-
[show abstract]
[hide abstract]
ABSTRACT: Frequently in clinical studies a primary outcome is formulated from a vector of binary events. Several methods exist to assess treatment effects on multiple correlated binary outcomes, including comparing groups on the occurrence of at least one among the outcomes ('collapsed composite'), on the count of outcomes observed per subject, on individual outcomes adjusting for multiplicity, or with multivariate tests postulating either common or distinct effects across outcomes. We focus on a 1-df distinct effects test in which the estimated outcome-specific treatment effects from a GEE model are simply averaged, and compare it with other methods on clinical and statistical grounds. Using a flexible method to simulate multivariate binary data, we show that the relative efficiencies of the assessed tests depend in a complex way on the magnitudes and variabilities of component incidences and treatment effects, as well as correlations among component events. While other tests are easily 'driven' by high-frequency components, the average effect GEE test is not, since it averages the log odds ratios unweighted by the component frequencies. Thus, the average effect test is relatively more powerful than other tests when lower frequency components have stronger associations with a treatment or other predictor, but less powerful when higher frequency components are more strongly associated. In studies when relative effects are at least as important as absolute effects, or when lower frequency components are clinically most important, this test may be preferred. Two clinical trials are discussed and analyzed, and recommendations for practice are made.
Statistics in Medicine 12/2010; 29(28):2890-904. · 1.88 Impact Factor
-
American journal of surgery 04/2010; 203(4):556-7. · 2.36 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: To compare the prevalence of urinary incontinence (UI) between Hispanic and non-Hispanic White women in a population-based study.
The prevalence of moderate to severe UI, defined as Sandvik severity score of >or=3, was assessed in relation to ethnicity by stratification, age adjustment and logistic regression models among 250 Hispanic and 491 non-Hispanic White women in Colorado, USA, who were participants in a breast cancer case-control study.
Hispanic women reported more stress UI (odds ratio 1.7, P = 0.005) and mixed UI (odds ratio 1.8, P = 0.005) than did non-Hispanic White women. These higher prevalences were largely associated with ethnic differences in parity, body mass index, diabetes, hysterectomy and bilateral oophorectomy.
The prevalence of moderate to severe UI in Colorado is higher among Hispanic women than among non-Hispanic white women. This difference is largely compatible with differences in reproductive history, adiposity and diabetes.
BJU International 03/2008; 101(5):575-9. · 2.84 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: To estimate the reliability and interobserver consistency of urodynamic interpretations of female bladder and urethral function.
Three urogynecologists and three female urologists at a tertiary care medical center reviewed masked, abstracted clinical and urodynamic information from 100 charts, selected for adequate completeness from a consecutive series of 135 women referred for urodynamic testing. For each of the 100 cases, the reviewers assigned International Continence Society filling and voiding phase diagnoses, and overall clinical diagnoses. Raw agreement proportions and weighted kappa chance-corrected agreement statistics (kappa) were used jointly to describe both reliability and interobserver agreement. Reliability was estimated from duplicate reviews, masked and separated by at least 4 months, of each case by each physician. Interobserver agreement was estimated from comparisons of all pairs of responses from different physicians.
For clinical diagnosis of stress incontinence (present, absent, indeterminate), the within- and across-physician weighted kappa's were, respectively, 0.78 and 0.68. Corresponding results were 0.40 and 0.13 for detrusor overactivity without incontinence, 0.58 and 0.38 for detrusor overactivity with incontinence, and 0.51 and 0.26 for voiding dysfunction. Standard errors of each kappa were between 0.023 and 0.043.
In our group, lower urinary tract diagnoses of stress urinary incontinence from both clinical and urodynamic data demonstrated substantial reliability and interobserver agreement. However, by conventional interpretation of kappa-statistics, reliability of diagnoses of detrusor overactivity or voiding dysfunction was only moderate, and interobserver agreement on these diagnoses was no better than fair. Urodynamic interpretations may not be satisfactorily reproducible for these diagnoses.
Obstetrics and Gynecology 09/2006; 108(2):315-23. · 4.73 Impact Factor
-
[show abstract]
[hide abstract]
ABSTRACT: Diabetic bladder dysfunction is among the most common and bothersome complications of diabetes mellitus. While bladder filling and voiding problems have been reported, the precise functional changes in diabetic bladders remain unclear. We investigated time dependent changes in bladder function in streptozotocin induced diabetic rats.
Cystometrograms and detrusor muscle contractility were examined in male age matched control and diabetic Sprague-Dawley rats (Harlan, Indianapolis, Indiana) 3, 6, 9, 12 and 20 weeks after diabetes induction with streptozotocin.
Diabetes decreased average body weight and increased bladder weight, capacity and compliance. Peak detrusor leak pressure increased gradually from weeks 3 to 6 to 9 in diabetic rats (mean +/- SEM 47.3 +/- 2.5, 50.8 +/- 3.0 and 56.0 +/- 3.6 cm H(2)O) and in controls (36.9 +/- 1.4, 37.7 +/- 1.5 and 41.6 +/- 1.81 cm H(2)O, respectively). However, at 12 and 20 weeks diabetic rats deviated strongly from this trend with peak detrusor leak pressure decreasing vs controls (41.6 +/- 2.8 and 37.3 +/- 0.9 vs 45.2 +/- 1.7 and 49.6 +/- 1.4 cm H(2)O, respectively) and post-void resting pressures increasing from 9-week levels vs controls (interactions p <0.0001). In contractility studies increased contractile force responses of diabetic animals to carbamylcholine chloride, potassium chloride, adenosine 5'-triphosphate and electric field stimulation peaked at 6 or 9 weeks but at 12 to 20 weeks they generally reverted toward those of controls (carbamylcholine chloride and electrical field stimulation interactions p = 0.0022 and 0.01, respectively).
Diabetic bladders may undergo a transition from a compensated to a decompensated state and transition in the streptozotocin rat model may begin 9 to 12 weeks after induction.
The Journal of Urology 08/2006; 176(1):380-6. · 3.75 Impact Factor
-
Peter B. Imrey
07/2005; , ISBN: 9780470011812
