Prediction of cumulative incidence function under the proportional hazards model

In the presence of dependent competing risks in survival analysis, the Cox model can be utilized to examine the covariate effects on the cause-specific hazard function for the failure type of interest. For this situation, the cumulative incidence function provides an intuitively appealing summary curve for marginal probabilities of this particular event. In this paper, we show how to construct confidence intervals and bands for such a function under the Cox model for future patients with certain covariates. Our proposals are illustrated with data from a prostate cancer trial.     

Applying k-sample tests to conditional probabilities for competing risks in a clinical trial

In the presence of competing risks, a full picture of the data can be developed considering the cumulative incidence function for each risk. If one risk type is of particular interest, the conditional probability of failure due to that risk, conditional on no failure due to the remaining competing risks, can be used to compare any number of samples. Kappa-sample tests of significance are derived and extended to stratified test statistics, allowing adjustment to be made for important prognostic factors. The methods are applied to a clinical trial involving patients with advanced breast cancer, where interest focused on progression of disease at old and new sites. They show that estrogen receptor status positive is an important prognostic factor in terms of time to progressive disease at a current tumour site, even when stratified for a potentially confounding measure of spread of disease, whereas progesterone receptor status positive is important with regard to disease progression at new sites only.     

Statistical modeling of carcinogenic risks in dogs that inhaled 238PuO2

Combined analyses of data on 260 life-span beagle dogs that inhaled 238PuO2 at the Inhalation Toxicology Research Institute (ITRI) and at Pacific Northwest National Laboratory (PNNL) were conducted. The hazard functions (age-specific risks) for incidence of lung, bone and liver tumors were modeled as a function of cumulative radiation dose, and estimates of lifetime risks based on the combined data were developed. For lung tumors, linear-quadratic functions provided an adequate fit to the data from both laboratories, and linear functions provided an adequate fit when analyses were restricted to doses less than 20 Gy. The estimated risk coefficients for these functions were significantly larger when based on ITRI data compared to PNNL data, and dosimetry biases are a possible explanation for this difference. There was also evidence that the bone tumor response functions differed for the two laboratories, although these differences occurred primarily at high doses. These functions were clearly nonlinear (even when restricted to average skeletal doses less than 1 Gy), and evidence of radiation-induced bone tumors was found for doses less than 0.5 Gy in both laboratories. Liver tumor risks were similar for the two laboratories, and linear functions provided an adequate fit to these data. Lifetime risk estimates for lung and bone tumors derived from these data had wide confidence intervals, but were consistent with estimates currently used in radiation protection. The dog-based lifetime liver tumor risk estimate was an order of magnitude larger than that used in radiation protection, but the latter also carries large uncertainties. The application of common statistical methodology to data from two studies has allowed the identification of differences in these studies and has provided a basis for common risk estimates based on both data sets.     

Modelling age-specific risk: application to dementia

We give up-to-date methods for estimating the age-specific incidence of a disease and for estimating the effect of risk factors. We recommend taking age as the basic time scale of the analysis; then, the hazard function can be interpreted as the age-specific incidence of the disease. This choice raises a delayed entry problem. We present three methods: the person-years method; the smoothed Nelson-Aalen estimator, and the penalized likelihood approach. When explanatory variables are available, the Poisson model and the Cox model with delayed entry may be used for estimating relative risks; the penalized likelihood approach can also be used. We apply these methods to estimate the age-specific incidence of dementia using data from a large cohort study, Paquid. This 5-year study followed a random initial sample of 3675 subjects with 190 incident cases of dementia. We compare the estimates based on the three possible methods. The estimated incidences computed separately for men and women cross and it is verified that a non-proportional hazards model for gender holds; women below 75 have a lower risk than men while women above 75 have a higher risk.     

Design and Application of Risk Adjusted Cumulative Sum for Strength Monitoring of Ready Mixed Concrete

The cumulative sum procedure is an effective statistical process control tool that can be used to monitor the quality of ready mixed concrete during its production process. In this paper, an attempt has been made to design and apply a new cumulative sum procedure for the ready mixed concrete industry, which takes care of the risks involved in and associated with the production of concrete. This new procedure can be termed as risk adjusted cumulative sum. The 28-day characteristic cube compressive strengths of the various grades of concrete and detailed information regarding the production process and the risks associated with production were collected from the operational ready mixed concrete plants in and around Ahmedabad and Delhi, two important cities in India. The risks were quantified using a likelihood-based scoring method. Finally, a risk adjusted cumulative sum model was developed by imposing the weighted score of the estimated risks on the conventional cumulative sum plot. This model is a more effective and realistic tool for monitoring the strength of ready mixed concrete.     

Lifetime risk of dementia and Alzheimer's disease. The impact of mortality on risk estimates in the Framingham Study

We estimated the remaining lifetime risks of developing Alzheimer's disease (AD) and dementia from all causes, based on data from longitudinal population studies. The risk of developing AD during one's lifetime depends on both disease incidence and life expectancy. Conventional estimates of cumulative incidence overestimate the risk when there is a substantial probability of mortality due to competing causes. A total of 2,611 cognitively intact subjects (1,061 men, 1,550 women; mean age, 66 +/- 7 years) were prospectively evaluated for the development of AD or other dementia. A modified survival analysis was used to estimate both cumulative incidence and the sex-specific remaining lifetime risk estimates for quinquennial age groups above age 65 years. Over a 20-year follow-up period, 198 subjects developed dementia (120 with AD). The remaining lifetime risk of AD or other dementia depended on sex, being higher in women, but varied little with age between 65 and 80 years. In a 65-year-old man, the remaining lifetime risk of AD was 6.3% (95% CI, 3.9 to 8.7) and the remaining lifetime risk of developing any dementing illness was 10.9% (95% CI, 8.0 to 13.8); corresponding risks for a 65-year-old woman were 12% (95% CI, 9.2 to 14.8) and 19% (95% CI, 17.2 to 22.5). The cumulative incidence between age 65 and 100 years was much higher: for AD, 25.5% in men and 28.1% in women; for dementia, 32.8% in men and 45% in women. The actual remaining lifetime risk of AD or dementia varies with age, sex, and life expectancy and is lower than the hypothetical risk estimated by a cumulative incidence in the same population.     

Utilization of manufacturers' implant card data to estimate heart valve failure

Heart valve manufacturers possess the most complete inventory of world-wide mechanical valve failures, but to convert failures to time-related risks requires estimates of patient follow up. Since manufacturers did not actively track patients, they needed a model that incorporates an assumed death rate to decrease the numbers of patients at risk. We present a method for using a manufacturer's implant card database to estimate time-related complication rates for patient subsets, and illustrate its use by examining the risk of outlet strut fracture (OSF) with the Bj?rk-Shiley 60 degrees Convexo-Concave valve (CC60). We developed a parametric model for valve patient survival based on actively followed valve patients from three centers using only variables typically available from implant cards. Using this survival model, a simulated lifetime was produced for each valve in the CC60 implant database for which the required covariates were known. These lifetimes were then used to analyze OSF as if they were true follow up times. This allowed the use of conventional methods of univariate and multivariate analysis for OSF, including parametric statistical models. According to the approximate linearity of the cumulative hazard functions, OSF risk over time appeared to be fairly constant. Several risk factors were identified, including valve size, patient age at implant and valve position. Using parametric models for both patient survival and OSF permits the estimation of the probability of OSF before death for an individual patient (as opposed to the usual actuarial probability of OSF given that the patient does not die). Because the patient may die before his valve would have failed, this cumulative incidence of OSF is always less than the actuarial risk. For all but the very highest risk patients, the cumulative incidence over their relatively short remaining lifetimes is very small.     

Risk Allocation by Law—Cumulative Impact of Change Orders

Change, defined as any event that results in a modification of the original scope, execution time, or cost of work, is inevitable on most construction projects due to the uniqueness of each project and the limited resources of time and money available for planning. There are many factors that may cause a change such as design errors, design changes, additions to the scope, or unknown conditions in the field. For each change, contractors are entitled to an equitable adjustment to the base contract price and schedule for all productivity impacts associated with the change. The focus of this paper is to outline the types of changes that can occur on a construction project and also to spell out the financial recovery possibilities that exist for the contractor for each type of change. There are many historical and current court decisions that shape the outcomes of such claims and determine who holds the risks associated with various project changes. Also, an effective cumulative impact claim contains certain vital elements upon which the final outcome will be determined by the legal system. Last, there are certain actions that a contractor and owner can do to either enhance or mitigate the effectiveness of a potential cumulative impact claim.     

Pre-AIDS mortality in the Edinburgh City Hospital HIV cohort

In this paper, we look at the incidence and predictive factors of pre-AIDS mortality among HIV-infected individuals, and injecting drug users (IDUs) in particular, and compare IDUs with non-IDUs. 627 patients (73 per cent IDUs) of the Edinburgh City Hospital HIV Cohort were enrolled pre-AIDS and followed up until September 1994. Analyses were performed using cumulative hazard and cumulative incidence estimators for a competing risks model, the Cox proportional hazards model and the non-parametric hazard estimator of Fusaro et al. (1993). The effects of age and CD4 T-lymphocyte cell count, progressively depleted during HIV progression, were investigated. 60 deaths occurred in AIDS-free patients during follow-up; 25 were drug-related deaths in IDUs. Pre-AIDS mortality was higher among IDUs than non-IDUs (p = 0.07). The cumulative incidences of pre-AIDS death after five years from enrollment were 11 per cent in IDUs and 6 per cent in non-IDUs; the cumulative AIDS incidences were, respectively, 19 per cent and 32 per cent. After eight years, cumulative pre-AIDS death incidence was 15 per cent among IDUs; cumulative AIDS incidence among IDUs was 35 per cent. Both groups had similar risks of medically-related (non-AIDS)-MRNA-death. Age and CD4 count were both individually predictive of MRNA death (relative risks (RRs); 2.1 per decade of life, p < 0.01; and 1.9 for each 100 cells per 100 microliters lost, p < 0.0001), although when used together age was less significant (RR 1.6, p = 0.07). Neither was statistically significant for drug-related mortality, although hazard may be lower in older individuals and may increase with falling CD4 count. The drug-related mortality was 1.1 per cent: 2.3 per cent in the first two years after enrollment, and 0.4 per cent thereafter. We conclude that older HIV-infected individuals are at greater risk of medically-related death before AIDS. This risk increases as CD4 count declines. Drug-related hazard may be greater in younger individuals and may increase as CD4 counts fall, but neither effect was formally significant.     

Predictors of depressive symptom trajectories in mothers of preterm or low birth weight infants.

Predictors of maternal depression trajectories were examined longitudinally in families with an infant born preterm or at a low birth weight. A total of 181 mother–infant dyads enrolled in the study before the infant’s neonatal intensive care unit (NICU) discharge. Maternal depressive symptoms were assessed at 5 timepoints, and contextual variables and infant risks were assessed at NICU discharge. Hierarchical linear models revealed that mothers who experienced more risk factors reported more depressive symptoms just before their infant’s NICU discharge and showed less decline in depressive symptoms in the months immediately following the child’s birth. Although cumulative risks predicted depression trajectories, this effect appeared driven by maternal and family sociodemographic risks rather than infant risks. Addition of family support as a covariate in the multilevel models with a subsample of families revealed that social support and depression covaried across time. However, most of the findings regarding the association between risk and depression remained consistent, whereas the effects of maternal race and multiple birth were slightly attenuated. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Internet paradox. A social technology that reduces social involvement and psychological well-being?

We present a method for estimating age- and time-specific HIV incidence using back-calculations of AIDS incidence data. Two-dimensional penalized likelihood is employed, using a flexible bivariate step function model of HIV incidence, together with a quadratic roughness penalty which leads to thin-plate spline smoothing. This allows incidence estimates to vary flexibly and smoothly in both age and time. We propose generalized cross-validation as a guide for choice of an appropriate level of smoothing and describe an EM algorithm for computing the estimates. We propose the method primarily for qualitative assessment of trends in age-specific incidence over time and apply it to a small Italian data set on men who have sex with men. The analysis suggests a trend over time of increasing relative incidence among younger individuals, consistent with incidence patterns observed in other countries.     

A nonparametric approach to a survival study with surrogate endpoints

A nonparametric estimator for the joint distribution of a survival time and surrogate response time, which may occur earlier during follow-up, is presented. In the absence of the surrogate response variable, the estimator reduces to the Kaplan Meier nonparametric estimator for the survival time alone. The estimator derived in this paper is done so in a particular novel way using an exchangeable process (reinforced random walks) to model individual observations. The methodology introduced in the paper is readily extended to modelling multiple state processes.     

Thyroid cancer rates and 131I doses from Nevada atmospheric nuclear bomb tests

BACKGROUND: We examined data on death from thyroid cancer across the continental United States and data on incidence from selected areas of the country for evidence of an association between this disease and exposure to radioactive iodine (131I) from nuclear tests in Nevada in the 1950s. METHODS: Analyses involving 4602 thyroid cancer deaths (1957-1994) and 12 657 incident cases of thyroid cancer (1973-1994) were performed. Excess relative risks (ERRs) per Gray (Gy) of radiation were estimated by relating age-, calendar year-, sex-, and county-specific rates to estimates of dose to the thyroid that take age at exposure into account. RESULTS: Analyses of cumulative dose yielded negative ERRs that were not statistically significant. An association was suggested for dose received by children under 1 year of age for both mortality data (ERR per Gy = 10.6; 95% confidence interval [CI] = -1.1 to 29) and incidence data (ERR per Gy = 2.4; 95% CI = -0.5 to 5.6); no association was found for dose received at older ages. For mortality data, but not incidence data, there was an elevated ERR in the 1950-1959 birth cohort of 12.0 (95% CI = 2.8 to 31) per Gy. CONCLUSIONS: Risk of thyroid cancer from exposure to 131I from atmospheric nuclear tests did not increase with cumulative dose or dose received at ages 1-15 years, but associations were suggested for individuals exposed under 1 year of age and for those in the 1950-1959 birth cohort. The absence of increased risk from dose received at ages 1-15 years is not consistent with studies of children exposed to external radiation sources. This inconsistency may result from the limitations and biases inherent in ecologic studies, including the error introduced when studying a mobile population. These problems preclude making a quantitative estimate of risk due to exposure; however, given such limitations, it is perhaps remarkable that any evidence of the effects of 131I emerges from this study.     

A method of evaluating drug efficacy by statistical analysis of healing speed of peptic ulcer

At present, the evaluation of anti-ulcer drugs is generally accomplished simply by calculating the cumulative healing rate at a certain point of time during treatment, which does not implicate any analysis of the healing speed of the ulcer. If the cumulative healing rate of an ulcer is expressed as a function of drug administration time, t, then it will be possible to calculate parameters concerning the healing speed of ulcers and thus evaluate drug efficacy as the time series analysis of the cumulative healing rate. A new method of evaluating anti-ulcer drugs by a statistical analysis of healing speed is proposed. A non-linear regression analysis was performed between two variables, t (time of drug administration: week) and y (non-healing rate: %), to obtain the exponential function y = Ae-kt. The theoretical values calculated from the exponential equation were in close proximity to the observed values. With this analysis, four parameters concerning the healing speed were defined, namely the healing rate constant, the initiation time of healing, the half-life of non-healing rate and the time necessary for 50% healing. With this method, the efficacy of drugs on peptic ulcer healing was dynamically analysed, the non-healing rate (y) being expressed as an exponential function of length of time (t) of treatment, thus obtaining digital parameters for healing speed.     

The impact of postoperative pain on the development of postoperative delirium

We performed a prospective observational study to examine the role of postoperative pain and its treatment on the development of postoperative delirium. Pain was measured in direct patient interviews using a visual analog scale (VAS) and was assessed for pain at rest, pain with movement, and maximal pain over the previous 24 h. Postoperative delirium was diagnosed during these interviews by using the confusion assessment method (CAM) and/or by using data from the medical record and the hospital's nursing intensity index. The method of postoperative analgesia, type of opioid, and cumulative opioid dose were also recorded. After controlling for known preoperative risk factors for delirium (age, alcohol abuse, cognitive function, physical function, serum chemistries, and type of surgery), higher pain scores at rest was associated with an increased risk of delirium over the first 3 postoperative days (adjusted risk ratio 1.20, P = 0.04). Pain with movement and maximal pain were not associated with delirium. Method of postoperative analgesia, type of opioid, and cumulative opioid dose were not associated with an increased risk of delirium. We conclude that more effective control of postoperative pain reduces the incidence of postoperative delirium. Implications: We performed daily interviews in a large population of patients undergoing noncardiac surgery to measure their level of pain and development of delirium. We found an association between higher pain levels at rest and the development of delirium. Our results suggest that better control of postoperative pain may reduce this serious complication.     

Fertility after hysteroscopic myomectomy

We have conducted a cohort study of cancer risks among 140,208 Swedish farmers in order to compare their cancer risks with those of the general male population. Since there were no individual data regarding exposure to agricultural chemicals and acquiring such data was not realistic, we obtained crude and hypothetical estimates for exposure by dividing the data into time periods, year-of-birth cohorts and geographical areas. The cohort was followed-up in the Cancer Environment Register from 1 January 1971 either until death or until 31 December 1987. The relative risk was computed as the ratio of the observed and expected number of cases (SIR = standardized incidence ratio). A total of 15,040 cases were observed vs 18,918 expected, resulting in a statistically significant decreased SIR of 0.80 (95% confidence interval: 0.78-0.81). The SIR was significantly decreased for several cancer sites, and the lowest value was found for tongue, lung, oesophagus, liver and urinary organs, which is in agreement with other studies on cancer risks among farmers. Other major cancer sites with decreased SIRs were the colon, rectum, pancreas and kidney. Lip cancer and multiple myeloma showed statistically significant increased risks. SIRs for stomach cancer, prostate cancer, skin carcinoma, malignant melanoma, tumours in connective tissue or muscle, malignant lymphomas and leukaemia were all close to unity, which is not consistent with several other studies that have shown increased risks for these sites. For malignant lymphomas the SIR increased over time, though not significantly, and was highest among younger farmers. The SIR for non-Hodgkin lymphoma was lowest in the northernmost region. This gives some support to the hypothesis that there is an association between non-Hodgkin lymphoma and exposure to pesticides and other agricultural chemicals. It is of note that the SIR for multiple myeloma was significantly increased in those parts of Sweden where the use of pesticides has been less frequent and in lower amounts.     

Technetium-99m-tetrofosmin uptake in brain tumors by SPECT: comparison with thallium-201 imaging

BACKGROUND/AIMS: There is growing interest worldwide in primary liver cancer. The aim of this study was to describe the incidence of this cancer over a 20-year period in a well-defined French population. METHODS: Time trends by 4-year period were studied by sex, age group, place of residence, histological type and associated cirrhosis. Trends were also analysed using the age-period-cohort model. RESULTS: Primary liver cancer incidence in men increased from 7.5/100000 for the period 1976-79 to 10.2/100000 for the period 1992-95. The mean annual variation was +2.2%, (p<0.05). The increase in incidence was seen mainly in the 55-64 and 65-74 age groups and concerned hepatocellular carcinomas. In men, the increase in incidence rates with time was observed mainly in rural areas, whereas incidence rates in urban areas remained stable. The rise in incidence was due mostly to an increase in primary liver cancer with cirrhosis, in relation to a progressive increase in post-hepatitic cirrhosis and a recent increase in alcoholic cirrhosis. The estimated cumulative risk for the life span 30-74 years increased from 0.8% for the 1904-1908 cohort to 2.1% for the 1934-1938 cohort. There was no significant trend in female rates. CONCLUSIONS: In France, incidence rates for primary liver cancer are increasing in men, whilst they are remaining stable in women. Our data confirm the primary importance of alcohol in the aetiology of this cancer. Further studies are necessary to unravel the respective roles of alcohol and hepatitis C virus in the increasing incidence of primary liver cancer.     

P53 protein immunohistochemical expression in colonic adenomas with and without associated carcinoma

On the basis of the positive outcome of animal experiments, several large placebo-controlled trials are underway and aiming for the first time at the prevention of an immune-mediated disease, type 1 diabetes. The first of these trials, The Deutsche Nicotinamide Intervention Study (DENIS), evaluated the clinical efficacy of high doses of nicotinamide in children at high risk for IDDM. Nicotinamide has been shown to protect beta-cells from inflammatory insults and to improve residual beta-cell function in patients after onset of IDDM. Individuals at high risk for developing IDDM within 3 years were identified by screening the siblings (age 3-12 years) of patients with IDDM for the presence of high titer (> or =20 Juvenile Diabetes Foundation [JDF] U) islet cell antibodies. Probands (n = 55) were randomized into placebo and nicotinamide (slow release, 1.2 g x m(-2) x day(-1)) receiving groups and followed prospectively in a controlled clinical trial using a sequential design. Rates of diabetes onset were similar in both groups throughout the observation period (maximum 3.8 years, median 2.1 years). This sequential design provides a 10% probability of a type II error against a reduction of the cumulative diabetes incidence at 3 years from 30 to 6% by nicotinamide. The trial was terminated when the second sequential interim analysis after the eleventh case of diabetes showed that the trial had failed to detect a reduction of the cumulative diabetes incidence at 3 years from 30 to 6% (P = 0.97). The group receiving nicotinamide exhibited decreased first-phase insulin secretion in response to intravenous glucose (P = 0.03). No other side effects were observed. We conclude that in this subgroup of diabetes-prone individuals at very high risk and with an assumed rapid disease progression, nicotinamide treatment did not cause a major decrease or delay of diabetes development. However, the data do not exclude the possibility of a less strong, but potentially meaningful, risk reduction in this cohort, or a major clinical effect of nicotinamide in individuals with less risk of progression to IDDM than studied here.     

Significant reduction in circuit pressure with modified plasma separation chamber for a heparin removal device

OBJECTIVE: This study aimed to determine the maximum dose of radiation the CLARION 1.2 cochlear implant can withstand safely. INTRODUCTION: Cochlear implants restore functional hearing to patients with sensorineural deafness. Because some patients may need radiation therapy, it is important to investigate the influence of ionizing radiation on cochlear implant function. METHODS: This study tested the function of four CLARION 1.2 implants (Advanced Bionics, Sylmar, CA, U.S.A.) after varying radiation treatments with gamma rays. The first implant received a cumulative dosage of 69 Gy over nine treatments (single doses between 0.1-30 Gy). The second was irradiated with a total of 90 Gy, receiving three treatments of 30 Gy each. The third and fourth received doses more typical of patient therapy (i.e., 2 Gy) approximately 30 times, for a cumulative dosage of approximately 60 Gy. Implant function was tested after every treatment; the CLARION implant incorporates a back-telemetry system, allowing impedance and current output testing. RESULTS: Despite the type of treatment, the results were quite consistent: difficulties in function occurred when the cumulative dosage inside the implant was approximately 60 Gy. The first implant recovered completely and the second recovered partially. DISCUSSION: The CLARION 1.2 cochlear implant seems to safely withstand approximately 60 Gy of radiation before experiencing functional difficulties. In a clinical situation, the implant would not likely be in the target volume irradiated, and thus the patient's therapeutic cumulative dosage might be higher.     

Multiple risk factors in the development of externalizing behavior problems: group and individual differences

The aim of this study was to test whether individual risk factors as well as the number of risk factors (cumulative risk) predicted children's externalizing behaviors over middle childhood. A sample of 466 European American and 100 African American boys and girls from a broad range of socioeconomic levels was followed from age 5 to 10 years. Twenty risk variables from four domains (child, sociocultural, parenting, and peer-related) were measured using in-home interviews at the beginning of the study, and annual assessments of externalizing behaviors were conducted. Consistent with past research, individual differences in externalizing behavior problems were stable over time and were related to individual risk factors as well as the number of risk factors present. Particular risks accounted for 36% to 45% of the variance, and the number of risks present (cumulative risk status) accounted for 19% to 32% of the variance, in externalizing outcomes. Cumulative risk was related to subsequent externalizing even after initial levels of externalizing had been statistically controlled. All four domains of risk variables made significant unique contributions to this statistical prediction, and there were multiple clusters of risks that led to similar outcomes. There was also evidence that this prediction was moderated by ethnic group status, most of the prediction of externalizing being found for European American children. However, this moderation effect varied depending on the predictor and outcome variables included in the model.