Basic epidemiologic and statistical methods in clinical research.

M Gottlieb; G Anderson; H Lepor

Home/Archive/PubMed

PubMed · 1441022

Basic epidemiologic and statistical methods in clinical research.

M Gottlieb·G Anderson·H Lepor

View original source ↗Export metadata ↓

Abstract

Various study designs and approaches to statistical analysis in clinical research have their own underlying rationales and limitations for interpretation. This information is presented in an intuitive and accessible manner, relying minimally on basic algebra. Epidemiologic concepts of study design and interpretation, bias and confounding, hypothesis testing, and sample size and power are explained. Statistical tests and their appropriate applications are discussed for mean comparisons (t tests and ANOVA), percentages (chi-square), survival analysis, and correlation and regression. Applicable nonparametric tests are also introduced.

Explore related subjects

Carcinoma, Transitional Cell Epidemiologic Methods Humans Research Design Statistics as Topic Urinary Bladder Neoplasms Journal Article Review

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

M Gottlieb, G Anderson, H Lepor. 1992. Basic epidemiologic and statistical methods in clinical research.. https://pubmed.ncbi.nlm.nih.gov/1441022/

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed1992

Structural chromosome 1 aberrations in transitional cell carcinoma of the bladder: interphase cytogenetics combining a centromeric, telomeric, and library DNA probe.

Fluorescence in situ hybridization (FISH) was used to study numerical and structural chromosome 1 aberrations in interphase nuclei of transitional cell carcinomas (TCCs) of the urinary bladder. One of the characteristic numerical aberrations, as detected previously in low-grade noninvasive TCCs, included trisomy for chromosome 1 (A. H. N. Hopman et al., Cancer Res., 51: 644-651, 1991). We examined in more detail 22 cases with a centromeric (1q12) and a telomeric associated (1p36) DNA probe and with a library DNA probe from sorted human chromosome 1 in single- and double-target FISH procedures. All flow cytometrically determined DNA diploid TCCs (13 cases), which showed three spots for 1q12 (6 cases), had two spots for 1p36. Since the library DNA probe showed three separate domains in the nuclei of these cases, the additional copy for 1q12 could be explained as an extra chromosome 1p-, containing the 1q12 target. In the flow cytometrically determined DNA tetraploid/aneuploid tumors, the results were more complex. In 6 of 9 cases, we observed an overrepresentation of 1q12 as compared to 1p36, also suggesting the presence of extra copies of 1p- chromosomes. The results of the present study demonstrate the utility of the FISH method to assess structural chromosome aberrations in interphase nuclei of solid tumors.

Carcinoma, Transitional Cell↗

PubMed1992

Independent clinical, histological and quantitative prognostic factors in transitional-cell bladder tumours, with special reference to mitotic frequency.

A cohort of 537 transitional-cell bladder cancers (TCC) was followed up for a mean of 9 years. Clinical stage, WHO grade, papillary status, 6 nuclear factors and volume-corrected mitotic index (M/V index) were related to progression and survival. Classic and quantitative prognostic factors were significantly interrelated (p less than 0.001). In Ta-Tl tumours M/V index predicted progression independently (p less than 0.001) and in the entire cohort progression was related independently to the M/V index (p = 0.0001) and to the WHO grade (p = 0.0022). In survival analysis, clinical stage (p less than 0.0001), M/V index (p less than 0.0001), WHO grade (p less than 0.0001), papillary status (p less than 0.0001) and nuclear factors (p less than 0.0001) were significant predictors. In papillary tumours, clinical stage (p less than 0.0001), M/V index (p less than 0.0001), WHO grade (p less than 0.0001) and nuclear factors (p = 0.0001-0.0133) were related to survival. In a multivariate analysis T-category (p less than 0.001), WHO grade (p less than 0.001), M/V index (p = 0.002) and papillary status (p = 0.034) predicted survival independently in the entire cohort whereas in papillary tumours T-category (p less than 0.001) and M/V index (p less than 0.001) were independent predictors. If tumours with pelvic lymph-node metastases or distant metastases at diagnosis were excluded from the analysis, T-category (p less than 0.001), M/V index (p less than 0.001) and WHO grade (p less than 0.001) were independent predictors. In papillary tumours T-category (p less than 0.001), M/V index (p less than 0.001) and WHO grade (p = 0.048) predicted survival. The results emphasize the importance of mitotic activity as a most important histological prognostic factor in TCC, second only to clinical stage. In Ta-TI tumours quantitative mitotic frequency analysis includes all the available independent prognostic information. Accordingly, TCC can be graded by mitotic frequency analysis in place of subjective grading systems.

Carcinoma, Transitional Cell↗

PubMed1992

Renal pelvic transitional cell carcinoma. The role of the kidney in tumor-node-metastasis staging.

Renal pelvic carcinoma is a rare tumor. There are several staging systems currently in use based on a bladder cancer staging model. An analysis of the American Cancer Society, Illinois Division, experience with 611 cases of renal pelvic transitional cell carcinoma suggested that the kidney parenchyma may be a determinant in the anatomic spread of disease. A tumor-node-metastasis system for renal pelvic (separate from ureteral) cancer is recommended.

Carcinoma, Transitional Cell↗