Optimal Hypothesis Testing: From Semi to Fully Bayes Factors.

We propose and examine statistical test-strategies that are somewhat between the maximum likelihood ratio and Bayes factor methods that are well addressed in the literature. The paper shows an optimality of the proposed tests of hypothesis. We demonstrate that our approach can be easily applied to practical studies, because execution of the tests does not require deriving of asymptotical analytical solutions regarding the type I error. However, when the proposed method is utilized, the classical significance level of tests can be controlled.

Race of male sex partners and occurrence of bacterial vaginosis.

BACKGROUND: Whether bacterial vaginosis (BV) is sexually transmitted is uncertain. Also it is unknown why BV is approximately twice as prevalent among black as among white women. An association of BV with a characteristic of the male sex partner, such as race, might support sexual transmission as well as account for the observed ethnic disparity in BV. METHODS: Three thousand six hundred twenty nonpregnant women 15 to 44 years of age were followed quarterly for 1 year. At each visit, extensive questionnaire data and vaginal swabs for Gram's staining were obtained. The outcome was transition from BV-negative to positive (Nugent's score > or =7) in an interval of 2 consecutive visits. RESULTS: BV occurred in 12.8% of 906 sexually active intervals to white women-24.8% of intervals when the woman reported a black partner and 10.7% when all partners were white. Among white women, there was a 2-fold increased risk for BV incidence with a black, compared with a white partner (risk ratio [RR] 2.3, 95% confidence interval 1.6-3.4; adjusted RR 2.2, 95% confidence interval 1.5-3.4), but differed according to condom use. In the presence of consistent condom use, the adjusted RR was 0.7 (0.3-2.4); it was 2.4 (1.0-6.2) in the presence of inconsistent use; and 2.7 (1.7-4.2) in the absence of condom use. Black women could not be studied, as there were insufficient numbers who reported only white male sex partners. CONCLUSION: The association of BV occurrence with partner's race, and its blunting by condom use, suggests that BV may have a core group component and may be sexually transmitted.

Personal hygienic behaviors and bacterial vaginosis.

BACKGROUND: Vaginal douching is consistently associated with bacterial vaginosis (BV), but whether it is a cause or result of BV remains unknown. The association between BV and other feminine hygienic behaviors is less studied; if BV symptoms caused behavior change then all hygiene behaviors might be more common among women with BV. Lack of association between nondouching hygiene behavior and BV would argue against reverse causation. METHODS: In the Longitudinal Study of Vaginal Flora 3620 women had 13,517 visits where BV (Nugent score) was assessed. Associations between hygienic behavior and BV were assessed by Poisson regression. RESULTS: After adjusting for demographic and sexual behavior factors, neither type of underwear (nylon vs. cotton prevalence ratio (PR) 1.05, 95% CI: 0.97-1.13), menstrual protection (tampons vs. pads; PR: 1.04, 95% CI: 0.95-1.12; pads and tampons vs. pads 1.00, 95% CI: 0.92-1.07), use of pads or panty liners when not menstruating (PR: 0.99, 95% CI: 0.95-1.05), nor weekly or greater use of hygiene spray (PR: 1.01, 95% CI: 0.94-1.09), powder (PR: 1.02, 95% CI: 0.96-1.07) or towlettes (PR: 1.03, 95% CI: 0.94-1.13) were strongly associated with BV. PR for daily versus less than daily bathing and showering were 1.06 (95% CI: 1.02-1.12) and 1.04 (95% CI: 1.00-1.09). Douching remained associated with BV (PR for weekly or greater vs. never 1.17, 95% CI: 1.09-1.26) and was not substantially impacted by adjustment for other hygienic behavior. CONCLUSIONS: Douching, but not other feminine hygiene behaviors, is significantly associated with BV, providing additional evidence that douching may be causally associated with BV and is not simply a response to BV symptoms.

Estimation of Variance Components in the Mixed-Effects Models: A Comparison Between Analysis of Variance and Spectral Decomposition.

The mixed-effects models with two variance components are often used to analyze longitudinal data. For these models, we compare two approaches to estimating the variance components, the analysis of variance approach and the spectral decomposition approach. We establish a necessary and sufficient condition for the two approaches to yield identical estimates, and some sufficient conditions for the superiority of one approach over the other, under the mean squared error criterion. Applications of the methods to circular models and longitudinal data are discussed. Furthermore, simulation results indicate that better estimates of variance components do not necessarily imply higher power of the tests or shorter confidence intervals.

Dietary intake of selected nutrients affects bacterial vaginosis in women.

Bacterial vaginosis (BV), a condition of altered vaginal flora, is associated with various adverse reproductive health outcomes. We evaluated the association between diet and the presence of BV in a subset of 1521 women (86% African-American) from a larger study of vaginal flora. Participants completed the Block Food Questionnaire and clinical assessments and self-report measures of sexual and hygiene behavior. A total of 42% of the women were classified as having BV (Nugent score > or = 7). Severe BV (Nugent score > or = 9 and vaginal pH > or = 5) was present in 14.9% of the women. BV was associated [adjusted OR (AOR)] with increased dietary fat (1.5, 1.1-2.4) after adjusting for other energy nutrients and behavioral and demographic covariates. Severe BV was associated with total fat (2.3, 1.3-4.3), saturated fat (2.1, 1.2-3.9), and monounsaturated fat (2.2, 1.2-4.1). Energy intake was only marginally associated (P = 0.05) with BV (1.4, 1.0-1.8). There were significant inverse associations between severe BV and intakes of folate (0.4, 0.2-0.8), vitamin E (0.4, 0.2-0.8), and calcium (0.4, 0.3-0.7). We conclude that increased dietary fat intake is associated with increased risk of BV and severe BV, whereas increased intake of folate, vitamin A, and calcium may decrease the risk of severe BV.

A pilot study of vaginal flora changes with randomization to cessation of douching.

OBJECTIVES: The objectives of this study were to determine whether women who douche regularly would enter a randomized trial of douching cessation or continuation, whether they would adhere to the assigned behavior, and whether there was a dramatic impact on vaginal flora. GOAL: The goal of this study was to determine the feasibility of a large, definitive trial of douching cessation. STUDY DESIGN: Women who douched at least weekly and who had either bacterial vaginosis or normal flora by Gram stain were assigned at random either to continue douching or to stop for 8 weeks. Vaginal Gram stains were obtained every 7 days from each woman. RESULTS: Forty-eight women were randomized. Those assigned to continue reported douching during 77% of study weeks; those assigned to stop denied douching in 94% of weeks. No dramatic differences in flora were observed between women in the continue versus stop groups. CONCLUSION: A large randomized trial of douching cessation is feasible.

The association of psychosocial stress and bacterial vaginosis in a longitudinal cohort.

OBJECTIVE: The purpose of this study was to assess the association of psychosocial stress with bacterial vaginosis in a longitudinal sample of nonpregnant women. STUDY DESIGN: A 1-year prospective longitudinal design was used. Nonpregnant women (n = 3614) aged 15 to 44 years were recruited at routine health care visits. Assessments were conducted quarterly for 1 year and included a standardized pelvic examination, an assessment of clinical symptoms, and an extensive self-report interview. RESULTS: Psychosocial stress was associated with overall prevalence of (adjusted odds ratio, 1.10; 95% CI, 1.01-1.20) and an increased incidence of (adjusted odds ratio, 1.29; 95% CI, 1.12-1.48) bacterial vaginosis. The association between stress and bacterial vaginosis incidence was not changed appreciably by the control for behavioral and demographic characteristics and was magnified (odds ratio, 2.05; 95% CI, 1.15-3.66) in a case-crossover analysis. CONCLUSION: Increased psychosocial stress is associated with greater bacterial vaginosis prevalence and incidence independent of other risk factors.

Vulvovaginal symptoms in women with bacterial vaginosis.

OBJECTIVE: A substantial, but highly variable, percentage of women with bacterial vaginosis are said to be asymptomatic. The purpose of this study was to estimate the prevalence of symptoms among women with bacterial vaginosis compared with women without bacterial vaginosis by direct, explicit, and detailed questioning of these women. METHODS: Women presenting for a routine health care visit at 12 health department clinics in Birmingham, Alabama, were recruited to participate in a longitudinal study of vaginal flora. At the first visit, they underwent a pelvic examination, lower genital tract microbiological evaluation, and an interview that included detailed questions regarding lower genital tract symptoms. The prevalence of symptoms among women with and without bacterial vaginosis (Gram stain score 7 or higher) was compared. RESULTS: Among 2,888 women without gonorrhea, Chlamydia, or trichomonas, 75% of women with and 82% of women without bacterial vaginosis never noted any vaginal odor in the past 6 months (P <.001). The corresponding values were 63% and 65% for never noting vaginal "wetness" (P =.02); 58% and 57% for vaginal discharge (P =.65); 91% and 86% for irritation (P =.004); 88% and 85% for itching (P =.64); and 96% and 94% for dysuria (P =.002), respectively. Cumulatively, 58% of women with bacterial vaginosis noted odor, discharge, and/or wetness in the past 6 months compared with 57% of women without bacterial vaginosis (P =.70). CONCLUSION: The 2 classic symptoms of bacterial vaginosis discharge and odor are each reported by a minority of women with bacterial vaginosis and are only slightly more prevalent than among women without bacterial vaginosis.

Regression models for unbalanced longitudinal ordinal data: computer software and a simulation study.

A computer program GGOREX in the form of a SAS macro is developed for the analysis of longitudinal ordinal data. It is extended from GEECAT and GEEGOR developed by Williamson, Lipsitz and Kim in their paper in 1999. An illustrative example with some preliminary data from a study conducted by the National Institute of Child Health and Human Development and the University of Alabama at Birmingham is given. Another set of computer programs is developed to make it possible to conduct a simulation study to validate the GGOREX procedure in a finite sample. A simple computer program in FORTRAN using the IMSL software library is developed to solve for the probability distribution of the longitudinal ordinal responses according to the correlation specification.