Factors Influencing Length of Stay in 2007 National Hospital Discharge Survey of Individuals Diagnosed with Type II Diabetes

Abstract

Type II diabetes has become a topic of overwhelming interest in our society.  Using the National Hospital Discharge Survey (NHDS) from 2007, we examine diabetes in relation to hospital care.  The NHDS includes a sample of persons discharged with the diagnosis of diabetes.  The sample size is approximately 36,000 persons.  We filter the sample from the NHDS aggregate with the ICD-9-CM codes: 250.00 and/or 250.02.  Our study focuses on the length of days of care in the hospital as our dependent variable and in relation to our predictive variables: sex, age, marital status, method of payment and type of admission.  We examine this data by conducting descriptive, explanatory and predictive analyses.  The latter set of analyses is by linear and by logistic regression.  Our hypotheses are that women, persons who are older, persons admitted by emergency, and persons for whom the payment is public are more likely to have a longer stay in the hospital than those who do not fit these criteria.  We found, for example, that, within 95% confidence intervals, women who are single, older, admitted as emergent, and for whom payment is public are more likely to have a longer length of stay than men with similar characteristics.  The significance of this study is that it enables hospitals to better prepare for the care of individuals admitted with the diagnosis of type II diabetes.