A Multiphase Design Strategy for Dealing with Participation Bias

Summary A recently funded study of the impact of oral contraceptive use on the risk of bone fracture employed the randomized recruitment scheme of Weinberg and Wacholder (1990, Biometrics 46, 963–975). One potential complication in the bone fracture study is the potential for differential response rates between cases and controls; participation rates in previous, related studies have been around 70%. Although data from randomized recruitment schemes may be analyzed within the two‐phase study framework, ignoring potential differential participation may lead to biased estimates of association. To overcome this, we build on the two‐phase framework and propose an extension by introducing an additional stage of data collection aimed specifically at addressing potential differential participation. Four estimators that correct for both sampling and participation bias are proposed; two are general purpose and two are for the special case where covariates underlying the participation mechanism are discrete. Because the fracture study is ongoing, we illustrate the methods using infant mortality data from North Carolina.     

Construct validity of the Short Form-36 Health Survey and its relationship with BMI in obese outpatients

Objective: To investigate the construct validity of the Short Form‐36 (SF‐36) Health Survey questionnaire in obese patients. Research Methods and Procedures: Our series consisted of 1735 obese patients (age, 44.7 ± 11.0 years; 1346 women) consecutively enrolled in the QUOVADIS study, an observational multicenter study of obese treatment‐seeking outpatients. The construct validity of the SF‐36 was assessed by main component analysis. Age‐, gender‐, and education‐adjusted general linear models were used to investigate the relationship between BMI and SF‐36 domains or factors identified by main component analysis. Results: BMI was significantly associated with poor health‐related quality of life in all eight SF‐36 domains, and the strongest association was observed with physical activity. Main components analysis generated a six‐factor solution explaining 59% of the observed variance. BMI was strongly associated with factors based on the loading of items regarding the physical activity domain and factors based on role‐physical and role‐emotional items or general health and bodily pain items. In contrast, mental health‐, vitality‐, and social functioning‐based factors were not related to BMI. Discussion: In obese treatment‐seeking outpatients, the clustering of SF‐36 items in main components is not significantly different from the domain‐based approach generally used, thus confirming the robustness of such a generic questionnaire in this specific condition. However, the peculiar clustering of some SF‐36 items and their relationship with BMI suggest that the health‐related quality of life profile of subjects belonging to that population may be better described with alternative aggregations of the SF‐36 items or with disease‐tailored questionnaires.     

Acute bronchitis therapy with ivy leaves extracts in a two-arm study. A double-blind, randomised study vs. an other ivy leaves extract

Ivy leaves extracts are authorised in medicinal products for the treatment of acute bronchitis. Different studies and the long experience on the market show safety and efficacy of this drug. A double-blind, randomised study was conducted to assess the efficacy and tolerability of ivy leaves soft extract with an other ivy leaves extract. 590 patients with acute bronchitis participated in this study. They were treated with test or comparator for 7 days (±1). The Bronchitis Severity Score (BSS) decreased gradually and to a similar extent from Day 1 to Day 7 in both treatment groups. Starting from values of 6.2-6.3 ± 1.2, the BSS decreased by approximately 4.7-4.9 points until Day 7, so that patients left the study with a mean BSS of 1.4-1.6. The BSS subscales cough, sputum, rhales/rhonchi, chest pain during coughing, and dyspnoea improved to a similar extent in both treatment groups. Overall, 2.7% of patients (per group and overall) experienced an adverse event, all of which were non-serious. Fewer patients younger than ten years had adverse events than would have been expected from their share of the study population (p = 0.015; Fisher's exact test). As a conclusion, the test product with ivy leaves soft extract proved to be non-inferior to the comparator ivy leaves extract in improving symptoms of acute bronchitis.     

Conformational control in structure-based drug design

In structure-based drug design, the basic goal is to design molecules that fit complementarily to a given binding pocket. Since such computationally modeled molecules may not adopt the intended bound conformation outside the binding pocket, one challenge is to ensure that the designed ligands adopt similar low energy conformations both inside and outside of the binding pocket. Computational chemistry methods and conformational preferences of small molecules from PDB and Cambridge Structural Database (CSD) can be used to predict the bound structures of the designed molecules. Herein, we review applications of conformational control in structure-based drug design using selected examples from the recent medicinal chemistry literature. The main purpose is to highlight some intriguing conformational features that can be applied to other drug discovery programs.     

A 45‐bp Insertion/Deletion Polymorphism of Uncoupling Protein 2 in Relation to Obesity in Tongans

We compared the current prevalence of increased BMI and type 2 diabetes in a representative group of Tongan subjects with measurements made in 1973, and we determined the distribution and possible interrelations with the UCP2 insertion/deletion (ins/del) polymorphism of these variables. We documented the BMI, glucose tolerance, and standard lipid variables in 1012 Tongan subjects (429 men and 583 women, ages 15 to 85 years) during 1998 and 2000 and compared the BMI findings with those of the 1973 survey. We also genotyped for the UCP2 ins/del polymorphism, assessed its association with obesity and type 2 diabetes, and compared its prevalence with those reported for other ethnic populations. The mean BMI ± SD was greatly increased in both men (30.2 ± 5.4 kg/m²) and women (33.8 ± 6.2 kg/m²), representing increases since 1973 of 11.9% and 19.4%, respectively. The genotype frequencies were 97% for the del/del genotype and 3% for the ins/del genotype; we found no ins/ins homozygotes. This distribution is strikingly different from those reported for white, South Indian, Pima Native‐American, and Asian populations (49 to 77% for del/del genotype). We conclude that there is a marked prevalence of obesity in Tonga, a prevalence that has increased since 1973. We also conclude that there is a unique, near‐uniform distribution of the UCP2 45‐bp ins/del polymorphism in Tongans. This may be the result of a founder effect and may be relevant to the prevalence of obesity and type 2 diabetes in Tonga.     

Childhood brain tumours,early infections and immune stimulation: A pooled analysis of the ESCALE and ESTELLE case-control studies (SFCE,France)

BackgroundFew studies have investigated whether early infections and factors potentially related to early immune stimulation might be involved in the aetiology of childhood brain tumours (CBT). In this study, we investigated the associations between CBT with early day-care attendance, history of early common infections, atopic conditions (asthma/wheezing, eczema, allergic rhinitis), early farm residence/visits and contact with animals.MethodsWe pooled data from two nationwide French case-control studies, the ESCALE and ESTELLE studies. Children with a CBT diagnosed between 1 and 14 years of age were identified directly from the French National Registry of Childhood Cancers, while population controls were recruited from telephone subscribers. Odds-ratios (OR) and 95% confidence intervals (CI) were estimated using logistic regression adjusted for potential confounders.ResultsThe analyses included 469 cases and 2719 controls. We found no association between attending a day-care centre (OR: 0.9, 95%CI: 0.7–1.2) or having had repeated common infections (OR: 0.9, 95%CI: 0.7–1.2) in the first year of life and the risk of CBT. There was also no association with a history of asthma/wheezing (OR: 0.8, 95%CI: 0.56–1.1). Farm visits (OR: 0.6, 95%CI: 0.5–0.8) as well as contact with pets (OR: 0.8, 95%CI: 0.6–1.0) in the first year of life were inversely associated with CBT.ConclusionsOur findings suggest a protective effect of early farm visits and contact with pets, but not with other markers of early immune stimulation. This might be related to immune stimulation but needs further investigation.