Several examples of categorized data from epidemiological studies are analyzed to illustrate that more informative analysis than tests of independence can be performed by fitting models. All of the analyses fit into a unified conceptual framework that can be performed by weighted least squares. The methods presented show how to calculate point estimate of parameters, asymptotic variances, and asymptotically valid chi 2 tests. The examples presented are analysis of relative risks estimated from several 2 x 2 tables, analysis of selected features of life tables, construction of synthetic life tables from cross-sectional studies, and analysis of dose-response curves.
This randomized clinical trial examined the feasibility of low-fat dietary interventions among postmenopausal women of diverse backgrounds. During 1992–1994, 2, 208 women aged 50–79 years, 28% of whom were black and 16% Hispanic, enrolled at clinics in Atlanta, Georgia, Birmingham, Alabama, and Miami, Florida. Intervention/support groups met periodically with a nutritionist to reduce fat intake to 20% of energy and to make other diet modifications. At 6 months postrandomization, the intervention group reduced fat intake from 39.7% of energy at baseline to 26.4%, a reduction of 13.3% of energy, compared with 2.3% among controls. Saturated fatty acid and cholesterol intakes were reduced, but intakes of fruits and vegetables, but not grain products, increased. Similar effects were observed at 12 and 18 months. Black and non-Hispanic white women had similar levels of reduction in fat, but the decrease in Hispanic women was less. Changes did not vary significantly by education. While bias in self-reported intakes may have resulted in somewhat overestimated changes in fat intake, the reported reduction was similar to the approximately 10% of energy decrease found in most trials and suggests that large changes in fat consumption can be attained in diverse study populations and in many subgroups. Am J Epidemiol 1999; 149:1104–12.
Several examples of categorized data from epidemiological studies are analyzed to illustrate that more informative analysis than tests of independence can be performed by fitting models. All of the analyses fit into a unified conceptual framework that can be performed by weighted least squares. The methods presented show how to calculate point estimate of parameters, asymptotic variances, and asymptotically valid chi 2 tests. The examples presented are analysis of relative risks estimated from several 2 x 2 tables, analysis of selected features of life tables, construction of synthetic life tables from cross-sectional studies, and analysis of dose-response curves.
Statistical methods for testing independence in multiway contingency tables which are based on the correspondence between the analysis of factorial experiments and tests of marginal independence are developed. This relationship simplifies the interpretation of analyses of contingency tables and leads to simplified tests for tables in which some of the probabilities are constrained to be zero. The linear models approach is used to calculate smoothed estimates of probabilities.
Twelve parallel group, randomized, double-blind studies of nomifensine's safety and efficacy in the treatment of depressed patients were combined into three pools according to common protocols. This approach permitted evaluation of 1) efficacy results for studies with moderate-sized pools of patients, 2) the degree to which efficacy was generalizable to depressed patients in the general population, and 3) the conditions under which pooled active vs. active (imipramine vs. nomifensine) studies could be regarded as pivotal in support of efficacy. Results showed that nomifensine's superiority over placebo was generalizable to patients with a wide range of characteristics, including age 60 years or older. An appropriate statistical profile of more pronounced nomifensine responders would include patients with a duration of present episode less than 4 months who are acutely depressed, exhibit more severe symptoms, and have been previously hospitalized or treated with other psychotropic medications. A comprehensive assessment and power analysis of the pooled active vs. active studies provided strong evidence for comparability of nomifensine and imipramine.
'%3e%3cpath fill='%23012169' d='M0 0v30h60V0z'/%3e%3cpath stroke='%23fff' stroke-width='6' d='m0 0 60 30m0-30L0 30'/%3e%3cpath stroke='%23C8102E' stroke-width='4' d='m0 0 60 30m0-30L0 30' clip-path='url(%23b)'/%3e%3cpath stroke='%23fff' stroke-width='10' d='M30 0v30M0 15h60'/%3e%3cpath stroke='%23C8102E' stroke-width='6' d='M30 0v30M0 15h60'/%3e%3c/g%3e%3c/svg%3e)