Epidemiology: Risk Factors for Chronic Disease

There are various measures of association applicable for this kind of study. With weighted reasons and definite conclusions drawn, an epidemiological measure of association is the suitable method. Various aspects of the epidemiological measure of association make it more appropriate for this study (Stewart, 2007). This is because, in this particular study, the objectives of the study match those of epidemiological measures of association.

Perse, the study tasks learners (or researchers) to identify changes or variations of relative risk against odds ratio as they occur. In addition, this study requires learners to compute the odds ratio between the exposure (agent causing disease) and relative risk. According to research, these are also the basic objectives of epidemiologic as a measure of association. With such considerations, an epidemiologic measure of association proves to be a suitable solution for carrying out the study. The nature of data required would necessitate acquisition through an epidemiologic measure of association

The table below shows the measure of association between chronic disease, A 17, B24, smoking, and C32. It also includes 95% percent confidence intervals. Additionally, it shows the results of the measure of association between the chronic disease and factors A17, B24, C24, and smoking at the different exposure levels as required.

Factor Adjust OR 95% CI
Smoking 0.22 0.30-0.63
Factor A17 0.75 0.57-1.30
Factor B24 1.63 0.87-2.19
Factor C32 1.92 1.05-3.59

Considering conclusions drawn from multivariate analysis, results of bivariate and those of multivariate have the description of being in ascending order. It is also important to note that 0.8, 1.05, and 1.98 precede each other in that order. Another evident observation is that 95 % confidence intervals increase systematically in both bivariate and multivariate analysis. Other variants that exhibit this ascending characteristic are those from the bivariate analysis.

The 95%confidence intervals of multivariate analysis and bivariate analysis occur in a systematic ascending order. Furthermore, another feature drawn from the table is the cumulative characteristic of some variants. When there is an analysis of both the exposed and unexposed groups, bivariate and multivariate analysis results exhibit cumulative characteristic (Meyer, 2010). The values increase by a common factor consecutively.

Results obtained from the bivariate analysis are important, as they are applicable in real-world situations. In research institutes, multivariate and bivariate analysis find application in the determination of incidence occurrence of the phenomenon under study or research to both parties that will experience direct and indirect effects (Chevrol, 2003). What is more, they can find the application of health experts in weighing the risks of coronary heart disease to both active and passive smokers.

Coronary heart disease is a heart disorder that has main causes by heavy intake of nicotine and is, therefore, a common health disorder among smokers. Other than the risks of smoking, there can be the usage of these results to determine more threats posed by exposure of non-smokers to those that smoke (Stewart, 2007). To achieve this there is a need of applying odds ratio and cohort data. Cohort data is a type of data that learners (researchers) use for study cases and research purposes making cohort data important especially in our case study. For an instant, a factor often measured using a combination of these two data is coronary heart disease in non-smokers. Using the two data is essential in the prediction of the occurrence of this ailment in non-smokers

Two areas always require prompt use of cohort data. The first is during the determination of the cumulative incidence of the disease (coronary heart disease) in the directly exposed persons (smokers) and those persons that are indirectly exposed (non-smokers). However, for cohort data to function efficiently and achieve accurate results, it must function concurrently with the odds ratio. On the other hand, cohort data has a nature of lacking consistency and the redundancy of having incomplete data.

Hence, although it is an efficient method of researching and solving case studies, it is dependent on other forms of data. The odds ratio in this case acts as a compliment (Chevrol, 2003). As such, it complements the data whenever used by providing the missing items of data. These kinds of complex modes of data combination are not applicable to all forms of analysis. Multivariate analysis is the common type of analysis where such data combination is common.

The tables below show the results of the multivariate and bivariate analysis as calculated. They are also a representation of the measure of association between the disease that is the subject matter and the risk factor at each level of exposure.

Factor Adjust OR 95% CI
Not received 0.08 0.28
Brand Y 0.06 0.13
Brand X-50 mg 0.15 0.25
Brand X-100 mg 0.25 0.18
Brand X-200 mg 0.46 0.16

The major similarity between the bivariate and multivariate analysis tables is that the figures are similar. The variants, however, have no order of arrangement, that is, for instance in both tables the percentage confidence intervals follow no sequence. The values are 0.28, 0.13to o.25.s


Chevrol, M. (2003). Epidemiologic Measure of association. New Jersey: John Wiley & Sons, Inc.

Meyer, R. (2010). Epidemiologic sources of data. Indiana: Pearson Education.

Stewart, J. (2007). Epidemiologic Measure of association. New York: Routledge.