AUTH/3869/12/23 - Complainant v Eli Lilly

Case Summary

This case was in relation to a Facebook advertisement posted by a medical publisher that linked to a Lilly article discussing new breast cancer treatment. 

There was an appeal by the complainant of one of the Panel’s rulings.

The outcome under the 2021 Code was:

This summary is not intended to be read in isolation.
For full details, please see the full case report below.

CASE/3869/12/23

COMPLAINANT v ELI LILLY

Allegations about social media posts

CASE SUMMARY

This case was in relation to a Facebook advertisement posted by a medical publisher that linked to a Lilly article discussing new breast cancer treatment. 

There was an appeal by the complainant of one of the Panel’s rulings.

The outcome under the 2021 Code was:

This summary is not intended to be read in isolation.
For full details, please see the full case report below.

FULL CASE REPORT

A complaint was received from a complainant who described themselves as working in Oncology about Eli Lilly and Company Limited.

COMPLAINT

The complaint wording is reproduced below with some typographical errors corrected:

“Last week, the 14th of December, I was scrolling through Facebook on a bus and a story come up, sponsored by [named third-party]. It had the heading ‘Treating High Risk Early BC’. The sub heading stated ‘Leading oncologists from Germany, Italy and Brazil discuss the latest research. As my mother is going [through] treatment [for] breast cancer I opened it. 

I was surprised the link opened an article on [named third-party] – ‘The Use of CDK4/6 Inhibitors in the Treatment of High-Risk, HR-Positive, HER2-Negative, Early Breast Cancer’ which was funded by Lilly. And in Table 1, they list 2 clinical trials with their medicine abemaciclib, one published and one ongoing. 

First I thought I was targeted on Facebook by [named third-party] because I work in Oncology. But when I clicked on the button ‘Why did you saw this ad’, it suggested because of 3 criteria:

1.    Been on a hashed list that [named third-party] used
2.    Set their age to 18 or higher
3.    Primary location in the United Kingdom.

This then made me wonder what a hashed list is. According to Facebook, a hashed list is a coded list that helps Facebook show you relevant ads, while also anonymising and protecting your information. Businesses and organisations that you've previously interacted with may want to contact you on Facebook. If you've given a company your contact information, such as your phone number or email address, and that company decides to advertise on Facebook, you may see one of their ads.

Companies upload a coded list of phone numbers or email addresses to optimise their ad delivery. These coded lists are called hashed lists. Hashed lists are lines of code that ensure that Facebook doesn't receive your actual contact information from the business. Facebook's system then takes the contact information that it already has and turns it into code to see if it matches the code uploaded from the company. If the code matches, Facebook may show a relevant ad. After the match process has been completed, Facebook will delete the hashed list.

It’s a bit creepy that I got targeted in this way and I wonder if Facebook saw the photos of my mother’s bald head after she lost her hair or used my check in at the oncology ward and sold that data to [named third-party]? It makes me very content knowing how I was hashed but don’t think it should be allowed to push information about medicines to unknowing Facebook users. Even more so if they are not doctors or health professionals. 

More than 530 people has since liked this advertisement, there are 6 comments on it and it’s been shared 26 times. One of the people that shared it is [name], who is a train driver from [England]. [name] replied – glory to God. [name] tagged [name] in a comment and [name] according to his Facebook profile is a stage 4 cancer thriver and digital creator. 

This [named third-party] advertisement has since last week Thursday popped up on my timeline a few times. It has even made its way to my Instagram.
I feel really angry seeing [named third-party] and Lilly push their breast cancer treatment to the general population this way. My mother sent me a link on Facebook only last week about the Duchess of York calling her new breast Derek and I do feel this [named third-party] advertisement triggers me. 

Please find the link to the article which you don’t need to log in to read. 
I also attach some of the screenshots of the advertisement as it appeared on my phone.”

When writing to Eli Lilly, the PMCPA asked it to consider the requirements of Clauses 3.6, 5.1, 5.5, 5.6, 15.5, 16.1, 26.1, 26.2 and 2 of the 2021 Code.

FURTHER INFORMATION FROM THE COMPLAINANT

“[Personal details redacted]

I am a medical doctor, yes. But I did not publish that information on Facebook. And I don’t use the same email account for Facebook as I did for [third-party medical publisher]. I also did not sign up to getting [third-party medical publisher] advertisements on Facebook.

And that’s what is triggering me. Facebook should not know that I am a doctor and but may know I have a special interest in breast cancer as my mother was treated for breast cancer in 2022/23. In my professional job I have no interest in breast cancer treatment as it’s not my line of expertise. [Third-party medical publisher] targeting me with advertisements on my Facebook timeline because of my mother’s condition and not my professional interests. Based on the comments by other people on the advertisements, they were not HCPs. One is a train driver and I think a second one is a teacher.

I believe [third-party medical publisher] spammed me with advertisements because of my mother’s treatment and based on the two comments I read which was posted by other people, they received the same advertisement and they were not HCPs. This is pushing family of people with cancer to ask about this specific treatment and gives hope of a new treatment which promises to be very effective for breast cancer. This is not right and should not be allowed. And I got spammed continuously. For example, I got the same advertisement on my Facebook feed on the following days and times:

[details of eleven dates and 71 times provided].

I have screenshots for each instance mentioned above and also for [six of] the dates. It’s been a bombardment of spam!”

ELI LILLY’S RESPONSE

The response from Eli Lilly is reproduced below:

“We acknowledge your letter dated 21 December 2023 and its attachments, detailing a complaint regarding allegations about social media posts. 

Lilly is deeply committed to the medical education of healthcare professionals and is equally committed to conducting this work consistent with established standards.  We recognise the seriousness of this complaint. The complaint is about posts arising from [third-party medical publisher]’s social media accounts on Facebook and Instagram with a link to an editorial article with the heading of ‘The Use of CDK 4/6 Inhibitors in the Treatment of High-Risk, HR-Positive, HER2-Negative, Early Breast Cancer’ sponsored by Lilly (‘Editorial Article’). On receipt of this complaint, as a matter of utmost caution, Lilly informed [third-party medical publisher] to cease all dissemination tactics for this Lilly sponsored editorial article to the UK.

[Third-party medical publisher] is the leading online global destination for physicians and healthcare professionals worldwide offering the latest medical news and expert perspectives, essential point-of-care drug and disease information and relevant continuing medical education. It is widely perceived as a credible and established source of authority for physicians and aligns with our Lilly goals to support health care professional education globally at scale.  [Third-party medical publisher] designed and initiated this project, which was developed by their own clinical strategy team, and approached Lilly with a request for funding to develop independent medical education programmes. 

Lilly investigated and understands this was an editorial article developed independently by [third-party medical publisher] as part of a proposal by [third-party medical publisher] and agreed between the Lilly International Business Unit (IBU) located outside the UK and [third-party medical publisher’s parent company] based in the USA. For reasons we’re still working to understand, Lilly IBU colleagues who partnered with [third-party medical publisher] on this project did not make the UK compliance team aware of this initiative and thus the local compliance team did not have the opportunity to sensitise the project owner to the potential social media risks associated with such a project in the UK.  Although we acknowledge that in the development of this project our own robust Lilly processes were not adequately followed and consequently a UK compliance review did not occur, and therefore we failed to maintain high standards, we believe that there are no breaches of Clauses 3.6, 5.5, 5.6, 15.5, 16.1, 26.1, 26.2, and Clause 2:

Clause 3.6: Materials and activities must not be disguised promotion.

It has never been the intention of Lilly or that of [third-party medical publisher] to create a promotional content and the processes followed by both parties is with the view that this content is strictly independent education based on balanced scientific data. As per the Agreement the target audience is [third-party medical publisher]’s professional network who indicate during registration (to [third-party medical publisher]) that they are physicians who specialize in oncology and because this is an independent activity it’s in [third-party medical publisher]’s sole discretion to utilise different editorial drivers and to decide on the third-party websites to deploy editorial drivers. Consequently, Lilly never instructed or expected [third-party medical publisher] to utilise Facebook or other social media for this educational activity. [Third-party medical publisher] confirmed to Lilly that the content would not be disseminated on Facebook/Instagram even to healthcare professionals and be gated behind a [third-party medical publisher] login. 

[Third-party medical publisher] is a well-recognized source of authority for healthcare professional use only and there is a clear statement at the top of every page to confirm this: ‘This site is intended for healthcare professionals’. The Editorial Article is published on [third-party medical publisher]’s web site in the form of [clinical programme] that is described by [third-party medical publisher] as ‘an arm’s length evidence-based review and discussion of a clinical topic or condition by a group of experts’. The editorial content, tone and language is written with highly scientific terminology by an independent group of HCP experts and represents a factual and balanced overview across numerous CDK4/6 inhibitors and other treatments too. The content is not at all promotional in tone nor there are any visuals that may represent a product brand. 
Additionally, the content intentionally did not go through Lilly approvals (not even through a technical accuracy review as suggested in the Agreement) given the strictly independent non-promotional nature of this agreement, which prevents control of the content. We note the PMCPA guidance on arm’s length agreements here that: ‘it is possible for a company to sponsor material produced by an independent organisation which mentions its own products and not be liable under the Code for its contents, but only if, inter alia, there has been a strictly arm’s length arrangement between the parties.’ 

Clause 5.5: Material relating to medicines and their uses, whether promotional or not, and information relating to human health or diseases which is sponsored by a pharmaceutical company or in which a pharmaceutical company has any other involvement, must clearly indicate the role of that pharmaceutical company.

A mandatory transparent disclaimer is provided for the readers at the top and bottom of the web site page where the Editorial Article is located:

‘Funded through sponsorship by Eli Lilly. [Third-party medical publisher] approached Eli Lilly to fund the production of this editorial article. Eli Lilly has had no influence over the selection of the authors or the content of the article. The sponsorship fee included an honorarium for the authors, who were contracted and paid by [third-party medical publisher] Editorial. The views and opinions of the authors are not necessarily those of Eli Lilly, or of [third-party medical publisher], its publisher, advisers, or advertisers. No part of this publication may be reproduced in any form without the permission of the publisher.’ 

Clause 5.6: Material should only be provided or made available to those groups of people whose need for or interest in it can reasonably be assumed. Material should be tailored to the audience to whom it is directed.

As it is undertaken by [third-party medical publisher] in the Agreement, the target audience for the Editorial Article is [third-party medical publisher] professional network who indicate during registration (to [third-party medical publisher]) that they are physicians who practice in the United Kingdom and specialise in Oncology. 

[Third-party medical publisher] has provided us more information on their targeting methodology:

-    The [third-party medical publisher] Facebook content that popped up on the complainant’s Facebook account is not a general post by [third-party medical publisher], it is a ‘paid advert’, labelled as ‘Sponsored’ which is targeted to very specific audience. The Facebook content which is a mere title ‘Treating High-Risk Early BC’ with a visual and a ‘Learn More’ button is not available to [third-party medical publisher] Facebook followers. 
-    [Third-party medical publisher] uses hashed lists and pixel audiences to accurately retarget known [third-party medical publisher] users and to ensure relevance to the target audience. 
-    In this instance [third-party medical publisher] used pixel audiences as the targeting methodology and not hashed lists (described by the complainant).  
-    For an advertised post to be made visible to a user of Facebook or Instagram, they must meet both criteria outlined below as also described in Figure 1. Although [third-party medical publisher] has >800,000 followers on Facebook and similar on Instagram, the two criteria below must be met for this paid post to be viewed which reflect a clear validated HCP target:
(i)    they MUST have visited [third-party medical publisher] AND accepted the Meta Tracking Pixel cookie.
AND
(ii)    they must also be registered to the [third-party medical publisher] platform as a validated HCP based on the criteria outlined below - referenced as ‘77184’ in Figure 1.

Figure 1: HCP Targeting infographic:

[Third-party medical publisher] informed us these techniques are well established methods of retargeting on social media. They note the Facebook paid advert that has the link to the Editorial Article has not been targeted to any user behavior on Facebook or photos posted by any of their connections as mentioned by the complainant. [Third-party medical publisher] believes that Facebook does not even have that feature to allow targeting by user behavior.

Clause 15.5: The telephone, text messages, email, faxes, automated calling systems and other digital communications must not be used for promotional purposes, except with the prior permission of the recipient.

As explained above, the Editorial Article is developed by a group of experts independently from Lilly, based on an arm’s length arrangement and the content is educational, represents a factual and balanced overview across numerous CDK4/6 inhibitors and other treatments and is not promotional in tone and content. Therefore, we consider that the activity of dissemination of the Editorial Article should not be considered as promotional. 

Notwithstanding the foregoing, we believe that the prior consent requirement was met because, as explained to us by [third-party medical publisher], for any recipient of the link to the Editorial Article via [third-party medical publisher] social media activities that person must have visited [third-party medical publisher] website AND accepted the Meta Tracking Pixel cookie.  

Clause 16.1: Promotional material about prescription only medicines directed to a UK audience which is provided on the internet must comply with all relevant requirements of the Code.

As stated above, the complained editorial article was independently developed by [third-party medical publisher] for educational purposes. It was written with highly scientific terminology by an independent group of HCP experts and represents a factual and balanced overview across numerous CDK4/6 inhibitors and other treatments; and is not promotional in tone and content. Therefore, we believe that the article should not be considered as a ‘promotional material’ and should not be subject to the relevant requirements of the Code.

Clause 26.1: Prescription only medicines must not be advertised to the public. This prohibition does not apply to vaccination and other campaigns carried out by companies and approved by the health ministers. 
Clause 26.2: Information about prescription only medicines which is made available to the public either directly or indirectly must be factual and presented in a balanced way. It must not raise unfounded hopes of successful treatment or be misleading with respect to the safety of the product. 
Statements must not be made for the purpose of encouraging members of the public to ask their health professional to prescribe a specific prescription only medicine.

As clarified above, it has never been the intention of Lilly or [third-party medical publisher] to present the Editorial Article to the public. The targeted audience was clearly defined in the Agreement as [third-party medical publisher] professional network who indicate during registration (to [third-party medical publisher]) that they are physicians who practice in the United Kingdom and specialize in Oncology. 

That said, as we investigated the matter in detail upon receipt of the complaint, we realized that because our internal processes were not properly followed the local compliance review was missed. Unfortunately, the regional team developing this project failed to consider the PMCPA Social Media Guidance and that has resulted in us having no provision in the Agreement requiring [third-party medical publisher] to be prudent about inadvertent exposure of the content to members of the public via any third-party platform. Whilst we accept this would have been a valuable addition to the Agreement, we also maintain that this Facebook paid advert was utilised entirely independently of Lilly with no expectations set by us to use social media. [Third-party medical publisher] made it very clear in the Agreement that they ‘will maintain sole and exclusive control of the editorial and associated drivers’.

Clause 2: Activities or materials must never be such as to bring discredit upon, or reduce confidence in, the pharmaceutical industry.

Although we acknowledge that we might not have maintained the high standards by not following our internal review processes, we believe that neither the content of the editorial article nor the intended and agreed method of dissemination of the content should result in breach of Clause 2. This is mainly because the article is developed independently by [third-party medical publisher] for educational purposes and is not promotional in tone and content; also as per the Agreement it is clear that the targeted audience is healthcare professionals, oncologists in particular, who registered themselves as such while using [third-party medical publisher] platforms. Thus, it has never been the intention of Lilly to allow the Editorial Article to be accessed by the general public; last but not least the Editorial Article was presented only to healthcare professionals who have accepted the [third-party medical publisher] Privacy Policy which includes the use of Meta Tracking Pixel cookie, in other words who have given prior permission for advertorial interactions from [third-party medical publisher]. 

Lilly’s Ethics & Compliance Program is extensive in the UK and around the world with a set of policies and procedures designed to ensure that the company operates with high ethical standards. The program is guided by the company’s values of integrity, excellence, and respect for people. All employees receive compliance training, and the company periodically reviews its ethics and compliance programs. 

Lilly regrets that the Complainant had an unwelcome experience on social media that involved a medical education editorial owned by [third-party medical publisher], targeted to oncologists and funded by Lilly. Lilly understands and fully respects the ABPI Code of Practice and strives to ensure that all of its activities are in adherence to the Code at all times. We have a dedicated compliance and medical signatory team alongside a clear governance process in place for handling funding requests for independent activities that would normally ensure a UK review for such activities to ensure it follows both the letter and the spirit of the local laws, regulations, policies, and procedures that govern our business. Following this complaint, we are taking corrective measures in consultation with our global ethics and compliance team to ensure that no global or regional educational activities with a UK nexus can occur without clear consultation and approval from the UK affiliate.”

PANEL RULING

The complainant’s allegations related to a Facebook advertisement posted by a medical publisher which contained a link to an Eli Lilly sponsored editorial article with the title “The Use of CDK4/6 Inhibitors in the Treatment of High-Risk, HR-Positive, HER2-Negative, Early Breast Cancer”. The complainant alleged that the Facebook post appeared on their social media feed and whilst they work in oncology, noted that members of the public who were not health professionals had interacted with the post by liking and/or commenting on it. Examples included a train driver from [UK] and a digital content creator who was a stage 4 cancer survivor.

The Panel noted that the PMCPA was dealing with a series of cases that involved [medical publishers] and various companies. The allegations and evidence provided in each case differed and thus consequently the rulings. Each case was considered independently on the evidence before each Panel.

The Facebook advertisement was posted by the medical publisher and labelled as “Sponsored” below the medical publisher’s name. The post was titled “Leading oncologists from Germany, Italy and Brazil discuss the latest research” followed by a picture of a woman smiling and exercising beneath the prominent heading “Treating High-Risk Early BC”. Below the picture was the website address of the medical publisher above “Treating high-risk early BC” with a “Learn more” button beside it. The “Learn more” button linked to the Eli Lilly sponsored editorial article mentioned above. 

The Panel noted Eli Lilly’s submission that the medical publisher had designed and initiated this project and had approached Eli Lilly with a request for funding to develop independent medical educational programmes. The Panel further noted that the editorial article was developed by the medical publisher as part of a proposal agreed between the Lilly International Business Unit (IBU) located outside of the UK, and the medical publisher which was based in the USA. 

The Panel noted Eli Lilly’s submission that its overseas IBU colleagues, who had partnered with the medical publisher third-party on this project, did not make the UK compliance team aware of this initiative and so they did not have the opportunity to consider the potential risks associated with such a project. The Panel also noted that the Work Order (WO) specified that the target audience for this instruction would be physicians specialising in oncology and based in the UK, as well as other specified countries and that the third-party would recruit target audience members to the sponsored article using editorial drivers such as third-party medical publisher invitations and graphic adverts.

The Panel noted that the WO was between Eli Lilly and Company Limited, and the US third-party medical publisher, and sat beneath a Master Agreement and queried whether the limited company was the UK company. The position was unclear. It referred to payment in US dollars. The Panel queried whether the WO was signed by a UK employee. Eli Lilly’s response implied that the role of the US IBU was such that the UK company had no relevant role, responsibility, or knowledge of the activity in question. The WO covered three activities: an algorithm, expert guidance, and the [clinical programme] in question. Eli Lilly’s response was not clear that the activity in question was part of what appeared to be an established program of sponsored activities and not an isolated unsolicited proposal from the third-party medical publisher.

It was a well-established principle that UK pharmaceutical companies were responsible for the activities of overseas affiliates if such activities fell within the scope of the Code. The Panel noted that the Facebook post was disseminated, amongst others, to UK health professionals and considered that irrespective of the UK company’s role the activity fell within the scope of the Code. 

The Panel noted that the article at issue appeared to be developed as a result of the [clinical programme] described in the WO and was described as a text-based evidence-based review on a specified topic which would be published across the third-party medical publisher’s professional network. 

The Panel considered whether Eli Lilly had a responsibility for the material. It was clear that both the article and the drivers to the article including those deployed on third party websites, were covered by the WO. The WO provided that Eli Lilly would have no influence over the content of the [clinical programme] other than to optionally review it for technical accuracy and the drivers were within the third-party medical publisher’s sole editorial control.

The Panel further considered the medical publisher’s proposal, which specified that it would be an “arm’s length” project, however, upon considering the WO agreement, it was apparent that the medical publisher would be providing Eli Lilly with monthly traffic reporting which included: the number of visits, the time spent on each page and traffic to key actions. The reporting would include breakdowns of: company target audience segments, physician traffic (up to three specialities and the rest reported in aggregate) and health professional traffic. 

In addition, the WO provided at Section B, [clinical programme], for the delivery of an estimated number of visits and a commitment for the medical publisher to reach 70% of specific country estimated visits. It was further stated that the estimated visits did not reflect the medical publisher’s “promise to achieve defined levels of coverage”. The phrase “defined levels of coverage” did not appear to be defined in the WO. In addition, the WO stated that Eli Lilly had the option to “review the [clinical programme] for technical accuracy only” although, according to Eli Lilly, this did not take place. In the Panel’s view, the overall arrangements were not at arms’ length noting in particular: the detailed traffic reports which went beyond due diligence to ensure that the sponsorship was used for the intended purpose; and the contractual expectations and requirements over the level of dissemination of the sponsored article. In reaching this decision, the Panel also bore in mind that the Proposal document referred to Eli Lilly’s product abemaciclib and compared its efficacy favourably to a competitor product in the same class which “did not demonstrate a significant benefit as add on therapy for early breast cancer”. The proposal document then stated that there was an opportunity for a [clinical programme] to contextualise these results and with a focus on defining how the pharmacology of the two agents or the patient population affected outcomes. The Panel therefore decided that Eli Lilly had a responsibility for the sponsored article and its dissemination including the related drivers such as the Facebook post.

Disguised promotion

A key question for the Panel to address was whether the article itself was promotional. The Panel noted the broad definition of promotion at Clause 1.17 of the Code, which referred to any activity which promotes the administration, consumption, prescription, purchase, recommendation, sale, supply or use of its medicines. The Panel noted the broad dissemination of the Facebook post and linked article to members of the public. In addition, the Eli Lilly-sponsored article, explicitly mentioned their medicine and the therapeutic indication. Noting these points, the Panel considered that the sponsored article constituted promotional material for which Eli Lilly would be responsible. The Panel noted its comments above that Eli Lilly was responsible for the Facebook post and further considered that the post and linked article would be considered as inextricably linked.

Clause 3.6 stated materials and activities must not be disguised promotion. The supplementary information to Clause 15.6 “Disguised Promotional Material”, the wording of which closely aligned to Clause 3.6, stated, amongst other things, that promotional material must not give the impression that it is non-promotional, and the identity of the responsible pharmaceutical company must be obvious.

Whilst the Panel noted that the article mentioned it was sponsored by Eli Lilly, the Panel considered that it was not clear that Eli Lilly was involved from the Facebook post which was the first item seen by the complainant. The Panel noted its description of the Facebook post above; it did not directly or indirectly refer to either Eli Lilly or the medicine. It was only on opening the link that the complainant would have become aware that they had been directed to Eli Lilly-sponsored material, the nature of which was promotional. The Panel considered that the promotional nature of the sponsored article had been disguised and a breach of Clause 3.6 was ruled in relation to the sponsored article.

Declaration of Involvement

Clause 5.5 stated that material relating to medicines and their uses, whether promotional or not, and information relating to human health or diseases which is sponsored by a pharmaceutical company or in which a pharmaceutical company has any other involvement, must clearly indicate the role of that pharmaceutical company. The supplementary information to this clause included that the declaration of sponsorship must be sufficiently prominent to ensure that readers of sponsored material are aware of it at the outset.

The Panel considered that Eli Lilly’s declaration of sponsorship should have been prominent on the Facebook post, for which Eli Lilly had a responsibility as set out above, but was only included in the linked article in small text. This omission meant that Eli Lilly had failed to ensure that readers were aware of Eli Lilly’s involvement at the outset. The Panel ruled a breach of Clause 5.5.

Promotion to the public

In relation to the alleged promotion of a prescription only medicine to the public, the Panel noted its decision that the Facebook post and linked article were promotional. The Panel noted the header of the article stated in small font that the site was for health professionals only; it was unclear whether there were additional safeguards when accessing the linked post. The Panel considered the post and associated article, if accessed by, and/or liked by, members of the public, may have encouraged them to enquire generally about treatment for breast cancer. In reaching this decision, the Panel bore in mind the complainant’s reference to a positive comment by a member of the public in response to the post. Noting its comments above, the Panel considered that the post, combined with the linked material, amounted to advertising of a prescription only medicine to the public and may, on the balance of probabilities, have encouraged members of the public to ask their health professional to prescribe a specific medicine; the Panel, therefore, ruled a breach of Clauses 26.1 and 26.2.

Tailoring towards the audience and consent

Clause 5.6 stated that material should only be provided or made available to those groups of people whose need for, or interest in, it can reasonably be assumed. Material should be tailored to the audience to whom it is directed. The Panel noted the content of the sponsored article and that it was intended for health professionals; it had been linked to a post on a public forum. The Panel noted that, in its view, the material had not been restricted to the audience to whom it was intended nor was the promotional nature of the material appropriate for the public and therefore the Panel ruled a breach of Clause 5.6.

The Panel noted that Clause 15.5 stated that, amongst other things, digital communications must not be used for promotional purposes except with the prior permission of the recipient. The Panel noted the complainant’s concern that they did not expect to be on the receiving end of Facebook posts promoting breast cancer treatment. The complainant submitted further information to explain that they do not use the same email account for Facebook as they did for the third-party medical publisher and did not sign up to getting third-party medical publisher advertisements on Facebook. The complainant added that they did not have any interest in breast cancer treatment within their profession but that their mother was treated for it in 2022/23. The Panel considered Eli Lilly’s response that two criteria needed to have been met for the post to be viewed. Firstly, the user must have visited the third-party website and accepted the “Meta Tracking Pixel” cookie and secondly, they must be registered to the medical publisher’s platform as a validated health professional. The Panel noted that the third-party medical publisher’s cookie policy explained that they utilised third parties such as Facebook to deliver advertisements based on user’s activities and interests inferred from their use of services. The Panel noted its comments above about Eli Lilly’s responsibility for the Facebook post and its dissemination, bearing in mind the contractual requirements and expectations. In such circumstances, the Panel considered that Eli Lilly should have satisfied itself that governance around consent met the requirements of Clause 15.5. On balance, the Panel did not consider that a pharmaceutical company could rely on a reference to Facebook in a cookie policy and the acceptance of a pixel cookie to demonstrate that it had met the requirements of Clause 15.5 and consent, in relation to the dissemination of a promotional Facebook post, to a health professional’s personal Facebook account. The Panel queried how many individuals read and understood such policies. Bearing all of these points in mind, the Panel considered the requirements of Clause 15.5 had not been met and ruled a breach accordingly.

Clause 16.1

The Panel noted that Clause 16.1 required promotional material about prescription only medicines directed to a UK audience which is provided on the Internet to comply with all the relevant requirements of the Code. The Panel further noted that the supplementary information to Clause 16.1, Website Access, stated that “Unless access to promotional material about prescription only medicines is limited to health professionals and other relevant decision makers, a pharmaceutical company website or a company sponsored website must provide information for the public as well as promotion to health professionals with the sections for each target audience clearly separated and the intended audience identified”.

The Panel noted its rulings of breaches of Clauses 26.1 and 26.2 above in relation to the sponsored article. The Panel also noted that the only reference to the intended audience, health professionals, was at the top of the linked article. The Panel considered that the article did not meet the requirements of Clause 16.1 and its supplementary information; a breach of Clause 16.1 was ruled.

Clauses 5.1 and 2

The Panel noted Eli Lilly’s submission that it accepted that a provision in the Agreement requiring [named third-party] to be prudent about inadvertent exposure of the content to members of the public via any third-party platform and it accepts this would have been a valuable addition to the Agreement. The Panel also noted Eli Lilly’s submission that it maintained that the Facebook-paid advertisement was utilised entirely independently of Eli Lilly with no expectations set by it to use social media. [Named third-party] made it very clear in the Agreement that they “will maintain sole and exclusive control of the editorial and associated drivers”. The Panel noted its comments above about Eli Lilly’s responsibilities in this regard and that the contract stated that the third-party medical publisher may deploy editorial drivers on third party websites. In the Panel’s view, Eli Lilly was therefore on notice that third parties such as Facebook might be used and therefore ought to have taken proactive steps to understand and minimize risk in relation to the Code. This was particularly so given the sponsored article concerned a sensitive issue, the treatment of early breast cancer, which ought not to have been disseminated to the general public. The Panel also bore in mind the complainant’s concerns about the frequency of delivery of the post. Bearing in mind these points and its rulings of breaches of the Code above, the Panel considered that high standards had not been maintained. A breach of Clause 5.1 was ruled.

The Panel noted that Clause 2 was used as a sign of particular censure and reserved for such use. Taking account of the above breach rulings, and with the aforementioned reasons, the Panel considered that the ruling of a breach of Clause 5.1 was a proportionate sanction and no breach of Clause 2 was ruled accordingly. 

APPEAL BY THE COMPLAINANT

The complainant’s written basis for appealing is reproduced below.

“I would like to appeal the [no breach of Clause 2] ruling of the PMCPA panel for the following main reasons:
1.    Industry sponsored [third-party medical publisher] adverts reach the eyes of non-HCP’s on Facebook and Meta. Social media is not the proper place to advertise medicines.
2.    There appears to be a lack of transparency in how Lilly and [third-party medical publisher] targeted individuals on the Meta platform (Facebook and Instagram). Factors such as my mother's recent breast cancer treatment may have influenced Meta's targeting. Promoting cancer treatments to people living under the cloud of cancer or other health concerns has implications for the industry's reputation.
3.    The ABPI Appeal Board [sic] should examine the business model behind this and similar Meta advertisements in the UK funded by the pharmaceutical industry. Several [third-party medical publisher] related cases are in considered by the Panel right now and each case is judged on individual merit. This appeal urges the ABPI Appeal Board [sic] to consider the broader implications of third-party publishers promoting medicines to (non-HCP) UK audiences disguised as education. Companies must declare transfer of value to HCPs and health care organisations but not to publishers.

Background information
[third-party medical publisher] 

[Third-party medical publisher] is a website providing access to medical information for clinicians and medical scientists; the organisation also provides continuing education for physicians and other health professionals. 

The site generates revenue primarily through advertising and sponsored content for pharmaceutical, biotech, and medical device companies, as well as hospitals, health insurance providers, and lifestyle and wellness brands. It is listed on the NASDAQ and reported a turnover of more than $750million last year.  

In 2013 an article in JAMA stated that [parent company] along with its sister site [third-party medical publisher] were the top recipients of industry dollars, a large and undisclosed conflict of interest. [a link was provided].

A study of [third-party medical publisher] and [parent company] also found both services to lack neutrality and exhibiting bias potentially based on very high payments (compared to their industry competitors) from the pharmaceutical industry. (Wikipedia)

In April 2024, [third-party medical publisher] was strongly criticized for running educational content funded by [tobacco company], with critics saying that the tobacco industry should abstain from any involvement in medical education. [a link was provided].

I would like to refer you to [third-party medical publisher] Medical Affairs, a business unit launched in 2021 to accelerate the adoption of life changing therapies into clinical practise. (PR Newswire) [a link was provided] It claims to ‘exclusively integrate your Brand within [third-party medical publisher]’s award winning interactive editorial content’. 

The Meta Pixel and microtargeting
Previously known as the Facebook pixel, the Meta Pixel is a JavaScript code designed to monitor user activity across various websites and compile the collected data into a unique identifier. Leveraging data from interactions, behaviours, and interests, Meta subsequently creates targeted advertisements. These may include health-related ads if the user's activity suggests an interest in cancer treatments, support groups, or similar topics.

Earlier this week, the advert below appeared on my Instagram feed. This was one of three variations for Vit D supplementation that appeared on my timeline. [an image was provided]. 

When asking Instagram why certain ads are displayed, the response consists of two parts. First part is the type of profile that the advertiser would like to target. In Eli Lilly’s response letter, its stated that this was set as a UK based health care professional with an interest in oncology. The second part is based on online activity such as reacting to posts or visiting websites about breast cancer.

Additionally, it includes whether other people with similar profiles frequently interact with the topic of breast cancer. Meta then predicts that the user may have an interest in breast cancer as well. [an image was provided]. According to the Meta Business Help Centre, the Meta Pixel is one of Facebook’s most powerful targeting capabilities. Instead of building your audience with the social network’s self-serve ad tool, you have the option to let the Meta Pixel do it for you. It makes the whole process faster, easier. But even better, it adds to the effectiveness of your advertising campaigns too.

The Meta Pixel contains extensive data about a person’s browsing history outside of the Meta platform. According to the Meta Pixel, [third-party medical publisher] tracks my browsing history with 57 other partners, capturing my browsing data on my device. [an image was provided] [Name] is a website with detailed guidance for marketeers on how to target their ideal audience on the Meta platform, written by a former Facebook Ads consultant - [a link and image was provided].

This guide suggests digging down based on demographics, interests and behaviours.

Demographics include: sex, location, ethnic affinity, fields of study, politics, relationships, work industry, life events

Interests listed include hobbies, choice of entertainment, activities, health interest
Behaviours include devices used, primary email domain, friends of certain people. 

Using all the data harvested and collected by the Meta Pixel, allows microtargeting. The Information Commissioners Office defines microtargeting as a form of online targeted advertising that analyses personal data to identify the interests of a specific audience or individual in order to influence their actions. [a link was provided]. However, the tool has come under scrutiny due to its potential to capture sensitive information, including health, financial, or personally identifiable data, often without explicit user consent.

The ICO gives the following example to explain: if you are researching buying a new car, you may receive microtargeted advertisements from third-party car manufacturers, organisations selling cars, or organisations selling car insurance products based on your research activities.

Many examples have been published where patients with a recent cancer diagnosis suddenly found their Meta pages flooded with adverts based on this diagnosis [links and names of four examples provided].

[One of the examples] developed software to track Facebook tracking activity as well as the advertisements Facebook places in front of its users. This allowed them to provide live visualisations of the data analysis. This research established a decade ago that Facebook uses your social relationships to make money. [a link was provided]. I refer to figure 2 of publication: [an image was provided].
 
To quote the author, ‘We were also surprised to see how the advertising Facebook targeted, via stories on a user's newsfeed, was drawing not just on data collected on a participant's Facebook page but from all other internet sites they use with a Facebook connection’. Facebook tracks all usage on computers/devices via its social plugins which have been placed on more than 13 million sites including health and government sites (Gibbs, 2015).

Reference to microtargetting in popular TV

Following the Cambridge Analytica scandal, the concept of microtargeting is so well known that it’s even referenced in the popular TV show, the Good Fight. [a link was provided]

Personal background of relevance

My mother received treatment for early breast cancer in 2022/23. I posted a photo of her ringing the bell in the chemo ward when her treatment was completed, with her signature bald head. I recall posting two other photos with her in the chemo ward as well as before or after follow up visits in the cancer care unit.

During this time, I did a lot of online research about the treatment of breast cancer. My [browsing] history would have a long list of visits to NICE, [third-party medical publisher], Clinicaltrials.gov to name a few.  My mother joined several breast cancer support groups on Facebook, sharing updates from them on her Facebook timeline. All of this would have been captured by the Meta Pixel.

Thinking back, I had dinner with her oncologist early in December 2023. Also, in 2023 I had another friend receiving radiation for her HR sensitive breast cancer, which we discussed on Facebook. I also discussed investigations and second line treatment options with a Radiologist in Facebook Messenger at that time. All of this data would be captured by the Meta Pixel.

The week before the adverts started appearing, my mother sent me a link on Facebook about the Duchess of York calling her new breast implant Derek.

Appeal Argument

I appeal the Panel's decision not to rule a breach of Clause 2. This volume of advertisements to individuals with loved ones suffering from breast cancer tarnishes the pharmaceutical industry's reputation. This advert received 530 likes and was shared 26 times in two weeks. A teacher with breast cancer commented "Glory to God," indicating she read the ad and it gave her hope for a cure. 

I question Eli Lilly and [third-party medical publisher] 's transparency in their response. It must be unusual for an international business unit to fund a project of this size without consulting the UK affiliate if the target audience is in the UK. The Work Order covered both the editorial and its drivers but no further information was given about the drivers/adverts or its cost. Eli Lilly claims [third-party medical publisher] had sole discretion over third-party websites but didn't expect them to use Meta. What lead to that expectation?  Nonetheless, [third-party medical publisher] did advertise on Meta, and the cost surely warranted a separate proposal line in the WO. 

According to the Ely Lilly response letter:

- In this instance [named third-party] used pixel audiences as the targeting methodology and not hashed lists (described by the complainant)’

[Third-party medical publisher] admits that my Meta Pixel was used to target me as someone interested in breast cancer, contradicting their reasoning in figure 1 of the Eli Lilly response. Their claim that only validated HCPs saw the advert is refuted by evidence of a teacher and train driver reacting to it. 

I also do not accept the statement that:

‘The Facebook paid advert that has the link to Editorial Article has not been targeted to any user behavior on Facebook or photos posted by any of their connections as mentioned by the complainant. [named third-party] believes that Facebook does not even have that feature to allow targeting by user behavior’.

That is exactly what the Meta pixel is – a unique identifier allowing microtargeting based on the interests identified based on their activity on and outside of the Meta platform. The claim that a NASDAQ listed company with a turnover of $750 million does not know this, is disingenuous. 

Regarding clause 15.5, I do not agree that I give permission for breast cancer adverts on my social media. Ely Lilly's argument that my acceptance of [third-party medical publisher]'s terms and conditions when I registered 15 years ago implies consent is a stretch. 

Eli Lilly's oversight failures in this project suggest a breach of Clause 2, given the multiple examples of poor oversight by Eli Lilly. Eli Lilly states that the UK affiliate was not consulted although the activity had a UK audience. The content did not go through Lilly review and approvals although the contract suggested it should have. The Work Order included a statement that the ‘third-party would recruit target audience members to the sponsored article using editorial drivers such as third-party medical publisher invitations and graphic adverts’. But no detail was given yet Eli Lilly states that [third-party medical publisher] promised not to use the Meta platform but did so anyway. 

I observed that the PMCPA is addressing several cases involving [media publishers]. The Panel evaluates each case independently based on evidence. I urge the Board to consider repeated breaches, indicating a business model misaligned with the spirit of the ABPI Code of Practice. Integrating branded messages within editorial content as the [third-party medical publisher] Medical Affairs business unit stives to do, intentionally blurs the line between promotion and education.’’

On being provided with a redacted copy of the Work Order (attachment 3 to Eli Lilly’s response to the complaint) the complainant’s written basis for further appeal is reproduced below.

“I would like to thank you for the opportunity to review the redacted Work Order and the transparency with which you diligently investigated this compliant.

I can see this Work Order forms part of a Master Agreement between Eli Lilly and
[parent company] with the project name [provided] Campaign.

The WO describes the following services and deliverables:
A. The algorithm for decision making
B. The [clinical programme] programme to be hosted by [parent company]
C. An [expert] programme to be hosted by [parent company]

Under the description of services and deliverables B, the [clinical programme], I note the
following points:

Section 2. [parent company] may deploy drivers on third party websites and have exclusive editorial control of all these drivers.

Section 3. Monthly reporting will include: visits, time spent on page, breakdown of
company target audience segments, physician traffic and HCP traffic.

Section 4. [parent company] plans to deliver 286 [clinical programme] visits from the UK. It also states that this commitment is not an essential part of the Company’s funding to the [clinical programme].

Section 5. The target audience shall be members of the [third-party medical publisher]’s Professional Network who registered on [third-party medical publisher] and indicated that they are physicians.

Reading the redacted Work Order, it suggests that Eli Lilly had no decision or control over the way [parent company] advertised the content and which drivers they used but graphic ads and the placement on third party websites were mentioned. Should Eli Lilly have asked which websites would be used? Should Eli Lilly have asked that [parent company] use restraint in the sheer number of times a single reader was spammed by the same ad on a specific website? I would argue yes. Pharmaceutical Companies and third parties acting on their behalf should not spam readers to the point of frustration. Should Eli Lilly have asked what steps are being taken to prevent these adverts from being seen by the general public in the UK? I would argue yes. 

Section 5 of the WO states that audience members shall only be HCPs registered on [third-party medical publisher]. Yet I offered evidence of two people who read the Facebook ad and clearly were not HCPs. Their comments were visible to others on Facebook, making the original advertisement visible to a wider audience and in doing so promoted a medicine to members of the public.

Should companies like Eli Lilly and [third-party medical publisher] actively advertise a medicine on a person’s social media such as Instagram or Facebook? The first thing I saw when I walked out of a concert and switched on my phone was an advertisement for breast cancer. It is not appropriate to remind me of my mother’s breast cancer as I walk out of a concert and opened Facebook to post a photo of the experience. In the US, it is acceptable to advertise breast cancer treatments to the public, but this is not the case in the UK and the rest of the EU. [A copy of the link was provided]

[Third-party medical publisher], by their own admission, targeted social media users which they identified by means of the Meta Pixel as people with an interest in breast cancer treatment. Contrary to what is stated in the WO, they were not all health care professionals but people who are vulnerable because they or a loved one has breast cancer. Making them more likely to click on the link to the [third-party medical publisher] website, read about the medicine and ask their doctor about it. That is not the actions of an ethical company. [Third-party medical publisher] admitted to using pixel audiences (now termed the Meta Pixel) but later in their response denied that Facebook as this feature. I simply cannot accept that statement as truthful or credible.

The way [third-party medical publisher] recruited readers on social media is brings discredit to the pharmaceutical industry. They acted on behalf of Eli Lilly and Eli Lilly should have had more competent processes in place to prevent something like this bringing their reputation into disrepute.

Again, I think you for the opportunity to review the WO and offer my thoughts”

RESPONSE FROM ELI LILLY

The Eli Lilly’s written basis for responding to the appeal is reproduced below.

“We acknowledge your letter dated 2 June 2025 and its attachments, detailing the complainant’s appeal regarding allegations about social media posts. 

Lilly regrets that the complainant, who is both a healthcare professional (HCP) and a relative of a breast cancer patient, had this experience and recognise the difficult emotional circumstances the complainant has described in connection with this case. This has never been Lilly’s intention through the delivery of online medical education to HCPs. Lilly remains committed to supporting the education of healthcare professionals and to conducting this work in accordance with established ethical and professional standards. The promotion of prescription only medicines to patients or members of the public is not permissible in the UK and Lilly do not permit such activities. It has never been Lilly UK’s intention to promote prescription only medicines on social media either. Breast Cancer Oncology treatment decisions in the UK are driven by sub-specialised oncologists in consultation with their hospital multi-disciplinary team. 

Lilly considered this complaint very seriously and as stated in our initial response, our internal processes were not adequately followed meaning this international-led project did not undergo appropriate UK review and risk assessment. Subsequently, we have accepted without appeal all the breaches ruled by the Panel and provided the requisite undertaking and assurance in respect of the Panel’s rulings of breaches of Clauses 3.6, 5.1, 5.5, 5.6, 15.5, 16.1, 26.1 and 26.2 of the Code. We believe the Panel has applied a sanction that is both sufficient and proportionate in this case and do not consider the need at the complainant’s appeal to revisit information on these specific breaches.

The appeal in this case is about whether a Clause 2 breach should be issued or not.  We respectfully agree with the Panel’s ruling that the circumstances of this case do not amount to conduct that “brings discredit upon or reduces confidence in the pharmaceutical industry,” and we set out our rationale below.

Eli Lilly’s robust governance approach 

Upon receipt of the initial complaint, Lilly treated the matter with the utmost seriousness and initiated a comprehensive global investigation. This process involved co-ordinated efforts across our Medical, Commercial, HR and Ethics & Compliance functions. Briefings were provided to senior leadership including the Head of the International Business Unit, the Chief Compliance Officer and the Chief Executive Officer. This approach reflects our strong leadership commitment to upholding the highest ethical standards across all our operations globally.

In direct response to the concerns raised we acted promptly and decisively by withdrawing all related materials from the [third-party medical publisher] platform as an immediate and precautionary measure. The comprehensive global investigation determined that a single employee did not adhere to the established procedures, which would have prompted UK compliance review. The individual was subject to a disciplinary process, and the insights gained were disseminated extensively among commercial teams both within and outside the UK. We have subsequently conducted a thorough review of relevant third-party agreements and internal procedures in addition, to strengthen governance and mitigate the risk of recurrence and put in additional safeguards to ensure that all global activities with a UK nexus receive a UK compliance review.  Such UK compliance reviews include assessment of all arrangements, funding and specific elements such as targeting, platform use and social media which are assessed to determine compliance with the ABPI code.

Protecting Patient Safety and Public Health

The article subject to this complaint was a scientific, factual and balanced review of the entire CDK4/6 inhibitor class in the treatment of breast cancer. It was authored in academic language and was designed for an expert oncologist audience. The authors were independently selected by [third-party medical publisher] and are internationally recognised leaders in oncology: [names and details provided]

An article of such high scientific complexity, developed independently by recognised medical leaders in oncology would not reasonably be expected to have a direct negative impact on patient safety or public health. 

Partnership with [third-party medical publisher] 

The complainant raised concerns regarding [third-party medical publisher], the organisation who conducted this educational activity. [Third-party medical publisher] is a leading global online platform that provides medical news, clinical information and continuing medical education (CME) to healthcare professionals. Based on publicly available estimates, it has a global membership of approximately 13 million HCPs. The complainant is a member themselves which we hope means this is a credible and trusted source of scientific information. 

This independently developed article was hosted on a platform specifically designed for healthcare professionals and not for patients or members of the public. It is common and accepted practice for pharmaceutical companies to collaborate with established medical education providers to support independent education for healthcare professionals. [Third-party medical publisher] reached out to Eli Lilly to request funding for their own independently developed medical education programme. 

While we acknowledge the complainant's perspective, we do not believe that concerns regarding [third-party medical publisher] ’s broader funding model fall within the scope of this case and subsequent appeal. The global pharmaceutical industry supports a significant proportion of independent medical education initiatives via third-party providers including private education companies, academic institutions and medical conferences. This is common around the world and many healthcare professionals rely on these independent medical educational resources for their continued professional development (CPD).  

Summary 

We regret the impact of this activity on the complainant as the wellbeing of healthcare professionals, patients and caregivers remains our highest priority. 

Lilly is committed to the highest ethical standards and has a long-standing global reputation for integrity. Eli Lilly has been named to Ethisphere’s list of the World’s Most Ethical Companies multiple times including in 2025.  Ethisphere evaluates companies based on criteria such as ethics and compliance programs, corporate citizenship, and leadership. We recognised early in this process that our high standards were not met and have taken this matter seriously, providing corrective coaching and reinforcing our already robust governance processes to prevent recurrence. We believe that patient safety and public health were not compromised.

We fully accepted the Panel’s rulings which we consider both sufficient and proportionate. Our prompt and thorough management of this complaint reflects our continued commitment to upholding trust and confidence in the pharmaceutical industry and we agree with the Panel Ruling of no Clause 2 breach in this regard.”

FINAL COMMENTS

The complainant had no final comments.

APPEAL BOARD RULING

The Appeal Board reminded itself of the wording and requirements of Clause 2. 

Eli Lilly’s investigation showed that, due to individual human error by an employee based outside of the UK, this activity had been miscategorised which resulted in lack of oversight by the UK affiliate. The Appeal Board was concerned that it had been possible for the project to be completed on that incorrect basis, without any other person identifying the error of checking that the correct process had been followed. In the Appeal Board’s view, a process where an individual human error could have led to this failure created unnecessary risk. The Appeal Board accepted Eli Lilly’s submission that if the correct procedures had been followed, and the UK company informed, the activity would not have been approved. 

The Appeal Board considered that Eli Lilly’s policies and procedures in relation to such funding requests required significant improvement to ensure greater oversight by the UK affiliate. 

The Appeal Board considered the unusual nature of this error, in that an individual outside of the UK had not followed Eli Lilly procedures and policies and approved this sponsorship in isolation. The Appeal Board took account of the Panel’s rulings of breaches of the Code including Clause 5.1, which Eli Lilly had accepted, and the Appeal Board determined that this isolated incident did not bring discredit upon the industry, as a whole.  The Appeal Board therefore upheld the Panel’s ruling of no breach of Clause 2 of the Code. The appeal on this point was unsuccessful. 

Complaint received    18 December 2023

Case completed    3 July 2025