Bayesian doubly optimal group sequential designs for randomized clinical trials

When conducting a randomized comparative clinical trial, ethical, scientific or economic considerations often motivate the use of interim decision rules after successive groups of patients have been treated. These decisions may pertain to the comparative efficacy or safety of the treatments under study, cost considerations, the desire to accelerate the drug evaluation process, or the likelihood of therapeutic benefit for future patients. At the time of each interim decision, an important question is whether patient enrollment should continue or be terminated; either due to a high probability that one treatment is superior to the other, or a low probability that the experimental treatment will ultimately prove to be superior. The use of frequentist group sequential decision rules has become routine in the conduct of phase III clinical trials. In this dissertation, we will present a new Bayesian decision-theoretic approach to the problem of designing a randomized group sequential clinical trial, focusing on two-arm trials with time-to-failure outcomes. Forward simulation is used to obtain optimal decision boundaries for each of a set of possible models. At each interim analysis, we use Bayesian model selection to adaptively choose the model having the largest posterior probability of being correct, and we then make the interim decision based on the boundaries that are optimal under the chosen model. We provide a simulation study to compare this method, which we call Bayesian Doubly Optimal Group Sequential (BDOGS), to corresponding frequentist designs using either O'Brien-Fleming (OF) or Pocock boundaries, as obtained from EaSt 2000. Our simulation results show that, over a wide variety of different cases, BDOGS either performs at least as well as both OF and Pocock, or on average provides a much smaller trial. ^

Trends in utilization of computer tomography (CT) scans for victims of motor vehicle collision (MVCs) in a level I trauma center from 1996 to 2006

Purpose. To evaluate trends in the utilization of head, abdominal, thoracic and other body regions CTs in the management of victims of MVC at a level I trauma center from 1996 to 2006.^ Method. From the trauma registry, I identified patients involved in MVC's in a level I trauma center and categorized them into three age groups of 13-18, 19-55 and ≥56. I used International Classification of Disease (ICD-9-CM) codes to find the type and number of CTs examinations performed for each patient. I plotted the mean number of CTs per patient against year of admission to find the crude estimate of change in utilization pattern for each type of CT. I used logistic regression to assess whether repetitive CTs (≥ 2) for head, abdomen, thorax and other body regions were associated with age group and year of admission for MVC patients. I adjusted the estimates for gender, ethnicity, insurance status, mechanism and severity of injury, intensive care unit admission status, patient disposition (dead or alive) and year of admission.^ Results. Utilization of head, abdominal, thoracic and other body regions CTs significantly increased over 11-year period. Utilization of head CT was greatest in the 13-18 age group, and increased from 0.58 CT/patient in 1996 to 1.37 CT/patient in 2006. Abdominal CTs were more common in the ≥56+ age group, and increased from 0.33 CT/patient in 1996 to 0.72 CT/patient in 2006. Utilization of thoracic CTs was higher in the 56+ age group, and increased from 0.01 CT/patient in 1996 to 0.42 CT/patient in 2006. Utilization of other CTs did not change materially during the study period for adolescents, adults or older adults. In the multivariable analysis, after adjustment for potential confounders, repetitive head CTs significantly increased in the 13-18 age group (95% CI: 1.29-1.87, p=<0.001) relative to the 19-55 age group. Repetitive thoracic CT use was lower in adolescents (95% CI: 0.22-0.70, p=<0.001) relative to the 19-55 age group.^ Conclusion. There has been a substantial increase in the utilization of head, abdominal, thoracic and other CTs in the management of MVC patients. Future studies need to identify if increased utilization of CTs have resulted in better health outcome for these patients. ^

Phase I clinical trials in 56 patients with thyroid cancer: the M. D. Anderson Cancer Center experience.

INTRODUCTION: Thyroid cancer is the most common endocrine malignancy. The outcomes of patients with relapsed thyroid cancer treated on early-phase clinical trials have not been systematically analyzed. PATIENTS AND METHODS: We reviewed the records of consecutive patients with metastatic thyroid cancer referred to the Phase I Clinical Trials Program from March 2006 to April 2008. Best response was assessed by Response Evaluation Criteria in Solid Tumors. RESULTS: Fifty-six patients were identified. The median age was 55 yr (range 35-79 yr). Of 49 patients evaluable for response, nine (18.4%) had a partial response, and 16 (32.7%) had stable disease for 6 months or longer. The median progression-free survival was 1.12 yr. With a median follow-up of 15.6 months, the 1-yr survival rate was 81%. In univariate analysis, factors predicting shorter survival were anaplastic histology (P = 0.0002) and albumin levels less than 3.5 g/dl (P = 0.05). Among 26 patients with tumor decreases, none died (median follow-up 1.3 yr), whereas 52% of patients with any tumor increase died by 1 yr (P = 0.0001). The median time to failure in our phase I clinical trials was 11.5 months vs. 4.1 months for the previous treatment (P = 0.04). CONCLUSION: Patients with advanced thyroid cancer treated on phase I clinical trials had high rates of partial response and prolonged stable disease. Time to failure was significantly longer on the first phase I trial compared with the prior conventional treatment. Patients with any tumor decrease had significantly longer survival than those with any tumor increase.

What Have We Learned from Articles Published in the Family Preservation Journal?

This exploratory descriptive study presents a content analyses of all (N=22) Family Preservation Journal (FPJ) articles published from its inception (1995) until today. Three raters independently used an analysis template to ascertain trends from these articles and assessed information about their purposes, methods, and findings/ implications. The main findings were less than half of the articles were deemed as 'research'; few used standardized or outcome measures; none compared family preservation to another method; descriptive knowledge was more likely to be generated; and the articles were primarily targeted to practitioners and other researchers. Given the relatively short history ofFPJ, the majority of these findings were considered typical and consistent with the literature. The recommendations call for more comprehensive practice descriptions, more research, and more rigorous research-oriented studies.

Does cigarette smoking modify the association between change in body mass index during young adulthood and risk of coronary mortality during middle-age in men? Twenty-five year follow-up results from the Chicago Western Electric Study

Many persons in the U.S. gain weight during young adulthood, and the prevalence of obesity has been increasing among young adults. Although obesity and physical inactivity are generally recognized as risk factors for coronary heart disease (CHD), the magnitude of their effect on risk may have been seriously underestimated due to failure to adequately handle the problem of cigarette smoking. Since cigarette smoking causes weight loss, physically inactive cigarette smokers may remain relatively lean because they smoke cigarettes. We hypothesize cigarette smoking modifies the association between weight gain during young adulthood and risk of coronary heart disease during middle age, and that the true effect of weight gain during young adulthood on risk of CHD can be assessed only in persons who have not smoked cigarettes. Specifically, we hypothesize that weight gain during young adulthood is positively associated with risk of CHD during middle-age in nonsmokers but that the association is much smaller or absent entirely among cigarette smokers. The purpose of this study was to test this hypothesis. The population for analysis was comprised of 1,934 middle-aged, employed men whose average age at the baseline examination was 48.7 years. Information collected at the baseline examinations in 1958 and 1959 included recalled weight at age 20, present weight, height, smoking status, and other CHD risk factors. To decrease the effect of intraindividual variation, the mean values of the 1958 and 1959 baseline examinations were used in analyses. Change in body mass index ($\Delta$BMI) during young adulthood was the primary exposure variable and was measured as BMI at baseline (kg/m$\sp2)$ minus BMI at age 20 (kg/m$\sp2).$ Proportional hazards regression analysis was used to generate relative risks of CHD mortality by category of $\Delta$BMI and cigarette smoking status after adjustment for age, family history of CVD, major organ system disease, BMI at age 20, and number of cigarettes smoked per day. Adjustment was not performed for systolic blood pressure or total serum cholesterol as these were regarded as intervening variables. Vital status was known for all men on the 25th anniversary of their baseline examinations. 705 deaths (including 319 CHD deaths) occurred over 40,136 person-years of experience. $\Delta$BMI was positively associated with risk of CHD mortality in never-smokers, but not in ever-smokers (p for interaction = 0.067). For never-smokers with $\Delta$BMI of stable, low gain, moderate gain, and high gain, adjusted relative risks were 1.00, 1.62, 1.61, and 2.78, respectively (p for trend = 0.010). For ever-smokers, with $\Delta$BMI of stable, low gain, moderate gain, and high gain, adjusted relative risks were 1.00, 0.74, 1.07, and 1.06, respectively (p for trend = 0.422). These results support the research hypothesis that cigarette smoking modifies the association between weight gain and CHD mortality. Current estimates of the magnitude of effect of obesity and physical inactivity on risk of coronary mortality may have been seriously underestimated due to inadequate handling of cigarette smoking. ^

Racial disparities in CD4 counts at initial HIV-1 diagnosis: Analysis of the Adult Spectrum of HIV disease dataset and public health implications

Introduction. The HIV/AIDS disease burden disproportionately affects minority populations, specifically African Americans. While sexual risk behaviors play a role in the observed HIV burden, other factors including gender, age, socioeconomics, and barriers to healthcare access may also be contributory. The goal of this study was to determine how far down the HIV/AIDS disease process people of different ethnicities first present for healthcare. The study specifically analyzed the differences in CD4 cell counts at the initial HIV-1 diagnosis with respect to ethnicity. The study also analyzed racial differences in HIV/AIDS risk factors. ^ Methods. This is a retrospective study using data from the Adult Spectrum of HIV Disease (ASD), collected by the City of Houston Department of Health. The ASD database contains information on newly reported HIV cases in the Harris County District Hospitals between 1989 and 2000. Each patient had an initial and a follow-up report. The extracted variables of interest from the ASD data set were CD4 counts at the initial HIV diagnosis, race, gender, age at HIV diagnosis and behavioral risk factors. One-way ANOVA was used to examine differences in baseline CD4 counts at HIV diagnosis between racial/ethnic groups. Chi square was used to analyze racial differences in risk factors. ^ Results. The analyzed study sample was 4767. The study population was 47% Black, 37% White and 16% Hispanic [p<0.05]. The mean and median CD4 counts at diagnosis were 254 and 193 cells per ml, respectively. At the initial HIV diagnosis Blacks had the highest average CD4 counts (285), followed by Whites (233) and Hispanics (212) [p<0.001 ]. These statistical differences, however, were only observed with CD4 counts above 350 [p<0.001], even when adjusted for age at diagnosis and gender [p<0.05]. Looking at risk factors, Blacks were mostly affected by intravenous drug use (IVDU) and heterosexuality, whereas Whites and Hispanics were more affected by male homosexuality [ p<0.05]. ^ Conclusion. (1) There were statistical differences in CD4 counts with respect to ethnicity, but these differences only existed for CD4 counts above 350. These differences however do not appear to have clinical significance. Antithetically, Blacks had the highest CD4 counts followed by Whites and Hispanics. (2) 50% of this study group clinically had AIDS at their initial HIV diagnosis (median=193), irrespective of ethnicity. It was not clear from data analysis if these observations were due to failure of early HIV surveillance, HIV testing policies or healthcare access. More studies need to be done to address this question. (3) Homosexuality and bisexuality were the biggest risk factors for Whites and Hispanics, whereas for Blacks were mostly affected by heterosexuality and IVDU, implying a need for different public health intervention strategies for these racial groups. ^

Persistently elevated troponin I in heart failure patients - a suitable prognosticator and a possible boon to public health?

Baseline elevation of troponin I (TnI) has been associated with worse outcomes in heart failure (HF). However, the prevalence of persistent TnI elevation and its association with clinical outcomes has not been well described. HF is a major public health issue due to its wide prevalence and prognosticators of this condition will have a significant impact on public health. Methods: A retrospective study was performed in 510 patients with an initial HF admission between 2002 to 2004, and all subsequent hospital admissions up to May 2009 were recorded in a de-identified database. Persistent TnI elevation was defined as a level ≥0.05 ng/ml on ≥3 HF admissions. Baseline characteristics, hospital readmissions and all cause mortality were compared between patients with persistent TnI elevation (Persistent), patients with no persistence of TnI (Nonpersistent) and patients who had less than three hospital admissions (admission <3) groups. Also the same data was analyzed using the mean method in which the mean value of all recorded troponin values of each patient was used to define persistence i.e. patients who had a mean troponin level ≥0.05 ng/ml were classified as persistent. Results: Mean age of our cohort was 68.4 years out of which 99.6% subjects were male, 62.4% had ischemic HF. 78.2% had NYHA class III to IV HF, mean LVEF was 25.9%. Persistent elevation of TnI was seen in 26% of the cohort and in 66% of patients with more than 3 hospital admissions. Mean TnI level was 0.67 ± 0.15 ng/ml in the 'Persistent' group. Mean TnI using the mean method was 1.11 ± 7.25 ng/ml. LVEF was significantly lower in persistent group. Hypertension, diabetes, chronic renal insufficiency and mean age did not differ between the two groups. 'Persistent' patients had higher mortality (HR = 1.26, 95% CI = 0.89–1.78, p = 0.199 when unadjusted and HR = 1.29, 95% CI = 0.89–1.86, p = 0.176 when adjusted for race, LVEF and ischemic etiology) HR for mortality in persistent patients was 1.99 (95% CI = 1.06–3.73, p = 0.03) using the mean method. The following results were found in those with ischemic cardiomyopathy (HR = 1.44034, 95% CI = 0.92–2.26, p = 0.113) and (HR = 1.89, 95% CI = 1.01–3.55, p = 0.046) by using the mean method. 2 out of three patients with HF who were readmitted three or more times had persistent elevation of troponin I levels. Patients with chronic persistence of troponin I elevation showed a trend towards lesser survival as compared to patients who did not have chronic persistence, however this did not reach statistical significance. This trend was seen more among ischemic patients than non ischemic patients, but did not reach statistical significance. With the mean method, patients with chronic persistence of troponin I elevation had significantly lesser survival than those without it. Also ischemic patients had significantly lesser survival than non ischemic patients. ^