Social media is changing the face, structure, and possibly even the outcomes of clinical trials. The ability to identify very specific groups of people with a relevant diagnosis or symptoms, the communication forums that connect people living with the same disease, and the opportunity to target advertisements are all becoming a part of successful clinical trials.
KPI recently had the opportunity to interview Sandra Shpilberg, who founded Seeker Health—an end-to-end solution for enrolling people in clinical trials by mining the vast information present on multiple social media platforms. She sold Seeker Health to Eversana in 2018 and recently authored the book New Startup Mindset.
In our interview, Shpilberg highlighted the power of patients and social media, noting that through social media they are educated, engaged, and focused on getting what they believe is the best treatment for themselves.
Even with all the benefits of digital recruitment there are also new challenges for clinical trials: patients who (unintentionally or intentionally) unblind themselves via social media, compromising the data integrity of clinical trials.
Unblinding via social media can happen rather easily. Patients are seeking out platforms where they can connect with others who have the same disease and are participating in the same clinical trial. It’s natural that discussions about personal experience and side effects occur—which can lead to accurate conclusions about who is receiving the treatment and who is receiving a placebo.
In addition to threatening the integrity of the clinical trial, unblinding through social media poses other risks. One of these is patients leaving the study if they suspect they’re receiving the placebo. As one mom of a child in a clinical study shared with Heide Ledford in a nature article, “Nobody wants to be in the control,” and when the patient involved has limited time, the costs of remaining in a trial without experiencing progress are high for participants—may be too high.
Another risk is the premature dissemination of trial results. Ledford reports that a study published in 2012 found patients using outside laboratories to test for results during studies and sharing them online, and participants comparing appearance and taste of drugs to try to determine if they were on a placebo. Information publicly shared on discussion boards was even used by at least one financial analyst to make a recommendation about the company’s stock.
Sharing clinical trial experiences online can also influence the experiences of the participants, particularly when it comes to subjective reporting about side effects.
Social Media and FDA Guidance
It’s crucial that pharma companies recognize the difference between the U.S. Food and Drug Administration (FDA) guidance for approved products and products undergoing clinical research and trials. When it comes to approved products and information provided via the internet or social media, the draft guidance is clear:
- for approved products, firms are “responsible for communications on the Internet and Internet-based platforms, such as social media, made by its employees or any agents acting on behalf of the firm to promote the firm’s product, and these communications must comply with any applicable regulatory requirements.”
- for approved products, firms must be compliant when they are writing, collaborating, or exerting control or influence on specific content about their product provided by a third party
- for approved products, firms are “not responsible for a product-related communication that appears on the firm’s own forum, an independent third-party website, or through social media” although the guidance recommends correcting misinformation in these areas
- for approved products, firms are “generally not responsible for third-party UGC [User generated Content]… truly independent of the firm” although content on a platform owned or operated by the firm “should include an overarching clear and conspicuous statement that the firm did not create or control the UGC” and again, it’s recommended that misinformation is corrected/addressed
- for approved products, “firms might be responsible for UGC that they solicit or influence, regardless of the forum”
However, pharma companies do have an ethical responsibility to correct misinformation they become aware of in UGC and must report any known adverse events mentioned in UGC.
FDA guidance on internet or social media platforms and clinical studies is not yet formalized but it will be very different than the guidance for approved products. Brian Bollwage and Marc Hoffman write for Chiltern in a November 2019 document that when it comes to the impact of social media on the integrity of a clinical trial, authorities globally will take a strong stance, even if the sponsor didn’t directly influence the trial’s integrity. “That is to say, while the sponsor may not have caused the unblinding to occur, if an unblinding event is significant enough to call into question the integrity of the trial data, the relevant authority can, and would be expected to, consider the disqualification of data from such trials” even leading to “losing” a trial.
Reducing the Risk of Patient Unblinding with a Patient Recruitment Agency
One solution for reducing the risk of patient unblinding via social media is to engage a patient recruitment agency. The agency acts to keep social media activities separate from the CRO and sponsor while running communities or social media pages that are owned and operated by the agency.
The agency can legally independently advertise the study as a third party within those social pages or communities and drive participants to the sponsor’s study site. Potential study participants then complete the pre-screening and referral process under full HIPAA and GDPR compliance.
The agency keeps patient pre-screening data completely separate from the sponsor and CRO. Non-enrolled patients (which the community will be geared towards) do not represent an unblinding risk. The agency generally charges a client fee for each referral into the study.
Investigators should never be given access to, or be informed of any specific social media communities where studies are being promoted. This is easier to accomplish with centralized media management by an agency that is not the CRO.
An agency can also be contracted to mine social media for information about studies and report conversations that may lead to unblinding. This may also help to counteract risks to future enrolment if inaccurate information is being shared, or messaging that risks patient safety, as the American Cancer Society Journals pointed out in an October 2019 editorial.
Reducing the Risk of Patient Unblinding with Informed Consent Forms (ICF)
Informed consent forms (ICF) may need to include language to discourage the use of social media as it relates to the patient’s participation in the study. This helps to prevent enrolled patients from unblinding themselves.
In an article for Chemical & Engineering News (c&en), Bibliana Campos Seijo acknowledges that this type of request can seem heavy-handed. A possible alternative or an addition to the request is to set up a moderated group for all the participants where you can guide discussions away from sharing any information that may compromise the trial. For example, Bristol-Myers Squibb has worked with Inspire to develop a moderated online community that allows trial participants to connect and support each other.
Ledford also notes the information on the ICF should be handled carefully. She quotes Craig Lipset, head of clinical innovation at Pfizer who talks about how including guidelines about communications can create worry or detract from a participant finding online support, “You can see in online communities where participants are scared that they have just signed a confidentiality agreement and will be thrown in jail for posting.”
When KPI Agency works with clients we sometimes advise using language that asks participants to limit their discussions about participation to family, friends, and health care providers, and to point out that social media sources can provide misleading information as they are not monitored, but to do so in non-coercive patient friendly terms. It’s more about explaining the rationale rather than scarring or forcing them.
Since different trials run different risks, ICFs should be written on a case-by-case basis that takes into consideration when and how social media use may impact data integrity.
Reducing the Risk of Patient Unblinding with Additional Safeguards
The risk of unblinding on any social media platform always exists, so it’s important to apply all possible safeguards to prevent influencing the study outcome. Ultimately, every clinical trial is about science and data integrity, and the human participants without whom there would not be a trial.
Clinical staff can also inadvertently unblind a patient. In an article by Luke Gelinas and Ramon Jones for Advarra, they give an example of how this can happen: patients on a forum are sharing and contrasting their symptoms while participating in a study and a staff member responds by sharing which patient received the placebo and which received the treatment. This is effectively addressed “By placing safeguards for the types of data shared, monitoring participant comment sections, and communicating rules for engagement” to patients and clinical staff.
Lipset recommends becoming “savvier” in setting up trials by narrowing eligibility criteria, so studies are less subjective and more difficult to game by participants.
Take the time to carefully craft information that helps participants and their families understand why certain information must remain confidential, and how blinding helps ensure accurate, reliable, and science-based results. As the ACS Journal’s Editorial says, “It is clear that the onus lies heavily with us as investigators to educate our patients on the appropriate use of social media as it relates to clinical trials.”
Every person participating in a clinical trial is facing health challenges that may change the quality or duration of their life. For some, the trial is their last chance to get better. In almost every case, the focus on individual health will take priority over the scientific results of the study. This needs to be considered as we take action to reduce the risk of patient unblinding. Ideally, the conclusion of the trial has protected data integrity without inflicting undue hardship on the participants.
