Faculty Bibliography

Public health interventions are often designed to target communities defined either geographically (e.g. cities, counties) or socially (e.g. schools or workplaces). The group randomized trial (GRT) is regarded as the gold standard for evaluating these interventions. However, community leaders may object to randomization as some groups may be denied a potentially beneficial intervention. Under a regression discontinuity design (RDD), individuals may be assigned to treatment based on the levels of a pretest measure, thereby allowing those most in need of the treatment to receive it. In this article, we consider analysis, power, and sample size issues in applying the RDD and related cutoff designs in community-based intervention studies. We examine the power of these designs as a function of intraclass correlation, number of groups, and number of members per group and compare results to the traditional GRT.

Adjuvant surgical oophorectomy is an effective and remarkably cost effective treatment for premenopausal women with hormone receptor positive operable breast cancer. Previously published secondary analyses indicated a survival benefit for patients whose surgery was performed in the luteal phase of the menstrual cycle as opposed to the follicular. This study utilizes additional follow-up and more fully examines this hypothesis and the general implications of long-term follow-up on trial design. Beginning in 1993 we recruited women to a multicenter randomized clinical trial of adjuvant surgical oophorectomy and tamoxifen for 5 years. We recorded the reported day 1 of the patients' last menstrual cycle on the day of their adjuvant surgery. We conducted secondary analyses of the association of history-estimated luteal or follicular phase oophorectomy surgery with disease-free and overall survival. In multivariable Cox analyses, disease-free survival (DFS) exhibited a positive trend and overall survival (OS) showed a significant improvement in patients whose surgery was estimated to have occurred in the luteal phase of the menstrual cycle compared to the follicular (HR for DFS: 0.66, 95% CI: 0.37-1.16; HR for OS: 0.49, 95% CI: 0.27-0.88). From the hazard function plots, it appears that the luteal phase surgery effect on DFS diminishes after 6 years of follow-up. In conclusion, adjuvant surgical oophorectomy during the luteal phase of the menstrual cycle resulted in a reduced hazard of recurrence as compared to oophorectomy in the follicular phase during the first 5.5 years of follow-up. The practical and biological implications of these findings deserve rigorous evaluation in clinical trials.

BACKGROUND:During the recruitment phase of a randomized breast cancer trial, investigating the time to recurrence, we found a strong suggestion that the failure probabilities used at the design stage were too high. Since most of the methodological research involving sample size re-estimation has focused on normal or binary outcomes, we developed a method which preserves blinding to re-estimate sample size in our time to event trial. PURPOSE/OBJECTIVE:A mistakenly high estimate of the failure rate at the design stage may reduce the power unacceptably for a clinically important hazard ratio. We describe an ongoing trial and an application of a sample size re-estimation method that combines current trial data with prior trial data or assumes a parametric model to re-estimate failure probabilities in a blinded fashion. METHODS:Using our current blinded trial data and additional information from prior studies, we re-estimate the failure probabilities to be used in sample size re-calculation. We employ bootstrap re-sampling to quantify uncertainty in the re-estimated sample sizes. RESULTS:At the time of re-estimation data from 278 patients were available, averaging 1.2 years of follow up. Using either method, we estimated a sample size increase of zero for the hazard ratio because the estimated failure probabilities at the time of re-estimation differed little from what was expected. We show that our method of blinded sample size re-estimation preserves the type I error rate. We show that when the initial guess of the failure probabilities are correct, the median increase in sample size is zero. LIMITATIONS/CONCLUSIONS:Either some prior knowledge of an appropriate survival distribution shape or prior data is needed for re-estimation. CONCLUSIONS:In trials when the accrual period is lengthy, blinded sample size re-estimation near the end of the planned accrual period should be considered. In our examples, when assumptions about failure probabilities and HRs are correct the methods usually do not increase sample size or otherwise increase it by very little. Clinical Trials 2010; 7: 219. http://ctj.sagepub.com.

OBJECTIVE:We describe factors, in the context of the Social Determinants of Health model, associated with receiving Pap smears within risk-appropriate guidelines (i.e., guidelines that specify screening intervals based upon a woman's individual risk of developing cervical cancer). METHODS:Completed in June 2006, we conducted a cross-sectional survey of women from 14 health clinics in Ohio Appalachia pertaining to psychosocial, demographic, biological, and health-related factors. A logistic regression model was constructed to predict whether or not a woman was within risk-appropriate cervical cancer screening guidelines. RESULTS:Of 562 women with a date of last Pap smear, 380 (68%) were within risk-appropriate guidelines. Logistic regression showed that, compared to women with low-level SES, women with middle- and high-level SES had 3.39 [1.85, 6.21] and 3.86 [2.03, 7.34] times the odds, respectively, of being within risk-appropriate guidelines. Odds of being within guidelines increased 1.09 [1.04, 1.15] fold for each decrease of one major life event. Additionally, women that were financially better off or financially worse off than their parents at the same age had lower odds (0.41 [0.23, 0.73] and 0.49 [0.24, 0.98], respectively) of being within guidelines than women who reported their finances were the same as their parents. Results also showed an interaction between marital status and age at first intercourse (p=0.001). CONCLUSION/CONCLUSIONS:The results suggest an impact of psychosocial factors on Pap smear testing behaviors, and illustrate the need to examine risk-appropriate interventions to improve screening.

BACKGROUND:Screening has become one of our best tools for early detection and prevention of cancer. The group-randomized trial is the most rigorous experimental design for evaluating multilevel interventions. However, identifying the proper sample size for a group-randomized trial requires reliable estimates of intraclass correlation (ICC) for screening outcomes, which are not available to researchers. We present crude and adjusted ICC estimates for cancer screening outcomes for various levels of aggregation (physician, clinic, and county) and provide an example of how these ICC estimates may be used in the design of a future trial. METHODS:Investigators working in the area of cancer screening were contacted and asked to provide crude and adjusted ICC estimates using the analysis of variance method estimator. RESULTS:Of the 29 investigators identified, estimates were obtained from 10 investigators who had relevant data. ICC estimates were calculated from 13 different studies, with more than half of the studies collecting information on colorectal screening. In the majority of cases, ICC estimates could be adjusted for age, education, and other demographic characteristics, leading to a reduction in the ICC. ICC estimates varied considerably by cancer site and level of aggregation of the groups. CONCLUSIONS:Previously, only two articles had published ICCs for cancer screening outcomes. We have complied more than 130 crude and adjusted ICC estimates covering breast, cervical, colon, and prostate screening and have detailed them by level of aggregation, screening measure, and study characteristics. We have also demonstrated their use in planning a future trial and the need for the evaluation of the proposed interval estimator for binary outcomes under conditions typically seen in GRTs.

We review design and analytic methods available for multilevel interventions in cancer research with particular attention to study design, sample size requirements, and potential to provide statistical evidence for causal inference. The most appropriate methods will depend on the stage of development of the research and whether randomization is possible. Early on, fractional factorial designs may be used to screen intervention components, particularly when randomization of individuals is possible. Quasi-experimental designs, including time-series and multiple baseline designs, can be useful once the intervention is designed because they require few sites and can provide the preliminary evidence to plan efficacy studies. In efficacy and effectiveness studies, group-randomized trials are preferred when randomization is possible and regression discontinuity designs are preferred otherwise if assignment based on a quantitative score is possible. Quasi-experimental designs may be used, especially when combined with recent developments in analytic methods to reduce bias in effect estimates.

INTRODUCTION/BACKGROUND:Data on overweight and obesity prevalence among children enable state and local officials to develop, target, fund, and evaluate policies and programs to address childhood overweight. During the 2004-2005 school year, the Ohio Department of Health (ODH) conducted surveillance of elementary school-aged children through coordination with the ODH oral health survey to create a system that would provide county and state estimates of obesity and overweight prevalence. METHODS:We used a stratified, cluster-sampling survey design. Schools were considered clusters and were sampled from strata determined by their county and by their participation rate in the Free and Reduced Price Meal program. We selected public elementary schools by probability proportional to size sampling without replacement. We requested consent from the guardian or parent of each third-grade student. Trained health care professionals used state-purchased equipment to weigh students and measure their height. We removed implausible observations and calculated sex-specific, body mass index (BMI)-for-age percentiles using Centers for Disease Control and Prevention growth charts. RESULTS:Of eligible schools, 374 agreed to height and weight screening; 41 were considered substitutes. Of 26,590 enrolled students, 17,557 (66.0%) returned consent forms, and 15,209 (57.2%) provided consent. BMI estimates were generated for 14,451 students, resulting in an overall response rate of 54.3%. The overall oral health response rate was 52.8%. CONCLUSION/CONCLUSIONS:By adding BMI screening to Ohio's third-grade oral health survey and incorporating trained volunteer screeners, the ODH successfully implemented overweight and obesity surveillance using minimal resources. Future efforts should focus on improving student response rate.

BACKGROUND:Mutations in the P53 gene are among the most common genetic abnormalities in human lung cancer. Codon 273 in the sequence-specific DNA binding domain is one of the most frequently mutated sites. METHODOLOGY/METHODS:To investigate the role of mutant p53 in lung tumorigenesis, a lung specific p53(273H) transgenic mouse model was developed. Rates of lung cancer formation in the transgenic animals and their littermates were evaluated by necropsy studies performed in progressive age cohorts ranging from 4 to 24 months. In order to establish the influence of other common genetic abnormalities in lung tumor formation in the animals, K-Ras gene mutation and p16INK4a (p16) promoter methylation were evaluated in a total of 281 transgenic mice and 189 non-transgenic littermates. PRINCIPAL FINDINGS/RESULTS:At the age extremes of 4-12 and 22-24 months no differences were observed, with very low prevalence of tumors in animals younger than 12 months, and a relatively high prevalence at age 22 months or older. However, the transgenic mice had a significant higher lung tumor rate than their non-transgenic counterparts during the age of 13-21 months, suggesting an age-related shift in lung tumor formation induced by the lung-specific expression of the human mutant p53. Histopathology suggested a more aggressive nature for the transgenic tumors. Older mice (>13 months) had a significantly higher rate of p16 promoter methylation (17% v 82%). In addition, an age related effect was observed for K-Ras codons 12 or 13 mutations, but not for codon 61 mutations. CONCLUSIONS/SIGNIFICANCE/CONCLUSIONS:These results would suggest that the mutant p53(273H) contributes to an acceleration in the development of spontaneous lung tumors in these mice. Combination with other genetic and epigenetic alterations occurring after the age of 13 months is intimately linked to its oncogenic potential.

We hypothesized that IFN-alpha would enhance the apoptotic activity of bortezomib on melanoma cells. Combined treatment with bortezomib and IFN-alpha induced synergistic apoptosis in melanoma and other solid tumor cell lines. Apoptosis was associated with processing of procaspase-3, procaspase-7, procaspase-8, and procaspase-9 and with cleavage of Bid and poly(ADP-ribose) polymerase. Bortezomib plus IFN-alpha was effective at inducing apoptosis in melanoma cells that overexpressed Bcl-2 or Mcl-1, suggesting that this treatment combination can overcome mitochondrial pathways of cell survival and resistance to apoptosis. The proapoptotic effects of this treatment combination were abrogated by a caspase-8 inhibitor, led to increased association of Fas and FADD before the onset of cell death, and were significantly reduced in cells transfected with a dominant-negative FADD construct or small interfering RNA targeting Fas. These data suggest that bortezomib and IFN-alpha act through the extrinsic pathway of apoptosis via FADD-induced caspase-8 activation to initiate cell death. Finally, bortezomib and IFN-alpha displayed statistically significant antitumor activity compared with either agent alone in both the B16 murine model of melanoma and in athymic mice bearing human A375 xenografts. These data support the future clinical development of bortezomib and IFN-alpha for malignant melanoma.