A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data
Hospitalizations in the United States cost almost 2 trillion dollars every year, which is one-third of the annual total healthcare costs. However, it is to be noted that a large percentage of hospital readmissions can potentially be avoided or prevented. At the University of Iowa Hospitals and Clini...
Main Author: | |
---|---|
Other Authors: | |
Format: | Others |
Language: | English |
Published: |
University of Iowa
2018
|
Subjects: | |
Online Access: | https://ir.uiowa.edu/etd/6607 https://ir.uiowa.edu/cgi/viewcontent.cgi?article=8106&context=etd |
id |
ndltd-uiowa.edu-oai-ir.uiowa.edu-etd-8106 |
---|---|
record_format |
oai_dc |
spelling |
ndltd-uiowa.edu-oai-ir.uiowa.edu-etd-81062019-10-13T04:41:31Z A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data Li, Qing Hospitalizations in the United States cost almost 2 trillion dollars every year, which is one-third of the annual total healthcare costs. However, it is to be noted that a large percentage of hospital readmissions can potentially be avoided or prevented. At the University of Iowa Hospitals and Clinics, a nurse-led transitional care team (TCT) intervention is deployed to attempt to prevent unnecessary hospital readmissions. TCT is designed in a way to provide patients with disease self-management, medical education, and clear instructions regarding discharge and hospital revisits. In this study, we use a quasi-randomization type of analysis based on propensity score matching to explore the intervention effect of TCT versus a control group with no preventative care. Previously, researchers chose 30-day and 90-day readmission rates as the outcomes to examine the performance of hospitalization readmissions, but these categorical outcomes have some limitations. By using the time from discharge to admission as an outcome, this dissertation presents a more precise measurement because it is a time-to-event outcome, which allows a patient multiple events. In this recurrent events data analysis setting, we developed a two-stage pseudo likelihood approach to estimation and inference for analyzing the differences in discharge to admission times between TCT and Control. We also extended this method into an “alternating recurrent events” setting, which takes length of stay factors into account. In this scenario, two events -- length of stay at the hospital and readmissions alternating occur. 2018-12-01T08:00:00Z dissertation application/pdf https://ir.uiowa.edu/etd/6607 https://ir.uiowa.edu/cgi/viewcontent.cgi?article=8106&context=etd Copyright © 2018 Qing Li Theses and Dissertations eng University of IowaZamba, Gideon Biostatistics |
collection |
NDLTD |
language |
English |
format |
Others
|
sources |
NDLTD |
topic |
Biostatistics |
spellingShingle |
Biostatistics Li, Qing A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
description |
Hospitalizations in the United States cost almost 2 trillion dollars every year, which is one-third of the annual total healthcare costs. However, it is to be noted that a large percentage of hospital readmissions can potentially be avoided or prevented. At the University of Iowa Hospitals and Clinics, a nurse-led transitional care team (TCT) intervention is deployed to attempt to prevent unnecessary hospital readmissions. TCT is designed in a way to provide patients with disease self-management, medical education, and clear instructions regarding discharge and hospital revisits.
In this study, we use a quasi-randomization type of analysis based on propensity score matching to explore the intervention effect of TCT versus a control group with no preventative care. Previously, researchers chose 30-day and 90-day readmission rates as the outcomes to examine the performance of hospitalization readmissions, but these categorical outcomes have some limitations.
By using the time from discharge to admission as an outcome, this dissertation presents a more precise measurement because it is a time-to-event outcome, which allows a patient multiple events. In this recurrent events data analysis setting, we developed a two-stage pseudo likelihood approach to estimation and inference for analyzing the differences in discharge to admission times between TCT and Control. We also extended this method into an “alternating recurrent events” setting, which takes length of stay factors into account. In this scenario, two events -- length of stay at the hospital and readmissions alternating occur. |
author2 |
Zamba, Gideon |
author_facet |
Zamba, Gideon Li, Qing |
author |
Li, Qing |
author_sort |
Li, Qing |
title |
A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
title_short |
A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
title_full |
A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
title_fullStr |
A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
title_full_unstemmed |
A two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
title_sort |
two-stage pseudo likelihood approach to estimation and inference for alternating recurrent events data |
publisher |
University of Iowa |
publishDate |
2018 |
url |
https://ir.uiowa.edu/etd/6607 https://ir.uiowa.edu/cgi/viewcontent.cgi?article=8106&context=etd |
work_keys_str_mv |
AT liqing atwostagepseudolikelihoodapproachtoestimationandinferenceforalternatingrecurrenteventsdata AT liqing twostagepseudolikelihoodapproachtoestimationandinferenceforalternatingrecurrenteventsdata |
_version_ |
1719264997466439680 |