Font Size: a A A

Development And Preliminary Application Of Symptom Assessment Scale Of Cancer Therapy-Cervix

Posted on:2021-11-14Degree:DoctorType:Dissertation
Country:ChinaCandidate:S Q WeiFull Text:PDF
GTID:1484306134455064Subject:Nursing
Abstract/Summary:PDF Full Text Request
Objectives 1.To develop a cancer-site-specific patient-reported symptoms measurement,the Symptom Assessment of Cancer Therapy-Cervix(SACT-Cx).2.To investigate the psychometric properties of SACT-Cx among Chinese cervical cancer patients based on classical test theory and item response theory.3.To describe the incidence and severity of symptoms in patients with cervical cancer during radiotherapy,and to classify them into symptom clusters.Then we proceed to explore the associations between symptom clusters and patients' demographic and clinical characteristics.Methods 1.A systematic scoping review from databases and a semi-structured interview with cervical cancer patients were conducted in order to comprehensively identify the common and frequently reported symptoms of cervical cancer.Additionally,published existing instruments in the context of cancer were also reviewed to extract potential items.Item selection was a consensus-based process using Delphi approaches.Then 50 patients with cervical cancer accepted the SACT-Cx investigation to modify the expression of each item.2.274 patients with cervical cancer accepted questionnaires investigation to evaluate the psychometric properties of SACT-Cx.Validity was examined by content,construct and convergent validity.Content validity was measured for 7 experts in the second Delphi approach by using I-CVI and S-CVI/Ave.To verify the construct validity,the exploratory factor analysis(EFA)was carried out to explore the structure of the scale and further reduce the items of the SACT-Cx.To assess the reliability,the internal consistency of the SACT-Cx was measured by calculating the Cronbach's Alpha.Use the same data with the part 2.EFA was carried to evaluate the unidimensionality of each dimension.The difficulty and discrimination of each items was estimated using Stata 15.0,and then drawn the item characteristics curve,item information curve of each item and test information function of each dimension.3.A total of 184 patients with cervical cancer during the radiotherapy eligible for this study.EFA with principal components analysis and varimax rotation with Kaiser normalization was performed to classify the symptom clusters.Symptoms with an incidence of over 20% were included in the EFA.The internal reliability of the identified clusters was calculated using Cronbach's alpha.One-way analysis of variance(ANOVA),independent t-test and Pearson correlation coefficient were used to analyze the associations between symptom clusters and patient's demographic and clinical characteristics.Results 1.In the SACT-Cx,each item was composed of two sub-questions.The first one was to detect the occurrence of symptoms(0= “never occur”,1= “occur”)and the other was to assess the distress of symptoms experienced by patients.The score for the second question presented the occurrence and distress of each symptom which ranged from 0(“none”)to 10(“too severe to bear”).18 items with I-CVI less than 0.78 were removed.The final SACT-Cx contained 19 physical and psychological symptom items.2.For content validity,consensus-based Delphi process yielded I-CVI ranging from 0.86 to 1.00,and S-CVI/Ave of 0.94.Convergent validity was measured comparing the SACT-Cx with the MSAS.The associations between the two measures was evaluated using Person correlation analysis,with coefficients ranging from 0.01 to 0.99,which indicated moderate to high correlation.However,the symptoms ‘abdominal distention or swelling of body',‘diarrhea',‘lack of appetite' and “weight loss” presented a low level of correlation(r<0.4).The total scores of SACT-Cx was moderately correlated with the MSAS(r=0.74,P<0.01).EFA was performed by means of Promax rotation.Five factors were found in the SACT-Cx accounted for 77.23% of the explained variance,which were ‘symptoms related to mental health',‘symptoms related to urogenital tract',‘symptoms related to chemotherapy',‘symptoms related to digestive system' and ‘pain-fatigue'.The Cronbach's Alpha coefficient for the overall scale was ?=0.858.With regard to internal consistency in domains,the Cronbach's Alpha coefficient was0.94,0.97,0.9,0.68 and 0.67,respectively.According to the results of EFA,4 dimensions of SACT-Cx met the unidimensionality requirement.The ranges of discrimination of item in ‘symptoms related to mental health',‘symptoms related to urogenital tract',‘symptoms related to chemotherapy',and ‘symptoms related to digestive system' were 0.61-0.71?0.68-0.72?0.72-1.48 and 0.68-3.25,respectively.Except for item 2,item 3,and item 19,the difficulty of other items was exceeded the-3~3.Only item information of item 2 was higher than 3,and other items were lower than 0.5.3.The top 5 symptoms with the highest incidence were diarrhea(88%),fatigue(82.1%),bloody stools(56.0%),abdominal distention or swelling in body(54.3%)and perianal pain or bleeding(53.3%).The most severe symptom was fatigue(3.33 ± 2.19),diarrhea(3.20 ± 2.06),perianal pain or bleeding(1.97 ± 2.80),abdominal distention or swelling in body(1.96 ± 2.22),bloody stools(1.80 ± 1.95)and pain(1.80 ± 2.48).14 symptoms with prevalence over 20% were included in EFA.Using principal component analysis,4 symptom clusters were identified with Eigen values greater than 1.0.These 4 clusters explained 78.87% of the total variance.Symptom clusters 1-5 were psychological symptom cluster,urogenital system symptom cluster,digestive system-pain symptom cluster and digestive system – fatigue symptom cluster respectively.The Cronbach' alpha of factor 1-5 was 0.96,0.94,0.75 and 0.67,respectively.There was a significant difference between patients were employed and not employed in psychological symptom cluster(t = 2.37,P<0.05).The difference between patients had surgery and hadn't in urogenital system symptom cluster was significant(t = 2.45,P<0.05).Conclusion 1.The SACT-Cx was consisted of 19 physical and psychological symptoms,and covered mental health,urogenital tract,chemotherapy and digestive system facts.2.The SACT-Cx has been proved with adequate reliability and validity.The ability of items of SACT-Cx to distinguish the different symptoms of patients with cervical cancer was good.However,the difficulties of items were too high.3.Diarrhea and fatigue were the most frequent and severe symptoms of patients with cervical cancer during the radiotherapy.There were 4 stable symptom clusters extracted.Employment status of patients and surgery were associated with symptom clusters in patients with cervical cancer.
Keywords/Search Tags:Cervical Cancer, Classical Test Theory, Item Response Theory, Validity and Reliability, Symptom Cluster, Symptom Management
PDF Full Text Request
Related items