Work

Logistic regression in medical decision making and epidemiology

Year: 1986
Type: dissertation
Abstract: In his recent textbook "Primer of Biostatistics", S, A, Glantz refers to the nowadays\ngrowing pressure on clinicians for more effective use of medical resources. He asserts that\nclinicians should be able to make better informed judgements about claims of medical\nefficacy. They can participate then more intelligently in the debate on how to allocate medical\nresources. These better informed judgements are the objective of "medical decision\nmaking", where the choices to be made in diagnostic and therapeutic strategies are studied\nMedical decision making is to be based for a great part on statistical reasoning. A statistical\napproach which has proven to be valuable in this field is a technique known as logistic\nregression analysis. The assessment of the performance of this logistic model is the subject of\nthis thesis. Applications in medical decision making are considered, mainly with respect to\nmedical diagnosis. Logistic regression analysis may be viewed as a sophisticated diagnostic\naid. Multiple test outcomes and patient characteristics are incorporated into a logistic model\nfor the probability that a patient belongs to a certain disease class.\nPerformance of the logistic model in etiologic-epidemiological studies is another area of\nstudy. In this context logistic regression is used for the detection of risk factors, ad jus ted for\nconfounding in order to obtain unbiased assessments.\nThis thesis includes ten chapters, consisting of papers which have either been published or\nwhich were recently submitted for publication by the author, many of them in cooperation\nwith various colleagues. The chapters have been grouped into three parts. Part 1 (chapter 1)\npresents a review of developments in logistic regression from 1970 up to 1986. It outlines\nstatistical aspects of the model such as estimation. hypothesis testing and model selection.\nPart 2 (chapters 2 - 7) deals with logistic discriminant analysis in medical diagnosis. An\nextensive evaluation of a logistic model for the diagnosis of Crohn's disease by agglutination\nreactions is presented in chapter 2. Actually, this chapter results from our evolving insights\nsince the first application of logistic discriminant analysis for the diagnosis of Crohn's disease\n(chapter 3). Comparison of logistic discrimination with some other discriminant analysis\nmethods is studied in chapter 4 through application to real data from clinical practice, and in\nchapters 5 and 6 through the use of simulated data. This comparative evaluation was\nperformed on datasets consisting of mixtures of continuous and discrete data. The underlying\ndistribution in the first simulation study (chapter 5) is a fourdimensional normal from which\ndiscrete variables were obtained by discretizing the continuous variables. The simulation\nstudy in chapter 6 is based on a location model. Within each outcome combination of the\ndiscrete variables a multivariate normal distribution is assumed. The next chapter concerns\nthe evaluation oflogistic discriminant analysis for modelling QSAR's, quantitative structure- activity relationships (chapter 7). It is used there as a technique for detecting which chemical\ncompounds will be useful for the development of new drugs. The third and final part of this\nthesis (chapters 8- 10) is concerned with the application and evaluation of the logistic model\nin etiologic-epidemiological studies, particularly in case-control studies. In chapter 8 the first\nepidemiologic application of the logistic model in TheN etherlands, which uses estimates of\nthe model parameters based on conditional likelihood, is presented. Risk factors for stroke\nare investigated in a case-control study conducted some years ago in Tilburg. In chapter 9 the\n(multiplicative) logistic model is compared with an (additive) linear model for case-control\nstudies with one continuous exposure factor and without consideration of confounding\nvariables. Some aspects concerning variables selection in epidemiologic studies, also relevant\nfor logistic regression, are discussed in chapter 10. The thesis is concluded with a summary.\n12
Author: Paul Schmitz
Language: English
Open Access status: green
OpenAlex ID: W2127076952
Cites: 139
Cited by: 8
Related to: 10