Penn Medicine Clinical Research

Informed Consent Development

Informed consent involves providing a potential participant with adequate information to allow for an informed decision about participation in the clinical investigation. The Penn IRB has protocol template for biomedical and social behavioral research. The informed consent templates provide all of Penn required considerations and language which should be incorporated with the language regarding the specifics of the trial https://irb.upenn.edu/homepage/formsandtemplates/

Refer to Penn Medicine One Research Informed Consent Process Guidance: This guidance is to clarify the roles and responsibilities of those involved in the execution of the informed consent process for prospective human participants consistent with federal and state regulations, and institutional policies.

If you need research consents translated (video, document, virtual), please use HUP Interpreter Services (link will redirect to UPHS' Share point website, will require UPHS login).

There are several technology options for obtaining informed consent at Penn. The following table summarizes available options, and more information can be found on the research technology page

E Sig Technology Validated E Signature Capability  Cost Recommendation
REDCap No Unique link/ log in No Recommended for non FDA regulated studies
MyChart Yes* Portal username and password No

Recommended for studies involving Penn patients/ portal users
Epic eConsents Yes* Typed Name Signature Line No Recommended for FDA and non FDA regulated studies in person
TrialX Yes Typed Name, Signature Line OCR- Operations                               Recommended for FDA and non-FDA regulated studies virtually and in person
DocuSign Yes Unique link/ username and password Yes/ $8 per envelope Recommended tool for consent in FDA regulated / regulatory application studies

*While the EMR is not validated to part 11 standards and signing credentials meet.  Onc certification standards and can be used in accordance with part 11. We recommend that you have an SOP and process mapped out for any one of these tools before use.  Additionally the signature pads/ stylus singing is not considered to be FDA regulated e signatures.

Table comparing electronic signature technology options at Penn Medicine for clinical research.

 

Artificial Intelligence (AI) is defined as, development of computer systems able to perform tasks that normally require human intelligence, such as visual perception, speech recognition, decision-making, and translation between languages. AI is an umbrella term used to encompass various technologies used to perform tasks such as machine learning, natural language processing and deep learning.

When conducting human subjects research, a researcher may engage with an AI tool in some of the following capacities, not all of which will always require consent:

  • The AI tool collects data from humans through interaction or intervention like a ChatBot;
  • The AI tool is used to obtain informed consent from patients/ research 500;
  • The AI tool is used to attain, explore, or otherwise access identifiable data about research subjects/ patients for example to ingest data and train a model or perform a screening function for a study; or
  • The AI tool acts as an extension or representative of the investigator/ study team by answering questions for research subjects/ patients during the trial or conducting some trial functions. This may be through note writing, ordering prompts, ChatBot activities, etc.

In accordance with HRPP, the Penn IRB strongly recommends, and in some cases may require, for studies in which there is prospective consent from patients and Artificial Intelligence (AI) is used, that subjects be informed. This allows patients to make informed decisions about whether to participate.

Template AI Language for Consent and Considerations

This protocol involves the use of Artificial Intelligence (AI.) AI is an umbrella term used to encompass various technologies use to perform tasks such as machine learning, natural language processing and deep learning. In this study the following data from you will be shared with the AI tool  (list out specifics for your study for example demographic data, answers to questionnaires, medications, medical record data, etc.)

Describe how the subject will be interacting with the AI; will their data be shared, are they chatting, etc.

Describe if the AI is generative in nature, meaning reacting to the persons answers or simply giving preset answers, as applicable.

Describe what the AI will do with the data, what limitations, if any, will be placed on it, whether or not data can be removed and how, how data will be shared, if applicable

Comment on the risks of the use of AI explaining to the subjects that there may be unknown or unforeseeable risks especially in that even with the removal of identifiers, in many cases with AI, it does not necessarily mean that private or sensitive information may not be disclosed and potentially connected back to the research subject.)

Note: If PHI is being accessed by AI should also be described in the HIPAA authorization as a tool where PHI will be disclosed.

This document does not cover the use of AI to translate or transcribe consents for patients/ research subjects. Use of this functionality should be reviewed, and the technology used vetted for data security and validity, in advance of IRB review.  Additionally refer to the Penn IRB SOP on consent.