Search PubMed⌕ Search

Biomedical subjects

T Holford

Publications and source records attributed to T Holford.

At least 19 recordsLinked to original sources
PubMed1997

Risk of tongue cancer associated with tobacco smoking and alcohol consumption: a case-control study.

Recent studies indicate that cancer of the tongue is increasing rapidly among the younger population in many parts of the world. Few studies, however, have directly examined the risk factors for the disease. A case-control study was conducted in Beijing, China to investigate risk factors for tongue cancer. A total of 111 cases and 111 controls aged 20-80 years were included in this study. The results show that risk of tongue cancer is significantly elevated among ex-smokers (OR = 2.24, 95% CI = 1.09-4.62) and among current smokers (OR = 2.73, 95% CI = 1.26-5.91). The risk increases with increasing tobacco consumption, as reflected by both cigarette equivalents smoked per day and lifetime pack-years of tobacco smoking. Quitting smoking was associated with a reduction of the risk of tongue cancer. The numbers of cases in the study, however, is small, preventing further analyses during the years after quitting smoking. Overall, alcohol drinking was not found to be significantly associated with the risk of tongue cancer in this study (OR = 1.20, 95% CI = 0.58-2.50 for current drinkers). However, a marginally significant association was found for those who drank spirits at least 5 days a week (OR = 2.34, 95% CI = 0.90-6.06). A suggestion of effect modification for smoking and alcohol drinking was observed in this study.

Adult↗
PubMed1996

A tree-based method of analysis for prospective studies.

Prospective studies often involve rare events as study outcomes, and a primary concern is to identify risk factors and risk groups associated with the outcomes. We discuss practical solutions to risk factor analyses in prospective studies and address strategies to determine tree structures, to estimate relative risks, and to manage missing data in connection with some important epidemiologic problems. Some of the basic ideas for our strategies follow from work of Breiman, Friedman, Olshen, and Stone, although we propose extensions to their methods to resolve some practical problems that arise in implementation of these methods in epidemiologic studies. To illustrate these ideas, we analyse low birthweight associated risk factors with use of a data set from the Yale Pregnancy Outcome Study.

Bias↗
PubMed1996

Second solid tumors in patients with Hodgkin's disease cured after radiation or chemotherapy plus adjuvant low-dose radiation.

PURPOSE: Late solid tumors (STs) are a significant cause of morbidity and mortality in long-term survivors of Hodgkin's disease. To investigate the carcinogenic potential of two different therapeutic approaches, we measured the relative risk (RR) of STs in patients with early-stage disease cured after primary full-dose (approximately 40 Gy) radiation therapy (RT) and in patients with advanced disease who were treated with chemotherapy followed by low-dose (15 to 30 Gy) involved-field radiation (CMT). PATIENTS AND METHODS: Because therapy-induced STs generally begin after a latency period of 5 to 10 years, we restricted our analysis to patients treated before 1986 who achieved durable remissions. Patients who required salvage chemotherapy or who died of Hodgkin's disease were excluded from analysis. The RR of STs was calculated by dividing the observed number of cases by the expected number in a matched population from the Connecticut Tumor Registry. The actuarial incidence of STs was also measured. RESULTS: A total of 197 patients formed the RT group and 116 the CMT group. The median follow-up period in the RT group was 12.8 years, versus 13.5 years in the CMT group. The overall RR of STs in the CMT group was 1.5 (95% confidence interval [CI], 0.6 to 3.5; P = .122). There were no cases of lung or breast cancer. In the RT group, the overall RR of STs was 3.3 (95% CI, 2.0 to 5.3; P < .001). There were seven cases of lung cancer (RR = 10.8; 95% CI, 5.3 to 22.2; P < .001) and two cases of breast cancer (RR = 2; 95% CI, 0.6 to 7.4; P = .07). All six benign tumors occurred in the RT group. CONCLUSION: In patients cured by initial treatment for Hodgkin's disease, RT was associated with a statistically significant increase in STs, particularly lung cancer. CMT was not associated with a significant increase in STs. These data may have important implications for the design of newer therapies for early-stage Hodgkin's disease.

Actuarial Analysis↗
PubMed1995

Second solid malignancies after combined modality therapy for Hodgkin's disease.

PURPOSE: To determine the actuarial incidence (AI) and relative risk (RR) of second solid malignancies (SSM; solid tumors and non-Hodgkin's lymphoma) in patients with Hodgkin's disease who were treated with chemotherapy and adjuvant, low-dose radiation (combined modality therapy; CMT). PATIENTS AND METHODS: From 1969 to 1983, 102 patients with previously untreated advanced Hodgkin's disease (group A) and 81 patients with recurrent disease after radiation (group B) were treated with CMT. Patients were observed for the development of solid tumors (ST) and non-Hodgkin's lymphoma (NHL), and the AI and RR were calculated. RESULTS: Nearly half of the patients entering remission were observed for greater than 15 years. At 20 years, the AI for SSM was 12% in group A versus 41% in group B (P = .009). The overall RR for developing a ST in group A was 1.88 (not significant) versus 8.84 in group B (95% confidence interval, 5.3 to 15.4). The difference in the RR between groups A and B was significant (P < .001). The RR for developing NHL was significantly increased in both groups, but the difference between groups was not significant. CONCLUSION: Previously untreated patients with advanced disease who were treated with CMT (group A) had a modest but not significant increase in the RR of ST; however, patients treated with CMT for recurrent disease (group B) had a highly significant increase in the RR of ST. Possible explanations for the increase in ST in group B include more cumulative radiation or a greater carcinogenic effect of chemotherapy in previously irradiated patients, but it also is possible that the increase is due to a longer follow-up time.

Adolescent↗
PubMed1993

Analysis as-randomized and the problem of non-adherence: an example from the Veterans Affairs Randomized Trial of Coronary Artery Bypass Surgery.

In most randomized clinical trials not all patients adhere to the therapy to which they were randomly assigned. Instead, they may receive the therapy assigned to another treatment group, or a therapy different from any prescribed in the protocol. When non-adherence occurs, problems occur with the analysis comparing the treatments under study. Rigorous statistical principles require attributing outcome events to the original random treatment assignment ('intent-to-treat' analysis). Using data from the Veterans Administration Cooperative Study of Coronary Artery Bypass Surgery, we report the intent-to-treat analysis and apply four other methods of analysis for analysing non-adherers: 1. exclude non-adherers from analysis; 2. transfer them to the alternative treatment group at the time of randomization; 3. censor them at the time of treatment change, and 4. transfer them to the alternative treatment group at the time of treatment change. Inherent problems and biases of these four other methods are discussed.

Angina Pectoris↗
PubMed1991

A prospective longitudinal study of growth in twin gestations compared with growth in singleton pregnancies. I. The fetal head.

Since the available data on growth in twin gestations have been derived from retrospective cross-sectional studies with varying results, a prospective longitudinal study was initiated to assess fetal head growth in twin gestations as compared to singleton pregnancies. In uncomplicated twin gestations, growth of the fetal head, based on the increment in growth over time and the rate of growth throughout pregnancy, was found not to be significantly different than in singleton pregnancies. In light of these findings, current nomograms derived from measurements obtained in singleton pregnancies remain useful for evaluating fetal head growth in twin gestations.

Cephalometry↗
PubMed1991

A prospective longitudinal study of growth in twin gestations compared with growth in singleton pregnancies. II. The fetal limbs.

The assessment of fetal growth is crucial in twin gestations, since the information gained often has an impact on pregnancy management. The measurement of the fetal anatomy by ultrasound enables us to follow the growth and development of the fetus. However, the pattern of fetal growth in twin gestations has not yet been precisely characterized in prospective studies. In this light, we initiated a prospective longitudinal study and sonographically examined 35 patients with twin gestations every 3 weeks from the 15th week until delivery. Multiple biometric parameters were measured, including the femur length, humerus length, ulna length, and tibia length. The results of our study showed that growth of these long bones was not significantly different between twins A and B throughout gestation and that the growth velocity between twins and singletons was not significantly different. The incremental growth, although significantly less in twins than in singletons, was so small that it was judged not to be of clinical importance to warrant the generation of separate nomograms for the evaluation of growth in twin gestations.

Anthropometry↗
PubMed1991

Intent-to-treat analysis and the problem of crossovers. An example from the Veterans Administration coronary bypass surgery study.

In randomized clinical trials of treatment for ischemic heart disease that compare medical with surgical treatment, many persons initially assigned to medical therapy eventually receive surgical intervention. For example, in the three major trials of bypass grafting for stable angina, crossover rates from medical to surgical therapy were approximately 25% at 5 years. For this reason, the classic intent-to-treat analyses have been criticized for their inability to evaluate the "true" effect of treatment. In this article we emphasize the concept of "initial treatment" as it applies to intent-to-treat analyses and examine four proposed alternative methods of analysis based on adherence with survival data from the Veterans Administration Cooperative Study to illustrate the concepts. The alternative methods include (1) censoring crossovers when treatment changes, (2) transferring crossovers from the original to the new treatment group when treatment changes, (3) excluding all crossovers from analysis, and (4) counting crossovers from the date of randomization in the treatment ultimately received group. We point out the biases attendant on analyses based on adherence and reaffirm the validity of intent-to-treat analysis.

Actuarial Analysis↗
PubMed1990

Sonographic assessment of growth of the fetal head in diabetic pregnancies compared with normal gestations.

A longitudinal ultrasound study was conducted in 45 insulin-dependent diabetic patients who maintained good glycemic control (mean plasma glucose less than 120 mg/dl) throughout most of their pregnancy in order to assess growth of the fetal head in the presence of euglycemia. Patients with and without vasculopathy were found to be comparable with regard to their glycemic control, medical and obstetric complications, as well as incremental growth and the velocity of growth of the fetal biparietal diameter (BPD). When compared with the control group, the velocity of growth of the BPD was not significantly different throughout pregnancy. However, the actual increment in BPD growth remained less than that of the control fetuses, especially during the second trimester when a significant statistical difference was found. Possible explanations may include delayed ovulation, reduced growth velocity in the first trimester, or constitutionally smaller embryos among the diabetic group. The pattern of BPD growth among diabetics was best described by a third degree polynomial regression equation. These results demonstrate that in well-controlled diabetics, although the increment in BPD was less than controls, the growth pattern of the fetal BPD was similar among the White classes B to FR, and the velocity of growth of the BPD was similar among diabetics and nondiabetics.

Blood Glucose↗
PubMed1990

A longitudinal study comparing growth in diabetic pregnancies with growth in normal gestations: I. The fetal weight.

Diabetes mellitus is associated with fetal growth acceleration and retardation. These aberrations in fetal growth seem to be influenced by a variety of factors including vascular disease, glycemic control, hypertension and smoking. In order to characterize fetal growth under the above conditions, longitudinal sonographic evaluations were performed in 52 pregnant, insulin-dependent diabetic women with the usual monitoring of the patients' metabolic control. Regression analyses revealed that vascular disease and glycemic conditions were the most important influences for growth, with manifestation beyond the second trimester. With stringent glucose control (mean whole blood less than or equal to 100 mg/dl) in the absence of vasculopathy (white classes A, B, C), fetal growth was similar to that in normal pregnancies. In the presence of vasculopathy (white classes D and FR), growth was reduced, especially when near-normal glycemic levels were achieved. Conversely, in poorly controlled diabetic women, enhanced fetal growth were observed in patients with and without vasculopathy. No aberrations in fetal growth were observed, however, before the third trimester. The findings of our study demonstrate that vasculopathy and glycemia are dominant and independent regulators of fetal growth. However, their influences are not manifested in growth changes before the third trimester.

Embryonic and Fetal Development↗
PubMed1988

Diabetic nephropathy: pregnancy performance and fetomaternal outcome.

A study of 31 continuing pregnancies complicated by diabetic nephropathy was conducted to determine the effects of diabetes-associated renal disease on maternal health and fetal outcome. Throughout pregnancy there was a significant increase in maternal blood pressure (p less than 0.001) and proteinuria (p less than 0.0001), with nephrotic syndrome (greater than 3.0 gm protein/day) developing in 71% of pregnancies. After birth, however, proteinuria reverted to levels not significantly different from values in early pregnancy. There was no apparent adverse effect of pregnancy on the natural course of the underlying renal disease. Stillbirths occurred in two patients (6%), and the remaining 29 pregnancies resulted in live-births at a mean gestational age of 36 weeks. Seventy percent of these infants were appropriate for gestational age, whereas 16% were small and 13% were large for gestational age. Birth weight was best correlated with gestational age and creatinine clearance (p less than 0.0001). Neonatal complications included respiratory distress syndrome (19%), hyperbilirubinemia (26), and congenital malformations (10%). The uncorrected perinatal survival rate was 94%. These data suggest that with contemporary methods of maternal evaluation and treatment, fetal surveillance, and neonatal care, the risks to patients with diabetic nephropathy during pregnancy are not excessive. The likelihood of a successful fetal and neonatal outcome is comparable to that in other patients with insulin-dependent diabetes.

Adolescent↗
PubMed1987

Screening for gestational diabetes: one-hour carbohydrate tolerance test performed by a virtually tasteless polymer of glucose.

Although the 1-hour 50 gm blood glucose screening test is an effective way of detecting diabetes in pregnancy, the taste of available glucose drinks often creates gastrointestinal symptoms and leads to refusal of the patient to be tested. The efficacy of a virtually tasteless glucose polymer in testing carbohydrate tolerance in pregnancy was determined. Sixty-one pregnant patients undergoing screening for gestational diabetes underwent a 1-hour carbohydrate tolerance test of both glucose and a glucose polymer within 3 days of each other. Analysis of the data revealed a high degree of agreement between the results of the 1-hour carbohydrate tolerance test (kappa = 0.62, p less than 0.0001). These data suggest that glucose polymer can be used effectively in screening for gestational diabetes.

Blood Glucose↗
PubMed1987

Comparison of the logistic and Cox regression models when outcome is determined in all patients after a fixed period of time.

This paper presents an evaluation of the logistic and Cox regression models for a prospective study when the outcome is binary and is determined in all patients after a fixed period of time. The similarities and differences between the regression coefficients and test statistics are given for the two-sample case. Extension of results to the multivariate case and under product binomial sampling are discussed. The results are illustrated using data from a clinical trial designed to evaluate the effect of a lipid lowering drug on progression of coronary artery disease.

Coronary Disease↗
PubMed1987

Estimated fetal weight in the evaluation of growth in twin gestations: a prospective longitudinal study.

A prospective longitudinal study was conducted in order to determine by sonographically estimated fetal weight the patterns of fetal growth in twins. Thirty-five healthy women with normal twin pregnancies were examined every three weeks from the 15th week of gestation to delivery. Among the measurements obtained were the biparietal diameter (BPD), the abdominal circumference, and the calculated fetal weight. From 15-28 weeks, the growth velocity of the BPD and abdominal circumference remained fairly constant, with a steady increase in incremental growth. Beyond this age, we observed a slowing in growth of the BPD, while the abdominal circumference continued at a constant rate. The growth velocity of the weight steadily increased throughout pregnancy. Although greater biologic variability in weight between twin A and B was observed as gestational age progressed, the overall mean weights of twin A and B were not statistically different. We have generated a nomogram of fetal weight gain throughout pregnancy.

Body Height↗
PubMed1987

Persistent pulmonary hypertension: assessment of perinatal risk factors.

Persistent pulmonary hypertension of the neonate, a disease of unknown etiology, is associated with a mortality rate as high as 50%. We conducted a retrospective study (1979-1983) to identify antenatal events that may be associated with this disease. Thirty-seven mothers of neonates with this disorder were compared with 150 randomly selected control patients. Logistic regression analysis revealed that meconium, maternal fever (caused by urinary tract infection, upper respiratory infection, chorioamnionitis, and unknown causes), maternal anemia (caused by abruptio placentae, placenta previa, and unknown causes), and maternal pulmonary disease (caused by asthma and/or pneumonia) were associated with persistent pulmonary hypertension of the neonate. The computed odds ratios for the variables associated with this disease were increased significantly above the controls. In addition, cesarean section, nonvertex presentation, fetal distress, and vaginal bleeding were found to be significant on univariate analyses. The cesarean section rate in the study group was 59%, with the primary indications of fetal distress and/or third-trimester bleeding in 77.3% of instances. These data suggest that antenatal risk factors and the computed odds ratio may help identify pregnancies likely to produce infants at high risk for persistent pulmonary hypertension of the neonate.

Asphyxia Neonatorum↗
PubMed1986

A randomized clinical trial of the insulin pump vs intensive conventional therapy in diabetic pregnancies.

Improved perinatal outcome is associated with the prevention of hyperglycemia during pregnancy in diabetic women. To determine whether the method of insulin administration influences the degree of diabetic control obtained, we randomized 22 pregnant diabetic women to intensive conventional insulin therapy (N = 11) and insulin pump therapy (N = 11). Frequent outpatient visits; home glucose monitoring, at least six times daily; and frequent telephone contact were offered to all subjects. Patients were hospitalized in the inpatient clinical research center each trimester for a 24-hour metabolic profile. There were no differences between the two treatment groups with respect to outpatient mean glucose levels, symptomatic hypoglycemia, or glycosylated hemoglobin levels, or with respect to inpatient mean glucose level, glycemic excursions, chemical hypoglycemia, or hyperglycemia. Excellent metabolic control was achieved with both treatment methods.

Adult↗
PubMed1986

Intrapartum fetal weight estimation: a comparison of three formulae.

To determine the relative accuracy of fetal weight estimation using the biparietal diameter (BPD), the abdominal circumference (AC), and the femur length (FL) in three formulae (BPD/AC, FL/AC, and BPD/AC/FL), 63 patients in labor were examined. All patients delivered within 24 hours of ultrasound examination. A good correlation was found between the estimated fetal weight and the actual birth weight, using the three formulae: BPD/AC (r = 0.96); FL/AC (r = 0.95); and BPD/AC/FL (r = 0.96). The FL/AC formula overestimated fetal weight (P less than 0.01), however, particularly in fetuses weighing more than 2000 g. The mean percentage error with the BPD/AC formula was 0.99 per cent, 3.82 per cent with the FL/AC, and 2.43 per cent with the BPD/AC/FL formula. This study showed that although all three formulae were comparable, the best estimation of the birth weight was obtained when either the BPD/AC or the BPD/AC/FL formulae were used. Additionally, the results demonstrate that reliable estimates of fetal weight can be made even at term or in laboring patients.

Birth Weight↗