Ethical Considerations When Designing and Conducting Internet

Download Report

Transcript Ethical Considerations When Designing and Conducting Internet

Human Studies Program
Office of Research Compliance
Vice Chancellor for Research and Graduate Education
CTE
DCS IRB Series – Event 3
December 1,2014
Denise Lin-DeShetler

Why do we need to discuss internet research?
◦ Research about internet and using internet is increasingly
popular and important
◦ Federal regulations were written before internet became
universal
 They are currently silent about the use of internet for research
involving human participants
 Some guidance available, but they are ever-evolving; based on
FAQs
Human Studies Program
Office of Research Compliance
“Internet Research” as a tool for research and as a locale or
venue of research:
◦ Research involving study of information already available on the
Internet
◦ Using Internet as a medium for recruiting or interacting with subjects
◦ Research about the Internet itself and its effects (e.g., popularity of
social media types)
◦ Research about Internet users
◦ Internet used as an experimental intervention
Human Studies Program
Office of Research Compliance
Research involving:
 Survey instruments, databases,
search engines, databanks
 May not involve direct
interaction with human
participants, but PII may still
be collected
Research (qualitative and
quantitative) that utilizes
Internet “spaces” such as:
As a tool…
As a locale….
Chat rooms, gaming world, other
other simulated locales
But….
Research methods using Internet phoning, video conferencing or online chat
(i.e., Facetime, Skype) can be both use as a tool and a locale.
Stages in a Research Project
1. Recruitment
2. Data Collection
As a tool….
Observations
Data Mining
3.
Publication of Results
Possible re-recruitment
Human Studies Program
Office of Research Compliance
Advertising
Consent Process
Enrolling Participants
Human Studies Program
Office of Research Compliance

The recruitment material placed on the internet should
follow similar rules of any other recruitment material:
◦ Are not unduly coercive
◦ Do not promise or imply favorable outcomes or other benefits apart
from what is written in the protocol
◦ Do not communicate that the study product (or intervention) is safe
or effective for the purposes under investigation
◦ Do not make claims that the product or intervention is known to be
equivalent or better to any other product or intervention
◦ Do not use language that regulatory authorities, such as the IRB,
have approved the research
◦ Do not emphasize payment
Human Studies Program
Office of Research Compliance





Paid advertisements
In-text advertisements
Advertisements on social
networks
Social network page to
promote study
Any study-specific direct
advertisements via blog,
blog post, tweet, or text
IRB Should Review



Advertisement of studyspecific information limited to
basic trial information (title,
purpose, and study locations)
Communication that are
education or provide
generation information (i.e.,
podcast describing symptoms
of HIV)
Study-specific publicity
intended for general audience
(i.e., news stories)
IRB Do Not Need to Review



Types of internet-based recruitment methods:
◦ Email
◦ Online Advertising
◦ Chatroom Postings
All recruitment materials need to be reviewed and
approved before use.
General listings of research studies on websites that
provides limited information do not need prior
approval from IRB (e.g., SONA, Clinicaltrials.gov)
1.
2.
3.
4.
Study Title
Purpose
Protocol Summary
Contact information
Human Studies Program
Office of Research Compliance

Sending promotional messages (ads) about a study
without using the medium in ways typical of a
common user.
◦ Is this appropriate use of the website?
◦ Intrusive to potential participants?

Advertising in social network sites or Twitter groups,
such as a cancer support group, may seem intrusive and
inappropriate to members of that group
Human Studies Program
Office of Research Compliance
Appropriate methods for documenting consent should reflect
the risk and complexity of the research
◦ For minimal risk: waiver of consent whereby participants click on
the “I agree” statement on consent form to proceed to research
activities
 Must meet the regulatory criteria for a waiver of informed consent
◦ For more than minimal risk:
 Traditional signed consent, returned to PI who will then give consenting
participants password to access research material
 E-signatures
 Consent process through chat, email, or other online venue
Human Studies Program
Office of Research Compliance

Child participants:
◦ Authentication of age with use of age verification systems
◦ When research involves participation of minors, both child assent
and parental consent still applies, unless research meets appropriate
criteria for waiver of parental permission

Readability
◦ It is more difficult for investigators to discern whether or not their
participants understand the consent because process can be done
asynchronous, and the opportunity to ask the researcher questions
about the research is usually slim.
Human Studies Program
Office of Research Compliance

When obtaining consent over the internet, we
assume that potential participants at least 18
years old
◦ Age verification systems
◦ Questions that are more easily answerable to one
who is at least 18 (i.e., identifying images of old
items)

To authenticate identity when research
involves more than one interaction:
◦ Provide study participants PIN to be used for
subsequent computer and internet-based data
collection.
Human Studies Program
Office of Research Compliance
Public vs. Private Domains
Human Studies Program
Office of Research Compliance
Data Mining
Informational Risks
Human Studies Program
Office of Research Compliance
Case 2: A researcher wants to collect data from an HIV
support group website on Facebook to examine how much
perceived support participants are receiving from
participating in different types of support (chat groups vs.
support group sessions vs. blogs) they learn from subscribing
to the website. The website allows users to make
suggestions about the usability of the different resources the
website provides.
Does this study require IRB approval if no names or other
identifiers are collected?
Can this study be done without PII?
Human Studies Program
Office of Research Compliance
Per 45 CFR 46.101(b), research using only secondary data
that is publicly available or de-identified do not need IRB
approval
However:
Research that utilizes data from blogs, newspaper responses,
and social media networks should not have the same rules
for IRB approval because the expectations of privacy by
participants varies depending on the venue
Human Studies Program
Office of Research Compliance
Is there unauthorized disclosure of illegal behavior, substance
abuse, or chronic illness?
 Legality/ Criminal acts
 Stigmatization
 Financial
 Reputation
We look at identifiable information that is being collected:
 User posting on study-specific website in which user may be
sharing PII
 De-identified quotes gathered from blogs that can inadvertently
be traced back to individual
 Repurposed photograph of individuals or group posted on
website without permission
Human Studies Program
Office of Research Compliance

Publicly available means:
◦ Available at no charge to anybody with a computer
◦ Available to anybody willing to pay the requisite fee
◦ Available to anybody who meets the terms of a use agreement

Data that is publicly available qualifies for exemption
Human Studies Program
Office of Research Compliance


Privacy: refers to individuals’ right to have control over
access to themselves and their information
Confidentiality: refers to how information that is
obtained from individuals is protected.
Human Studies Program
Office of Research Compliance
Private information (in research) is “information about
behavior that occurs in a context in which an individual can
reasonably expect that no observation or recording is taking
place, and information which has been provided for specific
purposes by an individual and which the individual can
reasonable expect will not be made public (e.g., medical
records).



If the individual intentionally posts information on the
internet, then this is public information, UNLESS the
policies and/or terms of service of the entity hosting the
information indicates this is “private.”
Some may assume their post is “private” but are in fact,
not.
If in doubt, always go conservative. That is, consider it
private information.
Human Studies Program
Office of Research Compliance