Statistical Analysis of Reliability and Validity of a Chinese Version of Dysfunctional Voiding Symptom Score (DVSS) Questionnaire
Date Issued
2011
Date
2011
Author(s)
Chang, Shang-Jen
Abstract
Dysfunctional voiding in children is a multifactorial disease presenting with a variety of symptoms including daytime incontinence, urgency, curtsying, difficult voiding and constipation. In spite of four different scales have been validated to measure pediatric dysfunctional voiding since 2000, there is still lacking of psychometrically-sound scales in Chinese to measure pediatric dysfunctional voiding, we performed an age- gender-matched case-control study to evaluate the reliability and validity of Dysfunctional Voiding Symptom Score (DVSS) in Chinese version. From Sep 2008 to Feb 2011, we enrolled 60 children (38 girls and 22 boys, case) with diagnosis of dysfunctional voiding by clinical evaluation and urodynamic studies in clinics of pediatric urology. Baseline characteristics were recorded including age, gender, body height, weight, urine analysis and parameters associated with urodynamic studies. The DVSS was completed at clinics by the parents who mainly take care of the child in the presence of the child. After one week, the DVSS was administered at home and returned to clinics for evaluation of test-retest reliability. For the control group, we enrolled 235 age- and gender- matched healthy children from kindergartens and elementary schools.
The DVSS consisted of 10 items with the score of each item scores ranging from 0 to 3. Among these items, seven were related to voiding, two were related to defecation and one was related to stressful events. The DVSS was translated into Chinese and did back translation into English. The DVSS in Chinese version was reviewed by five members in the committee of pediatric urology of Taiwan Urological Association with respect to content validity.
The reliability of DVSS questionnaires completed by cases was assessed with Cronbach’s alpha test (internal consistency) and intraclass correlation (ICC, test-retest reliability). The predictive validity was first analyzed with logistic regression and was further assessed by Receiver Operating Characteristic curve analysis. Factor analysis was used to classify the ten symptoms into latent factors and estimate factor score of individual latent variables. Finally, we applied Bayesian method to evaluating the posterior probability and odds of dysfunctional voiding in children screened with DVSS from areas of different prevalence rates.
Mean total scores of DVSS in case and control group were 9.65(±3.87) and 4.13(±2.60), respectively. The Cronbach’s alpha coefficient was 0.448 for the ten items among DVSS indicating a heterogeneous composition of symptoms in dysfunctional voiding. Intraclass correlation coefficient for test-retest reliability was 0.89. The cut-off for total DVSS score was 6.66, yielding 81.67% of sensitivity and 82.63% of specificity. Factor analysis revealed three latent variables (overactive bladder symptoms, difficult elimination symptoms, stress and dysuria symptoms). Using Bayesian method, the applications of DVSS in areas of different prevalence figures give an indication of significantly different probabilities of dysfunctional voiding, 8% and 63% in a community with 2 % prevalence and pediatric urology clinics with 30 % prevalence, respectively give a total score equal to 8.
Conclusions: DVSS in Chinese version is reliable and valid to predict children with dysfunctional voiding. Through Bayesian method, we found significantly different probabilities and odds of having dysfunction voiding in children with the same total DVSS scores from areas of different prevalence figures.
Subjects
pediatric dysfunctional voiding
questionnaire scale
reliability
validity
factor analysis
Bayesian theorem
Type
thesis
File(s)![Thumbnail Image]()
Loading...
Name
ntu-100-P98842007-1.pdf
Size
23.32 KB
Format
Adobe PDF
Checksum
(MD5):905217c2680298f4079d2ba4739bb5ee