Mining Online Communities and Social Networks for Safety Signals
Mining Online Communities and Social Networks for Safety Signals
2
About Perficient
Perficient is the leading digital transformation consulting firm serving Global 2000 and
enterprise customers throughout North America.
With unparalleled information technology, management consulting, and creative capabilities, Perficient and its Perficient Digital agency deliver vision, execution, and value with outstanding digital experience, business optimization, and industry solutions.
3
Perficient ProfileFounded in 1997
Public, NASDAQ: PRFT
2015 revenue $473.6 million
Major market locations:Allentown, Atlanta, Ann Arbor, Boston, Charlotte, Chattanooga, Chicago, Cincinnati, Columbus, Dallas, Denver, Detroit, Fairfax, Houston, Indianapolis, Lafayette, Milwaukee, Minneapolis, New York City, Northern California, Oxford (UK), Southern California, St. Louis, Toronto
Global delivery centers in China and India
3,000+ colleagues
Dedicated solution practices
~95% repeat business rate
Alliance partnerships with major technology vendors
Multiple vendor/industry technology and growth awards
Rodney LemeryDirector, Safety and PharmacovigilancePerficient• 20+ Years in Life Sciences• BS in Biotechnology• MPH in International Epidemiology• PhD in Epidemiology
5
• Overview of adverse drug reactions and signal detection
• Regulatory climate surrounding social media usage in pharmacovigilance
• Summary of literature on digital frameworks for using social media data in pharmacovigilance
• Limitations and challenges in using these digital frameworks for using social media data in pharmacovigilance
• Next steps
Agenda
6
• “Unintended, harmful response suspected to be caused by the drug taken under normal circumstances” (Lee, 2006)
• In the U.S. alone, ADRs are estimated to account for ~100,000 deaths annually (Lazarou, Pomeranz & Corey, 1998)
Overview of Adverse Drug Reactions and Signal Detection
7
Signals are considered to be both previously unknown associations and new aspects about an already known association (Harmark, et. Al., 2016)
Overview of Adverse Drug Reactions and Signal Detection
8
Overview of Adverse Drug Reactions and Signal Detection
One qualitative way to evaluate the signals we receive, is using the SNIP methodology:• Strength• Newness• Importance• Prevention
9
According to a WHO publication (2002), the changing face of pharmacovigilance includes the following:• Improve patient care • Improve public health and safety • Contribute to the risk/benefit• Promote understanding of pharmacovigilance to
the publicWHO. (2002). The Importance of Pharmacovigilance. Safety Monitoring of Medicinal Products. Geneva: World Health Organization.
Overview of Adverse Drug Reactions and Signal Detection
10
Overview of Adverse Drug Reactions and Signal Detection
Signals originate from clinical and post marketed data with limitations specific to each of these areas:• Clinical Trials
– Tend to be small– Not diverse
• Demographics (race, gender etc.)• Comorbidities• Concomitant products
• Post Marketing– Spontaneous reporting systems
• Under-reporting *– Electronic Health/Medical Records– Social Media**
11
Current Regulatory Climates for Use of Social Media in Pharmacovigilance
FDA• No regulatory requirements specific to mining social media• Guidance on analyzing patient reported outcomes• FDASIA (2012) and the release of a strategic plan that emphasizes innovative collection and analysis of
post-market dataEMA• GVP guideline (2012)
– In 2014 Module VI updated and mandates regularly screening of websites under its control– The same GVP stipulates that it is considered good practice for the MAH to monitor external sites such
as patient support or special diseases group sites – When made aware, the GVP suggests ADRs be handled in the same manner as a spontaneous report– In 2016 Module VI has been issued in DRAFT and changes the definition of a identifiable reporter
• Requires qualification (ie. physician, nurse, patient etc.) and only one of the following:• Name, address, phone
12
NON-REGULATORY Supporting Initiatives• Strengthening Collaboration for Operating
Pharmacovigilance in Europe (SCOPE) • Raise awareness of national reporting
systems for AE reporting by consumers in Europe
• http://www.scopejointaction.eu/• Innovative Medicines Initiative (IMI) funded the WEB-
RADR project• Conduct scientific research into the use of
social media networks and to develop dedicated applications (Apps) for reporting ADRs to the National Competent Authorities in Europe
• http://web-radr.eu/
Current Regulatory Climate for Useof Social Media in Pharmacovigilance
13
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
A recent meta-analysis of 22 studies published in the literature summarized the efforts and characteristics of social media pharmacovigilance activities and provided a comprehensive framework for conducting this type of research in the future (Sarkera, Ginn, Nikfarjama, et.al., 2015)
Sarkera, A., Ginn, R., Nikfarjama, A., O’Connora, K., Smithc, K., Jayaramanb, S., Upadhayab, T., Gonzaleza, G.. (2015). Utilizing social media data for pharmacovigilance: A review. Journal of Biomedical Informatics, 54; pp. 202–212
14
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
15
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
Table 2 – Identified Sources for the 22 Studies
16
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
17
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
Table 3 – Coding of Event Terms to Various Lexicons
Some studies used phonetic spelling dictionaries to try and ensure proper identification of medicinal products.
18
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
19
• “….wide awake…”
• “….it feels like the Sahara desert in my mouth.”
• “I take it for diarrhea.” While another may say, “Had to stop treatment, it was causing diarrhea.”
• “Well played tysabri...kicking butt #nosleep”
• This cipro is totally "killing" my tummy .. hiks..
• “Over-eaten again just before bed. Stuffed. Good chance I will choke on my own vomit during sleep. I blame #Olanzapine #timetochange #bipolar”
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
20
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
Even this rather robust model doesn’t incorporate the act of evaluating and potentially reporting on the identified ADR through regulatory channels.
21
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
• We are proposing an augmentation to this framework that would allow an organization to evaluate the quality of the identified ADR and assess its reportability to a regulatory authority or partner.
• Freifeld, Brownstein, Menone, et. Al. (2014) coined the phrase “Proto-AE” to explain identifiable event terms in social media that had not been confirmed as actual adverse drug reactions.
Freifeld, C.C., Brownstein, J.S., Menone, C.M., Bao, W., Filice, R., Kass-Hout, T., and Dasgupta, N.. (2014). Digital Drug Safety Surveillance: Monitoring Pharmaceutical Products in Twitter. Drug Safety 37:343–350
22
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
We suggest the term proto-AE could be a useful identifier to relate the pre-reporting terms selected through the ADR identification process.
23
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
24
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
Carbonell, P., Mayer, M.A., Bravo, A.. (2015). Exploring brand-name drug mentions on Twitter for pharmacovigilance. Studies in Health Technology and Informatics. 210:55-9.
Freifeld, C.C., Brownstein, J.S., Menone, C.M., Bao, W., Filice, R., Kass-Hout, T., and Dasgupta, N.. (2014).
Digital Drug Safety Surveillance: Monitoring Pharmaceutical Products in Twitter. Drug Safety
37:343–350
25
Summary of Literature on Digital Frameworks for Using Social Media Data in Pharmacovigilance
Sarkera, A., Ginn, R., Nikfarjama, A., O’Connora, K., Smithc, K., Jayaramanb, S., Upadhayab, T., Gonzaleza, G.. (2015). Utilizing social media data for pharmacovigilance: A review. Journal of Biomedical Informatics, 54; pp. 202–212
26
Limitations and Challenges in Using These Digital Frameworks
for Social Media Data in PharmacovigilanceETHICAL CONCERNS
Use of identifiable data like geocode location on posting, username and other potentially personally identifiable information
Neglect of under-represented members of the online community; less computer literate, lack access to the internet, or have their social media usage censored
CHALLENGESADRs may be referred to using creative idiomatic expressions or terms not found within
existing medical lexicons (“….it feels like the Sahara desert in my mouth.”)The informal nature of social media results in a prevalence of poor grammar, spelling
mistakes, abbreviations and slangDifferentiate between indication and adverse event
Drugs may be described by their brand names, active ingredients, colloquialisms or generic drug terms (e.g. ‘antibiotic’)
27
Next Steps – General Support• Perficient can assist with general strategy in implementing a methodology for social media
monitoring and reporting• Support the design and conduct analysis of a social media targeted project (by active
substance or event of interest)• Use of innovative technology to augment the social media framework your company
currently uses
QuestionsType your question into the chat box
29
• How to Review, Cleanse, and Transform Clinical Data in Oracle InForm | registerDecember 8, 2016
• Leveraging Oracle IDMP Enterprise Foundation Suite for Regulatory Compliance | registerJanuary 12, 2017
Follow Us Online• Perficient.com/SocialMedia• Facebook.com/Perficient• Twitter.com/Perficient_LS• Blogs.perficient.com/LifeSciences
Thank You
31
ReferencesCarbonell, P., Mayer, M.A., Bravo, A.. (2015). Exploring brand-name drug mentions on Twitter for pharmacovigilance. Studies in
Health Technology and Informatics. 210:55-9.Chokor, A., Sarker, A., Gonzalez, G. (2016). Mining the Web for Pharmacovigilance: the Case Study of Duloxetine and Venlafaxine.
Masters project report retrieved on November 2, 2016 from https://arxiv.org/abs/1610.02567Duh, M.S., Cremieux, P., Van Audenrode, M., Vekeman, F., Karner, P., Zhang, H., and Greenberg, P. (2016). Can social media data
lead to earlier detection of drug-related adverse events? Pharmacoepidemiology and Drug Safety, ePubForrow, S., Campion, D. M., Herrinton, L. J., Nair, V. P., Robb, M. A., Wilson, M., & Platt, R. (2012). The organizational structure and governing principles of the Food and Drug Administration's Mini‐Sentinel pilot program. Pharmacoepidemiology and drug safety,
21(S1), 12-17. Freifeld, C.C., Brownstein, J.S., Menone, C.M., Bao, W., Filice, R., Kass-Hout, T., and Dasgupta, N.. (2014). Digital Drug Safety
Surveillance: Monitoring Pharmaceutical Products in Twitter. Drug Safety 37:343–350Härmark, L., Raine, J., Leufkens, H., Edwards, I. R., Moretti, U., Sarinic, V. M., & Kant, A. (2016). Patient-Reported Safety Information:
A Renaissance of Pharmacovigilance?. Drug safety, 39(10), 883-890.Hazell, L., Shakir, S.A.. (2006). Under-reporting of adverse drug reactions : a systematic review. Drug Safety. 29(5):pp. 385-96.Lazarou, J, Pomeranz, B.H., Corey, P.N.. (1998). Incidence of adverse drug reactions in hospitalized patients: a meta-analysis of
prospective studies. JAMA. 279(15):pp. 1200-5.Lengsavath, M., Dal Pra, A., de Ferran, A. M., Brosch, S., Härmark, L., Newbould, V., & Goncalves, S. (2016). Social Media Monitoring
and Adverse Drug Reaction Reporting in Pharmacovigilance An Overview of the Regulatory Landscape. Therapeutic Innovation & Regulatory Science, 2168479016663264.
O’Connor, K., Pimpalkhute, P., Nikfarjam, A., Ginn, R., Smith, K. L., & Gonzalez, G. (2014). Pharmacovigilance on Twitter? Mining Tweets for Adverse Drug Reactions. AMIA Annual Symposium Proceedings, 924–933.
Topaz, M., Lai, K., Dhopeshwarkar, N., Seger, D.L., R., Sa’adon, Goss, F., Rozenblum, R., Zhou, L.. (2015). Clinicians’ Reports in Electronic Health Records Versus Patients’ Concerns in Social Media: A Pilot Study of Adverse Drug Reactions of Aspirin and
Atorvastatin Drug Safety