This thesis studies two aspects of the statistical analysis of complex data in survival and event history analysis. After a short introduction to survival and event history analysis in Chapter 1, we proposed a multivariate proportional intensity factor model for multivariate counting processes in Chapter 2. In an exploratory analysis on process data, a large number of possibly time-varying covariates maybe included. These covariates along with the high-dimensional counting processes often exhibit a low-dimensional structure that has meaningful interpretation. We explore such structure through specifying random coefficients in a low dimensional space through a factor model. For the estimation of the resulting model, we establish the asymptotic theory of the nonparametric maximum likelihood estimator (NPMLE). In particular, the NPMLE is consistent, asymptotically normal and asymptotically efficient with covariance matrix that can be consistently estimated by the inverse information matrix or the profile likelihood method under some suitable regularity conditions. Furthermore, to obtain a parsimonious model and to improve interpretation of parameters therein, variable selection and estimation for both fixed and random effects are developed by penalized likelihood. We illustrate the method using simulation studies as well as a real data application from The Programme for the International Assessment of Adult Competencies (PIAAC). Chapter 3 concerns rare events and sparse covariates in event history analysis. In large-scale longitudinal observational databases, the majority of subjects may not experience a particular event of interest. Furthermore, the associated covariate processes could also be zero for most of the subjects at any time. We formulate such setting of rare events and sparse covariates under the proportional intensity model and establish the validity of using the partial likelihood estimator and the observed information matrix for inference under this framework.

Background: Burkitt lymphoma (BL) is a rare, highly aggressive B-cell malignancy treated most successfully with brief-duration, high-intensity chemotherapeutic regimens. The benefit of the addition of rituximab to these regimens remains uncertain. We sought to examine the effectiveness of chemotherapy with and without rituximab in patients with BL. Methods: This study is a retrospective cohort study of all adult patients with BL diagnosed and treated with modern, dose-intense chemotherapeutic regimens from 1998–2008 at two tertiary care institutions. All cases were confirmed by application of WHO 2008 criteria by hematopathologists. Medical records were reviewed for patient-, disease-, and treatment- related factors as well as treatment response and survival. Factors associated with survival were analyzed using Cox proportional hazards modeling. Results: A total of 35 patients were analyzed: 18 patients received rituximab with chemotherapy (R-chemo) and 17 received chemotherapy (chemo) alone. The median age was 42 (range 20–74 years); 57% were male; 71% had Ann Arbor Stage IV disease; 33% had central nervous system involvement; 78% had an Eastern Cooperative Oncology Group (ECOG) performance status of 0–1. R-chemo was associated with significantly longer overall survival (OS) than chemo alone (5-year OS 70% and 29%, respectively, p = 0.040). On multivariate regression analysis, poor performance status and central nervous system involvement were associated with poorer survival. The addition of rituximab to chemotherapy was associated with improved OS in patients with Burkitt lymphoma. Poor performance status and central nervous system involvement were prognostically significant on multivariate analysis.

Over the past several decades, the incidence of thyroid cancer in the United States has increased substantially surmounting to higher levels of concern for physicians around the nation. This concern led to a research investigation surrounding the increased risk thyroid cancer patients may face by delaying their surgeries. OBJECTIVE: We aim to evaluate the impact of a delay in surgical intervention on survival in patients with a diagnosis of papillary thyroid cancer. METHODS: This is an observational retrospective study focusing on disease specific survival using SEER-Medicare data as well as data from the National Cancer Database to analyze whether a delay in surgical intervention leads to a decrease in mortality in patients with papillary thyroid cancer. This study focuses on specific thyroid cancer association beyond that of another research paper that found delaying time to surgery does decrease overall survival as discovered by Dr. Scott Fligor in 2021. For the purposes of this study, data was accessed between the years 1999-2018. A survival analysis was performed using the Cox- hazard ratio as well as Kaplan-Meier curves. RESULTS: Preliminary results detail the fact that delaying surgeries past 180 days for patients led to decreased survival over a course of 5, 10, and 15 years as was determined by the use of Kaplan Meyer curves and the Cox hazard ratio. A positive coefficient for a Cox hazard ratio indicates a worse prognosis whereas a negative coefficient indicates the opposite. The results of this study show that increasing time to surgery increases risk of mortality for patients as the hazard ratios suggest for this research. The hazard ratio for 90 – 180 days delay for patients versus patients who underwent surgery within 0 – 90 days was 1.18 (95% confidence internal, 0.96 – 1.45). This hazard ratio is lower than that of the group that underwent surgery after 180+ days whose hazard ratio was 1.21 (95% confidence internal, 0.89 – 1.66). Since the hazard ratio for patients who delayed surgery after 180+ days is higher than the 0 – 90 days as well as the 90 – 180 days, this indicates a worse prognosis over time for patients with increased delays. CONCLUSION: Delaying surgery for thyroid cancer patients deceases their overall survival over a period of 5, 10, and 15 years. Elective or lower risk surgeries over the past 3 years have been delayed due to the Covid-19 pandemic, and thus this data was excluded for the purposes of this study. Further research should be done on the effects the pandemic had on the overall survival for patients who had to delay their surgeries due to the pandemic.

No

<p>The aim of this study is to illustrate, adapt and develop methods of survival analysis in analysing and estimating customer survival time in subscription-based businesses. Two particular objectives are studied. The rst objective is to redene the existing survival analysis techniques in business terms and to discuss their uses in order to understand various issues related to the customer-rm relationship.</p>

<p>The aim of this study is to illustrate, adapt and develop methods of survival analysis in analysing and estimating customer survival time in subscription-based businesses. Two particular objectives are studied. The rst objective is to redene the existing survival analysis techniques in business terms and to discuss their uses in order to understand various issues related to the customer-rm relationship.</p>

Philosophiae Doctor - PhD / Subscription-based industries have seen a massive expansion in recent decades. In this type of industry the customer has to subscribe to be able to enjoy the service; there-fore, well-de ned start and end points of the customer relationship with the service provider are known. The length of this relationship, that is the time from subscription to service cancellation, is de ned as customer survival time. Unlike transaction-based businesses, where the emphasis is on the quality of a product and customer acquisition, subscription-based businesses focus on the customer and customer retention. A customer focus requires a new approach: managing according to customer equity (the value of a rm's customers) rather than brand equity (the value of a rm's brands). The concept of customer equity is attractive and straightforward, but the implementation and management of the customer equity approach do present some challenges. Amongst these challenges is that customer asset metric - customer lifetime value (the present value of all future pro ts generated from a customer) - depends upon assumptions about the expected survival time of the customer (Bell et al., 2002; Gupta and Lehmann, 2003). In addition, managing and valuing customers as an asset require extensive data and complex modelling. The aim of this study is to illustrate, adapt and develop methods of survival analysis in analysing and estimating customer survival time in subscription-based businesses. Two particular objectives are studied. The fi rst objective is to rede ne the existing survival analysis techniques in business terms and to discuss their uses in order to understand various issues related to the customer-fi rm relationship. The lesson to be learnt here is the ability of survival analysis techniques to extract important information on customers with regard to their loyalties, risk of cancellation of the service, and lifetime value. The ultimate outcome of this process of studying customer survival time will be to understand the dynamics and behaviour of customers with respect to their risk of cancellation, survival probability and lifetime value. The results of the estimates of customer mean survival time obtained from different nonparametric and parametric approaches; namely, the Kaplan-Meier method as well as exponential, Weibull and gamma regression models were found to vary greatly showing the importance of the assumption imposed on the distribution of the survival time. The second objective is to extrapolate the customer survival curve beyond the empirical distribution. The practical motivation for extrapolating the survival curve beyond the empirical distribution originates from two issues; that of calculating survival probabilities (retention rate) beyond the empirical data and of calculating the conditional survival probability and conditional mean survival time at a speci c point in time and for a speci c time window in the future. The survival probabilties are the main components needed to calculate customer lifetime value and thereafter customer equity. In this regard, we propose a survivor function that can be used to extrapolate the survival probabilities beyond the last observed failure time; the estimation of parameters of the newly proposed extrapolation function is based completely on the Kaplan-Meier estimate of the survival probabilities. The proposed function has shown a good mathematical accuracy. Furthermore, the standard error of the estimate of the extrapolation survival function has been derived. The function is ready to be used by business managers where the objective is to enhance customer retention and to emphasise a customer-centric approach. The extrapolation function can be applied and used beyond the customer survival time data to cover clinical trial applications. In general the survival analysis techniques were found to be valuable in understanding and managing a customer- rm relationship; yet, much still needs to be done in this area of research to make these techniques that are traditionally used in medical studies more useful and applicable in business settings. / South Africa

Title of thesis: Evaluation of psychologically and physically load in competitors of Winter Survival; including incorporation of knowhow to hold the competition. Aim of thesis: 1. Creation of a manual to order competition of Winter Survival type in civil and army. 2. Valorize and confront physically load in Winter Survival 2007, 2008, 2011. 3. Valorize and confront psychologically load in Winter Survival 2007, 2008, 2011. Methods: 1. For the creation of the manual was used qualitative research with exploitation of interview with abstract questions. 2. To valorizing and fixing of physically load was used quantitative study. 3. To valorizing and fixing of psychologically load was used quantitative study. Results: 1. We created the "manual" to ordering and organising of cometition for trainee officials of competitions like Winter Survival. 2. The results prove, that competitors of WS moved around middle zone of physical load. WS 2008 was, in light of physical load, much easier, than WS 2007 and 2011. Effect of the extension in 2. part of competition is the extension of physical work in middle zone of physical load to the prejudice moderate zone. 3. Mental condition is impressed with physical load in "state" and "activity" parameter, less connections are between physical load and "mood" parameter; the...

Translocation programmes, particularly when using the release of a captive reared population, often fail in their efforts to create a self-sustaining population. High mortality after release is a key issue and often associated with behavioural, physiological and cognitive deficiencies between the released population and their wild counterparts. Mitigation of these deficiencies is essential for successful translocation programmes. I showed that pheasants (Phasianus colchicus) that were reared in more naturalistic conditions early in life were better suited to the natural environment after they were released into the wild. Post release survival was improved with exposure to more naturalistic diets prior to release. We identified four mechanisms to explain this. Pheasants reared with more naturalistic diets: 1) foraged for less time and had a higher likelihood of performing vigilance behaviours; 2) were quicker at handling live prey items; 3) were less reliant on supplementary feed which could be withdrawn; 4) developed different gut morphology. Consequently birds reduced the risk of predation by reducing exposure time whilst foraging, while allowing more time to be vigilant, were better at handling and discriminating natural food items and not solely reliant on supplementary feed and had a better gut system to cope with the natural forage. Post release survival was also improved when pheasants were reared with access to perches. We identified three mechanisms to explain this. Pheasants reared with access to perches had: 1) a physiology to better enable the birds to fly to the higher branches and cope with prolonged roosting; 2) a higher propensity to roost off the ground at night; and 3) more accurate spatial memory. Consequently, birds were at a reduced risk of terrestrial predation by roosting at night, and accurately remember their new environment upon release. I also showed that these manipulations did not compromise the welfare of the individuals prior to release, as often feared when trying to create a naturalistic environment to a captive population. An additional mechanism that can affect the success of a translocation programmes, operating at the level of the population, considers the optimality of the mixture of released individuals that can influence a release programme. The personality of birds within a released population, tested prior to release into the wild, influenced their fate and dispersal. I suggest a number of release mechanisms that would aid the survival of a diverse range of behavioural types that are essential for the production of a self-sustaining population in a fluctuating environment. I showed that harem size is strongly influenced by the vigilance behaviour of its constituent members. Despite a shared interest in increasing harem size, their optimal size is influenced by trade-offs in individual vigilance behaviour, resulting in relatively small harems, perhaps leading to females associating with less preferred males, and males being surrounded by fewer females than they could mate with. The aim of this study was to provide the background to future work trying to promote developments to allow for better reproductive success. I finally discussed these results and how they add to the current knowledge of captive-rearing and release, and examine the wider implications of my results from the pheasant rearing system for reintroduction biology. I calculate the likely costs of interventions and extrapolated the potential economic and environmental benefits of implementing changes to the current methods of rearing.

Background: Peripartum cardiomyopathy (PPCM) is an incompletely understood disease that causes significant morbidity and mortality in many parts of the world, including Northern Nigeria. The aims of this Thesis were: [1] to determine if selenium deficiency, serum ceruloplasmin and traditional birth practices are risk factors for PPCM, in Kano, Nigeria; [2] to describe the one year survival and left ventricular reverse remodeling (LVRR) in a group of patients with PPCM from three referral hospitals in Kano, Nigeria; [3] to identify potential electrocardiographic (ECG) predictors of PPCM; and [4] to assess right ventricular systolic dysfunction (RVSD) and remodelling in a cohort of PPCM patients in Kano, Nigeria. Materials and Methods: The studies were carried out in 3 referral hospitals in Kano, Nigeria. Study 1: This was a case-control study. Critically low serum selenium concentration was defined as &lt;70μg/L. Study 2: This was a longitudinal study. LVRR was defined as absolute increase in LV ejection fraction (LVEF) by ≥10.0% and decrease in LV end-diastolic dimension indexed to body surface area (LVEDDi) ≤33.0 mm/m2, while recovered LV systolic function as LVEF ≥55%, at 12 months follow-up. Study 3: This was a case-control study. Logistic regression models and a risk score were developed to determine ECG predictors of PPCM. Study 4: This was a longitudinal study and patients were followed up for 12 months. RVSD was defined as the presence of either tricuspid annular plane systolic excursion (TAPSE) &lt;16mm or peak systolic wave (S’) tissue Doppler velocity of RV free wall &lt;10cm/s. Recovery of RV systolic function was defined as an improvement of reduced TAPSE to ≥16mm or S’ to ≥10cm/s, without falling to reduced levels again, during follow-up. Results: Study 1: Total of 39 PPCM patients and 50 controls were consecutively recruited after satisfying the inclusion criteria. Mean serum selenium in patients (61.7±14.9μg/L) was significantly lower than in controls (118.4±45.6μg/L) (p&lt;0.001). The prevalence of serum selenium &lt;70μg/L was significantly higher among patients (76.9%) than controls (22.0%) (p&lt;0.001). The mean ceruloplasmin and prevalence of socio-economic indices, multiparity, pregnancy-induced hypertension, obesity and twin pregnancy were not different between the groups (p&gt;0.05). Logistic regression showed that rural residency significantly increased the odds for serum selenium &lt;70μg/L by 2.773 fold (p=0.037). Study 2: A total of 33 patients were followed-up. Of the 17 survivors at 12 months, 8 patients (47.1%) satisfied the criteria for LVRR, of whom 5 (29.4%) had recovered LV systolic function, but LVRR was not predicted by any variable in the regression models. The prevalence of normal LV diastolic function increased from 11.1% at baseline to 35.3% at twelve months (p=0.02). At one year follow-up, 41.4% of patients had died (two thirds of them within the first 6 months), but mortality wasn’t predicted by any variable including LVRR. Study 3: A total of 54 PPCM and 77 controls were studied. A rise in heart rate by 1 beat/minute increased the odds of PPCM by 6.4% (p=0.001), while presence of ST-T-wave changes increased the odds of PPCM by 12.06 fold (p&lt;0.001). In patients, QRS duration modestly correlated (r=0.4; p&lt;0.003) with LV dimensions and end-systolic volume index (LVESVI), and was responsible for 19.9% of the variability of the latter (R2 = 0.199; p=0.003). A risk score of ≥2 had a sensitivity of 85.2%, specificity of 64.9%, negative predictive value of 86.2% and area under the curve of 83.8% (p&lt;0.0001) for potentially predicting PPCM. Study 4: A total of 45 patients were studied. RV systolic function recovery occurred in a total of 8 patients (8/45; 17.8%), of whom 6 (75.0%) recovered in 6 months after diagnosis. The prevalence of RVSD fell from 71.1% at baseline to 36.4% at 6 months (p=0.007) and 18.8% at one year (p=0.0008 vs baseline; p=0.41 vs 6 month). Although 83.3% of the deceased had RVSD, it didn’t predict mortality in the regression models (p&gt;0.05). Conclusion: These studies have shown that selenium deficiency seems to be a risk factor for PPCM in Kano, Nigeria, related to rural residency. However, serum ceruloplasmin, customary birth practices and some other characteristics were not associated with PPCM in the study area. They have also shown that PPCM patients had modest LVRR but high mortality at one year. In addition, using the ECG risk score could help to streamline the diagnosis of PPCM with significant accuracy, prior to confirmatory investigations in postpartum women. Finally, RVSD and reverse remodelling were common in Nigerians with PPCM, in whom the first 6 months after diagnosis seem to be critical for RV recovery and survival. / Summary