pharmacovigilance & social media a new approach to …docs2.health.vic.gov.au/docs/doc... · a...
TRANSCRIPT
A new approach to Safety Signal Detection: Potential & Issues Karen Whitelock, Drug Safety Responsible, Novartis Pharma Australia
Dr David Lewis, Global Head of Pharmacovigilance
Melbourne, May 2015
Pharmacovigilance & Social Media
Disclaimer
2
The information within this presentation is based on the presenters expertise and experience and represents the views of the presenter for the purpose of this presentation.
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
Overview: Pharmacovigilance of social media
3
Social media landscape and pharmaceutical medicines
Regulations & guidelines for PV of digital media
AE reporting via Social Media
Overview of Novartis PV of digital media
WEB-RADR IMI project
• Assessment of pharmacovigilance using social media
• Mobile reporting of suspected adverse reactions
• Digital Drug Safety Surveillance
• Data Protection for Health Apps - Data Privacy
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
Trends in Consumer Technology Computer & smartphone evolution
The digital media landscape... Internet-based applications that allow for the creation and exchange of user-generated content
From: Chopra R - Pharmacovigilance & Digital Media (http://de.slideshare.net/rkc78834/pharmacovigilance-digital-media)
4 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
Multiple threads of digital media
Electronic Health
(eHealth)
Using information and
communication
technologies for the
provision of health related
services (diagnosis,
monitoring treatment)
Telemedicine
Delivery of health
care at a distance
using information
and communication
technologies consumer
healthcare.
Mobile Health
(mHealth)
Using mobile
communication
systems for the
provision of health
related services
Digital Health
The intersection of the
digital revolution with
consumer healthcare,
includes genomics (use of
gene chips to store a
patient’s genetic
identifiers and responses)
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
5
@Facebook 1.25b users
@Twitter 302m users
@YouTube >1b users
@Pinterest 70m users
Engaging with patients via digital media
Social media Use to provide
information & engage
with stakeholders. Also
valuable within the
organisation to
encourage collaboration
Apps An increasingly wide
array in the market
ranging from information
only through to
sophisticated sensors
and medical devices
Gamification Applying game design
techniques and
mechanics to “real life”
applications in order to
make them more
engaging
Wearables and
smart sensors So far, largely only
wristbands and sensors
but the potential is for
much more
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
6
Pharmaceutical industry use of digital media MAHs main use of digital media is for marketing and sales
Sponsors are primarily using social media for commercial purposes to distribute information about:
• Medicines (to healthcare professionals and non-HCPs)
• Diseases, and the treatment of disease
• Company matters including announcements
• To listen to patient and professional conversations about marketed medicines, and not to support clinical research
A minority of companies use social and digital media for:
• Patient engagement
• Patient recruitment and retention within clinical trials
7 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only | Streamlining
Multi-Centre
Research May
TGA:Monitoring the internet or digital media
1.2.2
Sponsors should regularly screen internet or digital media under their management or responsibility, for potential reports of suspected ARs.
• includes digital media that is owned, paid for and/or controlled by the sponsor.
• frequency of screening allows for valid ARs to be reported within reporting timeframe
based on the date the information was posted on the internet site/digital medium. • Sponsors may utilising their websites to facilitate collection of suspected ARs.
If a sponsor becomes aware of a suspected AR described in non-company sponsored digital medium, • the report should be assessed to determine whether it qualifies for reporting.
Cases from the internet or digital media, the identifiability of the reporter refers to the existence of a real person, that is, it is possible to verify the contact details of the reporter • e.g., an email address under a valid format has been provided
8
| Streamlining
Multi-Centre
Research May
2015 Workshop |
Karen Whitelock &
Dave Lewis|6 May
2015 | PV &
Social Media|
Business Use
Only
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
VI.B.1.1.4. Information on suspected adverse reactions from the internet or digital media
MAHs should regularly screen internet or digital media
under their management or responsibility, for potential reports of suspected ADRs. In this aspect, digital media is considered to be company sponsored if it is owned, paid for and/or controlled by the MAH.
• The frequency of the screening should allow for potential valid ICSRs to be reported to the competent authorities within the appropriate reporting timeframe
• Marketing authorisation holders may also consider utilising their websites to facilitate the collection of suspected ADRs
EEA Good Pharmacovigilance Practice (GVP) VI PV guidance on digital media (July 2012) focuses on ICSRs
9 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
Digital media and safety issues (GVP Module VI)
VI.B.1.1.4. Information on ADRs from digital media (ctd.)
If a MAH becomes aware of a report of suspected adverse
reaction described in any non-company sponsored digital
medium, the report should be assessed to determine
whether it qualifies for reporting.
VI.C.2.2.6 Emerging safety issues
Good practice for the MAH to monitor special internet sites or
digital media (e.g. patients’ support or special diseases groups)
• Check if they describe significant safety issues which may
necessitate reporting in accordance with VI.C.2.2.6.
• Frequency of the monitoring depends on the risks associated to
the medicinal product
10 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only | Streamlining
Multi-Centre
Research May
2015 Workshop |
Karen Whitelock &
Dave Lewis|6 May
2015 | PV &
Social Media|
Business Use
Only
MAH governance of the use of digital media Most MAHs have developed controls governing use of digital media
Nearly all MAHs have developed guidelines to address use of social media
General guidelines include:
• Rules for discussing company business on personal sites, how to set up a page or site, and privacy issues;
• Guidelines concerning authorized and unauthorized uses by personnel of social media;
• Social Media Advisory Board-specified practices;
• Guidelines for posting video online;
• Directives that stipulate one-way communication between personnel and patients involved in clinical studies;
• Restrictions, e.g., for company business only.
11 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only | Streamlining
Multi-Centre
Research May
2015 Workshop |
Karen Whitelock &
Dave Lewis|6 May
2015 | PV &
Social Media|
Business Use
Only
AE reporting via social media
Source: White paper on Adverse Event Reporting by VisibleR , Oct. 2011, http://www.visibletechnologies.com/resources/white-papers/adverse-events/
All brand mentions
257,177 posts; 224 brands
100%
Filtered for relevance
24%
With AE terms
5%
Adverse Event reports
0.4% of all brand mentions
3.3% contain AE-specific data
Contain AE
keywords
12,530 posts
3.3% mentions
are case reports
1 in 7 = name
& contact info
| Streamlining
Multi-Centre
Research May
2015 Workshop |
Karen Whitelock &
Dave Lewis|6 May
2015 | PV &
Social Media|
Business Use
Only
Overview of Novartis PV of digital media Highlighting challenges of monitoring the different channels
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
13
Novartis Pharma
Social Media Programmes Novartis Pharma Social Media
Listening Programmes
Novartis associate/vendor
screening non-NVS controlled
platforms
Single Reports
• AEs with 4 minimum criteria
Aggregate Reports
• AEs with 2 minimum criteria
Single Case Safety Reports
• Within 24 hours
• Send to DSE in the country where AE occurred
(if UNK send to country where AE was
received or reviewed (Local Programs) or
Switzerland/US (Global Programs)
Aggregate Reports
• Monthly reports / End of programme
• Send to country of the Programme Owner
Novartis associate/vendor
screening
NVS controlled platforms
Single Reports
• AEs with 4 minimum criteria
• AEs with 2 minimum criteria
(Novartis product & ADR)
Single Case Safety Reports
• Within 24 hours
• Send to DSE in the country where
the AE took place (if UNK send to
country where AE was received or
reviewed (Local Programs) or
Switzerland/US (Global
Programs)
IMI WEB-RADR research on social media WEB-RAdR consortium will explore PV of social media
Hypothesis Data capture, collation, timely data mining and appropriate analysis can provide actionable intelligence in relation to protection of public health. Research can provide value to stakeholders including patients, healthcare professionals, regulators and to the pharmaceutical industry. Emerging communication technology is changing the way people
interact with their healthcare providers and products • Large body of health care data is being generated in social media • Mobile technology creates an environment where people are constantly
connected to the Internet
The value of such data is not fully established Mining and analysis of social media is an emerging science Regulatory guidance is behind the emergence of new technology
14 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
| Streamlining
Multi-Centre
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
A brief history of Web-RADR over time... Web-RADR is an IMI project involving academia, HAs and industry
New technologies in mobile devices & new apps are leading to an evolutionary change in the extent, geographies and modes of use of the internet
Information gathering and sharing of experiences, opinions and suggestions is becoming routine
Web-RADR is a ground-breaking EU Innovative Medicines Initiative ( ) -funded initiative to recommend policies, frameworks, tools & methods
The project will explore the value of the new developments to assess the insights versus traditional PV
WEB-RAdR IMI project Comprises experts from regulatory agencies, academia, WHO & industry
WEB-RADR consortium brings together expertise from world-leading organisations in their fields
Project structured to enable EFPIA partners to participate in all areas
WEB-RADR has the potential to shape the regulatory framework of the future, thereby augmenting traditional pharmacovigilance
WEB-RADR offers the maximum possible benefits from a regulatory, societal & scientific perspective
16 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
WEB-RADR Consortium structure
EFPIA
Applicant Consortium
Regulators
& Patients
Novartis (Lead)
J&J (Deputy)
Sanofi-Aventis
Amgen
AstraZeneca
UCB
Bayer
+ GSK
MHRA (PM)
EMA
Halmed
Lareb
Eurordis
Universities of London, Liverpool & Gröningen
WHO (UMC), Epidemico and SRDC 17
| Streamlining Multi-Centre Research
May 2015 Workshop | Karen
Whitelock & Dave Lewis|6 May 2015 |
PV & Social Media| Business Use Ony
| Dr DJ Lewis | Web-RADR | 14 April 2015 | Business Use Only
Overview of Web-RADR work packages Effective project management to link the work streams together
Mobile & social
media usage
patterns 2b Regulatory framework WP 1
Scientific impact
& new insights 4
Social media
platform 2a
Mobile ADR
reporting app
& safety
comms 3a
19
Spontaneous
Reporting
Electronic
Paper based
• ICSR quantity & appropriateness
• Quality (VigiGrade)
• Demography
• Duplication (VigiMatch)
• Signal generation
• Qualitative Analysis
• Integration
Comparisons
WEB-RAdR Scientific Impact Traditional pharmacovigilance vs. New
Adverse Drug
Reaction
Positive benefit risk assessment
for each patient
Traditional model
HCP-centric
• Signal generation
• Trustworthiness
• Integration
App-
based
Social
Media
WEB-RADR model
Patient-centric
Identify risks &
tolerability issues
WEB-RAdR Pharmacovigilance of social media Targeting advances beyond state of the art pharmacovigilance
Novel signal detection techniques
Indicators of completeness + quality
Advanced geographic recognition
Filtering capability using EMA IME
Filtering for health system interaction
International collaboration portal for all users
Data export compatibility
Multi-source spontaneous report comparisons
Indicators of completeness + quality (vigiGrade)
20 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
Data quality indicators: healthcare vs patients
Assessment of Novartis data from SRs, literature and POPs
Initial and follow up all time Literature
Spontane
ous
Marketing
Programs
n % n % n %
Formulation 181 13.3
TTO 132 9.7
Demographics 1169 86.2
Indication 936 69.0
Dosage 891 65.7
Action Taken 557 41.1
Duration 19 1.4
Medical History 867 63.9
Lab Data 710 52.4
Dechallenge 225 16.6
Rechallenge 19 1.4
Total no of cases 1356
Average score 4.86946903
4960 65.7
2838 37.6
7058 93.5
4606 61.0
5529 73.2
3389 44.9
2571 34.0
3587 47.5
1135 15.0
1059 14.0
69 0.9
7551
4.740077
1345 73.7
694 38.0
1802 98.8
1371 75.2
1591 87.2
863 47.3
873 47.9
835 45.8
194 10.6
155 8.5
11 0.6
1824
5.6579509
Social media: caveats for patient safety
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
22
New developments will lead to new challenges for pharmacovigilance
MAHs must obey the law
MAHs must operate ethically
We must not cross the boundaries of the doctor-patient relationship, or be misled by false data or spurious signals
But we should be encouraged by the opportunities that digital media offer...
Future of social media: From telling to engaging
23
New apps and mobile technologies will continue to evolve
Best practices include:
•Establish relationships through two-way dialogue and human connections that encourage participation
•Demonstrate patient and customer responsiveness through timely, transparent responses and request feedback
•Share content that makes the public want to interact with the company and specific brands
•Use of innovative strategies to leverage information and drive online influence
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
Digital Drug Safety Surveillance: Twitter Monitoring pharmaceutical products in Twitter (Freifeld et al 2014)
Existing post-marketing adverse event surveillance systems suffer from under-reporting and data processing lags
Social media services such as Twitter are seeing increasing adoption, and patients are using them to describe adverse experiences with medical products
An analysis of 4,401 of these ‘posts with resemblance to adverse events’ (‘Proto-AEs’) from Twitter found concordance with consumer-reported FDA Adverse Event Reporting System reports at the System Organ Class level
Further research is required to investigate this finding
Identifying instances of Proto-AEs in the UK Pilot study results exploring PV of social media
Data collection scheme for Twitter & FAERS
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
26
Study to compare AE reporting via Twitter and patient reports to FDA
Correlation by SOC proto-AEs in Twitter/FAERS
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
27
Rank order (log-log scale): Proto-AEs = posts with resemblance to AEs
Social Media ADR Comparison to Pivotal Trial Patient reporting can be of good quality and differs from HCP sources
Antibody abnormal*Herpes zoster (skin)*
Swelling*Migraine*
Paraesthesia*Weight decreased*Skin discomfort*
Headache*Condition aggravated*
AST increasedDyspepsiaVomiting
Lymphopenia
Albumin in urineErythema
RashNausea
DiarrheaPruritisFlushing
Abdominal pain
Social Media (n=42)
051015
Pivotal Trial (Tx arm N=769)
0 10 20 30 40
dimethyl fumarate
Dec 1, 2013 through Jan 14, 2014, Twitter & Facebook
drug:event pairs coded using MedDRA v16
≥2% higher than placebo
Trial Data: Tecfidera US PI
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
Social media reports vs FAERS Hypothesis generating for risk minimisation
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
Note: Until methods for signal detection in social media and scientific analysis has been fully developed, refined
and finalized, it is proposed that all data extracted from MedWatcher Social and initial findings will be treated as
potential hypotheses for research purposes only.
Pilot Social Data Analysis for Drug List
Data used for
refining algorithms
(WP2A) and
development of
signal detection
methods (WP2B)
Data exported
by MAHs for
evaluation and
analysis
Data used for
scientific impact
evaluation (WP4)
Proposed Process Flow
Validate
Findings from
research > • new insights
on drug
safety
profiles
• Drug Use
• Prescription
patterns
• Patient
compliance
• PV best
practices
Report to
Health
Authorities
,
Publication
s (All)
Signal
Analysis
and Risk
Managem
ent
(MAHs)
Share New
insights
with
Patient
Groups via
Mobile
(WP3A)
(HAs)
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
Key points: pharmacovigilance of social media
• Secondary data use – contact with reporter not foreseen
• Masking of pilot drugs for unbiased study
• Anonymous background data for comparative analysis
• Action after validation of process and signal algorithms
• Data sharing between Work Packages during pilot
• Use of findings to guide policy decisions
• Publication of results after completion of research
Proposal is aligned with secondary use of data per GVP VI
Relevance of Data Protection for Health-Apps
“The close interaction with the operating system allows apps to access significantly more data than a traditional internet browser.”
“Apps are able to collect large quantities of data from the device (location data, data stored on the device by the user and data from the different sensors) and process these in order to provide new and innovative services to the end user.”
Personal data
• “Any information relating to an identified or identifiable natural person (data subject)
Health data (sensitive, personal)
• Any personal data closely linked to the health status of a person, such as genetic data or data on a person’s consumption of medicines
Personal Data in Apps
• Automatically generated by the device, (e.g. geolocation data, network settings, IP address)
• Generated by the user through apps (photos, notes, contact lists)
• Generated by the apps (e.g. browsing history)
32
Data privacy - Who am I?
Age: 50-year-old
Gender: Male
Suspected ADR: Respiratory arrest
Suspect drugs: Diprivan, pethidine, alprazolam & sertraline
PMH: Low BMI, vitiligo, lupus
Outcome: Fatal
http://en.wikipedia.org/wiki/Death_of_Michael_Jackson
Occupation: Rockstar
Narrative: Patient was being treated by his personal physician at his mansion. Administration of propofol led to respiratory arrest and paramedics were called to assist. The resulting court case saw the attending doctor found guilty of involuntary manslaughter.
33 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
Who am I? Personal medical data
Age: 41 years
Gender: Male
Suspected ADR: Possible drug interaction with alcohol
Suspect drugs: Fluoxetine, tiapride, and albendazole
PMH: Alcoholism?
Outcome: Fatal
Occupation: Deputy Head of Security, Ritz Hotel, Paris
Narrative: Patient was driving a Mercedes at high speed in the Pont d’Alma tunnel in Paris. He lost control of the car and crashed, killing himself and two passengers.
34 | Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social Media|
Business Use Only
Summary and vision of the future... Early challenges are being overcome, now we need to generate more
substantive data! Patients are increasingly aware of safety information
Rapid growth of apps & digital media sites as MAHs exploit opportunities
Potential for 2-way interactions is hard to ignore, so mobile reporting & listening capabilities will increase
MAHs must offer real benefits or advantages to patients, and manage the risk of receiving AE reports
Important that pharmacovigilance is carefully managed
Web-RADR provides the opportunity to shape the future, and to work with stakeholders to improve public health
1995
2014
Social media management strategy
36
2016
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
Questions?
| Streamlining Multi-Centre Research May 2015 Workshop | Karen Whitelock & Dave Lewis|6 May 2015 | PV & Social
Media| Business Use Only
37