Information Technology Reference
In-Depth Information
black-box warning contraindicating its use in any patients with any kind of heart
failure. Using over 5,892 patient years of data, they observed no association between
Cilostazol use and major adverse cardiovascular events. The fi ndings recapitulate a
prospective clinical study, which had less than half the patient years of data and
failed to reach a statistically signifi cant conclusion. In addition, the data-mining
study was able to profi le a subset of patients with CHF who were prescribed
Cilostazol despite its black box warning and examine its safety in this high-risk
group of patients—something that could not be done prospectively, given the black-
box warning on the drug.
Analyses using “Big Data” go beyond learning biomedical insights as outlined in
the examples above. The personalization of cancer chemotherapy by examining the
genomic variations that drive specifi c tumor subtypes and the response of those
subtypes to specifi c chemotherapy drugs is already being used to personalize treat-
ment of breast cancer [
16
]
7
and is likely to be among the fi rst clinical success stories
about the application of data science to identify alternative treatment strategies.
As discussed in the chapter on personalized medicine, consider the example of
personalizing treatment based on genomic sequencing, where Dr. Howard Jacob's
team pinpointed a new casual mutation for the treatment of a 6 year-old boy with an
extreme form of infl ammatory bowel disease [
17
,
18
]. The authors diagnosed an
X-linked inhibitor of apoptosis defi ciency, based on which, they decided to perform
an allogeneic hematopoietic progenitor cell transplant. This case report demon-
strates the power of using genomic data to arrive at a molecular diagnosis in an
individual patient in the setting of a novel disease.
On the operational side—which is related to the practice and delivery of health-
care—the use of Big Data has gained a lot of momentum. Predictive-analytic
approaches, such as those designed to predict readmissions [
19
], are increasingly
gaining traction
8
and have direct implications for reducing cost, as well as maintain-
ing economic viability of health care delivery systems in the light of new regulation.
The main drivers for using Big Data on the operational side are [
20
]:
The Patient Protection and Affordable Care Act, and the creation of accountable
care organizations (ACOs), which require that healthcare systems have a higher
degree of “business intelligence”.
Ineffi ciencies, fraud, and waste; where big data can play a crucial role in perfor-
mance improvements.
Adoption of open-data policies by the U.S. Department of Health and Human
Services, which spark innovation and increase transparency in health care.
As more data are made available, and health systems are increasingly under pres-
sure to provide the same or better quality care at less cost, it is natural to use data to
increase effi ciency, design less costly care workfl ows, increase intervention
8
http://www.beckershospitalreview.com/healthcare-information-technology/4-steps-to-
Search WWH ::
Custom Search