Survival analysis of time-to-event data in respiratory health research studies

This article provides a review of techniques for the analysis of survival data arising from respiratory health studies. Popular techniques such as the Kaplan–Meier survival plot and the Cox proportional hazards model are presented and illustrated using data from a lung cancer study. Advanced issues are also discussed, including parametric proportional hazards models, accelerated failure time models, time-varying explanatory variables, simultaneous analysis of multiple types of outcome events and the restricted mean survival time, a novel measure of the effect of treatment.

Assessment of delay in age-appropriate vaccination using survival analysis

 Assessment of delay in age-appropriate vaccination provides more information about timeliness of vaccination than up-to-date vaccination coverage. The authors applied survival analysis methods to data from a vaccination coverage survey among children aged 13–59 months conducted in Argentina in 2002. By age 19 months, 43% of children (95% confidence interval (CI): 40, 46) were vaccinated with the fourth dose of diphtheria, tetanus, and pertussis (DTP4). By age 13 months, 55% of children (95% CI: 52, 57) were vaccinated with measles-containing vaccine. By age 7 months, 33% of children (95% CI: 27, 40) were vaccinated with the third dose of hepatitis B. Compared with firstborn children, third children were more likely to be delayed for DTP4 (relative risk (RR) = 1.41, 95% CI: 1.22, 1.62), measles-containing vaccine (RR = 1.54, 95% CI: 1.32, 1.78), and the third dose of hepatitis B (RR = 1.31, 95% CI: 1.03, 1.67). Children whose caregivers had completed secondary school were less likely to be delayed for DTP4 (RR = 0.68, 95% CI: 0.52, 0.90) compared with those whose caregivers had not completed primary school. Survival analysis methods were helpful in measuring vaccine uptake and should be considered in future surveys when assessing delay in age-appropriate vaccination. Copyright © 2006 by the Johns Hopkins Bloomberg School of Public Health All rights reserved.

A comparative analysis of decision trees vis-à-vis other computational data mining techniques in automotive insurance fraud detection

The development and application of computational data mining techniques in financial fraud detection and business failure prediction has become a popular cross-disciplinary research area in recent times involving financial economists, forensic accountants and computational modellers. Some of the computational techniques popularly used in the context of - financial fraud detection and business failure prediction can also be effectively applied in the detection of fraudulent insurance claims and therefore, can be of immense practical value to the insurance industry. We provide a comparative analysis of prediction performance of a battery of data mining techniques using real-life automotive insurance fraud data. While the data we have used in our paper is US-based, the computational techniques we have tested can be adapted and generally applied to detect similar insurance frauds in other countries as well where an organized automotive insurance industry exists.

Disparities in pediatric leukemia early survival in Argentina: a population-based study

To identify disparities-using recursive partitioning (RP)-in early survival for children with leukemias treated in Argentina, and to depict the main characteristics of the most vulnerable groups.

Board structure and survival of new economy IPO firms

Research Question/Issue: This study examines the relevance of currently accepted best practice recommendations regarding board structure on the survival likelihood of new economy initial public offering companies. We argue that industry context determines governance outcomes. Research Findings/Insights: We study 125 Australian new economy firms listed between 1994 and 2002. Each firm is tracked until the end of 2007 for monitoring their survival. We find that board independence is associated with an increase in the likelihood of corporate survival. We also find that the benefits of board independence increase at a decreasing rate. Theoretical/Academic Implications: The standard best practice recommendation of board independence stems from the monitoring role of directors and is based on agency theory. The results from our study suggest that the recommendation regarding board independence does not work well for new economy firms. While the agency theory based model implies a monotonic relation between board independence and performance, our research suggests that the relationship is nonlinear. This variation occurs because of increased monitoring costs faced by outsiders due to higher information asymmetry and complexity of new economy firms. Our empirical results suggest that inside directors play a complementary role to outsiders in mitigating firm failure. Practitioner/Policy Implications: Our research offers insights to policy makers who are interested in setting best practice standards regarding board structure. Our research suggests that firm/industry characteristics play a crucial role in determining the optimal board structure. In firms/industries where outsiders face significantly higher information processing costs, insiders can play a valuable complementary role to outsiders in enhancing the effectiveness of the board. Thus future hard or soft regulations related to board structure should consider industry context.

Impact of early valve surgery on outcome of Staphylococcus aureus prosthetic valve infective endocarditis: analysis in the International Collaboration of Endocarditis-Prospective Cohort Study

BACKGROUND: The impact of early valve surgery (EVS) on the outcome of Staphylococcus aureus (SA) prosthetic valve infective endocarditis (PVIE) is unresolved. The objective of this study was to evaluate the association between EVS, performed within the first 60 days of hospitalization, and outcome of SA PVIE within the International Collaboration on Endocarditis-Prospective Cohort Study. METHODS: Participants were enrolled between June 2000 and December 2006. Cox proportional hazards modeling that included surgery as a time-dependent covariate and propensity adjustment for likelihood to receive cardiac surgery was used to evaluate the impact of EVS and 1-year all-cause mortality on patients with definite left-sided S. aureus PVIE and no history of injection drug use. RESULTS: EVS was performed in 74 of the 168 (44.3%) patients. One-year mortality was significantly higher among patients with S. aureus PVIE than in patients with non-S. aureus PVIE (48.2% vs 32.9%; P = .003). Staphylococcus aureus PVIE patients who underwent EVS had a significantly lower 1-year mortality rate (33.8% vs 59.1%; P = .001). In multivariate, propensity-adjusted models, EVS was not associated with 1-year mortality (risk ratio, 0.67 [95% confidence interval, .39-1.15]; P = .15). CONCLUSIONS: In this prospective, multinational cohort of patients with S. aureus PVIE, EVS was not associated with reduced 1-year mortality. The decision to pursue EVS should be individualized for each patient, based upon infection-specific characteristics rather than solely upon the microbiology of the infection causing PVIE.

Hepatitis B Virus transmission and hepatocarcinogenesis: a 9 year retropective cohort of 13676 relatives with hepatocellular carcinoma

Background/Aims
Familial clustering of hepatitis B virus (HBV) infection is related to perinatal transmission, and is the main cause of familial-type hepatocellular carcinoma (HCC). The route of HBV transmission differs between the children and siblings of patients with HCC. This study examined the differences in HBV carrier rates and HCC-related mortality between two generations in HCC families.
Methods
From 1992 to 1997, relatives of individuals with HCC were screened prospectively with ultrasonography, alpha-fetoprotein, liver biochemistry tests and viral markers. Total HCC-related deaths during a 9-year period were compared between the generations of index patients and their children.
Results
The study included a total of 13 676 relatives in two generations. More HCC-related deaths occurred in the index patient generation than in the child generation. Furthermore, children of female index patients had higher rates of liver cancer related mortality than children of male index patients. The same was true when the analysis was limited to male HBV carriers. The prevalence of HBsAg in the offspring of HBsAg positive mothers was 66% in the child generation and 72% in the index patient generation. These high prevalences indicated high maternal HBV replication status.
Conclusions
Perinatal transmission and maternal viral load are important risk factors in hepatocarcinogenesis.

Potential applications of discrete-event simulation and fuzzy rule-based systems to structural reliability and availability

This chapter discusses and illustrates some potential applications of discrete-event simulation (DES) techniques in structural reliability and availability analysis, emphasizing the convenience of using probabilistic approaches in modern building and civil engineering practices. After reviewing existing literature on the topic, some advantages of probabilistic techniques over analytical ones are highlighted. Then, we introduce a general framework for performing structural reliability and availability analysis through DES. Our methodology proposes the use of statistical distributions and techniques – such as survival analysis – to model component-level reliability. Then, using failure- and repair-time distributions and information about the structural logical topology (which allows determination of the structural state from their components’ state), structural reliability, and availability information can be inferred. Two numerical examples illustrate some potential applications of the proposed methodology to achieving more reliable and structural designs. Finally, an alternative approach to model uncertainty at component level is also introduced as ongoing work. This new approach is based on the use of fuzzy rule-based systems and it allows the introduction of experts’ opinions and evaluations in our methodology.

A brief review of recent research trends on applications of computational and statistical techniques in financial and business intelligence

Artificial neural networks and statistical techniques like decision trees, discriminant analysis, logistic regression and survival analysis play a crucial role in Business Intelligence. These predictive analytical tools exploit patterns found in historical data to make predictions about future events. In this paper we have shown some recent developments of a few of these techniques in financial and business intelligence applications like fraud detection, bankruptcy prediction and credit rating scoring.

Post-traumatic stress disorder and quality of life in sexually abused Australian children

The study used publicly available data on post-traumatic stress disorder in a sample of the Australian population with a history of sexual abuse to demonstrate how this evidence can inform economic analyses. The 2007 Australian Mental Health Survey revealed that 8.3% of 993 adolescents experienced childhood sexual abuse, of which 40.2% were diagnosed with post-traumatic stress disorder. Post-traumatic stress disorder diagnosis corresponded to a significant loss of quality of life. Survival analysis was used to estimate the lifetime persistence of post-traumatic stress disorder symptoms. The average time between post-traumatic stress disorder onset and remission was 11.4 years. Results suggest that successful treatment of post-traumatic stress disorder will save 2.05 quality adjusted life years per child or adolescent with post-traumatic stress disorder.

A randomised controlled trial of the effectiveness of soft silicone multi-layered foam dressings in the prevention of sacral and heel pressure ulcers in trauma and critically ill patients: The border trial

The prevention of hospital acquired pressure ulcers in critically ill patients remains a significant clinical challenge. The aim of this trial was to investigate the effectiveness of multi-layered soft silicone foam dressings in preventing intensive care unit (ICU) pressure ulcers when applied in the emergency department to 440 trauma and critically ill patients. Intervention group patients (n = 219) had Mepilex^® Border Sacrum and Mepilex^® Heel dressings applied in the emergency department and maintained throughout their ICU stay. Results revealed that there were significantly fewer patients with pressure ulcers in the intervention group compared to the control group (5 versus 20, P = 0·001). This represented a 10% difference in incidence between the groups (3·1% versus 13·1%) and a number needed to treat of ten patients to prevent one pressure ulcer. Overall there were fewer sacral (2 versus 8, P = 0·05) and heel pressure ulcers (5 versus 19, P = 0·002) and pressure injuries overall (7 versus 27, P = 0·002) in interventions than in controls. The time to injury survival analysis indicated that intervention group patients had a hazard ratio of 0·19 (P = 0·002) compared to control group patients. We conclude that multi-layered soft silicone foam dressings are effective in preventing pressure ulcers in critically ill patients when applied in the emergency department prior to ICU transfer.

Multiple states of financially distressed companies: Tests using a competing-risks model

This study examines the determinants of multiple states of financial distress by applying a competing-risks model. It investigates the effect of financial ratios, market-based variables and company-specific variables, including company age, size and squared size on three different states of corporate financial distress: active companies; distressed external administration companies; and distressed takeover, merger or acquisition companies. A sample of 1,081 publicly listed Australian non-financial companies over the period 1989 to 2005 using a competing-risks model is used to determine the possible differences in the factors of entering various states of financial distress. It is found that specifically, distressed external administration companies have a higher leverage, lower past excess returns and a larger size; while distressed takeover, merger or acquisition companies have a lower leverage, a higher capital utilisation efficiency and a larger size compared to active companies. Comparing the results from both the single-risk model and the competing-risks model reveals the need to distinguish between financial distress states.

Retirement patterns of Australian doctors aged 65 years and older.

OBJECTIVE: To investigate retirements over a 4-year period among Australian general practitioners (UPs) and specialists aged 65 years and over, and factors influencing retirement. METHODS: Data from Medicine in Australia: Balancing Employment and Life (MABEL) for the years 2009-12 were analysed for 435 GPs and 643 specialists aged 65 years and over at the time of entry to the MABEL survey. Discrete time survival analysis was used. RESULTS: The retirement rates were 4.1% (2009), 5.1% (2010), 4.2% (2011) and 10.4% (2012). Retirement was associated with: (1) the intention to leave medical work in 2009 and 2010; (2) working fewer hours in private consulting rooms in 2010 and 2012; (3) having lower job satisfaction in 2009 and 2011; (4) being older in 2009; (5) working fewer hours in a public hospital in 2012; and (6) working fewer hours in a private hospital in 2010. Doctors who intended to reduce their working hours were less likely to retire in 2009. CONCLUSIONS: Strategies to support doctors at the late career stage to provide their valued contributions to the medical workforce for as long as possible may include increasing job satisfaction and addressing barriers to reducing work hours.