Information Technology Reference
In-Depth Information
recruitment. Some institutions have created research data warehouses based only on
data collected as part of standard care, existing in the EHRs. These research initia-
tives have limited the invasiveness of research recruitment down to just consent to
use the information. Examples include the many current translational informatics
initiatives currently requesting for patients to “donate” their data [ 7 - 9 ].
Even when the data collection is minimized by using existing data from the EHR
and not changing the intervention, there is still a burden to the patient for which
consent is needed. This burden is a privacy and confi dentiality risk. Health care
assessments and therapies are personal and sensitive, and the record of those inter-
actions should not be known outside those providing care, receiving care, or paying
for care. The Health Insurance Portability and Accountability Act (HIPAA) Privacy
Rule and the Health Information Technology for Economic and Clinical (HITECH)
Health Act specifi cally protect patients, by defi ning what data are protected, defi n-
ing rules for the use for the protected health information (PHI), and including penal-
ties for the misuse and loss of PHI. Institutional review boards (IRBs) also protect
patients in natural history studies against loss of privacy. While protection of confi -
dentiality and privacy does not require engagement of patients directly, it is by
intent a very patient-centered component of research.
A method to reduce the privacy and confi dentiality risk is to de-identify data by
removing identifying information, such that there are not certain indications of the
individual with the data. Though full anonymization of clinical data is not possible
[ 10 ], removal of specifi c identifying information can reduce the risk of
re- identifi cation to reduce the risk to acceptable levels [ 11 ]. For some institutions,
careful use of data de-identifi cation has allowed the creation of very large research
cohorts, that use both data from EHRs linked to biospecimens from discarded sam-
ples [ 12 ]. By reducing the risk to privacy and confi dentiality, they have fully reduced
the burden to patients needed to create a biobank.
9.2.3
Data Collection and Integration
Once a patient has consented for use of their EHR data for a natural history, second-
ary data use study, the patient engagement is basically done. For interventional stud-
ies or natural history studies that require primary data collection, patient participation
begins at consent. Subjects are required to participate in either intervention or data
collection activities multiple times throughout the study, for the initial data, for the
fi nal outcomes measurements, and for monitoring during the course of the study.
This can also be a burden for the patients, requiring time and inconvenience. Ideally,
data needed for research is the same as data needed for care, but often the research
and care processes are separate, requiring either different testing schedules or dif-
ferent data elements. To reduce patient inconvenience, some researchers have con-
sidered integrating scheduling data for clinical care and research calendars [ 13 ].
The result would be that where appropriate, research visits and clinical care visits
are coordinated, decreasing the patient burden for research participation.
Search WWH ::




Custom Search