Ingram Olkin’s research while affiliated with Stanford University and other places

Background and objectives: Given the increased number of treatment options for stage IA lung cancer patients, there is a growing body of literature that focuses on comparing each option's relative impact on quality of life (QoL). The current study seeks to further understand the differences in these patients' QoL according to surgical approach. Methods: Screening-diagnosed first primary pathologic stage IA non-small-cell lung cancer surgical patients from the I-ELCAP cohort who answered a baseline and 1-year follow-up QoL questionnaire (SF-12) were included in the analysis. Thoracotomy patients (N = 85) were compared with VATS patients (N = 15) using paired t-tests and analysis of variance tests. Results: Multivariate analyses indicated no differences in QoL change between the two groups from pre- to post-surgery. Physical and emotional role functioning significantly improved among VATS patients and worsened among thoracotomy patients. Among thoracotomy patients, a significant decrease in post-surgical physical QoL was observed only in those who underwent lobectomy (-3.3; 95% CI: -5.1,-1.5), not limited resection. Conclusions: Although the sample size is small, preliminary findings underscore that changes in overall QoL are similar in VATS and thoracotomy stage IA lung cancer patients. Extension of the resection may be a more relevant factor on QoL post-surgery. J. Surg. Oncol. © 2016 Wiley Periodicals, Inc.

This is a comprehensive review and re-analysis of available literature to assess the outcome of lung cancer presenting as nonsolid nodules (NSNs), a more indolent form of cancer. PubMed and EMBASE were searched for articles reporting on CT-detected lung cancers manifesting as NSNs published in English on or before July 17, 2015. Only studies including clinicopathologic data, lung cancer-specific survival, or overall survival were included. Data extraction was performed by three independent reviewers using prespecified criteria. Twenty-four articles from 5 countries met criteria and they included 704 subjects with 712 lung cancers manifesting as NSNs. Each article reported from 2 to 100 lung cancer cases with a median follow up of 18-51 months. All NSNs were Stage I adenocarcinoma without pathologic nodal involvement upon resection, except for one case in which the NSN progressed to become part-solid nodule after 6 years of follow-up. The five-year lung cancer-specific survival rate was 100%. These findings suggest an indolent course for lung cancers manifesting as NSNs.

BACKGROUND: There is a paucity of literature comparing quality of life (QoL) before and after surgery in stage IA lung cancer, where surgical resection is the recommended curative treatment. OBJECTIVE: To assess the impact of surgery on physical and mental health-related QoL in patients with stage IA lung cancer treated with surgical resection. METHODS: Participants in the I-ELCAP cohort who were diagnosed with their first primary pathologic stage IA non-small-cell lung cancer, underwent surgery, and provided follow-up information on QoL 1 year later were included in the present analysis (N = 107). QoL information was collected using the SF-12 (12-item Short Form Health Survey), which generates 2 component scores related to mental health and physical health. RESULTS: Statistical analyses indicated that physical health QoL was significantly worsened from before surgery to after surgery, whereas mental health QoL marginally improved from before to after surgery. Physical health QoL worsened for women from baseline to follow-up, but not for men. Only lobectomy (not limited resection) had an impact on QoL from before to after surgery. LIMITATIONS: Results are considered preliminary given the small sample size and multiple comparisons. CONCLUSIONS: The current study findings have implications for lung cancer health care professionals in regard to how they can most effectively present the possible impact of surgery on quality of life to this subset of patients in which disease has not yet significantly progressed.

Introduction: Surgery is the treatment of choice for early-stage lung cancer (LC). Although lobectomy (L) is the historic standard treatment, the issue of whether long-term outcomes of sublobar resection (SL) are comparable is still under debate. The objective of this study was to perform a review of the literature on 5-year survival rates after SL compared to L for patients with early-stage LC. Methods: A priori inclusion criteria were as follows: (1) observational studies, (2) L compared to SL for early-stage LC, (3) radiographic staging by computed tomography scan, and (4) 5-year survival reported. A Medline search through January 2015 resulted in 31 studies representing 23 distinct datasets. The absolute difference in 5-year survival was calculated and plotted for each study. Results: L was performed in 4564 patients and SL in 2287 patients. Of 19 studies reporting the reason for SL, 11 indicated that SL was performed because of comorbidities or impaired cardiopulmonary function. Four studies showed no difference in 5-year survival, 13 favored L, and six favored SL. One propensity score study favored L and the other favored SL. Of 20 studies reporting recurrence rate, 11 favored L and nine favored SL. Conclusions: Studies comparing 5-year survival rates of SL to L are sufficiently heterogeneous to prevent carrying out traditional meta-analysis. SL survival is often similar to L when adjustments are made for age, comorbidities, or impaired cardiopulmonary function. New approaches are needed for the comparison of L to SL.

This article presents a brief description of some of the characteristics of life distributions that arise in survival analysis and reliability theory. Alternative definitions of a distribution are discussed and related to a variety of stochastic orders: hazard rate order, likelihood ratio order, convex order, Lorenz order. Nonparametric families, particularly log-concave densities, completely monotone distributions, increasing hazard rate families, new-better-than-used families, and bathtub hazard rate families are analyzed. A taxonomy for semiparametric families is presented and the effect of introducing parameters on various stochastic orders is shown. Finally, the introduction of covariate models in these families is developed.

In my talk at the LAA meeting in honor of Hans Schneider, I gave a brief biography of my introduction to linear algebra and my interaction with some of the linear algebraists at that time. It was suggested that I write this up, and thus the following.

The beta distribution is a basic distribution serving several purposes. It is used to model data, and also, as a more flexible version of the uniform distribution, it serves as a prior distribution for a binomial probability. The bivariate beta distribution plays a similar role for two probabilities that have a bivariate binomial distribution. We provide a new multivariate distribution with beta marginal distributions, positive probability over the unit square, and correlations over the full range. We discuss its extension to three or more dimensions.