Videoing clinical trials: how to balance US HIPAA vs. visual data sharing
Visual data in healthcare is vital for optimising efficiency, protecting staff and patient safety, and providing more remote care. In clinical trials, video brings even more potential benefits.
It captures an objective truth - a useful comparison for post-review and analysis, instead of solely relying on notes or the subjective opinion of one or two specialists in the room. This means better quality data and retrospective re-evaluation of results.
Patients may need to talk, interact, or move in some way to evaluate the effectiveness of treatment; these are best recorded and preserved in a video to help garner the full picture for later on in the process. In fact, video is often a requirement for some trials.
As a result, sharing visual data from clinical trials to further research, secondary research and to help develop/patent new drugs, is common practice for clinical research organisations. Access to patient data is vital for the pharmaceutical industry and the wider medical field. But it needs to be captured safely and shared in a way where sensitive data is protected.
The challenges for clinical trials: quality of results, cost, and time spent
Trials are often expensive and time-consuming, and poor quality outputs can result in months of re-evaluating data, solving issues, going back and forth with the Food and Drug Administration (FDA), or potentially having to redo trials.
In the US, between the difficulty in recruiting, reduced health services, and fewer people with health coverage - the costs of clinical trials continue to rise.
A clinical trial can last between 2 to 8 years, with the average cost ranging between $50 million and $100 million. The cost of delay in drug sales can range from $0.5 million to $8 million per day.
Additionally, patents for new drugs last roughly 10 years - if the trials hold up most of this time, it leaves less time for the drug to be on the market and generate revenue.
The pharmaceutical industry needs rapid turnover to get trials and analyses conducted as quickly and efficiently as possible so they can get new drugs to the market. Thanks to the objectivity that video can provide, recording clinical trials can improve chances of success and help streamline some of these processes.
The challenge with sharing videos in the clinical trials setting? US HIPAA (the Health Insurance Portability and Accountability Act, 1996).
What does US HIPAA mean for clinical trials?
HIPAA sets a nationwide standard for the protection of patients’ personal health information (PHI). This means any individually identifiable health information held or transmitted by a “covered entity” or its business associate - whether it's electronic, video, on paper or oral.
Healthcare providers and relevant third parties have to store all medical data securely, and cannot share it without knowledge or consent. Whilst not all researchers are technically covered entities, many may rely on covered entities for research support or information sources.
For clinical trials that rely on the sharing of data, HIPAA has specific guidelines for research purposes:
“Research purposes” are a systematic investigation, including research development, testing, and evaluation, designed to develop or contribute to generalizable knowledge.”
To share data from federal or privately-funded clinical trials, clinical research organisations have to comply with HIPAA’s Common Rule (the rules of biomedical and behavioural research ethics in the US) and/or the FDA’s human subject protection regulations.
When research is submitted to the FDA for approval, they require information to be de-identified and stripped of any information that could identify, either indirectly or directly, the patients/participants. Under HIPAA, outside parties who access this data would also be required to remove any personal identifiers.
To comply with HIPAA and the FDA, the main takeaways for clinical research organisations are:
Authorisation and informed consent are needed from patients. This means they need to be told how their medical information will be disclosed.
Not having consent to share personal data is very rarely allowed. If researchers gain approval from an Institutional Review Board or Privacy Board, they can waive patient consent for sharing personally identifiable data. For example, if researchers cannot use anonymised information or the research could not be carried out properly with patient authorisation.
Data in this context can be disclosed - provided it is de-identified. For video, this means anonymisation of personal and identifiable data.
How can video from clinical trials be shared under HIPAA and FDA?
Data anonymisation, or video redaction, allows clinical research organisations to protect PHI in the video because it removes all personal identifiers so that data is effectively de-identified.
When it is no longer categorised as PHI, anonymised video can be used and shared responsibly under HIPAA and other data privacy laws.
However, the level of anonymisation is uncertain but based on the risk factor, i.e. how likely that a patient can be identified from the information and this needs to be looked at on a case-by-case basis. There is no generally accepted standard for anonymisation and the process is often context-dependent. Health agencies like the FDA do not require 100% anonymity.
The HIPAA Privacy Rule specifies the two de-identification methods:
a formal determination by a qualified expert, or
the removal of specified individual identifiers and the absence of actual knowledge by the covered entity that the remaining information could be used alone or in combination with other information to identify the individual.
Data sharing is fundamental to the clinical trial space and helps to improve healthcare and treatment. It’s clear that collected and shared trial data has to prioritise data privacy and protection, whilst also factoring in cost, time and quality of video for results.
Our video redaction solution, Secure Redact, means that clinical research organisations can do just that. Powered by machine learning, it can automatically detect over 99% of PII for redaction, saving you time and keeping costs low - all through a scalable and secure platform.