This thesis reports the results of a prospective multicentred case-control study of mesothelioma carried out in South Africa. The objectives of the study were: 1) to examine asbestos exposure of cases in detail with respect to source, risk occupations, fibre type and duration; 2) to determine relative risks for level (certainty) of exposure (definite, probable, possible, unlikely), for category of exposure (occupational, environmental), and for fibre type and skin colour; 3) to determine whether cases without recall of exposure were exposed to other non-asbestos putative agents; 4) to investigate the possible protective effect of certain dietary components. Previous studies of mesothelioma in South Africa had, with the exception of one incidence study, focused on particular occupational or case material, exposure data had been gathered in a non-systematic way, often indirectly from surrogates, and non-asbestos agents had not been investigated. In this case-control study these issues are all addressed. In addition, special efforts were made to minimise potential sources of bias (e.g. interviewer bias) and so to furnish reliable effect estimates. The study incorporated the following methodological features: 1) a prospective approach to gather exposure and dietary information directly from the cases and controls in life and so avoid the use of surrogates for this information; 2) the study was multicentred with study teams established in six cities, each with a major referral hospital, to maximise nation-wide coverage; 3) information was gathered with interviewers blind (at least at the beginning of the study) to study objectives and case control status at the time of the interview; 4) rigorous pathologic review was used to establish the diagnosis of mesothelioma; 5) two controls were selected for each case, a cancer and a non-cancer patient matched for hospital, sex, age and skin colour; 6) in analysis the case control datasets were treated separately (i.e cases and cancer controls, and cases and non-cancer controls were treated as two separate datasets). One hundred and twenty three cases were accepted into the study. No case was documented with purely chrysotile exposure nor exposure to a putative non-asbestos cause of the tumour without some evidence of asbestos exposure. A minimum of 22 cases (18%) had exclusively environmental exposure, 20 were from the NW Cape (a crocidolite mining region). Fifty eight percent had occupational exposure, three of whom had mined amosite. The relative risks associated environmental exposure in the NW Cape were larger than for environmental exposure in the NE Transvaal: 21.9 versus 7.1 for the cancer control dataset and 50.9 versus 12.0 for the medical control dataset. Increasing consumption of carotene rich fruit was found to be protective for mesothelioma when adjusted for asbestos exposure. The results confirm the high disease burden due to occupational exposure, the importance of environmental exposure in the crocidolite mining area of the NW Cape, the relative paucity of cases linked to amosite, the rarity of chrysotile cases, and are consistent with the view that there is a fibre gradient in mesotheliomagenic potential for South African asbestos with crocidolite > amosite > chrysotile. The evidence for a protective effect of carotene rich fruit is new in the South African context.

Thesis (Ph. D.)--University of Washington, 1995. / Vita. Includes bibliographical references (leaves [175]-192).

It is widely believed that risks of many complex diseases are determined by genetic susceptibilities, including environmental exposures, and their interaction. Chatterjee and Carroll (2005) have recently developed an efficient retrospective maximum-likelihood method for analysis of case-control studies that exploits an assumption of gene-environment independence and leaves the distribution of the environmental covariates to be completely nonparametric. We generalize the semiparametric maximum-likelihood approach to situations when some of the environmental covariates are measured with error and allow genetic information to be missing on some subjects, e.g., unphased haplotypes. Profile likelihood techniques and an EM algorithm are developed, resulting in a relatively simple procedure for parameter estimation. We prove consistency and derive the resulting asymptotic covariance matrix of parameter estimates when variance of measurement error is known and when it is estimated using replications. The performance of the proposed method is illustrated using simulation studies emphasizing the case when genetic information is in the form of a haplotype and missing data arises from haplotype-phase ambiguity and missing genetic data. Inference is performed via a likelihood-ratio type procedure, one that we show has better small-sample performance thanWald-type inferences. An application of this method is illustrated using a case-control study of an association of calcium intake with early stages of colorectal tumor development.

Intravascular brachytherapy is the primary treatment for coronary in-stent restenosis. We hypothesized that differences in dose delivered to target may contribute to treatment failures. We compared dose distribution between arteries that developed recurrent restenosis (treatment failures) and those that remained patent at nine-months (treatment success). A cohort of 207 patients receiving brachytherapy for coronary in-stent restenosis with four radiation delivery devices was followed to identify treatment failures and successes. This cohort was examined to establish which patient and lesion characteristics had an effect on outcome. A nested case-control construct was then used in which treatment failures (n=14) were compared 1:2 to treatment successes (n=28) matched by two variables: radiation delivery system and angiographic pattern of in-stent restenosis. At baseline, the groups had similar patient and lesion characteristics. The dose absorbed by 90% of the artery encompassed by the external elastic membrane (D90EEM) was calculated using intravascular ultrasound (IVUS) images taken at 2-mm intervals along the treated lesion. Dose calculations were performed using dose kernel integration techniques; the dose kernels were generated from Monte Carlo simulations. The mean D90EEM minimum dose in treatment failures was 7.46±1.98 Gy, while for treatment success the mean D90EEM minimum dose was significantly higher: 8.87±1.13 Gy (p=0.007). Using a dose threshold of 8.4 Gy, a D90EEM minimum dose < 8.4 Gy occurred in 13 (93%) patients with treatment failure, but in only 9 (32%) with treatment success (p<0.001). No confounding variables were found to be statistically significant. In conclusion, current brachytherapy dose prescriptions result in significant inter- and intra-lesion variation in dose at the EEM. Arteries that receive < 8.4 Gy at any point along the EEM are more likely to be treatment failures. IVUS guided dosimetry may be critical to assure adequate dose regardless of radiation delivery system.

Thesis (Ph. D.)--University of Washington, 1998. / Vita. Includes bibliographical references (leaves [84]-91).

Thesis (Ph. D.)--University of Washington, 2000. / Vita. Includes bibliographical references (p. 137-144).

El texto completo de este trabajo no está disponible en el Repositorio Académico UPC por restricciones de la casa editorial donde ha sido publicado. / This letter has the purpose to comment the article by Marwaha et al. regarding affective instability and mental disorders. We wish to highlight the importance to report the proper measures of association in case-control studies, and the impact of adjusting the results when finding associations with possible confounders in the bivariate analysis. / Revisión por pares / Revisión por pares

Thesis (Ph. D.)--University of Washington, 2000. / Vita. Includes bibliographical references (leaves 33-40).

Thesis (Ph. D.)--University of Washington, 1990. / Vita. Includes bibliographical references (leaves [78]-89).

Thesis (Ph. D.)--University of Washington, 1997. / Vita. Includes bibliographical references (p. [32]-37).