20 (p = 0·051).

Factorial validity of Ch-EdFED was determined by principal component analysis. The Kaiser–Meyer–Olkin (KMO), indicating the appropriateness of factor analysis for the data, was 0·86, and the approximate chi-square for Bartlett’s test was 1090 (p < 0·001), which revealed a statistically significant correlation among items and indicated that the data were appropriate for principal component analysis. As is shown in Table 5, three principal components were extracted with eigenvalues of 4·743, 2·778 and 1·637, respectively, and a total of 83·3% of overall variance was explained. An oblique rotation analysis was used to determine the loadings of variables on factors. As Table 6 shows, the derived solution obtained suggests that variables 5, 6, 7, 8, 9 and 10 load on factor 1; 1, 2 and 11 load on factor 2; and 3 and 4 load on factor 3, and the loadings of variables on its own factor were all greater than 0·70.

Table 5. Rotation sums of squared loadings of Ch-EdFED (n = 102)
ComponentEigen values (λ)% of varianceCumulative %
14·74343·12243·122
22·77825·25668·378
31·63714·88483·262
Table 6. Rotated component matrix of Ch-EdFED (n = 102)
ItemCommunalities (extraction)Component
123
  1. The loadings of items on its own factor that were greater than 0·70 are indicated in bold.

10·9380·253 0·934 0·052
20·9440·253 0·938 0·024
30·7640·1440·091 0·857
40·7420·1930·107 0·833
50·780 0·857 0·1490·152
60·876 0·908 0·1990·105
70·917 0·912 0·2480·152
80·839 0·855 0·2270·236
90·766 0·845 0·1770·144
100·757 0·811 0·3040·078
110·8360·248 0·841 0·258

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Item analysis

All item-to-total correlation coefficients reveal high correlation with the total scale and Cronbach’s α reveals good internal consistency of the Ch-EdFED. While all 11 items were entered for analysis, only six were retained as a Mokken scale, which was further demonstrated to be strong and reliable. Results obtained in this study were quite similar to those of the previous studies, apart from the ordering of item 6 (Watson 1996, Lin et al. 2008). The reason why only five of the six were in the same order is unknown but possibly related to the smaller sample size. The fact that IIO was very low does not detract from the use of the EdFED and merely indicates that, while it is good for ordering people in terms of feeding difficulty on the basis of the mean scores on items, not all participants responded to the items in the same way. IIO has been demonstrated in another study (Watson et al. 2012) but, while accuracy was higher than in the present study, it was <0·40. The lower accuracy in the present study may arise as a result of the larger range of dementia severity than in the other studies, and this may also account for the slightly different ordering of items here.

The ordering of items by mean value indicates that the appearance of only ‘refusing to eat’ would suggests the lowest level of feeding difficulty, while the demonstration of the most passive behaviour – leaving mouth open – indicates the most severe level of difficulty in feeding. In other words, people with dementia will probably suffer from feeding difficulty as the dementia progresses, especially in the final stage of dementia. Those who begin to ‘refuse to eat’ may have the least difficulty in feeding, and later on more passive behaviours, such as ‘refusing to open mouth’, ‘spitting’, ‘refusing to swallow’ and ‘turning head away’, may be observed as dementia progresses, which indicates that the level of feeding difficulty has gradually got worse and the greatest possible level of difficulty is manifested as the most negative feeding behaviour – leaving mouth open – is finally demonstrated.

Validity and reliability

Pearson correlation suggests moderate inter- (r = 0·81, p < 0·001) and intra-rater (r = 0·89, p < 0·001) reliability of the Ch-EdFED. The findings are exactly the same as those in previous studies, which reveals that different versions of the EdFED scale are equivalently reliable in English or Chinese cultural backgrounds.

Content validity (CV = 1·00) is excellent, which reveals that all items are measuring the same construct as the overall scale. As to the exploratory factor analysis of this study, three principal components were extracted, precisely the same as the findings of Watson and Deary (1994). These three factors were previously suggested as patient obstinacy or passivity, nursing intervention and indicator of feeding difficulty. Furthermore, the variables loaded on each factor, all with high loadings, were exactly the same as those obtained by Watson and Deary (1994). Thus, on the basis of this study, the Ch-EdFED reveals good factorial validity.

NRS-2002 and CNSAT, measuring similar contents, respectively, as those instruments used by Watson et al. (2001), were used to examine convergent and discriminant validity of the Ch-EdFED in this study. Convergent validity of the Ch-EdFED was suggested through its positive and statistically significant correlation with NRS-2002 (r = 0·37, p < 0·001), which was used to measure nutritional risks. Discriminant validity was demonstrated as it did not correlate statistically significantly (r = 0·20, p = 0·051), though positively, with CNSAT, which was used as a measurement of swallowing function of patients, different from that of the Ch-EdFED.

Significance of study

Since the early 1990s, the study of assessment of feeding difficulty in older people with dementia has been conducted extensively overseas, resulting with many research findings and instruments. However, the EdFED is considered as the only validated instrument for evaluating feeding difficulty in older people with dementia (Amella & Lawrence 2008). Recently, Lin introduced the EdFED scale into Taiwan and developed the C-EdFED-Q, with good validity and reliability established (Lin & Chang 2003, Lin et al. 2008). However, Lin’s study failed to investigate the inter- and intra-rater reliability or convergent and discriminant validity, which are considered indispensable for the establishment of validity and reliability of an instrument. The C-EdFED-Q also cannot be directly used in Mainland China because of geographical, linguistic and cultural differences. Thus, this study was conducted to establish a simplified Chinese version of the EdFED, with systematic cross-cultural translation and validation of excellent validity and reliability.

Moreover, clinical significance was further demonstrated in the development of the Ch-EdFED by the extended range of participants in age the survey was conducted with. The Ch-EdFED was also applied to people with dementia who were aged below 65 years, the age group that was not included in the original studies. The overall data collected from 102 participants, nevertheless, reveal good results. Analysis of data collected exclusively from participants aged below 65 years was not carried out separately as it was too small a sample for valid statistical analysis.

The majority of findings in this study are fundamentally equivalent to those reported by both Watson and Lin, which suggests that data collected with different versions of the EdFED scale in accordingly different cultural backgrounds can be comparable once it is obtained with the same design and method of study, and also exchange of data across different cultural settings can be permitted. Thus, research can be conducted among disparate ethnic groups among Chinese and English worldwide for advanced theoretic and practical significance in this aspect.

Limitations

The study was mainly conducted with people with dementia in local long-term care centres, which may limit the generalisation of the findings to other demented population, such as residents with dementia in communities and inpatients with dementia in hospitals. Confirmatory factor analysis, which requires data of a much larger sample, was not conducted owing to limitations of time and human resources.

Conclusion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

In this study, the adequacy of individual items, the inter- and intra-rater reliability, internal consistency reliability, content and construct validity, and convergent and discriminant validity of the Ch-EdFED were all fully demonstrated. It is substantiated from the findings that the Ch-EdFED is a practical, reliable and valid instrument. However, as is widely known, the validation of any instrument requires enormous endeavour, which is definitely considered as a continuing and endless process. Thus, further validation of the Ch-EdFED may be required.

Acknowledgements

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

The study was supported by a grant from Shandong University. We gratefully acknowledge Hu-die Liang and Xiao-jing Li for data collection, and Li-chan Lin for sharing experience of the validation of C-EdFED-Q.

Contributions

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References

Study design: F-LL, WL; data collection and analysis: WL, RW and manuscript preparation: F-LL, WL, RW.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Background
  5. Methods
  6. Results
  7. Discussion
  8. Conclusion
  9. Acknowledgements
  10. Contributions
  11. Conflict of interest
  12. References
  • Amella EJ & Lawrence JF (2007a) Try this: Eating and Feeding Issues in Older Adults with Dementia: Part I: Assessment. Available at: http://consultgerirn.org/uploads/File/trythis/try_this_d11_1.pdf (accessed 25 December 2010).
  • Amella EJ & Lawrence JF (2007b) Try this: Eating and Feeding Issues in Older Adults with Dementia: Part II: Interventions. Available at: http://consultgerirn.org/uploads/File/trythis/issue11_2.pdf (accessed 25 December 2010).
  • van der Ark LA (2007) Mokken scale analysis in R. Journal of Statistical Software20, 119.
  • Aselage MB (2010) Measuring mealtime difficulties: eating, feeding and meal behaviours in older adults with dementia. Journal of Clinical Nursing19, 621631.
  • Athlin E, Norberg A, Axelsson K, Möller A & Nordström G (1989) Aberrant eating behaviour in elderly Parkinsonian patients with and without dementia: analysis of video recorded meals. Research in Nursing & Health12, 4151.
  • Davies AD & Snaith PA (1980) Mealtime problems in a continuing care hospital for the elderly. Age and Ageing9, 100105.
  • Huang BY, Shen N, Li SL, Wu XJ & Liang T (2007a) Development of an instrument to assess swallowing function in patients with stroke. Chinese Journal of Rehabilitative Theory and Practice13, 371272.
  • Huang BY, Shen N, Li SL, Wu XJ & Liang T (2007b) Study of validity and reliability of Clinical Nursing Swallowing Assessment Tool. Chinese Journal of Nursing42, 127130.
  • Kondrup J, Allison SP, Elia M, Vellas B & Plauth M; Educational and Clinical Practice Committee & European Society of Parenteral and Enteral Nutrition (ESPEN) (2003) ESPEN Guidelines for nutrition screening 2002. Clinical Nutrition22, 415421.
  • Lee CC, Li D, Arai S & Puntillo K (2009) Ensuring cross-cultural equivalence in translation of research consents and clinical documents: a systematic process for translating English to Chinese. Journal of Transcultural Nursing20, 7782.
  • Ligtvoet R, van der Ark LA, te Marvelde JM & Sijtsma K (2010) Investigating an invariant item ordering for polytomously scored items. Educational and Psychological Measurement70, 578595.
  • Lin LC & Chang CC (2003) A Chinese Translation of the EdFED-Q and Assessment of Equivalence. Alzheimer Disease & Associated Disorders17, 230235.
  • Lin LC, Watson R, Lee YC, Chou YC & Wu SC (2008) Edinburgh Feeding Evaluation in Dementia (EdFED) scale: cross-cultural validation of the Chinese version. Journal of Advanced Nursing62, 116123.
  • Miller MB (1971) Unresolved feeding and nutrition problems of the chronically ill aged. Gerontologist11, 329336.
  • National Bureau of Statistics of China (2010) 2009 National Economic and Social Development Statistics of China. Available at: http://www.stats.gov.cn/tjgb/ndtjgb/qgndtjgb/t20100225_402622945.htm (accessed 25 December 2010).
  • Norberg A, Norberg B & Bexell G (1980a) Ethical problems in feeding patients with advanced dementia. British Medical Journal281, 847848.
  • Norberg A, Norberg B, Gippert H & Bexell G (1980b) Ethical conflicts in long-term care of the aged: nutritional problems and the patient–care worker relationship. British Medical Journal280, 377378.
  • Polit DF & Beck CT (2006) The content validity index: are you sure you know what’s being reported? Critique and recommendationsResearch in Nursing & Health29, 489497.
  • Reisberg B, Ferris SH & Crook T (1982) The global deterioration scale for assessment of primary degenerative dementia. American Journal of Psychiatry139, 1136.
  • Stockdell R & Amella EJ (2008) The Edinburgh Feeding Evaluation in Dementia Scale: determining how much help people with dementia need at mealtime. American Journal of Nursing108, 4654.
  • Watson R (1993) Measuring feeding difficulty in patients with dementia: perspectives and problems. Journal of Advanced Nursing18, 2531.
  • Watson R (1994a) Measuring feeding difficulty in patients with dementia: developing a scale. Journal of Advanced Nursing19, 257263.
  • Watson R (1994b) Measuring feeding difficulty in patients with dementia: replication and validation of the EdFED Scale #1. Journal of Advanced Nursing19, 850855.
  • Watson R (1996) The Mokken scaling procedure (MSP) applied to the measurement of feeding difficulty in elderly people with dementia. International Journal of Nursing Studies33, 385393.
  • Watson R (1997) Undernutrition, weight loss and feeding difficulty in elderly patients with dementia: a nursing perspective. Reviews in Clinical Gerontology7, 317326.
  • Watson R & Deary IJ (1994) Measuring feeding difficulty in elderly patients with dementia: multivariate analysis of feeding problems, nursing intervention and indicators of feeding difficulty. Journal of Advanced Nursing20, 283287.
  • Watson R & Deary IJ (1997) Feeding difficulty in elderly patients with dementia: confirmatory factor analysis. International Journal of Nursing Studies34, 405414.
  • Watson R & Thompson DR (2006) Use of factor analysis in Journal of Advanced Nursing: a literature review. Journal of Advanced Nursing55, 3303441.
  • Watson R, Green SM & Legg L (2001) The edinburgh feeding evaluation in dementia Scale#2 (EdFED#2): convergent and discriminant validity. Clinical Effectiveness in Nursing5, 4446.
  • Watson R, van der Ark LA, Lin L-C, Fieo R, Deary IJ & Meijer RR (2012) Item response theory: how Mokken scaling can be used in clinical practice. Journal of Clinical Nursing21, 27362746.

The Journal of Clinical Nursing (JCN) is an international, peer reviewed journal that aims to promote a high standard of clinically related scholarship which supports the practice and discipline of nursing.

For further information and full author guidelines, please visit JCN on the Wiley Online Library website: http://wileyonlinelibrary.com/journal/jocn

Reasons to submit your paper to JCN:

High-impact forum: one of the world’s most cited nursing journals, with an impact factor of 1·316 – ranked 21/101 (Nursing (Social Science)) and 25/103 Nursing (Science) in the 2012 Journal Citation Reports® (Thomson Reuters, 2012)

One of the most read nursing journals in the world: over 1·9 million full text accesses in 2011 and accessible in over 8000 libraries worldwide (including over 3500 in developing countries with free or low cost access).

Early View: fully citable online publication ahead of inclusion in an issue.

Fast and easy online submission: online submission at http://mc.manuscriptcentral.com/jcnur.

Positive publishing experience: rapid double-blind peer review with constructive feedback.

Online Open: the option to make your article freely and openly accessible to non-subscribers upon publication in Wiley Online Library, as well as the option to deposit the article in your preferred archive.