Shu-Qi Zhang，Hong-Mei Wang，Xiao-Yuan Yang，Wen-Qin Ye

aDepartment of Nursing，Emei Sanitarium of Chengdu Military Region，Emeishan，Sichuan 614200，China

bDepartment of Nursing，General Hospital of Chengdu Military Region，Chengdu，Sichuan 610083，China

cDepartment of Nursing，Changhai Hospital Affiliated to the Second Military Medical University，Shanghai 200433，China

Orighinal article

The Care Dependency Scale：psychometric testing of the Chinese version

Shu-Qi Zhanga，*，Hong-Mei Wangb，Xiao-Yuan Yangb，Wen-Qin Yec

aDepartment of Nursing，Emei Sanitarium of Chengdu Military Region，Emeishan，Sichuan 614200，China

bDepartment of Nursing，General Hospital of Chengdu Military Region，Chengdu，Sichuan 610083，China

cDepartment of Nursing，Changhai Hospital Affiliated to the Second Military Medical University，Shanghai 200433，China

A R T I C L E I N F O

Article history：

Received 17 January 2016

Received in revised form

27 February 2016

Accepted 19 March 2016

Available online 16 June 2016

Elderly

Geriatric care environment

Geriatrics

Nursing home

Nursing dependency

Objective：To determine the validity and reliability of the Chinese version of the Care Dependency Scale（CDS）for proxy assessments by nurses in Chinese nursing homes.The CDS is an excellent tool for measuring care dependency in many countries.The Chinese version of the CDS contributes additional data regarding the scale's use.

Methods：This was a cross-sectional study of 184 older adults；128 were re-evaluated after 2 weeks. Cronbach's alpha，the mean inter-item correlation，the inter-rater reliability and the test-retest reliability were used to analyze the reliability of the CDS.A factor analysis was carried out to find common dimensions between the CDS versions.

Results：The alpha coefficient of the CDS was 0.95.The subsequent inter-rater and test-retest reliabilities were analyzed in terms of the Kappa values，which were between 0.84 and 0.89；0.83 and 0.92，respectively.Three factors were identified in the factor analysis.

Conclusions：The CDS is a suitable tool for assessing care dependency among elderly Chinese residents in nursing homes.

©2016 Shanxi Medical Periodical Press.Publishing services by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http：//creativecommons.org/licenses/by-nc-nd/4.0/）.

1.Introduction

According to the Chinese National Bureau of Statistics，the number of people aged 60 years or older grew to 1.93 million in 2012，which accounts for 14.3%of the total population of China and is 0.59%more than in 2011.1As demographics change to include an increase in chronically ill adults，there is a simultaneous increase in care dependency，with many people suffering from urinary incontinence，dementia and mild cognitive impairment.2，3

Due to the rapid economic development and‘one child'policies in China，the traditional way of caring for the elderly at home has changed.4，5Nursing homes are now considered an important place for the elderly who need consistent care because of their illnesses and/or disabilities，and lack of family care.However，nursing home care is relatively new for the elderly and nurses in China and is still evolving.In China，people need nursing homes because they are completely dependent on help with their activities of daily living（ADL），and their adult children cannot provide care.The majority of nursing homes focus on daily living activities.Most of the daily living care is typically provided by nurse assistants.Medical care is provided bya limited number of physicians and registered nurses.6

In China，most nursing homes are owned by the government and Street or Residents'Committees.Sports and recreational activities are less frequently taken in nursing homes because of a shortage of funds and lack of appropriate sites.Whereas in the Netherlands，for example，people who are in need of long-term care often live in residential homes or nursing homes with 24-h supervision and a protected living environment.7

1.1.Background

The assessment of a patient's needs is an essential part of nursing practice and a significant amount of nursing research is conducted to improve assessment methods，such as the development of measurement instruments.8

Some assessment instruments，such as the Barthel Index，9were developed to assess patients'ADL.Nevertheless，nurses'views were not taken into account and patients'psychological and sociologicalneeds were not considered.10，11The（ADL）are usually used to measure the dependency level of older adults in China.11

http：//dx.doi.org/10.1016/j.cnre.2016.06.005

2095-7718/©2016 Shanxi Medical Periodical Press.Publishing services by Elsevier B.V.This is an open access article under the CC BY-NC-ND license（http：//creativecommons. org/licenses/by-nc-nd/4.0/）.

1.2.The Care Dependency Scale

The Care Dependency Scale（CDS），which is based on Henderson's 14 human needs，aims to confirm patients'nursing care dependency as part of the assessment.The CDS was originally developed in the Netherlands in 1994 and has been translated into more than 16 languages.12-14

The scale consists of 15 items，including：eating and drinking；continence；body posture；mobility；day and night patterns；getting dressed and undressed；body temperature；hygiene；danger avoidance；communication；contactwithothers；senseofrulesandvalues；daily activities；recreational activities，and learning abilities，which are all separatelyassessed.Each item has a brief description，and five care dependency criteria.A 5-point Likert scale assesses a patient's reliance，from completely dependent to completely independent. Therefore，the CDS total values are between 15 and 75 points.The lowerthevalue，themorethepatientdependsonothersforcare.15-17

Development and psychometric testing of the Dutch CDS is described in several studies.In the first study，the concept of care dependency was operationalized16and 44 experts in a Delphi survey established the content validity of the initial scale.18Reliability of the scale was determined as an inter-rater reliability（Kappa）of 0.40-0.64；test-retest reliability（Kappa）was 0.55-0.80，and Cronbach's alpha was 0.97.19Utility tests revealed that the CDS is easy to use and quick to complete，normally taking less than 5 min.19A factor analysis was carried out and resulted in a unifactorial solution（Eigenvalue 10.721）.In addition，the Mokken scale analysis for the whole scale demonstrated a high H-coefficient of 0.75.16，20Indications for the construct-related validity were found by comparing the CDS to three reasonably reliable and valid instruments.20

The instrument was first used to assess demented and mentally handicapped patients.18Patients'caredependency wasalso assessed in the geriatric，surgical and pediatric wards of hospitals in Germany.21A modified Arabic CDS for the self-assessment of older people in Egypt was used to measure nursing home residents，home care recipients and non-care recipients，and the results showed that the assessment of care dependency is useful among care recipients.22A modified CDS for rehabilitation was tested in Austria，and had a Cronbach's alpha of 0.97.23A modified CDS for pediatrics was tested in Germany and Egypt，and both had a strong internal consistency：Cronbach's alpha was 0.91 and 0.85 for the German and Arabic versions，respectively.24-26

The importance of the present study lies in further determining whether it is appropriate to compare the care dependency status of older patients in different settings and cultures using the same instrument.The aim of this study was to identify the psychometric properties of the CDS by analyzing data that was collected in China.

2.Methods

2.1.Translation procedures

The translated text of the English version into the Chinese version was examined to see whether the text was understood and considered valid by a nursing specialist in the original language，and whether it included items that were common and suitable for use across cultures.27Back translation is very common，and highly recommended for verifying the translation of an instrument.28

The initial forward translation and backward translation for the English version were performed by the language center of the Chinese Second Military Medical University.First，the instrument was translated from English to Chinese.Then，the Chinese version of the scale was translated back from Chinese to English.A native English-speaking person compared the original and backward translated English versions of the instrument，and changes to the item descriptions and criteria were made.A Chinese language expert checked the grammar and structure of the items.Finally，a group of Chinese nurses evaluated whether the items were relevant to the Chinese health care system.

2.2.Participants and settings

CDS data were obtained from a sample of 184 older adults from one nursing home in Shanghai.In July 2013，the final sample consisted of 103 women and 81 men.The patients involved in the study were aged 60 years and over，and their mean age was 81. Most of them suffered from a chronic illness（e.g.，cancer，chronic kidney disease，stroke，diabetes，dementia，or skeletal and circulatory system diseases）.

2.3.Procedures

The nurses most involved in the daily care of the patients rated their CDS status.All raters（n=15）were trained to administer CDS，and the aim of the study was explained.Because the nurses were to assess the scores independently，the principles of the assessment and instructions for the use of the CDS were elaborately stated.

2.4.Ethical considerations

The study was conducted with the cooperation of the nurses and their head nurses in the nursing home.The Bio-Ethical Committee at Changhai Hospital of Shanghai gave us permission to do this research in 2013.The patients were verbally informed about the purpose of the study.They were told that a questionnaire about their dependency status would be completed by the nurse most involved in their daily care and that this investigation was voluntary.Those who agreed to participate in the investigation and returned the informed consent were included in the final sample.

2.5.Reliability

Reliability is the degree of consistency or dependability with which an instrument is analyzed.29In this study，an item analysis（mean and SD）and Cronbach's alpha were used to assess the internal consistency of the Chinese version of the CDS.30，31

Repeated measurements involving the same raters and patients in identical circumstances were performed to determine the reliability of the measurement.The measurements were repeated in two different ways：first，inter-rater reliability（IRR）was determined by two raters.Two raters independently completed the questionnaire for the same patient at the same time（T1）for a total sample of 184；second，one of the raters completed the CDS for the same patient twice，with a 2-week time interval（T2），which includes only 128 patients due to loss from nursing home discharges or deaths.This method was based on two assumptions：that the health of the patient had not changed dramatically within the time period and that no learning effects had occurred in the observer in the period between the first and second measurement.29Cohen's Kappa was used to calculate inter-rater reliability and test-retest reliability（TRR），32as described by Landis and Koch.The verbal descriptions for the different ranges of the Kappa statistics are slight（＜0.20），fair（0.21-0.40），moderate（0.41-0.60），substantial（0.61-0.80）and almost perfect（＞0.80）.33

2.6.Validity

A factor analysis（principal component analysis）was used to examine construct validity.The CDS was assessed to determine whether it actually measures care dependency and does not contain any systematic errors.A principal component analysis was used to examine the underlying conceptual dimensions of the 15-item CDS，which presents values for factor loading；Eigenvalues and percentages of variance were also obtained.Factor loading express the correlations between each of the CDS items and the identified underlying factor.In this research，a factor loading＞0.40 was considered significant.All of the statistical analyses were performed with SPSS version 13.0 for Windows（SPSS，Chicago，IL，USA）.

3.Results

3.1.Sample characteristics

The mean age of the sample（n=184）in this study was 79.6（SD=9.2），and 56.0%were female.The CDS sum score was calculated by adding each of the 15 CDS item scores.The mean CDS sum score for males was 36.2（SD=17.5），30.9（SD=14.7）for females，and 36.1（SD=16.6）for the whole sample.Table 1 shows the mean scores and standard deviations for each of the 15 CDS items.

3.2.Reliability

The reliability of the CDS was analyzed in terms of internal consistency using Cronbach's alpha，mean inter-item correlation，inter-rater reliability and test-retest reliability.Cronbach's alpha was 0.95 by calculating at the scale level for the whole sample.The value of the mean inter-item correlation was 0.96.

As described by Landis and Koch，33the verbal descriptions for the different ranges of Kappa statistics are almost perfect（＞0.80），substantial（0.61-0.80），moderate（0.41-0.60），fair（0.21-0.40）and slight（＜0.20）.The Chinese data revealed that the IRR Kappa was between 0.84 and 0.89（Table 2）.Table 2 shows the TRR Kappa statistics ranging from 0.83 to 0.92 over an interval of 2 weeks.

3.3.Construct validity

A principal component analysis was used to find the common dimensions in the 15 caredependency items.InTable 3，the unrated principal component loadings of the scale items，Eigenvalues and percentages of explained variance are presented.For the Chinese sample，the first factor had an Eigenvalue of 9.3，and this factorexplained 62.2%of the variance，with the factor loading ranging from 0.13 to 0.81；four items had a loading lower than 0.40.Item five had a small correlationwith other items，and the factor loading was low in the first factor and the second factor，and thus it was deleted.Factor analysis performed for the first time is reported in Table 3.

Table 1 Means and standard deviations per item of the CDS.（n=184）.

Table 2 Inter-rater reliability（IRR）and test-retest reliability（TRR）（Kappa）of the CDS.

Table 3 Factor loadings，Eigenvalues and variance of the CDS.（n=184）.

We continued the principal component analysis.This time，we obtained one factor that had an Eigenvalue of 8.4 and explained 72.9%of the variance；the factor loading ranged from 0.78 to 0.90，and the result is presented in Table 4.

4.Discussion

The internal consistency of the CDS showed good results，and Cronbach's alpha was 0.95.The results of the validity testing were also satisfactory.According to Polit et al，29this indicates that the reliability of the scale is good enough toassess dependency，and the mean high inter-item correlation possibly points to a homogeneous population.

In terms of validity，the factor analysis produced one factor. The high factor loadings showed that all of the items were affected by the same care dependency concept，except for item five.The reason for this result may be that nurses do not take sleep seriously in the Chinese heath care setting.Patients go to bed at approximately 21：00，and most of them take medicine or just lay in bed if they cannot sleep，and the nurses do not disturb them.As a result，this item may not reflect a dependency of care.The results could be concluded that the psychometric properties regarding construct validity were acceptable and showed differences with earlier outcomes in other countries.14，20The analyses showed that one item had substantial residual correlations.The correlation of the item cannot be explained by the nursing care dependency concept，which means that other effects also influenced the item.

Table 4 Factor loadings，Eigenvalues and variance of the CDS.（n=184）.

In general，the CDS can be recommended as a dependency measurement in nursing homes in China.The CDS is a quick and easy instrument for nurses because most nurses graduated from college.After the assessment，appropriate care can be provided to make up for patients'self-care defects.The outcomes show perfect IRR Kappa values.The high Kappas were collected because the raters received good training had worked in the nursing home for a long time and knew the patients very well.A test-retest strategy was used to assess the stability of the scores，and the TRR Kappa statistics for the scores across the 2-week interval showed substantial to almost perfect Kappa values.The Kappa values in this study were in accordance with the previous Dutch results.15

5.Conclusions

An assessment from a nursing perspective enables nurses to give older patients care in nursing homes so that they can acquire care independence.Our findings provide evidence for the fact that the CDS components are measurable and can be used as an appropriate tool to assess care dependency in nursing homes in China.Further research is recommended regarding the psychometric properties of the CDS in other patient groups and Chinese health settings.

Relevance to clinical practice

The results of the Chinese version of the CDS are different from other versions，and may reflect the care dependency of the geriatric care environment of China.

Conflicts of interest

All contributing authors declare no conflicts of interest.

1.National Bureau of Statistics of China.Social Security and Elderly Life.2012；2013（lastaccessed 17 September2013）http：//www.stats.gov.cn/tjgb/ndtjgb/ qgndtjgb/t20130221_402874525.htm.

2.Chen YM，Hwang SJ，Chen LK，Chen DY，Lan CF.Urinary incontinence among institutionalized oldest old Chinese men in Taiwan.Neurourol Urodyn.2009；28：335-338.

3.Guo M，Gao L，Zhang G，et al.Prevalence of dementia and mild cognitive impairment in the elderly living in nursing and veteran care homes in Xi'an，China.J Neurol Sci.2012；312：39-44.

4.Cui LJ，Xu S，Wang XH.Ideas and models of elderly care.Chin J Gerontol.2000；20：3-5（in Chinese）.

5.Shang ZK.Nursing home care and administration in China.Chin J Popul Econ. 2008；167：50-54（in Chinese）.

6.Zhou XR，Liu G.Demands of nursing home care.Med J Chin People Health. 2007；19：597-598（in Chinese）.

7.Ministry of Health，Welfare and Sports.Long-term care in the Netherlands.The Exceptional Med Expenses Act.2009；10：11.

8.Mahoney FI，Barthel DW.Functional evaluation：the Barthel index.Md State Med J.1965；14：61-65.

9.Shelkey M，Wallace M.Katz index of independence in activities of daily living（ADL）.Director.2000；8：72-73.

10.Shelkey M，Wallace M.Katz index of independence in activities of daily living. J Gerontol Nurs.1999；25：8-9.

11.Feng Q，Zhen Z，Gu D，Wu B，Duncan PW，Purser JL.Trends in ADL and ADL disability in community-dwelling older adults in Shanghai，China，1998-2008. J Gerontol B Psychol Sci Soc Sci.2013；68：476-485.

12.Dijkstra A，Buist G，Dassen T.Nursing-care dependency.Development of an assessment scale for demented and mentally handicapped patients.Scand J Caring Sci.1996；10：137-143.

13.Dijkstra A.Care Dependency；an Assessment Instrument for Use in Long-term Care Facilities.Academic thesis.Groningen：De Regenboog；1998.

14.Dijkstra A，Coleman M，Dassen TW，Romoren TI，V¨alim¨aki M，Zanotti R.Care dependency of nursing home patients with dementia：assessment from European perspective.Tijdschr Gerontol Geriatr.2000；31：245-251.

15.Dijkstra A，Buist G，Moorer P，Dassen T.Construct validity of the nursing care dependency scale.J Clin Nurs.1999；8：380-388.

16.Dijkstra A，Tiesinga LJ，Plantinga L，Veltman G，Dassen TW.Diagnostic accuracy of the care dependency scale.J Adv Nurs.2005；50：410-416.

17.Dijkstra A，Brown L，Havens B，et al.An international psychometric testing of the care dependency scale.J Adv Nurs.2000；31：944-952.

18.Lohrmann C，Dijkstra A，Dassen T.The Care Dependency Scale：an assessment instrument for elderly patients in German hospitals.Geriatr Nurs.2003；24：40-43.

19.Dijkstra A，Coleman M，Tomas C，V¨alim¨aki M，Dassen T.Cross-cultural psychometric testing of the care dependency scale with data.J Adv Nurs.2003；43：181-187.

20.Dijkstra A，Muszalik M，Ke˛dziora-Kornatowska K，Kornatowski T.Care dependency scale-psychometric testing of the polish version.Scand J Caring Sci. 2010；24：62-66.

21.Lohrmann C，Balzer K，Dijkstra A，Dassen T.Care dependency in nursing homes-a psychometric study.Z Gerontol Geriatr.2003；36：255-259.

22.Boggatz T，Farid T，Mohammedin A，Dijkstra A，Lohrmann C，Dassen T.Psychometric properties of the extended Care Dependency Scale for older persons in Egypt.J Clin Nurs.2009；18：3280-3289.

23.Eichhorn-Kissel J，Dassen T，Kottner J，Lohrmann C.Psychometric testing of the modified care dependency scale for rehabilitation.Clin Rehabil.2010；24：363-372.

24.Tork H，Dassen T，Lohrmann C.Care dependency of hospitalized children：testing the Care Dependency Scale for Paediatrics in a cross-cultural comparison.J Adv Nurs.2009；65：435-442.

25.Tork H，Dassen T，Lohrmann C.Care dependency of children in Egypt.J Clin Nurs.2008；17：287-295.

26.Tork H，Lohrmann C，Dassen T.Psychometric testing of the modified Care Dependency Scale among hospitalized school-aged children in Germany.Nurs Health Sci.2008；10：17-22.

27.Chapman DW，Carter JF.Translation procedures for the cross-cultural use of measurements.Educ Eval Policy Anal.1979；3：71-76.

28.Jones E.Translation of quantitative measures for use in crosscultural research. Nurs Res.1987；36：324-327.

29.Polit DF，Beck CT，Hungler BP.Nursing Research：Principles and Methods.7th ed. Philadelphia：Lippincott，Williams&Wilkins；2004.

30.Cronbach LJ.Coefficient alpha and internal structure of tests.Psychometrics. 1951；3：297-334.

31.Nunally JC，Bernstein IH.Psychometric Theory.3rd ed.New York：McGraw-Hill Inc；1994.

32.Cohen J.A coefficient of agreement for nominal scales.Educ Psychol Meas. 1960；20：37-46.

33.Landis JR，Koch GG.The measurement of observer agreement for categorical data.Biometrics.1977；33：159-174.

How to cite this article：Zhang S-Q，Wang H-M，Yang X-Y，et al. The Care Dependency Scale：psychometric testing of the Chinese version.Chin Nurs Res.2016；3：62-65.http：//dx.doi.org/10.1016/ j.cnre.2016.06.005

*Corresponding author.

E-mail address：happypenny@126.com（S.-Q.Zhang）.

Peer review under responsibility of Shanxi Medical Periodical Press.