End of Life Renal Disease
The study compares two scenarios; non-compliance and complications within six months for people diagnosed with end-stage renal disease who have been through patient education and those that have not. Further, two age sets are considered, that is, for 20 and 40-year-olds diagnosed with end-stage renal disease. In this regard, age and education are determinants of non-compliance and complications within six months of diagnosis. The study will seek to find out if the variables, age and education (independent variables), can be used to predict cases of non-compliance and complications within the first six months.
The study will utilize quantitative analysis for this research, in which case, the research aims to evaluate the relation between education and age to non-compliance and complications recorded within the first six months of diagnosis. The proposed regression analysis would be used to examine the relationship between education and age to non-compliance and complications within the first six months of diagnosis. If this relationship is clearly represented, it can serve as a forecast. Regression analysis have two central objectives (David, 2017). They are supposed to quantify relationships and describe them using measured values and their graphical representation and provide forecasts and predictions. In this regard, the research will draw conclusions as to whether age and education patients with end of life renal diseases can be used to predict likelihood of non-compliance to treatment guidelines and complications exhibited within the first six months of being diagnosed with the disease.
The data collected are for the three variables of interest; education, age, and non-compliance and complications within the first six months of diagnosis. The study will collect data from hospitals that offer end of life renal diseases treatment services to various clients. Based on the records, the age of patients, either 20 or 40 years will be recorded. For each patient, the study will collect information as to whether or not the patient was subjected to patient education on management of end of life renal diseases upon diagnosis. This will also be recorded as either yes or no, denoted as 0 for no and 1 for yes. Correspondingly, the research will record the number of complications and cases of non-compliance recorded among the targeted patients within the first six months of diagnosis. The study will target a sample of 30 patients. Notably the study emphasizes the use of only documented data available in the clinic records for the respective patients in order to avoid cases whereby patients might give false information if directly interviewed. As a result, clinic records are the sole source of data used in the study. For purposes of confidentiality the data will be collected anonymously with no reference to patient name.
The study will use regression analysis to analyze how the two variables, age and education influence non-compliance and complications within the first six months of diagnosis. Regression analysis is a statistical method of modeling relationships between different variables (dependent and independent) (Waegeman, 2015). It is used to describe and analyze relationships between data. Regression analysis allows predictions to be made, the relationships between the data being used as a basis for predicting and designing a prediction model. A regression is based on the idea that a dependent variable, in this case, non-compliance and complications developed within first six months, is determined by one or more independent variables, in this case, age and education (David, 2017). Assuming that there is a causal relationship between the two variables, the value of the independent variable affects the value of the dependent variable. In order to study the developments and trends of variables, the data situation must be as complete and accurate as possible (Good & Hardin, 2016). Approximate calculations and plausibility checks are performed to verify the data. If records are missing, missing data techniques can be used, which is also known as imputation in the field of statistics (Meade & Islam, 2018). If the data and their relations are to be displayed graphically, this can be taken into account during the preparation. Some regression models require very special data formats, in which they must first be converted (Meade & Islam, 2018). This is the case, for example, of a linear regression, in which a linear relationship between variables is assumed.
Adaptation of the Model
Each regression model works with statistical error corrections in order to act with possible deviations. Functions that reduce deviations are sometimes determined by the model (Meade & Islam, 2018). Thus, a linear function is used in a linear regression to deal with deviations. Error values and approximations are calculated and integrated into the regression model.
Validation of the Model Used
Here, the study will examine whether the regression model describes the relationship between independent and dependent variables (Meade & Islam, 2018). Additionally, it will consider the extent to which the independent variables, that is, age and education, helps to determine non-compliance and complications experienced within the first six months of diagnosis.
David A. F. (2017). Statistical Models: Theory and Practice. Cambridge University Press.
Good, P. I. & Hardin, J. W. (2016). Common Errors in Statistics. Hoboken, New Jersey: Wiley. p. 211.
Meade, N. & Islam, T. (2018). Prediction intervals for growth curve forecasts. Journal of Forecasting. 14 (5), pp. 413–430
Waegeman, W. (2015). ROC analysis in ordinal regression learning. Pattern Recognition Letters, 29, pp. 1–9.