Kaplan-Meier Survival Function Research Articles

Understanding the gendered patterns of substance use initiation among adolescents living in rural, central Mexico

BackgroundLittle is known about the age of initiation and gender differences in substance use among adolescents in rural, central Mexico. MethodsThe cross-sectional data were collected from students enrolled in the Videobachillerato (VIBA) (video high school) programme in Guanajuato, Mexico. Questionnaires asked students about the age at which they had used alcohol, cigarettes, or cannabis for the first time. Kaplan–Meier Survival Functions were used to estimate if males and females were significantly different in their cumulative probabilities of initiating substances over time. ResultsOn average, alcohol is initiated at 14.7years of age, cigarettes at 15.1years of age, and cannabis at 16.5years of age. Over time, males had a significantly higher probability of initiating alcohol (Kaplan–Meier Failure Curve: X2=26.35, p<0.001), cigarettes (Kaplan–Meier Failure Curve: X2=41.90, p<0.001), and cannabis (Kaplan-Meier Failure Curve: X2=38.01, p<0.001) compared to females. ConclusionsThese results highlight the gendered patterns of substance use initiation among adolescents in rural, central Mexico and underscore the need for gendered substance use prevention interventions with these adolescents. By putting forth efforts to understand substance use initiation patterns of adolescents living in rural, central Mexico, culturally specific and efficacious prevention efforts can be tailor-made to create lasting differences.

Survival estimates for elite male and female Olympic athletes and tennis championship competitors

In this paper, we report survival estimates for male and female Olympic medal winners and for male and female finalists at the British and U S national tennis championships. We find a consistent longevity advantage of Olympic medal-winning female athletes over Olympic medal-winning male athletes competing separately in the same events since 1900 and for female finalists over male finalists competing separately in the finals of the national tennis championships of Britain and of the United States since the 1880s. This is the case for sample mean comparisons, for Kaplan-Meier survival function estimates, including life expectancy, and for Cox proportional hazard estimates, which show statistically significant lower hazard rates for women with birth year and other variables constant. The female longevity advantage over males is similar in the early period samples (birth years before 1920) and in the full period samples, and is 5-7 years.

SurvivalBIV: Estimation of the Bivariate Distribution Function for Sequentially Ordered Events Under Univariate Censoring

In many medical studies, patients can experience several events. The times between consecutive events (gap times) are often of interest and lead to problems that have received much attention recently. In this work we consider the estimation of the bivariate distribution function for censored gap times, using survivalBIV a software application for R. Some related problems such as the estimation of the marginal distribution of the second gap time is also discussed. It describes the capabilities of the program for estimating these quantities using four different approaches, all using the Kaplan-Meier estimator of survival. One of these estimators is based on Bayes’ theorem and Kaplan-Meier survival function. Two estimators were recently proposed using the Kaplan-Meier estimator pertaining to the distribution of the total time to weight the bivariate data (de Uña-Álvarez and Meira-Machado 2008 and de Uña-Álvarez and Amorim 2011). The software can also be used to implement the estimator proposed in Lin, Sun, and Ying (1999), which is based on inverse probability of censoring weighted. The software is illustrated using data from a bladder cancer study.

826: To push or not to push: managing the second stage in morbidly obese parturients

826 To push or not to push: managing the second stage in morbidly obese parturients Jennifer McNamara, Anthony Odibo, Anthony Shanks, Methodius Tuuli, Kimberly Roehl, Alison Cahill Washington University in St. Louis, Department of Obstetrics and Gynecology, St. Louis, MO OBJECTIVE: Labor management of morbidly obese women presents a unique challenge, yet almost no published data exists regarding the potential advantage of passive descent in the 2nd stage for women with a body mass index(BMI) 40. We aimed to estimate the effect of passive descent on time spent pushing and birth outcomes in a morbidly obese cohort. STUDY DESIGN: A retrospective cohort study of all consecutive women with BMI 40 with term labor that reached the 2nd stage of labor from 2004-2008 at our institution. Detailed information was collected on maternal demographics, labor progress, and maternal and neonatal outcomes. Women allowed to labor down for at least 30min (passive descent) were compared to those who pushed immediately for length of pushing, length of 2nd stage, and maternal and neonatal outcomes using univariable and multivariable analysis. Active pushing time was also analyzed with Kaplan-Meier survival function. Subgroup analysis was performed by parity. RESULTS: Of 558 women with BMI 40, 97(17.4%) labored down and 461(82.6%) pushed immediately. The overall time (min) spent pushing was significantly longer,(19[7-39] vs 7[3-17], p 0.001), for morbidly obese women as was their length of 2nd stage (81[57-127] vs 15[8-27], p 0.001). Morbidly obese women in the passive descent group were significantly more likely to push for at least 60 min compared to those who pushed immediately,(16.5% vs. 7.2%, aOR 2.56[1.35-4.87]). They were also significantly more likely to push for 90min. The groups had similar rates of operative delivery, but significantly increased maternal febrile morbidity(14.4% vs 7.6%, aOR 2.14[1.09-4.18]) was seen in the labored down cohort. Subgroup analysis of nulliparous patients yielded similar results with no evidence for benefit for those obese women who labored down. CONCLUSION: Morbidly obese parturients who labored down spent greater time pushing, and had higher rates of febrile morbidity compared with morbidly obese women who pushed immediately. Passive descent is not an effective strategy for the management of the 2nd stage at term in morbidly obese patients. 827 Does advancing gestation affect patientreported sleep quality and daytime sleepiness? John Owen, Tamula Patterson, Susan Harding, Suzanne Cliver, John Hauth Univ of Alabama at Birmingham, Obstetrics and Gynecology, Birmingham, AL, Univ of Alabama at Birmingham, Pulmonary and Critical Care Medicine, Birmingham, AL OBJECTIVE: To evaluate sleep quality and daytime sleepiness in pregnancy using the validated Pittsburg Sleep Quality Index (PSQI) and Epworth Sleepiness Scale (ESS). STUDY DESIGN: Prospective observational study of healthy women with normal gestations initiating care Jan, 2010 Jan, 2011. Sleep quality and daytime sleepiness were assessed in the early 2nd and 3rd trimesters (TM). The PSQI is a 10-question survey that assesses sleep hygiene, impediments to sleep, hypnotic use, daytime dysfunction, sleep partner and subjective sleep quality. Scores range from 0 to a maximum of 21, and scores 5 indicate poor sleep quality. The ESS grades the likelihood of dozing in 8 daily scenarios; scores range from 0 to 24, and 10 indicates significant daytime sleepiness. RESULTS: Of 495 consenting women, the mean age was 23 y, 53% were multiparous, and 81% were African American. They underwent initial evaluation at a mean (SD) gestational age (GA) of 15.7 (4.3) wks, and 458 (93%) had follow-up at a mean GA of 30.0 (1.0) wks. At the 2nd TM PSQI assessment, 201/495 (41%) of women had poor sleep quality, increasing to 216/458 (47%) in the 3rd TM. The mean PSQI also showed a small, but statistically significant, 0.5 unit increase (5.5 to 6.0) from the 2nd to the 3rd TM (p 0.01). Although the ESS score was not significantly different between the 2nd and 3rd TM (p 0.1), significant daytime sleepiness was observed in 27% at both assessments. Overall, 14% of women reported hypnotic use during either evaluation. Use of hypnotics was associated with a significantly worse sleep quality score ( 2 points on PSQI; p 0.0001) but had no effect on daytime sleepiness (p 0.6). CONCLUSION: Poor sleep quality and excessive daytime sleepiness are common in the 2nd and 3rd TM in an unselected obstetric population. Hypnotic use, reported in 14%, was associated with worsened sleep quality, but did not affect daytime sleepiness. Poster Session V Academic Issues, Antepartum Fetal Assessment, Genetics, Hypertension, Medical-Surgical-Disease www.AJOG.org

Abstract A51: Quantitative assessment of VEGF combined with known clinical variables predicts outcome in patients with diffuse large B-cell lymphoma (DLBCL): Implications for guiding future treatment selection.

Abstract Introduction: The pathogenesis of DLBCL is both complex and heterogeneous, and pathogenetic mechanisms remain largely unknown. A clinical index known as the international prognostic index (IPI) remains the ‘gold-standard’ for predicting response to therapy and overall survival; however, more recently, several groups have reported the utility of gene expression arrays to sub-classify patients and predict outcome. We sought to interrogate a series of biomarkers with multiplex quantitative immunofluorescence (QIF) to construct novel clinical algorithms which predict treatment response and overall survival (OS) post-treatment in a series of patients with DLBCL. Methods: Clinical data including outcome and tissue microarray (TMA) cores for 149 patients were evaluated with QIF for association with response to standard chemotherapy and rituximab. Multiplex QIF with CRI Nuance imaging software for MUM1, Ki67, CD20, CD34, BCL2, BCL6, CD68, VEGF, HLA-DR, and NfKB. Kaplan-Meier survival function curves, hazard ratio and concordance index (CoI) were used to associate marker expression with OS and response to therapy. Results: The cohort demographics for the 149 patients include a median age of 61 years; 43% female, 57% male and an IPI breakdown: Stage1 (30%), II (23%), III (25%) and IV (22%). Both low IPI and female gender were reliable predictors of good overall survival (p&amp;lt;0.001, respectively). 67 of 149 patients had complete clinical and QIF data for subsequent analysis. Increasing tumoral-CD68 positive cells were associated with a poor objective response to treatment (complete/partial vs. none; p=0.019). For OS, increasing levels of VEGF, on a continuous scale, were predictive of poor OS (VEGF: HR 1.23, 95%CI [.98–1.54], p=0.010, CoI 0.34 and when a cut-point of 0.18 is applied, the significance of VEGF improves, p=0.005. By contrast only the BCL6 cut-point (&amp;gt;3.84, p=0.003) was found to be significant in predicting poor OS. An expanded patient cohort will be needed to confirm both results, especially the BCL6 expression profiles. Optimal multivariate model to predict OS consisted of: Age, IPI index and VEGF levels (CoI 0.78, p=0.001), compared to a model with just age and IPI (CoI 0.72, p=0.02). Conclusion: CD68, VEGF and BCL6 appear to be useful markers for understanding patterns of response and outcome for patients with DLBCL, post-treatment. A combined age, IPI and VEGF clinical algorithm may guide future treatment options for selected individuals. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the AACR-NCI-EORTC International Conference: Molecular Targets and Cancer Therapeutics; 2011 Nov 12-16; San Francisco, CA. Philadelphia (PA): AACR; Mol Cancer Ther 2011;10(11 Suppl):Abstract nr A51.

655 A New Bowel Management Strategy Reduces Gastrointestinal Complications and Improves Patient Outcomes after Ventricular Assist Device Implantation

Purpose: Post-cardiotomy cardiogenic shock carries a substantial morbidity and mortality. Short term ventricular assist devices (VADs) are frequently used in patients with cardiogenic shock refractory to conventional therapy. Here we report the outcome of our experience. Methods and Materials: Retrospective analysis was performed for patients who received either left, right, or bi-ventricular assist device for postcardiotomy shock between 1/2001 to 5/2010. Results: During the study period, 37 patients required short term VAD (11 LVADs, 9 RVADs, and 17 BIVADs) support. 18 patients received Centrimag, 8 received Abiomed BVS, and 11 received other devices. Initial cardiac surgery were CABG (15 patients), CABG/Valve (15 patients), Valve (4 patients) and other procedures (3 patients). The overall one-year-survival was 25.4% by Kaplan-Meier survival function. VAD specific 1-year survival was 36.4% for LVAD, 33.3% for RVAD, 10.1% for BIVAD, and was not significantly different by log-rank test (p 0.69). Median interval to device implantation was 3 days (range 0 25 days). Patients that underwent VAD implantation on postoperative day 2 to 6 had better survival (6 of 15 survived) than those who underwent implantation within the first day (2 of 11 survived) or 7 days following the initial cardiotomy (1 of 11 survived). There was no improvement of survival over the study period from 2001 to 2010. Conclusions: The outcome of VAD support for postcardiotomy shock remains poor. Requirement of VAD support in early postoperative period, but not immediately after the cardiotomy, seems to be associated with better overall survival. 655

Bootstrap test for proportional hazard assumption on time-independent systematic effects of longevity data

This manuscript focuses on the development of a bootstrap test for validating the proportional hazard (PH) assumption in longevity data, avoiding parametric assumptions on baseline survival and hazard patterns, and subjective interpretations of previously developed graphical tests. Monte Carlo simulations are used to generate new data sets from the estimated Kaplan-Meier survival function, and inferences are then made on the coefficient of variation (CV) of the estimated hazard over time. One-tailed bootstrap intervals can be established, given that the CV could theoretically range between 0 (perfect PH) and +∞ (absolute loss of proportionality between hazard functions). This procedure was tested by simulation, and the obtained results suggested it as a useful statistical tool when Kaplan-Meier assumptions are satisfied. If not, this bootstrap test was robust for medium to large data sets, whereas it could suffer from statistical biases when testing small populations.

Exact Waiting Time Survival Function

Although Kaplan-Meier survival function is the most commonly used statistical technique of survival analysis, it possesses a disadvantage. It may occur that Kaplan-Meier gives same survival probabilities for two groups having the same number of events and censored observations, although time spans between consecutive events (i.e. waiting times) may considerable vary. Therefore, severity of a disease, in terms of survival times, has no role in the conventional concept of Kaplan-Meier. To overcome this problem, in this paper we propose an exact waiting time survival function by explicitly considering waiting times between events. A new variance estimator, reducing to binomial variance in case of data free from censoring and time differences between two consecutive events equalling to 1, is presented. In order to compare the performance of the new estimator with conventional Kaplan-Meier estimator for small to large sample sizes, as well as for small to heavy censoring, we conducted a simulation study. The outcome shows that on average Pitman Closeness Criteria gives results in favour of our new estimator and confidence intervals have higher coverage rates, as compared to that obtained by Kaplan-Meier estimator, especially for lower confidence limits. Furthermore widths of confidence intervals are smaller than those based on Kaplan-Meier and Greenwood standard error. The proposed procedures are applied to a lung cancer data set.

Long-term outcome in patients with a pathological complete response after chemoradiation for rectal cancer: a pooled analysis of individual patient data

Locally advanced rectal cancer is usually treated with preoperative chemoradiation. After chemoradiation and surgery, 15-27% of the patients have no residual viable tumour at pathological examination, a pathological complete response (pCR). This study established whether patients with pCR have better long-term outcome than do those without pCR. In PubMed, Medline, and Embase we identified 27 articles, based on 17 different datasets, for long-term outcome of patients with and without pCR. 14 investigators agreed to provide individual patient data. All patients underwent chemoradiation and total mesorectal excision. Primary outcome was 5-year disease-free survival. Kaplan-Meier survival functions were computed and hazard ratios (HRs) calculated, with the Cox proportional hazards model. Subgroup analyses were done to test for effect modification by other predicting factors. Interstudy heterogeneity was assessed for disease-free survival and overall survival with forest plots and the Q test. 484 of 3105 included patients had a pCR. Median follow-up for all patients was 48 months (range 0-277). 5-year crude disease-free survival was 83.3% (95% CI 78.8-87.0) for patients with pCR (61/419 patients had disease recurrence) and 65.6% (63.6-68.0) for those without pCR (747/2263; HR 0.44, 95% CI 0.34-0.57; p<0.0001). The Q test and forest plots did not suggest significant interstudy variation. The adjusted HR for pCR for failure was 0.54 (95% CI 0.40-0.73), indicating that patients with pCR had a significantly increased probability of disease-free survival. The adjusted HR for disease-free survival for administration of adjuvant chemotherapy was 0.91 (95% CI 0.73-1.12). The effect of pCR on disease-free survival was not modified by other prognostic factors. Patients with pCR after chemoradiation have better long-term outcome than do those without pCR. pCR might be indicative of a prognostically favourable biological tumour profile with less propensity for local or distant recurrence and improved survival. None.

Maternal obesity, health status during pregnancy, and breastfeeding initiation and duration

Objective. The purpose of this study was to determine whether maternal prepregnancy overweight/obesity has independent effects on breastfeeding initiation and duration and whether these effects are different for women who experience medical problems during pregnancy or labor/delivery complications in comparison with those who have no medical or labor/delivery complications.Methods. We used the early childhood longitudinal study-birth cohort data. Kaplan–Meier survival functions, logistic, and Cox regression modeling were used in the analyses.Results. Findings indicate that overweight/obese women with medical or labor/delivery complications were less likely to initiate breastfeeding in comparison with their counterparts of normal weight. We did not find an independent effect of prepregnancy overweight/obesity on breastfeeding initiation among women with no medical problems. This group of women, however, had an 11% increased risk of stopping breastfeeding with each additional month of breastfeeding duration in comparison to those of normal weight.Conclusions. It is important to evaluate the health history and pregnancy complications among overweight/obese mothers in developing interventions for successful initiation and duration of breastfeeding.

Effect of candidate gene polymorphisms on the course of attention deficit hyperactivity disorder

The main aim of this study was to examine the association between attention-deficit hyperactivity disorder (ADHD)-associated genes and the course of ADHD. Subjects were derived from identically designed case-control family studies of boys and girls with ADHD and a genetic linkage study of families with children with ADHD. Caucasian probands and family members with ADHD and with available genetic data were included in this analysis ( N = 563). The course of ADHD was compared in subjects with and without putative risk alleles ( DRD4 7-repeat allele, DAT1 10-repeat allele, and 5HTTLPR long allele). The persistence of ADHD (full or subthreshold diagnosis in the last month) was plotted using Kaplan-Meier survival functions and tested with Cox proportional hazard models. Survival analyses revealed that by 25 years of age 76% of subjects with a DRD4 7-repeat allele were estimated to have significantly more persistent ADHD compared with 66% of subjects without the risk allele. In contrast, there were no significant associations between the course of ADHD and the DAT1 10-repeat allele ( P = 0.94) and 5HTTLPR long allele. Our findings suggest that the DRD4 7-repeat allele is associated with a more persistent course of ADHD.

A study of duration of digital dermatitis lesions after treatment in a Danish dairy herd

Digital dermatitis (DD) is a contagious disease of cattle affecting the skin adjacent to the claws. Disease dynamics of DD have been described to some extend, but we still need to quantify the duration of lesions and look into non-treatment factors affecting this. The aim of this study was to estimate the duration of lesions due to DD and to evaluate parity and lactation stage as potential risk factors for longer duration of such lesions. An estimate of the duration of lesions will be a valuable parameter in the evaluation of the economic impact of the disease and will additionally allow estimates of incidence based on prevalence figures. From May 2007 until November 2008, lesions associated with DD in the hind legs of 151 Danish Holstein cows at the Danish Cattle Research Centre were clinically scored on fifteen occasions. The mean interval between recordings was 39 days. Onset and end of each new case were estimated as midway between recordings prior to a change in the presence or absence of a lesion. Kaplan-Meier survival functions and Cox proportional hazard regression were performed to estimate the duration and analyse differences in the duration of lesions between primi- and multiparous cows and between different stages in lactation at onset of the lesion. The median duration of lesions were estimated to be 42 days, less than most previous published estimates. The relatively aggressive regime of topical treatment in the study herd might have shortened the duration of the lesions. Furthermore the comparatively long interval between recordings introduced an element of uncertainty in this estimate. No significant effects of parity or days in milk at lesion onset on the duration of DD were found using these data though lesions developed earlier in the lactation may have a longer duration. Further data would be needed to confirm the latter.

Pretransplantation Patient Characteristics and Survival Following Combined Heart and Kidney Transplantation

Pretransplantation patient characteristics determine survival following combined heart and kidney transplantation (HKT). Time-to-event analysis. Academic research. The United Network for Organ Sharing provided deidentified patient-level data. Analysis included 19,373 heart transplant recipients from January 1, 1995, to December 31, 2005. Multivariate Cox proportional hazards regression analysis was performed to identify pretransplantation recipient characteristics associated with improved long-term survival following HKT. Kaplan-Meier survival functions and Cox proportional hazards regression were used for time-to-event analysis. Using the relative risks calculated in regression analysis, weights were assigned for each risk factor, allowing for the construction of a risk score. Among heart transplant recipients, 264 (1.4%) underwent HKT. Factors associated with diminished survival included peripheral vascular disease, recipient age older than 65 years, nonischemic etiology of heart failure, dialysis dependence at the time of transplantation, and bridge to transplantation using a ventricular assist device. After stratification by risk score, 1-year survival was 93.2% and 61.9% in the lowest- and highest-risk HKT groups, respectively. Further stratification by estimated glomerular filtration rate (eGFR) was performed based on a previous study showing decreased survival of patients undergoing orthotopic heart transplantation with a preoperative eGFR of less than 33 mL/min. Low-risk patients with an eGFR of less than 33 mL/min undergoing HKT constituted the only group that had significantly better survival compared with isolated patients undergoing orthotopic heart transplantation with eGFRs and risk scores in the same range (P = .006). When patients were stratified by risk score and by diminished eGFR (<33 mL/min), low-risk HKT recipients with a diminished eGFR had improved survival following HKT over isolated heart transplant recipients. Only low-risk patients with combined kidney failure (eGFR, <33 mL/min) and heart failure seem to gain a survival benefit from HKT.

Has Prenatal Screening Influenced the Prevalence of Comorbidities Associated With Down Syndrome and Subsequent Survival Rates?

With this study we aimed to compare survival rates for children with Down syndrome in 2 time periods, 1 before prenatal screening (1988-1990) and 1 contemporaneous with screening (1998-2000), and to examine the frequency of comorbidities and their influence on survival rates. Record-linkage was performed between the population-based Victorian Birth Defects Register and records of deaths in children up to 15 years of age collected under the auspice of the Consultative Council on Obstetric and Pediatric Mortality and Morbidity. Cases of Down syndrome were coded according to the presence or absence of comorbidities by using the International Classification of Diseases, Ninth Revision classification of birth defects. Kaplan-Meier survival functions and log rank tests for equality of survival distributions were performed. Of infants liveborn with Down syndrome in 1998-2000, 90% survived to 5 years of age, compared with 86% in the earlier cohort. With fetal deaths excluded, the proportion of isolated Down syndrome cases in the earlier cohort was 48.7% compared with 46.1% in the most recent cohort. In 1988-1990 there was at least 1 cardiac defect in 41.1% of cases and in 45.4% in 1998-2000. There was significant variation in survival rates for the different comorbidity groupings in the 1988-1990 cohort, but this was not so evident in the 1998-2000 cohort. Survival of children with Down syndrome continues to improve, and there is an overall survival figure of 90% to at least 5 years of age. It is clear from this study that prenatal screening technologies are not differentially ascertaining fetuses with Down syndrome and additional defects, because there has been no proportional increase in births of isolated cases with Down syndrome.

Long-term Outcomes of Intraoperative 5-Fluorouracil versus Intraoperative Mitomycin C in Primary Trabeculectomy Surgery

To compare the long-term efficacy of intraoperative 5-fluorouracil (5-FU) and mitomycin C (MMC) in primary trabeculectomy. Comparative case series and extension of a prospective, randomized controlled trial. One hundred fifteen eyes of 103 subjects undergoing primary trabeculectomy with either intraoperative 5-FU or MMC. This study is an extension of a 12-month, prospective, double-masked, randomized trial that previously was reported. Subjects were randomized to receive either intraoperative 5-FU (50 mg/ml for 5 minutes) or MMC (0.2 mg/ml for 2 minutes) during primary trabeculectomy. Follow-up data were collected from patient medical records after informed consent was obtained from the study patient. Attempts were made to contact subjects who had been lost to follow-up, and some consented to reexamination. The primary measure was Kaplan-Meier survival function, with failure defined as intraocular pressure (IOP) of more than 21 mmHg or less than 6 mmHg on 2 consecutive visits, less than 20% reduction from baseline IOP, loss of light perception vision, or additional glaucoma surgery to lower IOP (except bleb revision). Secondary measures included IOP, number of glaucoma medications, visual acuity, additional surgeries, and number and type of complications. Mean follow up was 53.4+/-31.4 months (interquartile range, 34-82 months) in the 5-FU group and 45.3+/-28.0 months (interquartile range, 19-70 months) in the MMC group (P = 0.15, t test). Kaplan-Meier success was 0.83 at 3 years and 0.76 at 5 years in the 5-FU group and 0.79 at 3 years and 0.66 at 5 years in the MMC group (P = 0.18, log-rank test). Bleb leakage was the most common complication in each group and developed in approximately 4% of subjects in each group per year (P = 0.33, log-rank test). There were no significant differences between topical 5-FU and topical MMC in reducing IOP of eyes undergoing primary trabeculectomy. Both antifibrosis agents may contribute to the development of bleb leakage. The author(s) have no proprietary or commercial interest in any materials discussed in this article.

Diagnóstico e prognóstico dos tumores pulmonares neuroendócrinos mediante microscopia eletrônica e análise multivariavel de agrupamento

To establish reproducible electron microscopic criteria for identifying the four major types of neuroendocrine tumors of the lung: carcinoid; atypical carcinoid; large cell neuroendocrine carcinoma; and small cell carcinoma. Measurements were made on electron micrographs using a digital image analyzer. Sixteen morphometric variables related to tumor cell differentiation were assessed in 27 tumors. The examination under electron microscopy revealed that all of the tumors could be classified as belonging to one of the four categories listed above. Cluster analysis of the morphometry variables was used to group the tumors into three clusters, and Kaplan-Meier survival function curves were employed in order to draw correlations between each cluster and survival. All three clusters of neuroendocrine carcinomas were found to be associated with survival curves, demonstrating the prognostic significance of electron microscopic features. The tumors fell into three well-defined clusters, which represent the spectrum of neuroendocrine differentiation: typical carcinoid (cluster 1); atypical carcinoid and large cell neuroendocrine carcinoma (cluster 2); and small cell carcinoma (cluster 3). Cluster 2 represents an intermediate step in neuroendocrine carcinogenesis, between typical carcinoid tumors and small cell carcinomas. Our findings confirm that electron microscopy is useful in making the diagnosis and prognosis in cases of lung tumor.

An Exploration of Factors Reducing Recidivism Rates of Formerly Incarcerated Youth with Disabilities Participating in a Re-Entry Intervention

Juvenile offenders are costly to our society in terms of the monetary and social expenditures from the legal system, victims’ person costs, and incarceration. The re-entry and community reintegration outcomes for formerly incarcerated youth with a disabling condition are bleak compared to peers without disabilities. In this study, we examined the factors––both static and intervention-related––that reduce recidivism rates for project participants possessing a mental health and/or special education diagnosis and were served by a facility-to-community re-entry intervention. Our sample was comprised of a total of 320 youth formerly incarcerated who received project services between August of 1999 through June of 2004. Project services include a transition specialist whose role is service coordination to youth from the facility to the community. Kaplan–Meier survival functions were calculated for participants on the outcome measure of the rates of recidivism. Additionally, Cox regression modeling was used to identify factors for the prediction of time to recidivate. Our findings indicate that the intervention may contribute to reduced recidivism rates but that a set of static demographic and pre-incarceration risks contribute more to the prediction of recidivism than community adjustment factors (e.g., employment or enrollment). With these findings, the intervention seems promising yet it remains unclear about which programmatic features contribute to reduced recidivism rates. Implications for research and practice are shared.

T10-O-05 Age at first sex: when does it really happen?

Objective The objective of this study is to promote a less distorted model for assessing age at first sex, especially in less developed countries (LDCs). This could lead to empowerment of sexual education policies globally. Evidence suggests that calculations for age at first sex have been under-estimated by neglecting virgins and censoring (Zaba et al). Methods Findings are based on the 2007 Durex Sexual Wellbeing Global Survey (carried out in July and August 2006 in 26 countries globally). Margin of error in each country was 3% (n=1,000) for a 95% confidence interval. Out of the 26,032 responses, 85.6% were sexually active. Median and mean survival time were calculated based on the Kaplan-Meier survival function and timetables. A Cox regression-based test checked for equality of survival curves by gender, income, age, education, and area of residency. Results Traditional models under-estimate age at first sex by at least twelve months. By allowing censoring and including virgins, the median age at first sex changes from 19.1 (based on arithmetic mean) to 20.2 years globally. Differences were also found in LDCs. Equality checks resulted in differences for age at first sex by area of residency, income, and education. Rural residents with income below-average and low education are more likely to experience earlier first sexual intercourse. Conclusions The implications of these results are: a) improvement of assessment mechanisms of when children and adolescents are in fact losing their virginity; and b) determination of when children and adolescents should be exposed to formal sexual education in schools.

Association of reduced relative dose intensity and survival in lymphoma patients receiving CHOP-21 chemotherapy

Dear Editor, Reductions in average relative chemotherapy dose intensity (ARDI; i.e. administered compared with planned) compromise patient outcomes [3], and a recent Belgian study showed that survival for non-Hodgkin lymphoma (NHL) patients receiving CHOP-21 was reduced when ARDI fell below 90% [2]. We support these findings with similar data from the UK Audit of Lymphoma Patients (n = 78 patients who received CHOP-21 chemotherapy in 1999–2000) and from the combined Belgian and UK data (n = 289) [4]. Patient, disease and treatment characteristics were similar between the two studies, except that the UK patients were younger (mean age ± SD was 55 ± 15 years versus 63 ± 14 years) and fewer UK patients received colony stimulating factor (CSF). First cycle CSF use was 4% in the UK study compared to 25% in the Belgian study. During an average time to death or censoring of 72 months, 35% of patients in the UK study died. In the Belgian study, the average observation time was 30 months, during which 31% of patients died. After adjusting for the higher mean age of the Belgian patients, Kaplan–Meier survival functions were similar between the two populations (log-rank test stratified by 10-year age groups, p = 0.38). Kaplan–Meier plots for patients with >90% ARDI versus ≤90% ARDI showed reduced survival for the patients with ≤90% ARDI (Fig. 1). A trend towards reduced survival was apparent in the UK dataset alone (Fig. 1a; log-rank test-based p = 0.090). For the combined dataset, the effect was statistically significant (p < 0.001; Fig. 1b), as for the Belgian data alone [2]. Fig. 1 Kaplan–Meier survivor estimates by RDI for UK dataset alone (a) and the combined UK and Belgian datasets (b). Asterisk Log-rank test-based p values Potential predictors of reduced survival were assessed using an extended Cox proportional hazards regression model with robust standard errors allowing for clustering by centre. Using the UK dataset, reduced survival was significantly associated with a higher Ann Arbor disease stage (hazard ratio (HR) at treatment initiation 2.59 per stage increase by 1, 95% CI 1.45–4.66, p = 0.001) and showed a trend towards association with age (HR 1.02 per year of age, CI 1.00–1.04, p = 0.058) and RDI ≤ 90% (HR 1.42, CI 0.88–2.28, p = 0.146). Using the combined dataset, reduced survival was associated with more advanced disease stage (HR 2.00, CI 1.44–2.77, p < 0.001), age (HR 1.03, 95% CI 1.01–1.05, p = 0.002) and RDI ≤ 90% (HR 1.77, CI 1.12–2.79, p = 0.014). In both models, the strength of the association with disease stage decreased over time. Approximately 23% and 30% of UK and Belgian patients, respectively, received ARDI ≤ 90% and were, therefore, at risk of reduced survival. There are many factors that result in a decision to reduce or delay chemotherapy, including local institutional practice. Particularly relevant is the higher proportion of elderly patients in the Belgian dataset; elderly patients are at high FN risk [1] and often receive lower doses of chemotherapy [3]. Despite this common practice, dose-dense CHOP-14 chemotherapy supported with G-CSF has been shown to be efficacious and well tolerated in both young and elderly NHL patients [5, 6]. This study highlights the potential impact of receiving ARDI ≤ 90% on survival. While further investigation is needed, delivering full chemotherapy dose intensity remains an important goal in NHL patients who receive CHOP-21 chemotherapy.

The risk of fetal loss following a prenatal diagnosis of trisomy 13 or trisomy 18

The objective of this study is to determine the risk of fetal loss (spontaneous abortion or stillbirth) following a prenatal diagnosis of trisomy 13 (T13; Patau syndrome) or trisomy 18 (T18; Edwards syndrome). Five regional congenital anomaly registers in England and Wales provided details on the outcomes of 198 pregnancies prenatally diagnosed with T13 and 538 prenatally diagnosed with T18. For each pregnancy the time from prenatal diagnosis until birth, miscarriage or termination occurred was calculated and these times were analyzed using Kaplan-Meier survival functions. Our results showed that between 12 weeks gestation and term an estimated 49% (95% CI: 29-73%) of pregnancies diagnosed with T13 and 72% (61-81%) of pregnancies diagnosed with T18 ended in a miscarriage or stillbirth. Between 18 weeks and term the proportions were 42% (18-72%) for T13 and 65% (57-79%) for T18 and between 24 weeks and term the proportions were 35% (5-70%) for T13 and 59% (49-77%) for T18. Male fetuses with T18 appeared to be more likely to be lost than female fetuses. These are the most precise estimates currently available for the risk of loss in a general population. These estimates should be useful in counseling women who are carrying an affected fetus and knowing the risk of fetal loss is essential to compare the performance of prenatal screening programs occurring in the first and second trimester.