KIDNEY FAILURE (KF) MODEL: WHY ARE KF CASES GETTING YOUNGER?

Abstract

The intent of determining the occurrence of renal failure among the younger population prodded the researchers to develop a predictive model that will aid healthcare providers. The purpose of this paper is to analyze the characteristics of the younger population of the present in relation to its increasing vulnerability in developing kidney failure such as, intake of power/energy drink (PD), body mass index (BMI), family history (FH) and presence of other pre-existing condition (PC). In establishing the extent by which these identified independent variables really contributed to the development of KF among the young ones, we did a simulation run with 1000 generated random data in the mini tab version 15 based on real behaviors. We formulated and tested the first linear model: Y₁ = _ß0 + _ß1PD + _ß2BMI + _ß3PC + _ß4FH and the results showed that RF (Y₁) = 28.9 - 4.02 PD - 0.0028 PC + 0.0459 FH. Notice that BMI has been removed from the equation because it is highly correlated with other independent variables. At this point, we could not say yet that this is a good model. So, we noted that the sum of square residual error (SSE) is 251. Knowing that the higher is the SSE, the larger is the degree of error of the theory and thus, there is a need for us to generate a new, better model. At this point, we constructed a reciprocal model: Y₂ = ?₀ + ?₁(1/PD) + ?₂(1/BMI) + ?₃PC + ?₄FH. Consistent with the result of the first trial, BMI is still omitted because it is highly correlated with other independent variables. As you can see, of the three (3) remaining variables, only PD was found to be a determinant of KF development i.e., KF = 50/PD. This means that our unknown is 50 and so to illustrate, if a teenager consumes 5 bottles of PD per day, then he can develop renal failure at the age of 10 (50/5). We can say that this is a better model compared with the first one because the SSE is 0. This implies that the model is very accurate. This result illustrates the real behavior because the younger population before did not consume a lot of PD while the rest of the identified independent variables already existed even then. As the proponents were just utilizing a hypothetical data, we are interested in applying the generated model in actual data. After analyzing the gathered data, the intake of power drinks was the only determinant of kidney failure development.