Date of Award

2019-10-21

Degree Name

PhD Nursing

Dissertation Committee

Mary Barger, PhD, MPH, CNM, Chairperson; Kathy Shadle James, DNSc, APRN, Committee; Pablo Velez, PhD, RN, Committee.

Keywords

Diabetes Nurse Practitioner

Abstract

Background: Medicare spends $17 billion yearly on 30-day readmissions. Hispanic adults have a higher prevalence of diabetes (12.6%) compared to non-Hispanic whites (9.4%). Those with a diagnosis of diabetes have the 17% higher rate (14-23%) for 30-day readmission. Little research has been conducted on Hispanics with diabetes relative to 30-day readmissions. Aims: Among Hispanics with type 2 diabetes: 1) measure the incidence of 30-day readmission by sociodemographic, behavioral and clinical factors;2) identify independent factors associated with 30-day readmission among Hispanic adults with type 2 diabetes accounting for potential covariates; 3) compare the ability of the Hispanic Diabetic Study Model plus ED, to the LACE model. Methods: The overall 30-day readmission rate for the three years of 14.7% (N=5985) and for those in the study cohort of 9.5% (N=3865). The Hispanic-Diabetic model plus ED had ACU of 0.67 with a high specificity of 99.5% and low sensitivity of 5%. However, the PPV (positive predictive value) was 57.4%, much better than 9.5%. Findings: The readmission rate in the cohort declined from 11.8% to 9.5% over time indicate studied hospital has purposefully worked to decrease 30-day readmissions and have been successful among Hispanics with diabetes. Employment status: only 15% were employed, disabled (20%) had an adjusted odd of 2.43, retired (49%, adj 1.68) increased odds of readmission. The number of ED visits prior to admission was the strongest predictor ranging from an adjusted odd of 6.04 for 1 visit to 14.56 for 3 visits. The length of stay OR=1.75 for 4-6 days or an increase of 5% odds for every day in the hospital and smoking marginally increased odds of readmission. Receiving a consult for a home health aide at discharge was an important factor. Surprisingly, the diabetes complications code decreased the odds of 30-day readmission even though HgA1C above 9% did slightly increase odds. The majority were Spanish-speaking (69%), and 70% of all participants were seen by a diabetes nurse practitioner. Implications: Identified key factors of 30-day readmission among Hispanics with diabetes may lead to future targeted interventions effective in reducing readmissions in this population.

Document Type

Dissertation: Open Access

Department

Nursing

Creative Commons License

Creative Commons Attribution-Noncommercial 4.0 License
This work is licensed under a CC BY-NC-SA License.

Download

Share

COinS