Development And Psychometric Testing Of The Self-management Scale For Continuous Ambulatory Peritoneal Dialysis Patients

Objective: To develop a scale of self-management for continuous ambulatoryperitoneal dialysis (CAPD) patients, which would apply to Chinese cultural background,and to examine the reliability and validity of the scale.Methods⑴The theoretical framework of the scale was based on the “health deviation self-carerequisites” of Orem,s theory of self-care.⑵The pretesting scale is formed by the following steps: concept defining, literaturereview of both domestic and foreign publications, referring to evidence-baced guideline ofkidney disease, semi-structured interviews, and two rounds of Delphi expert consultation;9CAPD patients with different ages, educational backgrounds and occupations werepurposefully selected as a pretest to assess the comprehensibility of the items; The expertpanel was asked to evaluate whether each item in the scale logically represent the contentto be measured. The contribution of individual items in each dimension was adjusted.Finally, an initial scale with5factors and44items was formed.⑶Totally313CAPD patients were investigated by using the initial scale. All datasetup the date base through epidata3.1, and was input into SPSS18.0. The items wereanalyzed and screened according to item distribution, coefficient of variation(CV), criticalration(CR), Pearson correlation coefficients and exploratory factor analysis. The reliabilityanalysis included Cronbach,s coefficient, Split-half reliability and test-retest reliability.The validity analysis included face validity, content validity and construct validity.Results⑴Item analysis: The formal scale of self-management for CAPD patients wasdeveloped. There were16original items deleted, and28items accepted after item analysisand exploratory factor analysis(EFA). The formal scale consists of5dimensions: technical operations of liquid replacement(7items), solution for unusual situation(4items), dietmanagement(5items), monitoring complications(8items), emotional regulation and socialparticipation(4items). The scale was designed as a Likert4scale, The scores for“Never”,“Sometimes”,“Often”, and“Always”are0,1,2, and3points respectively. Thescale dose not included any reverse items. The total score ranges from0to84, and higherscores suggest better self-management ability among CAPD patients.⑵Reliability test：①The Cronbach’s α of the whole scale was0.926, and0.875,0.903,0.650,0.908,0.802of the five factors respectively.②The split reliability of the whole scale was0.960, and0.891,0.909,0.640,0.925,0.849of the five factors respectively.③The test-rest reliability of the whole scale was0.937, and0.837,0.802,0.782,0.805,0.804of the five factors respectively. All the coefficients were statistically significant(P＜0.01).⑶Validity test：①Face validity: The initial scale was drafted based on literature review,semi-structured interviews with CAPD patients and Delphi expert consultation, and then itwas tested among small samples and thus showed good face validity.②Content validity: Content validity was evaluated by Content Validity Index(CVI).The CVI of the formal scale was0.963, and each factor ranges form0.9to1.0, and eachitem0.73to1.0.③Construct validity: The correlation coefficients between items and the total scaleranges from0.370to0.758. The correlation coefficients between items and its commonfactor ranges from0.546to0.918. The correlation coefficients between items and othercommon factors were ranges from0.177to0.559. The correlation coefficients amongfactors ranges from0.288to0.545. The correlation coefficients between factors and thetotal scale ranges from0.579to0.879. All the difference of correlation coefficients werestatistically significant(P＜0.01).There are5factors extracted by EFA, which explained for64.567%variances. The loading coefficient of items under common factors was over0.5,suggesting the factor model of the formal scale was reasonable and basically closed to thetheoretical construct. The scale was proved to be a reliable and valid tool for the assessment of self-management of CAPD patients.Conclusion: The self-management scale for CAPD patients was proved to be areliable and valid tool which can reflect essence of self-management precisely andscientifically. Its content conforms to our country’s social cultural background and thecharacteristics of CAPD patients’ self-management behavior. The instrument can be usedto evaluate specific self-management behaviors of CAPD patients.