Evaluation Of Healthcare Utilization In Patients With Metabolic Syndrome

Mohit Arora

marora1@umbc.edu

IS 804 вЂ" Final Project Report

Evaluation of Healthcare Utilization in Patients with

Metabolic Syndrome

Abstract

Metabolic Syndrome is a cluster of cardio-metabolic risk factors, including obesity, hyperglycemia, dyslipidaemia and hypertension, and has been linked with elevated risk of developing Cardiovascular Disease and Type 2 Diabetes. It is estimated that approximately 39% of the US adult population meets the criteria for Metabolic Syndrome. The aim of this project was to evaluate the role of positive diagnosis of Metabolic Syndrome on Healthcare Utilization вЂ" specifically on the annual number of hospitalizations and doctor’s office visits. Data on 16,632 subjects was extracted from the Third National Health and Nutrition Examination Survey dataset and used for analysis. It was concluded from Multivariate Multiple Regression model that the number of Hospitalizations and Doctors Office visits is significantly higher in subjects with Metabolic Syndrome accounting for differences in age, race and gender. It was also seen that the variables selected for analysis accounted for a very small percentage of the variance in number of hospitalizations and doctor’s office visits. It was hence concluded that further work was required to evaluate the influence of Metabolic Syndrome on healthcare utilization while accounting for these unknown factors.

1. Introduction

Metabolic Syndrome (MetS) is a cluster of cardio-metabolic risk factors that has shown to significantly increase the risk of developing Cardiovascular Disease (CVD) and Type 2 Diabetes Mellitus (T2DM). Risk factors include abdominal obesity, hyperglycemia (elevated blood glucose), dyslipidaemia (abnormal blood lipid levels) and hypertension (elevated blood pressure). It is estimated that 2500 Americans die from CVD each day with costs related to CVD approximating to $403 billion annually. Studies have shown that having MetS doubles the risk of developing CVD and increases the risk of developing T2DM by a factor of seven. A study in 2005 estimated that approximately 39% of the US adult population meets the criteria for Metabolic Syndrome.

The aim for this project was to compare Healthcare Utilization trends in patients with and without MetS. Healthcare utilization is defined as the level of usage of medical services and can be measured using variables such as Number of Hospitalizations, Number of Emergency Room Visits, Number of Doctors Office Visits, etc. Since MetS is associated with several co-morbidities, it is conceivable that it can lead to higher healthcare utilization.

2. Methods

2.1 Data Source

Data for the project was extracted from a dataset called the Third National Health and Nutrition Examination Survey (NHANES III). This survey was conducted by the National Center for Health Statistics between 1988 and 1994. It contains cross-sectional interview and laboratory examination data from a nationally representative sample of 33,944 US residents aged 2 months and older. This dataset is publicly available and can be downloaded from the NHANES website.

The NHANES III dataset contains two variables that can be used to measure Healthcare Utilization

1. Number of Hospitalizations in the last 12 months

2. Number of Doctors Office Visits in the last 12 months

In addition, data on demographic characteristics (Age, Race and Gender) was extracted from this dataset. Only data for adult subjects (ages 17 an over) was extracted. Based on the criteria provided by the International Diabetes Federation , a variable MetS was created to classify each patient with a positive or negative diagnosis for Metabolic Syndrome. In cases, where data for classification of MetS was missing, the subject data was excluded from the final dataset. The final dataset included data from 16,633 patients. Figure 1 shows a representation of the final dataset.

Figure 1 вЂ" Structure of Final Dataset

2.2 Hypotheses

Based on the concept of Healthcare Utilization and availability of data the following hypotheses were tested:

Null Hypothesis 1 (N10)

The number of hospitalizations for patients with Metabolic Syndrome is not higher than number of hospitalizations for patients without Metabolic Syndrome.

Alternative Hypothesis 1 (N1A)

The number of hospitalizations for patients with Metabolic Syndrome is higher than number of hospitalizations for patients without Metabolic Syndrome.

Null Hypothesis 2 (N20)

The number of doctors office visits for patients with Metabolic Syndrome is not higher than number of doctors office visits for patients without Metabolic Syndrome.

Alternative Hypothesis 2 (N2A)

The number of doctors office visits for patients with Metabolic Syndrome is higher than number of doctors office visits for patients without Metabolic Syndrome.

2.3 Analysis Plan

The variables extracted for analysis are summarized in table 1.

Variable Description

SEQN Unique Identifier

Age Age in years

Race 0=White 1=Black 2=Hispanic 3=Other

Gender 0=Male 2=Female

MetS Metabolic Syndrome Diagnosis 0=Negative 1=Positive

Hosp Number of Hospitalizations

Doc Number of Doctors Office Visits

Table 1 вЂ" Variables extracted from the NHANES III dataset

In an attempt to understand the data, the first step was to conduct a Univariate test for the continuous variables Hosp, Doc and Age and obtain Frequencies for the categorical Race and Gender.

The second