Papers by Theodore Colton
American Journal of Hypertension, Apr 1, 1994
American Journal of Epidemiology, 1985
Using data from the Diethylstilbestrol-Adenosis (DESAD) Project, a study of the effects of diethy... more Using data from the Diethylstilbestrol-Adenosis (DESAD) Project, a study of the effects of diethylstilbestrol (DES) exposure during fetal life, the authors compared prenatal records with obstetric history from mother's questionnaires completed 10 or more years after the birth of their daughters. Except for the history of hospitalization and trunk x-ray, no differences were observed in agreement (questionnaire compared with record) between the group of DES-exposed mothers identified through review of their prenatal records and the comparison group of mothers who were not exposed. The authors also compared data from mothers of DES-exposed daughters who initiated their own enrollment in the study (walk-ins and referrals). To obtain prenatal records for these women, physicians were contacted. They would usually supply drug exposure data but not the other obstetric history requested. Mothers of these walk-ins and referrals had slightly better agreement between questionnaire and records when compared with the two groups identified by review of prenatal records. In general, there was good to excellent agreement for all groups when mothers' recall of personal history (past miscarriage, past pregnancy, etc.) was compared with their medical records. However, for medical intervention such as drugs and x-rays, agreement was poor. Of the DES-exposed mothers identified through review of their prenatal records, 29% could not remember whether they took DES. An additional 8% said they did not take DES when it was recorded in their charts.
Clinical Pharmacology & Therapeutics, 1968
I have read with great interest Dr. Mainland'so "Statistical ward rounds-4" and I a... more I have read with great interest Dr. Mainland'so "Statistical ward rounds-4" and I am answering his call regarding sequential medical trials. I am not armed with barrels of disks and stacks of playing cards, but I would like to respond to some of his indictments of sequential trials. Obviously, the use of sequential trials is not a yes-orno issue and I certainly do not take the position that sequential procedures are the answer to all clinical trials. I think they are useful and applicable in many situations. In others they are not appropriate. I agree with Dr. Mainland in some of his criticisms of the method but I think he's painted much too dismal a picture and I do take issue with several of his statements. I would like to discuss these. If there is one central theme to my remarks it is that the charges against sequential analysis apply equally well to a fixed sample size trial, with sample size being determined from predetermined Type I and Type II errors, as indicated on p. 617 of Dr. Mainland's paper. Within the
Journal of Chronic Diseases, 1971
Wiley StatsRef: Statistics Reference Online, 2014
Background: A synthetic estrogen, diethylstilbestrol (DES), was widely prescribed to pregnant wom... more Background: A synthetic estrogen, diethylstilbestrol (DES), was widely prescribed to pregnant women during the 1950s and 1960s but was later discovered to be associated with an increased risk of clear-cell carcinoma of the vagina and cervix in female offspring. DES has not been linked to other cancers in female offspring, but studies of other prenatal factors such as twin gestation and pre-eclampsia have indicated that in-utero estrogen levels may influence breast cancer risk. We evaluated the relation of in-utero DES exposure to the risk of adult breast cancer. Methods: A cohort of 4821 exposed women and 2095 unexposed women, most of whom were first identified in the mid-1970s, were followed by mailed questionnaires for an average of 19 years. Reported cancer outcomes were validated by medical record review. Breast cancer incidence in DES-exposed daughters was compared with cancer incidence in unexposed daughters with use of Poisson regression analysis, adjusting for year of birth, age at menarche, age at first birth, and number of births. Findings: The rate ratio for incidence of invasive breast cancer in exposed versus unexposed women was 1.4 (95% confidence interval (CI) ¼ 0.7-2.6). DES exposure was not associated with an increased risk of breast cancer in women under 40 years, but among women aged 40 and older the rate ratio was 2.5 (95% CI ¼ 1.0-6.3). The rate ratio for the association of DES exposure with estrogen receptor-positive tumors was 1.9 (95% CI ¼ 0.8-4.5). Interpretation: While not statistically significant, the overall 40% excess risk, arising exclusively from the subset of estrogen receptor-positive cases, raises a concern calling for continued investigation.
Encyclopedia of Biostatistics, 2005
Encyclopedia of Biostatistics, 2005
Encyclopedia of Biostatistics, 2005
Statistics in Medicine, 1989
This paper discusses a two-period crossover design for the comparison of two active treatments (A... more This paper discusses a two-period crossover design for the comparison of two active treatments (A and B) and placebo (P) for relief of recurrent symptoms of a chronic health disorder. It is based on blocks of ten patients for which the treatment sequences A:B and B:A are each assigned to three patients and the sequences A:P, P:A, B:P, and P:B are each assigned to one patient; thus, treatment periods have a 2:2:1 allocation for A, B, and P. The principal model for analysis of this design involves additive subject effects, period effects, and treatment effects. Analysis of within-patient differences provides an estimate of the comparison between active treatments with variance (2vw/7r) and an estimate of the comparison between an active treatment and placebo with variance (4vw/7r); here vw is the within-patient variance and r is the number of blocks of ten patients. Analyses which address carryover effects and treatment effects adjusted for carryover effects are also described. An example using simulated data on relief of recurrent gastrointestinal pain illustrates the methodology.
Ophthalmology, 1998
The purpose of the study was to determine whether body mass index (BMI) is an independent risk fa... more The purpose of the study was to determine whether body mass index (BMI) is an independent risk factor for the development of nuclear, cortical, or posterior subcapsular lens opacities. A cohort study. Eye examinations were conducted on surviving members of the Framingham Heart Study Cohort from 1973 to 1975 (Framingham Eye Study I) and again from 1986 to 1989 (Framingham Eye Study II). Data from the Framingham Heart Study, including weight measurements collected biennially from 1948, were used to examine associations between BMI (mean BMI across examinations, slope of BMI over time, and fluctuations in BMI) and the development of lens opacities. This analysis included 714 individuals, aged 52-80 years, who were free of lens opacities at the first eye examination. Development of nuclear, cortical, and posterior subcapsular lens opacities. A total of 444 persons developed lens opacities during the approximately 13 years between eye examinations. In logistic regression analyses that controlled for age, sex, education, diabetes, and smoking, the risk of developing cortical opacity increased with higher BMI at the time of the first eye examination (P = 0.002). Risk of cortical opacities also increased, at a borderline level of significance, with higher average BMI (P = 0.09) across examinations and increasing BMI levels over time (P = 0.10). There was a strong association between increasing BMI over time and the development of posterior subcapsular lens opacities (P = 0.002). No associations were found for nuclear lens opacities. Although the mechanism explaining the association is unclear, these findings suggest that BMI, a potentially modifiable characteristic, is associated with the development of cortical and posterior subcapsular lens opacities.
New England Journal of Medicine, 1976
The Lancet, 1978
A review of death certificates in New Hampshire, Maine, and Massachusetts for 1959-77 yielded a t... more A review of death certificates in New Hampshire, Maine, and Massachusetts for 1959-77 yielded a total of 1722 deaths among former workers at the Portsmouth Naval Shipyard where nuclear submarines are repaired and refuelled. Next of kin were contacted for 592. All deaths under age 80 were classified as being in former nuclear or non-nuclear workers depending on information supplied by next of kin. With U.S. age-specific proportional cancer mortality for White males as a standard, the observed/expected ratio of leukaemia deaths was 5.62 (6 observed, 1.1 expected) among the 146 former nuclear workers. For all cancer deaths, this ratio was 1.78. Among non-nuclear workers there was no statistically significant increase in proportional mortality from either leukaemia or from all cancers. The excess proportional leukaemia and cancer mortality among nuclear workers exceeds predictions based on previous data of radiation effects in man.
Journal of the American Statistical Association, 1960
Abstract For samples from normal populations with unknown means and unknown common standard devia... more Abstract For samples from normal populations with unknown means and unknown common standard deviation, but where an upper bound to the standard deviation is known, tests based on the normal distribution (when dealing with one population) and on the Chi-square distribution (when dealing with several populations) are suggested as alternatives to, respectively, the t-test and the F-test. The tests suggested use the upper bound to the standard deviation in place of the sample estimate. Although the tests suggested have a smaller first kind of error, their power can, under certain conditions, exceed the power of the standard t-test and F-test. Tables are constructed which for degrees of freedom, first kind of error, and second kind of error give, percentagewise, the range within which the upper bound for the standard deviation must lie if these tests are to be more powerful than the respective t-test and F-test.
Journal of the American Statistical Association, 1976
Plans Compared with and Wald SPRT Plans Within the Neyman-Pearson framework of hypothesis testing... more Plans Compared with and Wald SPRT Plans Within the Neyman-Pearson framework of hypothesis testing with fixed-error-level specifications, two-stage designs are obtained such that sample size is minimized when the alternative hypothesis is true. Normally distributed variates with known variance and binomially distributed variates are considered. It is shown that when the alternative hypothesis is true, these optimal two-stage designs generally achieve between one-half and two-thirds of the ASN differential between the two extremes of analogous fixed-sample designs (maximum ASN) and item-by-item Wald SPRT design (minimum ASN when alternative hypothesis is true). type plans, namely, two-stage plans, and their ASN properties relative to these two extremes. In a two-stage plan one analyzes the data on only two occasions, after collection of n 1 first-stage observations, and, if the decision is to continue, after collection of an additional n2 second-stage observations. The ASN function for the two-stage plan may be seen very simply to be n1 + n 2 Pr (second stage required). (1.1)
Biometrics, 1965
In a previous paper (Colton [1963]) I considered a cost function approach to the design of a clin... more In a previous paper (Colton [1963]) I considered a cost function approach to the design of a clinical trial for the comparison of two medical treatments. I examined two types of plans, one with a fixed sample size and one with random pairing and sequential, one-pairat-a-time, observations. Anscombe [1963] considered a similar formulation, dealing with the sequential case. The results of these investigations indicated that the optimal sequential plan led to an overall smaller cost (or, alternatively, an overall greater gain) than the corresponding optimal fixed sample size plan. Numerical results showed that the overall net gain with the optimal sequential plan could be as much as 25 percent more than that for the corresponding optimal fixed sample size plan. Here I consider intermediate plans: in particular, two possible two-stage plans. The appropriate cost functions are determined, the optimal plans are obtained, and the corresponding net gains are compared with each other and with the previously reported fixed sample size and sequential results. The results show that although the twostage plans proposed are quite different, their optimal net gains are similar. The overall net gain of the two-stage optimal plans is at most 13 percent more than that of the corresponding optimal fixed sample size plan. However, the optimal two-stage plan can account for as much as 50 percent of the difference in overall net gain between the fixed sample size and sequential plans. Hence, within this cost formulation it appears that by going from the one extreme of a fixed sample size plan to a two-stage plan one can achieve about half of the gain that is attainable by the opposite extreme of a pair-by-pair fully sequential plan.
Archives of Environmental Health: An International Journal, 1966
Anesthesia & Analgesia, 1979
Gastric volume and pH were studied immediately after induction of anesthesia and endotracheal int... more Gastric volume and pH were studied immediately after induction of anesthesia and endotracheal intubation in 101 elective surgical patients. Of 44 patients not given cimetidine, 82% had a gastric pH less than 2.5 with a mean pH of 1.6; 45% of these patients had a gastric aspirate pH less than 2.5 associated with a volume exceeding 25 ml. In 57 patients premedicated with intravenous cimetidine at variable intervals (1 5 to 60 minutes) prior to induction of anesthesia, a significant time-dependent increase was noted in gastric pH (p c 0.001) together with a decline in gastric volume (p c 0.001 1. Of the patients given intravenous cimetidine (mean 4.5 mg/kg) 45 minutes prior to induction of anesthesia, 90% had a gastric pH greater than 2.5. The increase in gastric pH after cimetidine administration would result in a reduced chemical pulmonary reaction should aspiration occur during induction of anesthesia.
The American Statistician, 1982
Abstract This report summarizes the evidence regarding adverse effects of diethylstilbestrol (DES... more Abstract This report summarizes the evidence regarding adverse effects of diethylstilbestrol (DES) administration during pregnancy. From the association in 1969 of in utero DES exposure with subsequent increased risk of vaginal cancer in young women, the report includes the state of evidence regarding in utero DES exposure with other female genital tract abnormalities, with male genital tract abnormalities, with subsequent adverse reproductive experiences, and, among the exposed mothers, with increased risk of breast and other gynecologic cancers.
Uploads
Papers by Theodore Colton